This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
value to near 60 degrees. Glycine readily fits into all structures and for that reason does not favor helix formation in particular. Can one predict the secondary structure of proteins by usin g this knowl edge of the conformational preferences of amino acid residues? Predictions of secondary structure adopted by a stretch of six or fewer residues have proved to be from about 60% to 70% accurate. What stands in th e way of more accurate prediction ? Note that the confo rmational preferences of amino acid residues are not tipped all the way to one structure (see Table 2.3). For example, glutamate, one of the strongest h elix formers, prefers a helix to i3 strand by only a factor of two . The preference ratios of most other residues are smaller. Indeed , some penta - and hexapeptide sequences have been found to adopt one structure in on e protein and an entirely different structure in another (Figure 2.60). Hence, some amino acid sequences do not unique! y determine second ary structure. Tertiary interactions interactions between residues th at are far apart in th e sequence may be decisive in specifying the secondary structure of some segments. T he context is often cru cial in determining the con ~ Figure 2.60 Alternative conformations of a peptide fo rm ati onal outcom e. The conformation of a protein 'i.t sequence. Many sequences can ado pt alte rnati ve evolved to work in a particular environment or context. conforrnations in different prote ins. Here t he sequence VDLLKN Substantial improvem ents in secondary structure predicshown in red assumes an a helix in one protein cont ext (left) and tion can be achi eved by usin g famili es of related sea I> strand in anot her (right). [Drawn f rom (left ) 3WRP.pdb and quences, each of which adopts the sam e structure. (right) 2HLApdb. ]
Protein Misfolding and Aggregation Are Associated with Some Neurological Diseases
.V..
U ntil quite recently, all infectious diseases were believed to be tran smitted by eith er viruses or bacteri a. In one of th e great surprises in modern medicine, certain infectious n eurological diseases were found to be transmitted by agents that were similar in size to viruses but consisted only of protein. These diseases include bovine spongiform encephalopathy (com monly referred to as mad cow disease ) and the analogous diseases in other organisms, including Creutzf eldt-}akob disease (C] D ) in human beings and scrapie in sheep . The agents causin g these di seases are term ed prions. Th e leading proponent of the hypothesis that diseases can be tran smitted purely by proteins, Stanley Prusiner, was awarded the Nobel Prize in physiology or medicine in 1997. Examination of th ese infecti ous agents reveal ed the foll owing character• lsllcs: •
1. The tran smissible agent con sists of aggregated forms of a specific pro tein. The aggregates have a range of m olecular weights. 2. The protein aggregates are resistant to treatment with agents that degrade mos t proteins. 3. T he protein is largely or completely derived from a cellular protein, call ed PrP, that is normally presen t in the brain.
54 CHAPTER 2 Protein Composition and Structure
How does the structure of the protein in the aggregated form differ from that of the protein in its normal state in the brain? The structure of the normal cellul ar protein PrP contains extensive regions of a helix and relatively little l3 -strand structure. T he structure of the form of the protein present in infected brains, termed Prpsc, has not yet been determined because of chal lenges posed by its insoluble and heterogeneous nature. However, a variety of evidence indicates that some parts of the protein that had been in a-helical or turn conformations have been converted into l3 -strand conformations. Th e 13 strands of one protein link with those of another to form 13 sheets joining the two proteins and leading to the formation of aggregates. These fibrous protein aggregates are often referred to as amyloid forms. With the reali zation that the infectious agent in prion diseases is an aggregated form of a protein that is alread y present in the brain, a model for disease transmission emerges (Figure 2.61 ). Protein aggregates built of abnormal forms of PrP act as nuclei to which other PrP molecules attach. Prion di seases can thus be transferred from one individ ual organi sm to another through the transfer of an aggregated nucl eus, as likely happened in the mad cow disease outbreak in the U nited Kingdom in the 1990s. Cattle fed on animal feed containing material from diseased cows developed the disease in turn. PrPSC nucleus
DODD Figure 2.61 The protein-only model for prion~disease transmission. A nucleus consisting of proteins in an abnormal conformation grows by the addition of proteins from the norma l pool.
DDDDDDDDDDDDDD Normal PrP pool
Amyloid fibers are also seen in the brains of patients with certain noninfectious neurodegenerative diseases such as Alzheimer and Parkinson dis eases. For example, the brains of patients with Alzheimer disease contain protein aggregates called amyloid plaques that consist primarily of a single polypeptid e termed AI3 . This polypeptide is derived from a cellular protein amyloid precursor protein , or APP, through the action of specific proteases. Pol ypeptide AI3 is prone to form insoluble aggregates. Despite the difficul ties posed by the protein's insolubili ty, a detail ed structural model for AI3 has been deri ved through the use of NMR (nuclear magnetic resonance) techniques that can be applied to solids rather than materials in solution . As expected , the structure is rich in 13 strands, wh ich come together to form extended parallell3 -sheet structures (Figure 2.62).
Figure 2.62 A structure of amyloid fibers . A detailed model for Aj3 f ibrils deduced from soli d -state NMR studies shows that protein aggregati on is due to the formati on of large parallel j3 sheet s. [From A. T. Petkova. Y. Ishi i. J. J. Balbach, 0. N. An tzukin, R. D. Leapman, F. Delagio, and R. Tycko, Proc, Na t l . Acad. Sci. U.S.A.
99(2002): 16742-16747.]
How do such aggregates lead to the death of the cells that harbor them? The answer is still controversial. One hypothesis is that the large aggregates themselves are not toxic, but instead smaller aggregates of the same proteins may be the culprits, perhaps damaging cell membranes.
Protein Folding Is a Highly Cooperative Process
100
.-c:C1J ~
o
As stated earlier, proteins can be denatured by heat or by chemical denatu rants such as urea or guanidinium chloride. For many proteins, a comparison o ~==~~------of the degree of unfolding as the concentration of denaturant increases [Denaturantl reveals a relatively sharp transition from the folded, or native, form to the unfolded, or denatured form, suggesting that only these two conformational Figure 2.63 Transition from folded to states are present to any significant extent (Figure 2.63). A similar sharp unfolded state. Most proteins sho w a sharp tran sitio n from t he folded to the transition is observed if denaturants are removed from unfolded proteins, al unfo lded form o n treatment with lowing the proteins to fold. increasing concentrat ions of denat urant s. The sharp transition seen in Figure 2.63 suggests that protein folding and unfolding is an"all or none" process that results from a cooperative transition. For example, suppose that a protein is placed in conditions under which some part of the protein structure is thermodynamically unstable. As this part of the folded structure is disrupted, the interactions between it and the remainder of the protein will be lost. The loss of these interactions, in turn, will destabilize the remainder of the structure . Thus, conditions that lead to the disruption of any part of a protein structure are likely to unravel the protein completely. The stTuctural properties of proteins provide a clear rationale for the cooperative transition. The consequences of cooperative folding can be illus Unfolded 100 trated by considering the contents of a protein solution J!. under conditions corresponding to the middle of the tran sition between the folded and the unfolded forms . Under "-C1J these conditions, the protein is "half folded." Yet the solu "0 c: 50 tion will contain no half-folded molecules but, instead, will " c: .be a 50 / 50 mixture of fully folded and fully unfolded mol C1J 0 ecules (Figure 2.64). Although the protein may appear to "behave as if it exists in only two states, at an atomic level, Folded this simple two -state existence is an impossibility. o ~~~~--------[Denaturant] ~ Unstable, transient intermediate structures must exist be tween the native and denatured state (p . 56). Determining Figure 2.64 Components of a partially denatured protein the nature of these intermediate structures is an intense solution. In a ha lf-unfo lded protein so lution, half the mo lecules are fully folded and half are fu lly unfo lded. area of biochemical research. ~
~
~
Proteins Fold by Progressive Stabilization of Intermediates Rather Than by Random Search How does a protein make the transition from an unfolded structure to a unique conformation in the native form ? One possibility a priori would be that all possible conformations are tried out to find the energetically most favorable one. How long would such a random search take? Consider a small protein with 100 residues . Cyrus Levinthal calculated that, if each residue can assume three different conformations, the total number of struc tures would be 3 '00 , which is equal to 5 X 10 47 If it takes 10- 13 s to convert aile structure into another, the total search time would be 5 X 1047 X 10- 13 s, which is equal to 5 X 1034 s, or 1.6 X 10 27 years. Clearly, it would take much too long for even a small protein to fold properly by randomly trying out all possible conformations. The enormous difference between calculated and actual folding times is called Levinthal 's paradox , This para dox clearly reveals that proteins do not fold by trying every possible 55
56 CHAPTER 2 Protein Composition and Structure
1800 !1S,t hlQ 2000 ~h~hl n' 2200 2400
1
is
MKthinn:~s:*.~~~
2600 2800
is like a wease •
2883
200 10 - gIW4{{ c u!kO{d6 jS! N1EyUx)p 400 " W ·
~ \ "'R . I, and do not move when li p
= 1.
A technique called z onal, band, or most commonly gradient centrifugation can be used to separate proteins with different sedimentation coefficients. The first step is to form a density gradient in a centrifuge tube. Differing proportions of a low-density solution (such as 5% sucrose) and a high-density solution (such as 20% sucrose) are mixed to create a linear gradient of sucrose concentration ranging from 20% at the bottom of the tube to 5% at the top (Figure 3.15). The role of the gradient is to prevent convective now. A small volume of a solution containing the mixture of proteins to be separated is placed on top of the density gradient. When the rotor is spun, proteins move through the gradient and separate accordin g to their sedimentation coefficients. The time and speed of the centrifugation is determined empirically. The separated bands, or zones, of protein can be harvested by making a hole in the bottom of the tube and collecting drops . The drops can be measured for protein content and catalytic activity or another functional property. This sedimentation-velocity technique readily separates proteins differing in sedimentation coefficient by a factor of two or more. The mass of a protein can be directly determined by sedimentation equilibrium, in which a sample is centrifuged at relatively low speed so that sed imentation is counterbalan ced by diffusion . The sedimentation -equilibrium technique Jar determining mass is very accurate and can be applied without denaturing the protein. Thus the native quaternary structure oj multimeric proteins is preserved. In contrast , SDS- polyacrylamide gel electrophoresis
low-density solution
Figure 3.15 Zonal centrifugation. The steps are as follows: (A) form a density gradient. (6) layer the sample o n top of t he gradient, (C) place the tube in a swinging-bucket rotor and centrifuge it. and (D) collect the samples. [After D. Freifelder. Physical Biochemistry. 2d ed. (w. H. Freeman and Company. 1982). p . 397.)
High-density Separation by sedimentation
solution
coefficient layering of sample
Fractions collected through hole in bottom of tube
c')
• •
Rotor
Centrifuge tube ---.
•
Density gradient ___
(A)
(8)
(C)
(D)
78 CHAPTER 3 Exploring Proteins and Proteomes
provides an estimate of the mass of dissociated polypeptide chains under de naturing conditions. Note that, if we know the mass of the dissociated components of a multimeric protei n as determined by SDS- polyacrylamide analysis and the mass of the intact multimeric protein as determined by sedimentation -equilibrium analysis, we can determine the number of copies of each polypeptide chain present in the multimeric protein.
3.2
Amino Acid Sequences Can Be Determined by Automated Edman Degradation
After a protein has been purified to homogeneity, a determination of the protein's amino acid sequence, or primary structure, is often desirable. Let us examine first how we can sequence a simple peptide, such as Ala-Gly-Asp-Phe-Arg-Gly The first step is to determine the amino acid composition of the peptide. The peptide is hydrolyzed into its constituent amino acids by heating it in 6 M HCl at 110°C for 24 hours. Amino acids in hydrolysates can be separated by ion-exchange chromatography. The identity of the amino acid is revealed by its elution volume, which is the volume of buffer used to remove the amino acid from the column (Figure 3. 16), and its quantity is revealed by reaction with ninhydrin. Amino acids treated with ninhydrin give an intense blue color, except for proline, which gives a yellow color because it contains a secondary amino group. The concentration of an amino acid in a solution, after heating with ninhydrin, is proportional to the optical absorbance of the solution . This technique can detect a microgram (10 nmol) of an amino acid, which is about the amount present in a thumbprint. As little as a nanogram (10 pmol) of an amino acid can be detected by replacing ninhydrin withfiuorescamine, which reacts with the a -amino group to form a highly fluorescent product (Figure 3.17). A comparison of the chromatographic patterns of ou r sample hydrolysate with that of a standard mixture of amino acids would show that the amino acid composition of the peptide is (Ala, Arg, Asp, G lY2, Phe) The parentheses denote that this is th e amino acid composition of the pep tide, not its sequence. Figure 3.16 Determination of amino acid composition. Different amino acids in a peptide hydrolysate can be separated by ion-exchange chro matography on a sulfonated polystyrene resin (such as Do wex -50). Buffers (in this case, sodium citrate) of increasing pH are used to elute the amino acids from the column. The amount of each amino acid present is determined from the absorbance. Aspartate, which has an acidic side cha in, is first to emerge. whereas arginine, which has a basic side chain, is the last. The original peptide is revealed to be composed of one aspartate. one alanine, one phenylalanine, one arginine, and two glycine residues.
ELUTION PROFILE OF PEPTIDE HYDROLYSATE Gly
Ala
Asp
Arg
Phe
.,u
.
c: ..c ~
a
II
~ . !!> I
«:
~
-' IZ
ELUTION PROFILE OF STANDARD AMINO ACIDS ------------~)
pH 3.25 0.2 M Na citrate
I II
------------------------~) ------------~)
pH 5.28 0.35 M Na citrate
pH 4.25 0.2 M Na ci trate
Elution volume
'
79 3.2 Amino Acid Sequence Determination
o R
N Hz
"-----
o
o
)
Figure 3.17 Fluorescent derivatives of amino acids. Fluorescamine reacts with the a-amino group of an amino acid to form a fluorescent derivative.
OH Amine derivative
Fluorescamine
o
TIle next step is to identify the N -terminal amino acid. Pehr Edman devised a method for labeling the amino-terminal residue and cleaving it from the peptide without disrupting the peptide bonds between the other amino acid residues. The Edman degradation sequentially removes one residue at a time from the amino end of a peptide (Figure 3.18). Phenyl isothiocyanate re acts with the uncharged terminal amino group of the peptide to form a phenylthiocarbamoyl derivative. Then, under mildly acidic conditions, a cyclic derivative of the terminal amino acid is liberated, which leaves an intact peptide shortened by one amino acid. The cyclic compound is a phenylthiohydantoin (PTH ) amino acid, which can be identified by chromatographic procedures. The Edman procedure can then be repeated on the shortened peptide, yielding another PTH- amino acid, which can
OH OH
o Ninhydrin
H H
EDMAN OEGRADATION
0- 2
3
+
o
4r--2.
1
Gly
Phenyl isothiocyanate
Labeling First round
234
1
H N
Release
1
2
Labeling
H N S
345
"\ H , C~ H
1
H H / Asp- Phe - Arg - Gly
N"'-
I
H
o
Labeling
345
1
Release
>-
Second round
Release
HH NH N .___-" " H
-
"'1/
Asp - Phe- Arg - Gly
o
o PTH- alanine
Figure 3.18 The Edman degradation. The labeled amino-terminal residue (PTH - alanine in the first round) ca n be released without hydrolyzing the rest of the peptid e. Hence, the amino-terminal residue of the shortened peptide (Gly-Asp-Phe-Arg-Gly) can be determined in the second round . Three more rounds o f the Edman degradation reveal the complete sequence of the original peptide.
Peptide shortened by one residue
80
again be identified by chromatography. Three more rounds of the Edman degradation will reveal the complete sequence of the original hexapeptide. The development of automated sequencers has markedly decreased the time required to determine protein sequences. One cycle of the Edman degradation the cleavage of an amino acid from a peptide and its identification is carried out in less than 1 hour. By repeated degradations, the amino acid sequence of some 50 residues in a protein can be determined. Gas-phase sequenators can analyze picomole quantities of peptides and proteins, using high-pressure liquid chromatography to identify each amino acid as it is released (Figure 3.19). This high sensitivity makes it feasible to analyze the sequence of a protein sample eluted from a single band of an SDS-polyacrylamide gel.
CHAPTER 3 Exploring Proteins and
Proteomes
0.06
E c:
.,.on
-'" '"N
0.04
ClI
u
c:
.n 0.02 0 ~
.n
..:
o
4
8
12
16
Proteins Can Be Specifically Cleaved into Small Peptides to Facilitate Analysis
20
Elution time (minutes)
In principle, it should be possible to sequence an entire protein by using the Edman method. In practice, the peptides cannot be much longer than about 50 residues, because not all peptides in the reaction mixture release the amino acid derivative at each step. For instance, if the efficiency of release for each round were 98%, the proportion of "correct" amino acid released 60 after 60 rounds would be (0.98 ), or 0.3 a hopelessly impure mix. This obstacle can be circumvented by cleaving a protein into smaller peptides that can be sequenced. In essence, the strategy is to divide and conquer. The key is to cleave the protein into a small number of pure fragments. Specific cleavage at particular amino acid types can be achieved by chemical or enzymatic methods. For example, cyanogen bromide (CNBr) splits polypeptide chains only on the carboxyl side of methionine residues (Figure 3.20). A protein that has 10 methionine residues will usually yield 11 peptides on cleavage with CNBr. Highly specific cleavage is also obtained with trypsin, a proteolytic enzyme secreted by the pancreas. Trypsin cleaves polypeptide chains on the carboxyl side of arginine and lysine residues (Figure 3.21 and p . 246). A protein that contains 9 lysine and 7 arginine residues will usuaIIy yield 17 peptides on digestion with trypsin. Each of these tryptic peptides, except for the carboxyl-terminal peptide of the protein, will end with either arginjne or lysine. Table 3.3 gives several other ways of specifically cleaving polypeptide chains.
Figure 3.19 Separation of PTH-amino acids. PTH-amino acids can be rapidly separated by high-pressure liquid chromatography (HPLC). In this HPLC profile, a mixture of PTH- amino acids is cl early resolved into its components. An
unknown amino acid can be identified by its elution position relative to the known ones.
o Figure 3.20 Cleavage by cyanogen bromide. Cyanogen bromide cleaves polypeptides on the ca rboxyl side of
o
o
H
+ CNBr
N H
o H
>
o
Methionine
Homoserine
methionine residues.
Figure 3.21 Cleavage by trypsin. Trypsin hydrolyzes polypeptides on the carboxyl side of arginine and lysine
residues.
o
lactone
o
lysine
lysine
or
or
•
•
arginine
o
H
.l;'> H
R,
o
•
arginine
o
H
H
N N H
o
•
Trypsin
>
R,
H
N H
•
,
0
•
,0
-
+
T.BLE 3.3 Specific cleavage of polypeptides Cleavage sit e
Reagent
81 3.2 Amino Acid Sequence Determination
Chemical cleavage Cyanogen bromide
Carboxyl side o f methionine residues
O ~ lodosoben zoate
Carboxyl side of tryptophan residues
Hydroxylamine 2-Nit ro-5-thiocyanobenzoate
Asparagine-glycine bonds Amino side of cysteine residues
Enzymatic cleavage Trypsin
Carboxyl side of lysine and arginine residues Carboxyl side of arginine residues Carboxyl side of aspartate and glutamate residues (glutamate
Clostripajn Staphylococca l protease
only under certain conditions)
Thrombin Chymotrypsin
Carboxyl side of arginine Carboxyl side of tyrosi ne, tryptophan. phenylalanine, leucine,
Carboxypeptidase A
and methionine Amino side of (-term inal amino acid (not arginine. lysine. or
S,- - S
proline) R-__ (
The pep tides obtained by specific chemical or enzymatic cleavage are separated by some type of chromatography. The sequence of each purified peptide is then determined by the Edman method. At this point, the amino acid sequences of segments of the protein are known, but the order of these segments is not yet defined . How can we order the peptides to obtain the primary structure of the original protein? The necessary additional information is obtained from overlap peptides (Figure 3.22). A second enzyme is used to split the polypeptide chain at different linkages. For example, chymotrypsin cleaves preferentially on the carboxyl side of aromati c and some other bulky nonpolar residues (p . 247). Because these chymotryptic peptides overlap two or more tryptic peptides, they can be used to establish the order of the peptides. The entire am ino acid sequence of the polypeptide chain is then known .
c-
H, H2 Disulfide-linked chains SH
Chymotryptic peptide
Ala - Ala - Trp - Gly - lys
Va l- Lys - Ala - Ala - Trp
Thr - Phe
,
Val - lys
Tryptic peptide
l
HS
HO
OH
Dithiothreitol (excess)
5
5
HO SH
Tryptic peptides
R'
R __ - / C
OH
HS
+
-R
C
H, H, Separated reduced chains
Tryptic peptide
Thr - phe- Val - Lys - Ala - Ala - Trp - Gly - lys
I
/
H, c '--- c AU AU
AG U UA cG The linked A U AU nucleotides CG highlighted in AU part 8 AG Cu uc u Figure 4.20 Complex structure of an RNA molecule. A Single-stranded RNA molecule may f old back o n it self t o f o rm a compl ex stru cture. (A) The nucleotide sequence showing Watson - Crick base pa irs and o ther no nstandard base pairings in stem -loop stru ctures. [B) The three -dimensional structure and one important long-range interaction between three bases. In the three-dimensi o na l structure to the left, cytidine nucleotides are shown in blue, adenosine in red , guanosine in black. and uridine in green. Hydrogen bonds within th e Watson - Crick base pair are sho wn as dashed black lines; additio nal hydro gen bo nds are shown as dashed green lines.
117
stranded DNA molecule cannot. Indeed, the complexity of some RNA molecules rivals t h at of proteins, and these RNA molecules perform a number of functions that had formerly been thought the private domain of • protems.
4.3
4.3 DNA Replicat ion
DNA Is Replicated by Polymerases That Take Instructions from Templates
We now turn to t he molecu lar mechanism of DNA replication . The full repl ication machinery in a cell comprises more than 20 proteins engaged in intricate and coordinated interplay. In 1958, Arth ur Kornberg and his colleagues isolated from E. coli the first known of th e enzymes, called DNA polymerases, that promote the formation of the bonds joinin g units of the DNA backbone. E. coli has a number of DNA polymerases, d esignated by roman numerals, that participate in DNA repl ication and rep air (Chapter 28).
DNA Polymerase Catalyzes Phosphod iester-Bond Fo rmat ion DNA polymerases catalyze the step-by -step addition of deoxyribonucleotide units to a DNA chain ( Figure 4.21). The reaction catalyzed , in its simplest form, is (DNA)" + d NT P
(DNA)n+l + PP;
where dN TP stands for any d eoxy ribonucleotide and PP; is a pyrophosphate ion. DNA synthesis has the following characteristics: The reaction requires all four activated precursors that is, the deoxynucleoside 5' -triphosphates dATP, dGTfJ, dCT fJ, and TTP as well as 1.
1+ . Mg- LOn.
2. The new DNA chain is assembled directly on a preexisting DNA template.
DNA polymerases catalyze the formation of a phosphodiester bond efficiently only if the base on the incoming nucleoside triphosphate is complementary to the base on the template strand . Thus, D NA polym erase is a template-directed enzyme that synthesizes a product with a base sequence complementary to that of the template .
3. DNA po/ymemses require a primer to begin synthesis. A primer strand having a free 3 ' -OH group must be already bound to the template strand. The chain-elongation reaction catalyzed by D NA polymerases is a nucle ophilic attack by the 3' -OH terminus of th e growing chain on the innermost
5/
3'
P
dATP
J'
e,
c,
C
G P
T P
P
5'
P
5'
dGTP G , C
)
3'
C, G
A , T
P
figu re 4.21 Polymerization reaction catalyzed by DNA polymerases.
C P
P
5"
PPj
\ /
A
C
,
3'
5'
3'
P
PP,
\./
A
P
)
G, C
c,
A,
G
G
T
C
P
,
P
A P
5'
118 CHAPTER 4 DNA, RNA , and the Flow of Genetic Information
3'
Primer strand
3'
Primer strand
\
/
H2 C
~o
base
2P,
,
v---
0
0
base Cl Z
z
-3 "--
~
~
ct>
0>
ct>
l!i. ~
0>
"
"'-
~ ,
O.~ / :P
/\
° /° H,C
base base
HO
o
PP;
v--5'
0
base base -
HO
5'
Figure 4.22 Chain elongation reaction. DNA polymerases catalyze the formation of a phospho diester bridge,
phosphorus atom of the deoxynucleoside triphosphate (Figure 4.22) . A phosphodiester bridge is formed and pyrophosphate is released. The subsequent hydrolysis of pyrophosphate to yield two ions of orthophosphate (P;) by pyrophosphatase, a ubiquitous enzyme, helps drive the polymerization forward. Elongation of the DNA chain proceeds in the 5' -to-3' direction.
4. Many DNA polymerases are able to correct mistakes in DNA by removing mismatched nucleotides. These polymerases have a distinct nuclease activity that allows them to excise incorrect bases by a separate reaction. This nuclease activity contributes to the remarkably high fidelity of DNA replication, R which has an error rate ofless than 10 - per base pair. The Genes of Some Vi ru ses Are Made of RNA
Genes in all cellular organisms are made of DNA. The same is true for some viruses but, for others, the genetic material is RNA. Viruses are genetic elements enclosed in protein coats that can move from one cell to another but are not capable of independent growth. One well-studied example of an RNA virus is the tobacco mosaic virus, which infects the leaves of tobacco plants. This virus consists of a single strand of RNA (6390 nucleotides) surrounded by a protein coat of 2130 identical subunits. An RNA polymerase that takes direction from an RNA template, called an RNA -directed RNA polymerase, copies the viral RNA . Another important class of RNA virus comprises the retroviruses, so called because the genetic information flows from RNA to DNA rath er than from DNA to RNA. This class includes human immunodeficiency virus 1 (HIV-l), the cause of AIDS, as well as a number of RNA viruses that produce tumors in susceptible animals. Retrovirus particles contain two copies of a single-stranded RNA molecule. On entering the cell, the RNA is copied into DNA through the action of a viral enzyme called reverse tran scriptase (Figure 4. 23). The resulting double-helical DNA version of the viral genome can become incorporated into the chromosomal DNA of the host and is replicated along with the normal cellular DNA. At a later time, the integrated viral genome is expressed to form viral RNA and viral proteins, which assemble into new virus particles.
119 4.4 Gene Expression
Reverse transcriptase
Reverse transcriptase
)
)
Digestion of RNA
Synthesis of DNA complementary to RNA
DNA-RNA hybrid
Viral RNA
4.4
Reverse transcriptase ) Synthesis of second strand of DNA
Double-helical viral DNA
DNA transcript of viral RNA
Gene Expression Is the Transformation of DNA Information into Functional Molecules
The information stored as DNA becomes useful when it is expressed in the production of RNA and proteins. This rich and complex topic is the subject of several chapters later in this book, but here we introduce the basics of gene expression. DNA can be thought of as archival information, stored and manipulated judiciously to minimize damage (mutations). It is expressed in two steps. First, an RNA copy is made that encodes directions for protein synthesis. This messenger RNA can be thought of as a p hotocopy of the original information it can be made in multiple copies, used, and then dis posed of. Second, the information in messenger RNA is translated to syn thesize functio nal proteins. Other types of RNA molecules exist to facilitate this translation. Several Kinds of RNA Play Key Roles in Gene Expression
Scientists used to believe that RNA played a passive role in gene expression, as mere conveyors of information like messenger RNA. However, recent investigations have shown that RNA plays a variety of roles, from catalysis to regulation. Cells contain several kinds of RNA (Table 4.2): 1. Messenger RNA (mRNA) is the template for protein synthesis, or trans lation. An mRNA molecule may be produced for each gene or group of genes that is to be expressed in E coli, whereas a distinct mRNA is produced for each gene in eukaryotes. Consequently, mRNA is a heterogeneous class of molecules. Tn prokaryotes, the average length of an mRNA molecule is about 1.2 kilobases (kb). In eukaryotes, mRNA has structural features, such as stem -loop structures, that regulate the efficiency of translation and lifetime of the mRNA.
TABLE 4.2 RNA molecules in
E. coli
Relati ve am o unt
Sedimentation coefficient
Type
(%)
(5)
Ribosomal RNA (rRNA)
80
23 16 5
15 5
4
Transfer RNA (tRNA ) Messenger RNA (mRNA)
Mass (kd )
Num ber o f nucleotides
x x x x
3700 1700 120 75
1.2 0.55 3.6 2.5
10' 10' 10' 10'
Hetero geneous
Figure 4.23 Flow of information from RNA to DNA in retroviruses . The RNA genome of a retrovirus is converted into DNA by reverse tran scr ipta se, an enzyme brought into the cell by th e infecting vi ru s particle. Reverse tran scriptase possesses several activities and catalyzes the synthesis o f a complementary DN A strand, the digest ion o f th e RNA, and th e subsequent synth esis of the DNA strand.
Kilobase (kb) A unit of length equal to 1000 base pairs of
a double-stranded nucleic aCi d molecu le (or 1000 bases of a single-stranded molecul e). One kiloba se of double-stranded DNA has a contour length of 0.34 jJ.m and a mass of about 660 kd.
2. Transfer RNA (tRNA) carries amino acids in an activated form to the ribosome for peptide-bond formation , in a seq uence dictated by the mRNA template. There is at least one kind of tRNA for each of the 20 amino acids, Transfer RNA consists of about 75 nucleotides (having a mass of about 25 kd ). 3. Ribosomal RNA (rRNA) is the major component of ribosomes (Chapter 30). In prokaryotes there are three kinds of rRNA, called 23S, 16S, and 5S RNA because of their sed imentation behavior. O ne molecule of each of these species of rRNA is present in each ribosome. Ribosomal RNA was once believed to play only a structural role in ribosomes. W e now know that rRNA is the actual catalyst for protein synthesis. Ribosomal RNA is the most abundant of these three ty pes of RNA. Transfer RNA comes next, followed by messenger RNA, which constitutes only 5% of the total RNA. Eukaryotic cells contain additional small RNA molecules. 4. Small nuclear RNA (snRNA) molecules participate in the splicing of RNA exons. •
5. A small RNA molecule is an essenti al component of the signalrecognition particle, an RNA protein complex in the cytoplasm that helps target newly synthesized proteins to intracellular compartments and extracellular destinations.
6. Micro RNA (miRNA ) is a class of small (about 21 nucleotides) nOllcoding RNAs that bind to complementary mRNA molecules and inhibit their translation.
7. Small interfering RNA (siRNA) is a class of small RNA molecules that bind to mRNA and facilitate its degradation. Micro RNA and small interfering RNA also provide scientists with powerful experimental tools for inhibiting the expression of specific genes in the cell. 8. RNA is a component of telomerase, an enzyme that maintains the telomeres (ends) of chromosomes during DNA replication .
In this chapter, we will consider rRNA , mRNA, and tRNA. All Cellular RNA Is Synthesized by RNA Polymerases
The synthesis of RNA from a DNA template is called transcription and is catalyzed by the enzyme RNA polymerase (Figure 4.24). RNA polymerase
Figure 4.24 RNA polymerase. This large enzyme compri ses many subunits, including J3 (red) and W(yellow), which form a "cl aw" t hat holds the DNA t o be transcribed. N o tice that the active site includes a MgH ion (green) at the center of t he stru cture. The curved tubes making up the prote in in the image represent the backbone of the polypeptide chain. [Drawn from IL9Zpdb.]
120
catalyzes the initiation and elongation of RNA chains. The reaction catalyzed by this enzyme is
121 4.4 Gene Expression
, (RNA)n + ! + PP j
(RNA)n + ribonucleoside triphosphate
RNA polymerase requires the following components:
1. A template. The preferred template is double-stranded DNA. Singlestranded DNA also can serve as a template. RNA, whether singl e or double stranded, is not an effective template; nor are RNA- DNA hybrids . 2. Activated precursors. All four ribonucleoside triphosphates UTP, and C TP are required .
ATP, GTP,
2
3. A divalent metal ion. Either Mg2+ or Mn + is effective.
The synthesis of RNA is like that of DNA in several respects (Figure 4.2 5). First, the direction of synthesis is 5' ) 3'. Second, the mechanism of elongation is similar: the 3'-OH group at the terminus of the growing chain makes a nucl eophilic attack on the innermost phosphate of the incoming nucleoside triphosphate. Third, the synthesis is driven forward by the hydrolysis of pyrophosphate. In co ntrast with DNA polymerase, however, RNA polymerase does not require a primer. In addition , RNA polymerase lacks the ability of ])NA polymerase to excise mi smatched nucleotides. All three types of cellular RNA mRNA, tRNA, and rRNA are syn thesized in E. coli by the sam e RNA polymerase according to instructions given by a DNA template. In mammalian cells, there is a division of labor among several different kind s of RNA polymerases. We shall return to these RNA polymerases in Chapter 29.
3'
Primer strand
3'
Primer strand
\
\
/
/
H2 C
base
~O
H2 C
2 Pj
base
o
Z l>
0 Z l>
- H2 0
~
o
PPj
~)
'3" "'"-
OH
0 .'>,... /
-)\ ?
~
(1)
-'" '"
~
"c..
. 0
H2 C
base base ---1
base base
~O
HO
OH
HO
OH
5'
RNA Polymerases Take Instructions from DNA Templates R.\JA polymerase, like the DNA polymerases described earlier, takes instructions [rom a DNA template. The earliest evidence was the finding that the base composition of newly synthesized RNA is the complem ent of that of the DNA template strand, as exemplifi ed by th e RNA synthesized from a template of single-strand ed DNA from t he <j>X1 74 viru s (Table 4.3).
5'
Figure 4.25 Transcription mechanism of the chain-elongation reaction catalyzed by RNA polymerase.
TABLE 4.3 Base composition (percentage) of RNA synthesized from a viral DNA template DNA template (plu s, o r coding, strand of X174)
RNA product
A
25
25
T G C
33
32 23 20
24 18
U A
C G
Hybridization experiments also revealed that RNA synthesized by RNA polymerase is complementary to its DNA template. In these experiments, DNA is melted and allowed to reassociate in the presence of mRNA. RNA- DNA hybrids will form if the RNA and DNA have complementary sequences. The strongest evidence for the fideli ty of transcription came from basesequence studies showing that the RNA sequence is the precise complement of the DNA template sequence (Figure 4. 26).
5' - GCGGCGACGCGCAGUUAAUCCCACAGCCGCCAGUUCCGCUGGCGGCAU - 3'
mRNA
3' - CGCCGCTGCGCGTCAA TTAGGGTGTCGGCGGTCAAGGCGACCGCCGTA - S'
Template strand of DNA
5'- GCGGCGACGCGCAGTT AAT CCCACAGCCGCCAGT T CCGCT GGCGGCAT - 3'
Coding strand of DNA
Figure 4.26 Complementarity between mRNA and DNA. Th e base sequence of mRNA (red) is th e comp lement of t hat of t he DNA template strand (blue). The sequence sho wn here is fro m th e try pto phan o pero n. a segment of DNA containing the genes f or five enzymes that catal yze th e syntheSiS of try pto phan. The other strand of DNA (black) is ca lled th e coding strand because it has th e same sequence as th e RNA transcript except f or t hy m ine (T) in place of uracil (U).
Transcription Begins Near Promoter Sites and Ends at Terminator Sites Consensus sequence Not all ba se sequences o f promoter sites are
id entical. However, they do possess commo n featu res, w hich can be represented
by an
idealized co nsensus seq uence. Each base in
the consensus sequence TATAAT is f ound in
most prokaryoti c promoters. Nearly all promoter seq uences differ fro m thi s co nsensus
sequence at only o ne or two bases.
RNA polymerase must detect and transcribe discrete genes from within large stretches of DNA. What marks the beginning of the unit to be transcribed ? DNA templates contain regions called promoter sites that specifically bind RNA polymerase and determine where transcription begins. In bacteria, two sequences on the 5' (upstream) side of the first nucleotide to be transcribed function as promoter sites ( Figure 4. 27A). One of them, called the Pribnow box, has the consensus sequence TATAAT and is centered at - 10 (10 nucleotides on the 5' side of the first nucleotide transcribed, which is denoted by + 1). The other, called the - 35 region , has the consensus sequence TTGAC A. The first nucleotide transcribed is usually a purine. Eukaryotic genes encodi ng proteins have promoter sites with a TATAAA consensus sequence, called a TATA box or a Hogness box, centered at about - 25 (Figure 4.2 7B). Many eukaryotic promoters also have a
- TO
- 35 DNA template - --
- --
IIGACA - - - - --
- 35 region
CA)
TATAAT -
------
Pribnow box
Start of RNA
Prokaryotic promoter site
- 75 DNA template - - - --r GGNCAATCT
- 25 --r----~
CAAT box (sometimes present)
(B)
+1
+1
TATAAA
TATA box (Hogness box)
Start of RNA
Eukaryotic promoter site
Figure 4.27 Promoter sites for transcription in (A) prokaryotes and (B) eukaryotes. Co nsensus sequences are sho wn. The fir st nucleotide t o be tran scribed is num bered +l. The adjacent nucl eotide o n the 5' side is numbered - l. The sequences shown are th ose of th e coding strand o f DNA .
122
CMT box with a GGNCAATCT consensus sequence centered at about - 75. The transcription of eukaryotic genes is further stimulated by enhancer sequences, wh ich can be quite distant (as many as several kilobases) fro m the start site, on either its 5' or its 3' side. RNA polymerase proceeds along the DNA template, transcribing one of its strands until it synthesizes a terminator sequence. This sequence encodes a termination signal, which in E. coli is a base-paired hairpin on the newly synthesized RNA molecule (Figure 4.28). This hairpin is formed by basepairing of self-complementary sequences that are rich in G and C. Nascent RNA spontaneously dissociates from RNA polymerase when this hairpin is followed by a string ofD residues . Alternatively, RNA synthesis can b e terminated by the action of rho, a protein. Less is known about the termination of transcription in eukaryotes. A more detailed discussion of the initiation and termination of transcription will be given in Chapter 29. The important point now is that discrete start and stop signals for transcription are encoded
in the DNA template. In eukaryotes, the RNA transcript is modified (Figure 4 .29). A "cap" structure is attached to the 5' end, and a sequence of adenylates, the poly(A) tail, is added to the 3' end. These modifications will be presented in detail in Chapter 29 .
123 Gene Expression
4.4
,.----c~
U
C
U
G
(
)
' c. c/ I I A- U I I
e _G I I e _G I I G- e I I e -G I I e-G I I
....... G . C , 5' -C C A C A G
,.
A U U U U- OH
Figure 4.28 Base sequence of the 3' end of an mRNA transcript in E. coli. A stable hairpin structure is foll owed by a sequence of uri dine (U) residues
____ Poly(A} tail
~ Cap 5'
3'
I
I
Cod ing region Figure 4.29 Modification of mRNA. Messenger RNA in eukaryotes is modified after transcription. A nucleotide "cap" structure is added to the 5' end, and a poly(A) tail is added at the 3' end.
Transfer RNAs Are the Adapto r Mo lecule s in Protein Synthesis We have seen that mRNA is the template for protein synthesis. How then does it direct amino acids to become joined in the correct sequence to form a protein? In 1958, Francis Crick wrote: RNA presents mainly a sequence of sites where hydrogen bonding could occur. O ne would expect, therefore , that whatever went onto the template in a specific way did so by forming hydrogen bonds. It is therefore a natural hypothesis that the amino acid is carried to the template by an adaptor mol ecule, and that the adaptor is the part that actuall y fits onto the RNA. Tn its simpl est form, one would require twenty adaptors, one for each amino acid. This highly innovative hypothesis soon became established as fact . The adaptors in protein synthesis are transfer R NAs. The structure and reactions of these remarkable molecules will be considered in detail in C hapter 30 . For the moment, it suffices to note that tRNAs contain an amino acidattachment site and a template-recognition site. A tRNA molecule carries a specific amino acid in an activated form to the site of protein synthesis. The carboxyl group of this amino acid is esterified to the 3' - or 2' -hydroxyl group of the ribose unit at the 3' end of the tRNA chain (Figure 4 .30). The joini ng of an amino acid to a tRNA molecule to form an aminoacyl-tRNA is catalyzed by a specific enzyme called an aminoacyl-tRNA synthetase. This esterification reaction is driven by AT P cleavage. There is at least one specific synthetase for each of the 20 amino acids. The template-recognition site on tRNA is a sequence of three bases called an anticodon (Figure 4 .31). The anticodon on tRNA recognizes a complementary sequence of three bases, called a codon, on mRNA.
\
adenine
0, j
).
..( OH
Figure 4.30 Attachment of an amino acid to a tRNA molecule. The amino acid (shown in blue) is esterified to the 3' -hydroxyl group of the terminal adenylate of tRNA.
-
I~ I
124 CHAPTER 4 DNA, RNA, and the Flow of Genetic Information
Am ino acid-
•
attachment site
A
r0 Phosphorylated 5' terminus
• 5'
..
P
.. .. .. I
-\ :
Figure 4.31 General structure of an aminoacyltRNA. The amino acid is attached at the 3' end of the RNA. The anticodon is the template-recognition site. Notice that the tRNA has a cloverleaf structure with many hydrogen bonds (green dots) between bases.
4.5
~
>
•
:
:
' Coupling
Adivated monomer
base n
3'1--
OMT -
5'
5'
- 1
3'1-- 0 -
5'
5'
Phosphite triester intermediate
Growing chain
Oxidation
Repeat base n - l
base
fl CE,
by I,
n
base
HO
n-
0
o~/"o
1
3'
0
basen
0
•
reSin
•
®
OMT
Deprotection
0
Ip o~/"o
3'
3'
0
0
with dichloroacetic acid 5'
0
I>CE, O
Ip
3'
•
reSlIl
5'
Elongated chain figure 5.6 Solid-phase synthesis of a DNA chain by the phosphite triester method. The activated monomer added to the growing chain is a deoxyribonucleoside J' -phosphoramidite containing a DMT protect ing group o n its 5' -o xygen atom, a ~-cya noethyl (iKE) protecting group on its 3' -phosphory l oxygen atom, and a protect ing group on the base.
5'
5'
Phosphotriester intermediate
•
resin
This solid-phase approach is ideal for the synthesis of DNA, as it is for polypeptides, because the desired product stays on the insoluble support until the final release step. All the reactions take place in a single vessel, and excess soluble reagents can be added to drive reactions to completion . At the end of each step, soluble reagents and by-products are washed away from the resin that bears the growing chains. At the end of the synthesis, NH3 is added to remove all protecting groups and release the oligonucleotide from the solid support. Because elongation is never 100% complete, the new DNA chains are of diverse lengths the desired chain is the longest one. The sample can be purified by high -pressure liquid chromatography or by electrophoresis on polyacrylamide gels. DNA chains of as many as 100 nucleotides can be readily synthesized by this automated method. The ability to rapidly synthesize DNA chains of any selected sequence opens many experim ental avenues. For example, a synthesized oligonucleotide labeled at one end with 32p or a fl uorescent tag can be used to search for a complementary sequence in a very long DNA molecule or even in a genome consisting of many chromosomes . The use of labeled oligonucleotides as DNA probes is powerful and general. For example, a DNA probe that can base-pair to a known complementary sequence in a chromosome can serve as the starting point of an exploration of adjacent uncharted DNA. Such a probe can be used as a primer to initiate the replication of neighboring DNA by DNA polymerase. An exciting application of the solid-phase approach is the synthesis of new tailor-made genes. New proteins with novel properties can now be produced in abundance by the expression of synthetic genes .
140 CHAPTER 5 Exploring Genes and Genomes
Selected DNA Seq ue nces Can Be Greatly Amplified by the Polymerase Chain React io n
Flanking sequence
'-
Target sequence
,
......
(
,
I
I
I
I
I
{,"\
\...!.J
I
I
I
I
(3) I
I
Add excess primers Heat to separate strands
Cool to anneal primers
1. Strand separation. The two strands of the parent DNA molecule are separated by heating the solution to 95°C for 15 s .
I
..==--~- Primers --~. . I
I
[ (})
I
Figure 5.7 The first cycle in the polymerase chain reaction (PCR). A cycle con,ists of three steps: strand separation, the hybridization o f primers, and the extension of primers by DNA synthesis.
2. Hybridization of primers. T he solution is then abruptly cooled to 54°C to allow each primer to hybridize to a DNA strand . O ne primer hybridizes to the 3' end of the target on one strand, and the other primer hybridizes to the 3' end on the complementary target strand. Parent DNA duplexes do not form, because the primers are present in large excess. Primers are typically from 20 to 30 nucleotides long.
DNA synthesis. The solution is then heated to n oc, the optimal temperature for Taq DNA polymerase. This heat-stable polymerase comes from Thermus !M1uaticus, a thermophilic bacterium that lives in hot springs. 3.
The polymerase elongates both primers in the direction of the target sequence because DNA synthesis is in the 5'-to-3' direction. DNA synthesis takes place on both strands but extends beyond the target sequence.
These three steps strand separation, hybridization of primers, and DNA synthesis constitute one cycle of the PCR amplification and can be carried out repetitively just by changing the temperature of the reaction mixture. The thermostability of the polymerase makes it feasible to carry out peR in a closed container; no reagents are added after the first cycle. The duplexes are heated to begin the second cycle, which produces four duplexes, and then the third cycle is initiated (Figure 5.8). At the end of the third cycle, two short strands appear that constitute only the target sequence the sequence including and bounded by the primers . Subsequent cycles will amplify the target sequence exponentially. The larger strands increase in number arithmetically and serve as a source for the synthesis of more short strands. Ideally, after n cycles, the desired sequence is amplified 2n-fold . The amplification is a millionfold after 20 cycles and a billionfold after 30 cycles, which can be carried out in less than an hour. Several features of this remarkable method for amplifying DNA are noteworthy. First, the sequence of the target need not be known. All that is required is knowledge of the flanking sequences. Second, the target can be much larger than the primers. Targets larger than 10 kb have been amplified by peR. Third, primers do not have to be perfectly matched to flanking sequences to amplify targets. With the use of primers derived from a gene of known sequence, it is possible to search for variations on the theme. In this way, families of genes are being discovered by peR. Fourth, peR is highly specific because of the stringency of hybridization at relatively high temperature. Stringency is the required closeness of the match between primer and target, which can be controlled b y temperature and salt. At high temperatures, the only DNA that is amplified is that situated between primers that have hybridized . A gene constituting less than a millionth of the total DNA of a higher organism is accessible by peR. Fifth, peR is exquisitely sensitive. A single DNA molecule can be amplified and detected .
peR Is a Powerful Technique in Medical Diagnostics, Forensics, and Studies of Molecular Evo lution PCR can provide valuable diagnostic information in medicine. Bacteria and viruses can be readily detected with the lise of specific primers. For example, pe R can reveal the presence of human immunodeficiency virus in people who have not mounted an immune response to this pathogen and would therefore be missed with an antibody assay. Finding Mycobacterium tuberculosis bacilli in tissue specimens is slow
FIRST CYCLE BEGINS Flanking sequence ~
, Target s;quence \
I I
I I
I I
I I
Add excess primers Heat to separate Cool
I I IIII!I:::..~_ Primers I
I
I
,
I
-I I
Add heat-stable DNA polymerase Synthesize new DNA
I I
I
I
Heat to separate
Cool Excess primers still present
I
O2
>
° 2
'-
Figure 7.11 Concerted model. All molecules exist either in the T state or in the R stat e. At each level o f oxygen loading, an equilibrium exists between the T and R states. The equilibrium shifts from strongly favoring the T state w ith no oxygen bound to strongly favoring the R state when the molecule is ful ly loaded with oxygen. The R state has a greater affinity for oxygen than does the T state.
R-state binding curve ;;:. • -• •
- .:..0 - -~ - - ;.... - ~--
1.0 ~
c
.-o
•• • • •
0 .8
~
~
••
:::I
-;0 0.6
-'"
'".-co
Observed hemoglobinbinding curve
••
• •
0.4
tJ
• •
,•
~
• ~ 0.2 • >-
0.0
o
.•
.'
• • • '
.'
• •
••
••• •
.-----
•••••• T-state binding curve 50
100
150
200
pO, (torr) Figure 7.12 T-to·R transition. The observed binding curve for hemoglobin can be seen as a com bination o f t he binding curves t hat woul d be o bserved if all molecules remained in the T stat e or if all o f t he mo l ecules were in the R state. The sigmoidal curve is o bserved because molecu les convert from the T stat e int o the R state as o xygen molecules bind.
oxygen affinity of its sites increases. Additional oxygen molecules are now more likely to bind to the three unoccupied sites. Thus, the b ' nding curve is shallow at low oxygen concentrations when all of the molecules are in the T state, becomes steeper as the fraction of molecules in the R state increases, and flattens out again when all of the sites within the R -state molecules become filled (Figure 7.12). These events produce the sigmoid binding curve so important for efficient oxygen transport . In the concerted model, each tetramer can exist in only two states, the T state and the R state. In an alternative model, the sequential model , the binding of a ligand to one site in an assembly increases the binding affinity of neighboring sites without inducing a full conversion from the T into the R state (Figure 7. 13).
,
K, ~
O2
'\
,
K, >
0,
O2
Figure 7.13 Sequential model. The binding o f a ligand changes the confo rmati o n o f the subunit to w hich it b inds. Thi s confo rmati onal change induces changes in neighboring subunit s that increase their affin it y for t he ligand.
Is the cooperative binding of oxygen by hemoglobin best described by the concerted or the sequential model? Neither model in its pure form full y accounts for the behavior of hemoglobin . Instead, a combined model is required . H emoglobin behavior is concerted in that hemoglobin with three sites occupied by oxygen is almost always in the quaternary stru cture associated with the R state. The remaining open binding site has an affinity for oxygen more than 20-fo ld greater than that of fully deoxygenated hemoglobin binding its first oxygen. However, the behavior is not fully concerted, because hemoglobin with oxygen bound to only one of four sites remains primarily in the T-state quaternary structure. Yet, this molecule binds oxygen three times as strongly as does fully deoxygenated hemoglobin, an observation consistent only with a sequential model. These results highlight the fact that the concerted and sequential models represent idealized limiting cases, which real systems may approach but rarely attain . Structural Changes at the Heme Groups Are Transmitted to the cxlf31-cx2f32 Interface
0. , ~,- 0.2~2 interface
\ Deoxyhemoglobin
Oxyhemoglobin Figure 7.14 Conformational changes in hemoglobin. The movement of th e iro n ion o n oxygenat ion bri ngs t he iro nassociated histi d ine res idue toward the porphyrin ring. The associat ed movement of the h istid ine-con t ain ing a helix alters t he interface bet ween the a J3 dimers. inst igating o th er structural changes. For compari son, the deoxyhemoglo bin st ructure is shown in gray beh ind the oxyhemoglobin stru cture in co lo r.
We now examine how oxygen binding at one site is able to shift the equilibrium between the T and R states of the entire hemoglobin tetramer. As in myoglobin, oxygen binding causes each iron atom in hemoglobin to move from outside the plane of the porphyrin into the plane. When the iron atom moves, the histidine residue bound in the fifth coordination site moves with it. This histidine residue is part of an ex helix, which also moves (Figure 7.14). The carboxyl terminal end of this ex helix lies in the interface between the two ex[3 dimers. The change in position of the carboxyl terminal end of the helix favors the T -to -R transition. Consequently, the structural transition at the iron ion in one subunit is directly transmitted to the other subunits. T he rearrangement of the dimer interface provides a pathway for communication between subunits, enabling the cooperative binding of oxygen. 2,3-Bisphosphoglycerate in Red Cells Is Crucial in Determining the Oxygen Affinity of Hemoglobin
For hemoglobin to function effi ciently, a requirement is that the T state remain stable until the binding of sufficient oxygen has converted it into the
190 •
19 1
Rstate. The T state of hemoglobin is highly unstable, however, pushing the equilibrium so far toward the R state that little oxygen would be released in physiological conditions. Thus, an additional mechanism is needed to properly stabilize the T state. This mechanism was discovered by comparing the oxygen-binding properties of hemoglobin in red blood cells with fully purified hemoglobin (Figure 7.15). Pure hemoglobin binds oxygen much more tightly than does hemoglobin in red blood cells. This dramatic difference is due to the presence within these cells of 2,3-bisphosphoglycerate (2,3 -BPG; also known as 2,3 -diphosphoglycerate or 2,3-DPG) .
7.2 Cooperative Binding of Oxygen
Pure hemoglobin lungs (no 2,3-BPG)
Tissues 1.0
'2 o .-
-
Hemoglobin (in red celis, with 2,3-BPG)
0.8
~
66%
::J
- .0 ./ 0 ,-, H .-.--
~ 0.6
"'oc .-
0.4
---
U
~
~.
0 .2
>0.0
2,3-Bisphosphoglycerate (2.3-BPG)
20
50
100
150
200
p02 (torr)
This highly anionic compound is present in red blood cells at approximately the same concentration as that of hemoglobin (-2 mM). Without 2,3 -BPG, hemoglobin would be an extremely inefficient oxygen transporter, releasing only 8% of its cargo in the tissues. How does 2,3 -BPG lower the oxygen affinity of hemoglobin so significantly? Examination of the crystaJ structure of deoxyhemoglobin in the presence of2,3-BPG reveals that a si ngle molecule of 2,3 -BPG binds in the center of the telramer, in a pocket present only in the T form (Figure 7.16). On T-to- R transition, this pocket collapses and 2,3-BPG is released . Thus, in order for the structuraJ transition from T to R to take place, the bonds between hemoglobin and 2,3 -BPG must be broken. In the presence of 2,3 -BPG, more oxygen -binding sites within the hemoglobin tetramer must be occupied in order to induce the T-to-R transition, and so hemoglobi n remains in the lower-affinity T state until higher oxygen concentrations are reached. The regulation of hemoglobin by 2,3-BPG is remarkable because 2,3 -BPG does not in any way resemble oxygen, the molecule on which hemoglobin
~1 subunit
Figure 7.15 Oxygen binding by pure hemog lobin compared with hemoglobin in red blood cells. Pure hemoglobin binds oxygen more tightly than does hemoglobin in red bloo d cells. This difference is due to the presence o f 2,3-bisphosphoglycerate (2,3-BPG) in red blood cells.
N
o His 2
Lys 82
o His 143
~2 His 2
"ll
o
~ 2 subunit
Figure 7.16 Mode of binding of 2,3-BPG to human deoxyhemoglobin. 2,3-Bisphosphoglycerate binds to the central cavity o f deoxyhemoglobin (left). There. it interacts with three positively charged groups on each J3 chain (right). [Drawn from 1B86.pdb.]
1.0 ~
0 .-"
--'"'"
Materna l red cells
0.8
~
:J
carries out its primary function. 2,3-BPG is referred to as an allosteric effeC!01 (from alios, "other," and stereos, "structure" ). Regulation by a molecule structurally unrelated to oxygen is possible because the allosteric effector binds to a site that is completely distinct from that for oxygen. We will encounter allosteric effects again when we consider enzyme regulation in C hapter 10.
Fetal red cells
0.6
'""0
.- 0.4 U
02 flows from matern al oxyhemoglobin to fetal deoxyhemoglobin
-'"
::::.. 0.2
"0.0
0
50
100
p02 (torr) Figure 7.17 Oxygen affinity of fetal red blood cells. Fetal red blood cells have a higher oxygen affinity than maternal red blood cells because fetal hemoglobin does no t bind 2,3-BPG as well as materna l hemoglobin does.
The binding of 2,3 -BPG to hemoglobin has other crucial physiological consequences. The globin gene expressed by human fetuses differs from that expressed by adults;fetal hemoglobin tetramers include two a chains and two 'Y chains. The 'Y chain, a result of another gene duplication, is 72% identical in amino acid sequence with the 13 chain. One noteworthy change is the substitution of a serine residue for His 143 in the 13 chain, part of the 2,:1BPG-binding site. This change removes two positive charges from the 2,3BPG -binding site (one from each chain) and reduces the affinity of 2,3-BPG for fetal hemoglobin . Consequently, the oxygen-binding affinity of fetal hemoglobin is higher than that of maternal (adult) hemoglobin (Figure 7.1 7). This difference in oxygen affinity allows oxygen to be effectively transferred from maternal to fetal red blood cells. We have here an example in which gene duplication and specialization produced a ready solution to a biological challenge in this case, the transport of oxygen from mother to fetus .
7.3
Tissues
r-
1.0 ~
66%
"-
0 0.8 .-
--'" -'"'" :J
'""0
·e '" ~
-
Lungs
0.6
77%
0.4
----
0 .2
"-
0.0
o
100
20
p02 (torr) Figure 7.18 Effect of pH on the oxygen affinity of hemoglobin. Lowering the pH from 7.4 (red curve) t o 7.2 (blue curve) results in the relea se of O 2 f rom o xyhemoglo bin.
192
Hydrogen Ions and Carbon Dioxide Promote the Release of Oxygen: The Bohr Effect
W e have seen how hemoglobin's cooperative release of oxygen helps deliver oxygen to tissues where it is most needed, as revealed by their low oxygen partial pressure. This ability is enhanced by the ability of hemoglobin to respond to other cues in its physiological environment signaling the need for oxygen . Rapidly metabolizing tissues, such as contracting muscle, generate large amounts of hydrogen ions and carbon dioxide (pp. 447 and 448). So that oxygen is released where the need is greatest, hemoglobin has evolved to release oxygen more readily in response to higher levels of these substances. Like 2,3-BPG, hydrogen ions and carbon dioxide are allosteric effectors of hemoglobin that bind to sites on the molecule that are distinci from the oxygen-binding sites. The regulation of oxygen binding by hydrogen ions and carbon dioxide is called the Bohr effect after Christian Bohr, who described this phenomenon in 1904. The oxygen affinity of hemoglobin decreases as pH decreases from a value of?.4 (Figure 7 .18). Consequently, as hemoglobin moves into a region oflower pH, its tendency to release oxygen increases. For example, transport from the lungs, with pH 7.4 and an oxygen partial pressure of 100 torr, to active muscle, with a pH of 7.2 and an oxygen partial pressure of 20 torr, results in a release of oxygen amounting to 77% of total carrying capacity. Only 66% of the oxygen would be released in the absence of any change in pH Structural and chemical studies have revealed much about the chemical basis of the pH effect. At least two sets of chemical groups are responsible for the effect: the a-amino groups at the amino termini of the a chain and the side chains of histidines 13146 and a1 22, all of which have pKa values near pH 7. Consider histidine 13146, the residue at the C terminus of the 13 chain. In deoxyhemoglobin, the terminal carboxylate group of 13146 forms a salt bridge with a lysine residue in the a subunit of the other al3 dimer. This interaction locks the side chain of histidine 13146 in a position from which it can participate in a salt bridge with negatively charged aspartate 94 in the same chain, provided that the imidazole group of the histidine residue is protonated (Figure 7.19). The other groups also participate in salt bridges in the T state. The for-
mation of these salt bridges stabilizes the T state, leading to a greater tendency
193 7.3 The Bohr Effect !X2
Lys 40
C terminus Added proton
+
131 Asp 94
PI His 146
Figure 7.l9 Chemical basis of the Bohr effect. In deoxyhemoglobin, three amino acid residues form two salt bridges that stabilize the T quaternary structure. The formatio n of one of the salt bridges depends o n t he presence o f an added proton on histid ine 13146. The proximity of th e negative charge on aspartate 1394 in deoxyhemoglobin favors protonation of thi s histidine. Notice th at the salt bridge bet ween histidine 13146 and aspartat e 1394 is stabilized by a hydrogen bond (green dashed line).
for oxygen to be released. For example, at high pH, the side chain of histidine ~146 is not protonated and the salt bridge does not form. As the pH drops,
however, the side ch ain of histidine [3146 becomes protonated, the salt bridge with as partate [394 forms, and the T state is stabilized. Carbon dioxide, a neutral species, passes through the red-blood- cell membrane into the cell. This transport is also facilitated by membrane transporters including proteins associated with Rh blood types. Carbon dioxide stimulates oxygen release by two mechanisms. First, the presence of high con centrations of carbon dioxideleads to a drop in pH within the red blood cell (Figure 7.20).
o o
o o ,
CO 2
=
0 1\
y
Body tissue
y
,
Blood capillary
Figure 7.20 Carbon dioxide and pH. Carbon d iox ide in the ti ssues diffuses into red blood cells. Inside a red blood cell, carbon diox ide react s w ith water to form carbonic acid, in a reaction catalyzed by the enzym e carbonic anhydrase. Carbo nic acid dissociates to form HCO, - and H+, resulting in a drop in pH inside the red cell.
•
pH 7.4, no CO 2
-
pH 7 .2,
-
pH
no CO 2 7.2, 40 torr CO 2
Tissues
Lungs
1.0 ~
c:
-
Carbon dioxide reacts with water to form carbonic acid, H 2 C0 3 . This reaction is accelerated by carbonic anhydrase, an enzyme abundant in red blood cells that will be considered extensively in C hapter 9. Carbonic acid is a strong acid with a pKa of3.S. Thus, once formed, carbonic acid dissociates to form bicarbonate ion, HCO:l - , and H+, resulting in a drop in pH. This drop in pH stabilizes the T state by the mechanism discussed previou sly. In the second m echanism, a direct chemical interaction between carbon dioxide and hemoglobin stimulates oxygen release. The effect of carbon dioxide on oxygen affinity can be seen by comparing oxygen-bindin g curves in the absence and presence of carbon dioxide at a constant pH (Figure 7.21). In the presence of carbon dioxide at a partial pressure of 40 torr at pH 7.2 , the am ount of oxygen released approaches 90% of t he maxi • mum carrylllg capacity.
-
.-o -~'" -'" '"oc: .-ti :::I
0.8 0.6
88%
0.4
(!!
to:"... 0.2
----- --
"0.0
o
20
100
p02 (torr) Figure 7.21 Carbon dioxide effects. The presence of carbon dioxide decreases the affinity of hemoglobin for oxygen even beyond the effect due to a decrease in pH, resulting in even more efficient oxygen transport from the tissues to the lungs.
194 CHAPTER 7 Hemoglobin: Portrait of a Protein in Action
Carbon dioxide stabili zes deoxyhemoglobin by reacting with the terminal amino groups to form carbamate groups, which are negatively charged, in contrast with the neutral or positive charges on the free amino groups. R
\
,N- H H
° + °
R
II
C
•
-
0.0
o
50
100
150
200
pO, (torr) Figure 7.29 Oxygen-binding curves for several Hill coefficients. The curve labeled n = 2.8 closely resembles the curve for hemoglobin.
The Concerted Model
The concerted model can be formulated in quantitative terms. Only four parameters are required: (1) the number of binding sites (assumed to be equivalent) in the protein, (2) the ratio of the concentrations of the T and R states in the absence of bound ligands, (3 ) the affinity
This is the measure of how much more tightly a subunit for a protein in the R state binds a ligand compared with a subunit for a protein in the T state. Note that c < 1 beca use KR and KT are dissociation constants and tight binding corresponds to a small dissociation constant. What is the ratio of the concentration of T-state proteins with one ligand bound to the concentration of R-state proteins with one ligand bound? The dissociation constant for a single site in the R state is K R. For a protein with n sites, there are n possible sites for the first ligand to bind. This statistical factor favors ligand binding compared with a single-site protein. Thus, [Rd = n [Ro][S]/KR ' Similarly, [T d = 11 [T 0] [S]/KT ' Thus,
Appendix
Similar analysis reveals that, for states with i ligands bound, [Ti]/[R ] = e'L. In other words, the ratio of the concentrations of the T state to the R state is reduced by a factor of e for each ligand that binds. Let us define a convenient scale for the concentration of S:
1.0
20 1
r-
0.8
0.6
a = [S]!KR
This defillition is useful because it is the ratio of the concentration of S to the dissociation constant that de termines the extent of binding. Using this definition, we see that
[R,l
=
n[Rol[Sl KR
=
n[Rola
0.4
0 .2
0.0
a
50
150
200
p02 (torr)
Similal'iy,
[T, l
ll[Tol[Sl K[
=
=
llcL [Rola
What is the concentration ofR-state molecules with two li gands bound? Again, we must consider the statistical factor that is, the number of ways in which a second ligand can bind to a molecule with one site occupied. The number of ways is n - 1. H owever, because which ligand is the "first" and which is the "second" does not matter, we must divide by a factor of 2. Thus, / 11
[Rzl
=
[R ll[Sl
KR n- 1 [RIJa 2 n- 1 (n [ RoJa)"
=
2
n
n -
1
2
[Rol" z
We can derive similar equations for the case with i ligands bound and for T states. We can now calculate the fractional saturation, Y. This is the total concentration of sites with li gands bound divided by the total concentration of potential binding sites . Thus,
([ Rll + [T I]) + 2([R2l + [T z]) + ...
+ n([ Knl + [Tn ]) Y =--~~~----~~~~------~~~~~ ,,([Rol + [Tol + [ K.] + [T il + .. . + [R" l + [T"J) Substituting into this equation, we find
,,[Rola + llc[T "la + 2( n(" - 1)/2)[RoJa 2 2
Figure 7.30 Modeling oxygen binding with the concerted model. The fractional saturati on (Y) as a functi o n p02: L = 9000, C = 0.014. and KR = 2.5 torr. The fraction of molecules in the T state with zero, one, and two o xygen molecules bound (To, TI , and T, ) and the fraction of molecules in the R state with two, three, and four oxygen mo lecules bound (R" R3, and R.,) are shown. The fractions of molecules in other forms are t oo low to be shown .
Substituting (Tal - L[Rol and summi ng these series yields
1'
-
2
-
y=
100
Z
+ 2(" (,, - 1)/2)c [T ola + ... + n[RoJa" + llc"[T oJla " n([ Rol + lTol + n[Rola + nc[T ola + ... + [Rola " + c" [T ol,,")
y =
a (1 +
+ Lca(l + cu ),, - ' --'---.,---'---:-----,---'----'-(1 + a)" + L(l + cu)" U )" - I
We can now use this equation to fit the observed data for hemoglobin by varying the parameters L, e, and Ki{ (with n =4 ). An excell ent fit is obtained with L = 9000 , e = 0.014, and KR = 2 .5 torr (Figure 7.30). In addition to the fractional saturation, the concentra tions of the species T o, T j, T Z, R 2 , R 3 , and ~ are shown . The concentrations of all other species are very low. The addition of concentrations is a major difference between the analysis using the Hill equation and this analysis of the concerted model. The Hill equation gives only the fractional saturation, whereas the analysis of the concerted model yields concentrations for all species. In the present case, this analysis yields the expected ratio ofTstate proteins to R -state proteins at each stage of bind ing. This ratio changes from 9000 to 126 to 1.76 to 0.02 5 to 0.00035 with zero, one, two, three, and four oxygen m olecules bound . This ratio provides a quanti tative measure of the switching of the population of he moglobin molecules from the T state to the R state. The sequential model can also be formulated in quantitative terms. However, the formulation entails many more parameters, and many different sets of parameters often yield similar fits to the experimental data.
202
CHAPTER 7 Hemo globin: Portrait of a Protein in Action
Key Terms heme (p . 184 )
partial pressure (p . 187)
sic kle -cell anemia (p . 194 )
protoporphy rin ( p. 184 )
sigmoid (p . 187)
h e mog lobin S (p . 195)
proximal histidine (p . 185 )
cooperative (p . 187)
malaria (p.196)
func tio nal mag ne tic resonance imaging (fMR l) (p. 185)
T state (p . 188) R state (p . 188)
thalassemia (p . 196)
superoxid e anion (p. 185 ) m etmyoglobin (p . 185)
concerted mode l (MWC model) (p. 189)
thalassemia major (Cooley ane mia) (p. 196)
d istal histidine (p . 186)
sequential model (p . 190)
a chain (p . 186)
2,3 -bisphosphoglycerate (p . 191)
a - h em oglobin stabilizing prote in (AHSP) (p . 197)
j3 chain (p . 186)
fetal h e m oglo bin (p . 192 )
n e uroglo bin (p. 197)
g lobin fold (p . 186)
Bo hr effect (p . 192)
cytuglobin (p . 197 )
aj3 dime r (p . 187)
carbonic anhydrase (p. 193 )
Hill plot (p . 200)
oxygen-binding c ur ve (p . 187)
carbamate (p . 19 4 )
Hill coefficient (p . 200)
he moglobin H (p . 196)
fraction a l saturation (p . 187)
•
Selected Readings Where to Start Perutz, M . F. 1978. Hemoglobin structure and respiratory transport.
Sickle-Cell Anemia and Thalasssemia Herrick, J . ]). 1910. Peculiar elongated and sickle-shaped red blood cor·
Sci. Am. 239(6):92- 125. Perutz, M . F. 1980. ~tereochemical mechanism of oxygen transport by haemoglobin. Proc. R Soc. Lond. Biul. S the turnover number, . as described above. However, most enzymes are not normally saturated with substrate. Under physiological conditions, the [S]/KM ratio is typically between 0.01 and 1,0. When [S1< < K ~j, the enzymatic rate is much less than kcat because most of the active sites are unoccupied. Is there a number that characterizes the kinetics of an enzyme under these more typical cellular conditions? Indeed there is, as can be shown by combining equations 14 and 20 to give
Va
=
cat k [E][S]
KM
(34)
TABLE 8.5 Turnover numbers of some enzymes
Enzyme Carbonic anhydrase 3-Ketosteroid
isomerase Acety [chol inesterase Penicillinase Lactate
dehydrogenase Chymotrypsin DNA polymerase I Tryptophan synthetase Lysozyme
Turnover number (per second) 600,000 280.000 25.000 2.000 1.000 100 15 2 0.5
222
TABLE 8.6 Substrate preferences of chymotrypsin
CHAPTER 8 Enzymes: Basic Concepts and Kinetics
Amino acid in ester
Amino acid side chain
Glycine
-H
1.3 X 10- 1
/Hl Valine
-CH
\
2.0
CH ,
Norvaline
-CH,CH,CH ,
3.6 X 10'
Norleucine
-
3.0 X 10'
Phenylalanine
-CH,-
CH,CH,CH,CH,
1.0 X 10'
Source: After A. Fersht, Structure and Mechanism in Protein Science: A Guide to Enzyme Catalysis and Protein Folding (W. H. Freeman and Company, 1999), Table 7.3.
When [S]« K M , the concentration offree enzyme, [E], is nearly equal to the total concentration of enzyme [ElT; so
(35) Thus, when [S] « K M , the enzymatic velocity depends on the values of keat/KM , [S], and [Ely. Under these conditions, keat/KM is the rate constant for the interaction ofS and E. The rate constant keat/KM is a measure of catalytic efficiency because it takes into account both the rate of catalysis with a particular substrate (keat ) and the strength of the enzyme-substrate interaction (KM ). For instance, by using keat/KM values, one can compare an enzyme's preference for different substrates. Table 8.6 shows the keat/KM values for several different substrates of chymotrypsin. Chymotrypsin clearly has a preference for cleaving next to bulky, hydrophobic side chains. How efficient can an enzyme be? We can approach this question by determining whether there are any physical limits on the value of keat/K",. Note that this ratio depends on kl' k_I' and keat' as can be shown by substi· tuting for K M .
(36) TABLE 8.7 Enzymes for which kcat/KM is close to the diffusioncontrolled rate of encounter Enzyme Acetylcholi nesterase Carbonic anhydrase Catalase
1.6 X 10' 8.3 X 107
Crotonase
2.8 X 10 8
Fumarase
1.6 X 10 8
Triose phosphate isomerase
2.4 X 10' 1 X 108 9 7 X 10
i3-Lactamase Superoxide dismutase
4 X 10 7
Structure and Mechanism in Protein Science: A Guide to Enzyme Catalysis and Protein Folding (W. H. Freeman and Company, 1999), Table 4.5. Source: After A, Fersht,
Suppose that the rate of formation of product (keat ) is much faster than the rate of dissociation of the ES complex (k- j ). The value of keat/KM then approaches kl . Thus, the ultimate limit on the value of keat/KM is set by kj , the rate of formation of the ES complex. This rate cannot be faster than the diffusion-controlled encounter of an enzyme and its substrate. Diffusion limits 8 9 1 the value of kl and so it cannot be higher than between 10 and 10 s -I M9 8 1 Hence, the upperlimit on keat/KM is between 10 and 10 s -I M- . The keat/KM ratios of the enzymes superoxide dismutase, acetyl8 cholinesterase, and triosephosphate isomerase are between 10 and 9 I 10 S-I M- . Enzymes that have keat/KM ratios at the upper limits have attained kinetic perfection. Their catalytic velocity is restricted only by the rate at which they encounter substrate in the solution (Table 8.7). Any further gain in catalytic rate can come only by decreasing the time for diffusion. Remember that the active site is only a small part of the total
~1l.1'lme
structure. Yet, for catalytically perfect enzymes, every encounter decween enzyme and substrate is productive. In these cases, there may be attractive electrostatic forces on the enzyme that entice the substrate to the active site. These forces are sometimes referred to poetically as Circe
effects. Diffusion in solution can also be partly overcome by confining substrates and products in the limited volume of a multienzyme complex. Indeed, some series of enzymes are associated into organized assemblies so that the product of one enzyme is very rapidly found by the next enzyme. In effect, products are channeled from one enzyme to the next, much as in an assembly line.
Circe effect The uti lization of attractive forces to lure a
substrate into a site in which it undergoes a transformation of structure. as defined by William P. Jencks. an enzymologist. who
coined the term. A goddess of Greek mythology. Circe lured Odysseus's men to her house and then transformed them into pigs,
Most Biochemical Reactions Include Multiple Substrates Most reactions in biological systems start with two substrates and yield two products. They can be represented by the bisubstrate reaction:
A+B,
'P+Q
Many such reactions transfer a functional group, such as a phosphoryl or an ammonium group, from one substrate to the other. Those that are oxidation-reduction reactions transfer electrons between substrates. Multiple substrate reactions can be divided into two classes: sequential reactions and double-displacement reactions.
Sequential Reactions. In sequential reactions, all substrates must bind to the enzyme before any product is released. Consequently, in a bisubstrate reaction, a ternary complex of the enzyme and both substrates forms. Sequential mechanisms are of two types: ordered, in which the substrates bind the enzyme in a defined sequence, and random. Many enzymes that have NAD + or NADH as a substrate exhibit the ordered sequential mechanism. Consider lactate dehydrogenase, an important enzyme in glucose metabolism (p. 447). This enzyme reduces pyruvate to lactate while oxidizing NADH to NAD + . .
+ NADH + H+ "",==:' HO
II + NAD+
C
CH 3 Lactate
In the ordered sequential mechanism, the coenzyme always binds first and the lactate is always released first, This sequence can be represented by using a notation developed by W. Wallace Cleland: NADH
Enzyme
Pyruvate
Lactate
NAD+
----iJ__--'JL-___________-.J1L-_---LJ_ Enzyme =='
E (NADH) (pyruvate) :;:,
E (lactate) (NAD+)
The enzyme exists as a ternary complex consisting of, first, the enzyme and substrates and, after catalysis, the enzyme and products, In the random sequential mechanism, the order of the addition of substrates and the release of products is random, An example of a random sequential reaction is the formation of phosphocreatine and AD P from ATP and creatine, which is catalyzed by creatine kinase (p. 416). , t
223
224 CHAPTER 8 Enzymes: Basic Concepts and Kinetics
+ ADP
Creatine
Either creatine or ATP may bind first, and either phosphocreatine or ADP may be released first. Phosphocreatine is an important energy source in muscle. Sequential random reactions also can be depicted in the Cleland notation. ATP
Creatine
Phosphocreatine
Enzyme--
111
'"
40
[11=5K; 20
o [Substrate]
226
,
pathways in the inflammatory response . Statins are drugs that reduce high cholesterol levels by competitively inhibiting a key enzyme in cholesterol biosynthesis (p. 339). In uncompetitive inhibition, the inhibitor binds only to the ES, com plex. This enzyme- sub strate- inhibitor complex, ESI, does not go on to form any product. Because some unprodu ctive ESI complex will always be present, Vmax will be lower in the presence of inhibitor than in its absence (Figure 8.18) . The uncompetitive inhibitor lowers that apparent value of K M . This occurs since the inhibitor binds to ES to form ESI , depleting ES. To maintain th e equilibrium between E and ES, more S binds to E. Thus, a lower concentration of S is required to form half of the max imal concentration of ES and the apparent value of KM is reduced . The herbicide glycophosphate, also known as Roundup, is an uncompetitive inhibitor of an enzyme in the biosynthetic pathway for aromatic amino acids. In noncompetitive inhibition (Figure 8.19), substrate can still bind to the enzyme- inhibitor complex. However, the enzyme- inhibitor- substrate complex does not proceed to form product. The value of Vmax is decreased to a new value called V~fx> whereas the value of KM is unchanged. The maximal velocity in the presence of a pure noncompetitive inhibitor, V~~fx ' is given by
227 8.5 Enzyme Inhibition
E+I
S \.
) ES + 1--7) E+P
II
Ki ESI 100
)()
No inhibitor
80 OJ
~
'"OJ ~
60
[11 ; Ki
.->
~
-'" OJ
'"
40 I I
20
[11 ; lOKi
[1) ; 5Ki
o [Substrate]
,
KM for uninhibited enzyme
K~fP for (I] ; Kj Vapp _ _ _ V-,;m ",a;,,-x_ max
1 + [IJ;Ki
(37)
Why is Vmax lowered though KM remains unchanged? In essence, the inhibitor simply lowers the concentration of functional enzyme. The resulting solution behaves like a more dilute solution of enzyme. Noncompetitive
inhibition cannot be overcome by increasing the substrate concentration. Deoxycyc1ine, an antibiotic, functions at low concentrations as a noncom petitive inhibitor of a proteolytic enzyme (collagenase). It is used to treat periodontal disease. Some of the toxic effects of lead poisoning may be due to lead's ability to act as noncompetitive inhibitor of a host of enzymes. Lead reacts with crucial sulfhydryl groups in these enzymes. Double-reciprocal plots are especially useful for distinguishing between competitive, uncompetitive, and noncompetitive inhibitors. In competitive inhibition, the intercept on the y -axis of the plot of 11 Va versus I /[S] is the
S E + I ---'\~.,.) ES --7) E + P K·I
S
EI 100
_"--c
l
ESI
)( >
No inhibitor
80
•
• •" 60
.• ->
[I] ; Ki
•
-•• 40 '" 20
[I] ; 10K;
[I] ; 5K;
o [Substrate]
,
Figure 8.19 Kinetics of a noncompetitive inhibitor. The reacti on pathway shows th at the inhibit or binds both to free enzyme and t o enzyme compl ex . Consequently. as with uncompetitive competition. Vma. cannot be attained, KM remain s unchanged. and so the reaction rate increases more slow ly at low substrate concentraticns t han is the case for uncompetiti ve com petition.
Figure 8.18 Kinetics of an uncompetitive inhibito,', The rea ction pathway sho ws that the inhibitor binds only to the enzyme-substrate complex. Consequently. Vm " cannot be atta ined. even at high subst rate concentrations. The apparent value for KM is lowered. becoming smaller as more inhibitor is added.
+ Competitive inhibitor
l /V "'- No inhibitor present
same in the presence and in the absence of inhibitor, although the slope is increased (Figure 8.20). The intercept is unchanged because a competitive inhibitor does not alter Vmax' T he increase in the slope of the 11 Va versus 1/ [S] plot indicates the strength of binding of cOlnpetitive inhibitor. In the presence of a competitive inhibitor, equation 28 is replaced by 1
1
Vo o
1/ [51
Figure B.20 Competitive inhibition illustrated on a double-reciprocal plot. A double-reciprocal plo t of enzyme ki netics in t he presence and absence of a competit ive in hibit or illustrates that t he inhi bitor has no effect on Vma> but increases KM o
+ Unncompetitive inhibitor
,~ , --/ No inhibitor present
l /V
o
1/ [SI
Figure B.21 Uncompetitive inhibition illustrated by a double-reciprocal plot. An uncompetit ive inhibitor does not effect t he slo pe of the dou ble-reciprocal plot. Vma> and KM are reduced by equivalent amounts,
I Ij 1 + -'--'KI
1
(38)
[S]
In other words, the slope of the plot is increased by the factor (1 + [IJI K j ) in the presence of a competitive inhibitor. Consider an enzyme with a KM of 10- 4 4 M . In the absence of inhibitor, Vo = Vmaxi 2 when [S] = 10- M . In the pres· ence of 2 X 10 - 3 M competitive inhibitor that is bound to the enzyme with a 3 K; of 10 - M , the apparent KM (K~P) will be equal to K M(1 + [I] I Kj), or 3 X 10 - 4 M. Substitution of these values into equation 37 gives 4 Vo = Vmax / 4, when [S] = 10- M. The presence of the competitive inhibitar thus cuts the reaction rate in half at this substrate concentration. In uncompetitive inhibition ( Figure 8.2 1), the inhibitor combines only with the enzyme- substrate complex . The equation that describes the double- reciprocal plot for an uncompetitive inhibitor is 1
Va
1
1
+- -
Vm ax
+ lIj
(39)
KI
The slope of the line, K MI V", ,,,, is the same as that for the uninhibited enzyme, b ut the intercept on the y-axis will be increased by 1 + [I ]I Kj , C onsequently, the lines in dou ble-reciprocal plots will be parallel. In noncompetitive inhibition (Figure 8.22), the inhibitor can combine with either the enzyme or the enzyme- substrate complex. In pure noncompetitive inhibition , the values of the dissociation constants of the inhibitor and enzyme and of the inhibitor and enzyme- substrate complex are equal. The value of Vmax is d ecreased to the new value V~~fx , and so the intercept on the vertical axis is increased. T he new slope, which is equal to KIV/V ~;,;" is larger by the same factor. In contrast with Vm ax , KM is not affected by pure noncompetitive inhibition .
Irreversible Inhibitors Can Be Used to Map the Active Site In Chapter 9, we will examine the chemical details of how enzymes func· tion. The fi rst step in obtaining the chemical mechanism of an enzyme is to determine what functional groups are required for enzyme activity. How can we ascertain these functional groups? X -ray crystallography of the
+ Noncompetitive inhibitor
l /V Figure B.22 Noncompetitive inhibition illustrated on a double-reCiprocal plot. A double-reciproca l plot of enzyme kinetics in the presence and absence of a no ncompetit ive inhibito r shows that KM is unaltered and V" ,., is decreased.
228
"'- No inhibitor present
o
- 1/ [SI
229 8.5 Enzyme Inhibition
H
o
{-
-
.p
( P -1 P
P
P
P
l
TI T I
G
T
I
C
I
I
T
A
G
C
A
I
A
I
P (,.' py'v P
3'
I
G py' 5 '
P
v
p
A thymine
adenine
one oxygen atom with sulfur (producing a species called a phosphorothi· oate). Let us consider EcoRV endonuclease. This enzyme cleaves the phos· phodiester bond between the T and the A at the center of the recognition se· quence 5' -GATATC -3' . The first step is to synthesize an appropriate substrate for EcoRV containing phosphorothioates at the sites of cleavage (Figure 9.34) . The reaction is then performed in water that has been greatly 18 enriched in 0 to allow the incoming oxygen atom to be marked . The loca· 18 tion of the 0 label with respect to the sulfur atom indicates whether the reo action proceeds with inversion or retention of stereochemistry. The analysis revealed that the stereochemical configuration at the phosphorus atom was in· verted only once with cleavage. This result is consistent with a direct attack by water at phosphorus and rules out the formation of any covalently bound intermediate (Figure 9.35). Figure 9.35 Stereochemistry of cleaved DNA. Cleavage of DNA by EcoRV endonuclease result s in overall inversion of the stereochemica l configuration at the phosphorus atom, as indicated by the stereochemistry o f th e phosphorus atom bound t o one bridging oxygen atom, o ne 160, one 180 , and one sulfur atom. Thi s configuration strongly suggests that the hydrolysis takes place by water's direct attack at the phosphorus atom.
H l Bd
"Ii
0 , ~ 5 O__ P""'"
0 -....., C .... H2 0
thymine
0
\-- -0 -.....,
5
2-
18
o...jp/ \
H2
0
adenine
0
0 IBO
0 -.....,
(
2-
5 \ /
" ,'
\
0 -.....,
(
H2
0
adenine Inverted
(
H2
0
adenine Not inverted
(not observed)
Restriction Enzymes Requ ire Magnesium for Catalytic Activity
Many enzymes that act on phosphate-containing substrates require Mg2+ or some other similar divalent cation for activity. One or more Mg2+ (or similar) cations are essential to the function of restriction endonucleases. What are the functions of these metal ions? It has not been possible to directly visualize the complex between EcoRV 2 endonuclease and cognate DNA molecules in the presence ofMg + by crys· tallization because the enzyme cleaves the substrate under these circum· stances. Nonetheless, it has been possible to visualize metal ion complexes through several approaches. In one approach, crystals of EcoRV endonucle· ase are prepared bound to oligonucleotides that contain the enzyme's recog· nition sequence. These crystals are grown in the absence of magnesium to prevent cleavage; after their preparation, the crystals are soaked in solutions containing the metaL Alternatively, crystals have been grown with the use
26 3
of a mutated form of the enzyme that is less active. Finally, Mg2+ may be re placed by metal ion s such as ci+ that bind but do not result in much catalytic activity. In all cases, no cleavage takes place, and so the locations of the metal ion-binding sites are readily determined . As many as three metal ions h ave been fo und to be present per active site. The roles of these multiple metal ions is still under investigation . One ion-binding site is occupied in essentially all structures. This metal ion is coordinated to the protein through two aspartate residues and to one of the phosphoryl- group oxygen atoms near the site of cleavage. This m etal ion binds the water m olecule that attacks the phosphorus atom, helping to po2 sition and activate it in a manner similar to that for the Zn + ion of carbonic anhydrase (Figure 9 .36).
9.3 Restriction Enzymes
Scissile bond Asp 90 Thymine
,,
Asp 74
Figure 9.36 A magnesium ion-binding site in EcoRV endonuclease. The magnesi um ion helps t o act ivate a wat er molecule and positions it 50 that it can attack the phosphorus atom.
3' Adenine
The Complete Catalytic Apparatus Is Assembled Only Within Complexes of Cognate DNA Molecules, Ensuring Specificity We now return to the question of specificity, the defining feature of restric tion enzymes. T he recognition sequences for most restriction endonucleases are inverted repeats . T his arran gement gives t he three -dimensional structure of the recogn ition site a twofo ld rotational symmetry (Figure 9 .37). The restriction enzymes display a corresponding symmetry: they are dimers whose two subu nits are related by twofold rotational symmetry. The matching symmetry of the recognition sequence and the enzyme facil itates the recognition of cognate DNA by the enzyme. This similarity in structure has been confirmed by the determination of the structure of the complex between EcoRV endonuclease and DNA fragments containing its
(8)
(A)
,
, •
5'
5' 5 '~
GATATC
~3 '
J' ~
CTATAG
~ S'
D
t- - ---..
,Ii' •
d Symmetry axis
3'
!
A
3'
Figure 9.37 Structure of the re cognition site of EcoRV endonuclease. (Al The sequence of the reco gniti on site, w hi ch is symmetric around the axis o f ro tation deSignate d in green. (B) The inverted repeat w ith in the recognitio n sequence of EcoRV (and most o th er restriction endonucl eases) endo ws the DNA site w ith twofo ld ro tatio nal symmetry.
264
recognition sequen ce (Figure 9.38). The enzyme surrounds the D A in a ----------------CHAPTER 9 Catalytic Strategies tight embrace. An enzyme's binding affinity for substrates often determines specificity. Surprisingly, however, binding studies performed in the absence of magne· sium have demon strated that the EcoRV endonuclease binds to all seq uences, both cognate and noncognate, with approximately equal affinity. Why, then, does the enzyme cleave only cognate sequences? The answer lies in a unique set of interactions between t he enzyme and a cognate DNA sequence. \Vithin the 5' -GATATC-3' sequence, the G and A bases at the 5' end of each strand and their Watson- Crick partners directly contact the enzyme by hydrogen bonding with residues that are located in two loops, one projecting from the surface of each enzyme subunit (Figure 9.39). The most striking feature of this complex is the distortion of the DNA, wh ich is substantiall y kinked in the center (Figure 9.40) . The central two TA base pairs in the recognition sequence play a key role in producing the kink. They do not make contact Kink with the enzyme but appear to be required because of their ease of distortion. The 5'- TA- 3' sequence is known to be among the most easi ly deformed base pairs. The structures of complexes formed with noncognate D N A fragments are strikingly different from those formed with cognate DNA: the noncognate DNA conformation is not substantially distorted (Figure 9.41). This lack of distorDNA helix tion has important consequences with regard to catalysis. No Figure 9.38 EcoRV embracing a cognate DNA molecule. phosphate is positioned sufficiently close to the active-site asThis view o f th e structure o f EcoRV endonuclease partate residues to complete a magnesium ion -binding site (see bound t o a cognate DNA fragment is down the helical axis of Figure 9.36). Hence, the nonspecific complexes do not bind the DNA. The two prote in subun its are in yellow and blue. and the magnes ium ions and the complete catalytic apparatus is th e DNA backbone is in red. N o tice that the twofold axes of never assembled. The distorti on of the substrate and the the enzyme dimer and the DNA are aligned. [Drawn fro m 1RVB.pdb.] subsequent binding of the magnesium ion account for the (A)
(B) Gly 184 Gly 182
Cytosine
Guani ne
Thymine
Adenine
(C)
~ Figure 9.39 Hydrogen-bonding interactions between EcoRY endonuclease
and its DNA substrate. One of t he DNA-binding loops (in green) of EcoRY endonuclease is shown interact ing with the base pairs of its cognate DNA-binding site. Key am ino ac id residues are shown hydrogen bonding with (B) a CG base pair and (C) an AT base pa ir. [Drawn from 1RVB.pdb.]
-
Mg H-binding sites
~~ Figure
figure 9.40 Distortion of the recognition site. The DNA is represented as a ball-and-stick model. The path of the DNA helical axis. shown in red. is substantially distorted on binding to the enzyme. For the B form of DNA. the axis is straight (not shown).
9.41 Nonspecific and cognate DNA within EcoRV
endonuclease. A comparison of the positio ns o f the nonspecific (o range) and the cognate DNA (red) within EcoRV Notice that. in the nonspecific complex. the DNA backbone is too far from th e enzyme to complete the magnesium io n-binding sites. [Drawn from lRVBpdb.]
catalytic specificity of more than 1 ,OOO,OOO -foid that is observed for EcoRV endonculease. Thus, enzyme specificity may be determined by the specificity of
enzyme action rather than the specificity of substrate binding. We can now see the role of binding energy in this strategy for attaining catalytic specificity. The distorted DNA makes additional contacts with the enzyme, increasing the binding energy. However, the increase in binding energy is canceled by the energetic cost of distorting the DNA from its relaxed conformation (Figure 9.42) . Thus, for EcoRV endonuclease, there is little difference in binding affinity for cognate and nonspecific DNA fragments. However, the distortion in the cognate complex dramatically affects catalysis by completing the magnesium ion-binding site. This example il lustrates how enzymes can utilize available binding energy to deform substrates and poise them for chem ical transformation. Interactions that take place within the distorted substrate complex stabilize the transition state leading to DNA hydrolysis. The distortion in the DNA explains how methylation blocks catalysis and protects host-cell DNA. The host E. coli adds a methyl group to the Enzyme + nonspecific DNA
i;O ~
- 0.2 0.0
o
2
4
6
8
10
a Figure 10.15 Quantitative description of the MWC model. In thi s descriptio n of the MWC (Mo no d, Wy man, and Changeaux) mo del, fracti onal acti vity, Y, is the fracti on o f active si te s bo und t o substrate and is directly proportional to react ion ve locity; {X is th e rati o o f [5] t o the d issoc iation constant o f 5 with the enzyme in the R state; and L is th e rati o of the concentration of enzyme in the T state to t hat in the R state. The binding of the regulators ATP and CTP to ATCase changes t he value of L and thus the response to substrate concentratio n. To construct t hese curves, th e formula on page 200 was used, with c = 0.1 and n = 6.
>- '" ~E ... III
.'" 0 0-
~
1.0
--OE
-01'0._ '" 0
>- '" _OE &E .... III
+ 2 mM ATP
10
20
[Aspartatel, mM Figure 10,14 Effect of ATP on ATCase kinetics. ATP is an allosteric activator of aspartate transcarbamoy lase because it stabili zes the R state, making it easier for substrate to bind. As a result, the curve is shifted to the left, as sho wn in blue.
The increase in ATCase activity in response to increased ATP concen· tration has two potential physiological explanations . First, high ATP con· centration signals a high concentration of purine nucleotides in the cell; the increase in ATCase activity will tend to balance the purine and pyrimidine pools. Second, a high concentration of ATP indicates that energy is avail· able for m RN A synthesis and DNA replication and leads to the synthesis of pyrimidines needed for these processes. In the Appendix to Chapter 7, we developed a quantitative description of the concerted model. Although developed to describe a binding process, the model also applies to enzyme activity because the fraction of enzyme ac· tive sites with substrate bound is proportional to enzyme activity. A key as· pect of this model is the equilibrium between the T and the R states (p. 200), We defined L as the equilibrium constant between the R and the T forms,
R
T
[T J L= [RJ
The effects of CTP and ATP can be modeled simply by changing the value of L. For the CTP-saturated form, the value of L increases from 250 to 1250. Thus, it takes more substrate to shift the equilibrium appreciably to the R form . For the ATP saturated form, the value of L decreases to iO (Figure 10.15). Thus, the concerted model provides us with a good description of the kinetic behavior of ATCase in the presence of its key regulators .
(Al
( B) Heart
lDH-l LDH-2 lDH-3 LDH-4 lDH-S
•• •• -9
•• •• •• •• •• •• •• •• •• •• • • •• •• •• •• •• •• ••
H. H,M H2M2
-
Kidn ey
-
Red blood cell
-
-1
+12
+21
-
Leukocyte
Muscle
Liver
-
-
HM, M.
-5
Brain
Adult
10.2 Isozymes Provide a Means of Regulation Specific to Distinct Tissues and Developmental Stages !sozymes, or isoenzymes, are enzymes that differ in amino acid sequence yet catalyze the same reaction. Usually, these enzymes display different kinetic parameters, such as K M , or respond to different regulatory molecules _They are encoded by different genes, which usually arise through gene duplica tion and divergence. Isozymes can often be d istinguished from one another by biochemical properties such as electrophoretic mobility. The existence of isozymes permits the fine-tuning of metabolism to meet the needs of a given tissue or developmental stage. Consider the example of lactate dehydrogenase (LDH), an enzyme that catalyzes a step in anaerobic glu cose metabolism and glucose synthesis. Human beings have two isozymic polypeptide chains for this enzyme ; the H isozyme is highly expressed in heart muscle and the M isozyme is expressed in skeletal muscle. The amino acid sequences are 75% identical. Each functional enzyme is tetrameric, and many different combinations of the two isozymic polypeptide chains are possible. The H4 isozyme, found in the heart, has a higher affinity for sub strates than does the M4 isozyme. The two isozymes also differ in that high levels of pyruvate allosterically inhibit the H4 but not the M 4 isozyme. The other combinations, such as H 3M, have intermediate properties. We will consider these isozymes in their biological context in Chapter 16. The M4 isozyme functions optimally in the anaerobic environment of hard-working skeletal muscle, whereas the H4 isozyme does so in the aerobic environment of heart muscle. Indeed, the proportions of these isozymes are altered in the development of the rat heart as the tissue switches from an anaerobic environment to an aerobic one (Figure 10 .16A). Figure 10.16 B shows the tissue-specific forms of lactate dehydrogenase in adult rat tlssues.
Figure 10.16 Isozymes of lactate dehydrogenase_ (A) The rat heart LDH isozym e pro fil e changes in the course o f development. The H isozyme is represented by squares and the M isozyme by circles. The nega: ive and positi ve numbers deno t e the days before and after birth, respectively. (B) LDH isozyme content varies by ti ssue. [(A) After W -H. Li, M olecular Evo lution (Sinauer, 1997), p. 283; (B) after K. Urich,
Comparative Animal Biochemistry (Springer Verl ag, 1990), p. 542.]
•
W The appearance of some isozymes in the blood is a sign of tissue ~ damage, useful for clinical diagnosis . For instance, an increase in serum levels of H 4 relative to H 3M is an indication that a myocardial infarction, or heart attack, has damaged heart muscle cells, leading to the release of cellular material.
10.3 Covalent Modification Is a Means of Regulating Enzyme Activity he covalent attachment of another molecule can modify the activity of nzymes and many other proteins . Tn these instances, a donor molecule 283
284
TABLE 10.1 Common covalent modifications of protein activity
CHAPTER 10
Reg ulatory Strategies Modification Phospho rylation Acety lation Myristoylation ADP ribosylation Farnesylation -y-Carboxylation Sulfat ion Ubiquitination
I
HN
0
\
H N
/ 'H C
\ Acetylated lysine
CH,
Donor molecule ATP Acetyl CoA MyristoyI CoA NAD+ Farnesyl pyrophosphate
HCO, J-Phosphoadenosine-5'phosphosulfate
Ubiquit in
Example of modified pro te in Glycogen phosph orylase Histones Src RNA polymerase Ras Thrombin Fibri nogen Cycl in
Protein functi on
Glucose homeostasis: energy transducti on DNA packing; transcri ption
Signal transduction Tra nscri pti en Signal transduct ion Blood clotting Blood-clot format ion
Contro l of cell cycle
provides the functional moiety being attach ed. Most modifications are re· versible. Phosphorylation and dephosphorylation are the most common m eans of covalent modification . The attachment of acetyl groups and their removal are another. Histones proteins that are p ackaged with DNA into chromosomes are extensively acetylated and deacetylated in vivo (Section 31.3). More heavily acetylated histones are associated with genes that are bein g actively transcribed. The acetyl transferase and deacetylase enzymes are themselves regulated by phosphorylation, showing that the covalenl m odification of a protein may be controlled by the covalent modification of the modifying enzym es. Modification is not readily reversible in som e cases. The irreversible at· tachment of a lipid group causes some proteins in signal-transduction path· ways, such as Ras (a GTPase) and Src (a protein tyrosine kinase), to becomeaf· fixed to the cytoplasmic face of the plasma membrane. Fixed in this location, the protei ns are better able to receive and transmit information that is being passed along their signaling pathways (C hapter 14). Mutations in both Ras and Src are seen in a wide array of cancers. The attachment of the small protein ubiquitin is a signal that a protein is to be destroyed, the ultimate means of reg· ulation (Chapter 23). The protein cyclin must be ubiquitinated and destroyed before a cell can enter anaphase and proceed through the cell cycle. Virtually all the metabolic processes that we will examine are regulated in part by covalent modification . Indeed, the allosteric properties of many en zymes are modified by covalent modifi cation . Table 10.1 lists some of the common covalent modifications.
Phosphorylation Is a Highly Effective Means of Regulating the Acti vities of Target Proteins W e will see phosphorylation used as a regulatory m echanism in virtually every m etabolic process in eukaryotic cell s. Indeed, as much as 30% of eukaryotic proteins are phosphorylated . The enzymes catalyzing phospho· rylat ion reactions are called protein kinases. These enzym es constitute oneof the largest protein families known: there are m ore than 100 homologous protein kinases in yeast and more than 500 in human beings. This multi· plicity of enzymes allows regulation to be fine -tuned according to a specific tissue, tim e, or substrate. ATP is the m ost common donor of phosphoryl groups. The terminal bl phosphory I grou p of ATP is transferred to a specific amino acid . In eukary· otes, the acceptor is always one of the three containing a hydroxyl group in its side chain . Transfers to serine and threonine resid ues are handled by one class of protein kinases and to tyrosine residues by another. Tyrosine kinases, which are unique to multicellular organisms, play pivotal roles in growth regulation, and mutations in these enzymes are commonly observed in cancer cells.
2-
() - 0 - 0 " ,OH : : : 'V + O···· ;P, . . . , .p .p -0./ ' 0 "'--0 "'--0 I
,
Protei n kinase
,
,
"1 0
---j o
•
\
, HO
OH
AlP
Serine, threonine, Dr tyrosine
NH,
residue
o
2-
~O '-. Lo
N
~, ,,
o HO
OH
ADP
Phosphorylated protein
Table 10.2 lists a few of the known serine and t hreonine protein kinases, The acceptors in protein-phosphorylation reactions are located inside cells, where the phosphoryl-group don or ATP is abundant. Proteins that are entirely extracellular are not regulated by reversible phosphorylation, Protein phosphatases reverse the effects of kinases by catalyzin g the re moval of phosphoryl groups attached to proteins, The enzyme hydrolyzes the bond attaching the phosphoryl group ,
o ~ '-.
o /:.
.c.;.;0
P, ,,
2-
o
Protein
+
phosphatase H 20
,
•
2-
~O H +
•
o
o
Phosphorylated protein
Orthophosphate (PI)
The unmodified hydroxyl -containing side chain is regenerated and orthophosphate (Pi) is produced. These enzymes playa vital role in cells because they turn off the signaling pathways that are activated by kinases. One class of highly conserved phosphatase called PP2A suppresses the cancer· promoting activity of certain kinases. It is important to note tha t the phosphorylation and dephosphorylation reactions are not the reverse of one another; eac h is essentially irreversible TAllE10.2 Exa mptes of seri ne and t hreonine ki nases and t heir activating signa ts Signal C)tlic nucleotides Ca' and calmodulin AMP Diacylglycerol Metabolic Intermediates and other "local" effectors
Enzyme Cyclic AMP·dependent protein kinase Cyclic GMP·dependent protein kinase Ca 2 + - calmodul in protein kinase Phosphorylase kinase or glycogen synthase kinase 2 AMP-activated kinase Protein kinase C
Many target-spec ifi c enzymes, such as pyruvate dehydrogenase kinase and branched·chain ketoacid dehydrogenase kinase
Soorce: After D. FelL Understandlf18 the Control o f Metabolism (Portland Press, 1997). Table 7.2.
285
Protein-OH
+
AlP
1 "
~
Protein- OPO, 2-
1 Protein-OH
+ ADP
H20
+ HO PO, 2-
under physiological conditions. Furthermore, both reactions take place at negligible rates in the absence of enzymes . Thus, phosphorylation of a pro· tein substrate will take place only through the action of a specific protein kinase and at the expense of ATP cleavage, and dephosphorylation will take place only through the action of a phosphatase. The res ult is that tar· get protein s cycle unidirectionall y between unphosphorylated and phos· phorylated forms. The rate of cycling between the phosphorylated and the dephosphorylated states depends on the relative activities of kinases and phosphatases. Phosphorylation is a hi ghl y effective means of controlling the activity 01 proteins for several reasons: 1. A phosphoryl group adds two negative charges to a modified protein. These new charges may disrupt electrostatic interactions in the unmodified protein and allow new electrostatic interactions to be form ed. Such struc· tural changes can markedly alter substrate binding and catalytic activity. 2. A phosphoryl group can form three or more hydrogen bonds. The tetrahedral geometry of a phosphoryl group makes these bonds highly directional, allowing for specific interactions with hydrogen -bond donors. The free energy of phosphorylation is large. Of the -50 kJ mar' (-12 kcal mol I) provided by AT p, about half is consumed in making phosphorylation irreversible; th e other half is conserved in the phosphory· lated protein. A free -energy change of 5.69 kJ mol- 1 (1.36 kcal mol - I) cor· responds to a factor of l Oin an equilibrium constant (p. 210) . Hence, phos· phorylation can change the conformational equilibrium between different 4 functional states by a large factor, of the order of 10 . 3.
4. Phosphorylation and dephosphorylation can take place in less than a second or over a span of hours. The kinetics can be adjusted to meet the tim· ing needs of a physiological process. 5. Phosphorylation often evokes highly amplified effects . A single activated kinase can phosphorylate hundreds of target proteins in a short interval. If the target protein is an enzyme, it may in turn transform a large number of substrate molecules. 6. ATP is the cellular energy currency (C hapter 15). The use of this com· pound as a phosphoryl-group donor links the energy status of the cell to the regulation of metabolism. Protein kinases vary in their degree of specificity. Dedicated protein ki· nases phosp horylate a single protein or several closely related ones. Multifunctional protein kinases modify many different targets; they have a wide reach and can coordinate diverse processes . Comparisons of amino acid sequences of many phosphorylation sites show that a multifunctional kinase recognizes related sequences. For exa mple, the consensus sequence recognized by protein kinase A is Arg-Arg-X-Ser-Z or Arg-Arg-X- Thr-Z, in which X is a small residue, Z is a large hydrophobic one, and Ser or ThT is the site of phosphorylation. It should be noted that this sequence is not absolutely required . L ysine, for example, can substitute for one of the argi· nine residues but with some loss of affinity. Short syntheti c peptides can· taining a consensus motif are nearly always phosphorylated by serine· threonine protein kinases. Thus, the primary determinant of specificity is the amino acid sequence surrounding the serine or threonine phosphorylation site. However, distant residues ca n contribute to specificity. For instance, a 286
change in protein conformation may open or close access to a possible phosphorylation site.
287 10.3 Covalent Modifi cation
Cyclic AMP Activates Protein Kinase A by Altering the Quat ernary Structure The "flight or fi ght" response is common to many animals presented with a dangerous or exciting situ ation . Muscle becomes primed for action . T his priming is the resu lt of the activity of a particular protein kinase. In this case, the hormone epinephrine (adrenaline) triggers the formation of cyclic AMP (cAM P), an intracellular messenger formed by the cyclization of ATP. Cycl ic AM P subsequently activates a key enzyme: protein kinase A (PKA). The kinase alters the activities of target proteins by phosphorylating specific serine or threonine residues. The striking finding is that most effects of cAMP
,
0\
ineukaryotic cells are achieved through the activation by cAMP of PKA .
•
p_ , - 0
OH
0 "";" ,I - 0
PKA provides a clear example of the integration of allosteric regulation and phosphorylation . PKA is activated by cAMP concentrations near 10 nM. The activation mechanism is reminiscent of that of aspartate tran scarbannoylase. L ike that enzym e, PKA in muscle consists of two kinds of subunits: a 49-kd regulatory (R ) subunit and a 38 -kd catalytic (C) subunit . In the absence of cA M P, the regulatory and catalytic su bunits form an R 2 C 2 complex that is enzymatically inactive (Figure 10.17). T he b inding of two molecules of cAMP to each of the regu latory su bunits leads to the dissocia tion of R2C2 into an R 2 subunit and two C subunits. These free catalytic subunits are then enzy m atically active. Thus, the binding of cAMP to the regulatory subunit relieves its inhibition of the catalytic subunit. P K A and most other kinases exist in isozymic form s for fine -tuning regu lation to meet the need s of a specific cell or developmental stage. How does the b indi n g of cAMP activate the kinase? Each R chain con lains the seq uence Arg-Arg-G ly -A la -I1e, which matches t he con sen sus sequence for phosphorylation except for the presence of alanine in place of serine. In the R2CZ complex, this pseudosubstrate sequence ofR occupies th e catalytic site of C, thereby preventing th e entry of protein substrates (see Figure 10. 17). The binding of cAMP to the R chains allosterically m oves the pseudosubstrate sequences out of t he catalytic sites . The released C chains are then free to bind and phosphorylate substrate proteins.
Cyclic adenosine monophosphate
(cAMP)
cAMP ,,"udosubstrate sequence
c
""" ------R R
c
+
4 cAMP
)
R
c
+
R
Active
cAMP·binding domains Figure 10.17 Regulation of protein kinase A. The bi ndi ng of four molecules of cAM P activates protein kinase A by dissoc iat ing the inh ibited ho loenzyme (R, C, ) into a regulato ry subunit (R,) and t wo cata lytical ly act ive subunit s (e). Each R cha in includes cAMP-binding domains and a pseudosubstrate sequence.
+
c Active
ATP and the Target Protein Bind to a Deep Cleft in the Catalytic Subunit of Protein Kinase A
~ Figure 10.18 Protein kinase A bound
to an inhibitor. This space-f ill ing model shows a complex o f t he cat alyt ic subunit of prot ei n kinase A w ith an inhibitor bearing a pseudosubstrate sequence. No t ice that t he inhibito r (yellow) bind s in a cleft between the domains of the enzyme. The bound ATP. shown in red, is in t he act ive si t e adjacent t o the inh ibitor. [Drawn from 1ATP.pdb.)
X- ray crystall ograp hy revea led the three -dimensional structure of the catalytic subu nit ofPKA containing a bound 20-residue peptide inhibitor. T he 350 -residue catalytic subunit has AlP two lobes (Figure 10.18). ATP and part of the inhibitor fill a deep cleft between the lobes. The smaller lobe m akes many contacts with AT P Mg2+ , whereas the larger lobe binds the peptide and contributes the key catal ytic residu e~ . A s with other kinases (p . 269), the Inhibitor two lobes m ove closer to one another on substrate binding; mechanisms that restrict thi s domain closure provide a means of regul ating protein kinase activity. The PKA structure has broad significance because residues 40 to 21W constitute a conserved cataly tic core that ~ common to essentially all known protein kinases. W e see here an example of a successful biochemical solution to a problem (in this case, protein phosphorylation) being em ployed many times in the cour.of evolution . T he bound peptide in this crystal occupi es the active site because it contains t he pseud osubstrate sequen ce A rg-A rg-Asn-AlaIle (F igure 10.19). T he st ructure of the complex reveals the interactions by which the en zyme recogni zes th e consensus sequence. T he guanidinium group of the first arginine residue form s an ion pair with the carboxylate side chain of a glutamate residue (G lu 127) of the enzyme. The second arginine likewise interacts with two other carboxylate groups. The nonpolar side chain of isoleucine, which matches Z in the consensus sequence (see p . 285), fits snugly in a hydrophobic groove formed by two leucine residues of the enzyme.
Glu 127
Glu 170
AlP
Asn (side chain not shown) Ala
Figure 10.19 Binding of pseudosubst rat e to protein kinase A. No tice that the inhibit or makes mul tip le contacts w it h the enzyme. The two argi nine side chains o f the pseudosubstrate fo rm salt bridges w ith t hree glut amate carboxylate groups. Hydro phobic interact ions are also important in t he recognition of substrate. The isoleuci ne residue o f the pseudosubstrate is in contact w ith a pair o f leuci ne res idues of the enzyme.
Glu 230
lie ~
~Leu
198
Leu 205
•
10.4
Many Enzymes Are Activated by Specific Proteolytic Cleavage
W e turn now to a different mechani sm of enzyme regulation . Many enzymes acquire full enzymatic activity as t hey spon taneously fold into their characteri sti c t hree -dim ensional forms. In contrast, the folded forms of other enzymes are inactive until activated by cleavage of one or a few specific peptide bonds. The inactive precursor is called a zymogen or a proenzyme. An energy 288
289
[All£ 10.3 Gastric and pancreatic zymogens
Srre of synthesis
Stomach Paocreas Pancreas Pancreas Pancreas
Zymogen
Acti ve enzyme
Pepsinogen Chy motrypsi nogen Trypsi nogen Proca rboxypeptidase
Peps in Chymotrypsin Trypsi n Carboxypeptidase
Proelastase
Elastase
lOA Activation by Proteolytic Cleavage
source slich as ATP is not needed for cleavage. Therefore, in contrast with reversible regulation by phosphorylation, even protei ns located outside cells can be activated by this means. Another noteworthy d ifference is that proteolytic activation , in contrast with allosteric control and reversible covalent modification, occurs just once in th e life of an enzyme m olecule. Specific proteolysis is a common m ean s of activating enzymes and other proteins in biological systems. For example: I. The digestive enzymes that hydrolyze proteins are synthesized as zymo -
gens in the stomach and pancreas (Table 10.3 ).
2. Blood clotting is mediated by a cascade of proteolytic activations that ensures a rapid and amplified response to trauma.
3. Some protein hormones are synthesized as inactive precursors. For example, insulin is derived from pruinsulin by proteolytic removal of a peptide. 4. The fibrous protein collagen, the major constituent of skin and bone, is derived from procollagen, a solu ble precursor.
i. Many deve lopmental processes are controlled by the activation of zymo gens. FDr example, in the metamorphosis of a tadpole into a frog, large amounts of collagen are resorbed from the tail in the course of a few days. Likewise, much collagen is broken down in a mammalian uteru s after deliv ery. The conversion of procollagenase into co llagenase, t he active protease, is precisely timed in these remodeling processes.
6. Programmed cell death , or apoptusis, is m ediated by proteolytic enzymes called caspases, which are synthesized in precursor form as procaspases. When activated by various signals, caspases function to cause cell death in most organi sms, ranging from C. elegans to human beings. Apoptosis provides a means of sculpting the shapes of body parts in the course of d evelopment and a means of eliminating damaged or infected cells . We next exam ine the activation and control of zymogen s, usin g as examples several digestive en zym es as well as blood -clot for m ation.
Ribosomes attached to endoplasmic reticulum
Golgi complex
Zymogen granule
Chymotrypsinogen Is Activated by Specific Cleavage of aSingle Peptide Bond
Chymotrypsin is a digestive enzyme that hydrolyzes proteins in th e small in testine. Its mechan ism of action was described in detail in Chapter 9. Its inactive precursor, chymotrypsinogen, is synthesized in the pancreas, as are several other zymogens and digestive enzym es. Indeed , the pancreas is one of the most active organs in synthesizing and secretin g proteins . The enzymes and zymogen s are synthesized in the acinar cells of the pancreas and stored inside membrane -bounded granules (Figure 10 .20). The zymogen granules accu mulate at the apex of the acinar cell; when the cell is stimulated by a hormonal signal or a nerve impul se, the contents of the granules are released into a duct leading into the duodenum.
lumen Figure 10.20 Secretion of zymogens by an acinar cell of t he pancreas.
Chymotrypsinogen (i nactive)
Trypsin
n-Chymotrypsin (active) II
151 LII..:. 6 _ _ _ __ _ _-=2:...:4::J sl n-Chymotrypsin Two dipeptides
a-Chymotrypsin (active) II
13 1 ILI-6'-_ _ _ _1__4-'.j 6 1 1 149
A chain
B chain
245 1
C chain
Figure 10.21 Proteolytic activation of chymotrypsinogen. The three chains of ex-chy mo trYPsin are linked by t wo interchain d isulfide bonds (A t o B, and B t o C).
C hymotrypsinogen, a single polypeptide chain consisting of 245 amino acid residues, is virtually devoid of enzymatic activity. It is converted intoa full y active enzyme when the peptide bond joining arginine 15 and isoleucine 16 is cleaved by trypsin (F igure 10. 21). The resulting active enzyme, called 'IT-chymotrypsin, then acts on other 'IT -chymotrypsin molecules. Two dipep. tides are removed to yield a-chymotrypsin, the stable form of the enzyme. The three resulting chains in a-chymotrypsin remain linked to one another by two interchain disulfide bonds. The striking feature of this activation process is that cleavage of a single specific peptide bond transforms the protein from a catalytically inactive form into one that is fully active.
Proteolytic Activation of Chymotrypsinogen Leads to the Formation of a Substrate-Binding Site How does cleavage of a single peptide bond activate the zymogen? The cleavage of the peptide bond between amino acids 15 and lli triggers key conformational changes, which were revealed by the elucidation of the three-dimensional structure of chymotrypsinogen. The newly formed amino-terminal group of isoleucine 16 turns inward and forms an ionic bond with aspartate 194 in the interior of the chymotrypsin molecule (Figure 10.22). 1.
lie 16 (chymotrypsinogen) lie 16 (chymotrypsin) """'" Figure 10.22 '>V Conform ations of chymotrypsinogen (red) and cchymotrypsin (b lue). Notice the alteration of the pOSition of iso leuc ine 16 in chy motrYPSin. The electrostatic interacti o n between th e ex-amino group of isoleucine 16 and the carboxylate o f aspartate 194, essential for the stru cture o f active chymotrypsin. is possible only in chymotrYPsin. [Draw n from 1GCT.pdb and 2GCA.pdb.]
2. This electrostatic interaction triggers a number of conformational changes. Methionine 192 m oves from a deeply buried position in the zymo· gen to the surface of the active enzyme, and residues 187 and 193 become more extended. These changes result in the formation of the substrate· specificity site for aromatic and bulky nonpolar groups. O ne side of this site is made up of residues 1119 through 192. This cavity for binding part of the substrate is not fully formed in the zymogen. 3. The tetrahedral transition state in catalysis by chymotrypsin is stabi· li zed by hydrogen bonds between the negatively charged carbonyl oxygen atom of the substrate and two NH groups of the main chain of the enzyme (p . 247) . One of these NH groups is not appropriately located in chy· motrypsinogen, and so the oxyanion hole is incomplete in the zymogen. 4. The conformational changes elsewhere in the molecule are very small. Thus, the switching on of enzymatic activity in a protein can be accomplished by discrete, highly localized conformational changes that are triggered by the hydrolysis of a single peptide bond. 290
The Generati on of Trypsin f rom Trypsinogen Leads t o the Activati on of Other Zymogens
29 1 10.4 Activation by Proteolytic Cleavage
The structural changes accompanying the activation of trypsinogen, the precursor of the proteolytic enzyme trypsin, are somewhat different from those in the activation of chymotrypsinogen. X -ray analyses have shown that the conformation of four stretches of polypeptide, constituting about Jj%ofthe molecule, changes markedly on activation. These regions are very
flexible in the zymogen, whereas they have a well-defined conformation in trypsin. Furthermore, the oxyanion hole (p. 247) in trypsinogen is too far from histidine 57 to promote the formation of the tetrahedral transition state. The digestion of proteins in the duodenum requires the concurrent ac tion of several proteolytic enzymes, because each is specific for a limited number of side chains. Thus, the zymogens must be switched on at the sam e time. Coordinated control is achieved by the action of trypsin as the common activator of all the pancreatic zymogens trypsinogen, chymotrypsinogen, proelastase, procarboxypeptidase, and prolipase, a lipid degrading enzyme. To produce active trypsin, the cells that line the duodenum secrete an enzyme, enteropeptidase, which hydrolyzes a unique lysine- isoleucine peptide bond in trypsinogen as the zymogen enters the duodenum from the pancreas. The small amount of trypsin produced in this way activates more trypsinogen and the other zymogens (Figure 10.23). Thus, the formation of trypsin by enteropeptidase is the master activation step. Enteropeptidase
TrypSinogen
Trypsin
, Proelastase
•
/
Elastase
Procarboxypeptidase
Carboxypeptidase
,
,Chymotrypsinogen
Chymotrypsin
Prolipase
Lipase
Some Proteolyt ic Enzymes Have Spec ifi c Inhibitors The conversion of a zymogen into a protease by cleavage of a single peptide bond is a precise means of switching on enzymatic activity. However, this activation step is irreversible, and so a different m echanism is needed to stop proteolysis. Specific protease inhibitors accomplish this task. For example, pancreatic trypsin inhibitor, a 6-kd protein, inhibits trypsin by binding very tightly to its active site. The di ssociation constant of the complex is 0.1 pM, which corresponds to a standard free energy of binding of about -is kJ mol - I ( -18 kcal mol - I). Tn contrast with nearly all known protein assemblies, this complex is not dissociated into its constituent chains by treatment with d enaturing agents such as 8 M urea or 6 M guanidine hydrochloride. The reason for the exceptional stability of the complex is that pancreatic trypsin inhibitor is a very effective substrate analog. X -ray analyses showed that the inhibitor lies in the active site of the enzyme, positioned such that
Figure 10.23 Zymogen activation by proteolytiC cleavage. Enteropeptid ase initiates the acti vation of the pancreatic zymogens by activating trypsin. wh ich then activates other zymogens. Active enzymes are shown in yell ow ; zymogens are shown • In orange.
15
Asp 189
Trypsin - pancreatic trypsin
inhibitor complex
Free pancreatic trypsin inhibitor
the side chain of lysine 15 of this inhibitor interacts with the aspartate side chain in the specificity po.:ket of trypsin. In addition, there are many hydrogen bonds between the main chain of trypsin and that of its inhibitor. Furthermore, th e carbonyl group of lysine 15 and the sur· rounding atoms of the inhibitor fit snugly in the active site of the enzyme. Compari son of the structure of the in· hibitor bound to the enzyme with that of the free inhibitor reveals that the structure is essentially unchanged on binding to the enzyme (Figure 10.24). Thus, the inhibitor is preor· gani zed into a structure that is hi ghl y com plementary to the enzyme's active site. Indeed, the peptide bond between lysine 15 and alanine 16 in pancreatic trypsin inhibitor is cleaved but at a very slow rate: the half-life of the trypsin- inhibitor complex is several months. In essence, the inhibitor is a substrate, but its intrinsic structure is so nicely complementary to the enzyme's active site that it binds very tightly and is turned over slowly.
The amount of trypsin is much greater than the ~ Figure 10.24 Interaction of trypsin w ith its inhibitor. amount of inhibitor. Why does trypsin inhibitor Stru cture of a compl ex o f trypsin (yellow) and pancreatic exist? Recall that trypsin activates other zy mogens. trypsin inhibitor (red). Notice that lysi ne 15 of the inhibitor penetrates int o the acti ve site of the enzyme. There it forms Consequently, the prevention of even small amounts of a salt bridge wi th aspartate 189 in the active site. Notice also trypsin fro m initiating the inappropriately activated that bound inhibit o r and the free inhibitor are almost cascade prematurely is vital. Trypsin inhibitor binds to identical in structure. [Draw n from 2PTC.pdb.] trypsin molecules in the pancreas or pancreatic ducts. This inhibiti on prevents severe damage to those tissues, which could lead to acute pancreatitis. Pancreatic trypsin inhibitor is not the onl y important protease inhibitor. a r Antitrypsin (also called a l-antiproteinase ), a 53 -kd plasma protein, pro· tects tissues from digestion by elastase, a secretory product of neutrophils (white blood cell s that engulf bacteria). Antielastase would be a more accu· rate name for this inhibitor, because it blocks elastase much more effectively than it blocks trypsin . Like pancreatic trypsin inhibitor, aI-antitrypsin blocks the action of target enzymes by binding nearly irreversibl y to their active sites. Genetic disorders leading to a deficiency of <X l-antitrypsin show that this inhibitor is physiologically important . For example, the substitu· tion of lysine for glutamate at residue 53 in the type Z mutant slows the se· cretion of this inhibitor from liver cells. Serum levels of the inhibitor are about 15% of normal in people homozygous for this defect. The conse· quence is that excess elastase destroys alveolar walls in the lungs by digest· ing elastic fibers and other connective-tissue proteins. The resulting clinical condition is called emphysema (also known as destructive lung disease). People with emphysema must breathe much harder than normal people to exchange the sam e volume of air because their alveoli are much less resilient than normal. C igarette smoking markedly increases the likelihood that even a type Z heterozygote will develop em· physema. The reason is that smoke oxidizes methionine 358 of the inhibitor (Figure 10.25), a residue essential for binding elastase. Indeed, this methio· Oxidation nine side chain is the bait that selectively traps elastase. The methionine sui· H ,• H foxid e oxidation product, in contrast, does not lure elastase, a striking con· C/ '---N '---N sequence of the insertion of just one oxygen atom into a protein and a H H o 0 striking example of the effect of behavior on biochemistry. We will consider another protease inhibitor, antithrombin III, when we examine the control Figure 10.25 Oxidation of methionine to of blood clotting . meth ionine sulfoxide. )
292
INTRINSIC PATHWAY
Blood Clotting Is Accomplished by a Cascade of Zymogen Activations
Damaged surface
1
Enzymatic cascades are often employed in biochemical
systems to achieve a rapid response . In a cascade, an Kininogen initial signal institutes a series of steps, each of which Kallikrein is catalyzed by an enzyme. At each step, the signal is XII. amplified. For instance, if a signal molecule activates an enzyme that in turn activates 10 enzymes and each of the 10 enzymes in turn activates 10 additional enzymes, after four steps the original signal will have been amplified lO,OOO-fold. Blood clots are formed by a cascade of zymogen activations: the activated form of one clotting factor catalyzes the activation of the next (Figure 10.26 ). Thus, very small amounts of the initial factors suffice to trigger the cascade, ensuring a rapid response to trauma. Two means of initiating blood clotting have been described, the intrinsic pathway and the extrinsic path way. The intrinsic clotting pathway is activated by exposu re of anionic surfaces on rupture of the enFINAL dotheial lining of the blood vessels. The extrinsic COMMON PATHWAY pathway, which appears to be most crucial in blood clotting, is initiated when trauma exposes tissue fa ctor (Tf), an integral membrane glycoprotein. Shortly * = activated by thrombin after the tissue factor is exposed, small amounts of thrombin, the key protease in clotting, are generated. Thrombin then activates enzymes and factors that lead to the generation of yet more thrombin, an example of positive feedback. The extrinsic and in trinsic pathways converge on a common sequence of final steps to form a clot composed of the protein fibrin (Figure 10.26). Note that the active forms of the clotting factors are d esignated with a subscript "a."
Fibrinogen Is Converted by Thrombin into a Fibrin Clot The best-characterized part of the clotting process is the final step in the cascade: the conversion of fibrinogen into fibrin by thrombin, a proteolytic enzyme. Fibrinogen is made up of three globular units connected by two rods (Figure 10 .27) . This J40 -kd protein consists of six chains : two each of ACI. , B ~ , and 'Y . The rod region s are triple-strand ed ex-helical coiled coils, a recurring motif in proteins (p . 45). Thrombin cleaves four arginine- glycine peptide bonds in the central globular region of fibrino gen . On cleavage, an A peptide of 1 R residues is released from each of the two Au chains, as is
EXTRINSIC PATHWAY XI,
VII
VII,
IX.
Tissue factor .-.- - Trauma
N III
Xa
Prothrombin
Thrombin (i I.)
(II) Fibrinogen (I)
Fibrin (Ia) 'XIIi. Cross-linked fibrin clot
L
Figure 10.26 Blood-clotting cascade. A fibr in clot is formed by the interplay of the intrinsic. extrinsic, and final common pathways. The intrinsic pathway begins with the activation of factor XII (Hageman factor) by contact with abnormal su rfaces produced by in jury. The extrinsic pathway is triggered by trauma, which re leases tissue factor (TF). TF forms a complex with VII, which initiates a cascade-activating thrombin. Inactive fo rms of clotting factors are shown in red; their activated counterparts (indicated by the subscript "... ) are in yellow. St imulat o ry proteins that are not themselves enzymes are shown in blue boxes. A striking feature of this process is that the activated form of one clottin g factor catalyzes the activation of the next factor.
(A)
~
(B)
B-
_ -,-
B
Cleavage site
-----/
t
A-
.... A
Globular unit
Figure 10.27 Structure of a , fibrinogen molecule. (A) A ribbon diagram. The two rod regions are a-helical coiled coils. connected to a globular region at each end. The structure of the centra l globular region has not been determined. (B) A schematic representation shOWing the posit ions of the fibrinopeptides A and B. [Part A drawn from 1DEQpdb.]
293
a B peptide of 20 residu es from each of the two BI3 chains. These A and B pep tides are called fibrinopeptides. A fib· rinogen molecule devoid of these fibrinopeptides is called a fibrin monomer and has the subunit structure (ex l3'Y lz. Fibrin monomers spontaneously assemble into ordered fibro us arrays called fibrin . E lectron m icrographs and low· angle x-ray patterns show that fibrin has a periodic structure that repeats every 23 nm (Figure 10.28). Higher-resolution Figure 10.28 Electron micrograph of fibrin . The 23-nm period images reveal how the removal of the fibrinopeptides permits along the fiber axi s is half the length of a fibrinogen molecule. [Courtesy o f Dr. Henry Slayter.] the fibrin monomers to come together to form fibrin. The homologous 13 and 'Y chains have globular domains at the car· boxyl-terminal ends (Figure 10.29). These domains have binding "holes" that interact with peptides. The 13 domain is specific for sequences of t he form H 3N I -Gly-His-Arg-, whereas the 'Y domain binds H 3N+ -Gly-Pro-Arg-. Exactly these sequences (sometimes called "knobs") are exposed at the amino-terminal ends of the 13 and ex chains, respectively, on thrombin cleav· age. The knobs of the ex subunits fit into the holes on the 'Y subunits of another monomer to form a protofibril. This protofibril is extended when the knobs of the 13 subunits fit into the holes of 13 subunits of other protofibrils. Thus, analogous to the activation of chymotrypsinogen, peptide-bond cleavage exposes new amino termini that can participate in specific interactions. The newly formed "soft clot" is stabilized by the formation of amide bonds between the side chains oflysine and glutamine residues in different monomers. ,
• ,,
o
,,
•
,
o (,
,
Tra nsglutaminase .
, , ,
•
Lysine
Glutamine
, • ,
•
,
'( = 0
0
•
HN
\ ( /'--.., , H
H ",(
\
" 'N H
+ NH:
NH
,, ••
0= (•
,
•
•
•,
Cross-link
This cross -linking reaction is catalyzed by transglutaminase (factor XTTl a ), which itself is activated from the protransglutaminase form by thrombin , "'\ .J
.. r
Fibrinopeptid es
sequences
Polym erization )
Thrombin
CD Figure 10.29 Formation of a fibrin clot. (1) Thrombin cleaves fibrinopeptides A and B from the central globule of fibrinogen. (2) Globular domains at the carboxyl -t erminal ends of the i3 and -y cha ins interact with "knobs" exposed at the amino-terminal ends of the i3 and -y chains to form clots.
294
Gly-Pro-Arg sequences
295
Prothrombin Is Readied for Activation by a Vitamin K-Dependent Modif ication
10.4 Ac t ivation by Proteolytic Cleavage
Thrombin is synthesized as a zy mogen called prothrombin. The inactive molecule comprises four major domains, with the serine protease domain at its carboxyl terminu s. T he fir st d om ain is called a g la domain (a 'Ycarboxyglutamate-rich domain), and the second and third domains are called hingle domains (named after a D anish pastry that t hey resembl e; Figure 10.30). T hese domains work in co ncert to keep prothrombin in an in active fo rm and to target it to appropriate sites for its activation by factor X. (aserine protease) and factor Va (a stimulatory protein). Activation is begun by proteolytic cleavage of the bond between arginine 274 and t hreonine 275 to release a fragm en t con taining the first three d omains. C leavage of the bond between arginine 323 and isoleucine 324 (analogous to the key bond in chymotrypsinogen ) yields acti ve thrombin. Cleavage sites Gla
Kringle
Kringle
I
I
I
I
Seri ne protease
I
figure 10.30 Modu lar structure of prothrombi n. Cleavage o f two peptide bonds y ields thrombin. All t he -y -carbo xyglutamat e resid ues are in th e ela do ma in.
W Vitamin K (p. 295 and F igure 10.31) has been known for many years ~ to be essential for th e synthesis of prothrombin and several other clotting factors. Indeed, it is called vitamin K because a defi ciency in this vitamin results in defective blood koagulation (Scandinavian spelling). The results of studies of t he abnormal proth rom bin synthesized in the absence of vitamin K or in the presen ce of vitamin K antagonists, such as d icoumarol, revealed the vitamin's m od e of action . Dicoumaro l is found in spoiled sweet clover and causes a fatal hem orrhagic disease in cattle fed on this hay. T his coumarin derivative is used clinically as an anticoagulant to prevent thromboses in patients prone to clot formation. D icoumarol and such related vitamin K antagoni sts as warfarin also serve as effective rat poisons. Cows fed dicoLlmarol synthesize an abnormal prothrombin that does not bind ci+, in con trast with n ormal pro thrombin . This difference was puzzling for som e time because abnormal prothrombin has the same numberof amino acid residues as that of normal p rothrombin and gives the sam e amino acid an alysis after acid hyd rolysis. Nuclear magnetic resonan ce studies revealed that normal prothrombin contains y -carboxyglutamate, a form erly unknown residue that evaded detection because its second carboxyl group is lost on acid hydrolysis during amino acid analysis. The abnormal prothrombin formed subsequent to the administration of an ticoagulants lacks this m odified amino acid. In fact, the first 10 glutamate residues in th e amino- terminal region of p rothrombin
o :I o-· __ ·c
0
;:
\
,'
-
C
CH--- " '0 H •
- --N •
•
• ••
•
H
0 ,),eCarboxyglutamate residue
..... 0
0
o o
0 0
o
.'~
..
"
CH3 6
o
H "
CH 3 \Jitamin K
CH 3 Dicoumarol
figure 10.31 Structures of vitam in K and two antagonists, dicoumarol and warfarin.
H
\ =0
/ H3C Warfarin
are carboxylated to 'Y -carboxyglu tamate by a vitami n K -depend ent enzyme system (Figure 10.32 ). The vitamin K-dependent carboxylation reaction con· 2 verts glutamate, a weak chelator of Ca +, into y-carboxyglutamate, a much stronger chelator. Prothrombin is thus able to bind Ca2+ , but what i the 2 effect of this binding? The binding of Ca + b y prothrombin anchors the zy· mogen to phospholipid membranes d eri ved from blood platelets after injury. The binding of prothrombin to phospholipid surfaces is crucial be· cause it brings prothrombin into close proximity to two clotting protein that catalyze its conversion into thrombin . The calcium-binding domain is removed during activation, freeing the thrombin from the membrane so that it can cleave fibrinogen and other targets.
Hemophilia Revealed an Early Step in Clotting
Calcium ions
~ Figure 10.32 The calcium-binding
region of prothrombin. Proth ro mbin binds calc ium ions w ith t he mod ified am ino acid "/-carboxyglutamat e (red). [Drawn from 2PF2.pd b.j
An account of a hemorrhagic disposition
e xisting in certain families "About seventy or eighty years ago. a woman by the name of Smith settled in t he vicini t y of Plymout h. New Hampshire. and transm it ted the following idiosyncrasy to
her descendants. It is one, she observed, to which her fami ly is un fortu nately subj ect and has been the source not only o f great
sol ici tude. but frequently the cause of death. If the least scratch is made on the skin of some of them, as morta l a hemorrhage wil l eventually ensue as if the largest
wound is infli cted. . . . It is a surprising circumstance that rhe males only are subject to th is strange affection, and t hat all of t hem are not liable to it . , , , Altho ugh the females are exem pt. they are stil l capable of t rans mitting it to the ir male chil dren." JOl iN
Ono (1803)
Some important breakthroughs in the elucidation of clotting pathways have come from studies of patients with bleeding di~orders. Classic hemophilia, or hemophilia A, is the best -known clotting d efect. This disorder is genetically transmitted as a sex-linked recessive characteristic. In classic hemophilia, fa ctor VIiI (antihemophilic factor) of the intrinsic pathway is missing or has markedly reduced activity. Although factor VIII is not itself a protease, il marked ly stimu lates the activation of factor X, the final protease of the intrinsic pathway, by factor IX n , a serine protease (Figure 10 .33) . Thus, ac· ti vation of the intrinsic pathway is severely impaired in hemophil ia. In the past, hemophiliacs were treated with transfusions of a concen· trated plasma fracti on containing factor VIII. This therapy carried the risk of infection. Indeed , many hemophiliacs contracted hepatiti s and , more reo cently, AIDS. A safer preparation of factor VIII was urgently need ed . With the use of biochemical purification and recombinant DNA techniques, the gene for factor VIII was isolated and ex pressed in cells grown in cul ture. Recombinan t factor VIII purified from these cells has largely replaced plasma concentrates in treating hem ophili a.
Figure 10.33 Action of antihemophilic factor. Antihemophilic factor (Factor VIII) st im ulates th e activation o f fa ct or X by factor IX,. It is interesting to note that the activity of fa ct o r VII I is markedly increased by lim ited proteolysis by t hromb in. This posi t ive feedback amplifies th e clotting Si gnal and accelerates clot format ion after a threshold has been reached.
Anti hemophilic fa ctor
(VIII) +, ,
x. '
, . ,/ Proteo lYSIS ,
The Clotting Process Must Be Precisely Regulated
296
There is a f ine line between hem orrhage and thrombosis. C lots m ust form rapidly yet remain confined to the area of injury. What are the m echanisms that normall y limit clot format ion to t he site of injury? The lability of clotting factors co ntributes significantly to the control of clotting . Activated factors are short-li ved because they are di luted by blood flow, removed by the li ver , and degraded by proteases. For example, th e stimulatory protein fa ctors Va and VIII. are digested by protein C, a protease that is switched on by the action of thrombin . Thus, thrombin has a dua l fun ction: it catalyzes the formation of fibrin and it initiates the deactivation of the clotting cascade. Specific inhibitors of clotting factors are also critical in the termination of clotting. For instance, tissue factor pathway inhibitor (TFPI) inhibits the com· plex ofTF- VIl.- X •. Separate domains in TFPT inhibit VII. and Xa. Another key inhibitor is antithrombin III , a plasma protein that inactivates thrombin by forming an irreversible complex with it. Antithrombin III resembles
297
at -antitrypsin except that it inhibits thrombin much more strongly than it inhibits elastase (see Figure 10. 26). Antithrombin III al so blocks other serine proteases in the clotting cascade namely, factors XTl a, XI., IXa, and Xa. The inhibitory action of antithrombin III is enh anced by heparin, a negatively charged polysaccharide found in mast cells near the walls of blood vessels and on the surfaces of endothelial cells (Figure 10.3 4). H eparin acts as an anticoagulant by increasing the rate of formati on of irreversible complexes between antithrombin HI and the serine protease clotting factors. Antitrypsin and an tithrombin are serpins, a family of serine protease inhibitors.
W
T he importance of th e ratio of thrombin to antithrombin is illus~ trated in the case of a 14-year- old boy who died of a bleeding disor der because of a mutation in his <X I -antitrypsin, which normally inhibi ts elastase. Methionine 358 in <X I-antitrypsin's binding pocket for elastase was replaced by arginine, resulting in a chan ge in specificity from an elastase in hibitor to a th rombin inhibitor. <x t -A ntitryp sin activity normally increases markedly after injury to counteract excess elastase arising from stimul ated neutrophils. The mutant (X I -antitrypsin caused the patient's thrombin ac tivity to drop to such a low level th at hemorrhage ensued . We see here a
strihing example of how a change of a single residue in a protein can dramatically alter specificity and an example of the critical importance of having the right amount of a protease inhibitor. Antithrombin limits the extent of clot formation, but what happens to the clots themselves? C lots are not permanen t structures but are designed to desolve when the structural integrity of dam aged areas is restored . F ibrin is split by plasmin, a serine protease that hyd rolyzes peptide bonds in the coiled coil regions. P lasmin m olecules can di ffuse through aqueous channels in t he porous fibrin clot to cut the accessible connector rods. Plasmin is formed by Ihe proteolytic activation of plasminogen, an inactive precursor that has a high affinity fo r the fibrin clots. This conversion is carried out by tissue-type plasminogen activator (TPA), a 72- kd protein that has a domain structure closely related to that of prothrombin (Figure 10.3 5). fibrin binding
Kringle
-~=---
Kringle
10.4 Activation by Proteolytic Cleavage
I
)
/
...
-
I.
Figure 10.34 Electron micrograph of a mast cell. Heparin and other molecules in the dense granules are released into the extracellular space when the cel l is triggered to secrete. [Courtesy of Lynne Merce r.]
Serine protease
figure 10.35 Modular structure of tissue-type plasminogen activator (TPA).
However, a domain that targets TPA to fibrin clots replaces the m em branetargeting gla domain of prothrombin. The T PA bound to fibrin clots swiftly activates adhering plasminogen . In contrast, TPA activates free plasminogen very slowly. The gene for T PA has been cloned and expressed in cul tured mammalian cells. C linical studies have shown that TPA admini stered intravenously within an hour of the formation of a blood clot in a coronary artery markedly increases the likelihood of surviving a heart attack (Figure 10.36). (A)
(6) Figure 10.36 The effect of tissue-type plasminogen factor. TPA leads to t he di ssolution of blood clots, as shown by x-ray images of blood vessels in the heart (A) before and (B) 3 hours after the adm in ist rat ion of TPA. The positio n of t he clot is marked by the arrow in part A. [After F. Van de Werf. P. A. Ludbrook, S. R. Bergmann. A. J. Tiefenbrunn. K. A. A. Fox, H. de Geest, M. Verstraete. D. Collen, and B. E. Sobel. New Engl.}. Med.
310(1984)609- 613.]
298 CHAPTER 10 Regulatory Strategies
Summary 10.1 Aspartate Transcarbamoylase Is Allosterically Inhibited by the End Product of Its Pathway Allosteric proteins constitute an important class of proteins whose biological activity can be regulated _ Specific regulatory molecules can modulate the activity of allosteric proteins by binding to distinct regulatory sites, separate from the functional sites. These proteins have multiple functional sites, which display cooperation as evidenced by a sigmoidal dependence of function on substrate concentration. Aspartate transcarbamoylase (ATCase), one of the best-understood allosteric enzymes, catalyzes the synthesis of N-carbamoylaspartate, the first intermediate in the synthesis of pyrimidines. ATCase is feedback inhibited by cytidine triphosphate, the final product of the pathway. ATP reverses this inhibition . ATCase consists of separable catalytic (C3) subunits (which bind the substrates) and regulatory (r2) subunits (which bind CTP and ATP). The inhibitory effect of CTP, the stimu· latory action of ATP, and the cooperative binding of substrates are mediated by large changes in quaternary structure. On binding sub· strates, the c, subunits of the C6r6 enzyme move apart and reorient themselves. This allosteric transition is highly concerted. All subunits of an ATCase molecule simultaneously interconvert from the T (low. affinity ) to the R (high-affinity ) state. 10.2 Isozymes Provide a Means of Regulation Specific to Distinct Tissues and Developmental Stages Isozymes differ in structural characteristics but catalyze the same reaction. They provide a means of fine -tuning metabolism to meet the needs of a given tissue or developmental stage. The results of geneduplication events provide the means for subtle regulation of enzyme function. 10.3 Covalent Modification Is a Means of Regulating Enzyme Activity The covalent modification of proteins is a potent means of controlling the activity of enzymes and other proteins. Phosphorylation is the most common type of reversible covalent modification . Signals can be highly amplified by phosphorylation because a single kinase can act on many target molecules. The regulatory actions of protein kinases are reversed by protein phosphatases, which catalyze the hydrolysis of attached phosphoryl groups. Cyclic AMP serves as an intracellular messenger in the transduction of many hormonal and sensory stimuli. Cyclic AMP switches on protein kinase A, a major multifunctional kinase, by binding to the regulatory subunit of the enzyme, thereby releasing the active catalytic subunits of PKA. In the absence of cAMP, the catalytic sites of PKA are occupied by pseudosubstrate sequences of the regulatory subunit. lOA Many Enzymes Are Activated by Specific Proteolytic Cleavage
The activation of an enzyme by the proteolytic cleavage of one or a few peptide bonds is a recurring control mechanism seen in processes as di verse as the activation of digestive enzymes and blood clotting. The inactive precursor is a zymogen (proenzyme). Trypsinogen is activated by enteropeptidase or trypsin, and trypsin then activates a host of other zymogens, leading to the digestion of foodstuffs. For instance, trypsin converts chymotrypsinogen, a zymogen, into active chymotrypsin by hydrolyzing a single peptide bond. A striking feature of the clotting process is that it is accomplished by a cascade of zymogen conversions, in which the activated form of
one clotting factor catalyzes the activation of the next precursor. Many of the acti vated clotting factors are serine proteases. In the final step of clot formation, fibrinogen, a highly solu ble molecule in the plasma, is converted by thrombin into fibrin by the hydrolysis of four arginine- glycine bonds. The resulting fibrin monomer spontaneously forms long, insoluble fibers called fibrin . Zymogen activation is also essential in the lysis of clots . Plasminogen is converted into plasmin, a serine protease that cleaves fibrin, by tissue-type plasminogen activatOt . Although zymogen activation is irreversible, specific inhibitors of some proteases exert control. The irreversible protein inhibitor an ,ti.tbIom bin III holds blood clotting in check in the clotting cascade.
299 Selected Readings
Key Terms roopera6vity (p. 275)
h eter otropic effect (p . 282)
protein kinase A (PKA ) (p . 287)
feedback (end -product) inhibition (p . 277)
isozyme (isoenz y m e) (p. 283)
pseud osu bstrate seq u ence (p . 287)
allosteri c (regulatory ) site (p . 277)
cova le n t m odification (p . 283)
zymogen (proenzyme) (p. 288)
concerted mechan is m (p . 281)
protein kinase (p . 284)
en zymati c cascad e (p . 293)
sequen tial model (p . 281)
pro te in phosphatase (p. ZH5)
intrins ic pathway (p . 293)
homotropic effect (p . 281)
con sen s u s sequence (p . 286)
extrins ic pathway (p . 293)
Selected Readings Where to Start Kantrowitz, E. R., and Lipscomb. W . N . 1990. Escherichia coli aspartate (ransearbamoylase: The m olecular basis for a concerted allosteric.: (ransition . Trends Biochern. Sci. 15:53- 59 . Sehachman, t-I . K . 1988. Can a simple model acco un t for the al/osteric tram it ion of aspartate transcarbamoylase? j. BioI. Chern. 263: 18583 18586. Neurath. H . 1989 . Proteolytic processin g and phy.iological reg ul ation.
'irends Biochem. Sci. 14:268- 27 1. Ilode. W.. and H uber, R . 1992. Natural protein proteinase inhibitors and th eir interaction with protein as ... RUT. j. Biochem. 204 : 433 451.
Aspartate Transcarbamoylase and Allosteric Interactions lVest, J. M .. T suruta. H .. and Kantsrowitz. E. K 2004. A fluorescent probe-labeled Escherichia coli aspartate transcarbamoylase that monitors the allosteric conform ation state. ]. BioI. Chern. 279: 94;- 951. Endrizzi, J. A .. Beemink. P. T. . Alber. T. . and Schachman. H . K . 2000. Binding of bislibstrate analog promotes large stru ctural c hanges in the unregu lated catalytic trimer of aspartate transcarbamoylase: Implications for allosteric regulation . Proc. Natl. Acad. Sci. U. S. A. 97:5077- 5082. Beernink. P. T. Endrizzi. j. A ., A lber. T , and Schachman. H . K. 1999. Assessment of the allo!:iteric mechanism of aspartate transcarbarnoylase based on the crys talline stru cture of the unregulated catalytic subunit. Proc. Natl. Acad. Sci. U. S. A. 96:5388- 5393. \V,les. M. E.. Madison. L. L., G laser , S. S .. and Wild . ]. R. 1999. Dive rge nt allosteric patterns verify the regu latory paradi gm for aspartate (ranscarbarnoylase. f. Mol. BioI. 294: 1387- 1400. LiCata. V. J .. /.lurz. D . S .. Moerke. N. J .• and A llewell. J . M . 1998. The magnitude of the allosteric conformat ionaJ transltion of aspartate transcarbamoylase is altered by mutations. Biochemistry 37: 17381 17385. EiS< c E
~'" ,"00 0::'"
Data Interpretation Problems
14. Distinguishing between models. T he followin g graph shows
~
.Q'O
inhibitor. Which si ngle amino acid substitution would you recummend ?
50 100
C u
'" ~
~
o
1
234
5
6
PALA/ ATCase (mole ratio) Effect of PALA on ATCase.
75
'"c -'= '"u
00
''""
-'"
00
6. R versus T An allosteric enzyme that follows the concerted mechani sm has aT I R ratio of 300 in the absence o f substrate. Suppose that a mutation reversed the ratio. How would this mu tation affect t he relation between the rate of the reaction and substrate concentrati on ?
. Zymogen activation. Wh en very low concen t rations of pepsinogen are added to acid ic m euia, how does the half-time for activation depend on zymogen concentration ?
9. A revealing assay. Suppose that yo u have just examined a young boy with a bleeding disord er high ly su ggestive of classic hemophil ia (factor VIII deficiency). Becau se of the late hour, the laboratory that carries out specialized coagulati on assays is closed. However, you happen to have a sample of blood from a classic hemophiliac whom yo u admitted to the hospital an hour earlier. W hat is the simpl est and m ost rapid test t hat you can perform to determine whether your present patient also is deficient in factor VIIl activity? 10. Counterpoint. The synthesis of factor X, like that of prothrombin, requires vitamin K. fac tor X also conta ins '/ carboxyglutamate residu es in its a mino -ter minal reg ion . However, activated factor X, in contrast with th ro mbin , retains this region of the m olecule. What is a likely functional con sequence of this difference between the two activated specie, )
II. A discerning inhibitor. Antithrombin J II form s an irre versible complex with thrombin but not with prothrombin. What is the most likely reason for this difference in reactivity?
12. Repeating heptads. Each of the three types of fibrin chain s contai ns repeating h eptapeptide units (abcdefg) in w hic h residues a and d are hydrophobic. Propose a reason for this regularity.
13. Drug design. A drug company has decided to use recombi nant DNA method s to prepare a m odified a i-antitrypsin that will be more resistant to oxidation than is the naturall y occurring
50
c
u
~
'"
"-
25
o ~~~~~--~~ 2
i. Regulation energetics. The phosphorylation and dephospho-
rylation of proteins is a vital m eans of regulation . Protein kinases attach phosphoryl groups, whereas onl y a phosphatase will remove the phosphoryl group from the target protein. What is the epergy cost of this means of covalent regul ation ?
y
fR
10 5
10- '
10- '
10-
Substrate concentration (M) [From M. W. Kirschner and H. K. Schachman, Biochemistry 12(1966):2997- 3004.] 15. Reporting live from AT Case 1. ATease underwent reaction with tetranitromethane to form a colored n itrotyrosine group (Amax = 4 30 nm ) in each of its catalytic ch ains. The absorption by thi s reporter group depends on its immediate environment . An essential lysine resiuue at each catal ytic site also was m odified to block the bindin g o f substrate. Catalytic trimers from this d oubly modified enzyme were then com b ined with nat ive trimers to form a hybrid en zyme. The absorption by the nitrotyrosine group was m easured on addition of the substrate analog succinate. What is the significance of the alteration in the absorbance at 43 0 nm ?
/~, Succina te
~ +5
c
-6'" o ~ c
-e'"o
-5
~
.0
..:
350
450
550
Wavelength (nm) [After H. K. Schachman,;. BioI. Chem. 263(1988): 18583-18586.] 16 . Reporting live from A T Case 2. A different ATCase hybrid was cons tructed to test t he effects of allosteric acti vators and inhibitors. Norm a l regulatory s ubunits were combined with
302
CHAPTER 10 Reg ulatory Strategies
nitrotyrosine -containing catalytic subunits. The addition of ATP in the absence of substrate increased th e absorbance at 430 nm, the same change elicited by the addition of succinate (see the graph in Problem 15). Conversely, C TI:' in the absence of substrate decreased the absorbance at 430 nm . What is the significance of the changes in absorption of the reporter groups?
Chapter Integration Problem
17. Density matters. T he sedimenation value of aspartate tran· scarbamoylase decreases when the enzyme switches to the R state. On the basis of the allosteric properties of the enzyme, explain why the sedimentation value decreases. Mechanism Problems
18. A spartate transcarbamoylase. Write the mechanism (in de· tail ) for the conversion of aspartate and carbamoyl phosphate into N -carbamoylaspartate. Include a role for the histidine residue present in the active site.
~
§?
0
~
ClJ 00 C
'"
-'" u
+5 0
19. Protein kinases. Write a mechanism (in detail) for the phosphorylation of a serine residue by ATP catalyzed by a protein kinase. What groups might you expect to find in the enzyme's active site?
ClJ
u
c
-
.c '"
-5
crp
0
V>
.c - 10 ~
350
450
550
Wavele ngth (nm) [After H. K. Schachman,). Bioi. Chem. 263 (1988): 18583- 18586.]
Chapter
11
Carbohyd rates ....
~
•
.
Carbohydrates in food are important sources of energy. Starch, found in plant-derived food such as pasta, consists of chains of linked glucose molecules. These chains are bro ken down into ind ividual gl ucose molecules fo r eventual use in the generation of ATP and as bUi lding blocks for other mo lecules. [(Left) Superstock.]
et us take an overview of carbohydrates, one of the four major classes of biomolecules along with proteins, nucleic acids, and lipids. Carbohydrates are aldehydes or ketones with multiple hydroxyl groups. They make up most of the organic matter on Earth because of their extensive roles in all forms of life. First, carbohydrates serve as energy stores, fuels, and metabolic intermediates. Second, ribose and deoxyribose sugars form part of the structural framework of RNA and DNA. Third, polysaccharides are structural elements in the ce ll walls of bacteria and plants. In fact, cellu lose, the main constituent of plant cell walls, is one of the most abundant organic compounds in the biosphere . Fourth, carbohydrates are linked to many proteins and lipids. Such linked carbohydrates play key roles in cell- cell communication and in interactions between cell s and other ele ments in the cellular environment. A key property of carbohydrates in th eir role as mediators of cellular interactions is the tremendous structural diversity possible within this class of molecules. Carbohydrates are built from monosaccharides, small molecules, typically containing from three to nine carbon atoms, that vary in size and in the stereochemical configuration at one or more carbon centers. These monosaccharides may be linked together to form a large variety of oligosaccharide structures. The sheer number of possible oligosaccharides makes this class of molecules information rich . This information , when attached to proteins, can augment the alread y immense diversity of proteins.
I O utl i n e I~-------1l.l Monosaccharides Are Aldehydes or Ketones with Multiple Hydroxyl Groups 11.2 Complex Carbohydrates Are Formed by the Linkage of Monosaccharides 11.3 Carbohydrates Can be Attached to Proteins to Form Glycoproteins 11.4 Lectins Are Specific CarbohydrateBinding Proteins
303
3 04 CHAPTER 11 Carbohydrates
Unraveling oligosaccharide structures and elucidating the effects of their attachm ent to proteins constitute a trem endous challenge in the field of proteomics. Indeed, this subfield has been given its own name, glycomics.
11.1
Monosaccharides Are Aldehydes or Ketones with Multiple Hydroxyl Groups
Monosaccharides, the simplest carbohydrates, are aldehydes or ketones thai have two or more hydroxyl gro ups; the empirical formula of many is (C H 20)n, literally a "carbon hydrate." Monosaccharides are important fucl molecules as well as building blocks for nucleic acids. The smallest monosac· charides, for which n = 3, are d ihydroxyacetone and D- and L-glyceraldehvde.
o
O~
\ -H HO, /
~C H
H, / ",C
"C H' \
HO ' \
HO
HO
/ H2
/ H2
o-Glyceraldehyde (an aldose)
Dihydroxyacetone
(a ketose)
L-Glyceraldehyde (an aldose)
They are referred to as triuses (tri - for 3), Dihydroxyacetone is a ketose because it contains a keto group, whereas glyceraldehyde is an aldose be· cause it contains an aldehyde group . G lyceraldehyde has a single asymmet· ric carbon atom and, thus, there are two stereoisomers of this sugar: D-glyceraldehyde and L-glyceraldehyde. These two form s are enantiomers, or mirror images of each other. As mentioned in Chapter 2, the prefixes D and L designate the absolute configuration . Monosaccharides and other su gars will often be represented in this book by Fischer projectiuns (Figure 11.1 ). Recall that, in a Fischer projection of a molecule, atoms joined to an asymmetric carbon atom by horizontal bonds are in front of the plane of the page, and those joined by vertical bonds are behind the page (see the Appendix in C hapter 1). Fischer projection s provide clear and simple views of the stereochemistry at each carbon center. Si mpl e monosaccharid es with four, fi ve, six, and seven carbon atoms are called tetroses, pentoses, hexoses, and heptoses, respectively. These molecules have multiple asymmetri c carbon atom s and , for these monosaccharides, the symbols D and L designate the ahsolute configuration of the asymmetric carbon
H·
CH O --C OH -CH 2 0H
0 ", / H
"'c
H
IC
OH
CH2 0H D-Glyceraldehyde
0 ", / H
"' c
HO
IC H ICH 0 H 2
L-Glyceraldehyde
/
H2 OH
O= C .......CH2 OH Dihydroxyacetone
Figure 11.1 Fischer projections of trioses. The top structure reveals t he stereochemical relations assumed f or Fischer projecti ons.
CHO
1
H-
2
b- ,OH
ICH 0 H
3
2
D-Glyceraldehyde
--------CHO
1 2
H-
IC-
3
H-
b
CHO OH OH
HO HI-
CH 2 0H
4
4
CHO
H
C
OH
HO-
-
H
C
OH
H
C
H
C
OH
H
C
CHO
CHO
H- C
OH
HO-
C
H
H
C
OH
H
C
H
C
OH
H
IC
H
C
OH
H
b
CH 20H . -Allose
C
OH
HO
C
H
OH
HO
C
H
HO
C
H
OH
H
C
OH
H
C
OH
CH2 0 H D-Xylose
CHO
o-Lyxose
CHO
CHO
CHO
C
OH
HO
C
H
H-
C
OH
HO-
C
H
OH
HO
C
H
HO
C
H
H
C
OH
H
C
OH
OH
H
C
OH
H
b
OH
HO
C
H
HO
C
OH
H
C
OH
H
C
OH
H
C
OH
H
C
CH2 0 H o-Glucose
CH20 H D-Mannose
CH 20H D-Gulose
ICH 0 H 2
H-
CH2 0H D- Altrose
CHO
H
a-Arabinose
D-Ribose
CHO
H
CH 2 0H
CH2 0 H
5
C- OH
D-Threose
CHO
2
H
CH 20H
D-Erythrose
1
C-
C
OH
Ho-
b-
H
HO
C
H
HO
C
H
H
HO
C
H
HO
C
H
OH
H
b
OH
H
CH 20H D-Idose
H-
CHO
ICH
2
0H
o-Galactose
IC
OH
CH20 H o-Talose
Figure 11.2 D-Aldoses containing three, four, five, and six carbon atoms. A D-al dose (ontains an aldehyde group (shown in blue) and has the abso lute configu ratio n of o-glyceraldehyde at the asy mmetric center (shown in red) farthest f ro m the aldehyde group. The numbers ind icat e t he st and ard designations for each carbon at om.
atomfarthest fro m the aldehyde or keto group . In F igure 11. 2, for example, Ihe four -carbon ald oses D-erythrose an d D- threose h ave the sam e configuration at C-3 (because they are D su gars) but opposite configurations at C 2. They are diastereoisomers, not enantiom ers, b ecause they are not mirror images of each other. Figure 11 .2 shows the common D-aldose sugars. D Ribose, the carbohydrate componen t of RNA, is a five -carbon aldose. DGlucose, D- mannose, and D-galactose are abund ant six -carbon aldoses . Note that D-glucose and D-mannose differ in confi guration only at C-2 . . differing in configurat ion at a single asymmetric center are called epimers. T hus, D-glu cose and D- mannose are epimeric at C-2; D-glucose and D-galactose are epimeri c at C -4 . Di hydroxyacetone is the simplest ketose. T he stereochemical relation s between f)- ketoses containing as many as six carbon atoms are shown in Figure 11. 3. Note t hat ketoses have one fewer asy mmetric center t han d o aldoses with the same number of carbon s. D-Fructose is the m ost abundant ketohexose. 305
306 CHAPTER 11
1
Carbohydrates
2
CH 20H
3
Dihydroxyacetone
1
2 3
H
C
OH
CH 20H
4
o-Erythrulose
1 2
3
H
C
OH
HO
C
H
4
H
C
OH
H
C
OH
5
CH 2 0 H
CH 2 0 H
D- Ribulose
o-Xylulose
1
2
Figure 11.3 D-Ketoses cont aining three- four, five, and six carbon atoms. Th e ket o group is shown in blue. Th e asymmetric center farthe st from the keto group, which determines the D designation, is shown in red.
H
C
OH
HO
C
H
OH
HO
C
H
HO
C
H
OH
H
C
OH
H
C
OH
3
H
C
OH
HO
C
H
4
H
C
OH
H
C
5
H
C
OH
H
C
6
CH 2 0H
CH 2 0H
D-Psicose
o-Fructose
CH 2 0H o-Sorbose
CH 20H
o-Tagatose
Pentoses and Hexoses Cyclize to Form Furanose and Pyranose Rings
o
Pyron
The predominant forms of ribose, glucose, fructose, and many other sugars in solution are not open chains , Rather, the open-chain form s of these sugars cyclize into rings because the ring forms are energetically more stable, The basis for ring fo rmation is the fact that an aldehyde can react with an alcohol to form a hemiacetal.
/ R
~
c ,-
H
Aldehyde
+
HOR ' Hemiacetal
Alcohol
Furan
For an aldohexose such as glucose, the C -l aldehyde in the open-chain form of glucose reacts with the C-S hydroxyl group to form an intramolecular hemiacetal, The resulting six-membered ring is called pyranose because of its similarity to pyran (Figure 11.4 ). Similarly, a ketone can react with an alcohol to form a hemiketal.
/ R
HO
~
C
'---R'
Ketone
+ HOR" Alcohol
,
OR"
\/ C
>
R/
"'-- R'
Hemiketal
For a ketohexose such as fructose, the C -2 keto group in the open -chain form of fructose reacts with a hydroxyl group within the same molecule to
307 11.1 Monosaccharides 0 "", " H I
C
H H 2C HO
OH
3C
o-D-Glucopyranose
H
H 4C H
OH
OH
Figure 11.4 Pyranose formation . The o pen-chain fo rm of glucose cyclizes when the C-5 hyd ro xyl gro up attacks the o xygen atom of t he C-1 aldehyde gro up to form an int ramo lecular hemiacetal. Two anome ric fo rms, designated U' and ~ , can result .
OH
sC
6 CH 2 0H D-elucose (open-chain form)
H
OH
!3-D-Glucopyranose
lorm an intramolecular hemiketal. The C -2 keto group can react with either the C-6 hydroxyl group to form a six-membered ring or the C oS hydroxyl group to form a five- m embered ring (Figure 11. 5). Thefive-membered ring is called af uranose b ecau se of its similarity to furan .
HO
o 3( -
H
H l
-
OH
H J -
OH
,
H
,
H > O"""
H
OH OH
o-Frudose (open-chain form)
H
u-D-Frudofuranose (a cyclic form of fructose)
Figure 11.5 Furanose formation . The open-cha in form of fru ct ose cyclizes to a fi ve-membered ring whe n t he C-5 hydroxyl group attacks t he C-2 keto ne t o form an intramo lecular hem ike tal. Two ano mers are possible, but only the a ano mer is sho wn.
The depictions of glucopyranose and fru ctofuranose shown in F igures 11.4 and 11, S are Haworth prujections. In such projections, the carbon atoms in the ring are not explicitly shown , The approx im ate pl ane of the ring is perpendicular to the plane of the paper, with the heavy line on the ring projecting toward the read er. L ike F ischer projection s, H aworth projection s ~low easy depiction of the stereochemi stry of su gar s, An additional asymmetri c center is created wh en a cyclic h emiacetal is formed . In glu cose, Co l, th e carbonyl carbon atom in the open- chain form , becomes an asymmetric center in the ring form . Thus, two ring structures can be formed: a -D-glucop yranose and i3- o -glucop yranose (see Figure 11.4). For D sugars drawn as Haworth projections, the designation a means
that the hydroxyl group attached tu C- l is on the opposite side of the ring from the CH20 H at the carbon atom tha t determines whether the sugar is designated Dor L (the chiral center); f3 means that the hydroxyl group is on the same side as the CH 20 H at the chiml center, The C - l carbon atom is called the anomeric carbon atom, and the a and i3 form s are called anomers. An equilibrium mixture of glucose is approximately one -third IX anom er , two- third s ~ anomer, and < 1% open -chain form . The same nom encl ature applies to the furanose ring form of fru ctose, except that a and i3 refer t o the hydroxyl groups attach ed to C- 2, the anom eric carbon atom (see F igure 11 .5) . F ructose forms b oth p yranose and furanose rings. The pyranose form predominates in fru ctose free in sol uti on , and t he furanose form pred omina tes in many fru ctose derivatives (Figure 11. 6).
l" H
o
OH : H>-"
H
H
OH
OH
~-D-Ribose
/ H /
o
OH
' H'"
H
H OH
H
~ - 2-0eoxY-D-ribose
3 08 CHAPTER 11
HOH 2C
HOH 2C CH20 H ~ O _:-:::-.... H HO/1
Carbo hydrates
H
o
~HO---
OH OH
CH20H
H
OH
H
H CH 20 H
H
OH
H HO
HO HO
OH OH
H
~-D- Fructofuranose
a -D- Fructofuranose
Figure 11.6 Ring structures of fructose. Fructose can form both five-membered furanose and six-membered pyra nose rings. In each case. both a and ~ ano mers are possible.
OH
HO
CH20H OH
H
a-o-Fructopyranose
H
p-D-Fructopyranose
Pentoses such as D-ribose and 2-deoxy - D-ribose form furanose rings, aswe have seen in the structure of these units in RNA and DNA.
Py ra nose and Furanose Rings Can Assume Different Conformations
Steric hi
The six-membered pyranose ring is not planar, because of the tetrahedral geom etry of its saturated carbon atoms. Instead, pyranose rings adopt two classes of conform ations, termed chair and boat because of the resemblance to these objects (Figure 11.7). In t h e chair form, th e substituents on the ring carbon atoms have two orientations: axial and equatorial. Axial bonds are n earl y perpendicular to the average plane of the rin g, whereas equatorial bonds are nearly parallel to this plane. Axial substit uents steri cally hinder each other if they emerge on the same side of the rin g (e.g., 1,3 -diaxial groups). In contrast, equ atorial substitu ents are less crowd ed . The chair form of f3 -D-glucopyranose predominates because all axial positions are occu· pied hy hydrogen atoms. The b ulkier OH and CH 20H groups emerge at the less- hi nd ered periphery. T he boat form of glucose is disfavored be· cause it is quite steri cally h indered . a a
a
e a_ a
e
a
H CH20H OH
OH H C-3-endo
H C-2-endo
Figure 11.8 Envelope conformations of ~-D-ribose. The C-3-endo and C-2- endo forms of ~ - D -ribose are shown. The color indicates the fou r atoms that lie approximately in a plane.
e
-0 e
e •
•
a
Figure 11.7 Chair and boat forms of J3-o-glucopyranose. The chair form is mo re stable o wing t o less steric hindrance because the ax ia l pOSi tions are occupied by hydrogen atoms. Abbreviations: a, axial: e. equatorial.
e __ e
e
e a
a
a
a
HO H ___ CH20 HO HO_
.
~ --i, -
H
-0
H
HH OH Chair form
Boat form
Furanose rings, like pyranose rings, are not p lanar. They can be puckered so that four atoms are nearly coplanar and the f ifth is about 0.5 Aaway from this plane (Figure 11.8). This conformation is called an envelope form because t he structure resembles an opened envelope with the back flap raised . In the ribose moiety of most biomolecules, either C-2 or C-3 is olltof the plane on the same side as C oS. These conformations are called C-2-endo and C-3 -endo, respectively.
Monosaccharides Are Joined to Alcohols and Amines Through Glycosidic Bonds
Q-Glycosid ic bond
CH 2 0H
Monosaccharid es may react with alcohols and amines to form modified monosaccharides. For example, D-glucose will react with methanol in an acid-catalyzed process : the anomeric carbon atom C- l , which is part of a hemiacetal (p . 306), reacts with the hydroxyl group of meth anol to form a sugar acetal, also called a glycoside. The reaction forms two glycosides : methyl {Y-D-glucopyranoside and methyll3-o-g1ucopyranosid e. These two glucopyranosides differ in the configuration at the anomeric carbon atom. The bond formed between the anomer ic carbon atom of a sugar and the hydroxyl oxygen atom of an alcohol is called a glycosidic bond specifically, anO-glycosidic bond. Alternatively, the anomeric carbon atom of a sugar can be linked to the nitrogen atom of an amine to form an N -glycosidic bond. We have already encou ntered such reaction products; nucleosides are adducts between sugars, such as ribose, and amines, su ch as adenine (p. 109). Some other important modified su gars are sh own in Figure 11.9. Compounds such as methyl glucopyranoside differ in reactivity from the parent monosaccharide. For example, unmodified glucose reacts with 2 oxidizing agents such as cupric ion (Cu I ) because the open-chain form has afree aldehyde group that is readily oxidized.
H
•
H OH
-
HO
OH
C
Cu 2 +
C
H
•
Cu'
\. /
H
C
OH
H
C
OH
H
C
OH
HO
C
H
H
C
OH
H
C
OH
,1--0 H
H Methyl ,,-D-glucopyranoside
H
OH
M ethyl ~- D -g lucopyranosi de N-Glycosi di c bo nd
CH 2 0H
,1--0 NR2
H H
H
CU20 )
H20, HO
CH 2 0H
CH 2 0H
Glycosides such as methyl glucopyranoside do not react, because they are not readily interconverted with a form that includes a free aldehyde group. Solutions of cupric ion (known as Fehling's solution) provide a simple test for sugars, such as glucose, that can exist as a free aldehyde or ketone . Sugars that react are call ed reducing sugars; those that do not are called nonreducing sugars. Reducing sugars can often n onspecifically bind to other molecules. For instance, as a reducing sugar, glucose can react with hemoglobin to form
H
H
CH 2 0H H
OH
HO
H
COO-
OH
H OH
H H
H
OH
~ -D-Acetylgaladosamine
(Fuc)
(GaINAc)
Il-o-Acetylglucosamine (GIeNAc)
C-
OH
H-
C-
OH
=
OH
H
OH
P-l-Fucose
R
H-
H
CH2 0H
Sialic acid (Sia) (N-Acetylneuraminate)
"lure 11.9 Modified monosaccharides. Carbohydrates can be modified by the addition of sOOstituents (shown in red) other than hydroxyl groups. Such modified carbohyd rates are expressed on cell surfaces.
309
3 10 CHAPTER 11
glycosylated hemoglobin. C hanges in the amount of glycosylated hemoglo· bin can be used to monitor the effectiveness of treatments for diabetes mellitus, a condition characterized by high levels of blood glucose (p. 7i3). Reaction with glucose has no effect on the oxygen-binding ability of hemoglobin.
Ca rbohyd rates
Phosphorylated Sugars Are Key Intermediates in Energy Generation and Biosyntheses One sugar modification deserves special note because of its prominence in metabolism. The addition of phosphoryl groups is a common modifica· tion of sugars. For instance, the first step in the breakdown of glucose to ob· tain energy is its conversion into glucose 6-phosphate. Several subsequenl intermediates in this metabolic pathway, such as dihydroxyacetone phos· phate and glyceraldehyde 3-phosphate, are phosphorylated sugars. CH 20PO,2 -
H
0
'----c"""
H
-of'
H
OH
OH
Glucose 6-phosphate (G-6P)
H
C
OH
CH 20PO,2
phosphate
Glyceraldehyde 3-phosphate
(DHAP)
(GAP)
Dihydroxyacetone
Phosphorylation makes sugars anionic; the negative charge prevents these sugars from spontaneously leavi ng the cell by crossing lipid -bilayer membranes. Phosphorylation also creates reactive intermediates that will more readily form linkages to other molecules. For example, a multiply phosphorylated derivative of ribose plays key roles in the biosyntheses of purine and pyrimidine nucleotides (p . 71 2).
11.2
(J- l ,4-Glycosi dic bond
CH 2 0H
Complex Carbohydrates Are Formed by the Linkage of Monosaccharides
A--O H HO
OH I
I
H
H
4
I
OH
H
OH
Figure 11.10 Maltose, a disaccharide. Tw o molecules o f glucose are linked by an a-1,4-glycosid ic bond t o fo rm the d isaccharid e mal tose. The angles in the bonds t o th e central oxygen do not deno t e c arbon ato m s. The angles are added o nly for ea se o f illustrati o n.
G lycosidic bonds can join one monosaccharide to another. Oligosaccharide; are carbohydrates built by the linkage of two or more monosaccharides by O-glycosidic bonds (Figure 11.10). In maltose, for example, two D-glucose residues are joined by a glycosidic linkage between the C-1 carbon atom on one sugar and the hydroxyl oxygen atom on C-4 of the adjacent sugar. The sugar on the C-1 side of the link is in the 0. configuration. In other words, th e bond emerging from C-l lies below the plane of the ring when viewed in the standard orientation. Hence, the maltose linkage is called an a-l ,4· glycosidic bond. Because monosaccharides have multiple hydroxyl groups, various glycosidic linkages are possible. Indeed, the wide array of these link· ages in concert with the wide variety of monosaccharides and their many iso· meric forms makes complex carbohydrates structurally diverse molecules.
Sucrose, Lactose, and Maltose Are the Common Disaccharides A disaccharide consists of two su gars joined by an O -glycosidic bond. Three abundant disaccharides are sucrose, lactose, and maltose (Figure 11.1 1). Sucrose (common table sugar), a transport form of carbohydrates in plants, is obtained commercially from cane or beet . The anomeric carbon atomsof a glucose unit and a fructose unit are joined in this disaccharide; the config· uration of this glycosidic linkage is ex for glucose and J3 for fru ctose.
H
H
OH
HO H
OH
Lactose
Maltose
(p-D-Galactopyranosyl-( I -> 4 )-a-D-glucopyranose
(a-D-Glucopyranosyl-(I -> 4)-a-D-glucopyranose
Sucrose
la.D·Glucopyranosyl-(1 -> 2)-P-D-fructofuranose
H
OH
Consequently, sucrose is not a reducing sugar, because neither component monosaccharide is readily converted into an aldehyde or ketone, in contrast with most other sugars. Sucrose can be cleaved into its component monosac charides by the enzyme sucrase. Lactose, the disaccharide of milk, consists of galactose joined to glucose
Figure 11.11 Common disaccharides. Sucrose, lactose, and maltos e are commo n dietary components. The angles in the bonds to the central oxygens do not denote carbon atoms.
by a ~ - 1 , 4-g l ycosid ic linkage. Lactose is hydrolyzed to these monosaccha rides by lactase in human beings (p. 451 ) and by {3 -galactosidase in bacteria. In maltose, two glucose units are joined by an a-1,4-glycosidic linkage, as stated earlier. Maltose is produced by the hydrolysis of starch and is in turn hydrolyzed to glucose by maltase. Sucrase, lactase, and maltase are located on the Quter surfaces of epithelial cells lining the small intestine (Figure 11.12).
, Figure 11.12 Electron micrograph of a microvillus. Lactase and other enzymes that hydrolyze carbohydrates are present o n microvi lli that proj ect from the o uter face of theplasma membrane of intestinal epithelial ce lls. [From M. S. Mooseker and L. G. Titney, J Cell. Bioi. 67(1975}:725- 743.]
Glycogen and Starch Are Mobilizable Stores of Glucose Large polymeric oligosaccharides, formed by the linkage of multiple mono saccharides, are call ed polysaccharides. Polysaccharides play vital roles in energy storage and in maintaining the structural integrity of an organism. If all of the monosaccharides are the same, these polymers are called homopolymers. The most common homopolymer in animal cells is glycogen, the storageform of glucose. As will be considered in detail in Chapter 2 1, glycogen is a very large, branched polymer of glucose residues. Most of the glucose units in glycogen are linked b y a-1 ,4-glycosidic bonds. Branches are formed by ((·1 ,6-glycosidic bonds, present about once in 1 0 units (Figure 11.13).
H 0.
Y
a · ! ,6-Glycosidic bond
0, OH
6
CH,
0OH
H
OH
Figu re 11.13 Branch point in glycogen. Two chai ns of glucose mo lecules j oi ned by ",-1, 4-glycosidic bonds are linked by an (Y-1,6-glycosidi c bond to create a branch point. Such an ",-1 ,6glycosidic bond forms at approximately every 10 glucose units, mak ing glycogen a highly branched mo lecule.
311
312 CHAPTER 11 Carbohydrates
The nutritional reservoir of carbohydrates in plants is starch, of which there are two forms . Amylose, the unbranched type of starch, consists of glucose residues in a-1,4 linkage. Amylopectin, the branched form , has about one a-l ,6 linkage per 30 a-l, 4 linkages, and so it is like glycogen except fo r its lower degree of branching. More than half the carbohydrate ingested by human beings is starch . Roth amylopectin and amylose are rapidly hydrolyzed by a -amylase, an enzyme secreted by the salivary glands and the pancreas.
Cellulose, the Major Structural Polymer of Plants, Consists of Linear Chains of Glucose Units
COO -
H
Calacturonic acid
Ce llulose, the other major polysaccharide of glucose found in plants, serves a structural rath er than a nutriti onal role . Cellu lose is one of the most abun15 dant organic compounds in the biosphere. Some 10 kg of cellulose is syn· thesized and degraded on Earth each year. It is an unbranched polymer of glucose residues joined by [3 - 1,4 linkages . The [3 configuration allows eel· lulose to form very long, straight chains. Fibrils are formed by parallel chains that interact with one another through hydrogen bonds . The a- l,4 linkages in glycogen and starch produce a very different molec ul ar architecture from that of cellulose. A hollow helix is formed instead of a straight chain (Figure 11 .14). These differin g conseq uences of the a and [3 linkages are biologicall y important. The straight chain formed by [3 linkages is op· timal for th e constructi on of fibers havin g a hi gh tensil e strength . In contrast, the open helix formed by a linkages is well suited to forming an accessible store of sugar . Although mammals lack cellulases and therefore cannot digest wood and vegetable fibers, cellulose and other plant fibers are still an important constituent of our diet as a component of dietary fiber. Dietary fiber pro· d uces a feeling of satiety. Soluble fiber such as pectin (polygalacturonic acid) slows the movement of food through the gastrointestinal tract, allowing better digestion and absorption of nutrients. Insoluble fibers, such as cellulose, increase the rate at wh ich di gestion products pass through the large intes· tine. This increase in rate may minimi ze exposure to toxins in our di et.
Cellulose (P-I,4 linkages)
Figure 11.14 Glycosidic bonds determine polysaccharide structure. The 13-1.4 linkages favor st raight chains. which are o ptimal fo r struct ural purposes. The 0
GleNAe
Oligosaccharides Can Be "Sequenced" How is it possible to determin e the structure of a glycoprotein the oligosaccharide structures and their points of attachment? Most approaches m ake use of enzymes that cleave oligosacch arides at specific types of linkages. The first step is to detach the oligosacch.aride from th e protein . For example, N- linked oli gosaccharides can be released from proteins by an enzyme such as peptide N -glycosidase F, which cleaves the N -glycosidic bonds linking the oligosaccharide to the protein . T he oligosaccharides can then be isolated and analyzed . MALDI -T OF or other mass spectrometric techniques (Section 3.5) provide the m ass of an oligosaccharide fragm en t. However, many possible oligosaccharide structures are con sisten t with a given mass. More-complete info rmation can be obtained b y cleaving the oligosaccharide with enzym es of varying specificities. For example, f3 -1 ,4galactosidase cleaves f3- glycosidic bonds exclusively at galactose residues. The products can again be analyzed by mass spectrometry (Figure 11. 25). The repetition of this process with the use of an array of enzym es of differ ent specificity will eventually reveal the structure of the oligosacch aride. (A)
I
....
•uc
~
•
'"a ~
• ,.c • > .-• -• •
N
'0
C
eX> •
"''" "'"' ~
~
'"
(8)
I ~
c
•
,.c '0 C
..,.
'"
•
•
~
.•~ -••
~ ~
~
'" f.--~------- '-'"" ~~ -~- ~--~-----1000
1200
1400
1600
1800
2000
OH HO
OR
Mannose 6-phosphate residue
Figure 11.24 Formation of a man nose 6-phosphate marker. A glycopro tein destined for delivery to Iysosomes acqu ires a phosp hate marker in t he Golgi co mpartment in a two-step process. First, a phosphotransferase adds a phosphoN -acetylglucosamine unit to the 6-0H gro up of a man nose, and then a phosphodiesterase remo ves the added sugar t o generate a mannose 6-phosphate residue in the core o ligosaccha ri de.
Figure 11.25 Mass spectrometric "sequenc ing" of oligosaccharides. Carbohydrate-cleaving enzymes were used t o release and specifical ly cleave the o ligo sacchari de component of the glycoprotein fetu in fro m bo vine serum. Parts A and B show the masses obtained with MALDI-TO F spectrometry as well as t he correspo nd ing structures of the o ligosaccharide-digest ion products (using t he sa me scheme as that in Figure 11.18): (A) digestion wi t h pept ide N -glycosidase F (to re lease the oligosaccharide from t he pro tein) and neuraminidase: (6) digestion with peptide N -glycosi dase F, neuram inidase, and i3-1A-ga lactosidase. Kno w ledge of the enzyme specifi cit ies and the masses of the product s permits t he characteri zati o n of t he o ligosacchari de. See page 315 for t he carbo hydrate key. [After A. Varki, R. Cummings, J. Esko, H. Freeze, G. Hart, and J. M arth (Ed s.). Essentials of G/ycobiology (Cold Spring Harbo r Labo ratory Press, 1999), p. 596.]
319
320 CHAPTER 11 Carbohydrates
Proteases applied to glycoproteins can reveal the points of oligosaccha· ride attachment. Cleavage by a specific protease yields a characteristic pat· tern of peptide fragments that can be analyzed chromatographically. Fragments attached to oligosaccharides can be picked out because their chromatographic properties will change on glycosidase treatment. Mass spectrometric analysis or direct peptide sequencing can reveal the identity of the peptide in question and, with additional effort, the exact site of oligosaccharide attachment. Glycosylation greatly increases the complexity of the proteome. A given protein with several potential glycosylation sites can have many different glycosylated forms (sometimes called glycoforms), each of which may be generated only in a specific cell type or developmental stage. Now that the sequencing of the human genome is complete, the characterization of the much more complex proteome, including the biological roles of specifically modified proteins, can begin in earnest.
11.4
Lectins Are Specific Carbohydrate-Binding Proteins
The diversity and complexity of the carbohydrate units of glycoproteins suggest that they are functionally important. Nature does not construct complex patterns when simple ones suffice. Cellulose and starch, for exam· pie, are built solely from glucose units. In contrast, glycoproteins contain multiple types of residues joined by different kinds of glycosidic linkages. An enormous number of patterns in the composition and structure of surface sugars are possible because (1) different monosaccharides can be joined to one another through any of several OH groups, (2) the C-1 linkage can have either an a or a Jj configuration, and (3) extensive branching is possible. Indeed, many more different oligusaccharides can be formed from four sugaTl than can oligopep tides from four aminu acids. Why all this intricacy and diversity? It is becoming evident that carbo· hydrates are information-rich molecules that guide many biological processes. The diverse carbohydrate structures displayed on cell surfaces are well suited to serve as sites of interaction between cells and their envi· ronments. Proteins termed lectins (from the Latin legere, "to select") are the partners that bind specific carbohydrate structures on opposing cell surfaces. Lectins are ubiquitous: they are found in animals, plants, and • • m ICroorgamsms. Lectins Promote Interactions Between Cells
The ch ief function of lectins in animals is to facilitate cell- cell contact. A lectin usually contains two or more binding sites for carbohydrate units. The binding sites oflectins on the surface of one cell interact with arrays of carbohydrates displayed on the surface of another cell. Lectins and carbo· hydrates are linked by a number of weak interactions that ensure specificity yet permit unlinking as needed. The interactions between one cell surface and another resemble the action of Velcro; each interaction is weak but the • composite IS strong. Lectins can be divided into classes on the basis of their amino acid sequences and biochemical properties. One large class is the C type (for calcium-requiring) found in animals. These proteins each have a homologous domain of 120 amino acids that is responsible for carbohydrate binding. The structure of one such domain bound to a carbohydrate target is shown in Figure 11.26. •
~
Figure 11.26 Structure of a C-type carbohydrate-binding domain of an animal lectin. Notice that a calcium ion links a mannose res idue to the lectin. Selected intera ctions are shown, with some hydrogen atoms omitted for clarity.
Glu
•• •
Mannose
" calcium ion on the protein acts as a bridge between the protein and the sugar through direct interactions with sugar OH groups. In addition, two glutamate residues in the protein bind to both the calcium ion and the sugar, and other protein side chains form hydrogen bonds with other OH groups on Ihe carbohydrate. The carbohydrate-binding specificity of a particular lectin is determin ed by the amino acid residues that bind the carbohydrate . ~
Protein s termed selectins are m embers of the C- type family. l(;S Selectins bind immune-system cel ls to sites of injury in the inflammatory response (Figure 11 .27) . The L, E, and P forms of selectins bind specifically to carbohydrates on lym p h -node vessels, endothelium , or activated blood platelets, respectivel y. New therapeutic agents that control inflammation may em erge from a deep er understanding of how selecti ns bind and distinguish different carbohydrates. L-Selectin, originally thought to participate only in the immune response, is produced by embryos when Ihey are ready to attach to the endometrium of the mother's uterus. For a short period of time, the endometrial cells present an oligosaccharide on the cell surface. When the embryo attach es throu gh lecti ns, the attachment activates signal pathways in the endometrium to make implantation of the embryo possible. Plants also are rich in lectins. Although the exact role of lectins in plants is unclear, they can serve as potent insecticides. T he binding specificities of lectins fro m plants have been well characterized (Figure 11 .28) . Bacteria, too, contain lectins . Escherichia coli bacteria are able to adhere to the ep ithelial cells of the gastrointestinal tract because lectins on the E. coli surface recognize oligosaccharide units on the surfaces of target cells. These lectins are located on slender hairlike appendages called fimbriae (Pili). GlcNAc 1~- I.4
GlcNAc
I ~- 1 ,4
GlcNAc
Binds to wheat-germ agglutinin
Gal
1 ~- 1 ,3
GalNAc Binds to peanut lectin
Gal
Gal
GlcNAc
GlcNAc
I ~- 1.4
~-1.4 1
0- 16'-
....'
Man
/oL 12 I-'
•
Binds to phytohemagglutinin
Figure 11.27 Selectins mediate cell - cell interactions. The scanning electron micrograph sho ws lymphocytes adhering to the endothelial lining o f a lymph node. The L selectins on the lymphocyte surface bind spec ifically t o carbohydrates o n th e lining of the lymph-node vessels. (Co urtesy of Dr. Eugene Butcher.]
Figure 11.28 Binding selecti vities of plant lectins. The plant lectins wh eatgerm aggl utin in, peanut lectin, and phytohemagglutinin recognize different o ligosaccharides.
Influenza Virus Binds to Sialic Acid Residues
W Some viruses gain entry into specific host cells by adhering to cell surface carbohydrates. For example, influenza virus recognizes sialic acid residues present on cell -surface glycoproteins . The viral protein that binds to these sugars is called hemagglutinin (Figure 11 .29). l(;S
321
322 CHAPTER 11 Carbohydrates
Hemagglutinin Lipid bilayer Neuraminidase
Figure 11.29 Viral receptors. Influenza virus targets cell s by binding t o sialic ac id residues (purple d iamonds) located at the t ermini of oligosacchari des present on cell-surface glycoproteins and glycolipids. These carbohydrates are bound by hemagglutinin (interaction Circles), o ne of the major proteins expressed on the surface o f the vi rus. The other major vi ral surface prot ein, neuraminidase, is an enzyme that cleaves o ligosaccharide chains t o release t he viral particle at a later stage of the viral life cycle.
Host cell membrane
After the virus penetrates the cell membrane, another viral protein, neur· aminidase (sialidase), cleaves the glycosidic bonds to the sialic acid residues, freeing the virus to infect the cell. Inhibitors of this enzyme such as oseltamivir _ (Tamiflu) and zanamivir (Relenza) are important anti -influenza agents.
Summary 11.1 Monosaccharides Are Aldehydes or Ketones with Multiple • Hydroxyl Groups An aldose is a carbohydrate with an aldehyde group (as in glyceralde· hyde and glucose), whereas a ketose contains a keto group (as in dihy. droxyaceton e and fructose). A sugar belongs to the D series if the absolute configuration of its asymmetric carbon atom farthest from the aldehyde or keto group is the same as that of D-glyceraldehyde. Most naturally occurring sugars belong to the D series. The C- l aldehyde in the open-chain form of glucose reacts with the C -S hydroxyl group to form a six-membered pyranose ring. The C-2 keto group in the open· chain form of fructose reacts with the C -S hydroxyl group to form a five -membered furanose ring. Pentoses such as ribose and deoxyribose also form furanose rings. An additional asymm etric center is formed at the anomeric carbon atom (C- l in aldoses and C -2 in ketoses) in these cyclizations. The hydroxyl group attached to the anomeric carbon atom is on the opposite side of the ring from the CH 2 0 H group attached to the chiral center in the u anom er, whereas it is on the same side of the ring as the CH 2 0 H group in the 13 anomer. Not all the atoms in the rings lie in the same plane. Rather, pyranose rings usually adopt the chair can· formation, and furanose rings usually adopt the envelope conformation. Sugars are joined to alcohols and amines by glycosidic bonds from the anomeric carbon atom. For example, N -glycosidic bonds link sugars to purines and pyrimidines in nucleotides, RNA, and DNA.
11.2 Complex Carbohydrates Are Formed by the Linkage of Monosaccharides Sugars are linked to one another in disaccharides and polysaccharides by O-glycosidic bonds. Sucrose, lactose, and maltose are the common disaccharides. Sucrose (common table sugar) consists of a-glucose and 13fructose joined by a glycosidic linkage between their anomeric carbon atoms. Lactose (in milk) consists of galactose joined to glucose by a 13-1,4 linkage. Maltose (in starch) consists of two glucoses joined by an 0'-1,4 linkage. Starch is a polymeric form of glucose in plants, and glycogen serves a similar role in animals. Most of the glucose units in starch and glycogen are in 0'-1,4 linkage. Glycogen has more branch points formed by C1-1,6 linkages than does starch, and so glycogen is more soluble. Cellulose, the major structural polymer of plant cell walls, consists of glucose units joined by 13-1,4 linkages. These 13 linkages give rise to long straight chains that form fibrils with high tensile strength. In contrast, the CI linkages in starch and glycogen lead to open helices, in keeping with their roles as mobilizable energy stores. Cell surfaces and the extracellular matrices of animals contain polymers of repeating disaccharides called glycosarninoglycans. One of the units in each repeat is a derivative of glucosarnine or galactosamine. These highly anionic carbohydrates have a high density of carboxy1ate or sulfate groups. Proteins bearing covalently linked glycosaminoglycans are proteoglycans.
323 Key Terms
III Carbohydrates Can Attach to Proteins to Form Glycoproteins Specific enzymes link the oligosaccharide units on proteins either to the side-chain oxygen atom of a serine or threonine residue or to the side-chain amide nitrogen atom of an asparagine residue. Protein glycosylation takes place in the lumen of the endoplasmic reticulum. The N- linked oligosaccharides are synthesized on dolichol phosphate and subsequently transferred to the protein acceptor. Additional sugars are attached in the Golgi complex to form diverse patterns.
11.4 Lectins Are Specific Carbohydrate-Binding Proteins Carbohydrates on cell surfaces are recognized by proteins called lectins. In animals, the interplay of lectins and their sugar targets guides cell-cell contact. The viral protein hemagglutinin on the surface of the influenza virus recognizes sialic acid residues on the surfaces of cells invaded by the virus. A small number of carbohydrate residues can be joined in many different ways to form highly diverse patterns that can be distinguished by the lectin domains of protein receptors.
Key Terms monosaccharide (p. 304) triose (p. 304) ketose (p. 304) aldose (p. 304) enantiomer (p. 304) tetrose (p. 304) pentose (p. 304) hexose (p. 304) heptose (p. 304) diastereoisomer (p. 305 ) epimer (p. 305) hemiacetal (p. 306)
pyranose (p. 306) hemiketal (p . 306) furanose (p . 307) anomer (p. 307) glycosidic bond (p. 309) reducing sugar (p. 309) nonreducing sugar (p. 309) oligosaccharide (p. 310) disaccharide (p. 310) polysaccharide (p. 311) glycogen (p. 311 ) starch (p . 311)
cellulose (p. 312) proteoglycan (p. 3 12) glycosaminoglycan (p. 312) glycosyltransferase (p. 314) glycoprotein (p. 316) endoplasmic reticulum (p. 317) Goigi complex (p. 317) dolichol phosphate (p. 317) lysosome (p. 3 18) glycoform (p. 320) lectin (p. 320) selectin (p. 321)
324
CHAPTER 11 Carbohydrates
Se lected Readings Where to Start Sharon, N ., and Lis, H . 1993. Carbohydrates in cell recogni tion . Sci. Am. 268( 1):82 89. Lasky, L. A. 1992 . Selectins: Interpreters uf cell -specific carbo hyd rate information during infl ammati on . Science 258:964- 969. Weiss, P., and Ashwell , G. 1989. The asialogl ycuprotein recepto r: Prope rti es and modul atio n by liga nd . Prog. Clirl. BioI. I<es. 300: 169- 184. Paulson , j. C. 1989. G lycoprotein s: What are the sugar side chains fo r? Trends Biochem. Sci . 14:272- 276. Woods, R. j. 1995. T hree-d im ensional structures of oligosacc harides . CUrT. Opin. StTUct . BioI. 5:59 1- 59R. Books Va rki , A., Cu mmings, R., Esko, J.. Freeze, H. , Hart, G ., and M arth , j. 2002. Essentials oj Glycobiulugy. Cold 5pring Harbor Laboratory Press .
Fu kuda, M ., and Hindsga ul, O. 2000. Molecu lar Glycnhiology. IRL Press at Oxford Uni versity Press. EI Kh adem , H . S. 1988 . Carbohydrate Chemistry. Academic Press. G insburg, V. , and Robbin s, P. W . (Eds.). 198 1. Biology oj Carbohydrates (vob . 1- 3). Wil ey. Fukuda , M . (Ed .). 1992. Ce ll S urJace Carbohydrates a1ld Ce ll Develupment. C RC Press. Pre iss, j. (Ed .). 1988. The Biochemistry oj Plants: A Comprehensive Treatise : Carbohydrates. Academic Press.
Carbohydrate-Binding Proteins and Glycoproteins Yan, A ., and Lennarz, W. J. 2005. Unraveli ng the mechani sm of protein N -glycosylat ion . j. /Jiol. Chern. 280 :31 21- 3124. Qasba, P. K ., Ramakrishnan , 13., and Hoeggem an , E. 2005 . Substrateinduced conformatioi nal changes in glycosyltransferases. Trends Biochem. Sci. 30 :53- 62. Pratta, M . A ., Yao, W ., Decicco, c., Torto rella, M ., Liu, R.-Q., Copeland, R. A ., Magolda, R., Newton , R . C. , Trzaskos, J. M ., and Arner, E. r.. 2003 . Aggrecan protects cartil age coll agen [rom proteolyti c cleavage. }. BioI. Chern . 278 :45539 45545. Fisher, J . W . 2003 . Erythropoietin : Physiology and pharm aco logy update. Ex p. Bioi. Med. 22R: l - 14. Unli gil , U., and Rini, J. M . 2000. G lycosyltransferase structure and mechanism . Curro Opin. StTUct. Bioi. 10:5 10- 517. C heetham , J. C .. 5mi th, O. M ., Aoki, K . H ., Stevenson , J . L., H oeffel, T. J .. Syed, R. S., Eg rie, J., and Harvey, T. S. 199R . N MR stru cture of hu man erythropoietin and a comparison with its receptor bound co nfor mation. Nat. S tTUct. BioI. 5:861 - 866. Bouckaert, J., Hamelryck, T ., Wy n., L., and Loris, R. 1999 . Novel structures of plant lectins and their complexes with carhohydrates. CurroOpin. S truct. BioI. 9:572 577. Weis, W . I. , and O ri ckamer, K . 1996. Structural basis of lectin carbohyd rate recognition . A nna. Rev. Biochem. 65:44 1- 473.
Vyas , N. K. 199 1. Ato mic features of protein-carbohyd rate interac· tions. CUrT. Opin. StTUct. BioI. 1 :732- 740. Weis, W . I. , O rickamer, K., and Hend rickson, W . A . 1992. Structurc of a C -type mannose- binding protein complexed with an oligosaccha. ride. Nature 360: 127- 134 . Wright, C. 5. 1992. C rys tal structure of a wheat germ agglutinin / gly. cophorin -sialoglycopeptide receptor complex: Structural basis for cooperati ve lectin -cell binding. }. Bioi. Chern. 267: 1434 5- 14J52. Shaanan , B., Lis, H ., and Sharon , N . 1991. Structure of a legume lectin wi th an o rdered N -lin ked car bohydrate in complex with lactose. Science 254:862- 866 .
Glycoproteins lIern field , M ., Gotte, M ., Park , P. W ., Reizes. 0., F itzgerald, M. L., Lincecum, J., and Zako, M . 1999. Functions of cell surface hcp,ran sulfate proteoglycans. A nnu. Rev. Biochem. 68 :729- 777 . [ozzo, R. V. 1998 . M atri x proteoglyeans: Fro m molec ul ar design 10 cellular function . A nnu. ReTJ. Biochem. 67: 609- 652. Yanagis hita, M ., and !-Iascall, V. C. 1992 . Cell surface heparan sui fat' peateoglycan s. }. /3iol. Chern. 267:945 1- 9454. [ozzo, R. V. 1999. The biology of small leucine- ri ch proteoglycans. Functio nal network of intecacli ve proteins . .f. Bioi. Chern. 2i4: 1 RR4 J- l XH46. Carbohydrates in Recognition Processes Weis, W . I. 1997 . Cell -surface carbohydrate recognition by animal and virall ectins. Curro Opin. Strnc!. BioI. 7:624- 630. Sharon, N., and Lis, H . 19X9. Lectins as cell recognition molecules. Science 246:227- 234. Turner, M . L. 1992. Cell ad hesio n molecules: A unifying ap proach 10 topograp hic biology. Bioi. Reu Camb. Philos. Soc. 67 :3 59 377. Feizi, T. 1992. Blood group-related oligosaccharides are ligands in cell· adhesion events. Biochem. Soc. Trans. 20 :274- 278 . Jessell, T. M ., H ynes, M . A ., and Oodd, J . 1990 . C arbohydrates and carbo hydrate-binding proteins in the nervous system . Ann". Rev. Neurosci. 13:227 255. C lothia, c. , and Jones, E. V. 1997. T he molecular structure of cell adhe· sio n molecules. Annu. Rev. /3iochem. 66:823- 862. Carbohydrate Sequencing Venkataraman, G ., Shriver, Z .. Raman, R., and Sasisekharan, R. 1999. Sequencing complex polysaccharides. Science 286:537- 542. Zhao, Y, Kent, S. B. H ., and C hait, B. T. 1997. Rapid, sensitivestruc· ture analysis of oligosaccharides. Proc. Natl. Acad. Sci. U.S.A. 94: 1629- 1633. Rudd, P. M ., G uile, G. R., Ku ster, B., Harvey, D . J.. Opdenakker, G., and O wek, R. A. 1997. O ligosaccharide sequ encing technology. Na ture 388:205- 207.
Problems 1. Word origin. Acco unt for the o ri g in o f th e term carbohydrate.
2. Diversity. H o w m an y d iffe rent o ligosaccharides can be m ad e
(a ) (b ) (c ) (d ) (e)
D-glycera lde hyde a nd dih yc.lroxyacetone I)-glucose anc.l rl- mannose rl-g lucose and D- fruc tose a-rl-glucose and I3 - D- glucose D- ribose and D- ribul ose D-galac tose and D- glucose
b y linking o n e g lucose, o n e mannose, and o n e galactose? A ssume th a t each s ugar is in its p y ran ose fo rm . Compare this numbe r with the numbe r of tripe ptides that can b e m ad e from three differe nt a mino ac ids .
(f)
3. Couples. Indicate whe th e r eac h of the fo llo wing pairs o f s u g -
4. Mutarotation. The s p ec if ic ro tatio n s of th e at a n d
a rs con s ists o f a no m er s , e pime rs , or an aldose ketose p a ir :
of D- gluco se a r e
13 anomers
+ 11 2 d egrees a nd + 18.7 d egrees, resp ectively.
Pro blems 32 5 Specific rotation, [a ln, is defined as the observed rotation of light of wavelength 5S9 nm (the 0 line of a sud ium lamp) pass ~ ing through 10 em of a 1 g ml - 1 solu tion of a sample. When a crystalline sample of a~ I) ~ glucopyrano se is dissolved in water, the specific rotation d ecreases from 112 d egrees to an ellu ilib ~ rium value of 52.7 degrees. On the bas is of this result, what are the proportions of the u a nd f3 anomers at equ ilibri u m? Assume that the concentration of the open~c hain form is n eg~ ligible.
(c) f3 ~ Cal actosidase is an enzyme that will remove galactose residues from an oligosacch aride. What are t he products of f3 ~ galactosidase treatment of raffinose? HOH 2 C
HO~"../ HO'----'
HO
OH CH 2 0H
O,-Y
5. Telltale marher. Glucose reacts slowly with hemoglobin and other protei ns to form covalent compounds. Why is glucose active? What is the nature of the adduct formed ?
re~
6. Periodate cleavage. Compounds containing hydroxyl groups on adjacent carbon atoms undergo carbon arbon bond cleavage when treated with periodate ion (10 4 - ), F-fow can this reaction be used to distinguish between pyranosides and furanosides? i. Oxygen source. Does the oxygen atom attached to C~ 1 in methyl (>.Q.glucopyranoside come from glucose or methanol ?
Raffinose
12. Anomeric differences, a ~ [)~ Man nose is a sweet ~ tasting sugar. f3 ~u ~ Mannose .
on the other hand, tastes bitter. A pure solution of a ~ l) ~ mannose loses its sweet taste with time as it is converted into the f3 anomer. Draw the f3 anomer and explain how it is formed from the a anomer.
8, Sugar lineup. Identify the following four sugars, OH HO HOH 2 C OH
HOH2C H
(a) HO
HO
OH
,
H H
(b)
H
u-D-Mannose
H HO
H H
HO OH
.
H H
OHH
OH
(e) HOH 2C
H
(d) HO HO-
H
13. A taste of honey. Fructose in its f3 ~ U ~ pyran ose form accounts for the powerful sweetness of honey, The f3 ~ D ~ furanose form, a l ~ though sweet, is not as sweet as the pyranose form , The furanose form is the more stable form . Draw the two forms and explain why it may not always be wise to cook with honey.
-.......: ,---OH
14 . Making ends meet. (a) Compare the number of reducing
NH~
ends to nonreducing ends in a molecule of glycogen. (b ) As we will see in C hapter 21, glycogen is an important fuel ~ storage form that is rapid ly mobi lized. At which end the red ucing or nonreducing would you expect most metabolism to take place?
9, Cellular glue. A trisaccharide unit of a cell ~ surface
glycopro~
lein is postulated to playa critical role in mediating cell cell ad~ h",ion in a particular tissue, Design a simpl e experiment to test this hypothesis,
10, Mapping the molecu le. Each of the hyd roxyl groups of glu ~ cose can be me thy lated with reagents such as dimethylsulfate under basic conditions , Explain how exhaustive methylation fol ~ lowed by the comple te digestion of a known amount of glycogen would enable you to determine the number of branch points and reducing ends, II. Component parts. Raffinose is a trisaccharide and a minor constituent in sugar beets , (a) Is raffinose a reducing sugar? Explain . (b) What are the monosaccharides that compose raff inose?
15, Carbohydrates and proteomics, Suppose that a protein contains six potential N~ linked glycosylation sites. How many possib le proteins can be generated, depending on which of these sites is actually glycosylated ? Do not include the effects of div er~ sity within the carbohydrate added, Chapter Integration Problem
16, S tereospecificity. Sucrose, a major product of photosynthesis in green leaves, is synthesized by a battery of enzymes. The sub ~ strates for sucrose synthesis, t)~glucose and I) ~ fructose, are a mix~ lure of a and i3 anomers as well as acyclic compounds in solution , Nonetheless, sucrose consists of a ~ t) ~ glucose linked by its carbon ~ 1 atom to the carbo n ~2 atom of f3 ~ u ~ fructose. H ow can the sJX'Ci· ficity of sucrose be explained in light of the potential substrates?
Chapter
Lipids and Cell Membranes
0000000000000000000000 0 0
The surface of a soap bubble is a bilayer formed by det ergent molecu les. The polar heads (red) pack together, leaving t he hydrophobic groups (green) in contact with air on the inside and o utside of the bubble. Other bilayer structures define the boundary o f a cell. [(Left) Photonica.j
I O utl i n e l 12.1 Fatty Acids Are Key Constituents of Lipids 12.2 There Are Three Common Types of Membrane Lipids 12.3 Phospholipids and Glycolipids Readily Form Bimolecular Sheets in Aqueous Media 12.4 Proteins Carry Out Most Membrane Processes 12.5 Lipids and Many Membrane Proteins Diffuse Rapidly in the plane of the Membrane 12.6 Eukaryotic Cells Contain Compartments Bounded by Internal Membranes
326
he boundaries of cells are formed by biological membranes, the barriers that define the inside and the outside of a cell (Figure 12.1). These barriers prevent molecules generated inside the cell from leaking out and unwanted molecules from diffusing in; yet they also contain transport systems thai allow the cell to take up specific molecules and remove unwanted ones. Such transport systems confer on membranes the important property of selective permeability. Membranes are dynamic structures in which proteins float in a sea of lipids. The lipid components of the membrane form the barrier to permeability, and protein components act as a transport system of pumps and channels that allow selected molecules into and out of the cell. This transport system will be considered in the next chapter. In addition to an external cell membrane (called the plasma membrane), eukaryotic cells also contain internal membranes that form the boundaries of organelles such as mitochondria, chloroplasts, peroxi somes, and lysosomes. F unctional specialization in the course of evolution has been closely linked to the formation of such compartments_ Specific systems have evolved to allow the targeting of selected proteins into or through particular internal membranes and, hence, into specific organelles. External and internal membranes have essential features in common , and these essential features are the subject of this chapter. Biological membranes serve several additional functions indispensable for life, such as energy storage and information transduction, that are dictated by
the proteins associated with them. In this chapter, we will examine the general properties of membrane proteins how they can exist in the hydrophobic environment of the membrane while connecting two hydrophilic environments and defer a discussion of the functions of these proteins until the next and later chapters.
.
,
••
!• I
Many Common Features Underlie the Diversity
of Biological Membranes Membranes are as diverse in structure as they are in function. However, they do have in common a number of important attributes:
1. Membran es are sheetlike structures, only two molecules thick, that form dosed boundaries between different compartm ents . The thickness of most membranes is between 60 A. (6 nm) and 100 A(10 nm).
•
•
2. Membranes consist mainly of lipids and proteins. The mass ratio of lipids to proteins ranges from 1:4 to 4:1. Membranes also contain carbohydrates that are linked to lipids and proteins.
3. Membrane lipids are small molecules that have both hydrophilic and hydrophobic moieties. These lipids spontaneously form closed bimolecular sheets in aqueous media. These lipid bilayers are barriers to the flow of polar molecules.
4. Specific proteins mediate distinctive functions of membranes. Proteins serve as pumps, channels, receptors, energy transducers, and enzymes . Membrane proteins are embedded in lipid bi Iayers , which create suitable envirOIunents for their action.
•
• Figure 12.1 Red-blood -cell plasma mem brane. An electron 'micrograph of a preparatio n o f plasma membranes fro m red blood cells sho wing the membranes as seen "on edge:' in cross section. [Courtesy o f Dr. Vincent Marchesi.]
5. Membranes are noncovalent assemblies. The constituent protein and lipid molecules are held together by many noncovalent interactions, which act cooperatively,
6. Membranes are asymmetric. The two faces of biological membranes always differ from each other. 7. Membranes are fluid structures . Lipid molecules diffuse rapidly in the plane of the membrane, as do proteins, unless they are anchored by specific interactions. In contrast, lipid molecules and proteins do not readily rotate across the membrane. Membranes can be regarded as two-dimensional solutions of oriented proteins and lipids. 8. Most cell membranes are electrically polarized, such that the inside is negati ve [typically - 60 millivolts (m \1)J. Membrane potential plays a key role in transport, energy conversion, and excitability (Chapter 13).
12.1
Fatty Acids Are Key Constituents of Lipids
The hydrophobic properties of lipids are essential to their ability to form membranes. Most lipids owe t heir hydrophobic properties to on e component, their fatty acids.
Fatty Acid Names Are Based on Their Parent Hydrocarbons Fatty acids are long hydrocarbon chains of variou s lengths and degrees of unsaturation terminated with carboxylic acid groups. The systematic name for a fatty acid is derived from the name of its parent hydrocarbon by the substitution of oic for the final e. For example, the C I 8 saturated fatty acid is called octadecanoic acid because the parent hydrocarbon is octadecane. 327
328 CHAPTER 12 Lipids and Cell Membranes
, ,, ,
-
o Palmitate
(ionized form of palmitic acid)
Figure 12.2 Structures of two fatty acids. Palmitate is a l6-carbon, saturated fatty acid, and oleate is an l8-carbon fatty acid with a si ngle c is double bond.
p H, C
w
1: -
C C
2 -",C "
3 ~c-'"
H2 n
I "'-'0
H2
w-Carbon atom
H,C
(ionized form of oleic acid)
A C 18 fatty acid with one double bond is called octadecenoic acid; with two double bonds, octadecadienoic acid; and with three double bonds, octadeca· trienoic acid, The notation 18:0 denotes a C 1S fatty acid with no double bonds, whereas 18:2 signifies that there are two double bonds . The struc· tures of the ionized forms of two common fatty acids palmitic acid (16:0) and oleic acid (18: 1) are shown in Figure 12 ,2. Fatty acid carbon atoms are numbered starting at the carboxyl terminus, as shown in the margin. Carbon atoms 2 and 3 are often referred to as a and 13, respectively. The methyl carbon atom at the distal end of the chain is called the w -carbon atom. The position of a double bond is represented by the sym· 9 bol.l followed by a superscript number. For example, cis-.l means that there 2 is a cis double bond between carbon atoms 9 and 10; trans- .l means that there is a trans double bond between carbon atoms 2 and 3. Alternatively, the position of a doubie bond can be denoted by counting from the distal end, with the w-carbon atom (the methyl carbon) as number 1. An w- 3 fatty acid, for example, has the structure shown in the margin. Fatty acids are ionized al physiological pH, and so it is appropriate to refer to them according to their carboxylate form : for example, palmitate or hexadecanoate.
0
H2
Oleate
CH , w-3 double bond
(CH 2 )n COOAn w-l fatty acid
Fatty Aci ds Vary in Chain Length and Degree of Unsat uratio n Fatty acids in biological systems usually contain an even number of carbon atoms, typically between 14 and 24 (Table 12.1). The 16- and 18-carbon fatty acids are most common. The dominance of fatty acid chains contain· ing an even number of carbon atoms is in accord with the way in which TABLE 12.1 Some naturally occurri ng fatty acids in animals Number of carbons
Number of double bonds
12
0 0 0 0 0 0 0 1 1 2
14 16 18 20 22 24 16 18 18
Common name
Systematic name
Formula
Myrist at e
n-Dodecanoate n-Tetradecanoate
Palmitate
n-Hexadecanoate
Stearate
n-Octadecanoate
Arachidate
n-Eicosanoate
Behenate
n-Docosanoate n-Tetracosanoate cis - 6. I) -Hexadecenoate
CH,(CH, )lOCOO CH,(CH' )12COO CH3(CH,)"COO CH,(C H')16COO CH,(CH,)"COOCH3(C H, b oCOO CH,(CH, )"COOCH3 (CH, ),CH=CH(CH,),COOCH3 (CH, hCH=CH(C H, ),COOCH, (CH' )4(CH=CHCH, b(CH)6COO-
Laurate
Ugnocerate
Palmitaleate Oleate Linoleate
(;5 - /1 9 -Octadecenoate
cis.cis-/1 9, a 12 _ Octadecadienoate
18 20
3
4
Unolenate
aU-cis- 6 Y, 6 12 , 6 15 _
CH 3CH,(CH= CHCH, h(CH' )6COO-
Arachidonate
Octadecatrienoate all-cis /1 5, /1 t!. 6 11 • _6 14 Ei cosatetraenoate
CH3(C H, ),(CH= CHCH, ),(CH, hCOO '
329
fatty acids are biosynthesized (Chapter 26). T he hydrocarbon chain is almost invariably u nbranched in an imal fatty acids. The alkyl ch ain may be saturated or it may contain one or more dou ble bonds. The configuration of the double bonds in most unsaturated fatty acids is cis. The dou ble bonds in polyunsaturated fatty acids are separated by at least one methylene group. The properties of fatty acids and of lipids derived from them are markedly dependent on chain length and degree of saturation. Unsaturated fatty acids have lower melting points than saturated fatty acids of the same length. For example, the melting point of stearic acid is 69 .6°C, whereas that of oleic acid (which contains one cis double bond) is 13.4°C. The melting points of polyunsaturated fatty acids of t he C 18 series are even lower. Chain length also affects the melting point, as illustrated by the fact th at the melting temperature of palmitic acid (C 16 ) is 6.5 degrees lower t han that of stearic acid (C 18 ). Thus, short chain length and unsatumtion enhance the ./1.u-
12.2 Types of Me mbra ne Lipids
idity of fatty acids and of their derivatives.
12.2
There Are Three Common Types of Membrane Lipids
By definition, lipids are water-insoluble biomolecules that are highly soluble in organic solvents such as chloroform. Lipids have a variety of biological roles: they serve as fuel molecules, highly concentrated energy stores, signal molecules and messengers in signal-transduction pathways, and components of membranes. The first three roles of lipids will be considered in later chapters. Here, our focus is on lipids as membrane constituents . The three major kinds of membrane lipids are phospholipids, glycolipids, and cholesterol. We begin with lipids fo und in eukaryotes and bacteria. The lipids in archaea are distinct, although they have many features related to membrane formation incommon with lipids of other organisms.
Phospholi pids Are the Major Class of Membrane Lipids
Phospholipids are abundant in all biological membranes. A phospholipid
Fatty
G
acid I molecule is constructed from four components: one or more fatty acids, a y platform to which the fatty acids are attached, a phosphate, and an alcohol c Fatty attached to the phosphate (Figure 123). The fatty acid components provide e acid r a hydrophobic barrier, whereas the remainder of the molecule has hyPhosphate Alcohol 0 drophilic properties to enable interaction with the aqueous environment. I The platform on which phospholipids are built may be glycerol, a threeFigure 12.3 Schematic structure of a carbon alcohol, or sphingosine, a more complex alcohol. Phospholipids phospho lipid. derived from glycerol are called phosphoglycerides. A phosphoglyceride consists of a glycerol backbone to which are attached two fatty acid chains and aphosphorylated alcohol. In phosphoglycerides, the hydroxyl groups at C-1 and C- 2 of glycerol are esterified to the carboxyl groups of the two fatty acid • R) " / O ---!. C CH2 chains. The C -3 hydroxyl group of t he glycerol backbone is esterified Acyl groups with fatty acid to phosphoric acid. When no further additions are made, the resu lting o: 2hydrocarbon chains compou nd is phosphatidate (diacylglycerol 3-phosphate), the simplest C 3' : phosphoglyceride. Only smaIl amounts of phosphatidate are present H2 C"" /P\~O o 0 '0 in membranes. However, the molecule is a key intermediate in the biosynthesis of the other phospho glycerides (Section 26.1) . The abPhosphatidate (Diacylglycerol 3-phosphate) solute configuration of the glycerol 3-phosphate moiety of membrane lipids is shown in Figure 12.4. Figure 12.4 Structure of phosphatidat e The major phospho glycerides are derived from phosphatidate by (diacylglycerot 3-phosphate). The absolute configuration of the center carbo n (C-2) is shown. the formation of an ester bond between the phosphate group of
II
330
phosphatidate and the hydroxyl group of one of several alcohols. The com· mon alcohol moieties of phosphoglycerides are the amino acid serine, ethanolamine, choline, glycerol, and inositol.
CHAPTER 12 Lipids and Cell M embranes
HO
" c/
C " NH3+
HO
H2
~C ,, + ~CH3
HO"
c
C H2
"c~ " NH + 3
H2
H2 Serine
HO
H2
- OO\ } i
\/
N" \ " CH 3 CH 3
HO HO
~c" / OH
HO"
C H2
Choline
Ethanolamine
H C H2
OH OH
.
H
Glycerol
H
H
H
Inositol
The structural formu las of phosphatidylcholine and the other principal phosphoglycerides namely, p hosphatidylethanolamine, phosphatidylserine, phosphatidylinositol, and diphosphatidylglycerol are given in Figure 12.5. R,, ,
c/
0
" CH 2
o ......... c/
,
: -
N H2C" 0.......-P\ 0/ c,---c/ \" , CH, b H2 CH3
o
Phosphatidylcholine
Phosphatidylserine
-
o 0-' C-H "
c/
I
i H2 C"
O
q : -
o
~2
'°I: ,
NH3+
/ p\\-o/ " c/ ·b
H
H
1
H H .L
H2 C" o /-\\-0 \--'\-1 -. HO \ I
H
o
2
OH .\. OH H ()H H OH
Phosphatidylinositol
Phosphatidylethanolamine
Figure 12.5 Some common phosphoglycerides found in membranes.
H2
Diphosphatidylglycerol (cardiolipin)
Sphingomyelin is a phospholipid found in membranes that is not derived from glycerol. Instead, the backbone in sphingomyelin is sphingosine, an amino alcohol that contains a long, unsaturated hydrocarbon chain (Figure 12. 6). In sphingomyelin, the amino group of the sphingosine backbone is
\
HO Sphingosine
o
I
/ R, Figure 12.6 Structures of sphingosine and sphingomyelin. The sphingosine moiety of sphingomyelin is highlighted in blue.
C- NH
\ ,H
HO
H
Sphingomyelin
H2
:(-. 0 - '0
H
linked to a fatty acid by an amide bond . In add ition, the primary hydroxyl group of sphingosine is esterified to phosphorylcholine.
331 12.2 Types of Membrane Lipids
Membrane Lipids Can Include Carbohydrate Moieties Glycolipids, as their name implies, are sugar-containing lipids. L ike sphingomyelin, the glycolipids in animal cells are derived from sphingosine. The amino group of the sphingosine backbone is acylated by a fatty acid, as in sphingomyelin . G lycolipids differ from sphingomyelin in the identity of the unit that is linked to the primary hydroxyl group of the sphingosine backbone. In glycolipids, one or more sugars (rather than phosphorylcholine) are attached to this group. The simplest glycolipid, called a cerebroside, contains asingle sugar residue, either glucose or galactose.
Fatty acid unit
Sugar unit glucose
or HO
H
ga lactose
Cerebroside (a glycolipid)
More-complex glycolipids, such as gangliosides, contain a branched chai n of as many as seven sugar residues. Glycolipids are oriented in a completely asymmetric fashion with the sugar residues always on the extracellular side of
the membrane.
Cholesterol Is a Lipid Based on a Steroid Nucleus Cholesterol is a lipid with a structure quite different from that of phospholipids. It is a steroid, built from four linked hydrocarbon rings.
HO'/~/ Cholesterol
Ahydrocarbon ta.il is linked to the steroid at one end, and a hydroxyl group is attached at the other end. In membranes, the orientation of the molecule is parallel to the fatty acid chains of the phospholipids, and the hydroxyl group interacts with the nearby phospholipid head groups. Cholesterol is absent from prokaryotes but is found to varying degrees in virtually all animal membranes. It constitutes almost 25% of the membrane lipids in certain nerve cells but is essentially absent from some intracellular membranes.
Archaeal Membranes Are Built from Ether Lipids with Branched Chains 'I"'O - 168 ~-
--
(00-
figure 12.27 Locating the membrane-spanning helix of glycophorin. (A) Amino acid Jequence and transmembrane disposition of glycophorin A from the red-blood-cell membrane. rrfteen Q-linked carbohydrate units are shown as diamond shapes. and an N-linked unit is iOOwn as a lozenge shape. The hydrophobic residues (yellow) buried in the bilayer form a transmembrane (X helix. The carboxyl-terminal part of the molecule. located on the cytoplasmic SIde of the membrane. is rich in negatively charged (red) and positively charged (blue) residues. ~I Hydropathy plot for glycophorin. The free energy for transferring a helix of 20 residues from II'/! membrane to water is plotted as a funct ion of the position of the first residue of the helix ~ II'/! sequence of the protein. Peaks of greater than +84 kJ mol - I (+ 20 kcal mol - 1) in h)IIropathy plots are indicative of potential transmembrane helices. [(A) Courtesy of Dr. Vincent Marchesi; (8) after D. M. Engelman. T. A. Steitz. and A. Goldman. Identifying nonpolar mbilayer helices in amino acid sequences of membrane proteins. Annu. Rev. Biophys. Biophys. CMm. I~1986):321-353. Copyright © 1986 by Annual Reviews. Inc. All rights reserved.]
o
20
40
60
80
100
First amino acid residue in window
342
+ 168
--,
~
CHAPTER 12 Lipids and Cell Membranes
0
E + 84 :>1 ~
>
..c: ~
'"0.. 0 ~
"0
>-
+84 0
- 84
J:
- 168
20
400
First amino acid residue in wi ndow Cb) >
..c: ~
'"0
+ 168 +84 0
0.. ~
"0
>-
J:
- 84 _ 168 L--" :-~L--'-_ _-'-_'---_ _ _'----'----,:-'-,-_ 20 200
First amino acid residue in window
Chapter Integration Problem Data Interpretation Problems
11. Cholesterol effects. The red cu rve on the following graph shows the fluidity of the fatty acids of a phospholipid bilayer as a function of temperature. The blue curve shows the fluidity in the presence of cholesterol.
13. The proper environment . An understanding of the structure and function of membrane proteins has lagged behind that of other proteins. The primary reason is that membrane proteilll are more difficult to purify and crystallize . Why might this be the case?
Chapter
1
Membrane Channels and Pumps
.-
Closed
- - Open
The fl o w o f ions thro ugh a single membrane channel (c hannels a re sho wn in red in the illustration at the left) can be detect ed by the patc h-clamp technique, which records current changes as the channel transits between o pen and closed states. [(Left) After E. Neher and B. Sakmann. The patch clamp tec hnique. Copyright © 1992 by Scientific American, Inc. All rights reserved. (Right) Courtesy of Dr. Mauri cio MontaL]
he lipid bilayer of biological membranes is intrinsically impermeable to ions and polar molecules, yet certai n such species must be able to cross these membranes for normal cell function . Permeability is conferred by two classes of membrane proteins, pumps and channels. Pumps use a source of free energy such as ATP hydrolysis or light absorption to drive the thermodynamically uphill transport of ions or molecules. Pump action is an example of active transport . Channels, in contrast, enable ions to flow rapidly through membranes in a thermodynamically downhill direction. Channel action illustrates passive transport, or facilitated diffusion . Pumps are energy transducers in that they convert one form of free en ergy into another. Two types of ATP-driven pumps, P-type ATPases and the ATP-binding cassette (ABC) transporters, undergo conformational changes on ATP binding and hydrolysis that cause a bound ion to be trans ported across the membrane. A different mechanism of active transport utilizes the gradient of one ion to drive the active transport of another. An example of such a secondary transporter is the E. coli lactose transporter, a well -studi ed protein responsible for the uptake of a specific sugar from the environment of a bacterium . Many transporters of thi s class are present in the membranes of our cells. The expression of these transporters determines which metabolites a cell can import from the environment. Hence, adjust ing the level of transporter expression is a primary means of controlling metabolism. Pumps can establish persistent gradients of particular ions across membranes. Specific ion channels can allow these ions to flow rapidly across membranes down these gradients. These channels are among the most fascinating molecules in biochemistry in their ability to allow some ions to flow
O utl i n e 13.l
I
The Transport of Molecules Across a Membrane May Be Active or Passive
13.2 Two Families of Membrane Proteins Use ATP Hydrolysis to Pump Ions and Molecules Across Membranes 13.3 Lactose Permease Is an Archetype of Secondary Transporters That Use One Concentration Gradient to Power the Formation of Another 13.4 SpeCifiC Channels Can Rapidly Transport Ions Across Membranes 13.5 Gap Junctions Allow Ions and Small Molecules to Flow Between Communicating Cells 13.6 Specific Channels Increase the Permeability of Some Membranes to Water
3S1
352 CHAPTER 13 Membrane Channels and Pumps
freely through a membrane while blocking the flow of even closely related species. These gated ion channels are central to the functioning of our nerv· ous systems, acting as elaborately switched wires that allow the rapid now of current. We conclude with a discussion of a different class of channel: the cell· to-cell channel, or gap junction, allows the flow of metabolites or ions be· tween cells. For example, gap junctions are responsible for synchronizing muscle-cell contraction in the beating heart.
The Expression of Transporters Largely Defines the Metabolic Activities of a Given Cell Type Each cell type expresses a specific set of transporters in its plasma memo brane. The set of transporters expressed is crucial because these transporters largely determine the ionic composition inside cells and the compounds that can be taken up from the cell's environment. In some senses, the array of transporters ex pressed by a cell determines the cell's characteristics because a cell can execute only those biochemical reactions for which it has taken up the substrates. An example from glucose metaboli sm illustrates this point. As we will see in the discussion of glucose metabolism in Chapter 16, tissues differ in their ability to employ different molecules as energy sources . Which ti ssues can make use of glucose is largely governed by the expression of different members of a family of homologous glucose transporters call ed GLUT1, GLUT2, GLUT3, GLUT4, and GLUTS in different cell types. GLUT3, for example, is expressed only on neurons and a few other cell types. This transporter binds glucose relatively tightly so that these cells have first caU on glucose when it is present at relatively low concentrations. These are just the first of many examples that we will encounter demonstrating the critic~ rol e that transporter expression plays in the control and integration of metaboli sm.
13.1
The Transport of Molecules Across a Membrane May Be Active or Passive
We first consider some general principles of membrane transport. Two fac· tors determine whether a molecule will cross a membrane : (1) the perme· ability of the molecule in a lipid bilayer and (2) the availability of an energy source.
Many Molecules Require Protein Transporters to Cross Membranes As stated in Chapter 12, some molecules can pass through cell membranes because they dissolve in the lipid bilayer. Such molecules are called lipophilic molecules. The steroid hormones provide a physiological example. These cholesterol relatives can pass through a membrane in their path of movement, but what determines the direction in which they will move? Such molecules will pass through a membrane down their concentration gradient in a process called simple diffusion . In accord with the Second Law of Thermodynamics, molecules spontaneously move from a region of higher co ncentration to one of lower concentration . Matters become more complicated when the molecule is highly polar. For example, sodium ions are present at 143 mM outside a typical cell and at 14 mM inside the cell, yet sodium does not freely enter the ceil, because the charged ion cannot pass through the hydrophobic m embrane interior.
35 3
Insome circumstances, as during a nerve impulse, sodium ions must enter the cell. How are they able to do so? Sodium ions pass through specific channel s in the hydrophobic barrier formed by membrane proteins. This means of crossing the membrane is calledfaeilitated diffusion , because the diffusion across the membrane is facilitated by the channel. It is also called passive transport, because th e energy driving the ion movement originates from the ion gradient itself, without any contribution by the transport system. Channels, like enzymes, display substrate specificity in that they facilitate the transport of some ions, but not other, even closely related, ions . How is the sodium gradient established in the first place? In this case, sodium mu st move, or be pumped, against a concentration gradient. Because moving th e ion from a low concentration to a higher concentration results in adecrease in entropy, it requires an input of free energy. Protein transporters embedded in the membrane are capable of using an energy source to move the molecule up a concentration gradient. Because an input of energy from another source is required, this means of crossing the membrane is called
13.1 Active and Passive Transport Compared
active transport.
Free Energy Stored in Concentration Gradients Can Be Quantified An unequal distribution of molecules is an energy-rich condition because free energy is minimized when all concentrations are equal. Consequently, toattain such an unequal distribution of molecules, or concentration gradient, requires an input of free energy. Can we quantify the amount of energy required to generate a concentration gradient (Figure 13.1)? Consider an uncharged solute molecule. The free -energy change in transporting this species from side 1, where it is present at a concentration of el, to side 2, where it is present at concentration C2 , is l1G = RTln(C2/C l) = 2. 303RTlog1 0(C2/ Cl)
30
-Io E 20 l.)
"' E
+40
-
+20 I-
~
.-'"c
-
Q)
0
0 I-
- 20 Q) c
'" -" ~
- 40 I-
/
E Q) - 60 ::; - 80
Resting potential
J 1
2
Time (ms)
3
,
Figure 13.13 Action potential. Signals are sent alo ng neurons by the transient depolarization and repo larizati o n of the membrane.
One of the most important manifestations of ion -channel action is the nerve impulse, which is the fundamental means of communication in the nervous system . A nerve impulse is an electrical signal produced by the flow of ions across the plasma membrane of a neuron. The interior of a neuron, like that of most other cells, contains a high concentration of K + and a low concentration of N a +. These ionic gradients are generated by the N a + - K+ ATPase. The cell membrane has an electrical potential determined by the ratio of the internal to the external concentration of ions. In the resting state, the membrane potential is typically - 60 m V. A nerve impulse, or action potential, is generated when the membrane potential is depolarized beyond a critical threshold value (e.g., from -60 to - 40 m V). The membrane potential becomes positive within about a millisecond and attains a value of about + 30 mY before turning negative again. This amplified depolarization is propagated along the nerve terminal (Figure 13 .13). Ingenious experiments carried out by Alan Hodgkin and Andrew Huxley revealed that action potentials arise from large, transient changes in the permeability of the axon membrane to Na + and K + ions. The con ductance of the membrane to N a + changes first. Depolarization of the membrane beyond the threshold level leads to an increase in permeabil ity to sodium ions. Sodium ions begin to flow into the cell because of the large electrochemical gradient across the plasma membrane. The entry of Na + further depolarizes the membrane, leading to a further increase in N a + permeability. This positive feedback leads to a very rapid and large change in membrane potential, from about - 60 mV to +30 mY in a m i]] isecond. The membrane spontaneously becomes less permeable to sodium ions and more permeable to potassium ions. Consequently, potassium ions fl ow outward, and so the membrane potential returns to a negative value. The resting level of -60 mY is restored in a few milliseconds as the K + conductance decreases to the value characteristic of the unstimulated state. The wave of depolarization followed by repolarization moves rapidly along a nerve cell . The propagation of these waves allows a touch at the tip of your toe to be detected in your brain in a few milliseconds. This model for the action potential postulated the existence of ion channels specific for a + and K +. These channels must open in response to changes in membrane potential and then close after having remained open for a brief period of time. This bold hypothesis predicted the existence of molecules with a well-defined set of properties long before tools existed for their direct detection and characterization.
Closed
r
4 pA
1
Open
I
400 ms
4 ms
figure 13.14 Observing single channels. The results of a patch-clamp experi ment revealing a single ion channel undergoing transitions between closed and open states.
Patch-Clamp Conductance Measurements Reveal the Activities of Single Channels
Direct evidence for the existence of these channels was provided by the patch-clamp technique, which was introduced by Erwin Neher and Bert Sakmann in 1976 (Figure 13.14). This powerful technique enables the measurement of the ion conductance through a small patch of cell membrane. Remarkably, stepwise changes in membrane condu ctance are observed, corresponding to the opening and closing of individual ion channels. In this technique, a clean glass pipette with a tip diameter of about 111m is pressed against an intact cell to form a seal (Figure 13.15). Slight suction leads to the formation of a very tight seal so that the resistance between the inside of the pi~ette and the bathing solution is many gigaohms (1 gigaohm is equal to 10 ohms). Thus, a gigaohm seal (called a gigaseal) ensures that an electric current flowing through the pipette is identical with the current flowing through the membrane covered by the pipette. The gigaseal makes possible high-resolution current measurements while a known voltage is applied across the membrane. The flow of ions through a single channel and transitions between the open and the closed states of a channel can be monitored with a time resolution of microseconds.
• • •
•
•
•
• •
.... . . . '.. . . . .··... . . . " .. ·, ... . ..... ....'. .'. ....,.,"., . ... . . . . · . ."" . .•,., . . •
•
•
•
•
•
., ' "" .
Suction
Patch pipette Cell- . _
··. .. . . . .. ·."., . . ... .. '... ,. ' .. . . . ." . . ·· . .' . . • • ',. '. ". ' , " ' . .~ '., .' • • •• •
, "
•
•• '
, .' .': '. ' . '" ' , '"
..·
• . '•
..'
..
Suction
)
"
•
"
" '
,
' . '.' •
'
•
•
•
•
".'
' . '
Whole-cell mode
·..' . . . ... ' ·· .. ' .• .•' . . , .. • .• . '• . • • • • ...... • • • • •
••
•
. '
•
•
•
•
•
•
•
•
•
•
." •
.
•
•
Detachment
by pulling
low-resistance seal
Cell-attached mode (gigaohm seal)
Excised-patch mode (inside out)
figure 13.15 Patch-clamp modes. The patch-clamp technique for monitoring channel activity is highly versatile. A high-resistance seal (gigaseal) is formed between the pipette and a small patch of plasma membrane. This configuration is called cell attached mode. The breaking of the membrane patch by increased suction produces a low-resistance pathway between the pipette and the interior of the cell. The activity of the channels in the entire plasma membrane can be monitored in this whole-cell mode. To prepare a membrane in the excised-patch mode. the pipette is pulled away from the cell. A piece of plasma membrane with its cytoplasmic side now facing the medium is monitored by the patch pipette.
363
364 CHAPTER 13 Membrane Channels and Pumps
The Structure of a Potassium Ion Channel Is an Archetype for Many lon-Channel Structures
HO
HO-
F urthermore, the activity of a channel in its native membrane environlllent, even in an intact cell, can be directly observed . Patch -clam p m ethods provided one of the first views of single biomolecul es in action . Subsequently, other methods for observing single molecules were invented, opening ne\\! vistas on biochemistry at its most fundamental level.
"-r-, H
HO-
,NH OH HN . ," + ,, ,,
Tetrodotoxin
With the existence of ion channels firmly established by patch-clamp methods, scientists sought to identify the molecules that form ion channels. The Na + channel was first purified from the electric organ of electric eel, wh ich is a rich source of the protein forming this channel. That protein was pu rified on the basis of its ability to bind a specific neurotoxin. Tetrodotoxin, an organic compound isolated from the puffer fish , binds to N a + channels with great avidity (K i = 1 nM). The lethal dose of thi s poison for an adult human being is about 10 ng. T he isolated protei n is a single 260- kd chain. Subsequently, cDNAs encoding N a + channels were cloned and sequenced. Interestin gly, th e channel contains four internal repeats, each having a similar amino acid sequ ence, suggesting that gene duplication and divergence have produced the gene for thi s channel. H ydrophobicity profil es indicate that each repeat contains fi ve hydrophobic segments (51,52 , S3, S5, and S6) . Each repeat also contains a highl y positively charged S4 segment; positively charged arginine or lysine residues are present at nearly every third residue. It was proposed that segments 51 throu gh 56 are m embrane-spannin g a helices. The positively charged residues in S4 were proposed to act as the voltage sensors of the channel. The purification of K + channels proved to be much more difficult because of their low abundance and the lack of known hi gh -affinity ligands comparable to tetrodotoxin. Th e breakthro ugh came in studi es of mutant fru it fli es that shake violently when an esthetized wi th ether. T he mappin g and cloning of the gene, term ed shaker, responsible for thi s defect revealed the amino acid sequen ce encoded by a K +-channel gene. Shaker eDNA encodes a 70-kd protein that has fo ur subunits. R emarkably, each polypeptide contains sequ ences correspondin g to segm ents 5 1 through S6 in one of the repeated units of the N a + channel. Thus, a K +-channel subunit is homologous to one of the repeated units of N a + channels. Consistent with this hypothesis, four K +-channel subunits come together to fo rm a functional channel. More recently, bacteri al K+ channels were discovered that contain only the two m embrane-spanning regions corresponding to segments 55 and S6 . This and other information suggested that 5 5 and S6, including the region between them , form th e actual pore
Sodium channel Calcium channel
Pore
:;;J-C I 1
.-..- . "."'j--,I
Shaker potassium 1 r - -----.. -,• channel
Prokaryotic potassium channel
LJI
Figure 13.16 Sequence relations of ion channels. Like colors indicate st ruct urally similar regions of th e sod ium, calc ium, and po ta ssium channels. Each o f th ese channels ex hibits app roximat e fourfol d symmetry, eit her within one chain (sod ium, calcium channels) or by forming tetramers (pot assium channels).
1
36 5
--------'---'--
13.4 Ion Channels
-l'l
,,
ADP
+ Pi ~G o ' = - 30.S kJ mol - I (-7.3 kcal mol - I)
Glycerol 3-phosphate + H 2 0
glycerol + Pi I ~ G o ' = -9.2 kJmol - (- 2.2 kcalmol-
l
)
The magnitude of ~G o , for the hydrolysis of glycerol 3-phosphate is much smaller than that of ATP, which m eans that ATP has a stronger tendency to transfer its terminal phosphoryl group to water than does glycerol 3-phosphate. In other words, ATP has a higher phosphoryl-transJer potential (phosphoryl-group-transJer potential) than does glycerol 3-phosphate. The high phosphoryl -transfer potential of ATP can be explained by features of the ATP structure. Because ~G o' depends on the difJerence in free energies of the products and reactants, we need to examine the structures of both ATP and its hydrolysis products, ADP and Pi, to answer this question. Three factors are important: resonance stabilization, electrostatic repulsion, and stabilization due to hydration.
CH 2 0H
H-
IC-
OH
I
H2 C,
0
/'
2-
/ P:""
' 0 " \\ " 0
o
Glycerol 3-phosphate
1. Resonance Stabilization. ADP and, particularly, Pi, have greater reso nance stabilization than does ATP. Orthophosphate has a number of resonance forms of similar energy (Figure 15.4), whereas the 'Y phosphoryl group of ATP has a smaller number. Forms like that shown in Figure 15.5 are unfavorable because a positively charged oxygen atom is adjacent to a positively charged phosphorus atom, an electrostatically unfavorable juxtaposition.
~P HO, / "'-0 0-
... ( -....
rP HO/}/ "'-0 0
... .-
....
rP
HO~ ~o 0-
..... (-
....,
0-
0-
P
Ip+
+HcYl "'-0 0-
Figure 15.4 Resonance structures of orthophosphate.
2. Electrostatic R epulsion. At pH 7, the triphosphate unit of ATP carries about four negative charges. These charges repel one another because they are in close proximity. The repul sion between them is reduced when ATP is hydrolyzed .
3. Stabilization Due to Hydration. More water can bind more effectively to ADP and Pi than can bind to the phosphoanhydride part of ATP, stabilizing the A DP and Pi by hydration . ATP is often called a high -energy phosphate compound , and its phosphoanhydride bonds are referred to as high -energy bonds. Indeed , a "squiggle" (-P) is often used to indicate such a bond . Nonetheless, there is nothing special about the bonds themselves. Th ey are high -energy bonds in the sense that much Jree energy is released when they are hydrolyzed, for the reasons listed in factors 1 through 3.
0-
I ~p
RO~ "o ~ 0-
0-
" 0-
Figure 15.5 Improbable resonance structure. The structure cont ributes little to the terminal part o f ATP, because two pos itive charges are placed adjacent to each o ther.
416 --------~----------
CHAPTER 15 Metabolism: Basic Concepts and Design
H, ,/H
C
II
2-
0 1
o -">c/ C" o /p~~o - •:
0
•
o Phosphoenolpyruvate (PEP) NH
0
2-
I
H2
o ·,,> c/C"N'/C"N/P\~O - :1
o
H
0
CH 3 Creatine phosphate
2-
Phosphoryl-Transfer Potential Is an Important Form of Cellular Energy Transformation
The standard free energi es of hydrolysis provide a convenient means of comparing the phosphoryl-transfer potential of phosphorylated compounds. Such comparisons reveal that A TP is not the only compound with a high phosphoryl -transfer potential. Tn fact, some compounds in biological systems have a higher phosphoryl -transfer potential than thaI of ATP. These compounds include phosphoenolpyru vate (PEP), 1,3bi sphosphoglycerate (l ,3- BPG ), and creatine phosphate (Figure 15.6). Thus, PEP can transfer its phosphoryl group to ADP to form ATP. Indeed, this transfer is one of the ways in which ATP is generated in the breakdown of sugars (pp . 436 and 444) . It is significant that ATP has a phosphoryl -transfer potential that is intermediate among the biologically important phosphorylated molecules (Table 15 .1). This intermediate position enables ATP to function effici ently as a carrier of phosphory l groups. The amount of ATP in muscle suffices to sustain contractile acti vity for less than a second. Creatine phosphate in vertebrate muscle serves as a reservoir of high-potential phosphoryl groups that can be readily transferred to ATP. Indeed, we use creatine phosphate to regenerate ATP from ADP every time that we exercise strenuously. This reacti on is catalyzed by creatine kinase . CreHtine kinase
1.1-Bisphosphoglycerate (1.1-BPG) Figure 15.6 Compounds with high phosphoryl-transfer potential. These compou nds have a higher phosphoryltransfer potential than that of ATP and can be used to phosphorylate ADP to form ATP.
C reatine phosphate + ADP
pools is the basis of creatine's use as a dietary supplement by athletes in sports requiring short bursts of intense activity. After the creatine phosphate pool is depleted, ATP must be generated through metabolism (Figure 15.7). TABLE 15.1 Standard free energies of hydrolYSiS of some phosphorylated compounds kcal mol- 1
Compound Phosphoenolpyruvate l.3-Bisphosphoglycerate Creat ine phos phate ATP (to AOP) Glucose 1-phosphate Pyrophosphate Glucose 6-phosphate Glycerol 3-phosphate
- 61.9 - 49.4 - 43.1 - 30.5 - 20.9 -19.3 - 13.8 -9.2
-14.8 -11.8 - 10.3 - 7.3 - 5.0 - 4.6 - 3.3 - 2.2
/
417
Aerobic metabolism (Chapters 17 and 18)
ATP
15.3 The Oxidation of Carbon Fuels
Creatine phosphate
1
Anaerobic metabolism (Chapter 16)
~ OJ
c:
w
Seconds
15.3
,
Minutes
~
Hours
Figure 15.7 Sources of ATP during exercise. In t he initi al seconds. exercise is powered by existi ng high-p hosp horyl-transfer compo unds (ATP and creatine p hosphate). Subsequently. th e ATP must be regenerated by metabol ic pat hways.
)
The Oxidation of Carbon Fuels Is an Important Source of Cellular Energy
ATPserves as the principal immediate donor offree energy in biological systems rather than as a long -term storage form of free energy. In a typical cell , an ATP molecule is consumed within a minute of its formation. Although the total quantity of ATP in the body is limited to approximately 100 g, the turnover of thissmall quantity of ATP is very high. For example, a resting human being consumes about 40 kg of ATP in 24 hours. During strenuous exertion, the rate of utilization of ATP may be as high as 0.5 kg/ minute. For a 2-hour run , 60 kg (132 pounds) of ATP is utilized . Clearly, having mechanisms for regenerating ATP is vital. Motion, active transport, signal amplification , and biosynthesis can take place only if ATP is continually regenerated from ADP (Figure 15.8). The generation of ATP is one of the primary roles of catabolism . The carbon in fuel molecules such as glucose and fats is oxidized to C O 2 , and th e energy released is used to regenerate ATP from ADP and Pi. In aerobic organisms, th e ultimate electron acceptor in th e oxidation of carbon is O 2 and th e oxidation product is C O 2 . Con sequently, th e more reduced a carbon is to begin with, the m ore free energy is released by its oxidation . F igure 15 .9 shows the t.G OI of oxidation for one-carbon com pounds. Although fu el molecules are more compl ex (Fi gure 15 .10) than the single-carbon compounds depicted in Figure 15.9 , when a fuel is oxidized , the oxidation takes place one carbon at a ti me. The carbon -oxidation en ergy is used in some cases to create a co III pound with high phosphor y1- transfer potential and in other cases to create an ion gradient. In either case, the end point is the formation of A TP. Most energy - - - - - - - - --
-
C H/ \ .....H H
6C a' o~idation (kJ mol- ') /jGoloxidation
(kcal mol-' )
Oxidation of fuel molecules or Photosynthesis Figure 15.8 ATP- ADP cycle. This cycle is the fundamental mode of energy exchange in biological systems.
- - - - - - - - - - - . Least energy
o
OH
o H/
IC
Io
' H
Methane
Methanol
Formaldehyde
Formic acid
Carbon dioxide
-820
-703
-523
- 285
o
- 196
- 168
-125
-68
o
Figure 15.9 Free energy of oxidation of single-carbon compounds.
ADP
AlP '"
Figure 15.10 Prominent fuels. Fats are a more efficient fuel source than ca rbo hydrates such as glucose because the carbon in fats is more reduced.
H } --
H OH OH
HO
H
OH Fatty acid
Glucose
Compounds with High Phosphoryl-Transfer Potential Can Couple Carbon Oxidation to ATP Synthesis
H
C
OH
CH 20P032Glyceraldehyde 3-phosphate (GAP)
How is the energy released in the oxidation of a carbon compound converted into ATP? As an example, consider glyceraldehyde 3-phosphate (shown in the margin), which is a metabolite of glucose formed in the oxidation of that sugar. The C-l carbon (shown in red) is at the aldehyde-oxidation level and is not in its most oxidized state. Oxidation of the aldehyde to an acid will release energy.
o
"""C..... H
H-
IC-
Oxidation
OH
,
H
C
OH
CH20P032-
CH 20P0 32Glyc@raldehyde 3-phosphate
3-Phosphoglyceric acid
However, the oxidation does not take place directly. Instead, the carbon oxidation generates an acyl phosphate, 1 ,3 -bisphosphoglycerate_The electrons released are captured by NAD + , which we will consider shortly.
H
C
,
OH
H
C
OH
+ NADH + H'
CH20P032-
CH20P032-
1,3-Bisphosphoglycerate (1,3-BPG)
Glyceraldehyde 3-phosphate (GAP)
For reasons similar to those discussed for ATP, 1,3 -bisphosphoglycerate has a high phosphoryl-transfer potential. Thus, the cleavage of 1,3-BPG can be coupled to the synthesis of ATP.
H
C
OH
ICH20P032-
1,3-Bisphosphoglycerate
+ ADP
----»
H-
C
OH
ICH 0P0 2-
+
AlP
2 3 3-Phosphoglyceric acid
The energy of oxidation is initially trapped as a high-phosphoryl -transfer potential compound and then used to fOTm ATP. The oxidation energy of a carbon atom is transformed into phosphoryl-transfer potential, first as 1, 3-bisphosphoglycerate and ultimately as ATP. We will consider these reactions in mechanistic detail on p . 440.
Ion Gradients Across Membranes Provide an Important Form of Cellular Energy That Can Be Coupled to ATP Synthesis As described in C hapter 13, electrochemical potential is an effective means of storing free energy. Indeed, the electrochemical potential of ion gradients 41 8
CD Gradient created
419
Oxidation of fuels pumps protons out.
15.3 The Oxidation of Carbon Fuels
++ ++
- - - -
ADP ATP + P; '--....._ _7f + H2 0
-
Figure 15.11 Proton gradients. The o xidation of fuels can power the formation of proton grad ients by the action of specific proto n pumps. These proton grad ients can in turn drive the synthesis of ATP when the pro tons fl ow through an ATP synthesizing enzyme.
(2) Gradient used Influx of protons forms ATP.
across membranes, produced by the oxidation of fuel molecules or by photosynthesis, ultimately powers the synthesis of most of the ATP in cells. In general, ion gradients are versatile means of coupling thermodynamically unfavorable reactions to favorable ones. Indeed, in animals, proton gradients generated by the oxidation of carbon fuels account for more than 90% of ATP generation (Figure 15.11). This process is called oxidative phosphorylation (Chapter 18). ATP hydrolysis can then be used to form ion gradients of different types and functions. The electrochemical potential of a Na + 2 gradient, for example, can be tapped to pump Ca + out of cells or to transport nutrients such as sugars and amino acids into cells.
Energy from Foodstuffs Is Extracted in Three Stages Let us take an overall view of the processes of energy conversion in higher organisms before considering them in detail in subsequent chapters. Hans Krebs described three stages in the generation of energy from the oxidation of foodstuffs (Figure 15.1 2) .
In the first stage, large molecules in food are broken down into smaller units. This process is digestion. Proteins are hydrolyzed to their 20 different amino acids, polysaccharides are hydrolyzed to simple sugars such as glucose, and fats are hydrolyzed to glycerol and fatty acids. This stage is strictly a preparation stage; no lIseful energy is captured in this phase. In the second stage, these numerous small molecules are degraded to a few simple units that playa central role in metabolism. In fact, most of them sugars, fatty acids, glycerol, and several am ino acids are converted into the acetyl unit of acetyl CoA (p. 422). Some ATP is generated in this stage, but the amount is small compared with that obtained in the third stage. [n the third stage, ATP is produced from the complete oxidation of the acetyl unit of acetyl CoA. The third stage consists of the citric acid cycle and oxidative phosphorylation, which are the final common pathways in the oxidation of fuel molecules. Acetyl CoA brings acetyl units into the citric acid cycle [also called the tricarboxylic acid (TCA) cycle or Krebs cycle], where they are completely oxidized to CO 2 . Four pairs of electrons are transferred (three to NAD + and one to FAD) for each acetyl
FATS
POLYSACCHAR IDES
PROTEINS Stage I
Fatty acids and glycerol
Glucose and other sugars
Amino acids
Stage II
CoA
Citric acid cyde
2 CO 2
Stage III
Oxidative phosphorylation
H2 0 ATP Figure 15.12 Stages of catabolism. The extraction of energy from fuels can be di vided into three stages.
420 CHAPTER 15 Metabolism: Basic Concepts and Design
group that is oxidized . Then, a proton gradient is generated as electrons fl ow from the reduced forms of these carriers to 2 , and this gradi ent is used to synthesize ATP.
°
15.4
Metabolic Pathways Contain Many Recurring Motifs
At first glance, metabolism appears intimidating because of the sheer num· ber of reactants and reactions. N evertheless, there are unifying themes that make the comprehension of thi s compl exity m ore manageable. These uni· fying them es include common m etabolites, reactions, and regul atory schem es that stem from a common evolutionary heritage. Activated Carriers Exemplify the Modular Design and Economy of Metabolism
W e have seen that phosphoryl transfer can be used to drive otherwise endergonic reactions, alter the energy of co nformation of a protein , or serve as a signal to alter the activity of a protein . The phosphoryl-group donor in all of these reactions is ATP. In oth er word s, ATP is an activated carrier of phosphory l groups because phosphoryl transf er from ATP is an exergonic process. The use of activated carriers is a recurring motif in biochemistry, and we will consider several such carriers here. Many such activated carri· ers function as coenzymes:
Rea ctive site
H
~
N+
0 0 ,._ /
0
-.f "" o 0
NH,
/ o ...,P "
-rJ
H
HO
N
OHH
0
0,
HO
(7
,N
#
~N N=Z
OR
Figure 15.13 Structures of the oxidized fo rms of nicotinamide-derived electron carriers, Ni cotinamide ad enine dinucleotide (NAD I ) and nicotinamide adenine dinucleot ide phosphate (NADP+ ) are prom inent carriers o f high-energy electrons. In NAD I , R = H: in NADP+. R = PO,'-,
H
1. Activated Carriers of Electrons for Fuel Oxidation . In aerobic organisms, the ultimate electron acceptor in the oxidation of fu el molecules is O2, H owever , electrons are not transferred directly to 0 2' Instead , fuel molecules transfer electrons to special carri ers, which are either pyridine nucleotides or jlavins. The reduced forms of these carriers then transfer their hi gh-potential electrons to O 2 , N icotinamide adenine dinucl eotide is a major electron carrier in the oxi· dation of fuel molecules (Figure 15, 13). The reactive part of NAD I is its nicotinamide ring, a pyridine derivative synthesized from the vitamin niacin. In the oxidation of a substrate, the nicotinamide ring of NAD + accepts a hydrogen ion and two electrons, which are equivalent to a hydride ion (H:- ), The reduced form of this carrier is called NADH, Tn th e oxidized form , the nitrogen atom carries a positive charge, as indicated by NAD + , AD + is the electron acceptor in many reactions of the type OH .
+
NAD H
+
H+
In this dehydrogenation, one hydrogen atom of the substrate is directly transferred to NAD + , whereas the other appears in the solvent as a proton, Both electrons lost by the substrate are transferred to the nicotinamide ring, The other maj or electron carrier in the oxidation of fuel molecul es is the coenzyme fl avin adenine dinucleotide (Figure 15.14). The abbreviations for the ox idi zed and reduced forms of this carrier are FAD and FADH"respectively. FAD is the electron acceptor in reactions of the type
421
o - --,-~
15.4 Recurring M oti f s
H
~ NH
N
""'"
Reactive sites
I
~" ~ 7 ' N' "'0 H
HHH
(
H
I (I (I
, .-
OH OH
0-
( - OHO -
H2 C "
I
:I
0"/1
, p,
:I
~:p
Figure 15,14 5tructure of the oxidized form of flavin adenine dinucleotide (FAD). Th is electro n ca rrier consists o f a flavin mononucleo tide (FMN) unit (shown in blue) and an AMP unit (shown in black),
~
d"o~
'-o~
HO
OH
The reactive part ofFAD is its isoalloxazine ring, a derivative of the vitamin riboflavin (Figure 15 .15). FAD, like NAD +, can accept two electrons. In doing so, FAD, unlike NAD+, takes up two protons. These carriers of high-potential electrons as well as flavin mononucleotide (FMN), an electron carrier related to FAD, will be considered further in Chapter 18. 0
H
H H3 (
N
HJC
~
NH
+ 2 H+ + 2 e~
N
HJC H
N
0
pyruvate + AlP
+ AlP
or
Note: AG, the actua l fre e-energy change. has been calculated from AG"" and known concentrat ions reactants under typical physiologica l cond it ions. Glycolysis can proceed only if the o. G va lues of all reactions are negative. The small posit ive 6 G val ues o f three of the above reactions ind icate that the concentrations of metabolites in vivo in cells undergOing glycolYSiS are not precisely known.
another. This enzyme requir~s catalytic amounts of 2.3 -bisphosphoglycerate (2.3-BPG) to maintain an active-site histidine residue in a phosphory lated form. This phosphoryl group is transferred to 3-phosphoglycerate to re-form 2.3-bisphosphoglycerate. Enz-His-phosphate + 3-phosphoglycerate, >. Enz-His + 2.3-bisphosphoglycerate The mutase then functions as a phosphatase: it converts 2.3-bisphosphoglycerate into 2-phosphoglycerate. The mutase retains the phosphoryl group to regenerate the modified histidine. Enz-His + 2.3 -bisphosphoglycerate ~,= Enz-His-phosphate + 2-phosphoglycerate The sum of these reactions yields the mutase reaction: ' 2-phosphoglycerate
3-Phosphoglycerate,
In the next reaction. the dehydration of 2- phosphoglycerate introduces a double bond. creating an enol. Enolase catalyzes this formation of the enol phosphate phosphoenolpyruvate (PEP). This dehydration markedly elevates the transfer potential of the phosphoryl group. An enol phosphate has a high phosphoryl-transfer potential. whereas the phosphate ester of an ordinary al cohol. such as 2-phosphoglycerate. has a low one. The I1 C o' of the hydrolysis of a phosphate ester of an ordinary alcohol is - 13 kJ mol - I (- 3 kcal mol- I). whereas that of phosphoenolpyruvate is - 62 kJ mol - I (- 15 kcal mol- I). Why does phosphoenolpyruvate have such a high phosphoryl -transfer potential? The phosphoryl group traps the molecule in its unstable enol form. When the phosphoryl group has been donated to ATP. the enol un dergoes a conversion into the more stable ketone namely. pyruvate.
° 0."'(.. . . . / oPo, -
:'1
2-
(
AlP
-
o
o
:'1
0 "'( ........ / OH (
II / ("
H
-
/ ( "'"
H
H
Phosphenolpyruvate
;/( "", ...-::' 0
--.
I
:1
o·
(/
I
(H,
H
Pyruvate
Pyruvate
(enol form)
6.G0 in
~G in
kJ mo l- ' (kea l m ol - ')
1
Enzyme
Reaction type
kJ mo l- ' (kca l mo l- ' )
Hexokinase Phosphoglucose isomerase Phosphofructokinase Aldolase Triose phosphate isomerase Glyceraldehyde 3-phosphate dehydrogenase ~,osphoglycerate kinase Phosphoglycerate mutase Enolase Pyruvate kinase
Phosphoryl transfer Isomerization Phospho ryl transfe r Aldol cleavage Isomeri zation Phosphory latio n cou pled
- 16.7 (-4.0 ) +1.7 (+0.4) - 14.2 (- 3.4) + 23.8 (+ 5.7) + 7.5 (+1.8) + 6.3 (+1.5)
- 33.5 (- 8.0) - 2.5 (- 0 .6) - 22.2 (- 5.3) - /.3 (- 0.3) +2.5 (+0.6)
- 18.8 (- 4.5) + 4.6 (+ U)
+1.3 (+0.3) +0.8 (+0.2) - 3.3 (- 0.8) - 16.7 (- 4.0)
- 1.7 (- 0 04 )
to oxidation
Phos pho ryl transfe r Phosphoryl shift Dehydratio n Phosphory l t ransfer
+1.7 (+004 ) - 31.4 (- 7.5)
445 16.1 Glycolytic Pathway
446
Thus, the high phosphoryl-transfer potential of phosphoenolpyruvate arises primarily from the large driving force of the subsequent enol- ketone conversion. Hence, pyruvate is formed, and ATP is generated concomitantly. The virtually irreversible transfer of a phosphoryl group from phosphoenolpyruvate to ADP is catalyzed by pyruvate kinase. Because the molecules of ATP used in forming fructose 1,6-bisphosphate have already been regenerated, the two molecules of ATP generated from phosphoenolpyruvate are "profit."
CHA~P~IE;R~I~6~GJ.ly:c: o~ ly:si~s~an:dA Gluconeogenesis
Glucose ATP
AlP
Two ATP Molecules Are Formed in the Conversion of Glucose into Pyruvate
F-l ,6-BP
The net reaction in the transformation of glucose into pyruvate is DHAP
GAP
Glucose NAD+
+
2 Pi
+
2 ADP + 2 NAD I 2 pyruvate + 2 ATP
)
+
2 NADH
+
2H+
+ 2 H20
NADH-
Thus, two molecules of ATP are generated in the conversion of glucose into tux> molecules of pyruvate. The reactions of glycolysis are summarized in Table 16.1. Note that the energy released in the anaerobic conversion of glucose into two molecules of pyruvate is about -96 kJ mol- I (-23 kcal mol- I). We shall see in C hapters 17 and 1 8 that much more energy can be released from glucose in the presence of oxygen.
PEP
2x
2ATP
NADH_+ NAD+ -
+-
Ethanol Location of redox-balance steps. The generat io n and consumption of NADH. located within the glycolytic pathway.
Pyruvate NADH
co, Acetaldehyde
Lactate
NAD + Is Regenerated from the Metabolism of Pyruvate The conversion of glucose into two molecules of pyruvate has resulted in the net synthesis of ATP. However, an energy-converting pathway that stops at pyruvate will not proceed for long, because redox balance has not been maintained. As we have seen, the activity of glyceraldehyde 3-phosphate dehydrogenase, in addition to generating a compound with high phosphoryltransfer potential, of necessity leads to the reduction ofNAD+ to NADH. In the cell, there are limited amounts of NAD +, which is derived from the vitamin niacin, a dietary requirement for human beings. Consequently, NAD+ must be regenerated for glycolysis to proceed. Thus, the final process in the pathway is the regeneration ofNAD + through the metabolism of pyruvate. The sequence of reactions from glucose to pyruvate is similar in most 0[ganisms and most types of cells. In contrast, the fate of pyruvate is variable. Three reactions of pyruvate are of primary importance: conversion into ethanol, lactate, or carbon dioxide (Figure 16.9)_ The first two reactions are fermentations that take place in the absence of oxygen. In the presence of oxygen, the most common situation in multicellular organisms and in many unicellular ones, pyruvate is metabolized to carbon dioxide and water through the citric acid cycle and the electron-transport chain. We now take a closer look at cO 2 these three possible fates of pyruvate. Acetyl CoA
NADH
Ethanol
Further oxidation
Figure 16.9 Diverse fates of pyruvate. Ethanol and lactate can be formed by reactions that include NADH. Alternatively, a two-carbon unit from pyruvate can be coupled to coenzyme A (see p. 420) to form acetyl CoA.
1. Ethanol is formed from pyruvate in yeast and several other microorganisms. The first step is the decarboxylation of pyruvate. This reaction is catalyzed by pyruvate decarboxylase, which requires the coenzyme thiamine pyrophosphate. This coenzyme, derived from the vitamin thiamine (BI ), also participates in reactions catalyzed by other enzymes (p. 4711). The second step is the reduction of acetaldehyde to ethanol by NADH, in a reaction catalyzed by alcohol dehydrogenase. This process regenerates NAD +.
0
- :I
H+
C,,-- .& 0 0'/ C:?'
ICH,
CO 2
\, / Pyruvate deca rboxylase
Pyluvate
NADH + W
~O
H,
•
ICH,
NAD+
\, ./
C
•
Fermentation
H, / OH H- C
•
An ATP-generating process in which organic
compounds act as both donors and accep-
ICH,
Alcohol
dehydrogenase
tors of electrons. Fermentation can take
place in the absence of 0 , - Discovered by
Ethanol
Acetaldehyde
Louis Pasteur, who described fermentation as " la vie sans rair" ("a life without air"),
The active site of alcohol dehydrogenase contains a zinc ion that is coordinated to the sulfur atoms of two cysteine residu es and a nitrogen atom of histidine (Figure 16.10). This zinc ion polarizes the carbonyl group of the substrate to favor the transfer of a hydride from NADH. The conversion of glucose into ethanol is an example of alcoholic ferrnentation. The net result of this anaerobic process is Glucose
+ 2 Pi +
2 ADP
+
2H+
) 2 ethanol
+
2 CO 2
+
2 A TP
+
NADH
Hydride donor
Cys
2 H 20
Note that NAD + and NADH do not appear in this equation, even though they are crucial for the overall process. NADH generated by the oxidation of glyceraldehyde 3-phosphate is consumed in the reduction of acetaldehyde to ethanol. Thus, there is no net oxidation--'reduction in the conversion of glucose into ethanol (Figure 16.11). The ethanol formed in alcoholic fermen tation provides a key ingredient for brewing and winemaking.
Acetaldehyde His Hydride acceptor
2. Lactate is formed from pyruvate in a variety of microorganisms in a process called lactic acid fermentation . The reaction also takes place in the cells of higher organisms when the amount of oxygen is limiting, as in muscle cells during intense activity. The reduction of pyruvate by NADH to form lactate is catalyzed by lactate dehydrogenase.
Figure 16.10 Active site of alcohol dehydrogenase. The active site contains a zinc ion bound to two cysteine residues and one histidine residue. N otice that the zinc ion binds the acetaldehyde substrate through its oxygen atom, polarizing the substrate so that it more easily accepts a hydride from NAOH. Only the nicotinamide ring of NADH is shown.
-
NADH + H+
0 ", ./0 C
HO'-
I
C-
H
I
Lactate dehydrogenase
CH, Lactate Figure 16.11 Maintaining redox balance. The NAOH produced by the glyceraldehyde 3-phosphate d ehydrogenase reaction must be reoxidi zed to NAD+ for the glycolytic pathway to continue. In alcoholic fermentation, alcohol dehydrogenase oxidizes NADH and generates ethanol. In lactic acid fermentation (not shown), lactate dehydrogenase oxidizes NADH while generating lacti c acid.
The overall reaction in the conversion of gl ucose into lactate is Glucose
+
2 Pi
+
2 ADP --+) 2 lactate
+
2 A TP
+
2 H 20
As in alcoholic fermentation, there is no net oxidation- reduction. The NADH formed in the oxidation of glyceraldehyde 3-phosphate is consumed in the reduction of pyruvate. The regeneration of NAD I- in the reduction of
O~C/H H-
IC- OH ICH 0PO,2-
0 ~C /OPO,2Glyceraldehyde 3-phosphate
H-
I'C- OH ICH 0PO,2-
Glyceraldehyde
2 1.3.Sisphosphoglycerate
' -phosphate
(l,l-BPG)
2
dehydrogenase
-
•
•
•
°H
H+
.,;;-C,,--.,;::-O C
°
CH, Pyruvate
\. /
NAO '
CO2
•
H,
.,;::-0 C
ICH,
Acetaldehyde
Alcohol dehydrogenase
H, / OH H- C
ICH,
Ethanol
447
448
TABLE 16.2 Examples of pathogenic obligate anaerobes
CH APTER 16 Glycolysis and Gluconeogenesis
Bacterium
Result of infection
Clostridium tetoni Clostridium botulinum
Tetanus (lockjaw) Botulism (an especialiy severe type of food poi soning) Gas gangrene (gas is produced as an end point o f the fermentation. distorting and destroying the tissue) Cat scratch fever (flu-like symptoms) Abdominal, pelvic, pulmonary, and bl ood infectio ns
Clostridium perfringens
Bartonella hensela Baderoides (rogilis
pyruvate to lactate or ethanol sustains the continued process of glycolysis under anaerobic conditions.
3. Only a fraction of the energy of glucose is released in its anaerobic conversion into ethanol or lactate. Much more energy can be extracted aerobically by means of the citric acid cycle and the electron-transport chain. The entry point to this oxidative pathway is acetyl coenzyme A (acetyl CoAl, which is formed inside mitochondria by the oxidative d ecarboxylation of pyruvate. Pyruvate
+
NAD +
+
CoA -~) acetyl eoA
+
CO 2
+
NADH
This reaction, which is catalyzed by the pyruvate dehydrogenase complex, will be considered in detail in Chapter 17. The NAD + required for this reaction and for the oxidation of glyceraldehyde 3-phosphate is regenerated when NADH ultimately transfers its electrons to O 2 through the electrontransport chain in mitochondria.
Fermentations Provi de Usable Energy in the Absence of O xygen
TABLE 16.3 Glucose
Fermentations yield only a fraction of th e energy available from the complete combustion of glucose. Why is a relatively inefficient metabolic pathway so extensively used? The fundamental reason is that oxygen is not required. The ability to survive without oxygen affords a host of living accommodations such as soils, deep water, and skin pores. Some organisms, called obligate anaerobes, cannot survive in the presence of O 2 , a highly reactive compound. The bacterium Clostridium perjringens, the cause of gangrene, is an example of an obligate anaerobe. Other pathogenic obligate anaerobes are listed in Table 16.2 . Skeletal muscles in most animals can function anaerobically for short periods. For example, when animals perform bursts of intense exercise, their ATP needs rise faster than the ability of the body to provide oxygen to the muscle. The muscle functions anaerobically until fatigue sets in, which is caused, in part, by lactate buildup. Although we have considered only lactic acid and alcoholic fermentation, microorganisms are capable of generating a wide array of molecules as end points to fermentation (Table 16.3). Indeed, many food Starting and ending points of various fermentations products, including sour cream, yogurt, various cheeses, beer, wine, and sauerkraut, result from fermentation . lactat e ->
Lactate Glucose Ethanol Arginine Pyrimidines Purines
Ethylene glycol Threonine Leucine Phenylalanine
-> -> --> --> --> --> --> --> -->
acetate ethanol
acetate carbon dioxide carbon dioxide
format e acetate propionate 2-alkylacet ate
propionate
Note: The products o f some fermentations are the substra les fOl ot hers.
The Binding Site for N AD + Is Similar in M any Dehydrogenases The three dehydrogenases - glyceraldehyde 3-phosphate dehydrogenase, alcohol dehydrogenase, and lactate dehydrogenase have quite different three-dimensional structures. However, their NAD + binding domains are strikingly similar (Figure 16.12). This nucleotide-binding region is made up of four ex helices and a sheet of six
Nicotinamide-binding half
449 16.1 Glycolytic Pathway Nicotinamide
~ Figure 16.12 NAD I -binding region in dehydrogenases. Notice that the nicotinamide-binding half (yellow) is struc turally similar to the adenine-binding half (red). The two hal ves together form a structura l motif called a Rossmann fold. Th e NAD+ molecule binds in an extended confo rmation. [Drawn from 3LDH.pdb.]
Pyrophosphate Adenine-binding half Adenine
NAD
parallel ~ strands. Moreover, in all cases, the bound NAD+ displays nearly the same confonnation. This common structural domain was one of the first recurring structural domains to be discovered. It is often called a Rossmannfold after Michael Rossmann, who first recognized it. This fold likely represents a primordial dinucleotide-binding domain that recurs in the dehydrogenases of glycolysis and other enzymes because of their descent from a common ancestor.
Glucose
Glucose-6P (G-6P)
Galactose
,
Fructose (adipose tissue) "
Fructose and Galactose Are Converted into Glycolytic Intermediates
F-l ,6-BP
Although glucose is the most widely used monosaccharide, others also are important fuels . Let us consider how two abundant sugars fructose and galactose can be funneled into the glycolytic pathway (Figure 16.13). There are no catabolic pathways for metabolizing fructose or galactose, and so the strategy is to convert these sugars into a metabolite of glucose. Fructose can take one of two pathways to enter the glycolytic pathway. Much of the ingested fructose is metabolized by the liver, using the fructose l -phosphate pathway (Figure 16.14). The first step is the phosphorylation offructose to fructose l-phosphate by fructokitUlSe. Fructose l-phosphate is then split into glyceraldehyde and dihydroxyacetone phosphate, an intermediate in glycolysis. This aldol cleavage is catalyzed by a specific fructose l -phosphate aldolase. Glyceraldehyde is then phosphorylated to glyceraldehyde 3-phosphate, a glycolytic intermediate, by triose kinase. In other tissues, fructose can be phosphorylated to fructose 6-phosphate by hexokiTUIse. Galactose is converted into glucose 6-phosphate in four steps. The first reaction in the galactose- glucose interconversion pathway is the phosphorylation of galactose to galactose i -phosphate by galactokinase. ATP
CH 2 0H
ADP + H+
HO )
Galactokinase
OH
0y OH
p
1:"'-. I: 0
o Galadose
Galactose 1-phosphate
2-
Fructose (liver)
2x
Pyruvate Figure 16.13 Entry points in glycolysis for galactose and fructose.
(
Fructose
J
AlP Fructokin ase
Galactose l -phosphate then acquires a uridyl group from uridine diphosphate glucose (UDP-glucose), an intermediate in the synthesis of glycosidic linkages (p _314)_
ADP Fructose I-phosphate Fru ctose I-phosphate aldolase
Glyceraldehyde Triose ki na se
ATP
Dihydroxyacetone phosphate
-
ADP UDP-glucose
Galactose I-phosphate
Glyce ralde hyde 3-phosphate
-
Galadose I-phosphate uridyl transferase
Figure 16.14 Fructose metabolism. Fructose enters the glycolytic pathway in the liver through the fructose I-phosphate pathway.
HO
luridinel
+
"\ OH / HO \
J,
OH
2-
/ .0 '-p ,I: ~
"'
(j UDP-galactose
'-
0
Glucose I-phosphate
UDP-ga ladose 4-epi merase
UDP-glucose
The products of this reaction, which is catalyzed by galactose i-phosphate uridyl transferase, are UDP-galactose and glucose l -phosphate. The galactose moiety of UDP-galactose is then epimerized to glucose. The configuration of the hydroxyl group at carbon 4 is inverted by UDP -galactose
4 -epimerase. The sum of the reactions catalyzed by galactokinase, the transferase, and the epimerase is Galactose + A TP
--l»
glucose l -phosphate + ADP + H +
Note that UDP-glucose is not consumed in the conversion of galactose into glucose, because it is regenerated from UDP -galactose by the epimerase. This reaction is reversible, and the product of the reverse direction also is important. The conversion of UDP -glucose into UDP-galac-
tose is essential for the synthesis of galactosyl residues in complex polysaccharides and glycoproteins if the amount of galactose in the diet is inadequate to meet these needs. Finally, glucose l -phosphate, formed from galactose, is isomerized to glucose 6-phosphate by phosphoglucomutase. We shall return to this reaction when we consider the synthesis and degradation of glycogen, which proceeds through glucose l -phosphate, in C hapter 21. 450
451
Many Adults Are Intolerant of Milk Because They Are Deficient in Lactase
16.1 Glycolytic Pathway
W
Many adults are unable to metabolize the milk sugar lactose and ex~ perience gastrointestinal disturbances if they drink milk. Lactose intolerance, or hypolactasia, is most commonly caused by a d eficiency of the enzyme lactase, which cleaves lactose into glucose and galactose. CH, OH
CH,OH
CH,OH
HO
HO
} - -IO
lactase
o
+ HO
OH OH
OH Lactose
OH Galactose
OH OH Glucose
"Deficiency" is not quite the appropriate term, because a decrease in lactase is normal in the course of development in all mammals. As children are weaned and milk becomes less prominent in their diets, lactase activity normally declines to about 5 to 10% of the level at birth. This decrease is not as pronounced with some groups of people, most notably Northern Europeans, and people from these groups can continue to ingest milk without gastrointestinal difficulties. With the appearance of milk-producing domesticated animals, an adult with active lactase would hypothetically have a selective advantage in being able to consume calories from the read ily available milk. What happens to the lactose in the intestine of a lactase-deficient person? The lactose is a good energy source for microorganisms in the colon, and they ferment it to lactic acid while also generating methane (CH 4 ) and hydrogen gas (H2)' The gas produced creates the uncomfortable feeling of gut distension and the annoying problem of flatulence. The lactate produced by the microorganisms is osmotically active and draws water into the intestine, as does any undigested lactose, resulting in diarrhea. If severe enough, the gas and diarrhea hinder the absorption of other nutrients such as fats and proteins. The simplest treatment is to avoid the consumption of products containing much lactose. Alternatively, the enzyme lactase can be ingested with milk products. Galactose Is Highly Toxic If the Transferase Is Missing
W
Less common than lactose intolerance are disorders that interfere ~ with the metabolism of galactose. The disruption of galactose metabolism is referred to as galactosemia. The most common form, called classic galactosemia, is an inherited deficiency in galactose i-phosphate uridyl transferase activity. Afflicted infants fail to thrive. They vomit or have diarrhea after consuming milk, and enlargement of the liver and jaundice are common, sometimes progressing to cirrhosis. Cataracts will form, and lethargy and retarded mental development also are common. The bloodgalactose level is markedly elevated, and galactose is found in the urine. The absence of the transferase in red blood cells is a definitive diagnostic criterion. The most common treatment is to remove galactose (and lactose) from the diet. An enigma of galactosemia is that, although elimination of galactose from the diet prevents liver disease and cataract development, the majority of patients still suffer from central nervous system malfunction, most commonly a delayed acquisition oflanguage skills. Female patients also dis play ovarian failure.
Scanning electron micrograph of Lactobacillus. The anaerobic bacterium Lactobacillus is shown here (artificially colored) at a magnification of 22.24SX . As suggested by its name, this genus of bacteria ferments glucose into lactic acid and is widely used in the food industry. Lactobacillus is also a component of the normal human bacterial flora of the urogenital tract where. because of its ability to generate an acidic environment, it prevents the growth of harmful bacteria. [Dr. Denni s Kunkel/PhotoTake.]
452 CHAPTER 16 Glycolysis and Gluconeogenesis
Cataract formation is better understood . A cataract is the clouding of the normally clear lens of the eye. Tf the transferase is not active in the lens of the eye, the presence of aldose reductase causes the accumulating galactose to be reduced to galactitol. H HO,,-- / H
O~ / H C
H HO
C
IC
i-
OH
+ W
H
HO
C
H
H
C
OH
CH 20H Galactose
H
NADPH
HO
IC IC
H
C
NADP+
\,/ Aldose
reductase
C
HO )
OH H H OH
CH 2 0H Galactitol
Galactitol is osmotically active, and water will diffuse into the lens, instigating the formation of cataracts . In fact, there is a high incidence of cataract formation with age in populations that consume substantial am ounts of milk into adulthood.
16.2
The Glycolytic Pathway Is Tightly Controlled
The glycolytic pathway has a dual role: it degrades glucose to generate ATP and it provides building blocks for synthetic reactions, such as the format ion of fatty acids. The rate of conversion of glucose into pyruvate is regulated to meet these two major cellular needs. In metabolic pathways, enzymes catalyzing essentially irreversible reactions are potential sites of control. Jn glycolysis, the reactions catalyzed by hexokinase, phosphofructokinase, and pyruvate kinase are virtually irreversible; hence, these enzymes would be expected to have regulatory as well as catalytic roles. In fact, each of them serves as a control site. These enzymes become more active or less so in response to the reversible binding of allosteric effectors or covalent modification. In addition, the amounts of these important enzymes are varied by the regulation of transcription to m eet chan ging metabolic needs . The time reo quired for reversible allosteric control, regulation by phosphorylation, and transcriptional control is m easured typically in milliseconds, seconds, and hours, respectively. We will consider the control of glycolysis in two different tissues skeletal muscle and liver.
Glycolysis in Muscle Is Regulated to Meet the Need for ATP G lycolysis in skeletal muscle provides ATP primarily to power contraction. Consequently, the primary control of muscle glycolysis is the energy charge of the cell the ratio of ATP to AMP. Let us examine how each of the key regulatory enzymes responds to changes in the amounts of ATP and AMP present in the cell.
Phosphofructokinase.
Phosphofructokinase is the most important control site in the mammalian glycolytic pathway (Figure 16.1 5) . High levels of ATP allosterically inhibit the enzyme (a 340 -kd tetramer) . ATP binds to a specific regulatory site that is distinct from the catalytic site. The binding of ATP lowers the enzyme's affinity for fructose 6- phosphate. Thus, a high concentration of ATP converts the hyperbolic binding curve of fructose
453 16.2 Control of Glycolysis Catalytic sites
sites
~ Figure 16.15 Structure of
Catalytic sites
phosphofructokinase. lhe structure of phosphofructokinase fro m E. coli comprises a tetramer of four identical subunits. N otice the separation of the catalytic and allosteric sites. Each subunit of the human liver enzyme consist s of two domains that are similar to the E. coli enzyme. [Drawn fro m 1PFK.pdb.)
6-phosphate into a sigmoidal one (Figure 16.16). AMP reverses the in hibitory action of ATP, and so the activity of the enzyme increases when the ATP/ AMP ratio is luwered. In other words, glycolysis is stimulated as the energy charge falls. A decrease in pH also inhibits phosphofructokinase activity by augmenting the inhibitory effect of ATP. The pH might fall when muscle is functioning anaerobically, producing excessive quantities of lactic acid. The inhibitory effect protects the muscle from damage that would result from the accumulation of too much acid. Why is AMP and not ADP the positive regulator of phosphofructoki nase? When AT P is being utilized rapidly, the enzyme adenylate kinase (Section 9.4) can form ATP from AD P by the following reaction : ADP + ADP,
' ATP + AMP
Thus, some ATP is salvaged from ADP, and AMP becomes the signal for the low-energy state. Moreover, the use of AMP as an allosteric regulator provides an especially sensitive control. We can understand why by consid ering, first, that the total adenylate pool ([ATP), [ADP), [AMP)) in a cell is constant over the short term and, second , that the concentration of ATP is greater than that of ADP and the concentration of ADP is, in turn, greater than that of AMP. Consequently, small -percentage changes in [ATP) result in larger-percentage changes in the concentrations of the other adenylate nucleotides. This magnification of small changes in [AT PJto larger changes in [AMP] leads to tighter control by increasing the range of sensitivity of phosphofructokinase. Hexokinase. Phosphofructokinase is the most prominent regulatory enzyme in glycolysis, but it is not the only one. Hexokinase, the enzyme catalyzing the first step of glycolysis, is inhibited by its product, glucose
low [AIPI
1 .-v -o
~ c o .-
High [AIPI
[Fructose 6-phosphatel
'
Figure 16.16 Allosteri c regulation of phosphofructokinase. A high level of AlP inhibits t he enzyme by decrea sing its affinity fo r fru ct ose 6-phosphate. AMP diminishes and citrate enhances the inhibitory effect of AlP.
4S4
6-phosphate. High concentrations of this molecule signal that the cell no longer requires glucose for energy or for the synthesis of glycogen, a storage form of glucose (p. 311), and the glucose will be left in the blood. A rise in glucose 6-phosphate concentration is a means by which phosphofructokinase communicates with hexokinase. When phosphofructokinase is inactive, the concentration of fructose 6-phosphate rises. In turn, the level of glucose 6phosphate rises because it is in equilibrium with fructose 6-phosphate. Hence, the inhibition of phosphofructokinase leads to the inhibition of hexokinase. Why is phosphofructokinase rather than hexokinase the pacemaker of glycolysis? The reason becomes evident on noting that glucose 6-phosphate is not solely a glycolytic intermediate. In muscle, glucose 6-phosphate can also be converted into glycogen. The first irreversible reaction unique to the glycolytic pathway, the committed step (Section 10.1), is the phosphorylation of fructose 6-phosphate to fructose 1,6-bisphosphate. Thus, it is highly appropriate for phosphofructokinase to be the primary control site in glycolysis. In general, the enzyme catalyzing the committed step in a metabolic sequence is the most important control element in the pathway.
CHAPTER 16 Glycolysis and Gluconeogenesis
•
Pyruvate Kinase. Pyruvate kinase, the enzyme catalyzing the third irreversible step in glycolysis, controls the outflow from this pathway. This final step yields ATP and pyruvate, a central metabolic intermediate that can be oxidized further or used as a building block. ATP allosterically inhibits pyruvate kinase to slow glycolysis when the energy charge is high. Finally, alanine (synthesized in one step from pyruvate, p. 686) also allosterically inhibits pyruvate kinase in this case, to signal that building blocks are abundant. When the pace of glycolysis increases, fructose 1,6bisphosphate, the product of the preceding irreversible step in glycolysis, activates the kinase to enable it to keep pace with the oncoming high flu x of intermediates_ A summary of the regulation of glycolysis in resting and active muscle is shown in Figure 16.17.
Figure 16.17 Regulation of glycolysis in muscle. At rest (left). glycolysis is not very active (thin arrows). The high concentratio n of ATP inhibits phosphofructokinase (PFK). py ruvate kinase. and hexo kinase. G lucose 6phosphate is converted into glycogen (Chapter 21). During exercise (right). the decrease in the ATP/ AMP rati o resulting from muscle contractio n activates phosphofructokinase and hence glycolysis. The flux down the pathway is increased. as represented by the thick arrows.
AT REST (glycolysis inhibited)
DURING EXERCISE (glycolysis stimulated)
Glucose
Glucose
Hexokinase
8
Glycogen ...(~- Glucose 6-phosphate - -
Negative feedback
Hexokinase
Glycogen - -;.) Glucose 6-phosphate
fructose 6-phosphate
Fructose 6-phosphate
PFK
ATP (
Fructose l.6-bisphosphate
* *
ATP
/
PFK (
_
Pyruvate kinase
Pyruvate
ATP/AMP
High energy charge ATP/AMP
ATP
Phosphoenolpyruvate ATP
'\
Fructose 1.6-bisphosphate
FeedfolWard stimulation
*,,(., Relaxed muscle fiber
low energy charge
Phosphoenolpyruvate Musclefiber contraction
ATP (
"'" Pyruvate kinase
Pyruvate
CO 2 + H2 0 (long, slow run)
Lactate (sprint)
1 ",M F-2,6-6P
100
45• 5
1 ",M F-2,6-6P
16.2 Control of Glycolysis
80
.~ v -o
~ 60
o
.-~
o
20
o
1
2
3
4
5
o
[Fructose 6-phosphate] (mM)
(A)
1
2
3
4
5
[ATP] (mM)
(6)
Figure 16.18 Activation of phosphofructokinase by fructose 2, 6-bisphosphate. (A) Th e sigmoidal dependence of velocity o n substrate concentration becomes hyperboli c in the presence of 1 fLM fructose 2,6-bisphosphate. (B) ATP, acting as a substrate, initially stimulates the reaction. As the concentration of ATP increases, it acts as an alloster ic inhibitor. The inhibitory effect of ATP is reversed by fructose 2,6-bisphosphate. [After E. Van schaftingen, M. F. Jett, L. Hue, and H. G. Hers. Proe. Natl. Acad. Sci. U.s.A. 78(1981):3483- 3486.]
The Regulation of Glycolysis in the Liver Reflects the Biochemical Versatility of the Liver The liver has more-diverse biochemical functions than muscle. Significantly, the liver maintains blood-glucose levels : it stores glucose as glycogen when glucose is plentiful, and it releases glucose when supplies are low. It also uses glucose to generate reducing power for biosynthesis (p. 577) as well as to synthesize a host of biochemicals . So, although the liver has many of the regulatory features of muscle glycolysis, the regulation of glycolysis in the liver is more complex. Phosphofructokinase. Regulation with respect to ATP is the same in the liver as in muscle. Low pH is not a metabolic signal for the liver enzyme, because lactate is not normally produced in the liver. Indeed, as we will see, lactate is converted into glucose in the liver. Glycolysis also furnishes carbon skeletons for biosyntheses, and so a signal indicating whether building blocks are abundant or scarce should also regulate phosphofructokinase. In the liver, phosphofructokinase is inhibited by citrate, an early intermediate in the citric acid cycle (p. 482). A high level of citrate in the cytoplasm means that biosynthetic precursors are abundant, and so there is no need to degrade additional glucose for this purpose. Citrate inhibits phosphofructokinase by enhancing the inhibitory effect of ATP. One means by which glycolysis in the liver responds to changes in blood glucose is through the signal molecule fructose 2,6-bisphosphate (F -2,6-BP), a potent activator of phosphofructokinase (Figure 16.18). In the liver, the concentration of fructose 6-phosphate rises when blood-glucose concentration is high, and the abundance of fructose 6-phosphate accelerates the synthesis of F -2 ,6 -BP (Figure 16.19). Hence, an abundance of fructose 6phosphate leads to a higher concentration of F-2,6-BP. The binding offructose 2,6-bisphosphate increases the affinity of phosphofructokinase for fructose 6· phosphate and diminishes the inhibitory effect of ATP. Glycolysis is thus accelerated when glucose is abundant. Such a process is called feedforward stimulation. We will turn to the synthesis and degradation of this important regulatory molecule after we have considered gluconeogenesis.
Glucose
F-6P
F-2,6-6P activates PFK PFK <JJ-' T his reaction is catalyzed by th e cr. -ketoglutarate dehydrogenase T complex, an organized assemb ly of three kinds of enzymes that is homologous to the pyruvate dehydrogenase complex. In fact, the oxidative decarboxylation of a.-ketoglutarate closely resembles that of pyruvate, also an C( - ketoacid . P y ru vate d e h y urug l:Huse cumplex
Pyruvate + CoA + NAD '
)
acetyl CoA + CO 2 + NADH (I-Ketoglutarate
+ CoA + NAD +
Ci -
K t.'to g Jut",rctLC dclly d r o g enase compJ e."( )
sllccinyl CoA + CO 2 + NADH Both reactions include the decarboxylation of an a -ketoacid and the subsequent formation of a thioester linkage with CoA that has a high transfer potential. The reaction mechanisms are entirely analogous (p . 478).
A Compound with High Phosphoryl-Transfer Potential Is Generated from Succinyl Coenzyme A Succinyl CoA is an energy-rich thioester compound . The f1Go, for the hydrolysisofsuccinyl CoA is about -33.5 kJ mol I (- 8.0 kcal mol- I), which is comparable to that of ATP ( - 30.5 kJ mo] - I, or - 7.3 kcal mol - I) . In the citrate synthase reaction, the cleavage of the thioester bond powers the synthesis of the six-carbon citrate from the four-carbon oxaloacetate and
the two-carbon fragment. The cleavage of the thioester bond of succinyl CoA is coupled to the phosphorylation of a purine nucleoside diphosphate, usually GDP. This reaction is catalyzed by succinyl eoA synthetase (succinate thiokinase) .
486 CHAPTER 17 The Citric Acid Cycle
CoA
coo-
CH 2
ICH2 I
+ Pi + GDP - -
+ CoA + GTP
iH2
Icoo-
coo-
Succinyl CoA
Succinate
This reaction is the only step in the citric acid cycle that directly yields a compound with high phosphoryl-transfer potential. Some mammalian sue· cinyl CoA synthetases are specific for GDP and others for ADP. The E. coli enzyme uses either GDP or ADP as the phosphoryl-group acceptor. We have already seen that GTP is an important component of signal -transduction systems (Chapter 14). Alternatively, its -y phosphoryl group can be read· ily transferred to ADP to form ATP, in a reaction catalyzed by nucleoside diphosphokinase. GTP + ADP
,~'
GDP + ATP
Mechanism: Succinyl Coenzyme A Synthetase Transforms Types of Biochemical Energy
Figure 17.13 Reaction mechanism of succinyl CoA synthetase. The reaction proceeds through a phosphorylated enzyme intermediate. (1) Orthophosphate displaces coenzyme A. w hich generates another energy-rich compound. succinyl phosphate. (2) A histidine residue removes the phosphoryl group with the concomitant generati on of succinate and phosphohistidine. (3) The phosphohistidine residue then swings over to a bound nucleoside diphosphate. and (4) the phosphoryl group is transferred to form the nucleoside triphosphate.
The mechanism of this reaction is a clear example of an energy transformation: energy inherent in the thioester molec ule is transformed into phosphoryl -group -transfer potential (Figure 17.13). T he first step is the displacement of coenzyme A by orthophosphate, which generates another energy-rich compound , succinyl phosphate. A histidine residue plays a key role as a moving arm that detach es the phosphoryl group, then swings over to a bound nucleoside diphosphate and transfers the group to form the nucleoside triphosphate. The participation of high -energy compounds in all the steps is attested to by the fact that the reaction is readily reversible: t.G o , = - 3.4 kJ mol- I (-0 .8 kcal mol- I). The formation of GTP at the expense of succinyl CoA is an example of substrate-level phosphorylation. (00-
Succinyl phosphate
o
o 0 I.. 0 "'c""'- v,-\,>( I ~. COA
,
../ CD
CH 2
>
I CH 2 ICOO-
0
I(H 2 I (H 2 I(00-
2-
-N
~
NH
o 2-
o,'.'\.,:/'
c/
Succinate
--
,
., p .... ~
N + " NH
,
>
Succi"yl CoA
HN
1
"'1: ""
0 2,
N...... /
P "'0
,'/ I
o
GDP
>
GTP
48 7 17.2 Reactions of the Citric Acid Cycle His
CoA
~ Figure 17.14 Structure of succinyl
CoA synthetase. The enzyme is composed of two subunits. The a subunit contains a Rossmann fold that binds t he ADP component of CoA , and the 13 subunit contains a nucleotide-activating region called the ATP-grasp domain. The ATP-grasp domain is shown here binding a molecule of ADP. No tice the histid ine resi due is between the CoA and the ADP. Thi s histidine residue picks up the phosphoryl group from near the CoA and swings over to t ransfer it t o the nucleoti de bo und in the ATP-grasp domain. [Drawn from lCGLpdb.]
Rossmann fold a subunit
AlP grasp ~
subunit
~/
Succinyl CoA synthetase is an a 2132 heterodimer; the functional unit T is one al3 pair. The enzyme mechanism shows that a phosphoryl group is transferred first to succinyl CoA bound in the a subunit and then to a nucleoside diphosphate bound in the 13 subunit. Examination of the time-dimensional structure of succinyl CoA synthetase reveal s that each subunit comprises two domains (Figure 17.14). The amino-terminal domains of the two subunits have different structures, each characteristic of its role in the mechanism . The amino -terminal domain of the a subunit forms a Rossmann fold (p. 449), which binds the ADP component of succinyl CoA. The am ino-terminal domain of the 13 subunit is an ATP-grasp do main, found in many enzymes, which here binds and activates GDP. Succinyl CoA synthetase has evolved by adoptin g these domains and harnessing them to capture the energy associated with succinyl CoA cleavage, which is used to drive the generation of a nucleoside triphosphate.
Oxaloacetate Is Regenerated by the Oxidation of Succinate Reactions of four- carbon compounds constitute the final stage of the citric acid cycle: the regeneration of oxaloacetate. (00-
FAD
FAD H,
H
.....COO-
' C.....
HI-
(
II
IC
-ooc/" "H
(00Succinate
Fumarate
coo-
H, O
\-" ,
HO'H-
IC- IH IC- fjll Icoo-
NAD+
NADH
Malate
The reactions constitute a metabolic motif that we will see again in fatty acid synthesis and degradation as well as in the degradation of some amino acids. A methylene group (CH 2) is converted into a carbonyl group (C 0) in three steps: an oxidation, a hydration, and a second oxidation reaction . Oxaloacetate is thereby regenerated for another round of the cycle, and more energy is extracted in th e form of F ADH2 and NADH . Succinate is oxidized to fumarate by succinate dehydrogenase. T he hydrogen acceptor is FAD rather than AD + , which is used in the other three oxidation reactions in the cycle. FAD is the hydrogen acceptor in this reaction
+ H'
O~
/ coo-
"'c
H-
IC-
H
cooOxaloacetate
488 CHAPTER 17 The Citric Ac id Cycle
because the free -energy ch ange is insufficient to reduce NAD +. FAD is nearly always the electron acceptor in oxidations that remove two hydrogen atoms from a su bstrate. In su ccinate dehydrogenase, the isoalloxazine ring of FAD is covalently attached to a hi stidine side chain of the enzyme (denoted E-FAD ). E- FAD + succinate
OH COO-
H
Fumarate
OH
H
. coo-
H
L-Malate
, 900 kd) consisting of approximately 46 polypeptide chai ns. T hi s proton pump, like that of the other two in the respiratory chain, is encoded by genes residing in both the mitochondria and the nucleus. NADH-Q oxidoreductase is Lshaped, with a horizontal arm lying in the membrane and a vertical arm that projects into the matrix. The reaction catalyzed by this enzyme appears to be NADH + Q + 5 H;';,atrix -~) NAD + + QH z + 4 H ~ytoplasm The initial step is the binding of NADH and the transfer of its two highpotential electrons to the flavin mononucleotide (FMN) prosthetic group of
o
o
511 18.3 The Respiratory Chain
NH
NH
IR OH
H
OH
I
H
Flavin mononucleotide (reduced) (FMNH , )
HI- t - -OH H
o
CH 2 0P0 3 2Flavin mononucleotide (oxidized) (FMN)
Figure 18.8 Oxidation states of flavi ns.
this complex to give the reduced form , FMN H 2 (Figure 18.8). The electron acceptor of FMN, the isoalloxazine ri ng, is identical with t hat of FAD . Electrons are th en transferred from F M N H z to a series of iron- su lfur clusters, the second type of prosthetic group in N A D H -Q oxidoreductase. Fe·S cl usters in iron sulfur proteins (also called non heme iron pro teins ) playa critical role in a wide ran ge of redu ction reaction s in biological systems. Severa l types of Fe-S clusters are known (Figure 18.9). In the simpl est kin d , asingle iron ion is tetrahedrally coordinated to the sulfhydryl grou ps of fo ur cysteine residues of the protein . A second kind , denoted by 2Fe-2S, contains two iron ion s, two inorganic sulfides, and usually four cysteine resid ues . A third type, designated 4Fe-4S, contains fo ur iron ion s, four inorganic sulfides. and four cys teine residu es. NA D H -Q oxidoredu ctase contains both 2Fe-2S and 4Fe-4S clusters. Iron ions in these Fe-S complexes cycle between Fe2+ (reduced ) and F e3+ (oxidi zed ) states . Unlike q uinones and flavin s. iron- sulfur cl usters generall y undergo oxidation- reduction reactions without releasing or b indin g proton s. (AI
(B)
(C)
5
5 5 Cys
Figure 18.9 Iron-sulfur clusters. (A) A single iron io n bo und by fou r cystein e res idues. (8) 2Fe-2S cluster w it h iro n ions bridged by sulfide io ns. (e) 4Fe-4S cluster. Each o f th ese clusters can undergo oxidatio n- reduct ion reacti o ns.
All the redox reaction s take place in the extramembranous portion of NAD H-Q oxidoreductase. Although the d etails of electron tran sfer t hrough this complex remain the su bject of on-going investigation , NADH clearly binds to a site in the extramembranous dom ain. N AD H tran sfers its two electrons to FMN. T hese electrons flow through a series ofFe-S centers and then to coenzyme Q . The flow of two electrons from NADH to coenzyme Q through NADI-l-Q oxidoreductase leads to the pumping of four hydrogen ions out of the matrix of the mitochondrion. In accepting two electrons, Q takes up two protons
512
4W
CHAPTER 18 Ox idative Phosphorylation
Intermembrane space
Q pool
Figure 18.10 Coupled electron-proton transfer reactions through NADH-Q oxidoreductase. Electron s fl ow in Complex I from NADH through FMN and a series of iro n-sulfur clu ster to ubiquinone (Q). The el ectro n fl ow results in the pumping o f fo ur pro ton s and the uptake o f two prot ons fro m the mitochondrial matr ix. [Ba sed o n U. Brandt et al. FEBS Letters 545(200 3):9- 17, Figure 2.]
,
"
,~ NADH
- - ,IFMN
' 'H _
Q
1\
-' Matrix
NAD ' [Fe-5] [Fe-5]
from the matrix as it is reduced to QH 2 (Figure 18.10). The QH 2 leavesthe enzyme for the hydrophobic interior of the m embrane. Ubiquinol Is the Entry Point for Electrons from FADH2 of Flavoproteins
R ecall that F ADH2 is formed in the citric acid cycle, in the oxidation of succinate to fumarate by succinate dehydrogenase (p. 487). This en zyme is part of the succinate-Q reductase complex (Complex II), an integral membrane protein of the inner mitochondrial membrane. FADH2 does not leave the co mpl ex. R ather, its electrons are transferred to Fe-S centers and then to Q for entry into t he electron-transport chain . The succinate-Qreductase com· plex, in co ntrast with NADH-Q oxidoreductase, does not transport pro· tons. Consequentl y, less ATP is form ed from the oxidation ofFADH z than from NADH . Two other enzym es that we will en counter later, glycerol phosphate dehydrogenase (p . 528) and fatty acyl CoA dehydrogenase (p . 624), likewise trans· fer their high -potential electrons from FADH2 to Q to form ubiquinol (QH 2 ) , the reduced state of u biquinone. These enzymes oxidize glycerol and fats, respectively, providing electrons for oxidative phosphorylation, These en zy mes also do not pump protons . Electrons Flow from Ubiquinol to Cytochrome c Through Q-Cytochrome c Oxidoreductase
The second of the three proton pumps in the respiratory chain is Qcytochrome c oxidoreductase (also known as Complex III and as cytochrome reductase). The function of Q-cytochrome c oxidoreductase is to catalyze the transfer of electrons from QH 2 to oxidized cytochrome c (Cyt c), a watersoluble protein, and concomitantly pump protons out of the mitochondrial m atri x. The fl ow of a pair of electrons through this complex leads to the effective net transport of 2 H + to th e cytoplasmic side, half the yield obtained with NADH-Q redu ctase because of a smaller thermodynamic driving force. QH2
+ 2 Cyt c o x + 2 H ~atrix
-~) Q + 2 Cyt Cred + 4 H ~ytoplasm
Q-cytochrome c oxidoreductase itself contains two types of cytochromes, named band CI (Figure 18,1 1), A cytochrome is an electron-transferring protein that contains a heme prosthetic group, The iron ion of a cy tochrome alternates between a red uced ferrous ( + 2) state and an oxidized ferric ( +3) state durin g electron transport. The two cytochrome subunits of Qcytochrom e c oxidoreductase contain a total of three hemes : two hemes,
S13 Rieske iron-sulfur center
Met
.r
18.3 The Respirato ry Chain
Cys
Heme c1
His ~ Figure 18.l1 Structure of
Heme b L
His
His
termed hem e b L (L fo r low affini ty ) and heme bH ( H for high affinity ), within cytochrome b, and one hem e within cytochrome C1' The h em e pros thetic group in cy tochromes b, C1, and c is iron-protoporp hyrin IX , the same heme present in m yoglobin and h emoglobin (p . 184). T hese id entical hemes have different electron affiniti es because they are in d ifferent polypeptide environments. For example, hem e bL , which is located near the cytoplasmic face of the membrane, has lower affinity for an electron than does heme bH , which is near the m atrix side. T h is enzym e is also known as cytochrom e bC1 after its cytochrome groups. In addition to t he hemes, the enzyme contain s an iron- su lfu r protein with an 2fe -2S center. T his center , termed the Rieske center, is unu sual in t hat one of the iron ion s is coordinated by two histidine residues rat her t han two cysteine residues. T his coordination stabi li zes the center in its reduced form, raising its redu ction poten tial so th at it can readil y accept electrons from
QH2 The Q Cycle Funnels Electrons from a Two-Electron Carrier to a One-Electron Carrier and Pumps Protons The mechanism for the cou pling of electron t ransfer fro m Q to cytochrome eto transmembrane proton tran sport is know n as the Q cycle (Figure 18.12 ). Two QH 2 molecules bind to the compl ex consecutively, each giving up two electrons and two H +. These protons are released to the cytoplasmic side of the membrane. The two elect rons t ravel th rough the com plex to d ifferent destinations. O ne electron fl ows first to the Rieske 2Fe-2S clu ster, then to cytochrome c" and f inally to a m olecul e of oxid ized cytoch rom e c, converting
Q-
cytochrome c oxidoreductase (cytochrome bCl)' This enzyme is a ho mod imer w ith 11 distinct po lypeptide chains. Notice t hat the major prostheti c groups. three hemes and a 2Fe-2S cluster. are located either nea r th e cy topla smic edge of the complex bordering the intermembrane space (top) o r in the reg ion embedded in the membrane (u helices represented by vertical tubes). They are well positi oned to mediate the electron -tran sfer rea ctions between qui nones in the membrane and cytochrome c in t he intermembrane space. [Drawn from 1BCC.pdb.]
Figure 18.12 Q cycle. In t he first half of the cycle, t wo electrons o f a bound QH, are t ransf erred, one t o cyt ochrome c and t he o t her t o a bo und Q in a second bi nding sit e to form t he sem iquinone rad ical anio n Q ' - . The new ly formed Q d isso ciates and enters the Q pool. In the second half of the cycle, a second Q H, also gives up its electrons, o ne to a second mo lecule o f cy t ochrome c and the o t her t o reduce Q ' - to QH,. Thi s seco nd electron transf er res ult s in the uptake o f two pro t o ns from the matrix . The path of electro n transfer is shown in red.
Second half of Q cycle
First half of Q cycle
(yt c
2W
Q pool
-->~
Qpool
2 H'
)
Q pool
it in to its redu ced form . The reduced cytochrome c mol ecul e is free to dif· fuse away from the enzyme to continue down the respiratory chain . T he second electron passes through the two heme groups of cytochrome b to an oxidized ubi quin one bound in a second binding site. T his quinone (Q) molecul e is reduced to a semiquinone radical anion (Q . ) by the electron from the first Q H 2 molecul e. O n the addition of the electron fro m the second QH 2 molecul e, this quinone radical anion takes up two p roton s from the matrix side to form QH 2 . Th e remova l of these two protons fro m the matrix contributes to the formation of the proton gradient. In sum, fo ur protons are released on the cyto· plasmic side, and two protons are removed from the mitochondrial matrix. In one Q cycle, two Q H 2 molecules are oxidized to form two Q mole· cules, and then one Q molecule is reduced to QH 2 . Why thi s complexity? The formid abl e problem solved here is to efficiently funn el electrons from a two-electron carrier (QH 2 ) to a one-electron carrier (cytochrome c). The cytochrome b component of the reductase is in essence a recycling device that enables both electrons of QH 2 to be used effectively.
Cytochrome c Oxidase Catalyzes the Reduction of Molecular Oxygen to Water The last of the three proton -pumping assembli es of the respiratory chain is cytochrome c oxidase (Complex IV ) . Cytochrome oxidase catalyzes the transfer of electrons from the reduced form of cytochrome c to molecul ar oxygen, the final acceptor. 4 C yt Cred + 8 H ~atrix + O 2 -~) 4 Cyt c ox + 2 H 2 0 + 4 H ~ytoplasm
~
514
Tyr
T he requirement of oxygen for this reaction is what makes "aerobic" organisms aerobic. To obtain oxygen for this reaction is the reason that human beings must breath . Four electrons are funn eled to O 2 to completel y reduce it to H 20, and , concomitantly, protons are pumped from the m atrix to the cytoplasmic side of the inner mitochondrial membrane. This reaction is quite thermodynamicall y favorable. From the reduction potentials in Table 18.1. the standard free -energy change for this reaction is calculated to be 1 !l Go' = - 231.8 kJ mol- ( - 55 .4 kcal mol- I). A s much of this free energy as possible must be captured in the form of a proton gradient for subsequent use in AT P synthesis. Bovine cytochrome c oxidase is reasonably well understood at the structurallevel (Figure 18. 13). It consists of 13 subunits, of which 3 are encoded by the mitochondrion 's own genome. Cytochrome c oxidase contains two heme A groups and three copper ions, arranged as two copper centers, designated A and B. One center , CuAI CUA, contains two copper ions linked by two bridging cysteine residues, T his center initially accepts electrons from
515 18.3 The Respiratory Chain His
Heme a
His
His
~ Figure 18.13 Structure of cytochrome
c oxidase. This enzyme consists of 13
polypeptide chains. Notice that mos t of the complex , as well as two major prosthetic groups (heme a and heme arCus) are embedded in the membrane (0: helices represented by vertical tubes). Heme oJ- Cue is the site of the reduction of o xygen to water. The (uA/CuA prosthetic group is positio ned near the intermembrane space to better accept electrons from cytochrome c. CO(bb) is a carbonyl group of the peptide backbone. [Drawn fro m 20CC. pdb,]
reduced cytochrome c. The remaining copper ion, CUll, is coordinated by three histidine residues, one of which is modified by covalent linkage to a tyrosine residue . The copper centers alternate between the reduced C u + 2 (cuprous) form and the oxidi zed C u + (cupric) form as they accept and do nate electrons. There are two heme A molecules, called heme a and heme a3, in cytochrome c oxidase. Heme A differs from the heme in cytochrome c and CI in three ways: (1) a formyl group replaces a methyl group, (2) a G I 7 hydrocarbon chain replaces one of the vinyl groups, and (3) the heme is not covalently attached to the protein.
H
o o
----- --H
OH
o
• •,
_
"--' , , c-
, ,
,
o Heme A
5 16 CHAPTER 18 Oxidative Phosphorylation
1. Two m olecul es of cytochrome c sequentially transfer electrons t o reduce CUB and hem e 0 , . 2 Cytochrome c
2. Reduced CUB and Fe in hem e 0 , bind 0 2. which forms a peroxide bridge.
)
2 H2 0
Figure 18.14 Cytochrome ox idase mechanism. The cycle begins and ends w ith all prosthetic gro ups in their ox idized fo rms (shown in blue). Reduced forms are in red. Four cytochro me c mo lecules donate four electro ns. whi ch. in allowing the binding and cleavage of an 0 , molecule. also makes possible the import of four H+ from the matrix t o form two molecules o f H 2 0 . which are released from the enzyme t o regenerate the initial state.
2 Cytochrom e c
(
4. The addition of two more protons leads to the release of water.
3. The addition of two more electrons and two more protons cl eaves the peroxide bridge.
H em e a and heme a3 have di stinct red ox potentials becau se they are located in different environments wit hin cytochrom e c oxidase. An electron fl ows from cytochrome c to e UA/e UA, to hem e a to heme a.1 to C Url, and finally to O 2 , H eme a3 and C UB are directl y adj acent. Together, heme a, and CUB form the active center at which O 2 is reduced to H2 0 . Four molecul es of cytochrome c bind consecutively to the enzyme and tran sfer an electron to reduce one molecul e of O 2 to H 2 0 (Figure 18.14). E lectron s from two molecules of redu ced cytochrome c fl ow down the electron-tran sfer pathway, on e stopping at C UB and the other at heme a )Wit h both centers in the redu ced state, they togeth er can now bind an oxy· gen molecul e. 1.
Peroxide
2. A s molecular oxygen binds, it ab strac ts an electron from each of the nearby ions in the act ive center to form a peroxide (0 2 2- ) bridge between them (Figure 18.15).
3. T wo more m olecules of cytochrom e c bind and release electrons that travel to the active center. The additi on of an electron as well as H + to each oxygen atom reduces the two ion oxygen groups to C Ut.;2+ OH and 3 Fe + OH . Figure 18.15 Peroxide bridge. The oxygen bound t o heme 0 , is reduced t o peroxide by the presence of Cu • .
4. Reacti on with two more H + ion s allows the release of two molecules of H 2 0 and resets th e en zym e to its initial , full y oxidized form.
The four protons in this reaction come exclusively fr om the matrix. T hus , the consumpti on of th ese four protons co ntributes directly to the proton 1 grad ient. Recall that each proton contributes 2Ul kJ m ol- (5.2 kcal mol- I) to th e fr ee en ergy associated with t h e proton grad ient; so these 1 four protons contribute 87.2 kJ m ol- (2 0 .8 kcal mol- I), an am ount sub stan tiall y less than the free energy av ail able from the redu ction of oxygen to water. What is the fat e of thi s missing en ergy? R em ark abl y, cytochrome c oxidase uses this energy to pump f our additional pro tons from the matrix to the cytoplasmic side of the membrane in the course of each reaction cycle for II total of eight protons removed fro m the matrix (figure 18.16). T h e d e tails of how these proto ns are transp orted through the protein is still under stud y. However, two effects contribute to the m echanism . F irst , charge neutrality tends to be maintained in the interior of proteins. T hus, the addition of an electron to a site inside a p rotein tend s to favor the binding of .H + to a n earby site. Second , conformation al changes take place, particul arl y around the h em e a3-C uB center , in the co urse of the reaction cycle. Presumably, in on e conformation, protons m ay enter the protei n excl usively from the matrix side, whereas, in anoth er, th ey may exit ex clusively to the cy toplasmic side. T hu s, the overall process catal yzed by cytochrome c oxidase is
4 eyt Cred + 8 H ~atrix + O 2 Figure 18. 17 sum.marizes the fl ow of electrons from N ADH and FADH2 through the res piratory chain . This series of exergonic reactions is coupled to the pumping of protons from the matrix. A s we will see shortly, the energy inherent in the p roton gradient will be used to synthesize ATP.
Cyt Creduced
cyt Coxidized
4
4
4 H'
0 , ..... 2 H, O
4 H
Pumped protons
4 H'
Chemical protons
Figure 18.16 Proton transport by cytochrome c oxidase. Four protons are taken up from the matrix side to reduce one molecule of 0, to two molecules of H,O. These protons are called "chemical protons" because t hey participate in a clearly def ined react ion w ith 0,. Four additional "pumped" protons are transported out of t he matri x and re leased on the cytoplasmic side in the course of the reaction. The pumped protons double the efficiency o f free -ene rgy storage in the form of a proton gradient for this fina l step in the elect ron -transport chain.
Intermembrane space
III
I
IV II
Matrix
NADH
FADH ,
Q pool
O2
H2 0
Citric acid cycl e
Acetyl CoA
Figure 18.17 The electron-transport chain. High-energy electrons in t he form of NADH and FADH , are generated by the citric acid cycle. These electrons flow through the respiratory chain, which powers proton pumping and results in the reduction of 0 , .
Toxic Derivatives of Molecular Oxygen Such As Superoxide Radical Are Scavenged by Protective Enzymes As discussed earlier, m olecular oxygen is an ideal termin al electron acceptor, because its high affinit y for electrons p rovides a large th ermod ynamic driving force. H owever, danger lurks in the reduction of 0 2' T he transfer of fou r electrons lead s to safe products (two molecules of H 2 0), but partial reduction gen erates h azardo us compound s. In parti cul ar, the transfer of a 517
S18 CHAPTER 18 O xidati ve Phosphorylation
TABLE
18.3 Pathological and other conditions that may entail free -radical injury
Atherogenesis Emphysema: bronchitis Parkinson disease Duchenne muscular dystrophy Cervical cancer
Alcoholic liver disease Diabetes Acute renal fail ure Down syndrome Retrolental fibroplasia (conversion of the retina into a fibrous mass in premature infants) Cerebrovascutar disorders Ischemia: reperfusion injury SOUf ce: After D. B. Marks. A. D. Marks. and C. M. Smith. Basic M(?dicaJ Biochemistry: A Clinical Approach (WIU larns & Wilkins. 1996). p. 331.
single electron to O2 forms superoxide anion, whereas the transfer of two elec· trons yields peroxide.
0 -,
-
Peroxide
Superoxide •
Ion
Dismutation A reacti on in which a single reactant is converted into two djfferent products.
Both compounds are potentiall y destructive. The strategy for the safere· duction of O 2 is clear : the cata lyst does not release partly reduced intermedi· ates. Cytochrome c oxidase meets this crucial criterion by holding O 2 tightly between Fe and Cu ions. Although cytochrome c oxidase and other proteins t hat redu ce O 2 are re· markably successful in not releasing intermediates, small amounts of superoxide anion and hydrogen peroxide are unavoidably formed . Superoxide, hydrogen peroxide, and species that can be generated from them such as OH - are collectively referred to as reactive uxygen species or ROS. Oxidative damage caused by ROS has been implicated in the aging process as well as in a growing Jist of diseases (T able 18.3). What are the cellular defense strategies against oxidative damage by ROS? C hief among them is the en zy me superoxide dismutase. This enzyme scaven ges superoxide radicals by catalyzing the conversion of two of these radicals into hydrogen peroxide and m olecul ar oxygen. S\.lpcroxidc
2 O2~ + 2 H +
Figure 18.18 Superoxide dismutase mechanism. The oxidized form of superoxide dismutase (Mo,) reacts with one superoxide ion to form O 2 and generate the redu ced form of the enzyme (M,.d)' The redu ced form then reacts with a second superoxide and two protons to form hydrogen peroxide and regenerate the ox idized form of the enzyme.
dlsmutase
O 2 + H 20
Z
Eukaryotes contain two forms of this enzyme, a manganese-containing ver· sion located in mitochondria an d a copper- and zinc-dependent cytoplasmic form . These enzymes perform the di smu tation reaction b y a similar mechanism (Figure 18.18). The oxidized form of the enzyme is reduced by super· oxide to form oxygen. The reduced form of the enzy me, formed in this reo action, then reacts with a second superoxide ion to form peroxide, which takes up two protons along the reaction path to yield hydrogen peroxide. T he hydrogen peroxide formed by superoxid e dismutase and by other p rocesses is scavenged by catalase, a ubiquitous heme protein that catalyzes the di smutation of hydrogen peroxide into water and molecular oxygen. Cata lase,
0 < 2 H 22
0
H 0 2 +2 2
Superoxide dismutase and ca talase are remarkably efficient, performing their reactions at or near the diffu sion- limited rate (p . 221). Glutathione peroxidase also plays a role in scavenging H 2 0 2 (p. 587) . Other cellular de· fenses against oxidative damage include the anliuxidant vitamins, vitamins
51 9
Eand C. Because it is lipophilic, vitamin E is especially useful in protecting membranes from lipid peroxidation. One of the long-term benefits of exercise may be to increase the amount of superoxide dismutase in the cell. The elevated aerobic metabolism during exercise causes more ROS to be generated. In response, the cell synthesizes more protective enzymes. The net effect is one of protection, b ecause the increase in superoxide dismutase more effectively protects the cell during periods of rest .
18.3 The Respiratory Chain
Electrons Can Be Transferred Between Groups That Are Not in Contact How are electrons transferred b etween electron -carrying "C 14 C o groups of the respiratory chain ? This question is intriguu y. , ing because these groups are frequently buried in the inte'"'" 12 , \ .... " \ , '" OJ "-u \ , riorof a protein in fixed positions and are therefore not di~ ~ 10 \ " OJ 0 \ ,, rectly in contact with one another. Electrons can move "'OQ C C \ '" Approximate rate through proteins '" .8 Lthrough space, even through a vacuum. However, the rate > \ ,, .\ c::" of electron tran sfer through space falls off rapidly as the \ '" 0 _ \ ~ '" 6 \ ' electron donor and electron acceptor move apart from each U E OJ . _ \ " , \ "'''other, decreasing by a factor of 10 for each increase in sep\ ' 0 4 o \ " 0_ \ Rate through '" aration of 0 .8 A . The protein environment provides more OJ '" \ \ vacuum ' ... , 2 efficient pathways for electron conduction: typically, the '" \ ,, o , _ __ o ______ - L_____ _ ____J __ _ _ _ _ _ rate of electron transfer decreases by a factor of 10 every OQ o o ...J o 5 10 15 20 25 1) A (Figure 18.19). For groups in contact, electronDistance (A) van der transfer reactions can be quite fast, with rates of approxi Waals contact 13 mately 10 s - 1. Within proteins in the electron-transport Figure 18.19 Distance dependence of electron-transfer rate. The chain, electron- carrying groups are typically separated by rate of electron transfer decreases as t he electron donor and the 15 Abeyond their van der Waals contact distance. For electron accept o r move apart. In a vacuum, the rate decreases by such separation s, we expect electron-transfer rates of apa factor of 10 for every increase of 0.8 A. In pro t eins, the rate 4 proximately 10 s - \ (i.e., electron transfer in less than 1 decreases more gradually, by a fa ctor of 10 for eVery increase of ms), assuming that all other factors are optimal. Without 1.7 A. Th is rate is o nly approx imate because variations in the structure of the intervening protein medium can affect the rate. the mediation of the protein, an electron transfer over this distance would take approximately 1 day. The case is more complicated when electrons must be transferred between two distinct proteins, such as when cytochrome c accepts electrons from Complex III or passes them on to Complex IV. A series of hydropho bic interactions bring th e heme groups of cytochrome c and Cl to within Li Aof each other, with the iron atoms separated by 17.4 A. This distan ce 6 could allow cytochrome c reduction at a rate of 8.3 X 10 S - 1. Another important factor in d etermining the rate of electron transfer is the driving force, the free-energy change associated with the reaction (Figure 18. 20) . Like the rates of most reactions, those of electron-transfer reactions tend to increase as the free-energy change for the reaction becomes more favorable. Interestingly, however, each electron-transfer reaction has ~
~
--~
~
~
~~
o
o
~ u '" OJ
Vic::
"",10.-'" ~
- "C ""C
8
6
e~ U .-
-
~ ttl OJ_
- '"
4
o~
OJ"C
- c:: 2 ~ 0
~u
0'" g>:;; - '"
...J
"-
0 L -_____ _ _ _ _ __ _:-L-_ __ _.l.-_ _ ___ 0 .0 1.0 2.0
- tlG (volts)
Figure 18.20 Free-energy dependence of electron-transfer rate. The rat e of an electron-transfer react ion at first increases as the driving force for the reaction increases. The rate reaches a maximum and then decreases at very large dr iving forces.
5 20 CHAPTER 18 Oxidative Phosphorylation
an optimal driving force; making the reaction more favorable beyond this point decreases the rate of electron transfer. This so-called inverted region is important for the light reactions of photosynthesis, to be discussed in Chapter 19 . The Conformation of Cytochrome c Has Remained Essentially Constant for More Than a Billion Years
Cytochrome c is present in all organisms having mitochondrial respiratory chains: plants, animals, and eukaryotic microorganisms. This eJectron carrier evolved more than 1.5 billion years ago, before the divergence of plants and animals. Its fun ction has been conserved throughout this period , as evidenced by the fact that the cytochrome c of any eukaryotic species reacts in vitro with the cytochrome c oxidase of any other species tested thus far. For example, wheat-germ cytochrome c reacts with human CY' tochrome oxidase. Additionally, some prokaryotic cytochromes, such as cytochrome C2 from a photosyn thetic bacterium and cytochrome C550 frama denitrifying bacterium, closely resemble cytochrome c from tuna-heart mitochondria (Figure 18.21). This evidence attests to an efficient evolu tionary solution to electron transfer bestowed by the structural and fun ctional characteristics of cytochrome c_
~
Figure 18.21 Conservation of the "C) three-dimensional structure of cytoch ro me c. The side chains are shown for the 21 conserved ami no acids and the heme. [Drawn from 3CYT.pdb, 3C2C.pdb. and lSSC.pdb.]
Tuna
Rhodospirillum rubrum
Paracoccus denitrificam
Neurospora
Saccharomyces Screwwworm fly Penguin Chicken Pigeon Duck Kangaroo
Rabbit Pig
Donkey~\
Hor~~ ,
Tuna
Snake' I I Jurtle \
Candida
Moth
Dog HumanMon·r.:'e
being
Figure 18.22 Evol utionary tree constructed from sequences of cytochrome c. Branch lengths are proportional to the number o f amino acid changes that are believed to have occurred. This drawing is an adaptation of the work of Walter M . Fitch and Emanuel Margoliash.
The resemblance among cytochrome c molecul es extends to the level of amino acid sequence. Because of the molecule's relatively small size and ubiquity, the amino acid sequences of cytoch rome c from more than 80 widely ranging eukaryotic species have been determined by direct protein sequencing by Emil Sm ith, Emanuel Margoliash, and others. The striking finding is that 21 of 104 residues have been invctrilmt for more than one and Q half billiun years of evolution . A phylogenetic tree, constructed from the amino acid sequences of cytochrome c, reveals the evolutionary relationships between many animal species (Figure 18.22).
18.4
A Proton Gradient Powers the Synthesis Of ATP
Thus far, we have considered the flow of electrons from NADH to 020 an exergol11c process. •
NADH + V2 0 2 + H +
, H 2 0 + NAD+ 8.G OI = - 220.1kJmol - '(-52.6 kcalmon
Next, we consider how this process is coupled to th e synthesis of ATP, an endergonic process. ADP + P j + H +
AT!' + H 20 f1G o' = +30 .5 kJmol - 1 (+7 .3 kcal mOll )
Amolecular assembly in the inner mitochondrial m embrane carries out the synthesis of ATP. This enzyme complex was originally called the mitochondrial ATPase or PjFoATPase because it was discovered through its catalysis of the reverse reaction, the hydrolysis of ATP. ATP synthase, its preferred name, emphasizes its actual role in the mitochondrion. It is also call ed Complex V. How is the oxidation of NADH coupled to the phosphorylation of ADP! Electron transfer was f irst suggested to lead to the formation of a covalent high -energy intermediate that serves as a compound having a high phosphoryl-transfer potential, analogous to the generation of ATP by the formation of 1,3-bisphosphoglycerate in glycolysis. An alternative proposal was that electron transfer aids the formation of an activated protein conformation, which then drives ATP synthesis. The search for such intermediates for several d ecades proved fruitless. In 1961, Peter Mitchell suggested a radically different mechanism, the chelll'iosmotic hypothesis. He proposed that electron transport and ATP syn thesis are coupled by a proton gradient across the inner mitochondrial membrane. In his model, the transfer of electrons through the respiratory chain leads to the pumping of protons from the matrix to the cytoplasmic side of the inner mitochondrial n1.embran e. The H + concentration becom es lower inthe matrix, and an electric field with the matrix side negative is generated (Figure 18.23). Protons then flow back into the matrix to equalize the distribution. Mitchell's idea was that this flow of protons drives the synthesis of ATP by ATP synthase. The energy- rich unequal distribution of protons is cal led the proton-motive force. The proton -motive force can be thought of as being composed of two components: a chemical gradient and a charge gradient. The chemical gradient for protons can be represented as a pH gradient. The charge gradient is created by the positive charge on the unequall y distributed protons forming the chemical gradient . Mitchell proposed that both components power the synthesis of AT P.
Protons are pumped across this membran e as electrons flow through th e respiratory chain.
High [WI
+
-
+ + +
-
Inner mitochondrial membrane
++
-
-
Outer mitochondrial membrane
-
-
Low [WI
- + -
+
c...---- Intermembrane space '---- Matrix
+ +
Figure 18.23 Chemiosmotic hypothesis. Electron t ransfer thro ugh the respiratory chain leads to t he pumping of protons from t he mat rix t o the cytoplasmic side of the inner mitochondrial membrane. The pH gradient and membrane potential co nstitute a proto nmotive force that is used to drive ATP synthesis.
521
- - - -18.4 ATP Synthesis
522 CHAPTER 18 Oxidative Phosphorylation
Figure 18.24 Testing the chemiosmotic hypothesis. AT? is synthesized when reconstituted membrane vesicles containing bacteriorho do psin (a lightdriven proto n pump) and AT? synthase are illuminated . The o ri entation of ATP synthase in this reconstituted m embrane is the reverse o f that in the m itochondrion.
Bacteriorhodopsin in synthetic vesicle
Mitochondrial ATPase
•
ADP + Pi ATP
•
Proton-motive force (~p) = chemical gradient (~pH) + charge gradient (llo/)
~
-. ..-, • .. •
~
~
,"- -
"
•
c nng
F,
Mitchell's hi ghly innovative hypothesis that oxidation and phosphoryla· tion are coupled by a proton gradient is now supported by a wealth of evidence. Indeed, electron transport does generate a proton gradient across the inner mitochondrial membrane. The pH ou tside is 1.4 units lower than inside, and the membrane potential is 0.14 V, the outside being positive. As calculated on page 508, this membrane potential corresponds to a free en· ergy of 21.8 kJ (5. 2 kcal) per mole of protons. An artificial system was created to elegantly demonstrate the basic prin· ciple of the chemiosmotic hypothesis. The role of the respiratory chain was played by bacteriorhodopsin. This purple membrane protein from halobac· teria pumps protons when illuminated . Synthetic vesicles containing bacle· riorhodopsin and mitochondrial ATP synthase purified from beef heart were created (Figure 18.24). When the vesicles were exposed to light, ATP was formed. This key experiment clearly showed that the respiratory chain and ATP synthase are biochemically separate systems, linked only by a proton· motive force.
ATP Synthase Is Composed of a Proton-Conducting Unit and a Catalytic Unit
~ Figure 18.25 Structure of ATP
synt hase. A schematic structure is shown along w ith representati ons of the components for whi ch structures have been determined to high resolution. The ?-Ioop NTPase domains of the Ct and i3 subunits are indicated by purple shading. Notice that part of the enzyme complex is embedded in the inner mitochondrial membrane, whereas the remainder resides in the matrix. [Drawn from lE79.pdb and lCOvpdb.]
Biochemical, electron microscopic, and crystallographic studies of ATP synthase have revealed many details of its structure (Figure 18.25). It isa large, complex enzyme that looks like a ball on a stick. Much of the "stick" part, called the Fa subunit, is embedded in the inner mitochondrial memo o brane. The 85-A-diameter ball, called the F I subunit, protrudes into the mitochondrial matrix. The FJ subunit contain s the catalytic activity oflhe synthase. In fact, isolated F I subunits display ATPase activity. The F I subunit consists of five types of polypeptide chains (ex 1, 131, -y, 8, and e) with the indicated stoichiometry. The ex and 13 subunits, which make up the bulk of the F J, are arranged alternately in a hexameric ring; they are homologous to one another and are members of the P -Ioop NTPase family (p. 267). Both bind nucleotides but only the 13 subunits participate directly in catalysis. Beginning just below the ex and 13 subunits is a central stalk can· sisting of the "{ and e proteins. The "{ subunit includes a long helical coiled coil (p. 45) that extends into the center of the ex3133 hexamer. The 'Y subunit breahs the symmetry of the (X3f3 J hexamer: each of the f3 subunits is distinct by virtue of its interaction with a different face of y. Distinguishing the three ~ subunits is crucial for understanding the mechanism of ATP synthesis.
The F o subunit is a hydrophobic segment that spans the inner m ito chondrial membrane. F a contains the proton channel of the comp lex. T his channel consists of a ring comp rising from 10 to 14 c subun its that are embedded in the membrane. A single a subunit bi nds to the outside of the ring, The F 0 and F I subunits are connected in two ways : by the central -yf:: stalk and by an exterior column. The exterior column consists of one a subunit, two b subunits, and the /) su b unit. As will be discussed shortl y, we can think of the enzyme as consisting of a moving part and a stationary part: (1) the moving unit, or rotor, consists of th e c ring and the -yf:: stalk and (2) the stationary unit, or stator, is composed of the remainder of the m olecule.
523 18.4 ATP Sy nthesis
Proton Flow Through ATP Synthase Leads to the Release of Tightly Bound ATP: The Binding-Change Mechanism
ATP synthase catalyzes the formatio n of ATP from ADP and orthophosp hate.
The actual substrates are Mg2+ complexes of ADP and ATP, as in all known phosphoryl-transfer reactions with these nucleotides . A terminal oxygen atom of ADP attacks the phosphorus atom of P i to form a pentacovalent intermediate, which then dissociates into AT P and H 2 0 (Figure 18.26) . The attacking oxygen atom of ADP and th e departing oxygen atom of Pi occupy the apices of a trigonal bipyramid .
+ H2 0
ADP
AlP
Pentacovalent inte rmed iate
P;
Figure 18.26 ATP-synthesis mechanism. O ne o f t he o xygen at o m s of ADP attacks t he phosphorus at om of P; t o fo rm a pent acovalent interm ediate, which t hen forms ATP and releases a molecule o f H 2 0 .
How does the flow of protons drive the synthesis of ATP? Isotopicexchange experiments unexpectedly revealed th at enzyme- bound A TP f orms readily in the absence of a proton-motive force. When A DP and Pi were added 18 18 to AT P synthase in H 2 0, 0 became incorporated into P i through the synthesis of ATP and its subsequent hydrolysis (Figure 18 .27) . T he rate of 18 incorporation of 0 into P i showed that about equal amounts of bound ATP and A DP are in equilibrium at the catalytic site, even in the absence of a proton gradient . H owever , A TP does not leave th e catalytic site u nless
0
-
.0 0
\../
2
o - ,0
,0 -
\:!
"'==='====
R" O/ P" O/ p",·o ADP
P;
\,,,:/
Q
-,9
\\,~
Q
,0 2-
\/1
R"O/P"O/p"o/p~o
,
AlP
Figure 18.27 ATP forms without a proton-motive force but is not released. The results o f isotopic-exchange experiments ind icate th at enzyme-bound ATP is form ed from ADP and P; in the absence of a p roton-mo ti ve force.
ADP
"O-Iabeled P;
ATP
ATP
ATP ATP
jt ADP + Pi
T Figure 18.28 ATP synthase nucleotidebinding sites are not equivalent. The 'I subunit passes through the center of the "'3133 hexamer and makes the nucleoti d ebindi ng sites in t he 13 subunits dist inct from one another, N ote that each '" subun it conta ins bound ATP, but th ese nucleo tides do no t participate in any react ions.
Progressive alteration of the forms of the three active sites of ATP synthase Subunit 1 L --+ T --+ 0 --+ L --+ T ---+ 0 ....... .. Subunit 2 0 --+ L
T --+ 0 --> L ..... T.........
SubUnit 3 T ..... 0 --+ L ..... T ..... 0 ..... L...... .
o
L
protons flow through the enzyme. Thus, the role of the proton gradient is nol to form ATP but to release it from the synthase. On the basis of these observations and others, Paul Boyer proposed a binding-change mechanism for proton-driven ATP synthesis. This proposal states that a 13 subunit can perform each of three sequential steps in the function of ATP sy nthesis by changing co nformation. These steps are (1) ADP and Pi bi nding, (2) ATP synthesis, and (3) ATP release. The concepts of this initial proposal refined by more-recent crystallographic and other data yield a sati sfying m echanism for ATP synthesis, As already noted, in· teractions with the 'Y subunit make the three 13 subunits unequivalent (Figure 18.28). One 13 subunit can be in the L, or loose, conform ation, This conformation binds ADP and Pi , A second subunit can be in the T, or tight, conformation. This conformation binds ATP with great avidity, so much so that it will convert bound ADP and Pi into ATP. Both the T and L confor· mations are sufficiently constrained that they cannot release bound nu· cleotides. The final subunit will be in th e 0 , or open, form . T his form can exist with a bound nucleotide in a structure that is similar to those of theT and L for ms, but it can also convert to form a more open conformation and release a bound nucleotide. The rotation of th e 'Y subunit drives the interconversion of these three forms (Figure 18.29). ADP and P i bound in the subunit in the T form are transiently combining to form ATP. Suppose that the 'Y subunit is rotated by 120 degrees in a counterclockwise direction (as viewed from the top). This rotation converts the T -form site into an O-form site with the nucleotide bound as ATP. Concomitantly, the L-form site is converted into a T -form site, enabling the transformation of an additional ADP and Pi into ATP. The ATP in the O-form site can now depart from the enzy me to be replaced by ADP and Pi ' An additional 120-degree rotation converts this O-form site into an L-form site, trapping these substrates . Each subunit progresses from the T to the 0 to the L form with no two subunits ever present in the same conformational form. T his mechanism suggests that ATP can be synthesized and released by driving t he rotation of the 'Y sub· uni t in the appropriate directi on.
L
T
L
L
120 0 rotation
of y (CCW)
/
) ATP
)
ADP
+
ATP
p. I
T Figure 18.29 Binding-change mechanism fo r ATP synthase. The rotat ion of the 'I subunit interconvert s t he three 13 subunits, The subunit in the T (tight) form int erconverts ADP and P, and ATP but does no t allow ATP be relea sed. When the 'I subun it is rotated by 120 degrees in a counterclockwise (CCW) direction, the T-fo rm subun it is convert ed into the 0 form, all ow ing ATP release. ADP and P, can then bind t o the O -fo rm subunit. An additio na l 120-degree rotati o n (not sho wn) traps th ese substrates in an L-form subunit.
524
o
o
o
Rot at ional Catalysis Is t he World's Smallest Molecula r Motor Is it possible to observe the proposed rotation directly? Elegant experiments have demonstrated the rotation through the use of a simple experimental system consisting solely of cloned <X3133'Y subunits (Figure 18.30). The 13 subunits were engineered to contain amino-terminal poly histidine tags, which have a hi gh affin.i ty for nickel ions. T hi s property of the tags allowed the <X 3 ~J assembly to be immobilized on a glass surface that had been coated with nickel ions. The 'Y subunit was linked to a fluorescently labeled actin filament to provide a long segment that could be observed under a fluorescence microscope. Remarkably, the ad dition of ATP caused the actin filament to
S25 18.4 ATP Synthesis Actin fi lament
, ex
.
p
ADP + Pi
Figure 18.30 Direct observation of ATPdriven rotation in ATP synthase. The <X l [3l hexamer of ATP synthase is fixed to a surface. with the '1 subunit projecting upward and linked t o a fluo rescently labeled actin filament. The addition and subsequent hydrolysis of ATP result in the counterclockw ise rotation o f the '1 subunit. which can be directly seen under a fluorescence microscope.
rotate unidirectionally in a counterclockwise direction. The 'Y subunit was rotating, driven by the hydrolysis of ATP Thus, the catalytic activity of an individ ual molecule could be observed. The counterclockwise rotation is consistent with the predicted mechanism for hydrolysis because t he molecu le was viewed from below relative to the view shown in Figure 18. 30. More-detailed analysis in the presence of lower concentrations of A TP revealed that the "Y subunit rotates in 120-degree increments. Each increment corresponds to the hydrolysis of a single ATP molecule. In addition, from the results obtained by varying the length of t he actin filament and measuring the rate of rotation, the enzyme appears to operate near 100% efficiency; that is, essentially all of the energy released by ATP hydrolysis is converted into rotational motion . Proton Flow Around the C Ring Po wers ATP Synthes is The direct observation of rotary motion of the "Y subunit is strong evidence for the rotational mechanism for ATP synthesis. The last remaining question is: How does proton flow through F o drive the rotation of the"y sub unit? Howard Berg and George Oster proposed an elegant mechanism that provides a clear answer to this question . The mecharrism depends on the structures of the a and c subunits of F 0 (Figure tiU 1). T he a subunit di rectly abuts the membrane-spanning ring formed by 10 to 14 c subunits. Although the structure of the a subunit has not yet been experimentally d etermined, a variety of evidence is consistent with a structure that includes two hydrophilic half- channels that do not span the membrane (see Figure 18.31 ). Thus, protons can pass into either of these channels, but they cannot move completely across the m embrane. The a subunit is positioned such that each half-channel directly interacts with one c subunit. The structure of the c subunit was determined both by NMR methods and by x- ray crystallography. Each polypeptide chain forms a pair of a he lices that span the membrane . An aspart ic acid residue (Asp 61 ) is found in the middle of one of the helices. When the Asp 61 residues of the two c subunits are in contact with the hydrophilic environment of a half-channel, they can give up their protons so that they are in the charged aspartate form (Figure 18.32). The key to proton movement across the membrane is that, in a proton-rich environment, such as the cytoplasmic side of the mitochondrial membrane, a proton will enter a channel and bind the aspartate re idue. The subunit with the bound proton then rotates through the membrane until the aspartic acid is in a proton-poor environment of the other half channel, where the proton is released . The movement of protons through Ihe half-channels from the high proton concentration of the cytoplasm to the low proton concentration of the matrix powers the rotation of the c ring. Its rotation is favored by the ability of the newly protonated (neutralized)
.~
Aspartic acid
Subu nit c
Cytoplasmic half-channel
Matrix half-channel Subunit a Figure 18.31 Components of the protonconducting unit of ATP synthase. The c subunit consists of two <X helices that span the membrane. An aspartic acid residue in one of the helices lies on the center of the membrane. The structure of the a subunit has not yet been directly observed. but it appears to include two half-channels that allow protons to enter and pass partway but not completely through the membrane.
H+ H+
H+
W
W
H' H+ W
Intermembrane space W
H+
H+
H+ H+ H+ H'" W H' H+ W W H' W H' W
H' W
H' H+
H' W H'
H'
W
Cannot rotate in either direction
(
H+ W H' H W H' W W
:>
)
Matrix
H+
W
H'
Can rotate clockwise
(
Figure 18.32 Proton motion across the membrane drives rotation of the c ring. A proton enters from the intermembrane space into the cytoplasm ic half-channel t o neutralize the charge on an aspartate residue in a c subunit. With this charge neutralized, the c ring can rotate clockwise by one c subunit, moving an asparti c aci d residue o ut of the membrane into the matrix half-c hannel. This proton can move into the matrix, resetting the system t o its initial state.
Figure 18.33 Proton path through the membrane. Each proton enters the cytoplasmic half-channel, f ollows a complete ro tation of the c ring, and exits t hroug h the other half-channel into the matrix .
A little goes a long way Despite the various molecular machinations
and the vast numbers of ATPs syntheSized and protons pumped, a resting human bei ng requires surpri singly little power. Approximately
116 watts, the energy output of a typi cal light bulb, provides enough energy to sustain a resting person.
526
aspartic acid residue to occupy the hydrophobic environment of the membrane, Thus, the c subunit with the newly protonated aspartic acid moves from contact with the cytoplasmic half-channel into the membrane, and the other c subunits move in uni son. Each proton that enters the cytoplasmic half-channel moves through the membrane by riding around on the rotating c ring to exit through the matrix half-chanl1el into the proton-poor environment of the matrix (Figure 18,33). How does the rotation of the c ring lead to the synthesis of ATP? The c ring is tightly linked to the 'Y and c subunits. Thus, as the c ring turns, these subunits are turned inside the <X 3 f3 3 hexamer unit of Fl' The rotation of the 'Y subunit in turn promotes the synthesis of ATP through the bindingchange mechanism. The exterior column formed by the two b chains and the /) subunit prevents the <X 3 f3 3 hexamer from rotating. Recall that the number of c subunits in the c ring appears to range between 10 and 14. This number is significant because it determines the number of protons that must be transported to generate a molecule of ATP. Each 360-degree rotation of the 'Y su bunit leads to the synthesis and release of three molecules of ATP. Thus, if there are 10 c subunits in the ring (as was observed in a crystal structure of yeast mitochondrial ATP synthase), each ATP generated requires the transport of 10 /3 = 3.33 protons. For simplicity, we will assllme that three protons must flow into the matrix for each ATP formed, but we must keep in mind that the true value may differ. As we will see, the electrons from NADH pump enough protons to generate 2.5 molecules of ATP, whereas those from FADH2 yield 1.5 molecules of ATP. Let us return for a moment to the example with which we began this chapter. If a resting human being requires 85 kg of ATP per day for bodily
527 18.5 Mitochondrial Shuttles ADP + P;
AlP
AlP synth ase
Matrix
Intermembrane space Proton-motive force
I
I I I I
III
II
I I I I I I I I
I I I I
IV
° 2
H2O
Electron-transpo rt chain
Figure 18.34 Overview of oxidative phosphorylation. The electron-t ransport chain generates a prot o n gradient, w hich is used t o synthesize ATP.
25
fun ctions, then 3.3 X 10 protons must fl ow through the ATP synthase 21 per day, or 3.3 X 10 proto ns per second . Figure 18.34 summari zes the process of oxidative phosphorylation.
AlP Synthase and G Proteins Have Se ve ral Common Features
->:19- 44 . Klin genberg, M ., and Huan g, S. G . 1999. Structu re and function of the un coupling protein from b rown adipose tissue. Biochim. Biophys. Acta 141 3:271- 296. N icholls, D . G ., and Ri al, E. 1999. A history of the fi rst uncoupling protein , UCP1 . j. Bioenerl(. Biometobr. 31 :399- 406. Ricquier, 0., and Bouillaud , F. 2000 . The uncou pling protein homo logues: C l'l , UC P2, UC P3 , StUCP and AtUC P. lJiochem . j. 345 :16 1 179 . Walker, j. E. 1992. The mitochondria l transporter family. Cllrr. Opin . Struct. BioI. 2:5 19 526. Klingenberg. M . 1992. Structure -functi on of th e ADP / AT P carrier. Biochem. Soc. Tran. Dihydroxyacetone phosphate 3-phosphate ' NADP+ NADPI I
2 1,3·Bisphosphoglycerate ADP
ATP
2 3-Phosphoglycerate Figure 20.9 Hexose phosphate format ion. 3-Phosphogl ycerate is converted into f ructose 6-phos phate in a pathway parall el t o that o f gluconeogenesis.
glycolate oxidase, an enzyme with a flavin mononucleotide prosthetic group. The H zO z produced in this reaction is cleaved by catalase to H20 and O 2 , Transamination of glyoxylate then yields glycine . Two glycine molecules can unite to form serine, a potential precursor of glucose, with the release of CO 2 and ammonium ion (NH4 +). The ammon ium ion, used in the synthesis of nitrogen -containing compounds, is salvaged by a gluta· mine synthetase reaction (see Figure 20.7 and p. 661 ). This salvage pathway serves to recycle three of the four carbon atoms of two molecules of glycolate. However, one carbon atom is lost as CO 2 , This process is called photorespiration because O 2 is consumed and CO 2 is reo leased. Photorespiration is wasteful because organic carbon is converted into CO 2 without the production of ATP, NADPH, or another energy-rich metabolite. Evolutionary processes have presumably enhanced the prefer· ence of rubisco for carboxylation. For instance, the rubisco of higher plants is eightfold as specific for carboxylation as that of photosynthetic bacteria. However, some oxygenase activity may be an inevitable side effect of the carboxylase reaction mechanism.
Hexose Phosphates Are Made f rom Phosphoglycerate, and Ribulose 1,5-bisphosphate Is Regenerated The 3-phosphoglycerate product of rubisco is next converted into fru ctose 6-phosphate, which readily interconverts between its isomers glucose 1· phosphate and glucose 6-phosphate (pp. 450 and 595). The steps in this conversion (Figure 20 .9) are like those of the gluconeogenic pathway (p. 458), except that glyceraldehyde 3-phosphate dehydrogenase in chloroplasts, which generates glyceraldehyde 3-phosphate (GAP ), is specific for NADPH rather than NADH. These reactions and that catalyzed by rubisco bring CO 2 to the level of a hexose, converting CO 2 into a chemical fuel at the expense of NAD PH and A TP generated from the light reactions. The third phase of the Calvin cycle is the regeneration of ribulose 1,5· bisphosphate, the acceptor of CO 2 in the first step. The problem is to con· struct a five-carbon sugar from six -carbon and three- carbon sugars. A transketolase and an aldolase play the major role in the rearrangement of the carbon atoms. The transketnlase, which we will see again in the pentose phosphate pathway, requi res t he coenzyme thiamine pyrophosphate (TPP) to transfer a two -carbon unit (CO CHzOH) from a ketose to an aldose.
HO
C
Transketolase
+
H
+ HO
;, : : : :
R'
R Ketose
Cn carbons)
R
Aldose Cm carbons)
Cn
C
H
R'
Aldose - 2 carbons)
Ketose
Cm + 2 carbons)
Aldolase, which we have already encountered in glycolysis (p. 438), catalyzes an aldol condensation between dihydroxyacetone phosphate (DHAP) and an aldehyde. This enzyme is highly specific for dihydroxyace· tone phosphate, but it accepts a wide variety of aldehydes.
O~C/ H R
/
+ 0
H 2 OPO/ -
,
C \
Aldo lase
HO
C
H
H
C
OH
•
H 2 OH
R Aldose Cn carbons)
Dihydroxyacetone
Ketose
phosphate
(n + 3 carbons)
571 O~ / H
0"", /
(
HO
(
H
(
I( - OH I( - OH I( H, OP0 ' -
H Tran sketolase .
OH
H-
•
HH- (
OH
HOI-
+
I
1-
H- (
(H, OH
20.1 Th e (a lvin ( ycl e
H OH
3
CH, OP0 3 ' Fructose 6-phosphate
Glyceraldehyde 3-phosphate
Erythrose 4-phosphate
Xylulose S-phosphate
o
o C
o~ , H
H H-
IC OH IC- OH ICH, OPO,2-
C
HO
C
+ 0 - (
,
Aldolase
(
H-
HO
H
(-
H2 0
OH
H-
(-
OH OH
1,7-bisphosphate phosphata se
(H , OH H-
( :-
(H 20PO/ Erythrose 4-phosphate
phosphate
C HO H
I (
0 H
0 (
OH
+ H H
C
(
OH
OH CH 20PO,l
H
C
OH
0
•
Tran sketolase ,
CH20H C
H
(
OH
H
(
OH
H
(
OH
HO
(
+
H
,
H
(
OH
CH , OPO,'
(H 20PO,2-
CH, OPO, 2Sedoheptulose 7-phosphate
/H (
H
Glyceraldehyde 3-phosphate
Ribose
S-phosphate
(-
OH
•
HI-
I( I( -
'OH
H-
OH
Sedoheptulose 7-phosphate
(H 20H
0
HI-
(H , OPO,'-
Sedoheptulose 1,7-bisphosphate
Dihydroxyacetone
H
P
\ .J Sedoheptulose
,
(
Xylulose S-phosphate
figure 20,10 Formation of five-carbon sugars. First, t ra nsketolase convert s a six-carbon sugar and a t hree-carbon sugar into a f our- carbon sugar and a five-carbon sugar. Then, aldolase combi nes t he four-carbon product and a three-carbon sugar to fo rm a sevencarbon sugar, Fina lly, t his seven-carbo n sugar react s wi th ano t her three-c arbon sugar to form two add itional five-carbon sugars.
With these en zym es, the con struction of the five-carbon sugar proceeds as shown in F igure 20.10. Finally, ribose 5- phosphate is converted in to ri b ul ose 5-phosphate by phosphopentose isomerase while xylulose S-phosphate is converted into ribu lose 5-phosphate by phosphopentose epimerase, Ribul ose S-p hosphate is con verted in to ribul ose 1,S-bi sphosphate t h rou gh the action of phosphoribulose kinase (Figure 20,11 ). The sum of these reaction s is Fructose 6- phosphate + 2 glyceraldehyde 3- phos phate + dihydroxyacetone phosphate + :1 A TP ) :1 ribulose 1 ,S-bisphosphate + 3 ADP This seri es of reaction s compl etes the Calvin cycle (Figure 20 .12). The sum of all t he reactions results in the generation of a hexose and the regen eration of the starting compound , ribulose 1, 5- bisphosphate , Thus, onl y a limited amount of ribu lose 1,5 -bisphosphate is need ed to incorporate man y molecules of CO 2 into hexoses .
572
O" " , /H
C
CHAPTER 20 The Cal vin Cycle and the Pentose Phosphate Pathway
H
C
OH
Phosp hope nlose H
C
OH
H
C
OH
•
AlP
CH 2 0P0 3 2 -
Ribose s-phosphate
H Phosphopentose • ep, merase
C
ADP
-::----'\:---:;/ ---;---") HPhosp horibulose ki nase H-
OH
CH 2 0P03
C
H
H
C
OH
OH
C
OH
ICH
2-
2 0P0 3 ' -
Ribulose 1,s-bisphosphate
Ribulose
s -phosphate HO
C:-
CH 2 0 P0 3 2-
Xylulose s-phosphate Figure 20.11 Regeneration of ribulose 1,S-bisphosphate. Both ribose s-phosphate and xylulose s-phosphate are converted into ribul ose s-phosphate, w hich is t hen phosphorylated to complete the regeneration of ribulose 1,S-bisphosphate.
Ribulose S-phosphate
..---- 3ATP
3 ADP Ribose 5-phosphate
Xylulose S-phosphate
Ribulose 1,s-bisphosphate
3 CO 2
GAP
Sedoheptu lose 7-phosphate p., ' - - - H 20 Sedoheptulose 1,7-bisphosphate
Figure 20.12 Calvin cycle. The diagram shows the reactions necessa ry wi t h the correct st o ichiometry to convert t hree molecules of CO 2 into one molecule of d ihydroxyacet one phosphate (DHAP). The cycle is not as simp le as presented in Figure 20.1; rathe r, it ent ail s many reactions that lead ultimately to the synthesis of glucose and the regeneration of ribu lose 1.s-bisphosphat e. [After J. R. Bowyer and R. C. Leegood. "Photosynthesis," in Plant Biochemistry, P. M. Dey and J. B. Harborne, Ed s. (Academic Press, 1997), p. 85.]
DHAP
Erythrose 4-phosphate
GAP
.r- 6ATP Xylulose 5-phosphate
6 ADP
v-
( Fructose 6-phosphat e )
,
6NAOPH
p.
6 NADP'
1-- H2 0
6P I
Fructose 1,6-bisphosphate
DHAP
GAP
DHAP
Three ATP and Two NADPH Molecules Are Used to Bring Carbon Dioxide to the Level of a Hexose
What is the energy expenditure for synthesizing a hexose? Six rounds of the Calvin cycle are required, because one carbon atom is reduced in each round. Twelve molecules of ATP are expended in phosphorylating 12 molecules of 3- phosphoglycerate to 1,3 -bisphosphoglycerate, and 12 molecules
S73
ofNADPH are consumed in reducing 12 m olecules of 1,3 -bisphosphoglycerate to glyceraldehyde 3-phosphate. An additional six molecules of ATP are spent in regenerating ribulose 1,S -bisphosphate. We can now write a balanced equation for the net reaction of the Calvin cycle:
6C0 2
12NADPH + 12 H 2 0 ) Cr,H I 2 0 6 + 18ADP + 18P j + 12NADP +
20.1 The Calvin Cycle
+ 18ATP +
+
6H +
Thus, three molecul es of ATP and two m olecules of NADPH are consumed in incorporating a sin gle CO 2 molecule into a hexose such as glucose or fructose.
Starch and Sucro se Are the Major Carbohydrate Stores in Plants Plants contain two major storage forms of sugar : starch and sucrose. Starch , like its an imal counterpart glycogen , is a polymer of glucose residues, but it is less branched than glycogen because it contains a smaller proportion of (X1,6 -glycosid ic linkages (p. 311). Another differen ce is that ADP -glucose, not UDP-glucose, is the acti vated precursor. Starch is synth esized and stored in chloroplasts. In contrast, sucrose (common table sugar), a disaccharide, is synthesized in the cytoplasm. P lants lack the abi lity to transport hexose phosphates across the chloroplast m embrane, but they are able to transport triose phosphates from chloroplasts to the cytoplasm . T riose phosphate intermediates such as glyceraldehyde 3-phosphate cross into the cytoplasm in exchange for phosphate through the action of an abundant phosphate translocator. Fructose 6-phosphate formed from triose phosphates joins the glucose unit ofU DP-glucose to form sucrose 6-phosphate (F igure 20. 13). The hydrolysis of the phosphate ester yields sucrose, a readily transportable and mobili zable sugar that is stored in many plant cells, as in sugar beets and sugar cane.
o I
Triose phosphates (from chloroplasts)
HN
, HO
rOH
OH OH Fructose 6-phosphate
UDP-glucose
Sucrose 6-phosphate synthase
o CH 2 0H o HOH 2 C
OH
HN 0 HO
2-
+ O'~p/ O'-...p/O CH 2 0P0 3 2-
OH
;.\ ;:\ d '0 d '0
N/
0 0 ......
-
OH OH Sucrose 6-phosphate
UDP
figure 20.13 Synthesis of sucrose. Sucrose 6-phosphate is formed by the reaction between fructose 6-phosphate and the activated intermediate urid ine diphosphate glucose !UDP-glucose).
20.2 /' Mg2+
"
':~____~Th~Y~la~kO~id~____~/ Stroma
DARK
The Activ ity of the Calvin Cycle Depends on Environmental Conditions
The C alvin cycle operates during the day, whereas carbohydrate degradation to yield energy takes place primarily at night. How are synthesis ane degradation coordinately controlled? The light reactions lead to changes in the stroma namely, an increase in pH and in Mg2+, N ADPH, and reo duced ferredoxin con centration all of which contribute to the activation of certain Calvin cycle en zymes (Figure 20. 14).
NADPH Fdred
Rubisco Is Activated by Light-Driven Changes in Proton and Magnesium Ion Concentrations
/
LIGHT Figure 20.14 Light regulation of the Calvin cycle. The light reactions of photosynthesis transfer electrons out of the thylakoid lumen into the stroma and they transfer proton, from the stroma into the thylakoid lumen. As a consequence of these processes. the concentrat ions of NADPH. reduced ferredoxin (Fd). and MgH in the stroma are higher in the light than in the dark. whereas the concentration of H+ is lower in the dark. Each of these concentration changes helps couple the Ca lvin cycle reactions to the light reactions.
As stated earlier, the rate-limiting step in the Calvin cycle is the carboxylation of ribulose 1,5- bisphosphate to form two molecules of 3-phosphoglycerate. The activity of rubisco increases markedly on illumination because light facilitates the carbamate f ormation necessary to enzyme activity. In th e stroma, the pH increases from 7 to R, and the level of Mg2+ rises. Both effects are consequ ences of the light-driven pumping of protons into the thylakoid space. Mg2-1- ions from the thylakoid sp ace are released into the stroma to com· pensate for the influx of protons. Carbamate formation is favored at alkaline pH. CO 2 adds to a deprotonated from of lysine 20 1 of rubisco, and Mg2t ion binds to the carbamate to generate the active form of the enzyme. Thus, light leads to the generation of regulatory signals as well as ATP and NADPH. Thioredoxin Plays a Key Role in Regulating the Calvin Cycle
Light-driven reactions lead to electron transfer from water to ferredoxin and, eventually, to NADPH. The presence of reduced ferredoxin and NADPH are good signal s that conditions are right for biosynthesis. One way in which this information is conveyed to biosynthetic enzymes is by thioredoxin, a 12kd protein containing neighboring cysteine residues that cycle between are· duced sulfhydryl and an oxidized disulfide form (Figure 20.15). The reduced form of thioredoxin activates many biosynthetic enzymes by reducing disul· fide bridges that control their activity and inhibits several degradative enzymes b y the same means (Table 20.1 ). In chloroplasts, oxidized thioredoxin is reduced by ferredoxin in a reaction catalyzed by f erredoxin- thioredoxin reductase. This enzyme contains a 4Fe-4S cluster that couples two one-electron oxidations of reduced ferredoxin to the two-electron reduction of thioredoxin. Thus, the activities of the light and dark reactions of photosynthesis are coordinated through electron transfer from reduced ferredoxin to thioredoxin and
Disulfide bond
-"'" Figure 20.15 Thioredoxin. The IV oxid ized form of t hioredoxin contains a disu lfide bond. When thioredoxin is reduced by reduced ferredoxin, the disulfide bond is converted into t wo free sulfhydryl groups. Red uced thioredoxin can c leave disulfide bonds in enzymes, activating certain Calvin cycle enzymes and inactivating some degradative enzymes. [Drawn fro m 1F9M. pdb.l
574
TABLE 20.1 Enzymes regulated by thioredoxin
575 20.2 Control of the Calvin Cycle
Enzyme
Pathway
Rubisco
Carbon fixat ion in the Calvin cycle Gluconeogenesis Calv in cycle, glu coneogenesis, glycolysis
Fructose 1.6-bisphosphatase Glyceraldehyde 3-phosphate dehydrogenase Sedoheptulose l,7-bisphosphatase Glucose 6-phosphate dehydrogenase Phenylalanine ammonia lyase Ribulose 5' -phosphate kinase NADP+-malate dehydrogenase
Calvin cycle Pentose phosphate pathway lign in synthe sis
Calvin cycle C, pathway Light
--.. then to component enzymes containing regulatory disulfide bonds (Figure 20.16). We shall return to thioredoxin when we consider the reduction of ri bonucleotides (Section 25.3). NADPH is a signal molecule that activates two biosynthetic enzymes, phosphoribulose kinase and glyceraldehyde 3-phosphate dehydrogenase. In the dark, these enzymes are inhibited by association with a sm all protein called CP12. NADPH disrupts this association, leadi ng to the release of the actlve enzymes.
Ferredoxin,ed
Ferredoxin ox
•
Ferredoxin-thioredoxin reductase
The (4 Pathway of Tropical Plants Accelerates Photosynthesis by
Concentrating Carbon Dioxide The oxygenase activity of rubisco presents a biochemical challenge to trop ical plants because the oxygenase activity increases more rapidly with temperature than does the carboxylase activity. How, then, do plants, such as sugar cane, that grow in hot climates p revent very high rates of wasteful photorespiration? Their solution to this problem is to achieve a high local concentration of CO 2 at the site of the Calvin cycle in their photosynthetic cells. T he essence of this process, which was elucidated by Marshall Davidson Hatch and C. Roger Slack, is thatfour-carbon (C 4 ) compounds such as oxaloacetate and malate carry CO2 from mesophyll cells, which are in contact with air, to bundle-sheath cells, which are the major sites of photosynthe sis (Figure 20 .1 7). The decarboxylation of the four-carbon compound in a bundle-sheath cell maintains a high concentration of CO 2 at the site of the Calvin cycle. The three-carbon product returns to the mesophyll cell for an other round of carboxylation. The C 4 pathway for the transport of CO 2 starts in a mesophyll cell with the condensation of CO 2 and phosphoenolpyruvate to form oxaloacetate, in a reaction catalyzed by phosphoenolpyruvate carboxylase . In some species, oxaloacetate is converted into malate by an NADP+ -linked Air
Mesophyll cell Oxaloacetate
co,
CO2
Malate
pp.I +
p.I +
AMP
ATP
Phosphoenolpyruvate
Pyruvate
Bundle-sheath cell Malate Calvin cycle
cO 2 Pyruvate
Figure 20.17 C. pathway. Carbo n dioxide is concentrated in bundle-sheath cells by the expenditure of AlP in mesophyll cells.
SH SH
Thioredoxin
Inactive
5 I 5
Active
spontaneOU~5'_ _ _~ --.... oxidation
02
Figure 20.16 Enzyme activation by thioredoxin. Reduced t hioredo xin activates certain Calvin cycle enzymes by cleaving regulatory disulfide bonds.
576 CHAPTER 20 The Calvin Cycle and the Pentose Phosphate Pathway
malate dehydrogenase. Malate enters the bundle-sheath cell and is ox· idatively decarboxylated within the chl oroplasts b y an NADP + -linked m alate dehydrogenase. The released CO 2 enters the Calvin cycle in the usual way by cond ensing with ribulose l,5-bisphosphate. Pyruvate formed in this d ecarboxyl ation reaction returns to the m esophyll cell. Finally, phosphoenolpyruvate is fo rm ed from pyruvate by pyruvate-Pj dikinase . The net reaction of this C 4 pathway is CO 2 (in mesophyll cell) + ATP + 2 H 2 0 ) CO 2 (in bundle-sheath cell)
+
AMP
+
2 Pj
+ 2W
Thus, the energetic equivalent of two ATP molecules is consumed in transport· ing CO2 to the chloroplasts of the bundle-sheath cells. In essence, this process is acti ve transport: the pumping of CO 2 into the bundle-sheath cell is driven by the hydrolysis of one molecule of ATP to one molecule of AMP and two molecules of orthophosphate. The CO 2 concentration can be 20-fold as great in the bundle-sheath cells as in the mesophyll cells. When the C 4 pathway and the Calvin cycle operate together, the net reaction is 6 CO 2
+
30 ATP
+
12 NADPH + 24 H 2 0 ) C 6 H 12 0 G + 30ADP + 30P i + 12NADP +
+ 1RH+
Note that 30 molecules of ATP are consumed per hexose molecule formed when the C 4 pathway delivers CO 2 to the Calvin cycle, in contrast with 18 molecules of ATP per hexose molecule in the absence of the C 4 pathway. The high concentration of CO 2 in the bundle-sheath cells of C" plants, whi ch is d ue to the expenditure of the additional 12 molecules of ATP, is critical for their rapid photosynthetic rate, because CO 2 is limiting when light is abundant. A high CO 2 concentration also minimizes the energy loss caused by photo respiration. Trop ical plants with a C4 pathway do little photorespiration because the high concentration of CO 2 in their bundle-sheath ce lls accelerates the car· boxylase reaction relative to the oxygenase reaction. This effect is es pecially important at higher temperatures. The geographi c d istribution of plants having this pathway (C 4 plants) and those lacking it (C 3 plants) can now be understood in molecular terms. C 4 plants have the advantage in a hot enviro nm ent and under high illumination , which accounts for their prevalence in the tropics. C.1 plants, which consume only 18 m olecules of ATP per hexose molecule formed in the absence of photo respiration (compared with 30 molecules of ATP for C 4 plants), are more efficient at temperatures lower than about 28°C, and so they predominate in temperate environments. Rubisco is present in bacteria, eukaryotes, and even archaea, though other photosynthetic components have not been found in archaea. Thus, rubisco emerged early in evolution , when the atmosphere was rich in CO 2 and almost devoid of O 2 , The enz ym e was not originally selected to operate in an environment like t he present one, which is almost devoid of CO 2 and rich in O z. Photorespiration became significant about 600 million years ago, when the CO 2 concentration fell to present levels. The C 4 path· way is thought to have evolved in response no more than 30 million years ago and possibly as recently as 7 million years ago. It is interesting th at none of the enzymes are unique to C 4 plants, suggesting that this pathway made use of already existing enzymes .
577
Crassulacean Acid Metabolism Permits Growth in Arid Ecosystems Many plants growing in hot. dry climates keep the stomata of their leaves closed in the heat of the d ay to prevent water loss (Figure 20.18). A s a consequence. CO 2 cannot be absorbed during the daylight hours when it is needed for glucose synthesis. Rather. CO 2 en ters the leaf when the stomata open at the cooler temperatures of night. To store the CO 2 until it can be used during the day. such plants make use of an adaptation called crassulacean acid metabulism (CAM). n am ed after the genu s Crassulacea (the succulents). Carbon dioxide is fixed by the C 4 pathway into m alate. which is stored in vacuoles . During the day. m alate is decarboxylated and the CO 2 becomes available to the Calvin cycle. In contrast with C 4 plants. CAM plants separate CO 2 accumul ation fro m CO 2 utilization temporally rather than spatially.
20.3 The Pentose Phosphate Pathway
Figu re 20.18 Electron micrograph of an open stoma and a closed stoma. [Herb Charl es O hlmeyer/ Fran Heyl Associates.) •
20.3
The Pentose Phosphate Pathway Generates NADPH and Synthesizes Five-Carbon Sugars
The pentose phosphate pathway meets the need of all organisms for a source of NADPH to use in red uctive biosynthesis Crable 20.2 ). T his pathway consists of two phases : (1 ) the oxidative generation ofNADPH and (2) the nonoxidative interconversion of su gars (Figure 20.19). In the oxidative phase. NADPH is generated when glucose 6-phosphate is oxidized to ribose 5- phosphate . This five -carbon su gar and its derivatives are components of RNA and DNA. as well as ATP. NADH. FAD. and coenzyme A . ) Glucose 6-phosphate + 2 ADP + + H 2 0 ribose 5-phosphate + 2 NADPH + 2 H + + CO 2 In the non oxidative phase. the pathway catalyzes the interconversion of three-. four- . five-. six-. and seven -carbon sugars in a series of nonoxidative reactions. Excess five- carbo n sugars may be converted into intermed iates of the glycolytic pathway. A II these reactions take place in the cytoplasm. These interconversions rely on the same reactions that lead to the regeneration of ribulose 1. 5-bisphosphate in the Calvin cycle.
Two Molecules of NADPH Are Generated in the Conversion of Glucose 6-phosphate into Ribulose 5-phosphate The oxidative phase of th e pentose phosphate pathway starts with the dehydrogenation of glucose 6-phosphate at carbon 1. a reaction catalyzed by glucose 6-phosphate dehydrogenase (Figure 20.20). This enzyme is highly specific for NADP + ; the K M for NAD + is about a thousand times as great as that for NADP +. T he product is 6-phosphoglucono-8-lactone. wh ich is an intramolecular ester between the C-l carboxyl gro up and the C- 5 hydroxyl
TABLE 20.2 Pathways requiring NADPH Synthesis
Fatty acid
biosynthesi~
Cho lestero l biosyntheSi S Neurotransm itter bio syntheSIS N ucleo tid e bi osyntheSiS Detoxification Reduc t ion of oxid ized glutath ione Cy tochro me P450 monooxygenases
578 CHAPTER 20 The Calvin Cycle and the Pe ntose Phosphat e Pathway OH
H
H
OH
Glucose
6-phosphate
1 NADPH + Q,
~c
PHASE 1
(oxidative)
/ CH,OH
ICI( ICH
H-
H-
to J.
'OH
OH 2 2 0PO l -
Ribulose S-phosphate
~ /H
I
H-
C-
~ /
OH
IC- OH I C- OH ICH20pol
HH-
HOH-
-
(-
H
I( - 'OH ICH 0P0 1 3
Xylulose 5-phosphate (C.)
S-phosphate (Cs )
o~ /
o~ ,.... H
HOOH
H-
bH, OPO,'-
H-
CAP (C.)
CH, OH
I
(
H-
I
2
Ribose
I( -
CH,OH
(-
H
ICI( -
OH
I ICH
OH
H-C~ 'OH
2 0PO l
1-
Sedoheptulose 7-phosphate (C,)
~ / (H , OH
HOHH-
I
( - H
IC- OH I C- OH ICH 0PO 22
o~
o~ " H
J
I
H- C-
'OH
H- ( -
OH
I
HO H-
bH, OPO,'-
Fructose Erythrose 6-phosphate (C.) 4-phosphate (C.)
o~ (
b
HO
Figure 20.19 Pentose phosphate pathway. The pathway consists of (1) an oxidative phase that generates NADPH and (2) a nonoxidative phase that inter(onverts phosphorylated sugars.
H-
PHASE 2
(nonoxidative)
. . . . . .CH 2 0H
H-
H
IC- OH IC- OH ICH20POl2-
Frudose
6-phosphate (Co)
. . . . .CH2 0 H
II
H
(-
OH
ICH 0PO 1 2
l
xylulose 5-phosphate (C,)
0 "", - ,:;.0
"'C '"
H'
H-
+ NADP+ NADPH
H
OH H
0
--=::::::: \ =L = ===, ,-
H H
H
) --
Glucose 6-phosphate dehydrogenase
OH
H2 0
H
=
0
C-
OH NADP+ NADPH
=\ ::=:::= L ::=:,
HO- C-
IH
,
;=,
Lactonase
H-
C-
OH
H
C
OH
HO
\./
H
C
OH
6-Phosphogluconate dehydrogenase
H
C
OH
+ CO 2
OH
H
CH 20PO,26-Phosphoglueonoo-ladoRe
Glucose
6-phosphate
6-Phosphoglueonate
group. The next step is the hydrolysis of 6-phosphoglucono-o-Iactone by a specific lactonase to give 6-phosphogluconate. This six -carbon sugar is then oxidatively decarboxylated by 6-phosphogluconate dehydrogenase to yield ribulose 5-phosphate . N AD p + is again the electron acceptor. The final step in the synthesis of ribose 5-phosphate is the isomerization of ribulose 5phosphate by p hosphopentose isomerase (see Figure 20.1 1)
Ribulose
S-phosphate
Figure 20.20 Oxidative phase of the pentose phosphate pathway. Glucose 6-phosphat e is ox id ized t o 6-phosphoglucono-8-lactone to generat e o ne mo lecul e of NADPH. The lactone product is hydrolyzed to 6-phosphogluconate, which is OXi dat ively deca rboxylat ed to ribul ose 5-phosphate wit h the generat ion of a second molecule of NADPH.
The Pentose Phosphate Pathway and Glycolysis Are Linked by Transketolase and Transaldolase The preceding reactions yield two molecules ofNADPH and one molecule of ribose 5-phosphate for each molecule of glucose 6-phosphate oxidized . However, many cells need NADPH for reductive biosyntheses much more than they need ribose 5-phosphate for incorporation into nucleotides and nucleic acids . In these cases, ribose 5-phosphate is converted into glyceraldehyde 3-phosphate and fructose 6-phosphate by transketolase and transaldolase. These enzymes create a reversible link between the pentose phosphate pathway and glycolysis by catalyzing these three successive reactions.
Cs + C s C, + C 7 C4 + Cs
T ransketolasc
acetyl eoA + acyl eoA (shortened by C2)
6
' L-3-hydroxyacyl CoA
Carn itine acyltransferase (also called carniti ne palmi toyl transferase) Acyl CoA dehydrogenases (seve ra l isozymes having different cha in-length specific ity) Enoyl CoA hydratase (also called crotonase or 3-hydroxyacyl CoA hydro lyase) L-3-Hydroxyacyl CoA dehydrogenase [l -Ketothi olase (also called thiolase)
• An AMP-form ing ligase.
625
Approximately 2.5 molecules of ATP are generated when the respi· ratory chain oxidizes each of these NADH molecules, wh 1.~ molecules of ATP are formed for each FADH2 because their elec· trons enter the chain at the level of ubiquinol. Recall that the oxida· tion of acetyl CoA by the citric acid cycle yields 10 molecul es 01 ATP. Hence, the number of ATP molecules formed in the oxidation of palmitoyl CoA is 10.3 from the seven FADH 2 , 17.5 from the seven NADH, and 80 from the eight acetyl CoA molecules, which gives a total of 108. The equivalent of 2 molecules of ATP is con· sum ed in the activation of palmitate, in which ATP is split into AMP and two molecules of orthophosphate. Thus, the complete oxi·
dation of a molecule of palmitate yields 106 molecules of ATP.
22.3
o
o •
C
H,C
CoA
5
o
Unsaturated and Odd-Chain Fatty Acids Require Additional Steps for Degradation
The j3-oxidation pathway accomplishes the complete degradation of saturated fatty acids having an even number of carbon atoms. Most fatty acids have such structures because of their mode of synthesis (p. 636). However, not all fatty acids are so simple. The oxidation offatty acids containing double bonds requires additional steps, as does the oxidation of fatty acids containing an odd number of carbon atoms.
An Isomerase and a Reductase Are Required for the Oxidation of Unsaturated Fatty Acids Figure 22.10 First three rounds in the degradation of pa lmitate. Two-carbon units are sequentia lly removed from the carboxyl end of the fatty acid.
o Palmitoleoyl CoA
0
H •
H
'c /'
H3 C ,(CH 2ls
C"'"
4
3
2
/
CoA
5
C H2 1
2
ciS-J1'-Enoyl CoA ci5-~3_E noyl CoA •
Isomerase
o
l;l
(CH2ls
H3 C""
~C
/C~
"'c ,/
H
H2 4
/ "-.,
3
2
1
trDns-~2-Enoyl CoA
626
The oxidation of unsaturated fatty acids presents some difficulties, yet many such fatty acids are available in the diet. Most of the reactions are the same as those for saturated fatty acids. Tn fact, only two additional en· zymes an isomerase and a reductase are needed to degrade a wide range of unsaturated fatty acids. Consider the oxidation of palmitoleate. This C 16 unsaturated fatty acid, which has one double bond between C-9 and C -10, is activated and trans· ported across the inner mitochondrial membrane in the same way as saturated fatty acids. Palmitoleoyl CoA then undergoes three cycles of degradation, which are carried out by the same enzymes as those in the ox· idation of saturated fatty acids . However, the cis- ,13-enoyl CoA formed in the third round is not a substrate for acyl CoA dehydrogenase. The presence of a double bond between C -3 and C -4 prevents the formation of another double bond between C-2 and C-3. T his impasse is resolved by a new reac· 3 tion that shifts the position and configuration of the cis-,1 double bond.
.... CoA
5""
cis-Ll .1 Enoyl CoA isomerase converts this double bond into a trans -Ll double bond. The subsequent reactions are those of the saturated fatty acid oxida· 2 tion pathway, in which the trans-,1 -enoyl CoA is a regular substrate. Another problem arises with the oxidation of polyunsaturated fatty acids. Consider linoleate, a CI S polyunsaturated fatty acid with cis- Do 9 and 12 cis_,1 double bonds (Figure 22.11) . The cis-,1 3 double bond formed after three rounds of j3 oxidation is converted into a trans- ,1 2 double bond by the aforementioned isomerase. The acyl CoA produced by another round of ~ oxidation contains a cis- ,1 4 double bond. Dehydrogenation of this species by acyl CoA dehydrogenase yields a 2, 4-dienoyl intermediate, which is not a substrate for the next enzyme in the j3 -oxidation pathway. This impasse is circumvented by 2,4-dienoyl CoA reductase, an enzyme that uses NADPH to reduce the 2,4-dienoyl intermediate to trans - ,1 3-enoyl CoA. cis- t/ Enoyl CoA isomerase then converts trans - ,1 3-enoyl CoA into the trans-dl form, a cllstomary intermediate in the j3 -oxidation pathway. These catalytic
o
62 7 22 .3 Degradation of Unsaturated and Odd-Chain Fatty Acids
o t;i
H2
Linoleoyl CoA
H3 C"
/c~ /c~ (CH 2)4 C C
H
H2
5
4
32
1
tranS 112-En oyl CoA
o
w
cis-A' -Enoyl CoA •
Isomerase
o
cis-A'-Enoyl CoA
H2 ~ H3C" /C ~ .yC,,-(CH 2)4 C C H2
Isomerase
/
5
H
o
5
CoA
4321
trDns-~3_Enoyl CoA
2,4-Dienoyl CoA reductase NADPH + H+ FAD
0
FADH,
0
t;i
H,
/C~
(CH 2)4
H3C
" ,C H s
C
H
4
3
/
5
C H2
2
(CH 2)4
CoA H3C'
Acyl CoA dehydrogenase
1
" ,C H s
/C~ C H H 4
3
2
/
COA
5 1
2,4-Dienoyl CoA
strategies are elegant and economicaL On ly two extra enzymes are needed for the oxidation of any polyunsaturated fatty acid . Odd -numbered double bonds are handled by the isomerase, and even -numbered ones by the reductase and the isomerase.
Figure 22.n Oxidation of l inoleoyl CoA. The complete oxi dation of the diunsaturated fatty acid linoleate is facili t ated by the activity of enoyl CoA isomerase and 2,4-dienoyl CoA reductase.
Odd-Chain Fatty Acids Yield Propionyl CoA in the Final Thiolysis Step
o
Fatty acids having an odd number of carbon atoms are minor species. They are oxidized in the same way as fatty acids having an even number, except that propionyl CoA and acetyl CoA, rather than two molecules of acetyl eoA, are produced in the final round of degradation . The activated threecarbon unit in propionyl CoA enters the citric acid cycle after it has been converted into succinyl CoA. The pathway from propionyl CoA to succinyl CoA is es pecially inter esting becau se it entails a rearrangement that requires vitamin B12 (also known as cobalamin ). Propionyl CoA is carboxylated at the expense of the hydrolysis of a molecule of ATP to yield the D isomer of methyl malonyl CoA (Figure 22.12). This carboxylation reaction is catalyzed by HCO,+ ATP
o HJC""
/C~
~2
/ 5
Propionyl CoA
Propionyl CoA
p.,
+ ADP
COA __~~~_->
o - q
0
0
0
- :
I
H2
O/C~ C / C~ S/COA -==' O,/C~ C/C~S/COA -==" O'0,C/C~C/C~S/COA H ( 3
"H
o-Methylmalonyl CoA
,-
/ C
H3
\H
L-Methylmalonyl CoA
Figure 22.12 Conversion of propionyl CoA into succinyl CoA . Propio nyl CoA. generated from fatty acids wi th an o dd number of carbons as well as some amino ac ids, is converted into the citri c acid cycle intermediate succinyl CoA.
,-
-
J
H2 Succinyl CoA
628 CHA PTER 22
propionyl CoA carboxylase, a bi otin enzy me that has a catalytic mecha· ni sm like that of the homologous enzym e py ruvate carboxylase (p. 462). T he D isom er of m ethyl malon yl CoA is racemized to the L isomer, the su bstrate for a mutase that con verts it into succinyl CoA by an intramole· cular rearrangement. The CO S CoA group mi grates fro m C· 2 to m ethyl group in exchan ge for a hydrogen atom . T hi s very unu sual iso· m erization is catalyzed b y methy lmalonyl CoA mutase, which co ntain a d erivative of co balami n as its coenzym e.
Fatt y Acid M etabolism
Vitamin B12 Contains a Corrin Ring and a Cobalt Atom
Cobalamin enzym es, which are present in most organism s, catalyze three types of reactions : (1) intramolecular rearrangements; (2) methylations, as in the syn thesis of methionine (p. 691 ); and (3) the reduction of ribonucleotides to deoxyribonucleotides (Section 25 .3). In mammals, only two reactions are kn own to require coenzym e B 12 . The conversion of L-methylmalonyl CoA into succinyl CoA is one, and the formation of m ethi onin e by methylation of homocysteine is the other. The latter reaction is especially important be· cause m ethionine is required for the generation of coenzym es that partici· pate in the synthesis of purines and thymine, which are need ed for nucleic acid synthesis. The core of cobalamin consists of a corrin ring with a central cobalt atom (Figure 22. 13). T he corrin rin g, li ke a porphyrin , has four pyrrole units. Two of them are directly bonded to each oth er, whereas the others are joined by methine bridges, as in porphyrins, The co rrin ring is m ore reduced than that of porphyrins and the su bstituents are different. A cobalt atom is bonded to the four p yrrole nitrogens, The fifth substituen t linked to the cobalt atom is a deri vative of dimethy lbenzimidazole that contains ribose 3-p hosphate and amino isopropanol. O ne of the nitrogen atoms of di· m ethylbenzimidazole is linked to the cobalt atom , In coenzyme B 12, the sixth substituent linked to the cobalt atom is a 5' -deoxyadenosyl unit, This po· sition can also be occu pied by a cyano group, a m ethyl grou p , or other lig. ands, In all of these compounds, the cobalt is in the + 3 oxidation state,
o -----NH2 •
OH
..
,.::.
o
N
N
\:0
/ ~\ N , N
o
I
Coenzym e B12 (5 ' -Deoxyadenosylcobalamin)
""" " // " " , / /NH2
X ,
OH
_I
o Corrin rin g
N
x
,
~
-
I
• •• ,, ,, ,
---i
NH2
,•
N
~
0=
- -
o Benzimidazole
o
X
1=
CN ,
Cyanocoba lam in
Methylcobalam in
Figure 22,13 Structure of coenzyme B12 (S ' -deoxyadenosylcobalamin), The substitution of cyano and methyl groups creates cyanocoba lam in and methylcobalam in, respectivel y,
R
Mechan ism: Methylmalonyl CoA Mutase Catalyzes a Rearrangement to Form Succi ny l CoA The rearrangement reactions catalyzed by coenzyme BI 2 are exchanges of two groups attached to adjacent carbon atoms (Figure 22.14). A hydrogen atom migrates from one carbon atom to the next, and an R group (such as the - CO S CoA group of methylmalonyl CoA) concomitantly moves in the reverse direction. The first step in these intramolecular rearrangements is the cleavage of the carbon- cobalt bond of S'-deoxyadenosylcobal2 amin to generate the Co + form of coenzyme B' 2 and as' -deoxyadenosyl radical, C H 2 ' (Figure 22 ,1 5), In this homolytic cleavage reaction, one elec tron of the Co- C bond stays with Co (reducing it from the + 3 to the + 2 ox idation state) while the other stays with the carbon atom, generating a free radical. In contrast, nearly all other cleavage reactions in biological systems are heterolytic an electron pair is transferred to one of the two atoms that were bonded together,
R Homolytic bond cleavage,
c
c:-
H
H
-
c
c.-
R Figure 22.14 Rearrangement reaction catalyzed by cobalam in enzymes. The R group can be an amino group, a hydroxyl group, or a substituted carbon.
Figure 22.1S Formation of a S'-deoxyadenosyl radical. The methylmalonyl CoA mutase reaction begins with the ho mo lytic cl eavage of the bond joining C0 3+ to a carbon atom of the ribose o f the ad enosine moiety, The cleavage generates a 5'-deo xyaden osyl radical and lead s to the reduction of
~ CH • 1 .../~,,5 ' - Deoxyadenosyl
-
radical
.U QJ
"-
Citrate
'./
'"
~ o
.0
Figure 22.31 Dependence of the catalytiC activi t y of acetyl CoA carboxylase on the concentration of citrate. (A) Citrate can partl y activate the phospho rylated carboxylase. (B) The dephospho ry lated form of the carboxylase is highly active even when citrate is absent. Citrate partly overcomes the inhibition produced by phosphorylation. [After G. M. Mabrouk, I. M. Helm y, K. G. Thampy, and S. J. Wakil.
~
:g
Highly phosphorylated
:40407- 40415. Kalgutkar, A. S., C rews, B. C. , Rowlinson, S. W ., Garner , c., Seibert, K., and Marnett L. J, 1998. Aspirin -like molecules that covalentl y inactivate cyclooxygcnasc-2. Science 2gU :1268- 1270. Lands, W. E. 1991, Biosynthesis of prostaglandins. Annu. Rev, Nutr. 11 :41 - 60, Sigal. E, 1991. T he molecular biology of mammalian arachidonic acid metabolism. Am . f. Physiol. 260 :L13- L28. Weissmann, G . 1991. Aspirin . Sci. Am. 264(1):84- 90.
"
tection of inherited disorders of mitoc hond rial f3 ·oxidation .
Biochem. Soc. Trans. 26:145- 152. Pollitt, R. J. 1995. Disoruers of mitochondrial long -chain fatty acid oxidation . }. Inherited Metab. Dis. 18:473 490. Roe, C, R" anu Coates, P. M . 1995 . Mitochondrial fatty acid oxidation diso"lers, In The Metabolic Basis of Inherited Diseases (7th ed., pp. 1501- 1534), edited by C. R. Scriver, A. L. Beaudet, W. S. Sly, D. Valle, J, 1.l. Stan bury, J. B. Wyngaarden , and D. S. Fredrickson. McGraw- Hill .
.
Problems 1. After lipolysis. Write a balanced equation for the convers ion
Whic h carbo n atom o f the palmitate fo rmed by this system is m o re radioactive C- l o r C -14?
of glycerol into p yruvate, Which enzymes are requ ired in addi tion to those of t h e g lycolytic pathway '
6. Driven by decarboxylation. What is the role of decarboxylation
2. From fatty acid to ketone body, Write a balanced equation for the conversion of stearate in to acetoacetate ,
in fatty acid synthesis' Na m e a nothe r key reaction in a m etabolic pathway that e mploys thi s m echani stic motif. ,
3. Counterpoint. Compare and contrast fatty acid oxidation a nd synthesis with respect to site of t h e process. acyl carrier . reductants and oxida nts. stereoch emistry of the interm ed iates. direction of synthesis o r d egradation . (0 organization of the enzyme syst em . (a) (b) (e) (d) (e)
4, Sources, For each of the following unsaturated fatty acids, in dicate whether the biosynthetic precursor in animal s is palmitoleate, oleate, linoleate, o r linolenate, (a) 18: 1 cis -Il l I (d ) 20:3 cis-Il s , Il R, Il II 13 (b) 18:3 c .ls -Il 6 ,Il 9 , Il 12 ( e ) 22 :1 c is - ll (c) 20:2 c is- Il l I , 11 14 (f) 22:6cis -1l 4, 1l 7 , Il l", Il l.l, Ill t., 11 19
5, Tracing carbons. Con sider a cell e xtract that actively synthesizes palmitate . S uppose that a fatty acid synthase in this preparation forms o n e m o lecule of palmitate in about 5 minutes. A large amount of malonyl CoA labeled with 14C in each carbon atom of its malonyl un it is suddenl y added to this system, and fatty acid synthesis is sto pped a minute late r by alte ring the pH , The fatty a cid s in the supe rnatant are analyzed for radioactivity.
7. Kinase surfeit. Suppose that a promo te r mutatio n leads to the overproduction of protein kinase A in adipose cell s, H ow might fatty acid m e tabolism b e alte red by th is mutatio n ? 8. An unaccepting mutant, The serine res idue in acetyl CoA carboxylase th at is the target of the AMP -d ep endent protein kinase is m u tated t o alanine. What is a li kely con sequence o f this mutation?
9, Blocked assets. The presence of a fu el m o lecu le in the cytop lasm does not e n sure that the fuel m o lecul e can be effectively used. Give two examples of h ow impaired transport of metabolites between compartments leads to disease.
10 . Elegant inversion. P eroxisom es have an a lternative pathway for oxidizing polyunsaturated fatty acids. They contain a hydratase that converts D-3- h ydroxyacyl CoA into trans- 1l 2 -en oyl CoA. How can this enzyme be used to oxidize CoA s containi n g a cis doub le bond at an even - numbered carbon atom (e.g., t h e c is- Ill ' double b ond ofl ino leate)?
11 . Covalent catastrophe. What is a pote ntial disadvantage of havin g many catal y ti c sites together o n one very long poly pe p tide chain?
648
CHAPTER 22 Fatty Acid Metabolism (a) What is the effect of the mutation on en zy m e activity when the concentration of carnitine is varied (Fi gure A )I What are the KM and Vmox val ues for the wild-type and mutant enzymes)
12. Missing acyl CoA dehydrogenases. A number of genetic defi ciencies in acyl CoA dehydrogenases have been described. This deficiency presents early in life after a peri od of fasting . Symptom s include vomiting, lethargy, and sometimes coma. Not only are blood levels of gl ucose low (hypoglycemia), but starvation -induced ketos is is absent. Provid e a biochemical explanation for these last two observations.
-.-,
~
c::
15
E
-, ""E 10 -
13. Effects oj clofibrate. High blood levels of triacylglycerid es are associated with hea rt attacks and strokes. C lofibrate, a drug that increases the activity of peroxisomes, is sometimes used to treat patients with such a condition . What is the biochemical basis for thi s treatment?
Wild type
0
E
Mutant
c::
~
.> .-w -f-'"
5
u
0
~
a.
14. A different kind oj enzyme. Figure 22.3 1 shows the respon se o f acety l CoA carboxylase to vary ing amounts of citrate. Explain this effect in ligh t of the allosteric effects that citrate has on the enzyme. Predict t he effects of increasing con centrations of palmitoyl CoA.
250
500
(A)
750
1000
1250
[Carnitine], "M
(b) What is the effect when the experiment is repeated with varying concentration s of palmitoyl CoA (Figure B)? W hat are the KM and V""" values for the wi ld-type and mutant enzymes'
Mechanism Problems •
15 . Variation on a theme. Thiolase is homologous in structure to th e conden sin g enzym e. On the basis of this observation , p ropose a mechanism for the cleavage of 3- ketoacy l CoA by CoA.
16. Two plus three to make Jour. Propose a reaction m echanism for the condensation of an acetyl unit w ith a m alonyl unit to form an acetoacetyl unit in fatty acid synthesis.
-.-,
50
-,
40
~
c::
E
bO
E
-0
520
Chapter Integration Problems
17. Ill -adv ised di et. Suppo se that, for so me biz arre rea"m ,
-f-'"
(a) H ow would lack of carbohydrates affect your ability to uti lize fatsl (b) What wou ld your breath smell li ke? (c ) O ne of your best fri ends, after trying unsu ccessfull y to con vince you to abandon thi s d iet, makes you promise to con sume a healthy dose of odd -chain fatty acids . Does yo ur friend have yo ur hest interests at heart' Explain.
~
v
Mutant 10
Cl.
u 0 (B)
100
200
500
600
700
.-.. 100
-
0
~
~
c::
-
80
v
0
~
60
~
~
40
-f-'"
20
..-> w
Data Interpretation Problem
alyzes the conversion of long-chain acyl CoA into acyl carnitine, a prerequisite for transport into mitochondria and subsequent degradation. A mutant enzym e was constructed with a sin gle amino acid change at position :> of glutamic acid for alanine. Figures A through C show data from studies performed to identifY the effect of the mutation [data from ]. Shi, H . Zhu, D. N. A,vidson, and G. J. Woldegiorgis. j. BioI. Chern. 274(1999):9421
yo u d ecid ed to exis t on a diet of whale and seal blubber , excl u si ve l y.
Wild type
0
(e)
100
200
300
400
500
[Malonyl CoAl, "M
(d ) Suppose that t h e concentration of palmitoyl eoA = 100 I-LM, that of carnitine = 100 I-LM , and t hat of malonyl CoA = 10 I-LM. U nder these condition s, what is the most prominent effect of the mutation on the properties of the enzyme? (e) \Vhat can you conclude about the rol e of glutamate 3 in carnitine acy ltransferase I function ?
Chapt e r
Protein Turnover and Amino Acid Catabolism Arginine
Urea
1
Argininosuccinate Ornithine
Carbamoyl phosphate Citrulline
Degradation of cycl in B. This im portant prot ein in cell-cycle regulation is visible as the green areas in the images above (the protein was f used wi th green fl uorescent protein). (yelin B is prominent during metaphase but is degraded in anaphase t o prevent the premature initiation of anot her cell cycle. A large p rotease com plex ca lled th e proteasome digests the protein into peptides, wh ich are then degraded into ami no acids. These ami no acids are either reused or furt her processed so th at the carbon skeletons can be used as fuels or building blocks. The released amino group is converted into urea f or excretion by the urea cycle. [(Left) Courtesy o f Dr. Jonathan Pines, University of Cambridge, Wellcome/ CRC Inst itute of Cancer and Developmen t al Biology.] ,
he digestion of dietary proteins in the intestine and the degradation of proteins within the cell provide a steady sup ply of amino acids to the cell. Many cellular proteins are constantly d egraded and resynthesized in response to changin g metabolic d em ands. Others are misfolded or become damaged and they, too, must be d egraded . U nn eed ed or damaged proteins are marked for destruction b y the covalent attachment of chains of a sm all protein call ed ubiquitin and then degraded by a large, ATP -d ependent com plex called the proteasome. The primary use of amino acids provided through degradation or digestion is as building blocks f or the synthesis of proteins and other nitrogenous compounds such as nucleotide bases. Amino acids in excess of those needed for biosyn thesis can neither be stored, in contrast with fatty acids and glucose, nor excreted . Rather, sur plus amino acids are used as metabolic fuel . The a- amino group is removed, and the resulting carbon skeleton is converted into a major metabolic intermediate. Most of the amino groups harvested from surplus amino acids are converted into urea through the urea cycle, whereas their carbon skeletons are transform ed into acetyl CoA , acetoacetyl CoA , pyruvate, or one of the intermedi ates of the citric acid cycle. The principal fate of the carbon skeletons is conversion into glucose and glycogen .
Uutlin e 23.1 Proteins Are Degraded to Amino Acids 23.2 Prote in Turn over Is Tightly Regulated 23.3 The First Step in Amino Acid Degradation Is the Removal o f Nitrogen 23.4 Ammonium Ion Is Converted into Urea in Most Terrestrial Vertebrates 23.5 Carbon Atom s of Degraded Amino Acids Emerge As Major Metabolic Intermediate s 23.6 Inborn Errors of Metabolism Can Disrupt Amino Acid Degradation
649
650
Several coenzymes play key roles in amino acid degradation, fore· most among them is pyridoxal phosphate. This coenzyme forms Schiff· base intermediates that allow a-amino groups to be shuttled between amino acids and ketoacids. We will consider several genetic errors of amino acid degradation that lead to brain damage and mental retarda· tion unless remedial action is initiated soon after birth. Phenylketonuria, which is caused by a block in the conversion of phenylalanine into tyro· s ine, is readily diagnosed and can be treated by removing phenylalanine from the diet . The stud y of amino acid metabolism is especiall y reward· ing because it is rich in connections between basic biochemistry and clinical medicine .
CHAPTER 23 Protein Turnover and Amino Acid Catabolism
23.1 TABLE 23.1 Essential amino acids in human beings
Proteins Are Degraded to Amino Acids
Dietary protein is a vital source of amino acids. Especially important dietary proteins are those containing the essential amino acids amino acids that cannot be synthesized and must be acquired in the diet (Table 23.1). Proteins ingested in the diet are digested into amino acids or small peptides that can be absorbed by the inte.s tine and transported in the blood . Another crucial source of amino acids is the degradation of cellular proteins.
H istid ine
Isoleucine Leucine Lysine Methionine Phenyl alanine
The Digestion of Dietary Proteins Begins in the Stomach and Is Completed in the Intestine
Threo nine
Tryptophan Val ine
Figure 23.1 Digestion and absorption of proteins. Protein digestion is primarily a result of the activ it y of enzy mes secreted by the pancreas. Aminopeptidases associated with the intestinal epithelium further digest proteins. The amino acids and di- and tripept ides are absorbed into the intestinal cells by specific transporters. Free amino acids are then re leased into the blood for use by o ther ti ssues.
Protein digestion begins in the stomach, where the acidic environment fa· vors the denaturation of proteins into random coi ls. Denatured proteim are more accessible as substrates for proteolysis than are native proteins. T he primary proteolytic enzyme of the stomach is pepsin, a nonspecific protease that, remarkably, is maximally active at pH 2. Thus, pepsin can function in the highly acidic environment of the stomach that disables other proteins. Protein degradation continues in the lumen of the intestine. T he pan· creas secretes a variety of proteolytic enzymes into the intestinal lumen as inactive zymogens that are then converted into active enzymes (Sections 9.1 and lOA ). The battery of enzymes displays a wide array of specificity, and so the substrates are degraded into free amino acids as well as di- and tripep' tides. Digestion is further enhanced by proteolytic enzymes, such as aminopeptidase N, that are located in the plasma membrane of the intes· tinal cells. Aminopeptidases digest proteins from the amino-terminal end. Single amino acids, as well as di- and tripeptides, are transported into the in· testinal cells from the lumen and subsequently released into the blood for absorption by other tissues (Figure 23.1) .
LUMEN
INTESTINAL CELL
BLOOD
Amino acids
Amino acids Proteolytic enzymes
Peptidases Tripeptides Dipeptides
Proteins - - - - _
Oligopeptides
-f--'Amlnopep ' t'd lase
Cellular Proteins Are Degraded at Different Rates Protein turnover the degradation and resynthesis of proteins takes place constantly in cells. Although some proteins are very stable, many proteins are short lived, particularly those that participate in metabolic regulation. These proteins can be quickly degraded to activate or shut down a signaling pathway. In addition, cell s must eliminate damaged proteins. A significant proportion of newly synthesized protein molecules are defective because of errors in translation or misfolding. Even proteins that are normal when first synthesized may undergo oxidative damage or be altered in other ways with the passage of time. These proteins must be removed before they accumu late and aggregate. Indeed, a number of pathological conditions such as certain forms of Parkinson disease and Huntington disease are associated with protein aggregation. The half-lives of proteins range over several orders of magnitude (see Table 23. 2) . Ornithine decarboxylase, at approximately 11 minutes, has one of the shortest half-lives of any mammalian protein. This enzyme participates in the synthesis of polyamines, which are cellular cations essential for growth and differentiation. The life of hemoglobin, on the other hand, is limited only by the life of the red blood cell, and the lens protein, crystallin, by the life of the organism.
23.2
TABLE
23.2 Dependence of the half-lives of cytoplasmic yeast proteins on the identity of their aminoterminal residues
Highly st abilizing residues
(t1/2 > 20 hours) Ala
Cys
Gly
Met
Pro
Ser
Thr
Val
Intrinsically destabilizing resi dues
(t / 2 '
~
2 t o 30 minutes)
Arg
His
lie
Leu
Lys
phe
Trp
Tyr
Des ta bilizing res idues after chemical modification
(t1/2 = 3 to 30 minutes) Asn
Asp
Gin
Glu
Source: J. W. Tobias, T. E. Schrader. G. Rocap, and A. Varshavsky. Science 254( 1991):1374-1377.
Protein Turnover Is Tightly Regulated
How can a cell distinguish proteins that should be degraded ? Ubiquitin (Ub), a small (8.5- kd ) protein present in all eukaryotic cells, is a tag that marks proteins for destruction (Figure 23.2). Ubiquitin is the cellular equivalent of the "black spot" of Robert Louis Stevenson's Treasure Island: the signal for death.
Ubiquitin Tags Proteins for Destruction
Ub
~o "
HN
Isopeptlde bond
lys
Ubiquitin is highly conserved in eukaryotes: yeast and human ubiquitin differ at only 3 of 76 residues. The carboxyl-terminal glycine residue of ubiquitin becomes covalently attached to the I::-amino groups of severallysine residues on a protein destined to be degraded. The energy for the formation of these isopeptide bonds (iso because £- rather than a -amino groups are targeted) comes from ATP hydrolysis .
o Peptide bond
o
H N H
H N
(o
Peptide bond
~ Figure 23.2 Structure of ubiquitin .
Notice that ubiquitin has an
Ubiquitin
C terminus
extended carboxyl terminus, whi ch is activated and linked to proteins targeted for destructio n. Lysine re sidues are shown as ball-and-stick models, including lysine 48, the major site for linking additional ubiquitin mo lecules. [Drawn from lUBLpdb.]
651
652 CHAPTER 23 Protein Turnover and Amino Acid Catabolism E1
+
AMP
ATP E1
CD Figure 23.3 Ubiquitin conjugation. The ubiquitin-act ivating enzyme E1 adenylat es ubiq uit in (Ub) (1 ) and transfe rs t he ubiquit in to o ne of its own cysteine resi dues (2). Ubiq uit in is th en tran sf erred to a cyst eine resid ue in t he ubiquitin conjugating enzyme E2 (3). Fi nal ly. the ubiquit in-protein ligase E3 transfers the ubiquitin t o a lysi ne res idue on t he t arget prot ein (4a and 4b).
El E2
G)
E2
E1
E2
Y- NH r
"
Target
@
@
( 7
(
I
'\ E3
E2. E3
E3
Three enzymes participate in the attachment of ubiquitin to a protein (Figure 23 .3): ubiquitin-activating enzym e, or El; u biquitin -conjllgating enzyme, or E2 ; and ubiq uitin- protein ligase, or E3 . First, the C-terminal carboxylate group of ubiquitin becomes linked to a sulfhyd ryl group of E1 by a thioester bond . This ATP-dr iven reaction is reminiscent offatty acid activation (p. 62 2) . In this reaction, AT P is linked to the C-terminal carboxylate of ubiquitin with the release of pyrophosphate, and the ubiquitin is tran sferred to a sulfhydryl grou p of a key cysteine residue in E1. The activated ubiquitin is then shuttl ed to a sulfhydryl group of E2. Finally, E3 catalyzes the transfer of ubiquitin from E2 to an I> amino group on the tar• get protem . A chain of four or more ubiquitin molecules is especially effective in signaling the n eed for degradation (F igure 23.4). The ubiquitin ation reaction is processive: a chain of ubiquitin molecules can be generated by t he linkage of the E:-amino group of lysine residue 48 of one ubiquitin molecul e to th e terminal carboxylate of another. What determin es whether a protein becomes ubiquiti nated? One signal turns out to be unexpectedly simple. The half-life of a cytoplasmic protein is determined to a To target large extent by its amino -terminal residue (Table 23.2). This "",protein , dependency is referred to as t he N -terminal rule. A yeast protein with methionine at its N terminus typically has a half-life of more than 20 hours, whereas one with arginine at this position has a half-life of about 2 minutes. A highly -terminal residue su ch as arginine or d estabilizing leucine favors rapid ubiquitination, whereas a stabilizing residue such as methionine or proline does n ot. Other sign als thought to identify proteins for degradation include cyclin destruction boxes, which are amino acid sequences that mark cell-cycle proteins for destruction , and PEST sequ ences, whi ch contain the amino acid sequence proline Iso peptide bonds (P, single-letter abbreviation ), glutamic acid (E), serine (S), and threonine (T ). ~ Figure 23.4 Structure of tetraubiquitin_ Fo ur ubiqu iti n mo lecu les are li nked by isopeptide bo nds. No t ice that each E 3 enzymes are the readers of N -terminal residues. isopept ide bond is form ed by the linkage o f the carboxylate Although most eukaryotes have only one or a small 1111m bei group at the end of the ext end ed C term inus with the e-amino of distinct El enzymes, all eukaryotes have many .. group o f a lysine residue. Dashed lines indicat e the positions of E2 and E3 enzymes. Moreover, there appears to be only a t he extended C-termini that were no t observed in th e crystal single famil y of evol utionarily related E2 proteins but three structure. Th is unit is the primary signal f o r degradatio n when linked t o a t arget protein. [Draw n from 1TBE.pdb.] distinct families of E3 proteins, altogether consisting of
hundreds of members. Indeed, the E 3 famil y is one of the largest gene fam ili~~ in human beings. T he diversity of target proteins that must be tagged for destruction requires a large number of E3 proteins as readers.
653 23.2 Regulation of Prot ei n Tu rnover
Three examples demonstrate the importance of E3 proteins to nor mal cell fun ction . Proteins that are not broken d own owing to a d efective E3 may accumul ate to create a disease of protein aggregation such as juvenile and earl y -onset Parkinson disease. A d efect in another member of the E3 fam il y causes A ngelman syndrome, a severe neurological disorder characterized by mental retardation, absence of speech , uncoordinated movement, and hyperactivity. Conversely, un controlled protein turnover also can create dangerous pa thological conditions. For example, human pa pilloma virus (HPV ) encodes a protein that activates a specific E 3 enzym e. The enzyme ubiquitinates the tum or suppressor p3 3 and other proteins that control D N A repair, wh ich are then d estroyed . T he activation of this E3 enzyme is observed in more than 90% of cervical carcinomas . Thus, the inappropriate m arking of key regulatory proteins for destruction can trigger further events, leadin g to tumor formation .
The Proteasome Digests the
a subunits
Ubiquitin-Tagged Proteins ~ subu nits
If ubiquitin is the mark of death, what is the execu tioner? A large protease complex called the proteasome or the 26S pro~ subunits teasome digests the ubiquitinated proteins. This ATP-driven a subun its multisubunit p rotease spares ubiquitin, which is then recycled. The 26S proteasom e is a complex of two compon ents: a 20S catalytic u nit and a 19S regulatory unit. The 20S unit is constructed from two copies each of 14 homologous subunits and has a mass of 700 kd (Figure 23.:;). The subunits are arranged in four rings of 7 subunits that stack to fo rm a structure resem bling a barrel. The outer two rings of the barrel are made up of a subunits and the inner two rings of 13 subunits. The 20S catalytic core is a sealed barrel. Access to its in terior is controlled by a 19S regulatory unit, itself a 700- kd compl ex mad e up of 20 subunits. Two such 19S complexes bind to the 20S proteasome core, one at each end , to form the complete 26S proteasom e (Figure 23.6). T he 19S unit binds specif ically to polyubiquitin chains, thereby ensuring that only ubiq uitinated proteins are d egraded . Key components of the 19S com plex are six ATPases of a type called the AAA class (ATPase associated with various cellular activities). ATP hydrolysis likely assists the 19S complex to un fold the substrate and induce conform ati onal changes in the 20S catalytic core so that the substrate can be passed into the center of the complex . The proteolytic active sites are sequestered in the interior of the barrel to protect potential substrates un til they are directed into the barrel. There are three types of active sites in the 13 subunits, each with a different specif icity, but all employ an N -terminal threonine. The hyd roxyl group of the threo nine residue is converted into a nucleophile that attacks the carbonyl groups of peptide bonds to form acyl-enzym e intermediates (p . 244). Substrates are degraded in a p rocessive m anner without the release of d egradation intermediates, until the substrate is red uced to pep tides ranging in length from seven to nine residues. Finall y, an isopeptidase in the 19S unit cleaves off intact ubiquitin m olecules from th ese peptides. T he ubiquitin is recycled and the peptide products are further degraded by other cellular proteases to yield individual amino acids. T hus, the ubiquitination path way and the proteasome cooperate to d egrad e unwanted proteins. Figure 23.7 p resents an overview of the fa tes of amino acids following p roteasomal d igestion .
N-terminal threonine nucleophile ~ Figure 23.5 205 proteasome. The
205 proteasome comp rises 28 homologous subun its (u, red; 13. blue). arranged in four rings o f 7 subun it s each. Some o f the 13 subunits incl ude protease act ive sites at their am ino t ermin i. [Subun it d raw n from 1RYP.pdb.]
195 cap
205 catalytiC core
195 cap
Figure 23.6 265 proteasome. A 195 cap is attached t o each end o f the 20 5 cat alyt iC unit. [Fro m W. Baumeister, J. Walz. F. Zuh l. and E. Seemuller. Cell 92(1998):367- 380; courtesy o f Dr. Wo lfgang Ba umeister.]
654 CHAPTER 23 Protein Turnover and Amino Acid Catabolism
Figure 23.7 The proteasome and other proteases generate free amino acids. Ubiquitinated proteins are processed to peptide fragments from which the ubiquitin is subsequently removed and recycled. The peptide fragments are further d igested t o yield free amino acids. which can be used for biosynthetic reactions. most notably pro t ein synthesi s. Alternatively. t he amino group can be removed and processed t o urea (p. 661) and the carbon skeleton can be used to synthesize carbohydrat e or fat s or used directly as a f uel fo r cellular respiration.
Ubiquitinated protein
Proteasome
•
ilJ
•
Peptide fragments
Released ubiquitin
Proteolysis
I
I
I
I
I
I I
I
""
Amino acids Left intact for biosynthesis
Amino groups
Nitrogen disposal by the urea cycle
Carbon skeletons Glucose or glycogen synthesis
Fatty acid synthesis
Celiular respiration TABLE 23.3 Processes regulated by protein degradation Gene transc ription
Protein Degradation Can Be Used to Regulate Biological Function
Cell-cycle progression O rgan formation
Circad ian rhythms Inflammato ry response
Tumor suppression Cholesterol metabolism Antigen processing
o N ~
N
H
#
o
N
Bortelomib (a dipeptidyl boronic acid)
Table 23.3 lists a number of physiological processes that are controlled at least in part by protein degradation through the ubiquitin- proteaso me pathway_ In each case, the proteins being degraded are regulatory proteins_ Consider, for example, control of t he inflammatory response. A transcription factor called NF -KB (NF for nuclear facto r) initiates the expression of a num ber of the genes that take part in this response. This factor is itself activated by the degradation of an attached inhibitory protein, J-KB (I for inhibitor)_ In response to inflammatory signals that bind to membrane-bound receptors, [-KB is phosphorylated at two serine residues, creating an E3 binding site. The binding ofE3leads to the ubiquitination and degradation of I-KB, unleashing NF-KB. The liberated transcription factor migrates to the nucleus to stimulate the transcription of the target genes_The NF- KB- I-KB system~ lustrates the interplay of several key regulatory motifs: receptor-mediated signal transduction, phosphorylation , compartmentalization, controlled and specific degradation, and selective gene expression . The importance of _b ubiquitin- proteasome system for the regulation of gene expression is highlighted by the recent approval of bortezomib (Velcade), a potent inhibitor
of the proteasome, as a therapy for multiple myeloma. Bortezomib is a dipeptidyl boronic acid inhibitor of the proteasome.
The Ubiquitin Pathway and the Proteasome Have Prokaryotic Counterparts ->{Jy Both the ubiquitin pathway and the proteasome appear to be pres-
T ent in all eukaryotes. Homologs of the proteasom e are found in prokaryotes, although the physiological roles of these homologs have not been well established. The proteasomes of some archaea are quite similar in overall structure to their eukaryotic counterparts and similarly have 28 sub units (Figure 23.8). In the archaeal proteasome, however, all ex outer-ring subunits and all 13 inner-ring subunits are identical; in eukaryotes, each ex or 13 subunit is one of seven different isoforms . This specialization provides distinct substrate specificity. Although ubiquitin has not been found in prokaryotes, ubiquitin's molecular ancestors were recently identified in prokaryotes. Remarkably, these proteins take part not in protein modification but in biosynthesis of the coenzyme thiamine (p. 423) . A key enzym e in thiamine biosynthesis is ThiF, which activates the protein ThiS as an acyl adenylate and then adds a sulfide ion d erived from cysteine (Figure 23 .9) . ThiF is homologous to human El, which includes two tandem regions of 160 amino acids that are 28% identical in amino acid seq uence with a region of ThiF from E. coli.
Archaeal proteasome
Figure 23.8 Proteasome evolution. The archaeal proteasome cons ists of 14 identical 0: subunits and 14 identical f3 subunits. In the eukaryotic protea some. gene duplicati on and specia lization has led to 7 distinct subunits of each type. The overall archit ecture of the proteasome is conserved.
H3C pp.,
o/ c." _ "0
+
ATP
./
"
Thi F
"
ThiS
ThiS "SH "
C
O~ ' AMP
"
N
.y
AMP
\. -?\. ThiF
»
I C
O~ '
SH
Eukaryotic proteasome
N~
NH,
CH,OH
IN+
)
CH 3 Thiamine
Figure 23.9 Biosynthesis of thiam ine. The biosynthesis of thi amine begins with the addition of sulfide to the carboxyl terminus of the protein ThiS. This protein is activated by adenylation and conjugated in a manner analogous t o the first steps in the ubiquitin pathway.
The evolutionary relationships between th ese two pathways were cemented by the determination of the three-dimensional structure of ThiS, which re vealed a structure very similar to that of ubiquitin, despite being only 14% identical in amino acid sequen ce (Figure 23 .10). Thus, a eukaryotic system for protein m odification evolved from a preexisting prokaryotic pathway for coenzyme biosynthesis.
~ Figure 23.10 Structures of ThiS and
Ubiquitin ThiS C terminus
C terminus
ubiquitin compared. Notice that ThiS is structurally similar t o ubiquitin despite only 14% sequence identity. This observation suggests that a prokaryo tic prot ein such as ThiS evolved into ubiquitin. [Drawn from 1UBI.pdb and 1FOZ.pdb.]
655
•
656 CHAPTER 23 Protein Turnover and Amino Acid Catabolism
23.3
The First Step in Amino Acid Degradation Is the Removal of Nitrogen
What is the fate of amino acids released on protein digestion or turnover; The first call is for use as building blocks for biosynthetic reactions. However, any not needed as building blocks are degraded to compound~ able to enter the metabolic mainstream. The amino group is first removed, and then the remaining carbon skeleton is metabolized to glucose, one 01 several citric acid cycle intermediates, or to acetyl CoA. The major site 01 am ino acid degradation in m amm als is the li ver, although muscles readily degrade the branched-chain am ino acids (Leu, lIe, and Val ). The fate of the a -amino group will be considered first, followed by that of the carbon skeleton (Section 23.5).
Alpha-Amino Groups Are Converted into Ammonium Ions by the Oxidative Deamination of Glutamate The a-amino group of many amino acids is transferred to a-ketoglutaratl to form glutamate, which is then oxidatively deaminated to yield ammo nium ion (NH4 +). - OOC
R
)-1
+H3
H
•
COO-
+H3N
COO-
Glutamate
Amin o acid
Aminotransferases catalyze the transfer of an a-amino group from an a-amino acid to an a-ketoacid. These enzymes, also called transaminases, generally funne l a-amino groups from a variety of amino acids to a-ketoglutarate for conversion into NH4 +. j
. :-. H - OOC
0
0
H3
+
Rl
Am inotransferase
-ooe
R,
•
+H3N
•
- OOC
R,
+
-ooe
H
Rl
Aspartate aminotransferase, one of the most important of these enzymes, catalyzes the transfer of the amino group of aspartate to a-ketoglutarate. Aspartate + a -ketoglutarate ,
Carboxyphosphate
Bicarbonate
Carbamic acid
The active site for this reaction lies in a domain formed by the amino- terminal third of CPS. T his domain form s a structure, caIled an A TP -grasp f old, that surrounds AT P and hold s it in an orientation suitable for nucleophilic attack at the 'Y phosphoryl group. Proteins containing ATP-grasp folds catalyze the formation of carbon- nitrogen bon ds through acyl-phosphate intermediates. Such AT P-grasp folds are wi dely used in nucleotide biosynthesis. In the second step catalyzed by carbamoyl p hosphate synt hetase, car bamic acid is phosphorylated by another m olecule of ATP to form carbamoyl phosphate. ATP
\ / Carbamic acid
2- 0
AD P )
~
!I
O' j 'P"",- ......... C"-
o
site
p.
NH,
!
Glutamin e hydrolysis site
0
NH2
Carbamoyl phosphate
This reaction takes place in a second AT P -grasp domain within the en zyme. T he active sites leading to carbamic acid fo rmation and carbamoyl phosphate formation are very sim ilar , revealing that t his enzyme evolved by a gene du plication even t . Indeed, du plication of a gene encoding an ATP grasp domain fo llowed by specialization was cen tral to the evolu tion of nucleotide biosyn thetic processes (p . 715). The Side Chain of Gl utam ine Can Be Hydroly zed to Generate Ammon ia
Glutamine is the primary source of ammonia for carbam oyl phosphate syn thetase. In this case, a second polypeptide component of the en zym e hydrolyzes glu tamine to form amm onia and glu tamate . The active site of the glutamine- hydrolyzing componen t contains a catalytic dyad compri sing a cysteine and a histidine residue. Such a catalyti c d yad , remini scent of the acti ve site of cysteine proteases (p . 251 ), is conserved in a family of amidotransferases, incl ud ing CT P synthetase and GMP synthetase. Intermediates Can Move Between Active Sites by Channel ing
Carbamoyl phosphate sy nthetase contains three different active sites (see Figure 25.3), separated fro m one another by a total of 80 A (F igure 25 .4 ).
Carbamic acid phosphorylation site
-l:l
Figure 25.3 Structure of carbamoyl phosphate synthetase. Not ice that the enzyme contains sites for t hree react ions. Th is enzyme consist s o f t wo chai ns. The smaller chain (yellow) contains a site f o r glutamin e hyd rolysis t o generate ammonia. The larger chain incl udes two ATP-grasp do mai ns (blue and red). In o ne ATP-grasp doma in (blue), bicarbo nate is phospho ry lated t o carboxyphosphat e, w hich then reacts wi th ammonia t o generate carbami c acid. In the other ATPgra sp domai n. th e carbam ic acid is phospho ry lat ed to produce carbamoyl phosphat e. [Drawn fro m 1JDB.pdb.)
712 CHAPTER 25 Nucleotid e Bi osynthesis Glutamine
~
Figure 25.4 Substrate channel ing. The three acti ve sites of carbamoyl phosphate synthetase are linked by a channel (yel low) through whic h intermediates pass. Gl utam ine enters one active site. and carbamoyl phosphate. which incl udes the nitrogen ato m from the gl utamine side chain. leaves • another 80 A away. [Drawn from lJDB.pdb.]
Carbam ic acid
Carba moyl ph osphate
~---:
Intermediates generated at one site move to the next without leaving the enzyme. These intermediates move within the enzy me by means of substrate channeling, similar to the process described for tryptophan synthetase (p . 696). The ammonia generated in the glutamine-hydrolysis active site travels 4S A throu gh a channel within the enzyme to reach the site at which . The carbamic acid generated at this carboxyphosphate has been generated o site d iffuses an additional 3S A through an extension of the channel to reach the site at which carbamoyl phosphate is generated. This channeling serves two roles: (1) intermediates generated at one active site are captured with no loss caused by d iffusion and (2) lab ile intermediates, such as carboxyphosphate and carbamic acid (w hich decompose in less than 1 s at pH 7), are protected from hydrolysis. We will see additional examples of substrate channeling later in this chapter. Orotate Acquires a Ribose Ring from PRPP to Form a Pyrimidine Nucleotide and Is Converted into Uridylate
Carbamoyl phosphate reacts with aspartate to form carbamoylaspartate in a reaction catalyzed by aspartate transcarbamoylase (Section 10.1). C arbamoylaspartate then cyclizes to form di hydroorotate, which is then oxidized by N AD + to form orotate .
•
o
o
C
C
II
p.,
HN/
"'--NH 2
NADH
+
II
HN/
"'--NH
J
ooc-1 H Ca rbamoyl phosphale
A
H H
Carbamoylaspart.le
H
NAD+
" " '\
'~o
H H Dihydroorolale
,
W
\/
•
~o
OOC I
H Orotate
At this stage, orotate couples to ribose, in the form of 5-phosphoribosyl-lpyrophosphate ( P R PP), a form of ribose activated to accept nucleotide bases. PRPP is synthesized fro m ribose-S-phosphate, formed by the pentose phosphate pathway, by t he addition of pyrophosphate from ATP. Oro tate reacts with PRPP to form orotidylate, a pyrimidine nucleotide. T his reaction is driven by the hydrolysis of pyrophosphate. The enzyme
that catalyzes this addition, pyrimidine phosphoribosyltransferase, is homologous to a number of other phosphoribosyltransferases that add different groups to PRPP to form the other nucleotides. Orotidylate is then decarboxylated to form uridylate (UMP ), a major pyrimidine nucleo tide that is a precursor to RNA. This reaction is catalyzed by orotidylate decarboxylase.
o
C/o. 0
~
I o
H Orotate
+
HNI-
w
0=
co,
\
0= • 2-0 3POH 2C
/'
N'- -:
o
H
HO OH HO Orotidylate
OH
5-Phosphoribosyl-l-pyrophosphate (PRPP)
HO OH Uridylate
Orotidylate decarboxylase is one of the most proficient enzymes known. In its absence, decarboxylation is extremely slow and is estimated to take place once every 78 million years; with the enzyme present, it takes place approximately once per second, a rate enhancement of 10 17 -fold.
PP,
o 0=
Nucleotide Mono-, Di-, and Triphosphates Are Interconvertible
How is the other major pyrimidine ribonucleotide, cytidine, formed? It is synthesized from the uracil base of UMP, but the synthesis can take place only after UMP has been converted into UTP. Recall that the diphosphates and triphosphates are the active forms of nucleotides in biosynthesis and energy conversions. Nucleoside mono phosphates are converted into nucleoside triphosphates in stages . First, nucleoside mono phosphates are con verted into diphosphates by specific nucleoside monophosphate kinases that utilize ATP as the phosphoryl-group donor. For example, UMP is phosphorylated to UDP by UMP kinase. UMP + ATP
~,~'>
O~
HO
N--,{
OH
Orotidylate
UDP + ADP
Nucleoside diphosphates and triphosphates are interconverted by nucleoside diphosphate kinase, an enzyme that has broad specificity, in contrast with the monophosphate kinases. X and Y represent any of several ribonuc1eosides or even deoxyribonucleosides: XDP + YTP
~,~>
XTP + YDP
CTP Is Formed by Amination of UTP
After uridine triphosphate has been formed, it can be transformed into cytidine triphosphate by the replacement of a carbonyl group by an amino group. Gin + H2 0
t
GIU
NH,
0=
0,=
o
O~ ATP
N---'J
ADP
+
,
p.
HO
UTP
OH
HO
OH
CTP
713
Like the synthesis of carbamoyl phosphate, this reaction requires ATP and uses glutamine as the source of the amino group. The reaction proceeds through an analogous mechanism in which the 0 -4 atom is phosphorylated to form a reactive intermediate, and then the phosphate is displaced byammonia, freed from glutamine by hydrolysis. CTP can then be used in many biochemical processes, including RNA synthesis.
714 CHAPTER 25 Nucleotide Biosynthesis
C~' Aspartate
Glycine
{N
C N'O-Formyl'--> N(6 "';C- 7'- tetrahydrofolate 12 4 1 9 8 ( "'----C, 3 ,......C_ ( -Glutamine N NlO-Formyl- / N \ tetrahydrofolate ribose-P Purine
Glutamine
ring structure
IMP
ATP
25.2
Purine Bases Can Be Synthesized de Novo or Recycled by Salvage Pathways
Purine nucleotides can be synthesized in two distinct pathways. First, purines are synthesized de novo, beginning with simple starting materials such as amino acids and bicarbonate (Figure 25 .5)_ Unlike the bases of pyrimidines, the purine bases are assembled already attached to the ribose ring. Alternatively, purine bases, released by the hydrolytic degradation of nucleic acids and nucleotides, can be salvaged and recycled. Purine salvage pathways are especially noted for the energy that th ey save and the remarkable effects of their absence (p. 725).
GTP to RNA
Salvage Pathways Economize Intracellular Energy Expenditure dATP
dGTP to DNA
Figure 25.5 De novo pathway for purine nucleotide synthesis _ The origins o f t he atoms in the purine ring are indicated.
Free purine bases, derived from the turnover of nucleotides or from the diet, can be attached to PRPP to form purine nucleoside monophosphates, ill a reaction analogous to the formation of orotidylate. Two salvage enzymes with different specificities recover purine bases. Adenine phosphoribosyltransferase catalyzes the formation of adenylate (AMP ): Adenine + PRPP --+) adenylate + PP j whereas hypoxanthine-guanine phosphoribosyltransferase (HGPRT) catalyzes the formation of guanylate (GMP ) as well as inosinate (inosine monophosphate, IMP), a precursor of guanylate and adenylate _
o N
,;?'
"'---J N-
f '
N~
Hypoxanthine
HO Inosinate
OH
NH
Guanine + PRPP
) guanylate + PP j
Hypoxanthine + PRPP --+) inosinate + PP j Similar salvage pathways exist for pyrimidines . Pyrimidine phosphoribosyltransferase will reconnect uracil, but not cytosine, to PRPP. The Purine Ring System Is Assembled on Ribose Phosphate
De novo purine biosynthesis, like pyrimidine biosynthesis, requires PRPP but, for purines, PRPP provides the foundation on which the bases are constructed step by step. The initial committed step is th e displacement of pyrophosphate by ammonia, rather than by a preassembled base, to produce 5-phosphoribosyl-.l-amine, with the amine in the 13 configuration . Glutamine phosphoribosyl amidotransferase catalyzes this reaction. This enzyme comprises two domains: the first is homologous to the phosphori bosy Itransferases in salvage pathways, whereas the second produces ammonia from glutamine by hydrolysi s. However, this glutamine-hydrolysis domain is distinct from the domain that performs the same function in carbamoyl phosphate synthetase. In glutamine phosphoribosyl amidotransferase, a cysteine residue located at the amino terminus facilitates glutamine hydrolysis. To prevent wasteful hydrolysis of either substrate, the amidotransferase assumes the active configuration only on binding of both PRPP
and glutamine. As is the case with carbamoyl phosphate synthetase, the ammonia generated at the glu tamine-hydrolysis active site passes through a channel to reach PRPP without being released into solution .
0-........
HO
The Purine Ring Is Assembled by Successive Steps of Activation by Phosphorylation Followed by Displacement
OH PRPP
+ NH, --...
Glu
-> pp.o
+ H2 0
Gin
NH,
HO
OH
5-Phosphoribosyl -l -amine ATP
ADP
Nu
\/
Pi
\ /
,
Disp laceme nt
\ / Nu
De novo purine biosynthesis proceed s as follows (Figure 25 .6).
(3)
CD ATP
+ Gly
P·ribose-NH,
"'-
0) 0
ADP
II
NH3+
+
p.0
./
II
0
0
THF..... C
H ...........-N"-. .......... CH, P-rib ose C
H·,C." 'N· H H THF
"'-
II 0
./
+
p.0
ATP
,
H ...........-N"-. .......... CH, P-ribose C
II 0
Glydnamide
Formylglydnamide
•
ribonucleotide
ribonucleotide
H ...........-N"-. .......... CH,
~
P-ribose H,N
Phosphoribosylamin e
ADP
'H
H2O
+ ,
+ Gin
Glu
Formylglycinamidine ribonucleotide /
ATP ADP
....... + ATP
PI
+
o H,C
!N
H,C
N
\" C-
...........- "-. .{/ P-ribose C
0 C/· , -
b
--"'---- P-ribose C
®
NH, 5 ~ Amino i m i d az ol e
ribonucleotide
ATP
+
@
Asp ADP
+
p. 0
5-Aminoimidazole-
4-(N-succinylcarboxamide) ribonucleotide
Figure 25.6 De novo purine biosynthesis. P) Glycine is 1o coupled to t he amino gro up of phosphoribosylamine. (2) N _ Formyltetrahydrofolate (THF) transfers a formyl group to the amino group o f the glycine residue. (3) The inner amide group is phosphorylated and converted into an amidine by the addition of ammonia derived from glutamine. (4) An intramo lecular coupling reaction forms the five-membered imidazole ring. (5) Bicarbonate adds first to the exocyclic amino group and then to a carbon atom of the imidazole ring. (6) The imidazole carboxylate is phosphory lated, and the p hosphate is disp laced by the amino group of aspartate.
715
716 CHAPTER 25
1. The carboxylate group of a glycine residue is activated by phosphory. lation and then coupled to t he amino group of phosphoribosylamine. A new amide bond is formed , and the amino group of glycine is free to act as a nu· cleophile in the next step.
Nucleotide Bi osyn th esis
2. Formate is activated and then add ed to this amino group to form formylglycinamide ribonucleotide. In som e organism s, two distinct en· zym es can catal yze this step. O ne enzyme transfers the formyl group from 10 N -formy ltetrahydrofolate (p. 690). The other enzyme activates formate as formyl phosphate, which is added directly to the glycine amino group. The inner amide group is act ivated b y phosphorylation and then can· verted into an amidine by the add ition of ammonia d erived from glutamine.
3.
4. The prod uct of this reaction, formylglycinamidine ribonucleotide, cy· clizes to form the five -m embered imidazole ring found in purines. Although this cyclization is likely to be favorable thermodynamically, a molecul e of ATP is consumed to ensure irreversibility. The familiar pattern is repeated : a phosphoryl group from the ATP molecule activates the car· bonyl gro up and is displaced by the nitrogen atom attached to the ribose m olecul e. Cyclization is thus an intramolecular reaction in which the nucleophile and phosphate-activated carbon atom are present with in the same molecule.
5. Bi carbonate is activated by phosphorylation and t hen attacked by the exocyclic am ino group. The product of the reaction in step 5 rearranges to transfer the carboxylate group to the imidazole ring. Interestingly, mammals do not require ATP for this step; bicarbonate apparently attaches directly to the exocyclic amino group and is then transferred to the imidazole ring. 6. The imidazole carboxylate group is phosphorylated again and the phosphate group is d isplaced by the amino group of aspartate. Thus, a six· step process li nks glycine, formate, ammonia, bicarbonate, and aspartate to form an intermediate t hat contains all but two of the atoms necessa ry for the form ation of the purine ring. Three more steps complete ring construction (Figu re 25 .7). Fumarate, an intermediate in the citric acid cycle, is eli minated, leaving the nitrogen atom from aspartate joined to the imidazole ring . T he use of aspartate as an ooc. H
-
C.
!
N" P-ribose/
N
0
-H
OAG
(CH 2 )12 " CH 3
choline
N H
OH
".H "
Sphingomyelin HI-
OH OH (CH 2) 12
' CH 3
CH, OH
UOP-glucose
~ '
Gangliosides
Activated sugars
o
UOP
Ceramide
H
-'OJ
R
( CH VI2
' CH, Cerebroside
Figure 26.3 Synthesis of sphingolipids. Ceramide is t he starting point for the formation of sphingomyelin and gangliosides.
737
GalNAc
~4
~4
Gal
p3
--I Ceramide
(12,3
.r-... Gal
NAN Figure 26.4 Ganglioside GM , . This ganglioside consists of five monosaccharides linked to ceramide: o ne glucose (GIc) molecule, two galactose (Gal) molecules, one N -acetylgalactosamine (GaINAc) molecule, and o ne N -acetylneuraminate (NAN) molecule.
R2 = H, N-acetylneuraminate
R,
= OH, N-glycolylneuraminate
Gangliosides Are Carbohydrate-Rich Sphingolipids That Contain Acidic Sugars Gangliosides are the most complex sphingolipids. In a gangliuside, an oligosaccharide chain is linked to the terminal hydroxyl group of cerarnide by a glucose residue (Figure 26.4). This oligosaccharide chain contains at least one acidic sugar, either N -acetylneuraminate or N -glycolylneuraminate. These acidic sugars are called sialic acids. Their nine-carbon backbones are synthesized from phosphoenolpyruvate (a three-carbon unit) and N -acetylmannosamine 6-phosphate (a six-carbon unit). Gangliosides are synthesized by the ordered, step-by-step addition of sugar residues to ceramide. The synthesis of these complex lipids requires the activated sugars UDP-glucose, UDP-galactose, and UDP-N -acetylgalactosarnine, as well as the CMP derivative of N-acetylneurarninate. eMPN-acetylneuraminate is synthesized from CTP and N-acetylneuraminate. The sugar composition of the resulting ganglioside is determined by the specificity of the glycosyltransferases in the cell. More than 60 different gangliosides have been characterized (see Figure 26.3 for the structure of ganglioside GMt) . Ganglioside-binding by cholera toxin is the first step in the development of cholera, a pathological condition characterized by severe diarrhea. Gangliosides are also crucial for binding immune-system cells to sites of injury in the inflammatory response.
Sphingolipids Confer Diversity on Lipid Structure and Function
o (00-
OH
H
OH
H
H
(-
OH
II
(-
OH
The structures of sphingolipids and the more abundant glycerophospholipids are very similar. Given the structural similarity of these two types of lipids, why are sphingolipids required at all? Tndeed, the prefix "sphingo" was applied to capture the" sphinxlike" properties of this enigmatic class of lipids. Although the precise role of sphingolipids is not fi rmly established, progress toward sol ving the riddle of their function is being made. Most notably, sphingolipids may serve as a source of second messengers. For instance, ceramide derived from a sphingolipid may initiate programmed cell death in some cell types.
Respiratory Distress Syndrome and Tay-Sachs Disease Result from the Disruption of Lipid Metabolism ,
•
•
•
•
, ,
••
•
1
:> II'
,V,·
Figure 26.5 Lysosome with lipids. An electron micrograph of a lysosome containing an abnormal amount of lipid. [Courtesy of Dr. George Palade.]
738
Respiratory distress syndrome is a pathological condition resulting from a failure in the biosynthetic pathway for dipalmitoyl phos· phatidylcholine. This phospholipid, in conjunction with specific proteins and other phospholipids, is found in the extracellular fluid that sur· rounds the alveoli of the lung. Its function is to decrease the surface tension of the fluid to prevent lung collapse at the end of the expiration phase of breathing. Premature infants may suffer from respiratory dis· tress syndrome because their immature lungs do not synthesize enough dipalmitoyl phosphatidylcholine. Tay-Sachs disease is caused by a failure of lipid degradation: an inability to degrade gangliosides. Gangliosides are found in highest concentration in the nervous system, particularly in gray matter, where they constitute 6% of the lipids. Gangliosides are normally degraded inside Iysosomes by the sequen· tial removal of their terminal sugars but, in Tay-Sachs disease, this degradation does not take place. As a consequence, neurons become significantly swollen with lipid-filled Iysosomes (Figure 26.5). An affected infant dis· plays weakness and retarded psychomotor skills before 1 year of age. The child is demented and blind by age 2 and usually dies before age 3.
739
The ganglioside content of the brain of an infant with Tay-Sachs disease is greatly elevated. The concentration of ganglioside G M 2 is many times higher than normal because its terminal N -acetylgalactosamine residue is removed very slowly or not at all. The missing or deficient enzyme is a specific f3- Nacetylhexosaminidase.
26.2 Synthesis of Cholesterol
GalNac GalNac
'4 AA f-'-f'-{ Gal f-'-""" -
f--{ Ceramide
u2,3
Gal ~4
-4 Ceramide
u2,3 Cholesterol
NAN
NAN Canctioside CM>
Tay-Sachs disease can be diagnosed in the course of fetal development. Amniotic fluid is obtained by amniocentesis and assayed for J3-N-acetylhexosaminidase activity.
26.2
Cholesterol Is Synthesized from Acetyl Coenzyme A in Three Stages
We now turn our attention to the synthesis of the fundamental lipid cholesterol. This steroid modulates the fluidity of animal cell membranes (p. 343) and is the precursor of steroid hormones such as progesterone, testosterone. estradiol. and cortisol. All 27 carbon atoms of cholesterol are derived from acetyl GoA in a three-stage synthetic process (Figure 26.6).
"Cholesterol is the most highly decorated small molecule in biology. Thirteen Nobel Prizes have been awarded t o scientist s who devoted major parts of their careers to cho lest erol. Ever since it wa s isolated from gall st ones in 1784. cholestero l has exerted
an al most hypnotic fascinatio n for sc ie ntists fro m the most di verse areas of science and medicine. . .. Cho lesterol is a Janus-faced
molecule. The very property that makes it useful in cell membranes, namely its absolute insolubility in water, al so makes it lethal:' - MICHAEL BRO WN AND JOSEPH G OlDSITIN
Nobel Lectures (1985) ©
The Nobel Foundation. 1985
1. Stage one is the synthesis of isopentenyl pyrophosphate, an activated isoprene unit that is the key building block of cholesterol.
2. Stage two is the condensation of six molecules of isopentenyl pyrophosphate to form squalene. 3. Tn stage three. squalene cyclizes and the tetracyclic product is subsequently converted into cholesterol.
a
The first stage takes place in the cytoplasm. and the second two in the endoplasmic reticulum.
The Synthesis of Mevalonate. Which Is Activated as Isopentenyl
Pyrophosphate, Initiates the Synthesis of Cholesterol The first stage in the synthesis of cholesterol is the formation of isopentenyl pyrophosphate from acetyl CoA. This set of reactions starts with the for mation of 3-hydroxy -3-methylglutaryl CoA (HMG CoA) from acetyl CoA and acetoacetyl CoA. This intermediate is reduced to meva/onate for the synthesis of cholesterol (Figure 26.7) . Recall that, alternatively, mitochondrial 3-hydroxy-3-methylglutaryl CoA may be processed to form ketone bodies (p. 631) . The synthesis of mevalonate is the committed step in cholesterol formation . The enzyme catalyzing this irreversible step. 3-hydroxy-3-methylglutaryl GoA reductase (HMG -CoA reductase). is an important control site in cholesterol biosynthesis. as will be discussed shortly. 3-Hydroxy-3 -methylglutaryl CoA + 2 NADPH + 2 H + ) mevalonate + 2 NADP+
+
CoA
Figure 26.6 Labeling of cholesterol. Isot o p e- labeling experiments reveal the source of carbon atoms in c ho lestero l sy nthesized fro m acetate labeled in its methy l gro up (blue) or carboxylate atom (red).
CYTOPlASM
o
CoA
0
+
2H
,-/~'-.// --"" S/
H, O Acetoacetyl CoA
o /
,
-OOC
OH
o
Mev.lunate
~
o
3-Hydroxy3-methylglutaryl CoA (HMG-CoA)
COA
S
- OOC
2 NADPH
CoA
\, J,
+ "
2 NADP+
+
COA
H3
+
/
CoA
S
Acetyl CoA
Acetyl CoA
+ - OOC MITOCHONDRIA
~/
AcetCNtcetate
Figure 26.7 Fates of 3-hydroxy-3-methylglutaryl CoA. In the cytoplasm, HMG-CoA is converted into mevalo nate. In m itochondria, it is converted into acetyl CoA and acetoacetate.
HMG -CoA reductase is an integral membrane protein in the endoplasmic reticul um. Mevalonate is converted into 3 -isopenteny l pyrophosphate in three con· secutive reactions requiring ATP (Figure 26 .8). In the last step. the release of CO 2 yields isopentenyl pyrophosphate. an activated isoprene unit that is a key building block for many important biomolecules throughout the kingdoms of life.
Isoprene
cooATP
CH 3 '"
···· OH
Mevalonate
coo-
coo-
~
ADP
L
ATP
CH3 )
",'" .
OH
5-Phos phome'lalonate
~
L
cooATP
ADP )
",".
p.,
('>
~
ADP
L
,
co,
)
OH
5-Pyrophosphomevalonate
+
CH 3
"'"
/
,
H2 CH3
(;OP0 32-
3-lsopentenyl pyrophosphate
Figure 26.8 Synthesis of isopentenyl pyrophosphate. This activated intermediate is formed from mevalonate in three steps, t he last of which includes a decarboxylation.
Squalene (C 30) Is Synthesized from Six Molecules of Isopentenyi Pyrophosphate (Cs) Squalene is synthesized from isopentenyl pyrophosphate by the reaction sequence
This stage in the synthesis of cholesterol starts with the isomerization 01 isopentenyl pyrophosphate to dimethylallyl pyrophosphate.
Isopenlenyl pyrophosphate
740
Dimethylallyl pyrophosphate
l-Isopentenyl pyrophosphate
CH
OPO PO , -
I
'
2
,
"- C/ H2
... . + .. ·CH2 --'
R H Allytic substrate
_ _
CH, -jo
) R
H2
//'-~
H
CH,
H'
I
/
CH 2
- " " ' --
-jo )
R
H2C
I
~
~I
H
CH 2
H
H
Allylic carbocation
Ceranyl (or farnesyl) pyrophosphate
Figure 26.9 Condensation mechanism in cholesterol synthesis. The mechani sm for ioining dimethylallyl pyrophosphate and isopentenyl pyrophosphate to form geranyl pyrophosphate. The same m echani sm is used to add an additional isopentenyl pyrophosphate to form fa rnesy l pyrophosphate.
These two isomeric C s units (one of each type) condense to form a C lO compound : isopentenyl pyrophosphate attacks an allylic carbonium ion formed fro m dimethylallyl pyrophosphate to yield gerany l pyrophosphate (Figure 26.9). The same kind of reaction takes place again : ger anyl pyrophosphate is converted into an allylic carboniu m ion, w hich is attacked by isopentenyl pyrophosphate. T he resulting C 1 S compound is called farnesyl pyrophosphate. The same enzyme, gerany l transferase, catalyzes each of these condensations. The last step in the synthesis of squalene is a reductive tail-to-tail condensation of two molecules of farnesyl py rophosphate catalyzed by the endoplasmi c reticulum en zyme squalene synthase.
Dimethylanyl pyrophosphate
CH,
"""'/''-opo,opo,'~
2 Farnesyl pyrophosphate (C 1 S ) + NADPH ) squalene (C 30 ) + 2 PP j + NADP + + H +
,
pp.
lsopentenyl pyrophosphate
Geranyl pyrophosphate
CH,
The reactions leading from C s units to squalene, a C 10 isoprenoid, are summarized in Figure 26. 10.
"'----.//"
,
...... pp.
OPO, OPO,'-
lsopentenyl pyrophosphate
Squalene Cyclizes to Form Cholesterol The final stage of cholesterol biosynthesis starts with the cyclization of squalene (Figure 26.11) . Squalene is first activated by conversion into squalene epoxide (2 ,3 -oxidosqualene) in a reaction that uses O 2 and NADPH . Squalene epoxide is then cyclized to lanosterol by oxidosqualene cyclase. This remarkable transformation proceeds in a concerted fashion. The enzyme holds squalene epoxide in an appropriate conformation and initiates the reaction by protonating the epoxide oxygen. The carbocation formed spontaneou sly rearranges to produce lanosterol. Lanosterol is converted into cholesterol in a
CH,
H, Farnesyl pyrophosphate ____ Farnesyl pyrophosphate + NADPH
CH, CH,
Figure 26.10 Squalene synthesis. One molecule of dimethyallyl pyrophosphate and two molecules of isopentenyl pyrophosphate condense to form farnesyl pyro pho sphate. The tail -to- tail coupling of two molecules of farnesyl pyrophosphate yields squalene.
CH,
CH,
CH, Squalene
741
+
w +
•
H
NADPH
NADP+
+
+
0,
H2 0
\/
/',*--~ ' \ H \
,, )
H
\ HO
o Squalene
Squalene epoxide
H Protosterol cation
H
Figure 26.11 Squalene cyclization. The formati o n of the steroid nucleus from squalene begins with the formation of squalene epoxide. This intermediate is proto nated to form a carbocation that cyclizes to form a tetracyclic structure, which rearranges to form lanosterol.
HO
•
H
CH3 Lanosterol
multistep process by the removal of three methyl groups, the reduction of one double bond by NADPH, and the migration of the other double bond (Figure 26.12).
26.3
--•• •
H
--'.'.
H3
CH 3 Lanosterol
19 steps HCOOH + 2 CO 2
H Cholesterol
Figure 26.12 Cholesterol formation. Lanosterol is converted into cholesterol in a complex process.
742
The Complex Regulation of Cholesterol Biosynthesis Takes Place at Several Levels
Cholesterol can be obtained from the diet or it can be synthesized de novo, An adult on a low -ch olesterol diet typically synthesizes about 800 mg of cholesterol per day. The liver is the major site of cholesterol synthesis in mammals, although the intestine also forms significant amounts. The rate of cholesterol formation by these organs is highly responsive to the cellular level of cholesterol. This feedback regulation is mediated primarily by changes in the amount and activity of 3- hydroxy-3-methylglutaryl CoA reductase. As described earlier (p. 739), this enzyme catalyzes the formation of mevalonate, the committed step in chol esterol biosynthesis. HMG CoA reductase is controlled in multiple ways : 1. The rate of synthesis of reductase ml~NA is controlled by the sterol regulatory element binding protein (SREBP). This transcription factor binds to a short DNA sequence called the sterol regulatory element (SRE) on the 5' side of the reductase gene. It binds to the SRE when cholesterol levels are low and enhances transcription. In its inactive state, the SREBP resides in the endoplasmic reticul um membrane, where it is associated with the SRE13P cleavage activating protein (SCAP), an integral membrane protein. SCAP is the cholesterol sensor. When cholesterol levels fall, SCAP escorts SREBP in small membrane vesicles to the Golgi complex, where it is released from the membrane by two specific proteolytic cleavages (Figure 26.13). The released protein migrates to the nucleus and binds the SRE of the HMG -CoA reductase gene, as well as several other genes in the cholesterol biosynthetic
74 3
SREBP ~ /--.,
ER
SCAP
DNA-binding
26.3 Regulation of Cholesterol Synthesis
domain Cytoplasm Nucleus
Lumen
Cholesterol levels tan
Golgi
Metalloprotease Serine protease
pathway, to enhance transcription. When cholesterol levels rise, the proteolytic release of the SREBP is blocked, and the SREBP in the nucleus is rapidly degraded. These two events halt the transcription of genes of the cholesterol biosynthetic pathways.
2. The rate of translation of reductase mRNA is inhibited by nonsterol metabolites d erived from mevalonate as well as by dietary cholesterol. 3. The degradation of the reductase is stringently controlled. The enzyme is bipartite: its cytoplasmic domain carries out catalysis and its membrane domain senses signals that lead to its degradation . The membrane domain may undergo structural changes in response to increasing concentrations of sterols such as cholesterol that make the enzyme more susceptible to proteolysis. The reductase may be further degraded by ubiquitination and targeting to the 26S proteasome under some conditions (Section 23 .2). A combination of these three regulatory devices can alter the amount of enzyme in the cell more than 200 -fold .
4. Phosphorylation decreases the activity of the reductase. This enzyme, like acetyl eoA carboxylase (which catalyzes the committed step in fatty acid synthesis, Section 22 .5), is switched off by an AMP-activated protein kinase. Thus, cholesterol synthesis ceases when the ATP level is low. As we will see shortly, all four regulatory m echanisms are modulated by receptors that sense the presence of cholesterol in the blood .
Lipoproteins Transport Cholesterol and Triacylglycerols Throughout the Organism Cholesterol and triacylglycerols are transported in body fluids in the form of lipoprotein particles. Each particle consists of a core of hydrophobic lipids surrounded by a shell of more- polar lipids and proteins. The protein components of these macromolecular aggregates, called apoproteins, have two roles: they solubilize hydrophobic lipids and contain cell -targeting signals. Apolipoproteins are synthesized and secreted by the liver and the intestine.
Figure 26.13 The SREBP pathway. SREBP resides in the endoplasmic reticulum. where it is bound to SCAP by its regulatory (Reg) domain. When cholesterol levels fall , SCAP and SREBP move to the Golgi complex, where SREBP undergoes successive proteolytic cleavages by a serine protease and a metalloprotease. The released DNAbinding domain moves to the nucleus to alter gene expression. [After an illustration provided by Dr. M ichael Brown and Dr. Joseph Goldstein.]
TABLE 26.1 Pro perti es of plasma lipoproteins COMPOSITION (%) Plasma lipoproteins
Density (g ml - ')
Diameter (nm)
Apo lipoprotein
Physio logical ro le
TAG
< 0.95 0 .95- 1.006
75-1200 30- 80
848. C. E 6100, C. E
Dietar y rat transport
86 52
3 14
1 7
8 18
2 8
1.0 06 - 1.01 9
15- 35
8lDO, E
LDL precursor
38
30
8
23
II
1.019- 1.063 1.063-1.21
18-25 7.5- 20
8100 A
Cholest erol t ransport Reverse cholesterol
10 5- 10
38 14- 21
8 3- 7
22 19-29
21
Chylomicron Very low densi ty
Endogenous fat transport
lipoprotein
Ifltermed iate-density lipoprotein Low-densi ty lipoprotein High-density lipoprotein
CE
C
PL
P
33-57
transport Abbreviations: TAG, lliacylglyerol: CEocholesterol ester: C. free cholesterol; PL, phospholipid: P. protein.
Lipop rotein particles are classified according to increasing d ensity (Table
I
I
500 nm
Figure 26.14 Site of cholesterol synthesis. Electron micrograph o f a part o f a liver cell actively engaged in t he synthesi s and secreti o n o f very low density li poprotein (VLDL). The arrow po ints to a vesicl e t hat is rel easing its co ntent o f VLDL particles. [Co urtesy of Dr. George Palade.)
Un esterified cholesterol Phospholipid Cholesteryl ester Apoprotein 8- 100
744
26.1 ): chylomicrons, chylomicron remnants, very low density lipoproteins (VLDLs), intennediate-density lipoproteins (IDLs), low-density lipoproteins (LD L s), and high-density lipoproteins (HDLs). T riacylglycerols, cholesterol , and other lipids obtained from the diet are carried away from the intestine in the form of large chylomicrons. T hese particles have a very low density because triacylglycerols constitute ~ 90% of t heir content . Apolipop rotein ~ - 48 (apo B-48), a large protein (240 kd), form s an amphipathic spherical shell around the fat globule; the external face of this shell is hydrophilic, The triacylglycerols in chylomicrons are released through hydrolysis by lipoprotein lipases. These enzymes are located on the lining of blood vessels in muscle and other tiss ues that use fatty acids as fuels and in the syn thesis of lipids. T he li ver t hen takes up the cholesterol-rich residues, known as chylomicron remnants. T he liver is a m ajor site of triacylglycerol and cholesterol synthesis (Figure 26 .14). T ri acylglycerols and cholesterol in excess of the liver's own need s are exported into the blood in the form of very low density lipoproteins. T hese particles are stabilized by two apolipoproteins apo 8 -100 and apo E (34 kd). Apo B-100, one of the largest proteins known (513 kd), isa longer version of apo B-48, Both apo R proteins are encoded by the same gene and produced from the same initial RN A transcript. In the intestine, RNA editing (Section 29.3) modifies the transcript to generate the mRNA for apo B-48, the truncated form. T riacy lglycerols in very low density lipop roteins, as in chylomicrons, are hydrolyzed by lipases on capillary surfaces . T he resulting remnants, which are rich in cholesteryl esters, are called intermediate-density lipoproteins. These particles have two fates. Half of them are taken up by the liver for processing, and half are converted into low-density lipoprotein by the removal of m ore triacylglycerol. Low-density lipoprotein is the major carrier of cholesterol in blood (Figure 26.15 ). It contains a core of some 1500 cholesterol molecules esteri fied to fatty acids; the most common fatty acid chain in these esters is linoleate, a polyunsat urated fatty acid. A shell of phospholipids and unesterified cholesterol molecules surrounds this hi ghl y hydrophobic core. The shell also contains a single cop y of apo B-100, whi ch is recognized b y target cells. The role of L D L is to transport cholesterol to peripheral tissues and regulate de novo cholesterol synthesis at these sites, as described on page 745 . A different purpose is served by high-density lipoprotein , which picks up cholesterol released into t he plasm a from dying cells and from membranes undergoing turnover, a process termed reverse cholesterol transport. An acyltransferase in HDL esterif ies these cholesterols, which are then returned by H DL to the liver (Figure 26. 16). ~ Figure 26.15 Schematic model of low-denSity lipoprotein. The LDL particle is o
appro ximately 22 nm (220 A) in d iamet er.
r-, Intestine
745 Dietary fat + cholesterol
(
26.3 Regulation of Cholesterol Synthesis
» Chylomicron _ _
_ _ _~) Chylomicrons _ _"""_ _ "" FFA Adipose ..-...... tissue Bile acids
.,----
remnants
HDL
Peripheral tissues (with LOL receptors)
FFA k'" FFA
lIVER
--'») VLOL --./""----) IDL (
) LOL
I
Figure 26.16 An overview of lipoprotein particle metabolism. Fatty acids are abbreviated FFA. [After J. G. Hardman (Ed.). L. L. Limbird (Ed.). and A. G. Gilman (Consult. Ed.). Goodman and Gilman's The Pharmacological Basis of Therapeutics. 10th ed. (McGraw-Hili. 2001). p. 975. Fig. 361]
The Blood Levels of Certain Lipoproteins Can Serve Diagnostic Purposes
High serum levels of cholesterol cause disease and death by contributing to the formation of atherosclerotic plaques in arteries throughout the body. This excess cholesterol is present in the form of the low-density lipoprotein particle. so-called bad cholesterol. High-density lipoprotein is sometimes referred to as "good cholesterol." HDL functions as a shuttle that moves cholesterol throughout the body. HDL binds and esterifies cholesterol released from the peripheral tissues and then transfers cholesteryl esters to the liver or to tissues that use cholesterol to synthesize steroid hormones. A specific receptor mediates the docking of the HDL to these tissues. The exact nature of the protective effect of HDL levels is not known; however. a possible mechanism will be examined on page 747. The ratio of cholesterol in the form ofLDL to that in the form ofHDL can be used to evaluate susceptibility to the development of heart disease. For a healthy person. the HDL/LDL ratio is 3.5. Low-Density Lipoproteins Playa Central Role in Cholesterol Metabolism
Cholesterol metabolism must be precisely regulated to prevent atherosclerosis. The mode of control in the liver. the primary site of cholesterol syn thesis. has already been considered: dietary cholesterol reduces the activity and amount of 3-hydroxy -3-methylglutaryl eoA reductase. the enzyme catalyzing the committed step. Studies by Michael Brown and Joseph Goldstein are sources of insight into the control of cholesterol metabolism in nonhepatic cells. In general. cells outside the liver and intestine obtain cholesterol from the plasma rather than synthesizing it de novo. Specifically. their primary source of cholesterol is the low -density lipoprotein. The process ofLDL uptake. called receptor-mediated endocytosis. serves as a paradigm for the uptake of many molecules. The steps in the receptor-mediated endocytosis of LDL are as follows (see Figure 12.37).
746 CHAPTER 26 The Biosynthesis of Membrane Lipids and Steroids
1. Apolipoprotein B-100 on the surface of an LDL particle binds to a spe· cific receptor protein on the plasma membrane of nonhepatic cells. The recep· tors for LDL are localized in specialized regions called coated pits, which con· tain a specialized protein called clathrin. 2. The receptor- LDL complex is internalized by endocytosis; that is, the plasma membrane in the vicinity of the complex invaginates and then fuses to form an endocytic vesicle (Figure 26.17). 3. These vesicles, containing LDL, subsequently fuse with lysosames, acidic vesicles that carry a wide array of degradative enzymes. The protein component ofLDL is hydrolyzed to free amino acids. The cholesteryl esters in LDL are hydrolyzed by a lysosomal acid lipase. The LDL receptor itself usually returns unscathed to the plasma membrane. The round-trip time for a receptor is about 10 minutes; in its lifetime of about a day, it may bring many LDL particles into the cell.
(A)
(B) Figure 26.17 Endocytosis of LDL bound to its receptor. (A) Electron micrograph showing LDL (conjugated to ferritin for visualization. dark spots) bound to a coated-pit region on the surface of a cultured human fibroblast cell. (B) Micrograph showing this region invaginating and fUSing to form an endocytic vesicle. [From R. G. W. Anderson. M. S. Brown. and J. l. Goldstein.
CeI/10(1977);3S1- 364.]
Figure 26.18 LDL receptor domains. A schematic representation of the amino acid sequence of the LDL receptor showing six types of domain.
4. The released unesterified cholesterol can then be used for membrane biosynthesis. Alternatively, it can be reesterified for storage inside the cell. In fact, free cholesterol activates acyl CoA:cholesterol acyltransferase (ACAT), the enzyme catalyzing this reaction_ Reesterified cholesterol contains mainly oleate and palmitoleate, which are monounsaturated fatty acids, in contrast with the cholesterol esters in LDL, which are rich in linoleate, a polyunsaturated fatty acid (see Table 12_1). It is imperative that the cholesterol be reesterified. High concentrations of unesterified cholesterol disrupt the integrity of cell membranes. The synthesis of the LDL receptor is itself subject to feedback regulation. Studies of cultured fibroblasts show that, when-cholesterol is abundant inside the cell , new LDL receptors are not synthesized, and so the uptake of additional cholesterol from plasma LDL is blocked. The gene for the LDL receptor, like that for the reductase, is regulated by SREBP, which binds to a sterol regulatory element that controls the rate of mRNA synthesis.
The LDL Receptor Is a Transmembrane Protein Having Six Different Functional Regions The amino acid sequence of the human LDL receptor reveals the mosaic structure of this 115-kd protein, which is composed of six different types of domains (Figure 26.18). The amino-terminal region of the mature receptor is the site ofLDL binding. It consists of a cysteine-rich sequence of about 40 residues that is repeated, with some variation, seven times. A second type of domain in the LDL receptor is homologous to one found in the epidermal growth factor (EGF). This domain is repeated three times, and in between the second and third repeat are six repeats of a third domain that is similar to the blades of the transducin J3 subunit (p. 270). The six repeats form a propelJer-like structure that packs against one of the EGF-like domains (Figure 26.19). An aspartate residue forms hydrogen bonds that hold each blade to the rest of the str ucture. Exposure to the lowpH environment of the lysosomes causes the propeller-like structures to interact with the LDL-binding domain. This interaction displaces the LDL, which is then digested by the lysosome.
D Cysteine-rich
D Hydrophobic
o
EGF-like Cytoplasmic
Blade
DO-linked glycosylated
The final three domains are represented by a single copy apiece. The fourth domain, which is very rich in serine and threonine residues, contains O-linked sugars. These oligosaccharides may function as struts to keep the receptor extended from the membrane so that the LDL-binding domain is accessible to LDL. The fifth type of domain consists of 22 hydrophobic Key Asp residues that span the plasma membrane. The sixth and final do- residues main consists of 50 residues and emerges on the cytoplasmic side of the membrane, where it controls the interaction of the receptor with coated pits and participates in endocytosis. The gene for the LDL receptor consists of 18 exons, which correspond closely to the structural units of the protein. The LDL receptor is a striking example of a mosaic protein encoded by a gene that was assembled by exon shuffling. The Absence of the LDL Receptor Leads to Hypercholesterolemia and Atherosclerosis
•
EGF-like domain
W
Brown and Goldstein's pioneering studies of familial hypercholesI;ji terolemia revealed the physiological importance of the LDL receptor. The total concentration of cholesterol and LDL in the blood plasma is markedly elevated in this genetic disorder, which results from a mutation at asingle autosomal locus. The cholesterol level in the plasma of homozygotes is typically 680 mg dl - I , compared with 300 mg dl - I in heterozygotes (clinica! assay results are often expressed in milligrams per deciliter, which is I equal to milligrams per 100 milliliters). A value of < 200 mg dl - is regarded as desirable, but many people have higher levels. In familial hypercholesterolemia, cholesterol is deposited in various tissues because of the high concentration of LDL cholesterol in the plasma. Nodules of cholesterol called xanthomas are prominent in skin and tendons. Of particular concern is the oxidation of the excess blood LDL to form oxidized LDL (oxLDL). The oxLDL is taken up by immune-system cells called macrophages, which become engorged to form foam cells. These foam cells become trapped in the walls of the blood vessels and contribute to the formation of atherosclerotic plaques that cause arterial narrowing and lead to heart attacks (Figure 26.20). In fact, most homozygotes die of coronary artery disease in childhood. The disease in heterozygotes (1 in 500 people) has a milder and more variable clinical course. A serum esterase that degrades oxidized lipids is found in association with HDL. Possibly, the HDL-associated
(A)
(8)
""1m. Figure 16.19 Structure of propeller '0 domain. Notice the six-bladed
propeller- like shape of this domain (red). The propeller domain is adjacent to an EGF- like domain (blue) of the LDL receptor. [Drawn from lIJQ.pdb.]
Figure 16.10 The effects of excess cholesterol. Cross secti o n or (A) a normal artery and (B) an artery blocked by a cholesterol -rich plaque. [SPL/ Photo Researchers.]
747
748 CHAPTER 26 The Biosynthesis of Membrane Lipids and Steroids
protein destroys the oxLDL, accounting for HDL's ability to protect against coronary disease. The molecular defect in most cases offamilial hypercholesterolemia is an abo sence or deficiency of functional receptors for LDL. Receptor mutations that disrupt each of the stages in the endocytotic pathway have been identified. Homozygotes have almost no functional receptors for LDL, whereas het· erozygotes have about half the normal number. Consequently, the entry of LDL into liver and other cells is impaired, leading to an increased level of LDL in the blood plasma. Furthermore, less IDL enters liver cells because IDL entry, too, is mediated by the LDL receptor. Consequently, I DL stays in the blood longer in familial hypercholesterolemia, and more of it is con· verted into LDL than in normal people. All deleterious consequences of an absence or deficiency of the LDL receptor can be attributed to the ensuing elevated level ofLDL cholesterol in the blood . The Clinical Management of Cholesterol Levels Can Be Understood at a Biochemical Level
HJ.... y~ coo- ~
OH
-
o
••• • •
H •• •
-
Lovastatin
Figure 26.21 Lovastatin, a competitive inhibitor of HMG-CoA reductase. The part of the structure that resembles the 3hydroxy-3-methylglutaryl moiety is shown in red.
Homozygous familial hypercholesterolemia can be treated only by a liver transplant. A more generally applicable therapy is available for heterozygotes and others with high levels of cholesterol. The goal is to reduce the amount of cholesterol in the blood by stimulating the single normal gene to produce more than the customary number of LDL receptors. We have already observed that the production of LDL receptors is controlled by the cell's need for cholesterol. Therefore, in essence, the strategy is to de· prive the cell of ready sources of cholesterol. When cholesterol is required, the amount of mRNA for the LDL receptor rises and more receptor is found on the cell surface . This state can be induced by a two-pronged ap· proach. First, the reabsorption of bile salts from the intestine is inhibited. Bile salts are cholesterol derivatives that promote the absorption of dietary cholesterol and dietary fats (p. 619). Second, de novo synthesis of choles· terol is blocked. The reabsorption of bile is impeded by oral administration of positively charged polymers, such as cholestyramine, that bind negatively charged bile salts and are not themsel ves absorbed. Cholesterol synthesis can be ef. fectively blocked by a class of compounds called statins. A well-known ex· ample of such a compound is lovastatin, which is also called mevacor (Figure 26.21). These compounds are potent competitive inhibitors (K; = 1 nM) of HMG -CoA reductase, the essential control point in the biosynthetic pathway. Plasma cholesterol levels decrease by 50% in many patients given both lovastatin and inhibitors of bile-salt reabsorption. Lovastatin and other inhibitors of HMG-CoA reductase are widely used to lower the plasma-cholesterol level in people who have atherosclerosis, which is the leading cause of death in industrialized societies. The development of statins as effective drugs is further described in Chapter 35.
26.4
Important Derivatives of Cholesterol Include Bile Salts and Steroid Hormones
Cholesterol is a precursor for other important steroid molecules: the bile salts, steroid horIllones, and vitamin D. Bile Salts. Bile salts are polar derivatives of cholesterol. These compounds are highly effective detergents because they contain both polar and nonpolar regions. Bile salts are synthesized in the liver, stored and concentrated in the
749
Cholesterol
26.4 Synthesis of Steroid Hormones
o
o N H
CH,
'"·"···OH
HO
.' .'
."
~/ H:'----/"····· 0 H
Glycocholate
gall bladder, and then released into the small intestine. Bile salts, the major constituent of bile, solubilize dietary lipids (p. 619). Solubilization increases the effective surface area of lipids with two consequences: (1) more surface area is exposed to the digestive action oflipases and (2) lipids are more read ily absorbed by the intestine. Bile salts are also the major breakdown products of cholesterol. The bile salts glycocholate, the primary bile salt, and taurocholate are shown in Figure 26.22.
Taurocholate
Figure 26.22 Synthesis of bile salts. The OH groups in red are added to cholesterol. as are the groups shown blue.
Steroid Hormones. Cholesterol is the precursor of the five major classes of steroid hormones: progestagens, glucocorticoids, mineralocorticoids, androgens, and estrogens (Figure 26. 23). These hormones are powerful signal molecules that regulate a host of organismal functions. Progesterone, a progestagen, prepares the lining of the uterus for the implantation of an ovum. Progesterone is also essential for the maintenance of pregnancy. Androgens (such as testosterone) are responsible for the development of male secondary sex characteristics, whereas estrogens (such as estrone) are required for the Cholesterol (C 27) development of female secondary sex characteristics. Estrogens, along with progesterone, also participate in the ovarian cycle. Glucocorticoids (such as cortisol) promote gluconeogenesis and the formation of glycogen, enhance Pregnenolone (C 2 ,) the degradation of fat and protein, and inhibit the inflammatory response. They enable animals to respond to stress; indeed, the absence of glucocorticaids can be fatal . Mineralocorticoids (primarily aldosterone) act on the distal Progestagens (C2 ,) tubules of the kidney to increase the reabsorption ofNa + and the excretion of K + and H + , which leads to an increase in blood volume and blood pressure. The major sites of synthesis of these classes of hormones are the corpus luteum, for progestagens; the testes, for androgens; the ovaries, for estrogens; and the adrenal cortex, for glucocorticoids and Glucocorticoids Androgens (C2 ,) Mineralocorticoids mineralocorticoids. (C' 9) (C ,) 2 Steroid hormones bind to and activate receptor molecules that serve as transcription factors to regulate gene Estrogens expression (Section 31.3). These small similar molecules (C 18) are able to have greatly differing effects because the slight structural differences among them allow interactions Figure 26.23 Biosynthetic relations of classes of steroid hormones with specific receptor molecules. and cholesterol.
H,C21 ""'" 22 1B CH, 20
750 CHAPTER 26 The Biosynthesis of Membrane Lipids and Steroids
12 11
19CH,
.........-IIi'-....,. /
9
2
A
10
3
4
5 ~
H
B
C
23
24
25
CH, 26
27 CH, 13
14
B 7
6
Figure 26.24 Cholesterol carbon numbering. The numbe ring scheme for the carbon atoms in cholesterol and other ste roids.
Hyd roxyl group above plane
Letters Identify the Steroid Rings and Numbers Identify the Carbon Atoms CH,
/'./ ,~
HO
lll-Hydroxy
Hydroxyl group below plane
CH,
r ...""
./
_
HO'
l a-Hydroxy
C arbon atoms in steroids are numbered, as shown for cholesterol in Figure 26.24. The rings in steroids are denoted by the letters A, B, C , and D. C holesterol contains two angular methyl groups: the C-19 methyl group is attached to C-10, and the C -18 methyl group is attached to C- 13. TheC-18 and C -19 methyl groups of cholesterol lie abmJe the plane containing the four rings. A substituent that is above the plane is termed f3 oriented, whereas a substituent that is below the plane is a oriented. If a hydrogen atom is attached to C -S, it can ,be either a or 13 oriented. The A and B steroid rings are fused in a trans conformation if the C-S hydrogen is a oriented, and cis if it is f3 oriented. The absence of a Greek letter for the C-S hydrogen atom on the steroid nucleus implies a trans fu sion. The C-S hydrogen atom is a oriented in aU steroid horIllones that contain a hydrogen atom in that position. I n contrast, bile salts have a l3-oriented hydrogen atom at C -S. Thus, a cis fusion is characteristic of the bile salts, whereas a trans fusion is characteristic of all steroid h01mones that possess a hydrogen atom at C-S . A trans fusion yields a nearly planar structure, whereas a cis fusion gives a buckled structure.
~
CH, ./
.---t-.i'-....,./ H
. H
-
51l-Hydrogen
Sa-Hydrogen
(cis fusion)
(trans fusion)
Steroids Are Hydroxylated by Cytochrome P450 Monooxygenases That Use NADPH and O 2 The addition of OH groups plays an important role in t he synthesis of cholesterol from squalene and in the conversion of cholesterol into steroid hormones and bile salts. AU these hydroxylations require NADPH and O2, The oxygen atom of the incorporated hydroxyl group comes from O 2 rather than from H 2 0 . Whereas one oxygen atom of the O 2 molecule goes into the substrate, the other is reduced to water. The enzymes catalyzing these reactions are called monooxygenases (or mixed-function oxygenases). Recall that a monooxygenase also participates in the hydroxylation of aromatic amino acids (p. 671 ).
75 1
ROH
26.4 Synthesis of Steroid Hormones
Substrate
< Fe3
+,
RH RH
RH0 3 + ,... II ~
/::-_--- ( Adrenodoxin ) e- 4( . .
H20
RH
0/
0 2-
RH
I
< Fe3+ ,
e( Adrenodoxin ) ___.?f
Hydroxylation requires the activation of oxygen. In the synthesis of steroid hormones and bile salts, activation is accomplished by members of the cytochrome P450 family, a family of cytochromes that absorb light maximally at 450 nm when complexed in vitro with exogenous carbon monoxide. These membrane-anchored proteins (- 50 kd ) contain a heme prosthetic group. Oxygen is activated through its binding to the iron atom in the heme group. Because the hydroxylation reactions promoted by P450 enzymes are oxidation reactions, it. is at first glance surprising that they also consume the reductant NADPH. NADPH transfers its high-potential electrons to a flavoprotein, which transfers them, one at a time, to adrenodoxin, a nonheme iron protein. Adrenodoxin transfers one electron to reduce the ferric 3 2 (Fe I ) form ofP450 to the ferrous (Fe + ) form (Figure 26.25). Without the addition of this electron, P450 will not bind oxygen. Recall that only the ferrous form of myoglobin binds oxygen (p. 185). The binding of O2 to the heme is followed by the acceptance of a second electron from adrenodoxin . The acceptance of this second electron leads to cleavage of the 0-0 bond. One of the oxygen atoms is then protonated and released as water. The remaining oxygen atom forms a highly reactive ferryl Fe 0 intermediate. This intermediate abstracts a hydrogen atom from the substrate RH to form R". This transient free radical captures the OH group from the iron atom to form ROH, the hydroxylated product, returning the iron atom to the ferric state.
The Cytochrome P450 System Is Widespread and Performs a
Protective Function The cytochrome P450 system, which in mammals is located primarily in the endoplasmic reticulum of the liver and small intestine, is also important in the detoxification of foreign substances (xenobiotic compounds). For example, the hydroxylation of phenobarbital, a barbiturate, increases its solubility and facilitates its excretion. Likewise, polycyclic aromatic hydrocarbons that are ingested by drinking contaminated water are hydroxylated by P450, providing sites for conjugation with highly polar units (e.g., glueuronate or sulfate) that markedly increase the solubility of the modified
Figure 26.25 Cytochrome P450 mechanism. These enzymes bind O 2 and use one oxygen atom to hydroxylate their substrates.
752
aromatic molecule. One of the most relevant functions of the cytochrome P450 system to human beings is its role in metabolizing drugs such as caffeine and ibuprofen (Chapter 35). Some members of the cytochrome P450 system also metabolize ethanol (Section 27 .5). The duration of action of many medications depends on their rate of inactivation by the P450 system. Despite its general protective role in the removal of foreign chemicals, the action of the P450 system is not always beneficial. Some of the most powerful carcinogens are generated from harmless compounds by the P450 system in vivo in the process of metabolic activation. In plants, the cytochrome P450 system plays a role in the synthesis of toxic compounds as well as the pigments of flowers_
CHAP I ER 26 The Biosynthesis of Membrane lipids and Steroids
The cytochrome P450 system is a ubiquitous superfamily of monooxygenases that is present in plants, animals, and prokaryotes. The human genome encodes more than 50 members of the family, whereas the genome of the plant Arabidopsis encodes more than 250 members. All members of this large family arose by gene duplication fol lowed by subsequent divergence, which generated a range of substrate specificity. The specificity of these enzymes is encoded in delimited regions of the pri mary structure, and the substrate specificity of closely related members is often defined by a few critical residues or even a single amino acid.
Pregnenolone, a Precursor for Many Other Steroids, Is Formed from Cholesterol by Cleavage of Its Side Chain ,
Steroid hormones contain 21 or fewer carbon atoms, whereas cholesterol contains 27 . Thus, the first stage in the synthesis of steroid hormones is the removal of a six -carbon unit from the side chain of cholesterol to form pregnenolone. The side chain of cholesterol is hydroxylated at C -20 and then at C-22, and the bond between these carbon atoms is subsequently cleaved by desmolase . Three molecules of N ADPH and three molecules of O 2 are consumed in this remarkable six-electron oxidation. OH HO
-D
)
Cholesterol
)
20a.22P-Dihydroxycholesteroi
Plegl1enolone
Progesterone and Corticosteroids Are SyntheSized from Pregnenolone Progesterone is synthesized from pregnenolone in two steps. The 3-hydroxyl group of pregnenolone is oxidized to a 3-keto group, and the ~ 5 double 4 bond is isomerized to a ~ double bond (Figure 26.26). Cortisol , the major glucocorticoid, is synthesized from progesterone by hydroxylations at C-17, C -21, and C - II; C- 17 must be hydroxylated before C -2 1 is hydroxylated, whereas C - II can be hydroxylated at any stage_ The enzymes catalyzing these hydroxy lations are highly specific, as shown by some inherited disorders. The initial step in the synthesis of aldosterone, the major mineralocorticoid, is the hydroxylation of progesterone at C -21. The resulting deoxycorticosterone is hydroxylated at C-II. The oxidation oftheC-18 angular methyl group to an aldehyde then yields aldosterone.
21
o
HOH 2 CH J
"".... OH ----:: 17
II
,
, 3
H Pregnenolone
Cortisol
Progesterone
21
21
HO H 2
HO H2
O~ C/ H
CH J
H
H
II ... . •••
1I
CH J
CH J
11
OH
18
, Figure 26.26 Pathways for the formation of progesterone, cortisol, and aldosterone.
Corticosterone
Aldosterone
Androgens and Estrogens Are Synthesized from Pregnenolone
Androgens and estrogens also are synthesized from pregnenolone through the intermediate progesterone. Androgens contain 19 carbon atoms. The synthesis of al!drogens starts with the hydroxylation of progesterone at C-17 (Figure 26.27). The side chain consisting of C-20 and C- 21 is then cleaved to yield androstenedione, an androgen. Testosterone, another androgen, is formed by the reduction of the 17 -keto group of androstenedione.
Progesterone
,
o
,
, Androstenedione
17o-Hydroxyprogesterone
Figure 26.27 Pathways for the formation for androgens and estrogens.
H
Testosterone
HO Estrone
Estradiol
753
Testosterone, through its actions in the brain, is paramount in the develop· ment of male sexual behavior. It is also important for the maintenance of the testes and the development of muscle mass. Owing to the latter activity, testosterone is referred to as an anabolic steroid. Testosterone is reduced by Sa- reductase to yield dihydrotestosterone (DHT), a powerful embryonic an· drogen that instigates the development and differentiation of the male phe· notype. Estrogens are synthesized from androgens by the loss of the C·1 9 angular methyl group and the formation of an aromatic A ring. Estrone, an estrogen, is derived from androstenedione, whereas estradiol , another estro· gen, is formed from testosterone.
754 CHAPTER 26 The Biosynthesis of Membrane Lipids and Steroids
Vitamin D Is Derived f rom Cholestero l by the Ring-Splitting Activity of Light C holesterol is also the precursor of vitamin D, which plays an essential role in the control of calcium and phosphorus metabolism . 7-Dehydrocholesterol (provi tamin D 3) is photolyzed by the ultraviolet light of sunlight to previtamin 0 3 , which spontaneously isomerizes to vitamin 0 3 (Figure 26. 28) . Vitamin D3 (cholecalciferol) is converted into calcitriol (1 ,25-dihydroxycholecalciferol), the active hormone, by hydroxylation reactions in the liver and kidneys. Although not a steroid, vitamin D acts in an analogous fashion. It binds to a receptor, structurally similar to the steroid receptors, to form a complex that functions as a transcription factor, regulating gene expression . Vitamin D deficiency in childhood produces rickets, a disease char· acterized by inadequate calcification of cartilage and bone. Rickets was so common in seventeenth-century England that it was called the "chil·
HC 3 . .., '
"1,
CH 3 CH 3
CH 3
,/
CH3 /'--..., •
H
~
Ultraviolet
(
light
•
)
7-Dehydroehole5terol
Previtamin D]
(
CH 2
HO
.... ""'''-v/
~
CH 2
~
HO '"
OH Calcitriol
(' .25-Dihydroxycholecalciferol)
,','
.....
~/
Vitamin D] (Cholecalciferol)
Figure 26,28 Vitamin D synthesis. The pathway for the conversion o f 7-dehydrocholesterol into vitamin Dj and then into ca!citrio l, the active hormone,
dren 's disease of the English." The 7 -dehydrocholesterol in the skin of these children was not photolyzed to previtamin lJ 3 , because there was little sunlight for many months of the year. Furthermore, their diets provided little vitamin D , because most naturally occurring foods have a low content of this vitamin. Fish-liver oils are a notable exception. Cod -liver oil, abhorred by generations of children because of its unpleasant taste, was used in the past as a rich source of vitamin D . Today, the most reliable dietary sources of vitamin D are fortified foods. Milk, for example, is fortified to a level of 400 international units per quart (10 f.Lg per quart) . The recommended daily intake of vitamin D is 200 international units until age 50, after which it increases with age. In adults, vitamin D deficiency leads to softening and weakening of bones, a condition called osteomalacia. The occurrence of osteomalacia in Muslim women who are clothed so that only their eyes are exposed to sunlight is a striking reminder that vitamin D is needed by adults as well as by children .
Summary 26.1 Phosphatidate Is a Common Intermediate in the Synthesis of Phospholipids and Triacylglycerols Phosphatidate is formed by successive acylations of glycerol3-phosphate by acyl CoA . Hydrolysis of its phosphoryl group followed by acylation yields a triacylglycerol. CDP-diacylglycerol, the activated intermediate in the de novo synthesis of several phospholipids, is formed from phosphatidate and CTP. The activated phosphatidyl unit is then transferred to the hydroxyl group of a polar alcohol, such as inositol, to form a phospholipid such as phosphatidylinositol. In mammals, phospha,tidylethanolamine is formed by CDP -ethanolamine and diacylglycerol. Phosphatidylethanolamine is methylated by Sadenosylmethionine to form phosphatidylcholine. In mammals, this phosphoglyceride can also be synthesized by a pathway that utilizes dietary choline. CDP -choline is the activated intermediate in this route. Sphingolipids are synthesized from ceramide, which is formed by the acylation of sphingosine. Gangliosides are sphingolipids that contain an oligosaccharide unit having at least one residue of N -acetylneuraminate or a related sialic acid . They are synthesized by the step-by -step addition of activated sugars, such as UDP-glucose, to ceramide. 26.2 Cholesterol Is Synthesized from Acetyl Coenzyme A in Three Stages Cholesterol is a steroid component of animal membranes and a precursor of steroid hormones. The conunitted step in its synthesis is the formation of mevalonate from 3-hydroxy-3-methylglutaryl CoA (derived from acetyl CoA and acetoacetyl CoA). Mevalonate is converted into isopentenyl pyrophosphate (C s), which condenses with its isomer, dimethylallyl pyrophosphate (C s), to form geranyl pyrophosphate (C IO)' The addition of a second molecule of isopentenyl pyrophosphate yields farnesyl pyrophosphate (CI S), which condenses with itself to form squalene (C 30 ). This intermediate cyclizes to lanosterol (C 30 ) , which is modified to yield cholesterol (Cd. 26.3 The Complex Regulation of Cholesterol Biosynthesis Takes Place at Several Levels In the liver, cholesterol synthesis is regulated by changes in the amount and activity of 3-hydroxy-3-methylglutaryl C oA reductase.
755 Summary
756 CHAPTER 26 The Biosynthesis of Membrane Lipids and Steroids
Transcription of the gene, translation of the mRNA, and degradation of the enzyme are stringently controlled. In addition, the activity of the reductase is regulated by phosphorylation. Triacylglycerols exported by the intestine are carried by chylomicrons and then hydrolyzed by lipases lining the capillaries of target tissues. C holesterol and other lipids in excess of those needed by the liver are exported in the form of very low density lipoprotein. After delivering its content of triacylglycerols to adipose tissue and other peripheral tissue, VLDL is converted into interm ediate-density lipoprotein and then into low-density lipoprotein. IDL and LDL carry cholesteryl esters, primarily cholesteryllinoleate. Liver and pe· ripheral tissue cells take up LDL by receptor-mediated endocytosis. The LDL receptor, a protein spanning the plasma membrane of the target cell, binds LD L and m ediates its entry into the cell. Absence of the LDL receptor in the homozygous form of familial hypercholes· terolemia leads to a markedly elevated plasma level of LDL choles· terol and the deposition of cholesterol on blood -vessel walls, which in turn may result in childhood heart attacks. Apolipoprotein B, a very large protein, is a key structural component of chylomicrons, VLDL, and LDL. High-density lipoproteins transport cholesterol from the peripheral tissues to the liver. 26.4 Important Derivatives of Cholesterol Include Bile Salts and Steroid Hormones In addition to bile salts, which facilitate the digestion of lipids, five major classes of steroid hormones are derived from cholesterol: progestagens, glucocorticoids, mineralocorticoids, androgens, and estrogens. Hydroxylations by P450 mOlJooxygenases that use NADPH and O 2 play an important role in the synthesis of steroid hormones and bile salts from cholesterol. P450 enzymes, a large suo perfamily, also participate in the detoxification of drugs and other foreign substances. Pregnenolone (C 21 ) is an essential intermediate in the synthesis of steroids. This steroid is formed by scission of th e side chain of cholesterol. Progesterone (C 2 1), sy nthesized from pregnenolone, is the precursor of cortisol and aldosterone. Hydroxylation of progesterone and cleavage of its side chain yields androstenedione, an androgen (C I 9 )· Estrogens (CIS) are synthesized from androgens by the loss of an angular methyl group and the formation of an aromatic A ring. Vitamin D, which is important in the control of calcium and phos· phorus metabolism, is formed from a derivative of cholesterol by the action of light.
Key Terms phosphatidate (p. 733) triacylglycerol (p. 733) phospholipid (p. 734)
cholesterol (p. 739) mevalonate (p . 739)
low -den sity lipoprotein (LDL) (p. 744) high -density lipoprotein (HDL) (p. 744)
3- hydroxy-3 -methylglutaryl CoA reductase (HMG · CoA reductase) (p. 739) 3-isopentenyl pyrophosphate (p. 740)
receptor-mediated endocytosis (p. 745) bile salt (p. 748) steroid hormone (p. 749)
cerebroside (p. 737 )
sterol regulatory element binding protein (SREI3P) (p. 742)
cytochrome P450 monooxygenase (p . 750)
ganglioside (p. 738)
lipoprotein particles (p. 743 )
pregnenolone (p. 752)
cytidine diphosphodiacylglycerol (CDP·diacylglycerol) (p . 73 4) sphingolipid (p. 736) ceramide (p. 736)
reverse cholesterol transport (p.744 )
Selected Readings 757
Selected Readings Where to Start Gimpl, G ., Burger, K ., and Fahrenholz, F. 2002 . 1\ closer look at t he cholesterol sensor. Trends Biochem. Sci. 27: 59 5 599. Oram, j . F. 2002. Molec ular basis of cholesterol homeostasis: Lessons fromTangierdiseaseandABCI\1. Trends Mol. Med . 8:1681 73. Vance, D. E., and Van den Rosch , H. 2000. C holesterol in th e year 2000 . Iliochim. Biophy.,. Acta 1529: 1- R. Brown, M . S., and Goldstein, j. L. 19R6. A receptor-mediated pathway for chol esterol homeostasis. Science 232:34 47. Brown, M . S., and Goldstein, J . L. 19R4. How LDL receptors influence cholesterol and atherosclerosis. Sci. Am. 251(5):58- 66. Endo, A. 1992. The discovery and development of HMG -CoA reduclase inhibitors. j. Lipid Res. 33: 1569- 1582.
Books
J.
E. (Eds.). 1996. Biochemistry oj Lipids, Lipoproteins and Membranes. Elsevier. Scriver, C . R. (Ed .), Sly, w. S. (Ed. ). Childs, B., Beaudet, A . L. , Valle, D., Kinzl er , K. W .. and Vogelstein. B. 2000 . The M etabolic Basis oj Inherited Diseases (7th ed .). McGraw-Hill. Vance, D . E., and Vance,
Phospholipids and Sphingolipids Huwiler. A ., Kolter. T , Pfeilschifter, J ., and Sandhoff, K. 2000. Physiology and pathophysiology of sphingolipid m etabolism and signaling. Biochim. Biophys. Acta 14R5:63- 99 . Lykidis, A ., and J:k kowski, S. 2000 . Regulation of mammalian cell membrane biosynthesis. Prog. Nucleic A cid Res. Mol . Bioi. 65:36 1- 393 . Carman, G . M ., and Zeimetz, G. M . 1996. Regulation o f phospho lipid biosynthesis in the yeast Saccharomyces cerevisiae. j . [Jiol . Chem. 271: 13293- 13296. Henry, S. A ., and Patton -VOg!. J . L. 1998. Genetic regulation of phospholipid metabolism : Yeast as a model eukaryote. Prog. N ucleic Acid Res. Mol. Bioi. 61 :133-179. Kent. C. 199 5. Eukaryotic phospholipid biosynthesis. Annu. Rev. Biochem. 64: 315- 343. Prescott, S. M ., Z immerman, G . A ., Stafforin i, D . M ., and McIntyre, T. M . 2000 . Platelet-activating factor and related lipid mediators.
Annu. Rev. Biochem. 69:419-445.
Biosynthesis of Cholesterol and Steroids Hampton, R. Y. 2002. Proteolysis and sterol regulation . Annu. Rev. Cell Dev. Bioi. 18:345- 378. Kelley, R. l., and H erman, G. E. 2001 . Inborn errors of sterol biosynthesis. Annu. }{ev. Genom. 2:299- 341. Goldstein, J. L., and Brown. M . S. 1990. Regulation of the mevalonate pathway. Nature 343 :425-430. Gardner R. G ., Shan, H ., Matsuda, S. P. T, and Hampton, R . Y. 2001. An oxysterol-d erived positive signal for 3-hydroxy-3 -methyl gl u taryl -CoA reductase degradation in yeast . j. BioI. Chern. 276: R6R l - R694 . Istvan, E. S. , and Deisenhofer , J . 2001 . Structural mechanism for statin inhibition of HMG -CoA reductase. Science 292: 11 60- 1164. Ness, G . C ., and C hambers, C. M . 2000. Feedback and hormonal regulation of hepatic 3-hydroxy -3 -methylglutaryl coenzyme A rcuuctase: The concept of cholesterol buffering capacity. Proc. Soc. Exp.
Bioi. Med. 224 :8- 19. Libby, P.. Aikawa, M .. and Schonbeek, U. 2000. C holesterol and atherosclerosis. Biochim. Biopltys. Acta 1529 :299 309 . Yokoya ma. S. 2000. Release of cellular cholesterol: Molecular mechanism fo r cholesterol homeostasis in cells and in the body. Biochim. Biophys. Acta 1529:231- 244 . Cronin, S. R .. Khoury. A ., Ferry, D . K ., and Hampton. R. Y. 2000 . Regul ation of HMG -CoA red uctase d egradati on requires the P-type ATPase Codl p / Spfl p . j. Cell Bini. 14R:915- 924.
Edwards, 1'. A ., Tabor, D ., Kast, H . R ., and Venkateswaran , A . 2000. Regulation of gene ex pression by SREBP and SCAP. Biochim.
Biophys. Acta 1529: 103- 113. Istvan, E . S., Palnitkar, M ., tiuchanan, S. K ., and Dcisenhofcr , j . 2000. Crystal structure of the catalytic portion of human HMG -CoA reductase: Insights into regulation of activity and catalysis. EM B() j . 19:819- 830. Tabernero, L., Bachar, D . A .. Rodwell, V. W .. and Stauffacher. C . V. 1999 . Substrate-induced closure of th e flap domain in the ternary complex structures provides insights into the mechanism of cataly sis by 3- hydroxy -3-m ethylglutaryl -CoA reductase. Proc. Natl. Acad. Sci. U. S . A. 96 :7 167- 7171. Fass, D ., Blackl ow, S., Kim , P. S., and Berger, j. M . 1997. Molecul ar basis of fami lial hypercholesterolaemia from structure of LDL receptor module. Nature 388:69 1- 693. jmn, H., Meng, W ., Takagi, J., Eck, M. j., Springer, T. A ., anu Blacklow, S. C. 2001 . Implieations for familial hypercho lesterolemia from the structure of the LDL receptor YWTD-EGr domain pair. Nat. Stroct. [Jiol. 8:499- 504.
Lipoproteins and Their Receptors Jeon, H ., and Blacklow, S. C. 2005. Structure and physiologic function of the low -density lipoprotein receptor. Annu . Rev. Biochcm. 74 :535- 562. Brouillette, C. G ., Anantharamaiah, G. M. , Engler , J . A ., and Borhani , D . W . 2001. Stru ctura'! models of human apolipoprotein A - I: A critical analysis and review. Biochem. Biophys. Acta 1531 :4 46. H evonoja, T , Pentikainen , M . 0 ., H yvonen, M. T, Kovanen, P. T., and Ala-Korpela, M . 2000 . Structure of low density li poprotein (LDL) particles: Basis for understanding molecular changes in modified LDL. Biochim. Biophys. Acta 1488:189- 210. Silver, O. L., Jiang, X. C., Arai, T, Bruce, C ., and 'fall , A . R. 2000. Receptors and lipid transfer proteins in HDL meta bolism . A,III. N. Y. Acad. Sci. 902 : 103 111. Nimpf, J., and Schneid er . W . ]. 2000. From cholesterol transport to signal transduction: Low density li poprotein receptor, very low den· sity lipoprotein receptor, and apolipoprotein E receptor-2. Biochim.
Biophys. Acta 1529:287- 298 . Borhani , D . W ., Rogers. D . P.. Engler , ]. A ., and Brouillette, C. G . 1997 . C rystal stru cture of truncated human apolipoprotein A - I suggests a lipid -bound conformation . Proc. Nat!. Acad. Sci. U. S. A. 94:1229 1- 12296. Wilson, C ., Wardell, M . R., W eisgraber, K. H ., Mahley, R. W ., and Agard, D . A. 1991 . Three-dimensional structure of the LDL receptor-binding domain of human apolipoprotein E. Science 252 :1817- 1822. Plump, A. S. , Smith, j . D., Hayek, T, Aalto-Setiil ii, K., Walsh, A., Verstuyft. j . G ., Rubin, E. M .. and tireslow. J . L. 1992. Severe hypercholesterolemia and atherosclerosis in apolipoprotein E-deficient mice created by ho mo logo us recombination in ES cells. Cell 71 :343 353 .
Oxygen Activation and P450 Catalysis Williams, 1'. A ., Cosm e, ]., Vinkovic, D . M ., Ward, A., An gove, H. C., Day, 1'. ]., Von rhein, C., Tickl e, 1. J .. and Jhoti, H. 2004. Crystal structure of human cytochrome 1'450 3A4 bound to mety rapone and progesterone. Science 305:683- 686 . Ingelman -Sundberg, M ., O scarson, M ., and McLellan, R. A . 1999. Poly morphic human cytochrome 1'4 50 enzymes: An oppurtunity for individuali~ed drug treatment. Trends Pharmaco l. Sci . 20:342 349 . Nelson, D . R . 1999. Cytochrome P4 50 and the indi viduality of species. Arch. Biochem. Biophys. 369 : 1 10. Wong. L. L. 1998 . Cytochrome 1'450 monooxygenases. Curro Opin. Chem. Bioi. 2 :263- 26H.
758
CHAPTER 26 The Biosynthesis of Me mbrane lipids and Steroids
Denison , M . S., and Whitlock, J. P. 1995. Xenobiotic-inducible transcription of cytochrome P4 50 genes. }. Bioi. C hern. 270 : 1~ 1 75- 1 8 1 7 8 . Poulos, T. L. 1995. Cytochrome P4 50. CUrT. Opin. Stm ct. Riol. 5:767- 774.
Vaz, A. D ., and Coon, M . J. 1994. O n the mechanism of action of cy. tochrome P450 : Evaluation of hydrogen abstraction in oxygen· dependent alcohol oxidation. Biochemistry 33:6442- 6449 . Gonzalez, F. J.. and Nehert, n .W . 1991l. Evolution ofthe P4 50 genesu· perfamily: Animal- plant "warfare," molecul ar drive and human genetic differences in drug oxidation. Trends Genet. 6:182- 186.
Problems 1. Making fat . Write a balanced equation for the synthesis of a triacylglycerol, starting from glycerol and fatty acids.
2. Making a phospholipid. Write a balanced equation for the synthesis of phosphat idylethanolamine by the de n ovo pathway, starting from ethanolamine, glycerol, and fatty acids.
3. Activated donors. What is the activated reactant in each of the followin g b iosyn theses? (a) (b) (c) (d) (e) (f ) (g)
Phosphatidylinositol from inositol Phosphatidylethanolamine from ethanolamine Ceramide from sphingosine Sphingomyeli n from ceramide Cerebroside fro m cerami de Ganglioside G M I from gangli oside G M 2 Farnesyl pyrophosphate from geranyl pyrophosphate
4. Telltale labels. What is the distribution of isotopic labeling in cholesterol synthesized from each of the following precursors?
nantly female external genitalia. These people are usually reared as girls. At puberty, they masculinize because the testosterone level ri ses . The testes of these reductase-deficient men are nor· mal , whereas their prostate glands remain small. How mightthis information be used to design a drug to treat benign prostatic hy· pertrophy, a common consequence of the normal aging process in men? A majority of m en older than age 55 have some degree of prostatic en largement, which often leads to urinary obstruction. S. Drug idiosyncrasies. Debrisoquine, a p -adrenergic blocking agent, has been used to treat hypertension. The optimal dose varies greatly (20 400 m g daily) in a population of patients. The urine of most patients taking the drug contains a high level of 4· hydroxydebrisoquine. However, those most sensitive to the drug (about 8% of the group studied ) excrete debrisoquine and very lit· ti e of the 4-hydroxy derivative. Propose a molecular basis for this drug idiosyncrasy. Why should caution be exercised in giving other drugs to patients who are very sensitive to debrisoquine?
(a) Mevalonate la beled with 14C in its carboxyl carbon atom (b) Malonyl CoA labeled with 14C in its carboxyl carbon atom S. Familial hypercholesterolemia. Several classes of LDL-receptor mutations have been identifi ed as causes of th is d isease. Suppose that you have been given cells from patients with different mu tations, an antibody specific for the LDL receptor that can be seen with an electron microscope, and access to an electron mi croscope. What differences in antibody distribution might you expect to find in the cells from different patients?
6. RNA editing. A shortened version (apo £-48) of apolipoprotein H is formed by the intestine, whereas the full-length protein (apo B -100) is synthesized by the liver. A glutamine codon (CAA) is chan ged into a stop codon . Propose a simple mechanism for this change.
Oebrisoquine
9. Removal of odorants. Many odorant molecules are highly hy. drophobic and concentrate within the olfactory epithelium They would give a persistent signal independent of their con· centration in the environment if they were not rapidly modified Propose a mechanism for converting hydrophobic odorants intc water -soluble derivatives that can be rapidly eliminated.
10. Development difficulties. Propecia (finasteride) is a synthetic steroid that functions as a competitive and specific inhibitor 01 Sa -reductase, the enzyme responsible for the synthesis of dihy. drotestosterone from testosterone.
7. [nspirationfor drug design. Some actions of androgens are me diated by dihydrotestosteron e, which is formed by the reduction of testosterone. T hi s finishing touch is catalyzed by an NA D PH -dependent Sa-reductase.
CH,
CH,OH NADPH + H+
Testosterone
N H
NADP+
\ /
,
5a -Reductase
H
5a -Dihydrotestosterone
C hromosomal XY males with a genetic deficiency of this reductase are born with a male internal urogenital tract but predomi -
Finasteride
It is now widely used to retard the development or male patcern hair loss. Pregnant women are advised to avoid handling this drug. Why is it vitally important that pregnant women avoid contact with Propecia?
11. Life -style consequences. Human beings and the plant Arabidopsis evolved from the same distant ancestor possessing a
Problems 75 9
o Actin mRNA o HMG-CoA reductase mR NA
small number of cytochrome P450 genes. Human beings have approximately 50 such genes, whereas Arabidopsis has more than 250 of them. Propose a role for the large number of P450 isozymes in plants.
12. Personalized medicine. The cytochrome P450 system metabolizes many medicinally usefu l drugs. Although al l human bei ngs have the same number of P450 genes, individual polymorphisms exist that alter the specificity and efficiency of the proteins encoded by the genes. How could knowledge of individual polymorphisms be useful clinically?
1.00 -
c( ~
z'" "" ., ., 0.67E §
o ~
- '" c
0-"
E ;;; 0.33
c( ~
Mechanism Problems
o L-"- _ _
13. An interfering phosphate. In the course of the overall reaction catalyzed by HMG -CoA reductase, a histidine residue protonates a coenzyme A thiolate, CoA-S - , generated in a previous step. CoA
~
:l .-::
5-
r+~H
HNJI se:i HO
Control • mice
(A)
= _ Control • mice
_
,--l--L- L- , - - , -- ' -- ' -- L -
Cholesterol-fed •
mice
Cholesterol-fed • mice
(a) What is the effect of chol esterol feeding on the amount of HMG-CoA reductase mRNA ? (b) What is the purpose of also isolating the m RNA for the protein actin, which is not under the control of the sterol regulatory element?
His
1 CoA
SH
q-NH N #
HOI
HMG -CoA reductase protein was isolated by precipitation with a monoclonal antibody to HMG-CoA reductase . The amount of H MG-CoA protein in each group is shown in graph B. 1.00
A nearby serine residue can be phosphorylated by AMPdependent kinase, which results in a loss of activity. Propose an explanation for why phosphorylation of the serine residue inhibits enzyme activity.
0.67
14. Demethylation. Methyl amines are often demethylated by cytochrom e P450 enzymes . Propose a mechanism for the formalion of methy lamine from dimethylamine catalyzed by cytochrome P450 . What is the other product?
0 .33
o L...-, Control mice
Data Interpretation and Chapter Integration Problem
IS. Clwlesterolfeeding. Mice were divided into four groups, two of which were fed a normal diet and two of which were fed a cholesterol- rich diet. HMG-CoA reductase mRNA and protein [rom liver were then isolated and quantified . G raph A shows the results of the mRNA isolation .
(8)
Cholesterol-fed • mice
(c) What is the effect of the cholesterol diet on the amount of HMG -CoA reductase protein? (d) Why is this result surprising in light of the results in graph A ? (e) Suggest possible explanations for the results in graph B.
Chapter
The Integration of Metabolism
Glucose
o~
o
0
AlP •
lhe image at the left shows a detail of runners o n a Greek amphora painted in the sixth century S.c. Athletic feats, as well as others as seemingly simple as the maintenance of blood-glucose levels, require elaborate metabolic integration. The schematic representation illustrates the ox idation o f glucose to yield AlP in a process requiring interplay &etween glycolysis, the citric acid cycle, and ox idative phosphorylation. These are a few of the many metabo lic pathways that must be coordi nated to meet the demands of living. [(Left) Metropolitan Museum of Art, Rogers Fund, 1914 (14.130.12). Copyri ght © 1977 by the Metropolitan Museum of Art.)
I O utlin e l 27.1 Metabolism Consists of Highly Interconnected Pathways 27.2 Each Organ Has a Unique Metabolic Profile 27.3 Food Intake and Starvation Induce Metabolic Changes 27.4 Fuel Choice During Exercise Is Determined by the Intensity and Duration of Activity 27.5 Ethanol Alters Energy Metabolism in the Liver
760
e have been examining the biochemjstry of metabolism one pathway at a time but, in living systems, many pathways are operating simul· taneously. Each pathway must be able to sense the status of the others to function optimally to meet the needs of an organism. How is the intricate network of reactions in metabolism coordinated? This chapter presents some of the principles underlying the integration of metabolism in mammals. W e begin with a recapitulation of the strategy of metabolism and of recur· ring motifs in its regulation. W e then turn to the interplay of different path· ways as we examine the flow of molecules at three key crossroads: glucose 6-phosphate, pyruvate, and acetyl CoA. W e consider the differences in the metabolic patterns of the brain, muscle, adipose tissue, kjdn ey, and liver. Finally, we examine how th e interplay between these tissues is altered in a variety of metabolic perturbations. These considerations of metabolism will illustrate how biochemical knowledge illuminates the fun ctioning of the • orgamsm.
27.1
Metabolism Consists of Highly Interconnected Pathways
The basic strategy of catabolic metabolism is to form ATP, reducing power, and building blocks for biosyntheses. Let us briefly review these central themes: I. ATP is the universal currency of energy. The high phosphoryl-transfer
potential of ATP enables it to serve as the energy source in muscle contraction, active transport, signal amplification, and biosyntheses. In the cell, the hydrolysis of an ATP molecule changes the equilibrium ratio of products to 8 reactants in a coupled reaction by a factor of about 10 . Hence, a thermody-
761 27.1 Interconnection of Metabolic Pathways "To every thing there is a season, and a time to every purpose under the hea....en: A time t o be born, and a time to die; a
time to plant, and a time to pluck up that which is planted; A time to kill, and a time t o heal; a time to break down, and a time to build up." ECCLESIASTES 3:1- 3
namically unfavorable reaction sequence can be made highly favorable by coupling it to the hydrolysis of a sufficient number of A TP molecules. 2. ATP is generated by the oxidation of fuel molecules such as glucose, fatty acids, and amino acids. The common intermediate in most of these oxidations is acetyl CoA. The carbon atoms of the acetyl unit are completely oxidized to CO 2 by the citric acid cycle with the concomitant formation of NADH and F ADH 2 . These electron carriers then transfer their highpotential electrons to the respiratory chain. The subsequent flow of electrons to O 2 leads to the pumping of protons across the inner mitochondrial membrane' (Figure 27 .1) . This proton gradient is then used to synthesize ATP. Glycolysis also generates ATP, but the amount formed is much smaller than that formed by oxidative phosphorylation. The oxidation of glucose to pyruvate yields only 2 molecules of ATP, whereas the complete oxidation of glucose to CO 2 yields 30 molecules of A TP.
Pasteur eHect The inhibition of glycolysis by respiration. discovered by Louis Pasteur in studying fermentation by yeast. The consumption o f carbohydrate is about sevenfold lower under aerobic conditions than under anaer-
obic ones. The inhibition of phosphofructokinase by citrate and AT? accounts for
much of the Pasteur effect.
•
3. NADPH is the major electron donor in reductive biosyntheses. In most biosyntheses, the products are more reduced than the precursors, and so reductive power is needed as well as ATP. The high-potential electrons required to drive these reactions are usually provided by NADPH. The pentose phosphate pathway supplies much of the required NADPH.
4. Biomolecules are constructed from a small set of building blocks. The highly diverse molecules of life are synthesized from a much smaller numberof precursors. The metabolic pathways that generate ATP and NADPH also provide building blocks for the biosynthesis of more-complex molecules. For example, acetyl CoA, the common intermediate in the breakdown of most fuels, supplies a two-carbon unit in a wide variety of biosyntheses, such as those leading to fatty acids, prostaglandins, and cholesterol. Thus, the central metabolic pathways have anabolic as well as catabolic roles.
,
5. Biosynthetic and degradative pathways are almost always distinct. For example, the pathway for the synthesis of fatty acids is different from that for their degradation. This separation enables both biosynthetic and degradative pathways to be thermodynamically favorable at all times. A biosynthetic pathway is made exergonic by coupling it to the hydrolysis of a sufficient number of ATP molecules. The separation of biosynthetic and degradative pathways contributes greatly to the effectiveness of metabolic control.
Recurring Motifs are Common in Metabolic Regulation Anabolism and catabolism must be precisely coordinated. Metabolic networks sense and respond to information on the status of their component pathways. The information is received and metabolism is controlled in several ways:
Figure 27.1 Electron micrograph of mitochondria. Numerous mitochondria occupy the inner segment of retinal rod cells. These photoreceptor cells generate large amounts of ATP and are highly dependent on a continuous supply o f 0 ,. [Courtesy of Dr. Michael Hogan.]
762
1. Allosteric Interactions. The flow of molecules in most metabolic pathways is determined primarily by the activities of certain enzymes rather than by the amount of substrate available. Enzymes that catalyze essentially irreversible reactions are likely control sites, and the first irreversible reaction in a pathway (the committed step) is nearly always tightly controlled. Enzymes catalyzing committed steps are allosterically regulated, as exemplified by phosphofructokinase (PFK) in glycolysis and acetyl eoA carboxylase in fatty acid synthesis. Allosteric interactions enable such enzymes to rapidly detect diverse signals and to adjust their activity accordingly.
CHAPTER 27 The Integration of Metabolism
- Tyr - OH
- Ser- OH /
ATP
/
pp.,
ADP
- Ser - O - PO/~
(A)
ATP
- Tyr - O- AMP (8)
Figure 27.2 Covalent modifications. Covalent mo difications. Examples of re versible covale nt modifications o f pro teins: (A) phosphorylation and (B) adenylation.
2. Covalent Modification. Some regulatory enzymes are controlled by covalent modification in addition to allosteric interactions. For example, the catalytic activity of glycogen phosphorylase is enhanced by phosphorylation, whereas that of glycogen synthase is diminished. Specific enzymes catalyze the addition and removal of these modifying groups (Figure 27.2). Phosphorylation often takes place in response to hormonal signals. For instance, insulin, glucagon, and epinephrine stimulate protein kinases. Why is covalent modification used in addition to noncovalent allosteric control? The covalent modification of an essential enzyme in a pathway is often the final step in an amplifying cascade and allows metabolic pathways to be rapidly switched on or off by very low concentrations of triggering signals. In addition, covalent modifications usually last longer (from seconds to minutes) than do reversible allosteric interactions (from milliseconds to seconds). 3. Adjustment of Enzyme Levels. The amounts of enzymes, as well as their activities, are controlled . The rates of synthesis and degradation of many regulatory enzymes are altered by hormones. The basics of this control will be considered in Chapter 29; we will return to the topic in Chapter 31.
r Cytoplosm: Glycolysis Pentose phosphate pathway Fatty acid synthesis Inner mitochondriol membrane: Oxidative phosphorylation Mitochondrial matrix: Citric acid cycle fl-Oxidation of fatty acids Ketone-body formation
~
Interplay of both compartments: Gluconeogenesis Urea synthesis
Figure 27.3 Compartmentation of the major pathways of metabolism.
4. Compartmentation. The metabolic patterns of eukaryotic cells are markedly affected by the presence of compa! tments (Figure 27.3). The fates of certain molecules depend on whether they are in the cytoplasm or in mitochondria, and so their flow across the inner mitochondrial membrane is often regulated. For example, fatty acids are transported into mitochondria for degradation only when energy is required, whereas fatty acids in the cytoplasm are esterified or exported. 5. Metabolic Specializations of Organs. Regulation in higher eukaryotes is enhanced by the existence of organs with different metabolic roles. Metabolic specialization is the result of differential gene expression.
Major Metabolic Pathways Have Specific Control Sites
Let us now review the roles of the major pathways of metabolism and the principal sites for their control: 1. Glycolysis. This sequence of reactions in the cytoplasm converts one molecule of glucose into two molecules of pyruvate with the concomitanl generation of two molecules each of ATP and NADH. The NAD + consumed in the reaction catalyzed by glyceraldehyde 3-phosphate dehydrogenase must be regenerated for glycolysis to proceed. Under anaerobic conditions, as in highly active skeletal muscle, this regeneration is accomplished by the reduction of pyruvate to lactate. Alternatively, under aerobic conditions, N AD + is regenerated by the transfer of electrons from NADH to O2 through the electron-transport chain. Glycolysis serves two main purposes: it degrades glucose to generate ATP, and it provides carbon skeletons for biosyntheses_
Phosphofructokinase, which catalyzes the committed step in glycolysis, is the most important control site. ATP is both a substrate in the phosphorylgroup-transfer reaction and a regulatory molecule. A high level of ATP inhibits phosphofructokinase: the regulatory sites are distinct from the substrate-binding sites and have a lower affinity for the nucleotide. This in hibitory effect is enhanced by citrate and reversed by AMP (Figure 27.4). Thus, the rate of glycolysis depends on the need for ATP, as signaled by the ATP I AMP ratio, and on the availability of building blocks, as signaled by the level of citrate. In the liver, the most important regulator of phosphofructokinase activity is fructose 2, 6-bisphosphate (F -2,6-BP). Recall that the level ofF-2,6-BP is determined by the activity of the kinase that forms it from fructose 6-phosphate and of the phosphatase that hydrolyzes the 2-phosphoryl group (p. 466). When the blood-glucose level is low, a glucagon-triggered cascade leads to the activation of the phosphatase and the inhibition ofthe kinase in the liver. The level of F -2,6-BP declines and, consequently, so does phosphofructokinase activity. Hence, glycolysis is slowed, and the spared glucose is released into the blood for use by other tissues. Pyruvate kinase, which controls the outflow of glycolysis, also is an important regu latory site. It is stimulated by fructose 1,6-bisphosphate, a product of the PFK reaction, and inhibited by ATP. In the liver, pyruvate kinase is phosphorylated by the glucagon-stimulated cAMP cascade, diminishing the en, .. zyme s achvlty. 2. Citric Acid Cycle arul Oxidative Phosphorylation. The reactions of this common pathway for the oxidation of fuel molecules carbohydrates, fatty acids, and amino acids take place inside mitochondria. Most fuels enter the cycle as acetyl CoA. The complete oxidation of an acetyl unit by the citric acid cycle generates one molecule of GTP and results in the reduction of three molecules of NAD + to NADH and one molecule ofFAD to F ADH 2 . These electrons are transferred to O 2 through the electron-transport chain, which results in the formation of a proton gradient that drives the synthesis of nine molecules of A TP. The electron donors are oxidized and recycled back to the citric acid cycle only if ADP is simultaneously phosphorylated to ATP. This tight coupling, called respiratory control, ensures that the rate of the citric acid cycle matches the need for ATP An abundance of ATP also dimini shes the activities of two enzymes in the cycle isocitrate dehydrogenase and a -ketoglutarate dehydrogenase. The citric acid cycle has an anabolic role as well. In concert with pyruvate carboxylase, the citric acid cycle provides intermediates for biosyntheses, such as succinyl CoA for the formation of porphyrins and citrate for the formation of fatty acids.
3. Pyruvate Dehydrogenase Complex. This enzyme complex is a key regu latory site because it is the irreversible link between glycolysis and the citric acid cycle. The pyruvate dehydrogenase complex catalyzes the conversion of pyruvate into acetyl CoA. This reaction, which takes place inside mitochondria, is a decisive reaction in metabolism: it commits the carbon atoms of carbohydrates and amino acids to oxidation by the citric acid cycle or to the synthesis of lipids. The pyruvate dehydrogenase complex is stringently regulated by multiple allosteric interactions and covalent modifications. Pyruvate is rapidly converted into acetyl CoA only if ATP is needed or if two-carbon fragments are required for the synthesis of lipids. Pentose Phosphate Pathway. This series of reactions, which takes place in the cytoplasm, consists of two stages. The first stage is the oxidative decarboxylation of glucose 6-phosphate. Its purpose is the production of NADPH for reductive biosyntheses and the formation of ri bose 5-phosphate for the synthesis of nucleotides. Two molecules of 4.
763 27.1 Interconnection of Metabolic Pathways
Fructose 6-phosphate AlP
Phosphofructokinase
Activated by F-2.6-8P Activated by AMP ADP
Inhibited by AlP and citrate
Fructose 1,6-bisphosphate Figure 27.4 Regulation of glycolYSis. Phosphofructokinase is the key enzyme in the regulation of glycolysis.
764 CHAPTER 27 The Integration of Metabolism
Glucose 6-phosphale Glucose 6-phosphale dehydrogenase
NADPH 6-Phosphoglucono-&-lactone
ladonase
6-Phosphogluconate Figure 27.5 Regulation of the pentose phosphate pathway. The dehydrogenation of glucose 6-phosphate is the committed step in the pentose phosphate pathway.
NADPH are generated in the conversion of glucose 6-phosphate into rio bose 5-phosphate. The dehydrogenation of glucose 6 -phosphate is the committed step in this pathway. This reaction is controlled by the level of NADP + , the electron acceptor (Figure 27 .5). The second stage of the pentose phosphate pathway is the nonoxidative, reversible metabolism of five-carbon phosphosugars into phosphorylated three-carbon and six-carbon glycolytic intermediates. Thus, the nonoxida· tive branch can either introduce riboses into glycolysis for catabolism or generate riboses• from glycolytic intermediates for biosyntheses.
5.
Gluconeogenesis. Glucose can be synthesized primarily by the liver from
noncarbohydrate precursors such as lactate, glycerol, and amino acids. The major entry point of this pathway is pyruvate, which is carboxylated to ox· aloacetate in mitochondria. Oxaloacetate is then metabolized in the cyto· plasm to form phosphoenolpyruvate. The other distinctive reactions used by gluconeogenesis are two hydrolytic steps that bypass the irreversible reo actions of glycolysis. Gluconeogenesis and glycolysis are usually reciprocally regulated so that one pathway is minimally active while the other is highly ac· tive. For example, AMP inhibits and citrate activates fructose 1,6-bisphos· phatase, an essential enzyme in gluconeogenesis, whereas these molecules have opposite effects on phosphofructokinase, the pacemaker of glycolysis (Figure 27.6). Fructose-2,6-bisphosphate also coordinates these processes by inhibiting fructose 1,6-bisphosphatase. Hence, when glucose is abun· dant, the high level of F -2,6-BP inhibits gluconeogenesis and activates glycolysis. Fructose 1,6-bisphosphate H2 0
Fructose 1.6-bisphosphatase Activated by. citrate
Figure 27.6 Regulation of gluconeogenesis. Fructose 1,6bisphosphatase is the principal enzyme controlling the rate of gluconeogenesis.
6.
Inhibited by AMP P;
Inhibited by F·2.6· BP
Fructose 6-phosphate
Glycogen Synthesis and Degradation. Glycogen, a readily mobilizable
fuel store, is a branched polymer of glucose residues (Figure 27.7). In glyco. gen degradation, a phosphorylase catalyzes the cleavage of glycogen by or· thophosphate to yield glucose i -phosphate, which is rapidly converted into glucose 6-phosphate for further metabolism. In glycogen synthesis, the ac· tivated intermediate is UDP -glucose, which is formed from glucose i · phos· phate and UTP. Glycogen synthase catalyzes the transfer of glucose from UDP -glucose to the terminal glucose residue of a growing strand . Glycogen
degradation and synthesis are coordinately controlled by a hormone-triggered amplifying cascade so that the phosphorylase is active when synthase is inactive and vice versa. Phosphorylation and noncovalent allosteric interactions reg· ulate these enzymes.
7.
I
I
500nm Figure 27.7 Glycogen granules. The electro n micro graph shows part of a liver cell containing glycogen particles. [Courtesy of Dr. George Palade.]
Fatty Acid Synthesis and Degradation. Fatty acids are synthesized in the cytoplasm by the addition of two-carbon units to a growing chain on an acyl carrier protein. Malonyl eoA, the activated intermediate, is formed by the carboxylation of acetyl eoA. Acetyl groups are carried from mitochondria to the cytoplasm as citrate by the citrate- malate shuttle. In the cytoplasm, citrate is cleaved to yield acetyl eoA. In addition to transporting acetyl eoA, citrate in the cytoplasm stimulates acetyl GoA carboxylase, the enzyme
catalyzing the committed step. When AT? and acetyl eoA are abundant, the level of citrate increases, which accelerates the rate of fatty acid synthesis (figure 27.8). A different pathway in a different compartment degrades fatty acids. Carnitine transports fatty acids into mitochondria, where they are d egraded to acetyl eoA in the mitochondrial matrix by 13 oxidation. T he acetyl eoA then enters the citric acid cycle if the supply of oxaloacetate is sufficient. Alternatively, acetyl eoA can give rise to ketone bodies . The fAD H 2 and NADH formed in the l3-oxidation pathway transfer their electrons to O 2 through the electron-transport chain. Like the citric acid cycle, 13 oxidation can continue only if NAD + and FAD are regenerated. Hence, the rate of fatty acid degradation also is coupled to the need for AT? Malonyl eoA, the precursor for fatty acid synthesis, inhibits fatty acid degradation by inhibiting the formation of acyl carnitine by carnitine acyltransferase I, thus preventin g the translocation of fatty acids into mitochondria (Figure 27.9) . Glucose 6-phosphate, Pyruvate, and Acetyl CoA Are Key Junctions in Metabolism
o .......... CoA
5 Acetyl CoA Acetyl CoA carboxylase
HCO, - + AlP
Activated by citra te Inhibited by palmitoyl CoA
o
Molonyl CoA Figure 27.8 Regulation of fatty acid synthesis. Acetyl CoA carboxylase is the key contro l site in fatty acid synthesi s.
The factors governing the flow of molecules in metabolism can be further understoocl by examining three important molecules: glucose 6-phosphate, pyruvate, and acetyl eoA. Each of these molecules has several contrasting fates: Glucose 6-phosphate. Glucose entering a cell is rapidly phosphory lated to glucose 6-phosphate and is subsequently sto red as glycogen, d e graded to pyruvate, or converted into ribose 5-phosphate (Figure 27. 10). Glycogen is formed when glucose 6-phosphate and ATP are abundant. In contrast, gl ucose 6-phosphate flows into the glycolytic pathway when ATP or carbon skeletons for biosy ntheses are required. Thus, the conversion of glucose 6- phosphate into pyruvate can be anabolic as well as catabolic. The third major fate of glucose 6-phosphate, to flow through the pentose phosphate pathway, provides NADPH for reductive biosyntheses and ribose :i -phosphate for the synthesis of nucleotides. Glucose 6-phosphate can be formed by the mobilization of glycogen or it can be synthesized from pyruvate and glucogenic amino acids by the gluco neogenic pathway. I.
Carnitine /'" Acyl CoA Carnitine
acyltrons!eros. I
Inhibited by malonyl CoA CoASH
Acyl carnitine Figure 27.9 Control of fatty ac id degradation. Malo nyl Co A inhibits f atty acid degradation by inhibiting the formation of acyl carnitine.
Glucose
Glucose 6-phosphate
Glucose l -phosphate
Fructose 6-phosphate
Glycogen Pyruvate
6-Phosphogluconate
Ribose 5-phosphate
f igure 27.10 Metabol ic fates of glucose 6-phosphate.
765
766
Glucose 6-phosphate
CHAPTER 27 The Integration of Metabolism
Lactate
Pyruvate
Oxaloacetate
3-Hydroxy-3-methylglutaryl CoA
Figure 27.11 Major metabolic fates of pyruvate and acetyl CoA in mammals.
Cholesterol
Ketone bodies
Alanine
Acetyl CoA
Fatty acids
CO 2
2. Pyruvate. This three -carbon a-ketoacid is another major metabolic junction (Figure 27. 11 ). Pyruvate is derived primarily from glucose 6- phosphate, alanine, and lactate. Pyruvate can be reduced to lactate by lactate dehydrogenase to regenerate N AD + . This reaction enables glycolysis to proceed transiently under anaerobic conditions in active tissues such as contracting muscle. The lactate formed in active tissue is subsequently oxidized back to pyruvate in other tissues. T he essence of this interconversion is that it buys time and shifts part of the metabolic burden of active muscle to other tissues . Another readil y reversible reaction in the cytoplasm is the transamination of pyruvate, an a-ketoacid , to alanine, the corresponding amino acid. Conversely, several amino acids can be converted into pyruvate. Thus, transamination is a major link between amino acid and carbohydrate metabolism. A third fate of pyruvate is its carboxylation to oxaloacetate inside mitochondria, the fir st step in gluconeogenesis. This reaction and the subseq uent conversion of oxaloacetate into phosphoenolpyruvate bypass an irreversibl e step of glycolysis and hence enable glucose to be synthesized from pyruvate. The carboxylation of pyruvate is also important for replenishing intermediates of the citric acid cycle. Acetyl CoA activates pyruvate carboxylase, enhan cing the synthesis of oxaloacetate, when the citric acid cycle is slowed by a paucity of this intermediate. A fourth fate of pyruvate is its oxidative decarboxylation to acetyl CoA, as described on page 763. 3. Acetyl GoA. The major sources of this activated two-carbon unit are the oxidative decarboxylation of pyruvate and the 1'\ oxidation of fatt y acids (see Figure 27. 11 ). Acetyl CoA is also derived from ketogenic amino acids. The fate of acetyl CoA, in contrast with that of many molecul es in metabolism, is quite restricted. The acetyl unit can be completely oxid ized to CO 2 by the citric acid cycle. Alternatively, 3-hydroxy -3-methylglutaryl CoA can be form ed fro m three molecules of acetyl CoA. This six -carbon uni t is a precursor of cholesterol and of ketune budies, which are transport form s of acetyl units released from the liver for use by some peripheral tissues. A third major fate of acetyl CoA is its export to the cytoplasm in the form of citrate for the synthesis of fatty acids.
27.2
Each Organ Has a Unique Metabolic Profile
The metabolic patterns of the brain, mu scle, adipose tissue, kidney, and liver are strikingly different. Let us consider how these organs differ in their use of fuels to meet their energy needs:
1. Brain. Glucose is virtually the sole fuel for the human brain, except during prolonged starvation. The brain lacks fuel stores and hence requires a continuous supply of glucose. It consumes about 120 g daily, which corresponds to an energy input of about 1760 kJ (420 kcal), accounting for some 60% of the utilization of glucose by the whole body in the resting state. Much of the energy, estimates suggest from 60% to 70%, is used to power transport mechanisms that maintain the Na + - K + membrane potential required for the transmission of the nerve impulses. The brain must also synthesize neurotransmitters and their receptors to propagate nerve impulses. Overall, glucose metabolism remains unchanged during mental activity, although local increases are detected when a subject performs certain tasks. Glucose is transported into brain cells by the glucose transporter GLUT3 . This transporter has a low value of KM for glucose (1.6 mM), which means that it is nearly saturated under most conditions, given that the plasma concentration of glucose during fasting is 4.7 mM (84. 7 mg/ d l). Under these conditions, the concentration of glucose in the brain is about 1 mM. Glycolysis slows down when the glucose level approaches the K M value of hexokinase (~ 50 ,....,M), the enzyme that traps glucose in the cell (p. 435 ). This danger point is reached when the plasma-glucose level drops below about 2.2 mM (39.6 mg/ dl) and thus approaches the K M value of GLUT3 . Fatty acids do not serve as fuel for the brain but rather are utilized for membrane synthesis . In starvation, ketone bodies generated by the liver partly replace glucose as fuel for the brain.
2. Muscle. The major fuels for muscle are fatty acids, glucose, and ketone bodies. In resting muscle, fatty acids are the major fuel, meeting RS% of the energy needs. Muscle differs from the brain in having a large store of glyco gen (5000 kJ, or 1200 kcal ). In fact, about three -fourths of all the glycogen in the body is stored in muscle (Table 27 .1 ). This glycogen is readily con verted into glucose 6-phosphate for use within muscle cells. Muscle, like the brain, lacks glucose 6-phosphatase, and so it does not export glucose. Rather, muscle retains glucose, its preferred fuel for bursts of activity. In vigorously contracting skeletal muscle, the rate of glycolysis far exceeds that of the citric acid cycle, and much of the pyruvate formed is reduced to lactate, some of which flows to the liver, where it is converted into glucose (Figure 27 .1 2) . These interchanges, known as the Cori cycle (p. 46R), shift part of the metabolic burden of muscle to the liver. In addition, a large amount of alanine is formed in active muscle by the tratlSatl1ination of pyru vate. Alanine, like lactate, can be converted into glucose by the liver. Why does the muscle release alanine? Muscle can absorb and transaminate branched-chain amino acids in order to use the carbon skeletons as fuel; however, it cannot form urea. Consequently, the nitrogen is released into the blood as alanine. The liver absorbs the alanine, removes the nitrogen for disposal as urea, and processes the pyruvate to glucose or fatty acids .
TASlE 27.1 Fuel reserves in a typica l 70-kg m an A VAILABLE EN ERGY IN K ILOJOULES (KCAL)
Organ
Blood liver Brain Muscle Adi pose ti ssue
Gl ucose or glycogen
Triacylglycero ls
250 (60) 1700 (400) 30 (8) 5000 (1200) 330 (80)
200 (45) 2000 (450) 0 (0) 2000 (450) 560,000 (135,000)
Source: After G. F. CahilL Jr. Clin. Endocr inol. M cn ab. 5(1976):398.
M obi lizable prot eins (0) 0 1700 (400) 0 (0) 100,000 (24,000) 170 (40)
767 27.2 Metabolic Profi les of Organs
768
MUSCLE
LIVER
CHAPTER 27 The Integration of Metabolism
Glucose 6-phosphate Glycogen
Glucose
Glucose Glycolysis
Gluconeogenesis
Pyruvate
Lactate
.,-
Pyruvate
Lactate
Alanine
Figure 27.12 Metabolic interchanges between muscle and the liver.
Alanine
Protein degradation
Unlike skeletal muscle, h eart muscle functions almost exclu sively aerobically, as ev idenced by the density of mitocho ndria in h eart muscle. Moreover, the heart has virtually no glycogen reserves. Fatty acids are the h eart's m ain source of fuel, although ketone bodi es as well as lactate can serve as fuel for heart muscle.
3. Adipose Tissue. Th e triacylglycerols stored in adipose tissue are an enor· mous reservoir of metabolic fuel (see Table 27.1). In a typical 70 -kg man, the
Glucose VLDL (from the liver) (from the liver) ~
Glucose
•
Glycerol 3-phosphate ~
Chylomicron (from the intestine)
Fatty acids
Fatty acyl CoA
Triacylglycerols Hormone-
sensitive lipase Glycerol
Fatty acids
Glycerol Fatty acid- albumin (to the liver) complexes (to peripheral tissue) Figure 27.13 SyntheSiS and degradation of triacylglycerols by adipose tissue. Fatty ac ids fro m the liver are delivered t o adipose cells in th e for m of triacylglycero ls cont ained in very low density lipo p ro teins (VLDLs). Fatty acids fro m t he diet are transport ed in chylo microns.
13 kg of triacylglycer o ls have an en ergy content of 565 ,000 kJ (135,000 kcal). Adipose tissue is specialized for the esterification offatty acids to form triacylglycerols and for their release from triacylglycerols. In human beings, the liver is the major site of fatty acid synthesis, although, in the developed world, most people obtain most of their fatty acids from their diets. Dietary fat s are delivered to the adipose ti ssu e from the intestines by chy lo microns. Fatty acids in the liver are est erifi ed to glycerol phosphate to form triacylgl ycerol and are transported to the adipose tissue in lipoprotein particles , such as very low d ensity lipoproteins (p. 743). Triacylglycerols are not taken up by adipocytes; rather, th ey are first hydrolyzed by an extracel· lular lipoprotein lipase for uptake. This lipase is stimulated by processes ini· tiated by insulin. After the fatty acids enter the cell, the principal task of adi· pose tissue is to activate these fatty acid s and transfer the . resulting eoA derivatives to glycerol in the form of glycerol 3-phosphate . This essential intermediate in lipid biosynthesis com es from the redu ction of the glycolyti c intermediate dihydroxyacetone phosphate. Thus, adipose cells need glucose for the synthesis uf triacylglycerols (Figure 27.13). Triacylglycerols are hydrolyzed to fatty acids and glycerol by intracellu· lar lipases. The release of the first fatt y acid from a triacylglycerol, the ratelimiting step , is catalyzed by a hormone-sen sitive lipase that is reversibly phosphorylated. The hormone epinephrine stimulates the formation of cyclic AMP, the intracellular m essenger in the amplifying cascade, which activates a protein kinase a rec urring theme in hormon e action. Triacylglycerols in adipose cells are continually being hydrolyzed and resynthesized. Glycerol d erived from their hydroly sis is exported to the liver. Most of the fatt y acids formed on hydrolysis are reesterified if glycerol 3-phosphate is abundant. In contrast, they are released into the plasma if glycerol 3-ph osphate is scarce because of a paucity of glucose. Thus, the availability of glucose inside adipuse cells is a major factor in determining
whether fatty acids are released intu the blood. 4.
The Kidneys. The major purpose of the kidneys is to produce urine, which
serves as a vehicle for excretin g metabolic waste products and for maintaining the osmolarity of the body fluid s. The blood plasma is filtered nearly 60
times each d ay in the renal tubules. Most of the m aterial filtered ou t of th e blood is reabsorbed ; so onl y 1 to 2 liters of urine is p roduced . Water -solubl e materials in th e plasma, such as glucose, and water itself are reabsorbed to prevent wasteful loss . T he kidneys require large am ounts of energy to ac complish t he reabsorption . A lthough constitu ting onl y 0.5% of body m ass, the kidneys con sume 10% of the oxygen used in cellular respiration . M uch of the glucose that is reabsorbed is carried into t he kidney cells by the sodium- glucose co transporter. This transporter is powered by the N a -I-_K -Igradient, which is itself maintained by the N a -I-- K -I- ATPa~e (Section 13.4 ). During starvation , the kidney becom es an important site of gluconeogenesis and -may contribute as much as half of the blood glucose . 5.
The Liver. The metabolic activities of the liver are essential f or providing fuel to the brain, muscle, and other perip heral organs. Indeed , the liver , which can be from 2% to 4% of body we ight, is an organism's m etabolic hub (Figure 27. 14) . Most compound s absorbed by the intestine first pass through the liver, which is thus abl e to regulate the level of many m etabo lites in the blood . Let us first consider how the liver m etabolizes carbohydrates . T he liver removes two- thirds of the glucose from the blood and all of th e rem aining monosaccharides after m eals. Some glucose is left in th e blood for use by other tissues. T he absorbed glucose is converted into glucose 6-phosphate by hexokin ase and the liver -specific glucokinase. G lucose 6-phosphate, as already stated , has a variety of fates, although th e li ver uses little of it to meet its own energy needs. M uch of the glucose 6-phosphate is converted into glycogen . As much as 1700 kJ (400 kcal ) can be stored in this form in the liver. Excess glucose 6- phosphate is m etabolized to acetyl CoA , which is used to form fatty aci ds, cholesterol, and bile salts. T he pen tose phos phate pathway, anoth er m eans of processing glucose 6-phosphate, supplies the NADPH for these reductive biosyntheses. The liver can produce glucose for release into the blood by breaking down its store of glycogen and by carrying out gluconeogenesis. The main precursors for gluco neogenesis are lactate and alanine from muscle, glycerol from adipose ti ssue, and gluco geni c amino acids from the diet. T he liver also plays a central role in the regul ati on of lipid metabolism. When fuels are abundant, fa tty acids deri ved from the diet or synthesized by the li ver are esterified and secreted into the blood in the form of very low density lipop rotein (see Figure 27. 13). H owever, in the fasting sta te, the liver converts fatty acids into ketone bodies. H ow is the fate of liver fatty acids determined ? The selection is m ad e according to whether the fatty acids en ter the mi tochondrial m atrix. Recall that long-chain fatty acids traverse the inner mitochondrial m em brane only if they are esterifi ed to car nitine. Carnit ine acyltransferase I (also known as carnitine palmitoyl transferase I), which catal yzes the formation of acyl carnitine, is inhibited by malonyl CoA , the committed interm ediate in the synthesis of fatty acids (see Figure 27.9 ). Thus, when malonyl CoA is abundant , long-chain fa tty acids are prevented from entering the mitochondria l matrix, the compartment of f3 oxidation and ketone-body f ormation. Instead, fatty acids are exported to adipose tissue f or incorporation into triacylglycero ls. In contrast, the level of malonyl CoA is low when fuels are scarce. U nder these conditions, fatty acids liberated from ad ipose tissues enter the mitochond rial matri x for con version into ketone bodies. The liver also plays an essential role in dietary amino acid metabolism . The liver absorbs the m ajority of amino acids, leaving som e in the blood for peripheral tissues. T he priority use of amino acids is for protein synthesis rather than catabolism . By what m ea ns are amino acids directed to protein
Figure 27.14 Electron micrograph of liver cells. The liver plays an essentia l role in the int egrat ion of metabo lism. [Courtesy of Dr. Ann Hubbard.]
R
) =0
Fatty acyl carnitine
769
770 CHAPTER 27 The Integration of Metabol ism
synthesis in preference to use as a fuel? The KM value for the aminoacyl tRNA synthetases is lower than that for the en zymes taking part in amine acid catabolism. Thus, amino acids are used to synthesize aminoacyl tRNAs before they are cataboli zed . When catabolism does take place, the first step is the removal of nitrogen , which is subseq uently processed to urea. The liver secretes from 20 to 30 g of urea a day. The a- ketoacids are then used for gluconeogenesis or fatty acid synthesis. Tnterestingly, the liver cannot remove nitrogen from the branched-chain amino acids (leucine, isoleucine, and valine). Transamination of these amino acids takes place in the muscle. How does the liver meet its own energy needs? a-Ketoacids derived from the degradation of amin o acids are the liver's own fuel. In fact, the main role of glycolysis in the liver is to form building blocks for biosyntheses. Furthermore, the liver cannot use acetoacetate as a fuel, because it has little of the transferase needed for acetoacetate's act ivation to acetyl eoA. T hus, the liver eschews the fuels that it exports to muscle and the brain.
27.3
Food Intake and Starvation Induce Metabolic Changes
We shall now consider the biochemical responses to a series of physiological conditions. O ur first example is the starved- fed cycle, which we all experience in the hours after an evening meal and through the night's fast. This nightly starved- fed cycle has three stages: the postabsorptive state after a meal, the early fasting during the night, and the refed state after breakfast. A major goal of the many biochemical alterations in this period is to maintain glucose homeostasis that is, a constant bl ood -glucose level.
Figure 27.15 Insulin secretion . The electron micrograph shows t he release of insulin from a pancreatic J3 cell. One secreto ry granule is on the verge of fusing with the plasma membrane and releasing insulin into the extracellular space. and the other has already re leased th e hormone. [Courtesy o f Dr Lel io Orc;, L. Orci, J.-D. Va ssall i, and A. Perrelet. Sci. Am. 259 (Sept ember 1988):85- 94.]
1. The Well-Fed, or Postprandial, State. After we consum e and digest an evening m eal, glucose and amino acids are transported from the intestine to the blood . The dietary lipids are packaged into chylomicrons and transported to the blood by th e lymphatic system . This fed condition leads to the secretion of insulin, which is one of the two most important regulators offuel m etabolism, the other regulator being glucagon . The secretion of the hormone insulin b y the!) cells of the pancreas is stimulated b y glucose and the parasympathetic nervous system (Figure 27 .1 5). Tn essence, insulin signals the fed state; it stimulates the storage of fuels and the synthesis of proteins in a variety of ways. For instance, insulin initiates protein kinase cascades. These cascades stimulate glycogen synthesis in both muscle and the liver and suppresses gluconeogenesis by the liver. Insulin also accelerates glycolysis in the liver, which in turn increases the synthesis of fatty acids. The liver helps to limit the amount of glucose in the blood during times of plenty by storing it as glycogen so as to be able to release glucose in times of scarcity. How is the excess blood glucose present after a meal removed? The liver is able to trap large quantities of glucose because it possesses an isozyme of hexokinase called glucokinase. Recall that glucokinase has a high KM value and is thus active only when blood-gl ucose levels are high. Furthermore, glucokinase is not inhibited by glucose 6-phosphate as is the hexokinase with the low KM value. Consequently, the liver forms glucose 6phosphate more rapidly as the blood-glucose level rises. The increase in glucose 6-phosphate coupled with insulin action leads to a buildup of glycogen stores. The hormonal effects on glycogen synthesis and storage are reinforced by a direct action of glucose itself. Phosphorylase a is a glucose sensor in addition to being the enzyme that cleaves glycogen. When the glucose level is high, the
binding of glucose to phosphorylase a renders the enzyme susceptible to the action of a phosphatase that converts it into phosphorylase b, which does not readily d egrad e glycogen. Thus , glucose allosterically shifts the glycogen system f rom a degradative to a synthetic mode. The high insulin level in the fed state also p rom otes the entry of glucose into muscle and adipose tissue. Insulin stimulates the syn thesis of glycogen by muscle as well as by the liver. The entry of glucose into adipose tissue provides glycerol :l -phosphate for the synthesis of triacylglycerols. The ac tion of insulin also extends to amino acid and protein metabolism . Insulin promotes the uptake of branched -chain amino acids (valine, leucine, and isoleucine) by muscle. Indeed , insulin has a general stimulating effect on protein synthesis, which favors a building up of muscle p rotein. In addition , it inhibits the intracellular degradation of proteins.
2. The Early Fasting, or Postabsorptive, S tate. The blood- glucose level begins to drop several hours after a meal, leading to a d ecrease in insulin secretion and a ri se in glucagon secretion ; glucagon is secreted by the Ct cells of the pancreas in respon se to a low blood- sugar level in the fasting state. Just as insulin signals the fed state, glucagon signals the starved state. It serves to mobilize glycogen stores when there is no dietary intake of glucose. The main target organ of glucagon is the liver. G lucagon stimulates glycogen breakdown and inhibits glycogen synthesis by tri ggering the cyclic A M P cascade leading to the phosphorylation and activation of phosphorylase and the inhibition of glycogen synthase (Section 21.5) . G lucagon also inhibits fatty acid synthesis by diminishing the production of pyruvate and by lowering the acti vity of acetyl CoA carboxylase by m aintaining it in a phosphorylated state. In addi tion , glucagon stimulates gluconeogenesis in the liver and blocks glycol ysis by lowering the level of F -2,6-BP. All known actions of glucagon are m ediated b y protein kinases that are activated by cyclic AMP. T he activation of the cyclic A M P cascad e results in a higher level of phosphorylase a activity an d a lower level of glycogen synthase a activity. G lucagon 's effect on thi s cascade is reinforced by the diminished binding of glucose to phosphorylase a, w hich makes the enzy me less susceptible to the hyd rolytic acti on of the phosphatase . Instead, the phosphatase rem ains bound to phosphoryl ase a, and so the synthase stays in the inactive phosphorylated form . Consequently, there is a rapid m obilization of glycogen . The large am ount of glucose form ed by the hydrolysis of glucose 6- phosphate derived from glycogen is then released from the liver into the blood. The entry of glucose into muscle and adipose tissue decreases in response to a low insulin level. T he diminished utilization of glucose by muscle and adi pose tissue also contributes to the maintenance of the blood -glucose level. The net result of these actions of glucagon is to markedly increase the release of glucose by the liver. Both muscle and the liver use fatty acids as fuel when the blood -glucose level drops . Thus, the blood-glucose level is kept at or above 4.4 M (80 mg/ dl) by three major factors: (1) the mobilization of glycogen and the release of glucose by the liver, (2 ) the release of fatty acids by adipose tissue, and (3) the shift in the fuel used from glucose to fa tty acids by muscle and the liver. W hat is th e resul t of the depletion of the li ver's glycogen stores? Gluconeogenesis from lactate and alanine con tinues, bu t this p rocess merely repl aces glucose that had already been converted into lactate and alanine by the peripheral tissues . Moreover, the brain oxidizes glucose com pletely to CO 2 and H 2 0. T hus, for the net synt hesis of glucose to take place, another source of carbon is required . G lycerol released from adipose tissue on lipolysis provides some of the carbon atom s, with the rem aining carbon atoms coming from the hydrolysis of muscle proteins.
771 27.3 Food Intake and Starvation
772
3.
CHAPTER 27 The Integration of Metabolism
fast ? Fat is processed exactly as it is processed in the normal fed state. However, this is n ot the case for glucose. The liver does not initially absorb glucose from the blood, but, instead , leaves it for the peripheral tissues. Moreover, the liver remains in a gluconeogenic mode. Now, however, the newly synthesized glucose is u sed to replenish the liver's glycogen stores. As th e blood-glucose levels continue to rise, the liver completes the replenish· m ent of its glycogen stores and begins to process the remainin g excess glu· cose for fatty acid synthesis.
The Refed State. What are the biochemical responses to a hearty break·
Metabolic Adaptations in Prolonged Starvation Minimize Protein Degradation
....
6
Ketone bodies
-OJ [j '"E '" 0:
- - - Glucose
3 2
~
/ _---------- Fatty :acids
1
o
2
4
6
B
Days of starvation Figure 27.16 Fuel choice during starvation. The plasma levels of fatty acids and keto ne bodies increase in starvation, whereas t hat of glucose decreases.
What are the adaptations if fasting is prolonged to the point of star· vation ? A typical well -n ourish ed 70-kg man has fuel reserves totaling about 670,000 kJ (16 1,000 kcal; see Table 27. 1). The en ergy need for a 24 -hour period ranges from about 6700 kJ (1600 kcal) to 25, 000 kJ (6000 kcal), depe nding on the extent of activity. Thus, stored fuels suffice to meet calori c need s in starvation for 1 to 3 m onths. H owever, the carbohydrate reo serves are exh au sted in only a day. Even under starvation conditions, the blood-glucose level must be main· tained above 2.2 mM (40 mg/ dl). The first priority of metabolism in starvation is to provide sufficient glucose to the brain and other tissues (such as red blood cells) that are absolutely dependent on this fuel. H owever , precursors of glucose are not abundant. Most energy is stored in the fatty acyl moieties of triacylglycerols. Recall that fatty acids cannot be converted into glucose, because acetyl CoA cannot be transformed into pyruvate (p . 634). The glycerol moiety of triacylglycerol can be converted into glucose, but only a limited amount is available. The only other potential source of glucose is amino acids derived from the breakdown of proteins. However , proteins are not stored, and so any breakdown will necessitate a loss of function . Thus, the second priority of me· tabolism in starvation is to preserve protein, which is accomplished by shifting the fuel being used from glucose to fatty acids and ketone bodies (Figure 27. 16). The metabolic changes on the first day of starvation are like those after an overnight fast. The low blood -s ugar level leads to decreased secretion of insulin and increased secretion of glucagon. The dominant metabolic processes are the mobilization of triacylglycerols in adipose tissue and gluconeogenesis by
the liver. The liver obtains energy for its own needs by oxidizing fatty acids released from adipose tissue. The concentrations of acetyl CoA and citrate consequently increase, which switches off glycolysis. The uptake of glucose by muscle is markedly diminished becau se of the low insulin level, whereas fatty acids enter freely. Consequently, muscle uses no glucose and relies exclusively on fatty acids for fuel. The J3 oxidation of fatty acids by muscle halts the conversion of pyruvate into acetyl CoA, because acetyl CoA stimulates the phosphorylation of the pyruvate dehydrogenase complex, which renders it inactive (p . 490). H ence, pyruvate, lactate, and alanine are exported to the liver for conversion into glu cose. Glycerol derived from the cleavage of tri acylglycerols is another raw material for the syn th esis of glucose by the liver. Proteol ysis also provides carbon skeletons for gluconeogenesis; During starvation, degraded proteins are not replenished and serve as carbon sources for glucose synthesis. Initial sources of protein are those that turn over rapidly, such as proteins of the intestinal epithelium and the secretions of the pancreas . Proteolysis of muscle protein provides som e of threecarbon precursors of glucose. However, survival for most animals depends on being able to move rapidly, which requires a large muscle m ass, and so muscle loss must b e minimized .
o
TABLE 27.2 Fue l m eta bo lism in sta r vat io n A MOUNT FORM ED O R C ONSUMED IN
fuel exchanges and co nsumption Fuel use by the brain Glucose Ketone bodies All other use of glucose
3d day
24
H O URS (GRAMS)
Acetyl CoA
40th day COA
100 50 50
40 100 40
o
o /
CoA
5 Fuel mobi li zation
Ad ipose-t issue lipolysis Muscle-"protein degradation Fuel output of t he liver Glucose Ketone bod ies
Acet oacetyl CoA
180
180
75
20
150 150
80
____ Acetyl CoA + H, O
CoA
150
OH 0 - OOC
How is the loss of muscle curtailed? After about 3 days of starvation, the liver forms large amounts of acetoacetate and u -3 -hydroxybutyrate (ketone bodies; Figu re 27. 17). Their synthesis from acety l CoA increases markedly because the citric acid cycle is unable to oxid ize all the acetyl u nits generated by the degradation of fatty acids. Gluconeogenesis depletes the supply of oxaloacetate, which is essential for the entry of acetyl CoA into the citric acid cycle. Consequently, the liver produces large q u antities of ketone bodies, which are released into the blood. At this time, the brain begins tv cunsume significant amounts of acetoacetate in place of glucvse. After 3 days of starvation, about a t hird of the energy needs of th e brain are met by ketone bodies (Table 27 .2) . The heart also uses ketone bodies as fuel.
' ~"-.../
bodies are equivalents vf fatty acids that are an accessible fuel source for the brain . Only 40 g of glucose is then needed per day for the brain, compared with about 120 g in the f irst day of starvation . The effective conversion vf fatty acids into ketone bodies by the liver and their use by the brain markedly diminishes the need for glucose. Hence, less muscle is degraded than in the first days of starvation . The breakdown of 20 g of muscle daily compared with 7S g early in starvation is most important for survival. A person's survival time is mainly determined by the size of the triacylglycerol depot. What happens after depletion of the triacylglycerol stores? The only source of fuel that remains is protein. Protein d egradation accelerates, and death inevitably results from a loss of heart, liver, or kidney function.
COA
5
l -Hydroxy-l -methylgl utaryl CoA (HMG. CoA)
Acetyl CoA
CH, -OOC Acetoacetate /
After several weeks of starvation, ketvne bodies become the major fuel of the brain. Acetoacetate is activated by the transfer of CoA from succinyl CoA to give acetoacetyl CoA (Figure 27. 18). C leavage by thiolase t h en yields two molecules of acetyl CoA, which enter the citric acid cycle. In essence, ketone
/
NADH +H+
OH OH - OOC D-3- Hydroxybutyrate
Figure 27.17 Synthesis of ketone bodies by the liver.
o ~.........
COO-
Acetoacetate
/
Succinyl CoA
CoA transfera se
Succinate
o
Metabolic Derangements in Diabetes Result f rom Relative Insulin Insufficiency and Glucagon Excess
W
W e now consider diabetes mellitus, a complex disease characterized ~ by grossly abnormal fuel usage: glucose is overproduced by the liver and underutilized by other organs. The incidence of diabetes mellitus (usually referred to simply as diabetes) is about 5% of the population. Indeed, diabetes is the most common serious metabolic disease in the world; it affects hundreds of millions. Type 1 diabetes, or insulin-dependent diabetes mellitus (IDDM), is caused by the autoimmune destruction of the insulin -secreting 13 cells in the pancreas and usually begins before age 20 . Insulin dependency means that the affected person requ ires the administration of insulin to li ve. Most diabetics, in contrast, have a normal or even higher level of insulin in their blood , but
0 ......... COA
H,C
C
5
H,
Acetoacetyl eoA COA
Thiolase
o 2 H,C
......... CoA
5
Acetyl CoA
Figure 27.18 Entry of ketone bodies into the c itric acid cycle.
773
Diabetes Named for the excessive urination in the disease. Aretaeu s. a Cappadocian physician of the second century A.D., wrote: "The epithet diabetes has been assigned to the disorder. being sornething like passing of water by a Siphon." He perceptively charactefi z:ed
diabetes as "being a melti ng-dow n o f the fle sh and lirnbs in to uri ne." Mellitus From Latin, meaning "sweetened with honey." Refers to the presence of sugar in the urine of patients having th e disease. Mellitus distinguishes thi s disease from diabetes insipidus. which is caused by impaired renal reabsorption of water.
they are quite unresponsive to the hormone. This form of the di sease, known as type 2, or non-insulin-dependent diabetes mellitus (NIDDM ), typically arises later in life than does the insulin-dependent form. In type 1 diabetes, insulin production is insufficient and consequently glucagon is present at higher-than -normallevels. In essence, the diabetic person is in biochemical starvation mode despite a high concentration of blood glucose. Because insulin is deficient, the entry of glucose into adipose and muscle cells is impaired . The liver becomes stuck in a gluconeogenic and ketogenic state. The excessive level of glucagon relative to that of insulin leads to a decrease in th e amount of F -2, 6- BP in the liver . H ence, glycolysis is inhibited and gluconeogenesis is stimulated because of th e opposite effects of F -2,6- BP on phosphofructokinase and fructose -1, 6-bisphosphatase (Section 16.4; see also Figures 27.4 and 27 .6). The high glucagonlinsulin ratio in diabetes also promotes glycogen breakdown. Hence, an excessive amount of glucose is produced by the liver and released into the blood. Glucose is excreted in the urine (hence the name mellitus) when its concentration in the blood exceeds the reabsorptive capacity of the renal tubules. Water accompanies the excreted glucose, and so an untreated diabetic in the acute phase of the disease is hungry and thirsty_ Because carbohydrate utilization is impaired, a lack of insulin leads to the uncontrolled breakdown of lipids and proteins. Large amounts of acetyl CoA are then produced by 13 oxidation. However, much of the acetyl CoA cannot enter the citric acid cycle, because there is insufficient oxaloacetate for the condensation step. Recall that mammals can synthesize oxaloacetate from pyruvate, a product of glycolysis, but not from acetyl CoA; instead, they generate ketone bodies. A striking feature of diabetes is the shift in fuel usage from carbohydrates to fats; glucose, more abundant than ever, is spurned. In high concentrations, ketone bodies overwhelm the kidney's capacity to maintain acid- base balance. The untreated diabeti c can go into a coma because of a lowered blood-pH level and dehydration. Type 2, or non-insulin-dependent, diabetes accounts for more than 90% of the diabetes cases and is the most common metabolic disease in the world . In the United States, it is the leading cause of blindness, kidney failure, and amputation. The hallmark of type 2 diabetes is insulin resistance. The 13 cells of the pancreas secrete normal or even greater -than -normal amounts of insulin, but the tissues do not respond to the hormone despite the fact that the insulin receptor is functional. Sometimes the 13 cells fail , leading to type 1 diabetes. The exact cause of type 2 diabetes remains to be elucidated; obesity is a significant predisposing factor. Caloric Homeostasis Is a Means of Regulating Body Weight
..
In the United States, obesity has become an epidemic, with nearly .. 30% of adults classified as obese. Obesity is identified as a risk factor in a host of pathological conditions including diabetes mellitus, hypertension, and cardiovascular disease. The cause of obesi ty is quite simple in the vast majority of cases: more food is consumed than is needed, and the excess calories are stored as fat . Although the proximal cause of obesity is simple, the biochemical means by which caloric homeostasis and appetite contiol are usually mairitained is enormously complex, but two important signal molecules are insulin and leptin_ A protein consisting of 146 amino acids, leptin is a hormone secreted by adipocytes in direct proportion to fat mass. Leptin acts through a membrane receptor, related in structure and mechanism of action to the growthhormone receptor, in the hypothalamus to generate satiation signals. During periods when more energy is expended than ingested (the starved 774
state), adipose tissue loses mass. Under these conditions, the secretion of both leptin and insulin declines, fuel utilization is increased, and energy stores are used. The converse is true when calories are consumed in excess. The importance of leptin to obesity is dramatically illustrated in mice. Mice lacking leptin are obese and will lose weight if given leptin. Mice that lack the leptin receptor are insensitive to leptin administration . Preliminary evidence.indicates that leptin and its receptor playa role in human obesity, but the results are not as clear-cut as in the mOLise. As stated earlier, obesity is a predisposing factor for type 2 diabetes. What is the biochemical basis for this relation? Although much remains to be determined, recent research suggests that the adipocytes secrete a hormone called resistin (resistance to insulin) that renders tissues insensitive to insulin. Moreover, the amount of resistin secreted is directly proportional to fat mass. The precise physiological role of resistin remains to be determined . Insulin and leptin can be thought of as long-term regulators of caloric homeostasis, but short -duration hormones also have a role. For instance, cholecystokinin is released by the gastrointestinal tract during eating and binds to specific receptors in the brain, promoting a sense of fullness. On the other hand, the appetite-stimulating gastric peptide gherlin is secreted when the stomach is empty. The complex interplay of the genes and their products of the neuroendocrine system that control energy balance will be an exciting area of research for some ti me to come.
27.4
Fuel Choice During Exercise Is Determined by the Intensity and Duration of Activity
The fuels used in anaerobic exercises sprinting, for example differ from those used in aerobic exercises such as distance running. The selection of fuels during these different forms of exercise ill ustrates many important facets of energy transduction and metabolic integration. ATP directly powers myosin, the protein immediately responsible for converting chemical energy into movement (Chapter 34). However, the amount of ATP in muscle is small. Hence, the power output and, in turn, the velocity of running depend on the rate of ATP production from other fuels. As shown in Table 27.3, creatine phosphate (phosphocreatine) can swiftly transfer its high-potential phosphoryl group to ADP to generate ATP (p . 416). However, the amou nt of creatine phosphate, like that of ATP itself, is limited . Creatine phosphate and ATP can power intense muscle contraction for ~ to 6 s. Maximum speed in a sprint can thus be maintained for only 5 to 6 s (see Figure 15.7). Thus, the winner in a 1OO-meter sprint is the runner who both achieves the highest initial velocity and then slows down the least. fABLE
27.3 Fuel sources for muscle contraction
cuel sou rce
'-1uscle ATP :reatine phosphate :onversion of muscle glycogen into lactat e
:onversion of muscle glycogen into CO2 :onvers;on of liver glycogen into CO 2 ::::onversion of adipose· tissue fatty acids into CO 2
Maximal rate of ATP productio n (mmol $- ')
Tota l - P available (mmo l)
223
73.3
446
39.1 16.7
6,700 84,000
6.2 6.7
19,000
'lJote: Fuels stored are estimated for a 70-kg person having a muscle mass of 28
4,00 0,00 0
kg.
50urce: After E. Hultman and R. C. Harris. In Principles of Exercise Biochemistry, edited by J. R. Poortmans (Karger. 2004).
,p.78- 119.
775 27.4 Fuel Choice During Exercise
776 CHAPTER 27 The Integration of Metabolism
A lOO -meter sprint is powered by stored ATF', creatine phosphate, and the anaerobic glycolysis of muscle glycogen. The conversion of muscle glycogen into lactate can generate a good deal more ATP, but the rate is slower than that of phosphoryl-group transfer from creatine phosphate. During a - 10-second sprint, the ATP level in muscle d rops from 5. 2 to 3.7 mM, and that of creatine phosphate d ecreases from 9. 1 to 2. 6 m M. The essential role of anaerobic glycolysis is manifested in the elevation of the blood -lactate level from 1.6 to 8. 3 mM. The release of H + from the intensely active muscle concomitantly lowers the blood pH fro m 7.42 to 7.24. This pace cannot be sustained in a lOOO -m eter run (- 132 s) for two reasons. First, creatine phosphate is consum ed within a few seconds . Second, the lactate produced would cause acidosis. Thus, alternative fuel sources are needed. T he complete oxidation of muscle glycogen to CO 2 substantially increases the en ergy yield, but this aerobic process is a good deal slower than anaerobic glycolysis. H owever, as th e distance of a run in creases, aerobic respiration, or oxidative phosphorylation, becom es increasingly important. For instance, part of the ATP consumed in a lOOO-meter run must come from oxidative phosphurylation. Because ATP is produced more slowly by oxidative phosphorylation than by glycolysis (see Table 27.3), the pace is necessaril y slower than in a 1OO-meter sprint. T he championship velocity for the lOOO -meter run is about 7.6 m is , compared with approximately 10.2 m!s for the 1DO -meter event (F igure 27 .1ll ). T he runnin g of a marathon (2 6 miles 385 yard s, or 42 ,200 meters) requires a different selection of fuels and is characterized by cooperation be· tween muscle, liver , and adipose tissue. Liver glycogen complements mus· cle glycogen as an energy store that can be tapped. However, the total body glycogen stores (103 m ol of ATP at best) are insufficient to provide the 150 mol of ATP needed for this grueling -2- hour event. Much larger quantiti es of A TP can be obtained by the oxidation of fatty acids derived from the breakdown offat in adipose tissue, but t he maxinlal rate of ATP generation is slower yet than that of gl ycogen oxidation and is more than 10 -fold slower than that with creatine phosphate. Thus, ATP is generated much more slowly from high-capacity stores than from limited ones , accounting for the different velocities of anaerobic and aerobic events. ATP generationfromfatty acids is essentialfur distance running. However, a marathon would take about 6 hours to run if all the ATP cam e from fatty acid oxidation, because it is much slower than glycogen oxidation. Elite run· ners consume about equal amounts of glycogen and fatty acids during a marathon to achieve a mean velocity of 5.5 m i s, about h alf that of a 100m eter sprint. How is an optimal mix of these fue ls achieved? A luw bloodsugar level leads to a high glucagonlinsu lin ratio, which in turn mubilizes fatty
lOa-meter sprint 11
9.77 s
!
10
~ E 9
1OOO-meter sprint 131.96 s •
/
.~ 8 u -o
7
Marathon run (42,000 meters) 2 hr 4 min 55 s
6
\
Q)
>
Figure 27.19 Dependence of the velocity of running on the duration of the race. The va lues shown are wo rld t rack records.
5
- 1,000
'-~--::'-:-~~--:-:'-:-~~-::-:
I
10
100
Running time (s)
-
---'.10,000
acids from adipose tissue. Fatty acids readily enter muscle, where they are de graded by 13 oxidation to acetyl CoA and then to CO 2 , The elevated acetyl CoA level decreases the activity of the pyruvate dehydrogenase complex to block the conversion of pyruvate into acetyl CoA. Hence, fatty acid oxidation decreases the funneling of glucose into the citric acid cycle and oxidative phosphorylation. G lucose is spared so that just enough remains avail able at the end of the marathon. The simultaneous use of both fuels gives a higher mean velocity than would be attained if glycogen were totally con sumed before the start of fatty acid oxidation.
27.5 Ethanol Alters Energy Metabolism in the Liver
W
Ethanol has been a part of the human diet for centuries. However, its i;jl consumption in excess can result in a number of health problems, most notably liver damage. What is the biochemical basis of these health problems?
Ethanol Metabolism Leads to an Excess of NADH Ethanol cannot be excreted and must be metabolized, primarily by the liver. This metabolism is accomplished by two pathways. The first pathway comprises two steps. The first step, catalyzed by the enzyme alcohol dehy drogenase, takes place in the cytoplasm: A lcohol
CH 3 CH 2 0H + NAD +
dchy"cogcn.", )
Ethanol
CH 3 CHO + NADH + H + Acetaldehyde
The second step, catalyzed by aldehyde dehydrogenase, takes place in mitochondria: .A.ldehydc
CH3CHO + NAD + + H 2 0 Acetaldehyde
dehydrogenase )
CH 3 C()(r
+ NADH + H +
Acetate
Note that ethanol consumption leads to an accumulation of NADH. This high concentration ofNADH inhibits gluconeogenesis by preventing the oxidation of lactate to pyruvate. In fact, the high concentration of NADH will cause the reverse reaction to predominate, and lactate will accumulate. The consequences may be hypoglycemia and lactic acidosis . The overabundance of NADH also inhibits fatty acid oxidation. The metabolic purpose of fatty acid oxidation is to generate NADH for ATP generation by oxidative phosphorylation, but an alcohol consumer's }JADH needs are met by ethanol metabolism. In fact, the excess NADH signals that conditions are right for fatty acid synthesis. Hence, triacylglycerol s accumulate in the liver, leading to a condition known as "fatty liver." The second pathway for ethanol metabolism is called the ethanolinducible microsomal ethanol-oxidizing system (MEOS). This cytochrome P450-dependent pathway (p. 750) generates acetaldehyde and subsequently acetate while oxidizing biosynthetic reducing power, NADPH, to NADP + Because it uses oxygen, this pathway generates free radicals that damage tissues. Moreover, because the system consumes NADPH, the antioxida.nt glutathione cannot be regenerated (Section 20 .5), exacerbating the oxidative stress .
777 27.5 Ethanol
778 CHAPTER 27 The Integration of Metabolism
What are the effects of the other metabolites of ethanol ? Liver mitochondria can convert acetate into acetyl CoA in a reaction requiring ATP. The enzyme is the thiokinase that normally activates short-chain fatty acids. Acetate
+
coenzyme A
+
A TP
-~)
acetyl CoA
pp.
-~)
2P
•
+
AMP
+ PP j
•
However, further processing of the acetyl CoA by the citric acid cycle is blocked, because NADH inhibits two important regulatory enzymesisocitrate dehydrogenase and a-ketoglutarate dehydrogenase. The accumulation of acetyl CoA has several consequences. First, ketone bodies will form and be released into the blood , aggravating the acidic condition already resulting from the high lactate concentration. The processing of the acetate in the liver becomes inefficient, leadi ng to a buildup of acetaldehyde. This very reactive compound forms covalent bonds with many important functional groups in proteins, impairing protein function . If ethanol is consistently consumed at high levels, the acetaldehyde can significantly damage the liver, eventually leading to cell death. Liver damage from excessive ethanol consumption occurs in three stages. The first stage is the aforementioned development of fatty liver. In the second stage alcoholic hepatitis groups of cells die and inflammation results. This stage can itself be fatal. In stage three cirrhosis fibrous structure and scar tissue are produced around the dead cells_ Ci rrhosis impairs many of the liver's biochemical functions. The cirrhotic liver is unable to convert ammonia into urea, and blood levels of ammonia rise. Ammonia is toxic to the nervous system and can cause coma and death_ Cirrhosis of the liver arises in about 25% of alcoholics, and about 75% of all cases ofliver cirrhosis are the result of alcoholism. Viral hepatitis is a nonalcoholic cause of liver cirrhosis_ Excess Ethanol Consumption Disrupts Vitamin Metabolism
The adverse effects of ethanol are not limited to the metabolism of ethanol itself. Vitamin A (retinol ) is converted into retinoic acid, an important signal molecule for growth and d evelopment in vertebrates, by the same dehydrogenases that metabolize ethanol. Consequently, this activation does not take place in the presence of ethanol, which acts as a competitive inhibitor. Moreover, the MEOS system induced by ethanol inactivates retinoic acid. These disruptions in the retinoic acid signaling pathway are believed to be responsibl e, at least in part, for fetal alcohol syndrome as well as the development of a variety of cancers. The disruption of vitamin A metabolism is a direct result of the biochemical changes induced by excess ethanol consumption. Other disruptions in metabolism result from another common characteristic of alcoholics malnutrition . A dramatic neurological disorder, referred to as Wernicke- Korsakoff syndrome, results from insufficient intake of the vitamin thiamine. Symptoms include mental confusion, unsteady gait, and lack of fine motor skills. The symptoms of W ernicke- Korsakoff syndrome are similar to those of beriberi because both conditions result from a lack ofthiamine. Thiamine is converted into the coenzyme thiamine pyrophosphate, a key constituent of the pyruvate dehydrogenase complex . Recall that this complex links glycolysis with the citric acid cycle_ Disru ptions in the pyruvate dehydrogenase complex are most evident as neurological disorders because the brain is normally dependent on glucose for energy generation.
779
0. • •
0.
~.
0.
Q.,
••
~
Prolyl hydroxylase + ascorbate
+ 0., + 0. =
H H Prolyl residue
(j
•
-
a-Ketoglutarate
....o
+ (0., +
)
0. · . • ,
Summary
-
o ., H QH
4-Hydroxyprolyl
-
,
•,
'Q
Succinate
residue ,
f igure 27.20 Formation of 4-hydroxyproline. Proline is hydroxylated at (-4 by the action of prolyl hydroxylase, an enzyme that acti vates molecular o xygen.
Alcoholic sc urvy is occasionally observed because of an insufficient ingestion of vitami n C. Vitamin C is required for the formation of stable collagen fibers , The sym ptoms of scurvy include skin lesions and bloodvessel fragility. Most notable are bleeding gums, the loss of teeth , and periodontal infections, Gums are especially sensitive to a lack of vitam in C because .the coll agen in gums turns over rapidly, What is the biochemical basis for scurvy? Vitamin C is required for the continued activity of prolyl hydroxylase, This enzyme sy nthesizes 4-hydroxyproline, an amino acid that is required in collagen . To form this unusual amino ac id , proline residues on the amino side of glycine residues in nascent collagen chains become hydroxylated . O ne oxygen atom from O 2 becomes attached to C4 of proline while the other oxygen atom is taken up by ex-ketoglutarate, which is converted into succinate (Figure 27,20) , T his reaction is catalyzed by proly l hydroxylase, a dioxygenase, which requires an Fe 2 + ion to activate O 2 , T he enzy me also converts ex-ketoglutarate into succi nate without hydroxylating proline, In this partial reaction, an oxid ized iron complex is formed, which inactivates the enzyme. How is the active enzyme regenerated? Ascorbate (vitamin C) comes to the rescue by reducing the ferr ic ion of the inacti vated enzy me, In the recovery process, ascorbate is oxidized to dehydroascorbic ac id (Figure 27.21 ). Thus, ascorbate serves here as a specific antioxidant, Why does impaired hydroxylation have such devastating consequences? Collagen synthesized in the absence of ascorbate is less stable than the normal protein, H ydroxyp roline stabilizes the collagen triple helix by forming interstrand hydrogen bonds, The abnormal fibers formed by insufficiently hydroxylated collagen account for the symptoms of scurvy,
Summary 27,' Metabolism Consists of Highly Interconnected Pathways The basic strategy of metabolism is simpl e: the formation of ATP, red ucing power, and building blocks for biosyntheses, This complex network of reactions is controll ed by the allosteric interactions and reversible covalent modifications of enzymes and changes in their amou nts, by compartmentatio n, and by interactions between metabolically distinct organs, T he enzyme catalyzing the committed step in a pathway is usuall y the m ost important control site . Opposing pathways such as gluconeogenesis and glycolysis are reciprocally regulated so that one pathway is usually less act ive when the other is highly active.
,"
HQ
.'•
QH
Ascorbic acid
,.'
""
HQ
0
Ascorbate
Oehydroascorbic acid
Figure 27.21 Forms of ascorbic acid (vitamin C). Ascorbate is the ionized form of vitamin C, and dehydroascorbic acid is the oxidized f orm of ascorbate.
780 CHAPTER 27 The Integration of Metabolism
27.2 Each Organ Has a Unique Metabolic Profile T he metaboli c patterns of the brain , muscle, adipose tissue, kidney, and liver are very different. G lucose is essentially the sole fuel for the brain in a well -fed person. D uring starvation, ketone bodies (acetoac· etate and 3-hydroxybu tyrate) become the predominant fuel of the brain . Adipose tissue is specialized for the synthesis, sto rage, and mo· bilization of triacylglycero ls. T he kidney produces urin e and reabsorbs glucose. T he di verse m etaboli c activities of the liver support the other organs. The liver can rapidl y mobilize glycogen and carry out gluconeogenesis to meet the glucose needs of other organs. It plays a central role in th e regulation of lipid m etabolism . When fu els are abundant, fatty acids are sy nthesized, esterified , and sent from the liver to adi· pose tissue. In the fasting state, however, fatty acids are converted into k etone bodies by the liver. 27.3 Food Intake and Starvation Induce Metabolic Changes Insulin signals the fed state; it stimul ates t he fo rmation of glycoger and triacy lglycerols and the syn thesis of proteins. In contrast. gl ucagon signals a low blood -glucose level; it stimulates glycogen breakdown and gluconeogenesis by the liver and triacylglycerol hydrolysis by adipose t iss ue. A fter a m eal, the ri se in the blood· glucose level lead s to an in creased secretion of insulin and de· creased secretion of glucagon . Con seq uentl y, gl ycogen is synthe· si zed in mu scle and the liver. Wh en the blood- glucose level drops several hours later , glucose is th en fo rmed by the degradation of glycogen and b y the gluconeogeni c pathway, and fatt y acids are released by the hyd rolysis of triacy lglycerols. The li ver and muscle then increasin gly use fa tty ac ids inst ead of gl ucose to meet their ow n energy need s so that glucose is conserved for use by the brain and the red blood cell s. T he metabolic adaptations in starvation serve to minimize protein degradation . L arge amoun ts of ketone bodies are formed by the liver fro m fa tty acids and released into th e blood within a few days after the onset of star vation . After several weeks of starvation , ketone bodies become the major fuel of the brain . The diminished need for glucose decreases the rate of muscle breakdown, and so the likelihood of survival is enh anced . Diabetes mellitus, the most common serious metabolic disease, is d ue to metabolic derangem ents resulting in an insufficiency of insulin and an excess of glucagon relative to the needs of the person. The reo sui t is an elevated blood -glucose level, the mobilization of tri acylglyc· erols, and excessive ketone-bod y formation . Accelerated ketone- body for mation can lead to acidosis, co ma, and death in untreated insulin· dependent diabetics . 27.4 Fuel Choice During Exercise Is Determined by the Intensity and Duration of Activity Sprinting and marathon running are powered by di fferent fuels to max imize power output. T he 1OO -meter sprint is powered by stored AT P, creatine phosphate, and anaerobic glycolysis. In contrast, the oxidation of both muscle glycogen and fa tty acids derived from adipose tissue is essential in the runnin g of a marathon , a highl y aerobic process . 27.5 Ethanol Alters Energy Metabolism in the Liver The oxidation of ethanol resul ts in an unregulated overprod uction of N ADH , whi ch has several consequences. A rise in the blood levels of
lactic acid and ketone bod ies causes a fall in blood pH, or acidosis . T he liver is damaged because the excess NADH causes excessive fat formation as well as the generation of acetaldehyde, a reactive molecule. Severe liver damage can result .
781 Selected Read ings
Key Terms allosteric interaction (p. 762)
glu coneogenesis (p. 764)
starved- fed cycle (p . 770)
covalent m odification (p. 762)
glycogen synthesis and degradation (p . 764)
glucose homeostasis (p. 770)
glycolysis (p . 762)
iosulin (p. 770)
phosphofruc tokinase (p. 763)
fatty acid synthesis and degradation (p . 764)
glucagon (p . 770 )
citric acid cycle (p . 76:1 )
glucose 6 - phosphate (p . 765 )
ca loric homeostasi s (p. 774)
oxidative phos pho ry lation (p . 763)
pyruvate (p . 766)
leptin (p. 774 )
pyruvate dehydrogenase complex (p. 763)
acety l eoA (p. 766)
resis tin (p. 775)
pentose phosph ate path way (p. 763)
ketone body (p . 767)
creatine phosphate (p. 775 )
Selected Readings Books Fell, D. 1997. Understanding the Control of Metabolism. Portland Press. Frayn, K. N. 1996. Metabolic Regulation : A Human Perspective. Portland Press. Hargreaves, lv1., and Thompson, lvl. (Eds.). 1999. Biochemistry of Exercise X . Human Kineti cs. Poortmans, J . R. (Ed .). 2004 . Principles of Exercise 13iochemistry. Karger. Iiarris, R. A.. and Crabb, O . W . 2006. Melabolic interrelationships. In 'Iextbook of Biochemistry with Clinical CorrelatiollS (pp. 84Y- 890), edited by T. M. Devlin. Wiley- Liss.
Fuel Metabolism Rolland, F., Winderickx, ]., and Thevelein, j . M . 2001. Glucose-sensing mechanism in eukaryotic cells. Trends Biochem. Sci. 26:310- 317. Rasmussen, B. B., anu Wolfe, R. R. 1999. Regulation offatty aciu oxidation in skeletal muscle. Ann". n ev. Nut •. 19:463- 484 . Hochachka, P. W. 2000. Oxygen, homeostasis, and metabolic regulation. Adv. Exp. Med. Diol. 47 5:3 11 - 3.l5. Holm, E., Scdlaczek, 0., and Grips. E . 1999. Amino aciu metabolism in liver disease. Curro Opin. Clin. Nutr. Metab. Care 2:4 7- 53. Wagcnmakers, A . J. 1998. Protein and amino acio metabolism in human muscle. Adv. Exp. Med. BioI. 441 :307- 319.
Metabolic Adaptations in Starvation Baverel. G ., Ferrier, R. , and Martin , M. 1995. Fuel selection by the kidney: Adaptation to starvation . Proc. Nutr. Soc. 54:197- 212. MacDonald. I. A ., and Webber, J . 1995. Feeding. fasting and starvation: Factors affecting fuel utilization. Proc. Nutr. Soc. 54:267- 274. (,~hil1 , G . F. , jr. 1976. Starvation in man . Clin . Endocrinol. Metab. 5::197- 415. Sugden, M . c., Holness . M . J .. and Palmer, T. N. 1989. Fuel selection and carbon flux during the starved-to-fed transition. 13iochem. j . 263:313- 323.
Diabetes Mellitus Lowe!. B. B., and Shulman , G. 2005. Mitochondrial dysfunction and type 2 diabetes. Science 307:384- 387. Rutter. G . A . 2000 . Diabetes: The importance of the liver. Curro Riol. 10: R736- R 738. Saltiel. A. R. 2001. New perspectives into the molecular pathoge nesis and treatment of type 2 diabetes. Cell 104:5 17-329. Bell, C. I., Pilikis, S. J., Weber. I. T .. and Polonsky, K. S. 1996. Glucokinase mutations, insulin secretion , and diabetes mellitus. AmlU. Rev. Physiol. 58: 171 - 186.
Withers, D . J., and White. M . 2000. Perspective: The insulin signaling system a common link in the pathogenesis of type 2 diabetes. Endocrinology 141 :191 7- 1921 . Taylo r, S. I. 1995. Diabetes mellitus. In The Metabolic Dasis of Inh erited Diseases (7th ed ., pp. 935- 936), edited by C. R. Scriver, A . L. Bea udet. W. S. Sly, D. Valle, j. 13. Stanbury, j. B. Wyngaarden , and D. S. Fred rickson . McGraw - L-Lill .
Exercise Metabolism Shulman, R. G., anu Rothman, D . L. 2001. The "glycogen shunt" in exercising muscle: A role for glycogen in mll~cle energetics and fa · tigue. Pruc. Nat l. Acad. Sci. U. S. A. 9B:457- 461. G leason. T. 1996. Post-exercise lactate metabolism : A comparative review of sites. pathways, and regulation. Annu. Rev. Physiol. 58:556 58 1. Holloszy, J . 0., and Kohrt, W . M. 1996. Regulation of carbo hyd rate and fat metabolism durin g and after exercise. Annu. nev. Nutr. 16:12 1- 138. H ochachka. P. W ., and McClelland, G. 1:l. 1997. Cellular metabolic homeostasis during large-scale change in ATP turnover rates in muscles. j. Exp. Riol. 200:381- 386 Horowitz , J. F., and Klein, S. 2000. Lipid metabolism during endurance exercise. Am . .J. Clin. Nutr. 72:558S 563S. Wagenmakers, A. ]. 1999. Muscle amino acid metabolism at rest anu du ring exercise. Diabetes N utr. Metab. 12:3 16- 322.
Ethanol Metabol ism Molotkov, A. , and Ouester, G . 2002. Reti nol/ethanol orug interaction during acute alcohol intoxication involves inhibition of retinol metabolism to retinoi c acid by alcohol dehydrogenase. j. Bioi. Chem. 277:22553 22537. Stewart, S., jones, D ., and Day, C. P. 2001. Alcoholic liver disease: New insights into mechanisms anu preventive strategies. Trends A401.
Med.7:40R-41 3. Lieber, C. S. 2000. Alcohol : Its m etabolism and interaction with nutri ents. Annu. Rev. Nutr. 20:395 430. Niemela, 0.1999. Aldehyde-protein adducts in the liver as a result of ethanol- induced oxidati ve stress. Front. Rinsci. 1: D 506- 0513 . Riveros- Rosas, H., Jul ian -Sanchez, A., and Pin., E. 1997. Enzymology of ethanol and acetaldehyde metabolism in mammal s. Arch. M.d.
Res. 28:453-47 1.
782
CHAPTER 27 The Integration of Metabolism
Problems 1. Distinctive organs. What a re the key enzymatic differences between the liver , kidney, muscle, and brain that account for their differing utilization of m etabolic fuels '
2. Missing enzymes. Predict the m ajor consequence of each of the following enzymati c defi cien cies: (a) (b ) (c) (d ) (e) (f)
Hexokinase in adipose ti ssue G lucose 6-phosphatase in liver Carnitine acyl tran sferase 1 in skeletal muscle G lucokinase in li ver Thiolase in brain Kina se in liver that synthesizes fructose 2,6-bisphosphate
9 . An effect of diabetes. In sulin-dependent diabetes is often ac· companied b y hypertriglyceridemia, which is an excess blood level of triacylglycerols in the form of very low density lipoproteins. Suggest a biochemical explanation .
10. Sharing the wealth. The hormone glucagon signifies the starved state, yet it inhibits glycolysis in the liver. How does this inhibition of an energy-production pathway benefit the organism?
11. Compartmentation. G lycolysis takes place in the cytoplasm, whereas fatty acid degradation takes place in mitochondria. What m etabolic pathways depend on the interplay of reactions that take place in both compartments?
3. Contrasting milieux. Cerebrosp inal fluid has a low content of
12. Kwashiorkor. The most common form of m alnutrition in chil·
albu min and other proteins compared with plasma.
dren in the world, kwashiorkor, is caused by a diet having ample calories but little protei n. T he high levels of carbohydrate result in high levels of ins ulin. What is the effect of high levels of insulin on
(a) What effect does this lower content have on the con centration of fatty acids in the extracellular m edium of the brain? (b) Propose a plau sible reason for the selection by the brain of glucose rather than fa tty acids as the prime fue l. (c) H ow does the fu el p reference of muscle compl ement that of the brain ?
(a) lipid utilization ? (b) protein metabolism ? (c) Children suffering from kwashiorkor often have large distended bellies caused by water from the b lood leaking in to extracellular spaces. Suggest a biochemical basis for th is condition.
4. Metabolic energy and power. The rate of energy expenditure o f a typical 70-kg person at rest is about 70 watts (W ), like that of a li ght bulb. (a) Express thi s rate in kilojo ul es per second and in kilocalories per second . (b) How many electrons fl ow through the mitochondrial electron transport chain per second und er these conditions? (c) Estimate the corresp onding rate of ATP production. (d ) The total ATP content of the body is about 50 g. Estimate how often an ATP molecul e turns over in a person at rest.
5. Respiratory quotient (RQ ). This classic m etabolic index is defined as the volum e of CO 2 released divided by the volume of O 2 con sum ed . (a) Ca lcul ate the R Q values for the complete oxidation of glucose and of tripalmitoylglycerol. (b) What do RQmeasurements reveal about the contribu tions of different energy sources during intense exercise? (Assume that protein d egradation is negligible .)
6. Camel's hump. Compare the H , O yield from the complete oxidation of 1 g of glucose with that of 1 g of tripalmitoylglyceral. Relate these values to the evolu tionary selection of the contents of a camel 's hump .
7. The wages of sin. H ow long does one have to jog to offset the calories obtained from eating 10 macadamia nuts (7 5 kJ, or 18 kcal, per nut)? (A ssume an incremental power consumption of 400 W. )
8. Sweet hazard. Ingestin g large amounts of glucose before a marathon might seem to be a good way of increasing the fu el stores . However, experienced runners do not ingest glucose before a race. What is the biochemical reason for their avoidance of this potential fuel' ( Hint: Con sider the effect of glucose inges tion on the level of insulin .)
13. Oxygen de)icit. After light exercise, the oxygen consumed in recovery is approximately eq ual to the oxygen deficit, which is the a mount of additional oxygen that would have been con· sumed had oxygen consumption reached stead y state immedi· ately. How is the oxygen consumed in recovery used?
14 . Excess postexercise oxygen consumption. The oxygen consumed after strenuous exercise stops is signifi cantly greater than the oxygen deficit and is termed excess postexercise oxygen con sumption (EPOC ). Why is so much more oxygen required after intense exercise?
15. Psychotropic effects. Ethanol is unusual in thilt it is freely soluble in both water and lipids. Thus, it has access to all regions of the highly vasculari zed brain. A lthough the molecular basis of ethanol action in the brain is not clear, ethanol evidently influences a number of neurotransmitter receptors and ion channels. Suggest a bio· chemical explanation for the di verse effects of ethanol.
16. Fiber type. Skeletal muscle has several distinct fiber types. Type I is used primarily for aerobic activity, whereas type II is specialized for short, inten se bursts of activity. How could you distingui sh between these types of muscl e fiber if you viewed them with an electron microscope?
17. Tour de France. Cyclists in the Tour de France (more than 2000 miles in 3 weeks) require abo ut 836, 000 kJ (2 00,000 kcal) of energy, or 41, 840 kJ (10,000 kcal) day - I (a resting male requires = R368 kJ, or 2000 kcal, day - 1). (a) With the assumptions that the energy yield of ATP is about 50.2 kJ (12 kcal) m ol- I and that ATP has a molecular weight of 503 g mol - I, how much ATP would be expended by a Tour de France cycl ist? (b ) Pure ATP can be purchased at the cost of approximately $150 per gram. How much wou ld it cost to power a cyclist through the T o ur de France if the ATP had to be purchased'
Chapter
DNA Replication, Repair,
and .Recombination ~
O!ut1fr .ptuB titt1dl~ he tor~,oo( roGer ft tttl'lttJ'cn f),U"',it:" .r.pntt-' J' D;lJ;lw.o t1ftJaa (11'( blluilntill o::th::6 fQJtt- fiidJfC-
t»
t"
ft;z ..
tftrtWUattt'· (i:ru11,.:mh; lttl ~l~
Newly synthesized DNA tC.ldUOlt itubttulnlhl
t&-rndt'"bld 1K tl.rr 4'cttnt m,R'"..p...(tft1 l1 t4t'fIl em n:tb:J:...til\c, tOI !cOtt,"lIm (Ii:' t(b"OIU mltt4.'111tftdc:.ptml 100hLiifU tum tee l'& tC!:TOt Cft(:liU~rrb1lt1lnt41tl 'i:) Amt~ft"tltMU.>t-Htr\tl\' bttrrolci.t Mtn 11 DtU btftr Otd"lml)f:l ~tl m c A:!roUltt"
o ntfc:e bd: fUQMe Ie ( IU to1tr ~ ,';uttu J ~1i" cdl¢ 111 '1"ut.-t:"c:fcfht en 'dI:~"UI fl c fat ad ~........ ~ -t)~
.,IIB
fn 11.9 qtrc ,1M&... '*""..-bilMQ" ~'Itn~ ,n'''' cilGQ\tltlillllltf lilt
Template DNA
Faithful copying is essential to th e storage of geneti c information. With the precision of a diligent monk copying an illuminated manuscript, a DNA polymerase (above) copies DNA strands. preserving the precise sequence of bases wi th very few errors. [(Left)The Pierpont Morgan Library/Art Resource.]
erhaps the most exciting aspect of the structure of DNA deduced by Watson and Crick was, as expressed in their words, that the "specific pairing we have postulated immediately suggests a possible copying mech anism for the genetic material." A double helix separated into two single strands can be replicated because each strand serves as a template on which its complementary strand can be assembled (Figure 2R. l ). To preserve the information encoded in DNA through many cell divisions, copying of the genetic information must be extremely faithful. To replicate the human genome without mistakes, an error rate of less than 1 bp per 3 X lOY bp must be achieved. Such remarkable accuracy is achieved through a multilayered system of accurate DNA synthesis (which has an error rate of 1 per 4 3 10 _10 bases inserted ), proofreading during DNA synthesis (which 7 6 reduces that error rate to approximately 1 per 10 _ 10 bp), and postreplication mismatch repair (which reduces the error rate to approximately 1 per 109_10 10 bp). Even after DNA has been initially replicated, the genome is still not safe. Although DNA is remarkably robust, ultraviolet light as well as a range of chemical species can damage DNA, introducing changes in the DNA sequence (mutations) or lesions that can block further DNA replication
I Outlin e l 28.1 DNA Can Assume a Variety of Structural Forms 28.2 Double-Stranded DNA Can Wrap Around Itself to Form Supercoiled
Structures 28.3 DNA Replication Proceeds by the Polymerization of Deoxyribonucleoside Triphosphates Along a Template 28.4 DNA Replication Is Highly Coordinated 28.5 Many Types of DNA Damage Can Be Repaired 28.6 DNA Recombination Plays Important Roles in Replication , Repair, and Other Processes
783
Replication
) •
Damage
Repair (
Original parent molecule
First-generation daughter molecules Figure 28.1 DNA replication . Each strand of one double hel ix (shown in blue) act s as a t emplate for the synthesis o f a new complemen tary strand (shown in red).
Figure 28.2 DNA Repl ication, damage, and repair. Some errors (shown as a black dot) may occur in t he replicat ion processes. Additiona l defect s (shown in yellow) including modified bases, crosslinks, and single- and double-strand breaks are introduced into DNA by subsequent DNA-damaging reactions. Many of the errors are detected and subsequently repaired .
(Figure 28.2). All organisms contain D NA repair systems that detect DNA dam age and act to preserve the original seq uence. M utations in genes that encode components of DNA repair systems are key factors in the development of cancer. Among the most potentiall y devastating types of DNA damage are double-stranded breaks in DN A. With both strands of the double helix broken in a local region, neither strand is intact to act as a template for futur e DNA synthesis. A mechanism used to repair such lesions relies on DNA recombination that is, the reassortment of D NA sequ ences present on two different doubl e helices. In addition to its role in DNA repair, recombination is crucial for th e generation of genetic diversity in meiosis. Recombination is also t he key to generating a highly diverse repertoire of genes for key molecules in the immune system (Chapter 33). W e begin with a thorough examination of the structural properties of DNA.
28.1
DNA Can Assume a Variety of Structural Forms
T he doubl e- helical structure of D N A deduced by Watson and C rick immediately suggested how genetic information is stored and replicated. As was discussed earlier (Section 4.2), the essential features of th eir model are: 1. T wo polynucleotide chains running in opposite directions coil around a common axis to form a ri ght-handed double helix. 2. Purine and pyrimidine bases are on the inside of the helix, whereas phosphate and deoxyribose units are on th e outside. 3. A denine (A ) is paired with thymine (T ), and guanine (G) with cytosine (C). An A- T base pair is held together by two hydrogen bonds, whereas a G C base pair is held together by three such bonds. The A-DNA Double Helix Is Shorter and Wider Than the More Common B-DNA Double Helix
W atson and C rick based their model (known as the B-DNA helix ) on x-ray diffraction patterns of highly hydrated DN A fibers, which provided information about properties of the double helix th at are averaged over its constituent resid ues. X-ray diffraction studies of less- hydrated DNA fibers revealed a different form called A -DNA, which appears when the relative humidity is reduced to less than about 75%. A -DNA , like B-DNA, is a right-handed double heli x made up of antiparallel strands held together by Watson- C rick base-pairing. The A helix is wider and shorter than the B helix, and its base pairs are tilted rather than perpendicular to the helix axis (Figure 28.3). 784
785 Top • view
28.1 Forms of DNA
Side • view
~ Figure 28.3 B-form and A-form
Bform
A form
Man y of the structural di ffe rences between B-DNA and A-DNA arise trom d ifferent puckerin gs of their ribose units (Figure 28 .4). In A- DNA, C-3' lies ou t of the plane (a conformation referred to as C -3 ' endo) formed by the other four atoms of the fu ranose ring; in B-D N A , C -2' lies out of the plane (a conformation called C -2' endo). The C -3' -endo puckering in A-DNA leads to an I1-degree tilting of the base pairs away from the norm al to the helix. T he phosphates and other gro ups in the A helix bind fewer H20 molecules than do those in B-DNA. H ence, d ehydrati on favo rs the A form . Cellular DNA is generall y B form . H owever, the A helix is not confined to dehydrated DNA . Double-stranded regions of RNA and at least some RNA- DNA hybrid s adopt a double-helical form very similar to that of A-Dl A. T he position of the 2' -hydroxyl group of ribose prevents R NA from fo rm ing a classic Watson- C rick B helix because of steric hindrance : the 2' -oxygen atom would com e too close to three atom s of the adj oining phosphate group and one atom in the next base . In an A- ty pe helix, in co ntrast, the 2' -oxygen projects ou tward , away from other atoms. The Major and Minor Grooves Are Lined by Sequence-Specific Hydrogen-Bonding Groups Double-helical.nucleic acid molecules con tain two grooves, called the major groove and the minor groove. These grooves arise because the glycosid ic bonds of a base pair are not diametrically opposite each other (Figure 2il .5). The minor groove contains the p yrimidine 0- 2 and the purine N -1 of th e base pair, and the major groove is on the opposite side of the pair. T he methyl group of thymine lies in the major groove. In B-D N A , the m ajor
DNA. Space-fi ll ing mo dels o f ten base pa irs of B-form and A-form DNA d epi ct th eir right-handed helical structures. N o t ice that the B-fo rm heli x is lo nger and narrower than the A-form hel ix. The carbon atoms of the bac kbone are shown in w hite. [Drawn from 1BNA.pdb and 1DNZ.pdb.]
COl' ."do (A form )
C-2'
COl ' ."do (8 fo rm)
Figure 28.4 Sugar pucker. In A-form DNA, the C-3' carbon atom lies above t he approximate plane defined by the four o ther sugar nonhydrogen atoms (called C-3' end o). In B-form DNA, each ribose is in a C-2' -endo confo rmatio n. in which C-2 ' lies out of the plane.
786
Major-groove side
Major-groove side
CHAPTER 28 DNA Replication, Repair, and Recombination H. Figure 28.5 Major- and mi nor-groove sides. Because the t wo glycosidic bonds are not diametrically opposite each other, each base pair has a larger side that defines the major groove and a smaller side that defines t he minor groove. The grooves are lined by pot ential hydrogenbond donors (b lue) and acceptors (red).
Major groove
WH
'_'" Q
H--(N /
Glycosidic bond
CH 3
, ~
N
H
Q
H , Q . . - . ·H-N
H
Glycosidic bond
,.---1,
H ""N
\
Glyco;;dic bond
H
H
N=--\N_ W" ' O Gly20sidiC
H
-
Minor-groove side
Minor-groove side
Adenine-Thymine
Guanine-Cytosine
bond
groove is wider (12 versus 6 A) and deeper (8.5 versus 7.5 A) than the minor groove (Figure 28.6) . Each groove is lined by potential hydrogen-bond donor and acceptor atoms that enable specific interactions with proteins (see Figure 28.5). In the minor groove, N-3 of adenine or guanine and 0 -2 of thymine or cyto· sine can serve as hydrogen -bond acceptors, and the amino group attached to C-2 of guanin e can be a hydrogen-bond donor. In the m ajor groove, N-7 of guanine or adenine is a potential acceptor, as are 0 -4 of thymine and 0-6 of guanine. The amino groups attached to C -6 of adenine and C -4 of cytosine can serve as hydrogen -bond donors. Note that the major groove displays more features that distinguish one base pair from another than does the minor groove. The larger size of the major groove in B-DNA makes it more accessi· ble for interactions with proteins that recognize specific DNA sequences. Studies of Single Crystals of DNA Revealed Local Variations in DNA Structure
Minor groove
Figure 28.6 Major and minor grooves in B-form DNA . Notice the presence of the major groove (dep icted in orange) and th e narrower minor groove (depicted in yellow). The carbon at oms of the backbone are shown in w hite.
X-ray analyses of single crystal s of DNA oligomers had to await the devel· opment of techniques for synthesizing large amounts of DNA fragments with defined base sequences. X -ray analyses of single crystals of DNA at atomic resolution revealed that DNA exhibits much more structural variabil· ity and diversity than formerly envisaged. The x-ray analysis of a crystallized DNA dodecamer by Richard Dickerson and his coworkers revealed that its overall structure is very much like a B-form Watson- Crick double helix. However, the dod ecamer differs from the Watson- Crick model in not being uniform; there are rather large local deviations from th e average structure. The Watson- C rick model has 10 residues per complete turn , and so a residue is related to the next along a chain by a rotation of 36 degrees. In Dickerson's dodecamer, the rotation angles range from 28 degrees (less tightly wound) to 42 degrees (more tightly wound). Furthermore, the two bases of many base pairs are not per· fectly coplanar (Figure 28.7). Rather, they are arranged like the blades of a propeller. This deviation from the idealized structure, called propeller twist, enhances the stacking of bases along a strand. These local variations of the
C-8 -t-----...
-- ----- -
-
.- ..... - "-.
-
------
C-1 ' -l
Figure 28.7 Propeller twist. The bases o f a DNA base pair are o ften not preci sel y copl anar. They are twisted with respect to each other, like the blades of a propeller.
Propeller twist
787 28.1 Forms of DNA
~
Top view
Figure 28.8 Z-DNA. DNA o ligo mers such as dCGCGCG adopt an alternative conformation under some cond itions. This conformati on is ca ll ed Z-DNA because the phosphate groups zigzag along the backbo ne. [D rawn from l3lD.pdb.]
Side vi ew
double helix and others depend on base sequence. A protein searching for a specific target sequence in DNA may sense its presence partly through its effect on the precise shape of the double helix.
Z-DNA Is a Left-Handed Double Helix in Which Backbone Phosphates Zigzag
Alexander Rich and his associates discovered a third type of D NA helix when they solved the structure of dCGCGCG. They found that th is hexanucleotide forms a duplex of antiparallel strands held together by Watson- C rick base-pairing, as expected. What was surprising, however , was that this double helix was left- handed, in contrast with the right -handed screw sense of the A and B helices . Furthermore, the phosphates in the back bone zigzagged; hence, they called this new form Z-ONA (Figure 28.8). The Z -DNA form is adopted by short oligonucleotides that have sequences of alternating pyrimidines and purines. High salt concentrations are required to reduce electrostatic repulsion between the backbone phosphates, which are closer to one another than in A - and B-DNA. Under physiological conditions, most DNA is in the B form. Nonetheless, protein domains have been discovered that bind nucleic acids specifically in the Z-form. This observation strongly suggests that such structures are present in cells and perform specific functions. The properties of A-, B-, and Z-ON A are compared in Table 28. 1. TABLE 28.1 Comparison of A-, B-, and Z-DNA H ELIX TYPE
Shape Rise per base pair Helix diamet er
A
B
z
Broadest
Interrnediate
Narrowest
Screw sense
• 2.3 A • 25.5 A Right-handed
23.7 A Right-handed
Glycosidi c bond*
an ti
anti
Base pa irs per turn of helix Pitch per turn of helix
11 • 25.3 A 19'
10.4 35.4 l'
Narrow and very deep Very broad and shallow
Wide and quite deep Narrow and quite deep
Tilt of
bas~
pairs from
•
3.4 A •
• 3.8 A • 18.4 A Left-handed Al tern ating anti and syn
12
A
•
45.6 A
9'
normal t o helix ax is Major groove Minor groove
*Syn and ami refer t o the o rientation of the N-g!ycosidic bond between the base and deoxyribose, In the ami orientation, the base extends away from the deoxy ribose, In the syn orientati on, the base is above the deoxyribose. Pyrimidine can be only in ant; orientat ions, while purines ca n be anti o r syn.
Flat Very narrow and deep
788
Double-Stranded DNA Can Wrap Around Itself to Form Supercoiled Structures
28.2
CHAPTER 28 DNA Replication, Repair, and Recombination
Thus far, we have been considering the secondary stru cture of DNA. DNA double helices can fold up on themselves to form tertiary structures created by supercoiling. Supercoiling is mo ~t readily understood by considering covalently closed DNA molecules, but it also applies to DNA molecules constrained to be in loops by other means. Most D N A molecule. inside cells are subject to supercoi ling. Consider a linear 260-bp DNA duplex in the B-DNA form (Figure 28.9). Because the number of base pairs per turn in an unstressed DNA molecule averages 10.4, this linear DNA molecule has 25 (2 60/10.4) turns. The ends of this helix can be joined to produce a relaxed circular DNA (Figure 28.9B). A different circular DNA can be formed by unwinding the linear dupl ex by two turns before joining its ends (Figure 28.9C). What is the structural consequence of unwinding before ligation ? Two limiting con·
Figure 28.9 Linki ng num ber. The relations between the linking number (Lk), twist ing number (Tw), and writhing number (Wr) of a circul ar DNA molecule revealed schematically. [Aft er w. Saenger, Principles of Nucleic Acid Structure (Springer Verlag,
1984), p. 452.J
1
5
10
15
20
25
I
I
I
I
I
I
eN"" ·"'''-'''-",-",,,-'"'Oe
(A)
25 \
,
1
_,.: -:.-0..-'"'--'"'0",_
#
-~~
r, "
'\"\:
r5
20 _§
~
§ ~
§ ~"\:~"\;:.,,_ I
(B)
Lk
_ • ,-
\
Wr
=0
-,,"~-s:/,· -
15
= 25,
Tw
= 25,
/
10
Relaxed ci rcle
1 I
10
5 I
O-" ....... ___ "'-,",o 3' direction. The leading polymerization at the nucleotide level to give rise to overall strand is syntheSized continuo usly. wherea s t he lagging st rand growth in the 3' -+ 5' direction. is synthesized in short p ieces termed Okazaki fragment s. DNA Ligase Joins Ends of DNA in Duplex Regions
Figure 28.22 DNA ligase reaction. DNA ligase catalyzes the joining of one DNA strand with a free 3' -hydroxyl group t o another with a free 5' -phosphate group. In eukaryotes and archaea. ATP is cleaved t o AMP and PP, t o drive thi s reaction. In bacteria. NAD + is cleaved to AMP and nicot inamide mononucleotide (NM N ).
The joining of Okazaki fragm ents requires an enzyme that catalyzes the joining of the ends of two DNA chains. The existence of circular D A molecules also points to the existence of such an enzyme. In 1967, scientists in several laboratories simultaneously discovered DNA ligase. This enzyme catalyzes the formation of a phosphodiester bond between the 3' hydroxyl group at the end of one DNA chain and the 5 '-phosphate group at the end of the other (Figure 28 .22) . An energy source is required to drive this thermodynamicall y uphill reaction. In eukaryo tes and archaea, A TP is the energy source. In bacteria, NAD + typically plays this role. DNA ligase cannot link two molecules of single-stranded DNA or circularize single-stranded DNA. Rather, ligase seals breaks in double-stranded DNA molecules. The enzyme from E. coli ordinarily forms a phosphodiester bridge only if there are at least a few bases of single-stranded D A on the end of a d ouble -stranded fragment that can come together with those on another fragment to form base pairs. Ligase encod ed by T4 bacteriop hage can link two blunt -ended double- helical fragments, a capability that is exploited in recombinant DNA technology. (NMN)
base_______
base
------~----OH
(NAD+)
ATP
+
pp., + AMP
base
DNA ligase
____ 0
0"",
797
The Separation of DNA Strands Requires Specific Helicases and ATP Hydrolysis
28.3 DNA Polymerization
For a double-stranded DNA molecule to replicate, the two strands of the double helix must be separated from each Domdlll B2 other, at least locally. This separation allows each strand to act as a template on which a new polynucleotide chain can be assembled . Specific enzymes, termed helicases, utilize the energy of ATP hydrolysis to power strand separation. The detailed mechanisms of heli cases are still under investigation. However, the determination of the three___ Domain A2 dimensional structures of several helicases has been a source of insight. We will begin with a bacterial helicase called PcrA because it has been extensively studied, even though it differs from most helicases important to DNA replication in being a monomer. PcrA comprises fo ur domains, hereafter referred to as domains A 1 , A2, Bl, and B2 (Figure 28.23 ). Domain Al contains a Ploop NTPase fold, as was expected from amino acid se quence analysis. This domain participates in ATP bind ing and hydrolysis. Domain BI is homologous to Domain Al Dom ain B 1 domain Al but lacks a P -loop. Domains A2 and B2 have unique structures . From an analysis of a set of helicase crystal structures P-Ioop bo und to nu cleotide analogs and appropriate double~ Figure 28.23 Helicase structure. The bacteria l helicase and single -stranded DNA molecules, a mechanism for PcrA comprises four domains: AI. A2. BI . and B2. Notice the action of these enzymes was proposed (Figure 28.24). that the Al domain incl udes a P- Ioop NTPase fold {i ndicated by the purple shading w ith t he P loop shown in green}. whereas Domains Al and BI are capable of binding si n gle the BI d omain has a sim ilar overall structure but lacks a P-Ioop stranded DNA. In the absence of bound ATP, both do and does not bi nd nucleotides. Single-stranded DNA b inds to mains are bound to DNA. The binding of ATP triggers the Al and BI domains near the interfaces with domains A2 and conformational changes in the P -loop and adjacent B2. [Drawn from 3PJR.pdb] regions that lead to the closure of the cleft between these two domains . To achieve this movement, domain A I re leases the DNA and slides along the DNA strand, moving closer to domain BI. The enzyme then catalyzes the hydrolysis of ATP to form Figure 28.24 Helicase mechanism. ADP and orthophosphate. On product release, the cleft between do Initially. both domains Al and BI of PcrA mains A and B springs open . In this state, however, domain A 1 has a bind single-stranded DNA. On binding of tighter grip on the DNA than does domain BI, and so the DNA is pulled ATP, t he cleft between these domains across domain BI toward domain AI. The result is the translocation of closes and domain Al slides along the the enzyme along the DNA strand in a manner si milar to the way in DNA. On ATP hyd rolYSiS, the cleft opens up, pulli ng the DNA from domain Bl which an inchworm moves . The PcrA enzyme translocates in the 3' ) 5' toward doma in AI. As thi s p rocess is direction . When the helicase encounters a region of double-stranded repeated, double-stranded DNA is DNA, it continues to move along one strand and displaces the opposite unwound. Notice that the dots on the red DNA strand as it progresses. Interactions with specific pockets on the strand . re presenting two locations on the helicase help destabilize the DNA duplex, aided by ATP-induced strand, move as the double helix is conformational changes. unwound.
,
p.
+ -3'
Bl
ATP
\,. :>
ADP
) -~ )
)
~ Figure 28.25 Conserved res idues among helicases. A compari son of the amino acid
sequences of hundreds of helicases revealed seven regions of strong sequence conservation (shown in color). Notice that. when mapped onto the structure of PcrA, these conserved regions lie along the interface between the A1 and B1 do mains and along the ATP-binding surface. [Drawn from 3PJR.pdb.]
Amino acid sequence
Al
81
ATP-binding surface
Processive enzyme
From the Latin procedere, "to go forward."
An e nzyme that ca talyzes multi ple rou nd s of t he elongat io n o r digesti on of a
polymer whil e t he polymer stays bound. A dis tributive enzyme, in contrast, releases its pol ymeric substrate between successive catalytic steps.
DNA-enclosing site
~ Figure 28_26 Structure of a sliding
DNA clamp. The dimeric 13 subunit of DNA polymerase III forms a ring that surrounds the DNA duplex. Notice the central cavity through whi ch the DNA t emplate slides. Clasping the DNA molecule in t he ring. the polymerase enzyme is able t o move w ithout falling off the DNA substrate. [Drawn fro m 2POL.pdb.]
798
Helicases constitute a large and diverse class of enzymes. Some of these enzymes move in a 5' ) 3' direction, whereas others unwind RNA rather than DNA and participate in processes such as RNA splicing and the initiation of mRNA translation. A comparison of the amino acid sequences of hundreds of these enzymes reveals seven regions of striking conservation (Figure 28.25). Mapping these regions onto the PcrA structure shows that they line the A TP -binding site and the cleft between the two domains, consistent with the notion that other helicases undergo conformational changes analogous to those found in PcrA. However, whereas PcrA appears to function as a monomer, other members of the helicase family function as oligomers. The hexameric structures of one important group are similar to that of the F 1 component of ATP synthase (Section 18.4), suggesting potential mechanistic similarities. In particular, usin g a mechanism similar to the binding-change m echanism, the subunits within the helicase may act in a concerted fashion to unwind d ouble-strand ed DNA as one strand is pulled through the center of the hexameri c rin g while the other remains outside. These hexameric helicases include P -Ioops and are members of a class of ATPases called the AAA family .
•
28.4
DNA Replication Is Highly Coordinated
DNA replication must be very rapid , given the sizes of the genomes and the rates of cell division. The E. coli genome contains 4.6 million base pairs and is copied in less than 40 minutes . Thus, 2000 bases are incorporated per second . Enzyme activities must be highly coordinated to replicate entire genomes precisely and rapidly. We begin our consideration of the coordination of DNA replication by looking at E. co li , which has been extensively studi ed_ For this organism with a relatively small genome, replication begins at a single site and continues around the circular chromosome. The coordination of eukaryotic DNA replication is more complex, because there are many initiation sites throughout the genome and because an additional enzyme is needed to replicate the ends of linear chromosomes. DNA Replication Requires Highly Processive Polymerases
Replicative polymerases are characterized by their very high catalytic potency, fidelity, and processivity. Processivity refers to the ability of an enzyme to catalyze many consecutive reactions without releasing its substrate. These pol ymerases are assemblies of many subunits that have evolved to grasp their templates and not let go until many nucleotides have been added. The source of the processivity was revealed by the determination of the three-dimensional structure of the (3 2 subunit of the E. coli replicative polymerase called DNA polymerase III (Figure 28.26). This unit keeps the polymerase associated with the DNA double helix. It has the form of a staro shaped ring. A 1S -A -diameter hole in its center can readily accommodate a duplex DNA mol ecul e, yet leaves enough space between the DNA and the protein to allow rapid sliding during replication . To achieve a catalytic rate
of 1000 nucleotides polymerized per second requires that 100 turns of duplex DNA (a le ngth of 3400 A, or 0.34 mm) slide throu gh the central hole of 132 per second . Thus, f32 plays a key role in replication by serving as a sliding DNA clamp. How does DNA become entrapped inside the sliding clamp? Replicative polymerases also include assemblies of subunits that fu nction as clamp load ers. These enzymes grasp the sliding clamp and, utilizing the energy of ATP binding, pull ap
,II
t
Binding sites for DnaA protein
j
j
I I
I I
~~II~I'--'~------~'------'~---'~--~
I
I I
I I
5'- GA T C TNT TN T T T T - 3' 3'- eTA G A N A A N A A A A -S ' Consensus sequence
Figure 28.30 Origin of replication in E. coli. The oriC locus has a length of 245 bp. It contains a tandem array of three nearly identical 13-nucleotide sequences (green) and fi ve bind ing sites (yellow) for the DnaA protein.
The binding of DnaA proteins to DNA is the first step in the preparation for rep lication. DnaA is a member of the P-loop NTPase family and, more specifically, an AAA ATPase (p . 653). Each DnaA monomer comprises an ATPase domain linked to a DNA-binding domain at its C -terminus. DnaA molecules are able to bind to each other through th eir ATPase domains; a group of bound DnaA molecules will break apart on the binding and hydrolysis of A TP. The binding of DnaA mo lecules to one another signals the start of the preparatory phase, and their breaking apart signals the end of that phase. The DnaA proteins bind to the five high-affinity sites in oriC and then com e together with DnaA molecules bound to lower-affinity sites to form an oligomer, possibly a cyclic hexamer. The DNA is wrapped around the outside of the DnaA hexamer (Figure 28.3 1). 1.
2. Single DNA strands are exposed in the prepriming complex. With DNA wrapped around a DnaA hexamer, additional proteins are brought into play. The hexameric helicase DnaB is loaded arou nd th e DNA with the help of the helicase loader protein DnaC. Local regions of oriC , including the AT regions, are unwound and trapped by single-strand -binding protein . The result of this process is the generation of a structure call ed the prepriming complex, which makes singl e-stranded DNA accessible to other proteins. Significantly, the primase, DnaG, is now able to insert the R NA pnmer. •
3. The polymerase holoenzyme assembles. The DNA polym erase lIT holoen zyme assembles on the prepriming complex, initiated by interactions between DnaB and the sliding clamp subunit of DNA polymerase III . These interactions also trigger ATP hydrolys is w ithin the DnaA subunits, signaling the initiation of DNA replication. The breakup of the DnaA assembly prevents additional rounds of replication from beginning at the replication origin.
DnaA hexamer Figure 28.31 Assembly of dnaA . Monomers o f DnaA bind to their binding sites (shown in yell ow) in oriC and come together to form a complex structure. possibly the cyclic hexamer shown here. Thi s structure marks the o ri gin of rep l ication and favors DNA strand separat ion in the AT-rich sites (green).
802 CHAPTER 2B DNA Replication. Repair. and Recombination
DNA Synthesis in Eukaryotes Is Initiated at Multiple Sites Replication in eukaryotes is mechanistically similar to replication in prokaryotes but is more challenging for a number of reasons. One of them is sheer size: E. coli must replicate 4.6 million base pairs, whereas a human diploid cell must replicate more than 6 billion base pairs. Second, the genetic information for E. coli is contained on 1 chromosome, whereas, in human beings, 23 pairs of chromosomes must be replicated . Finally, whereas the E. coli chromosome is circular, human chromosomes are linear. Unless coun· termeasures are taken, linear chromosomes are subject to shortening with each round of replication. The first two chall enges are met by the use of multiple origins of replication. In human beings, replication requires about 30,000 origins of replication, with each chromosome containing several hundred . Each origin of replication is the starting site for a replication unit, or replicon. In contrast with E. coli, the origins of replication in human beings do not contain regions of sharply defined sequence. Instead, m ore broadly defined AT-rich se· quences are the sites around which the origin of replication complexes (ORCs) are assembled.
1.
The assembly of the ORC is the first step in the preparation for replication.
In human beings, the ORC is composed of six different proteins, each ho· mologous to DnaA. These proteins likely come together to form a hexam· eric structure analogous to the assembly formed by DnaA.
2.
Licensing factors recruit a helicase that exposes single strands of DNA.
After the ORC has been assembled, additional proteins are recruited, in· cluding Cdc6, a homolog of the ORC subunits, and Cdtl. These proteins, in turn, recruit a hexameric helicase with six distinct subunits called Mcm2·7. These proteins, including the helicase, are sometimes called licensing factors because they permit the formation of the initiation complex. After the ini· tiation complex has formed, Mcm2-7 separates the parental DNA strands, and the single strands are stabilized by the binding of replication protein A, a single-stranded -DNA- binding protein.
3.
Two distinct polymerases are needed to copy a eukaryotic replicon . An ini· tiator polymerase called polymerase a begins replication but is soon replaced by a more processive enzyme. This process is called polymerase switching because one polymerase has replaced another. This second enzyme, called DNA polymerase 5, is the principal repli cative polymerase in eukaryotes (Table 28.2). TABLE 28.2 Some types of DNA polymerases Name Prokaryotic Polymerases DNA polymerase I DNA polymerase II (error ' prone polymerase) DNA polymerase III Eukaryotic Pol ymerases DNA polymera se CI. Primase subunit DNA polymerase unit DNA polymerase 13 (error-prone polymerase) DNA polymerase 8
Function
Erases primer and fills in gaps on lagging strand DNA repair Primary enzyme of DNA synthesis
Initiator polymerase Synthesizes the RNA primer Adds stretch of about 20 nucleotides to the primer DNA repair Primary enzyme of DNA synthesis
Replication begins with the binding of DNA polymerase ex. This enzyme includes a primase subunit, used to synthesize the RNA primer, as well as an active DNA polymerase. After this polymerase has added a stretch of about 20 d eoxynucleotides to the primer, another replication protein, called replication factor C (RFC ), displaces DNA polymerase ex. RFC attracts a sliding clamp call ed proliferating cell nuclear antigen (PCNA), which is homologous to the ~ 2 subunit of E. coli polymerase III . The binding of PCNA to DNA polymerase /) renders the enzyme highly processive and suitable for long stretches of replication. Replication continues in both directions from the origin of replication until adjacent replicons meet and fuse. R NA primers are removed and the DNA fragments are ligated by DNA ligase. The use of multiple origins of replication requires m echanisms for ensuring that each sequence is replicated once and only once. The events of eukaryotic DNA replication are linked to the eukaryotic cell cycle (Figure 28.32). The processes of DNA synthesis and cell division are coordinated in the cell cycle so that the replication of all DNA sequences is complete before the cell progresses into the next phase of the cycle. This coordination requires several checkpoints that control the progression along the cycle. A family of small proteins termed cyclins are synthesized and degraded by proteasomal digestion in the course of the cell cycle. Cyclins act b y binding to specific cyclic-dependent protein kinases and activating them. One such kinase, cyclin-dependent kinase 2 (cdk2) binds to assemblies at origins of replication and regulates replication through a number of interlocking mechanisms.
803 28.4 Coordination of DNA Replication
Start
~
G,
5
M
G2 MitosiS.........- - - / begins Figure 28.32 Eukaryotic cell cycle. DNA replicati on and cell division must take place in a highly coordinated fas hion in eukaryotes. Mitosis (M) takes place only after DNA synthesis (5). Two gaps (G, and G,) in time separate the two processes.
Telomeres Are Unique Structures at the Ends of Linear Chromosomes Whereas the genomes of essentially all prokaryotes are circular, the chromosomes of human beings and other eukaryotes are linear. The free ends oflin ear DNA molecules introduce several compli cati ons that must be resolved by special enzymes . In particular, complete replication of DNA ends is difficult because polymerases act only in th e 5' ) 3' direction. T he lagging strand would have an incomplete 5' end after the removal of the RNA primer. Each round of replication would further shorten the chromosome. The first clue to how this problem is resolved came from sequence analyses of the end s of chromosom es, which are called telomeres (from the Greek telos, "an end" ). Telomeric DNA contains hundreds of tandem repeats of a six -nucleotide sequence. O ne of the strands is G rich at the 3' end, and it is slightly longer than the other strand . In human beings, the repeating G -rich sequence is AGGGTT. The structure adopted by telomeres has been extensively investigated . Recent evidence suggests that they may form large duplex loops (Figure 28.33) . It has been proposed that the single-stranded region at the very end of the structure loops back to form a DNA dupl ex with another part of the repeated sequence, displacing a part of the original tel om eric duplex . This looplike structure is formed and stabilized by specific telom ere-binding proteins. Such structures would nicely mask and protect the end of the chromosom e.
G-ric h strand
)
Figure 28.33 Proposed model for telomeres. A Single-stranded segment of t he G-rich strand extends fro m the end of the telomere. In one mo del for telomeres, this single-stranded region invades the duplex to form a large duplex loop.
I I I 5'
Telomeres Are Replicated by Telomerase, a Specialized Polymerase That Carries Its Own RNA Template
OH 3'
Telomere
AGGG q i AT C CCA
cA
A
How are the repeated sequences generated? An enzyme, termed telomerase, that executes this function has been purified and characterized . When a primer ending in GGTT is added to the human enzyme in the presence of deoxynucleoside triphosphates, th e sequences GGTT AGGGTT and GGTTAGGGTTAGGGTT, as well as longer products, are generated. Elizabeth Blackburn and Carol G reider di scovered that the enzyme adding the repeats contains an RNA molecule that serves as the template for elongation of the G -rich strand (Figure 28.34). Thus, the enzyme carries the information necessary to generate the telomere sequ ences. The exact number of repeated sequences is not crucial. Subsequently, a protein component of telomerases also was identified. From its amino acid sequence, this component is clearly related to reverse transcriptases, enzymes first discovered in retroviruses that copy RNA into DNA. Thus, telomerase is a specialized reverse transcriptase that carries its own template. Telomerase is generall y ex pressed at high levels only in rapidly growing cells. Thus, telomeres and telomerase can play important roles in cancer -cell biology and in cell aging .
Telomerase RNA
............. 3· 5'/ Elongation TT AGGG T "TI'" I '" I - - AGGG I I I I I I I T ' OH 3' 5' AATCCCA I C A
' ' '3'
5'
Translocation
____ OH 3'
"T"T-
11 5'
AGGGTIAGG GTI
CA
T
AT CCCA
A
......... 3· 5'' - '
Elongation
"T,"-
I I 5'
TAGGG AGGGTIAGG GT I I I I TI T,
+
I A C CC
CA
I
OH 3 '
AA
' -3'
5'/
Transl ocation
T - OH 3' T"AGGGTT AGGGTTAGGGT I I ~ s' AATC CCA
C
......... 3·
5'
Figure 28.34 Telomere for mation. M echanism o f synthesis o f the G-rich st rand o f tel omeric DNA. The RNA t emplate o f tel o mera se is sho w n in blue and the nucl eot ides added t o the G-rich strand o f the primer are sho wn in red. [Aft er E. H. Blackburn. N a ture 350 (1991):569-573.]
804
A
28.5
Many Types of DNA Damage Can Be Repaired
We have examined how even very large and complex genomes can, in principle, be replicated with co nsiderable fidelity. H owever, DNA does become damaged, both in th e course of replication and through other processes. Damage to DNA can be as simple as the misincorporation of a single base or it can take more complex form s such as the chemical modification of bases, chemical cross-links between the two strands of the double helix, or breaks in one or both of the phosphodi ester backbones. The results may be cell death or cell transformation, changes in th e DNA sequence that can be inherited by future generations, or blockage of the DNA replication process itself. A vari ety of DNA-repair systems have evolved that can recognize these defects and , in many cases, restore th e DNA molecule to its undamaged form . We begin with some of the sources of DNA damage. Errors Can Occur in DNA Replication
Errors introduced in the replication process are the simples t source of damage in the double helix. With the addition • of each base, there is the possibility that an incorrect base might be incorporated, forming a nonWatson- Crick base pair. These non- Watson- C rick base pairs can locall y distort the DNA double helix. Furth ermore, such mismatches can bemutagenic; that is, they can result in perm anent changes in the DNA sequence. When a double helix containing a non- Watson- Crick base pair is replicated, the two daughter double helices will have different sequences because the mismatched base is very likely to pair with its Watson- C rick partner. Errors other than mismatches include insertions, deletions, and breaks in one or both strands. Furthermore, replicative polymerases can stall or even fall off a damaged template entirel y. As a consequence, replication of the genome may halt before it is complete. A variety of mechanisms have evolved to deal with such interruptions, including specialized D NA polymerases that can replicate DNA across many lesions. A drawback is that such polymerases are substantiall y more error prone than are normal replicative polymerases. Nonetheless, these error-prOlle
po/ymerases allow the completion of a draft sequence of the genome that can be at least partl y repaired by D N A -repair processes. D N A recombination (Section 28.6) provides an additional m echanism for salvaging interruptions in DNA replicati on.
805 28.5 Repair of DNA Damage
Some Genetic Diseases Are Caused by the Expansion of Repeats of Three Nucleotides ~
Some ge neti c di seases are cau sed b y the prese nce of D N A se~ qu ences t hat are inheren tl y prone to errors in the course of repl ication. A particul arly important class of such diseases is characterized by the presence of long tandem arr ays of repeats of three nucl eot id es . An example is Huntington di sease, an au tosomal d ominant neurological d is order with a variable age of on set. The mu tated ge ne in th is disease ex presses a protein in the brain call ed huntingtin, w hi ch con tains a stretch of consec uti ve glu tamine residues. These glutam ine resid ues are en coded b y a tand em array of C AG sequences wit hin the gene. In unaffected persons, this array is between 6 and 31 repeat s, whereas, in those with the disease, the array is between 36 and 82 repeats or lon ger. Moreover , the array tend s to beco m e longer from one generation to the next. The co nseq uence is a phenom enon called anticipation : the children of an affected parent t end t o sho w sy mptom s of th e d isease at an earli er age than did t he paren t . T he tendency of these trinucleutide repeats to expand is explained by the fo rmation of alternative structures in DNA replication (F igure 28.35). Part of the array within a template strand can loo p out without disrupting base pairing ou tside this region . In replicati on , D N A polymerase extends this strand through the rem ainder of the array by a poorly understood m echanism, lead ing to an increase in th e number of copies of the trinucleotide sequence. cAG
G A
C A
CG CAGCAGCAGCAGCAGCAGCA GTC GTC GTC GTC GTC GTC GT
A number of oth er neurological diseases are characterized by ex pandin g arrays of trinucl eotide repeats. H ow do these long stretches of repeated amino acids cause disease? For h untingtin, it appears that the polyglutamine stretches become increasingly pro ne to aggregate as their len gth increases; the additional con sequences of such aggregation are still under • • lllvestlgatlOn . •
Bases Can Be Damaged by Oxidizing Agents, Alkylating Agents, and Light
A variety of chemical agents can alter specif ic bases within D N A after repli cation is compl ete. Such mutagens include reactive oxygen species such as hyd roxyl radical. For exampl e, hyd roxyl rad ical reacts with guanine to fo rm -oxoguanin e. 8-0xoguanine is mutagenetic beca use it often pairs with adenine rather than cytosine in D NA replication . Its choice of pairing part ner differs from that of guanine because it uses a di fferent ed ge of the base
Figure 28.35 Triplet repeat expansion. Sequences containing tandem arrays of repeated triplet sequences can be expanded to include mo re repeats by the looping out of some of the repeats befo re replication. The double helix formed from the red template strand will contain additional sequences encompassing the looped-out region.
806 CHAPTER 28 DNA Replication, Repair, and Recombination
Figure 28.36 Oxoguanineadenine base pair. When guan ine is o xidi zed t o 8-oxo guanine, the damaged base can f o rm a base pair with adenine t hrough an ed ge o f the base tha t d oes no t normally parti c ipate in base-pair format ion.
r w·....
H
~
N
O.. • .... H~N"
I N
N N
/
H
'- The presence in DNA of thymine rather than uraci l was an enigma T for many years. Both bases pair with adenine. The only difference between them is a methy l group in thymine in place of the C -5 hydrogen atom in uracil. Why is a methylated base employed in DNA and not in RNA? The existence of an active repair system to correct the deamination of cytosine provides a convincing solution to this puzzle. Cytosine in DNA spontaneously deaminates at a perceptible rate to form uracil. The deamination of cytosine is potentially mutagenic because uracil pairs with adenine, and so one of the daughter strands will contain a U- A base pair rather than the original C- G base pair. This mutation is prevented by a repair system that recognizes uracil to be foreign to DNA (Figure 28 .45 ). The repair enzyme, uracil DNA glycosylase, is homologous to AlkA. The enzyme hydrolyzes the glycosidic bond between the uracil and deoxyribose moieties but d oes not attack thymine-cont ainin g nucleotides. The AP site generated is repaired to reinsert cytosine. Thus, the methyl group on thymine is a tag that distinguishes thymine from deaminated cytosine. If thymine were not used in DNA, uracil correctly in place would be indistinguishable from urac il formed by deamination . The de fect would persist unnoticed, and so a C G base pair would necessarily be mutated to U- A in one of the daughter DNA molec ul es. This mutation is prevented by a repair system that searches for uracil and leaves thymine alone. Thymine is used instead of uraci l in DNA to enhance the fidelity of the genetIc message.
Uracil DNA glycosidase
A•
G
A• • •
•
•
T
T
AP endonuclease
A •
G
A • •
• •
•
T
T
~ DNA polymerase I DNA ligase
•
A
G
• •
•
T
C
A• • •
T
Figure 28.45 Uracil repair. Uridine bases in DNA, formed by the deamination of cytid ine, are excised and repla ced by cytidine.
809
Many Cancers Are Caused by the Defective Repair of DNA
810 CHAPTER 28 DNA Replication, Repair, and Re combination
CI
II
-- p - ~
~H
CI Cyclophosphamide
Cisplatin
As described in C h apter 14, cancers are ca used by mutations in genes associated with growth control. Defects in DNA-repair systems increase the overall frequency of mutations and , hence, the likelihood of cancer-causing mutations. Indeed, the sy nergy between stud ies of mutations that predispose people to cancer and studies of DNA repair in model organisms has been tremendous in revealing the biochemistry of DNA -repair pathways. Genes for DNA -repair proteins are often tumor -supp ressor genes ; that is, they suppress tumor development when at least one copy of t he gene is free of a deleterious mutation. When both copies of a gene are mutated, however, tumors develop at rates greater than those for the population at large. People who inherit defects in a single tumor-suppressor allele do not necessarily develop cancer but are susceptible to developing the disease because only the one remaining normal copy of the gene must develop a new defect to further the development of cancer. Consider, for example, xeroderma pigmentusum, a rare human skin disease _The skin in an affected person is extremely sensitive to sunli ght or ultraviolet light. In infancy, severe changes in the skin become evident and worsen with time. The skin becomes dry, and there is a marked atrophy of the dermis. Keratoses appear, the eyelids become scarred, and the cornea ulcerates. Skin cancer usually develops at several sites. Many patients die before age 30 from metastases of these malignant skin tumors. Studies of xeroderma pigmentosum patients have revealed that mutations occur in genes for a number of different proteins. T hese proteins are components of the human nucleotide-excision-repair pathway, including homologs of the UvrABC subunits. Defects in other repair systems can increase the frequency of other tumors. For example, hereditary nonpolyposis co lorecta l cancer (HNPCC, or Lynch syndrome ) results from defective DNA mismatch repair. HNPCC is not rare as many as 1 in 200 people will develop this fo rm of cancer. Mutations in two genes, call ed hMSH2 and hMLHl, account for most cases of this hereditary predisposition to cancer. The striking find ing is that these genes encode the hum an cou nterparts of M utS and MutL of E. co li. M utations in hMSH2 and hM LHI seem likely to allow mutations to accumulate throughout the genome. Tn time, genes important in controllin g cell proliferation become altered , resulting in the onset of cancer. Not all tumor -suppressor genes are specific to particular types of cancer. The gene for a protein called p53 is mutated in more than half of all tumurs. The pS3 protein helps control the fate of damaged cells. First, it plays a central role in sen si ng DNA damage, especially doubl estranded breaks. Then, after sensing damage, the protein either promotes a DNA -repair pathway or act ivates the apoptosis pathway, leading to cell death. Most mutations in the p53 gene are sporatic; that is, they occur in somatic cells rather than being inherited. People who inherit a deleterious mutation in one copy of the pS3 ge ne suffer from LiFraumeni syndrome and have a high probability of developing several types of cancer. Cancer cells often have two characteristics that make them especially vulnerable to agents that damage DNA molecules. First, they divide frequently, and so their DNA replication pathways are more active than they are in most cells. Second, as already noted, cancer cells often have defects in DNA- repair pathways. Several agents widely used in cancer chemotherapy, including cyclophosphamide and cisplatin, act by damaging DNA.
Many Potential Carcinogens Can Be Detected by Their Mutagenic Acti on on Bacteria
Many human cancers are caused by exposure to chemicals that cause mutations. It is important to identify such compounds that can cause mutations and ascertain their potency so that human exposure to them can be minimized . Bruce Ames devised a simple and sensitive test for detecting chem9 ical mutagens. In the Ames test, a thin layer of agar containing about 10 bacteria of a specially constructed tester strain of Salmonella is placed on a petri plate. These bacteria are unable to grow in the absence of histidine, because a mutation is present in one of the genes for the biosynthesis of this amino acid. The addition of a chemical mutagen to the center of the plate re sults in many new mutations . A small proportion of them reverse the original mutation, and histidine can be synthesized . These revertants multiply in the absence of an external source of histidine and appear as discrete colonies after the plate has been incubated at 37 °C for 2 days (Figure 28.46) . For example, 0.5 fJ..g of 2-aminoanthracene gives 11,000 revertant colonies, compared with only 30 spontaneous revertants in its absence. A series of concentrations of a chemical can be readily tested to generate a doseresponse curve. These curves are usually linear, which suggests that there is no threshold concentration for mutagenesis.
(A)
(8)
Figure 28.46 Ames test. (A) A petri plate containing about 109 Salmonella bacteria that cannot synthesize histidine and (B) a petri plate containing a filter-paper disc with a mutagen, which produces a large number o f revertants that can synthesize histidin e. After 2 days, the revertants appear as rings of colonies around the disc. The small number o f visible colonies in pl ate A are spo ntaneous revertants. [Fro m B. N. Am es, J. M cCann, and E. Yamasake. Mutat. Res. 31(1975):347- 364.]
Some of the tester strains are responsive to base-pair substitutions, whereas others detect deletions or additions of base pairs (frameshifts). The sensitivity of these specially designed strains has been enhanced by the genetic deletion of their excision-repair systems. Potential mutagens enter the tester strains easily because the lipopolysaccharide barrier that normally coats the surface of Salmonella is incomplete in these strains. A key feature of this detection system is the inclusion of a mammalian liver homogenate. Recall that some potential carcinogens such as aflatoxin are converted into their active forms by enzyme systems in the liver or other mammalian tissues. Bacteria lack these enzymes, and so the test plate requires a few milligrams of a liver homogenate to activate this group of mutagens. The Salmonella test is extensively used to help evaluate the mutagenic and carcinogenic risks of a large number of chemicals. This rapid and inexpensive bacterial assay for mutagenicity complements epidemiological surveys and animal tests that are necessarily slower, more laborious, and far more expensive. The Salmonella test for mutagenicity is an outgrowth of
811 28.5 Repair of DN A Damage
812 CHAPTER 28 DNA Replicat ion, Repair, and Recombinati on
studies of gene- protein relations in bacteria. rt is a striking example of how fundamental research in molecular biology can lead directly to important advances in public healt h.
28.6
DNA Recombination Plays Important Roles in Replication , Repair, and Other Processes
M ost processes associated with DNA repl icati on function to copy the genetic message as faithfully as possible. H owever, several biochemical processes require the recombination of genetic material between two DN.A molecules . In genetic recombination , two daughter molecules are formed by the exchange of genetic materi al between two parent molecul es (Figure 28.47). Recombination is essential in the fo llowing processes. 1. W hen replication stalls, recom bination processes can reset the replica· tion machinery so that replication can cont inue. 2.
Some double-stranded breaks in DNA are repaired by recombination.
3. In meiosis, the li mited exchange of geneti c material between paired chromosomes provides a simple mechanism for generating genetic diversity in a pop ulation . 4. As we shall see in Chapter 33, recombination plays a crucial role in generating molecular diversity for antibodies and some other molecules in the immune system . S. Some viru ses employ recombinati on pathways to integrate their genetic material into the D A of a host cell. 6. Recombination is used to manipulate genes in, for example, the gener· ation of "gene knockout" mice (p . 155). Recombination is most efficient b etween D N A sequences that are similar in sequence. In homologous recombination , parent DNA duplexes align at regions of sequence similarity, and new DNA molecules are formed by the breakage and joining of homologous segm ents.
Recombination
)
Figure 28.47 Recombination. Two DNA molecules can recombine with each other to form new DNA molecules that have segments from both pa rent molecules.
RecA Can Initiate Recombination by Promoting Strand Invasion
In m any recombination pathways, a DNA molecule with a free end recombines with a D NA molecu le having no free ends available fo r in teraction . DNA molecules with free end s are the common result of doublestranded DNA breaks, but they m ay also be gen erated in DNA replication if the replication complex stall s. This type of recombination has been studied extensively in E. coli , but it also takes place in other 0 [ ganism s throu gh the action of proteins homologous to those of E. coli. O ften dozens of proteins part icipate in the complete recombination process . H owever, the key protein is RecA, another member of the AAA
813
5'- - - - - -
3'---------
28.6 DNA Recombination
Strand invasion
)
5'================ J'
Figure 28.48 Strand invasion. This process, pro mo t ed by proteins such as RecA , can initiate recombinatio n.
ATPase family. To accomplish the exchange, the single-stranded DNA displaces one of the strands of the double hel ix (Figure 28.48). The resulting three-stranded structure is called a displacement loop or D-Ioop . This process is often referred to as strand invasion. Because a free 3' end is now bound to a contiguous strand of DNA , the 3' end can act as a primer to initiate new DNA synthesis . Strand invasion can initiate many processes, including the repair of double-stranded breaks and the reinitiation of replication after the replication apparatus has come off its template. In the repair of a break, the recombination partner is an intact DNA molecule with an overlapping sequence.
Some Recombination Reactions Proceed Through Holliday-Junction Intermediates
In recombination pathways for meiosis and some other processes, interme diates form that are composed of four polynucleotide chains in a crosslike structure. Intermediates with these crosslike structures are often referred to as Holliday junctions, after Robin Holliday, who proposed their role in recombination in 1964. Such intermediates have been characterized by a wide range of techniques including x-ray crystallography. Specific enzymes, termed Tecombinases, bind to these structures and resolve them into separated DNA duplexes. The ere recombinase from bacteriophage Pi has been extensively studied. The mechanism begins with the recombinase binding to the DNA substrates (Figure 28.49 ).
Figure 28.49 Recombination mechanism. Recombination begins as two DN A molecu les come t ogether to f orm a reco m bination synapse. One strand fro m each duplex is cleaved by the recombi nase enzyme; the 3' end o f each o f t he cleaved strands is linked to a t yrOSine (Y) res idue o n th e recomb inase enzyme. New phosphodiest er bond s are f o rm ed when a S' end of th e o ther cl eaved strand in t he comp lex attacks these tyrosi ne- DN A adducts. Aft er isom erizat ion, the se steps are repeated to form t he recombined p ro duct s.
Bond formation
Cleavage '>
Holliday junction
Recombinati on synapse
Isomerization
Bond
formation
Cleavage
(
. • ..
--
• .. 32P-labeled fragments
Gel-electrop ho resis pattern
32 P-labeled fragments
A striking pattern is evident when the sequences of many prokaryoti c promoters are compared. Two common motifs are present on the 5' (upstream) side of the transcription start site. They are known as the - 10 sequence and the - 35 sequence because they are centered at about 10 and 35 nucleotides upstream of the start site. The region containing these sequences, which spans as many as 40 nucleotides upstream from the start site, is called the core promoter. The -10 and - 35 sequences are each 6 bp long. Their consensus (average) sequences, deduced from analyses of many promoters (Figure 29.4), are - 35
- 10
+1
5' -T T G A C A .. ...................... T A T A A T .. " ...... Start site
The first nucleotide (the start site) of a transcribed DNA sequence is denoted as + 1 and the second one as + 2; the nucleotide preceding the start site is denoted as -1. These designations refer to the coding strand of DNA. Recall that the sequence of the template strand of DNA is the complement of that of the RNA transcript (see Figure 4.26). Tn contrast, the coding strand of DNA has the same sequence as that of the RNA transcript except for thymine (T) in place of uracil (U) . The coding strand is also known as the sense ( + ) strand, and the template strand as the antisense (- ) strand. Promoters differ markedly in their efficacy. Genes with strong promoters are transcribed freq uently as often as every 2 seconds in E. coli . In contrast, genes with very weak promoters are transcribed abou t once in 10 minutes . The -10 and -35 regions of most strong promoters have seq uences that correspond closely to the consensus sequences, whereas weak promoters tend to have multiple substitution s at these sites. Indeed, mutation of a single base in either the - 10 sequence or the -3 5 sequence can diminish promoter activity. The distance between these conserved sequences also is important; a separation of 17 nucleotides is optimal. Thus, the efficiency or strength of a promoter sequence serves to regulate transcription. Regulatory proteins that bind to specific sequences near promoter sites and interact with RNA polymerase (Chapter 31 ) also markedly influence the frequency of transcription of many genes. Outside the core promoter in a subset of highly expressed genes is the upstream element (also called the UP element for upstream element) . This sequence is present from 40 to 60 nucleotides upstream of the transcription start site. T he UP element is bound by the Ci subunit of RNA polymerase and serves to increase the efficiency of transcription by creating an additional binding site for the polymerase.
Transcription starts here
1 5'
- 10
3'
(A) (6) (C)
GT TAACTA GTACGC A
(D)
GT GATACT GAGCAC A
(E)
GT TTTCAT GCCTCC A
CG TATGTT GTGTGG A GC TATGGT TATTTC A
TATAAT Figure 29.4 Prokaryotic promoter sequences. A compari son of five sequences from prokaryo t ic promoters reveal s a recurring sequence o f TATAAT centered on positio n - 10. The - 10 co nsensus sequence (in red) wa s deduced from a large number of pro moter sequences. The sequences are fro m the (A) lac. (B) gal , and (C) trp o perons of E. coli; fro m (D) " phage; and fro m (E) (~X174 phage.
Sigma Subunits of RNA Polymerase Recognize Promoter Sites
To initiate transcription, the Ci21313' core of RNA polymerase must bind the promoter. However, it is the (J' subunit that makes this binding possible by enabling RNA polymerase to recognize promoter sites. In the presence of the (J subunit, the RNA polymerase binds weakly to the DNA and slides along the doubl e helix until it dissociates or enco unters a promoter. The (J' su bunit recognizes the promoter through several interactions with the nucleotide bases of the promoter DNA. Although each interaction by itself is weak, the combined effect is a strong sequence -specific interaction overall. A recent crystal structure of the RNA polymerase holoenzyme bound to a promoter site shows the (]' su bunit interacting with DNA at the - 10 and - 35 regions essential to promoter recognition. (Figure 29.5). Therefore, the (J' subunit is responsible for the specific binding of the RNA polymerase to a promoter site on the temp late DNA. 825
826 CHAPTER 29 RNA Synthesis and Processing
-35 element -40
-20
-30
-10 element
5' .:;.. 3'
cr
Figure 29.5 RNA polymerase holoenzyme complex. Notice that the (J' subunit (o range) of t he bact eri al RNA poly merase ho loenzy me makes sequence-specific contacts wi th t he -10 and - 35 pro mo ter sequences (yell o w). [From K. S. Murakami. S. Masuda. E. A. Campbell. 0. Muzzin. and S. A. Darst. Science 296(20.0.2):1285- 1290..)
A s the holoenzyme moves along the double helix in search of a pro· moter, it forms transient hydrogen bonds with exposed hydrogen-bond donor and acceptor groups on the base pairs . The search is rapid because RNA polymerase slides along D A instead of repeatedly binding and dis· sociating from it. In other words, the promoter site is encountered by a random walk in one dimension rather than in three dimensions. The observed rate can· stant for the binding of the RNA polymerase holoenzyme to promoter se· 10 quences is 10 M 1S - 1, more than 100 times larger than that expected for repeated encounters moving on and off the DNA. The (J' subunit is released when the nascent RNA chain reaches 9 or 10 nucleotides in length . f.fter its release, it can assist initiation by another core enzyme. Thus, the (J' subunil acts catalytically . E. coli contains multiple (J' factors to recognize several types of promoter sequences contained in E. coli DNA. The type that recognizes the consensus sequences described earlier is called (1 70 because it has a mass of 70 kd. A di fferent (1 factor comes into play when the temperature is raised abruptly. E. coli responds by synthesizing (J'~2 , which recognizes the pro· moters of heat-shock genes. These promoters exhibit - 10 sequences that are somewhat different from the - 10 sequence for standard promoters (Figure 29.6). The increased transcription of heat-shock genes leads to the coordinated synthesis of a series of protecti ve proteins. Other (J' factors respond to environmental conditions, such as nitrogen starvation. These findings demonstrate that (J' plays a key role in determining where RNA polymerase ini· tiates transcription.
Figure 29.6 Alternative promoter sequences. A compari son o f the consensus sequences o f standard. heat -shock. and nitrogen-starvatio n pro moters o f E. coli. These pro moters are recognized by (J'70 . (J'3' . and (J'' ' . respectivel y.
- 35
- 10
5··",,· T TGACA ................ TA TAA T .......... 3 ' 5'" . TN N C N C N C T T G A A ................ ·C C CAT NT'" """"' 3' 5 ..... C TGGGNA
......
TTGCA
3'
Standard promoter Heat-shock promoter Nitrogen-starvation promoter
RNA Polymerase Must Unwind the Template Double Helix for
827
Transcription to Take Place
29.1 RNA Polymerase
Although RNA polymerase can search for promoter sites when bound to double- helical DNA, a segment of the helix must be unwound before synthesis can begin. A region of duplex DNA must be unpaired so that nucleotides on one of its strands become accessible for base-pairing with in coming ribonucleoside triphosphates. The DNA template strand selects the correct ribonucleoside triphosphate by forming a Watson- Crick base pair with it (p. 11 2), as in DNA synthesis . How much of the template DNA is unwound by the polymerase ? Because unwinding increases the negative supercoiling of the DNA (p. 789), this question was answered by analyzing the supercoiling of a circular duplex DNA exposed to varying amounts of RNA polymerase. Topoisom erase I , an enzyme catalyzing the concerted cleavage and resealing of duplex DNA (p . 790), was then added to relax the part of circular DNA not in contact with polymerase molecules. These DNA sampl es were analyzed by gel electrophoresis after the removal of bound protein. The degree of negative supercoiling increased in proportion to the number of RNA polymerase molecules bound per template DNA, showing that the enzyme unwinds DNA Each bound polymerase molecule unwinds a 17-bp segment of DNA, which corresponds to 1.6 turns of B-DNA helix (Figure 29.7). Negative supercoiling of circular DNA favors the transcription of genes because it facilitates unwinding. Thus, the introduction of negative supercoils into DNA by topoisomerase II can increase the efficiency of promoters located at di stant sites. However, not all promoter sites are stimulated by negative supercoiling. The promoter site for topoi somerase II itself is a noteworthy exception. Negative supercoil in g decreases the rate of transcription of this gene, an elegant feedback control ensuring that DNA d oes not become excessively supercoiled . Negative supercoiling could decrease the efficiency of this promoter by changin g the structural relation of the - 10 and - 3 5 regions. The transition from the closed promoter complex (in which DNA is dou ble helical) to the open promoter complex (in which a DNA segment is un wound ) is an essential event in transcription. The stage is now set for the formation of the first phosphodiester bond of the new RNA chain.
RNA Chains Are Formed de Novo and Grow in the 5' -to-3' Direction
In contrast with DNA synthesis, RNA synthesis can start de novo, without the requirement for a primer. Most newly synthesized RNA chains carry a highly distinctive tag on the 5' end : the first base at that end is either pppG or pppA. Tag on the 5 ' end 2-
0.. .0.
\/
. /p",-- / 0. 0.
(A or G) o. ~
--
0.. .0.
\/
0. .0.
p",-- / 0.
\/
p",--
/
0.
Base
0.
---.J-.--./ O, H
o.H
The presence of the triphosphate moiety suggests that RNA synthesis starts at the 5' end. T he results of labeling experiments with 1'_ 32p substrates confirmed that RNA chai ns, like DNA chains, grow in the 5' ) 3' direction.
Double-helical DNA
Unwound DNA ~( 17 bp opened)
RNA polymerase Figure 29.7 DNA unwinding. RNA polymerase unwinds about 17 base pa irs of template DNA.
x
828
x
CHAPTER 29 RNA Synthesis and Processing
YTP
P
P
pp.,
\/,
OH
x
y ZTP
P)-(P
Z
PP,
\ / ,
p
y
P
OH l
P 5'
5' --> 3' growth
Elongation Takes Place at Transcription Bubbles That Move Along t he DNA Template The elongation phase of RNA synthesis begins after the formation of the first phosphodiester bond . An important change is the loss of u ; without u, the core enzyme binds more strongly to the DNA template . Indeed, RNA polymerase stays bound to its template until a termination signal is reached. The region containing RNA polymerase, DNA, and nascent RNA is called a transcription bubble because it contains a locally melted "bubble" of DNA (Figure 29 .8) . The newly synthesized RNA forms a hybrid helix with the template DNA strand. This RNA- DNA helix is about R bp long, which corresponds to nearly one turn of a double helix (p. 112). The 3' -hydroxyl group of the RNA in this hybrid helix is positioned so that it can attack the a-phosphorus atom of an incoming ribonucleoside triphosphate. The core enzyme also contains a binding site for the other DNA strand. About 17 bp of DNA are unwound throughout the elongation phase, as in the initiation phase. The transcription bubble moves a distance of 170 A (17 nm) in a second, which corresponds to a rate of elongation of about 50 nucleotides per second. Although rapid, it is much slower than the rate of DNA synthesis, which is ROO nucleotides per second . /
RNA polymerase
Coding strand
Template strand
Rewin ding
Unwinding
5'
3' ......
3':"LL..J.J."
5'.J.Nascent
RNA
5' ppp . . /
(A)
Figure 29.8 Transcription bubble. (A) A schemati c representat io n of a transcri ption bubble in the elongation of an RNA transcript. Duplex DNA is unwound at the forward end of RNA polymerase and rewound at its rea r end. The RNA - DNA hybrid rotates during elongation. (B) A surface model based on the crystal stru cture o f t he RNA po lymerase holoenzyme shows the unwound DNA (yellow and green) fo rming the transcription bubble. Notice that the template strand (green) is in contact w ith the catalytic Mg2+ (pink). [(B) From K. S. Murakami, S. Masuda, E. A. Campbell, O. Muzzin, and S. A. Darst.
Science 296(2002):1285-1290.]
RNA-DNA h-YbLr.J.. idY helix
Movement of polymerase
" " Elongation site
)
Nontem plate ;...--- stra nd
(B)
l'
The lengths of the RNA- DNA hybrid and of the unwound region of DNA stay rather constant as RNA polymerase moves along the D NA template. This finding indicates that DNA is rewound at about the same rate at the rear of RNA polymerase as it is unwound at the front of the enzyme. The RNA- DNA hybrid must also rotate each time a nucleotide is added so that the 3' -OH end of the RNA stays at the catalytic site. The length of the RNA- DNA hybrid is determined by a structure within the enzyme that forces the RNA- DNA hybrid to separate, allowing the RNA chain to exit from the enzyme and the DNA chain to rejoin its DNA partner (Figure 29.9). For many years, RNA pol ymerase was thought not to proofread the RNA transcript . However, recent studies have indicated that RNA polymerases do show proofreading nuclease acti vity, particular! y in the presence of accessory proteins. Studies of single molecules of RNA polymerase reveal that the enzymes hesitate and backtrack to correct errors. The error rate of 4 the ord er of one mistake per 10 or lOS nucleotides is higher than that for DNA replication , including all error-correcting mechanisms. The lower fi de ity of R NA synth esis can be tolerated because mistakes are not transmit ted to progeny. For most genes, many RNA transcripts are synthesized; a few defective transcripts are unlikely to be harmful. Sequences Within the Newly Transcribed RNA Signal Term ination
The termination of transcription is as precisely controlled as its initiation . In the termination phase of transcription, the formation of phosphodiester bonds ceases, the RNA- DNA hybrid dissociates, the melted region of DNA rewinds, and RNA polymerase releases the DNA. What determines where transcription is terminated ? The transcribed regions of DNA templates contain stop signals. The simplest one is a palindromic GC-rich region fol lowed by an AT-rich region . The RNA transcript of this DNA palindrome is self-complementary (Fi gure 29 .10). H ence, its bases can pair to form a hairpin structure with a stem and loop, a structure favored by its high content of G and C residues. Guanine cytosine base pairs are more stable than adenine- thymine pairs because of the extra hydrogen bond in the base pair. This stable hairpin is followed by a sequence of four or more uracil residu es, which also are crucial for termination. The RN A transcript ends within or just after them. How does this combination hairpin-oligo(U) structure terminate transcription? First, it seem s likely that RNA polymerase pauses immediately after it has synthesized a stretch of RNA that folds into a hairpin . Furthermore, the RNA- DNA hybrid helix produced after the hairpin is unstable because its r U- dA base pairs are th e weakest of the four kinds. Figure 29.10 Termination signal. A termination signal found at the 3' end of an mRNA transcri pt consists of a series of bases that form a stable stem-loop structure and a series of U residues.
• DNA
RNA
~ Figure 29.9 RNA- DNA hybrid
separation. A structure w ithin RNA polymerase forces the separation of the RNA- DNA hybrid. Notice that the DNA st rand exits in one direct ion and the RNA product exits in another. [Drawn from 116H.pdb.]
U ~ C ""G I
I
U
/G
'G . C I I
A· U
I I e·G I I e·G I I G· C I I e·G I I e·G I I G· C
- U- A-A - U- C- C- C- A- C- A/ 5'
\.A-1luE3!u~-:Ju[3u~OH 3'
829
830
Initiation
CHAPTER 29 RNA Synthesis and Processing
Termination in absence
or p
DNA template - - ' --,...,...,...,... - - - - - ' ' - - - - p
sites
5'
RNA transcripts Figure 29.11 Effect of p protein on the size of RNA transcripts.
s· s· 5'
3'
r
3'
No p
> 235 species
present at start of synthesis ~ added 30 seconds later
3'
3'
padded 2 minutes later
> 105 species > 135 species > 175 species
Hence, the pau se in transcription caused by the hairpin permits the weakly bound nascent RNA to dissociate from the DNA template and then from the enzyme. The sol itary DNA template strand rejoins its partner to re-form the DNA duplex, and the tran scription bubble closes. The rho Protein Helps to Terminate the Transcription of Some Genes
RNA polym erase needs no help to terminate transcription at a hairpin followed by several U residues . At other sites, however, termination requires the participation of an additional factor . This disco very was prompted by the observation that some RNA molecules synthesized in vitro by RNA polymerase acting alone are longer than those made in vivo. The missing factor, a protein that caused the correct termination, was isolated and named rho (p). Additional information about th e action of p was obtained by adding this termination factor to an incubation mixture at various times after the initiation of RNA synthesis (Figure 29. 11 ). RNAs with sedimentation coefficients of lOS, 13S, and 1 7S were obtained when p was added at initiation, a few seconds after initiation, and 2 minutes after initiation, respectively_ If no p was added, transcription yielded a 23S RNA product. It is evident that the template co ntains at least three termination sites that respond to p (yielding 10S, 13S, and 17S RNA) and one termination site that does not (yielding 23S RNA ). Thus, specific termination at a site producing 23S RNA can occur in the absen ce of p. However, p detects additlOnal termination signals that are not recognized by RNA polymerase alone. How does p provoke the termination of RNA synthesis? A key clue is the finding that p hydrolyzes ATP in the
presence of single-stranded RNA but not in the presence of DNA or duplex RNA. Hexameric p, which is structurally similar and homologous to ATP synthase (p . 522) , specificall y binds single-stranded RNA; a stretch of 72 nuc1eotides is bound in such a way that the RNA passes through the center of the structure (Figure 29. 12). The p
I
I "'" RNA polymerase
protein is brought into action by sequences located in the nascent RNA that are rich in cytosine and poor in guanine. The
Figure 29.12 Mechanism for the termination of transcription by p protein. This protein is an ATP-dependent helicase that binds the nascent RNA chain and pulls it away fro m RNA po lymerase and the DNA template.
ATPase activity of p enables the protein to pull the nascent RNA while pursuing I-1 711. Scott, W . G . 199R. RNA catalysis . Curro Opill. S tmet. Bio I. 8: 720- 72 6.
Problems I. Complements. Th e sequence of part of an rn RNA is j' AUGGGGAACAGCAAGAGUGGGGCCCUGUCCAAGGAG ·3' Whe.t is the sequence of the DNA codin g strand? Of the DNA template strand ?
(1) DNA Coding strand ACGG '\ G AC CG s'·GGA TACT T ACAGCCA TG A AT ACT CCA T T... 3 3'·CC T A TGAA TG TCGG T ACC TG TGCCGC T T A TGAGGT AA ... S
/ (2) Template strand
2. Checking Jar errors. Why is RNA synthesis not as carefull y monitored for errors as is DNA synthesis ?
3. Speed is not of the essence. Why is it advantageous for DNA synthesis to be more rapid than RNA synthesis?
4. Potent inhibitor. H eparin inhibits transcription by binding to RNA polymerase. What properties of heparin allow it to bind so effectively to R N A polymerase? 5. A loose cannon. Sigma protein by itself does not bind to promoter sites. Predict the effect of a mutation enabl ing ( f to bind to the -10 region in the absence o f other suhunits of f{NA polymerase.
6. Stuck sigma. What wo uld be the likely effect of a mutation that wo uld prevent merase core?
IF
from di RSociatin g from the RI'\A poly .
i. Transcription time. What is the minimum len gth of time required for t he synthesis b y E. coli polymerase of an mRNA encod ing a 100· kd protein '
8. Iletween bubbles. How far apart are transcription bubbles on E. coli gen es that are being transcribed at a maximal rate? 9. A revealing bubble. Consider t he syntheti c R NA- DNA t ran · scription bubble illu strated h ere. Le t u s refer to the cod ing DNA strand, the te mplate strand , and the RNA strand as strand s 1, 2, and 3, respectivel y.
• . · . . .
·• .• .•
.• .• . .
..• ..• ..• ..• ..• ..•
UUGGACACGGCGAA UU UU
s" UU (3) RNA strand
(a) Suppose that strand 3 is labeled with 32p a t its 5' end and that pol yacrylamide gel eiectrophore,is i, carri ed o u t unde r non · d enaturin g conditions . Pred ict the autoradiograph ic pattern for (i) strand 3 alone, (ii) strands 1 and 3, (iii ) strands 2 a nd 3, (iv) strands 1, 2, and 3 , and (v ) strands 1, 2, a n d 3 and core RNA polymerase. (b ) What is the likely effect of rifampicin on Rl\A synthesis in this system ' (c) Heparin blocks elongation of the RNA primer if it is added to core RNA polymerase befo re the o nset of transcription but not if added after transcription starts. Account for thi s difference. (d ) Suppose that synthesis is carried out in the presen ce of ATP, CTP, and UTP. Compare the le n gth of the longest prod · uct obtained with that expected when all four ribonucleoside triphosphates are present.
10. Abortive cycling. Di · and trinucleotides are occasio nally reo leased from RNA polymerase at the very start of t ranscriptio n , a process called abortive cycling. This process req uires the restart of transcription . Suggest a plausible explanation for abortive cycling.
11. Polymerase inhibition. Cordycepin inhibits poly(A) synthesis at low con centrations and RNA synthesis at higher concentrations.
856
CHAPTER 29 RNA Synthesis and Processi ng
16 . Separation technique. Suggest a means by which you could separate mRNA from the other types of RNA in a eukaryotic
NH2 N
:7 N
0
H
H
j
cell.
~N
~
Data Interpretation Problems
N-
OH
Cordycepin (3'-deoxyadenosine)
(a) What is the basis of inhibition by cordycepin ? (b) Why is poly(A) synthesis more sensitive to the presence of cordycepin? (c) Does cordycepin need to be modified to exert its effect?
17. R un-off experiment. N uclei were isolated from brain , liver, and muscle. The nuclei were then incubated with a -[12P1UTP under conditions that allow RNA synthesis, except that an inhibitor of RNA initiation was present. The radioactive RK!\ was isolated and annealed to various DNA sequences that had been attached to a gene chip. In the adjoining graphs, the intensity of the shading indicates ro ughly how much mRNA was attached to each RNA sequence.
12. An extra piece. In one type of mutation leading to a form of thalassemia, the mutation of a single base (G to A) generates a new 3' splice site (blue in the illu stration below) ak in to the normal one (yellow ) bu t farther upstream.
Liver
Muscle
Brain
Normal 3 ' end
/
of intron
5' CCTATT G GTCT ATTTTCC ACCC TTAG GC T GCTG 3'
1
5' CC TA TA
TCTATTTTCCACCCTTAG G CTG C TG 3'
What is the amino acid sequence of the extra segment of protein synthesized in a thalassemic patient having a mutation lead ing to aberrant splicing? The reading frame after the splice site begins with T C T. 13 . A long-tailed messenger. Another thalassemic patient had a mutation leading to the production of an mRNA for the I:l chain of hemoglobin that was 900 nucleotides longer than the normal one. The poly(A ) tail of thi s mutant mRNA was located a few nucleotides after the onl y AAUAAA sequence in the additional sequence. Propose a mutation that would lead to the prod uction of thi s altered m RNA.
(a) Why does the intensity of hybrid ization differ between genes?
(b ) What is the significan ce of the fact that some of the RKA molecules display different hybridization patterns in different tissues) (c) Som e genes are expressed in all three tissues. What would you guess is the nature of these genes? (d) Suggest a reason why an initiation inhibitor was included in the reaction Inixture.
1R. Christmas trees. The adjoining autoradiograph depicts several bacterial genes undergoing transcription . Identify the DNA. What are the strands of increasing length? Where is the beginning of transcription' The end of transcription' On the page, what is the direction of RNA synthesis? What can you conclude about the number of enzym es participating in RNA synthesis on a given gene?
Mechanism Problem
14 . R NA editing. Many uridine molecules are inserted into some mitochondria l mRNAs in trypanosomes. The uridine residu es com e from the poly(U ) tail of a donor strand. Nucleoside triphosphates do not participate in this reaction . Propose a reaction mechani sm that accounts for these findings. (Hint: Relate RNA editing to RNA splicing. ) Chapter Integration Problems
15 . Proteome complexity. What processes considered in this chapter make the proteome more complex than the genome? What processes might further enhance this complexity?
•
Chapter
Protein Synthesis Polypeptide 505
305
3'
The ribosome, shown at t he right. is a fa cto ry for the manufacture of polypeptides. Amino acids are carried into t he ribosome. one at a ti me. connected to transfe r RNA molecu les. Each amino acid is joi ned to the grow ing polypeptide cha in. which detaches from t he ribosome only after the polypeptide has been completed. This assembly-li ne approach allows even very long polypeptide cha ins to be assembled rapidly and with impressive accuracy. [(Left) Doug Martin/ Photo Researchers.]
en etic information is m ost important becau se of the proteins that it en cod es, in that proteins p lay m ost of the functiona l roles in cells. In C hapters 28 and 29, we examined h ow DNA is replicated and how DNA is transcribed into RN A . We now turn to the m ech anism of protein synthesis. a process called translatiun b e cause th e fo ur -le tter alph abet o f nucl e ic acids is translated in to t h e e ntirely differ ent twenty- letter alphabet of proteins. Tra n slation is a conceptually m ore complex process than e ither replication o r transcription , both of which take place within t h e framework of a comm o n base -pairing langu age . As befits its positio n linking the nucleic acid and protein lan guages, the process of protein synthesis d ep end s criti cally on both nucleic acid and protein factors . Protein synthesis takes place on ribosomes e n ormou s complexes containing three large RNA m olecules and m or e than 50 proteins . Among the great tri u mphs in b iochemistry in r ecent years h as been the determi n ation of t he structure of the ribosome a n d its components so that its fu nction can b e examined in atomic deta il. Perhaps the most significant con cl usion fro m th ese studies is that the ribosome is a ribozyme; th at is, the RNA co mponents play the m ost fundamental roles . These ob servations strongly su pport the n otion th at life evolved throug h an RNA world, an d the ribosom e is a surviv ing inhabitant of that world .
I O utlin e l 30.1 Protein Synthesis Requires the Translation of Nucl eotide Sequences into Amino Acid Sequences 30.2 Aminoacyl-Transfer RNA Synthetases Read the Genetic Code 30.3 A Ribosome Is a Ribonucleoprotein Particle (70S) Made of a Small (305) and a Large (50S) Subunit 30.4 Protein Factors Play Key Roles in Protein Synthesis 30.S Eu karyotic Protei n Synthesis Di ffers from Prokaryotic Prote in Synthesis Primarily in Translat ion Initiation 30.6 Ribosomes Bound to the Endoplasmic Reticulum Manufacture Secretory and Membra ne Prote ins 30.7 A Variety of Antib iotics and Toxins Can Inh ibit Protein SyntheS is
857
858 CHAPTER 30 Protein Synthesis
Transfer RNA molecules (tRNAs ), messenger RNA (mRNA) and many proteins participate in protein synthesis along with ribosomes. The link between amino acids and nucleic acids is first made by enzymes called aminoacyl-tRNA synthetases. By specificall y linking a particular amino acid to each tRNA, these enzymes translate the genetic code. This chapter focuses primarily on protein synthesis in prokaryotes because it illustrates many general principles and is well understood. Some distinctive features of protein synthesis in eukaryotes also are presented.
30.1
Protein Synthesis Requires the Translation of Nucleotide Sequences into Amino Acid Sequences
The basics of protein synthesis are the same across all kingdoms of lifeevidence that the protein-synthesis system arose very early in evolution. A protein is synthesized in the amino-to-carboxyl direction by the sequential addition of amino acids to the carboxyl end of the growing peptide chain (Figure 30.1). The amino acids arrive at the growing chain in activated form as aminoacyl -tRNAs, created by joining the carboxyl group of an amino acid to the 3' end of a transfer RN A molecule. Thelinking of an amino acid to its corresponding tRNA is catalyzed by an aminoacyl-tRNA synthetase. ATP cleavage drives this activation reaction. For each amino acid, there is usually one activating enzyme and at least one kind of tRNA.
NH, +
. '"
R, ..
H
0
C Hi R,
1,,_, H
H
Ol==
GUG > UUG). In approximately one-half of E. coli proteins, N-formylmethionine is removed when the nascent chain exits the ribosome. Methionine is linked to these two kinds of tRNAs by the same aminoacyltRNA synthetase. A specific enzyme then formylates the amino grou p of methionine attached to tRNA r (Figure 30 .1 7). The activated formyl donor 10 in this reaction is N -formyltetrahydrofolate (p . 689). It is significant that free methionine and methionyl-tRNA m are not substrates for this trans formy lase.
H
0 -=
o I
tRNAf Methionyl-tRNA, (Met-tRNA,)
N1 0-Formyltetrahydrofo late Transformylase
Tetrahydrofolate
~ NH
H ""
0:
H
o I
Ribosomes Have Three tRNA-Binding Sites That Bridge the 30S and 50S Subunits
tRNAf
A snapshot of a significant moment in protein synthesis was obtained by determining the structure of the 70S ribosome bound to three t RNA molecules and a fragment of mRNA (Figure 30. 18). As expected, t he mRNA fragment is bound within the 30S subunit. Each of the tKNA molecules bridges between the 30S and the 50S subunits, At the 30S end, two of the three tRNA molecules are bou nd to the mRNA fragment through
Formylmethionyl-tRNA, (IMet-tRNA,)
Figure 30.17 Formylation of methionyltRNA. Initi at o r tRNA (tRNA,) is first charged with methionine, and then a formyl group is transferred to the methiony l-tRNA r from N 'oformyltetrahydrofolate.
(B)
(A)
mRNA E site
P site
A site
E site
~ Figure 30.18 Transfer RNA-binding sites. (A) Three tRNA-binding sites are present
on t he 70S ribosome. They are ca ll ed the A (for am inoacyl), P (for peptidyl), and E (for exit) sites, Each tRNA molecule contacts both the 305 and the 50S subunit. (B)The tRNA molecules in si tes A and P are base-paired w ith mRNA. [(8) Drawn from 1JGPpdb.]
P site
A site
872 CHAPTER 30
anticodon- codon base pairs. These binding sites are called the A site (for amino acyl) and the P site (for peptidyJ ). The third tRNA molecule is bound to an adjacent site called the E site (for exit). The other end of each tR A molecul e interacts with the 50S subunit. The acceptor stems of the tRNA molecul es occupying the A site and the P site converge at a site where a peptide bond is formed. Further examination of th is site reveals that a tunnel connects this site to the back of the ribosome. The growing polypeptide chain escapes the ribosome through this tunnel during synthesis.
Protein Synthesis
The Growing Polypeptide Chain Is Transferred Between tRNAs on Peptide- Bond Formation
Protein synthesis begins with the interaction of the 30S subunit and mRNA through th e Shin e- Dalgarno sequence. On formation of this compl ex, the initiator tR A charged with fonnylmethionine binds to the initiator A G codon, and the 50S subunit binds to the 30S subunit to form the complete 70S ribosome. H ow does the polypeptide chain increase in length (Figure 30 .1 9)? The three sites in our snapshot of protein synthesis provide a clue. The initiator tRNA is bound in the P site on the ribosome. A charged tRNA with an anticodon complementary to the codon in th e A site then binds. The stage is set for the formation of a peptide bond: th e formyl methionine molecul e linked , ) the initiator tRNA will be transferred to th e amino group of the amino acid in the A site. The formation of the peptide bond, one of the most important reactions in life, is a thermodynamically spontan eous reaction that is catalyzed by a site on the 23S rRNA called the peptidyl transferase center. The amino group of the aminoacyl-tRNA in the A site is well positioned to attack the ester linkage between the initiator tRNA and the formylmethionine molecule (Figure 30.20). T he peptidyl tran sferase center includes bases that promote this reaction by helping to form an NH2 group on the A-site aminoacyl-tRNA and by helping to stabilize the tetrahedral intermediate
50S
o o
E"",
305
>
Am inoacyl-tRNA
Peptide-bond
binding
formation
.---- GTP Translocation
Elongation factor G
GOP + Pi
Figure 30.19 Mechanism of protein synthesis. The cycle begins with peptidyl tRNA in the P site. An aminoacyl-tRNA binds in the A site. With both sites occupied, a new peptide bond is f o rmed. The t RNAs and the mRNA are translocated through the action of elongation factor G, which moves the deacy lated tRNA to t he E site. O nce there. the tRNA is free to dissociate to complet e t he cycle.
o tRNA
dissociation
\NH
873
\NH
\NH
30.3 The Ribosome
R; .
0
0 HN
HN
HN
Ri+ I '"
0v 0 I
0
tRNA (P site)
-(0
W
H
./
\
~N Ri+2
,
O~
H
? tRNA (A site)
'..
•
n 0
Ri+ 1
H
"---R I+2
tRNA
•
H
0 -
'.
W
NH
I
Ri+ l
0
•
OH +
NH
I
RI+2
tRNA
""
O~
o
o
tRNA Tetrahedral
tRNA
I
H
H
I
intermediate
Figure 30.20 Peptide-bond formation. The amino group of the aminoacyl-tRNA attacks the carbonyl group of the ester linkage of the peptidyl-t RNA to form a tetrahedral intermediate. This intermediate collapses to form the peptide bond and relea se the deacylated tRNA . •
that forms . This reaction is, in many ways, analogous to the reverse of the reaction catalyzed by serine proteases such as chymotrypsin (p . 247). The peptidyl-tRNA is analogous to the acy l-enzyme form of a serine protease. In a serine protease, the acyl-enzyme is generated with the use of the free energy associated with cleaving an amide bond. In the ribosome, the free energy necessary to form the analogous species, an aminoacyl-tRNA, comes from the AT P that is cleaved by the aminoacyl-tRNA synthetase before the arrival of the tRNA at the ribosome. With the peptide bond formed, the peptide chain is now attached to the tRNA in the A site on the 30S subunit while a chan ge in the in teraction with the 50S subunit has placed that tRNA and its peptide in the P site of the large subunit. The tRNA in th e P site of the 30S subunit is now uncharged . For translation to proceed, the mRN A must be moved (or translocated) so that the codon for the next amino acid to be added is in the A site. This translocation is assisted by a protein enzyme called elongation factor G (p. 876), driven by the hydrolysis of GTP. On completion of this step, the peptidyl-tRNA is now fully in the P site, and the uncharged initiator tRNA is in the E site and has been disengaged from the mRNA. O n dissociation of the initiator tRNA , the ribosome has returned to its initial state except that the peptide chain is attached to a different tRNA, the one corresponding to the first codon past the initiating AUG. Note that the peptide chain remains in the P site on the 50S subunit, at the entrance to the exit channel , throughout this cycle, presumab ly growi ng into the tunneL T his cycle is repeated as new am inoacyl-tRNAs move into the A site, allowing the polypeptide to be elongated until the cycle is terminated ,
Only the Codon-Anticodon Interactions Determine the Amino Ac id That Is Incorporated
On the basis of the mechanism described on page 871 , the base- pairing interaction between th e anticodon on the incoming tRNA and the codon in the A site on m RNA determines which amino acid is added to the polypep tide chain. Does the amino acid attached to the tRNA play any role in this process? This question was answered in the following way. First, cysteine was attached to its cognate tRNA. The attached cystein e unit was then converted into alanine by removing the sulfor atom from the side chain in cysteine in a reaction catalyzed by Ran ey nickel; the reaction removed
874 CHAPTER 30 Protein Synthesis
Cysteine
the sulfur atom from the cysteine resid ue without affecting its linkage to t R A. T hus, a mischarged aminoacyl-tR NA was p roduced in which alanine was covalen tly attached to a tR NA specif ic for cysteine. H2 0
2 P,
+
+
AlP
AMP
+ tR NAcys -----"'~"-Cysteinyl-tRNA synthetase
H
o
> +H,
C " tRNA "
o Cy.-IHNAc"
o Ala-IRNA c"
Does this m ischarged tR NA recogni ze t he codon for cysteine or for alanin e? The answer came when the tR A was added to a cell -free proteinsy nthesizing system . T he template was a random copolym er of U and G in the ratio of 5:1, which normally incorporates cysteine (encoded by UG U) but not alanine (encoded by GCN ). However, alanine was incorporated into a polypeptide when Ala-tRNA C y. was added to the incu bation mixture. T he same result was obtained when mRNA fo r hemoglobin served as the templ ate and C4C]alanyl-tH. N A Cys was used as the mischarged aminoacyltR JA. When the hemoglobin was di gested wi th try psin, the only radioactive peptide prod uced was one that norm ally contain ed cysteine but not alanine. Thus, the amino acid in aminoacy l - t~ NA does not playa role in selecting a codon. In recent years, the ability of mischarged tRNAs to transfer their amino acid cargo to a growing polypeptide chain has been used to synthesize peptides with amino acids not found in proteins incorporated into specific sites in a p rotein . Aminoacyl- tRNAs are first linked to these unnatural amino acids by chemical methods. T hese mischarged aminoacyl- tRNAs are added to a cell -free protein-synthesizin g system along with speciall y engineered mRNA that contains codons correspond in g to the an ticodons of the mischarged aminoacyl-tRN As in the d esired positions. The proteins produced have unnatural amino acids in the expected positions. M ore than 100 differen t unnatural amino acids have been incorporated in this way. However, only L-am ino acids can be used; apparentl y this stereochem istry is requi red fo r peptide -bond formation to take place. Some Transfer RNA Molecules Recognize More Than One Codon . Because of Wobble in Base-Pairing
Anticodon
3'
5'
- X'- Y'- Z/• • •
X
5'
• • •
• • •
Y- Z3'
Codon
W hat are the rules that govern the recogni tion of a codon by the anticodon of at RNA ? A simple hypothesis is that each of the bases of the codon forms a W atson-Crick type of base pair with a complementary base on the anti· codon. The cod on and anticodon would then be lined up in an antiparallel fas hi on . tn the diagram in the m argin, the prime denotes the complemen· tary base. Thus X and X' wo uld be either A and U (or U and A ) or G and C (or C and G). According to thi s model, a particular anticodon can recognize only one cod on . T he facts are otherwise. As found experimen tally, some pure tR NA mol· ecules can recognize more than one codon. For example, the yeast alanyl tR A binds to three codons: G CU, G CC, and GCA . The first two bases of these codons are the same, whereas the third is diffe rent . C oul d it be that recogniti on of the third base of a codon is sometimes less di scriminating than recogni t ion of the other two ? T he pattern of degeneracy of the genetic code indicates that this might be so. XYU and XYC always encode the same amino acid ; X YA and XYG usuall y do. F rancis C rick surmised from these
data that the steric criteria might be less stringent for pairing of the third base than for the other two. Models of various base pairs were built to determine which ones are similar to the standard A . U and G . C base pairs with regard to the distance and angle between the glycosidic bonds. Inosine was included in this study because it appeared in several anticodons. With the assumption of some steric freedom (" wobble" ) in the pairing of the third base of the codon , the combination~ shown in Table 30.3 seemed plausible. The wobble hypothesis is now firmly established. The anticodons of tRNAs of known sequence bind to the cod ons predi cted by this hypothesis. For example, the anticodon of yeast alanyl-tRNA is JGC. This tRN A recognizes the cod ons GCU, GCC, and GCA. Recall that, by convention, nucleotide sequences are written in the 5' --+ 3' direction unless otherwise noted. Hence, I (the 5' base of this anticodon) pairs with U, C, or A (the 3' base of the cod on ), as predicted .
o HN ~
/ -----N
N
\.
ribose Inosine
TABLE 30.3 Allowed pairings at the third base of the codon according to the wobble hypothesis First base of anticodon
Third base of codon G U A orG Uor C U, C. or A
C A
•
U G ribose "
ribose~
,_'" N
N
0 -"'"
o,
,, , ,
,,
•
• • ,
,
NJ
1
, ,
~
/ N~
~
H, , , ,
,
,, , ,
,, ,•
o
N
ribose/ ' Inosine-cytidine base pair
Inosine-adenosine base pair
Two generalizations concerning the codon anticodon interaction can be made:
1. The first two bases of a cod on pair in the standard way. Recognition is precise. Hence, codons that differ in either uf their first two bases must be recognized by different tR NAs. For example, both UUA and CUA encode leucine but are read by different tR As. 2. The first base of an anticodon determines whether a particular tRNA molecule reads one, two, or three kind s of codons: C or A (one codon), U or G (two cod ons), or I (three codons). Thus, part of the degeneracy of the genetic code arises from imprecision (wobble) in the pairing of the third base uf the codon with the first base of the anticodon. We see here a strong reason for the frequent appearance of inosine, one of the unusual nucleosides, in anticodons. Inosine maximizes the number of coduns that can be read by a particular tRNA molecule. The inosines in tRNA are formed by the d eaminat ion of adenosine after the synthesis of the primary transcript. Why is wobble tolerated in the third position of the codon but not in the first two? The 30S subunit has three uni versally conserved bases adenine 1492, adenine 1493, and guanine 530 in the l oS RNA that form hyd ro gen bonds on the minor- groove side bu t only with correctly formed base pairs of the cod on- anticodon duplex (Figure 30.21 ). These interactions serve to check whether W atson- Crick base pairs are present in the first two
bS RNA A 1493:..--
Armcc 101 A ~6
Codon U 1
Figure 30.21 165 rRNA monitors basepairing between the codon and the anticodon. Adenine 1493, one of three un iversally conserved bases in 165 rRNA. forms hydrogen bonds with the bases in both the codon and the anticodo n only if the codon and ant icod on are correctl y paired . [Fro m J M . O gle and V. Ramakri shnan. Annu. Re v. Biochem . 74 (2005):129- 177, Fig. 2a.]
875
30S ribosomal subunit /" Initiation factors
positions of the codon- anticodon duplex. No such inspection device is pres· ent for t he third position; so more· varied base pairs are tolerated. Th is mechanism for ensuring fidelity is analogous to the minor -groove interac· tions utilized by DNA polymerase for a similar purpose (p. 794). Thus, the ribosome plays an active role in decoding the codon- anticodon interactions.
30S ' IFHF3 IF2 (CTP)' IM et -tRNA,
+ mRNA
30.4
fM et GTP
5'
mRNA
3 05 initiation comptex
IFl + IF3
50S subunit + H2 0
IF2, CDP + P;
fMet
70S initiation complex
Figu re 30.22 Translation initiation in prokaryotes. Initiation facto rs aid the assembly f irst of th e 305 in it iation complex and th en of the 70S initiation compl ex.
Protein Factors Play Key Roles in Protein Synthesis
Although r RNA is paramount in the process of tran slation, protein factors also are required for the efficient synthesis of a protein. Protein factors par· ticipate in the initiation , elongation, and termination of protein synthesis. P-Ioop NTPases of the G-protein family play particu larly important roles. Recall that these proteins serve as molecular switches as they cycle between a GTP-bound form and a GD P -bound form (p . 387).
Formylmeth ionyl-tRNA f Is Placed in the P Site of the Ribosome in the Formation of the 70S Initiation Complex Messenger RNA and formylmethionyl-tRNA r mu st be brought to the rio bosome for protein synthesis to begin . How is this accomplished? Three protein initiation factors (IF1 , IF2, and TF3 ) are essential. The 30S riboso· mal subunit first forms a complex with 1F1 and IF3 (Figure 30.22 ). Binding of IF3 to the 30S subunit prevents it from prematurel y joining the 50S sub· unit to form a dead-end 70S complex, d evoid of m RNA and fMet·tRNA r. 1F1 binds near the A site and thereby directs the fMet -RNA r to the P site. IF2, a member of the G-protein family, binds GTP, and the concomitant conformational change enables IF2 to associate with formylmethion yltRNA r. The IF2- GTP- initiator-tRN A complex binds with mRNA (cor· rectly positioned by t he Shine- Dalgarno sequence interaction w ith the 16S rRNA) and the 30S subu nit to form the 305 initiation complex. Structural changes then lead to the ejection of TF1 and IF3. IF2 stimulates the associ· ation of the 50S subunit to the comp lex. The G TP bound to IF2 is hydrolyzed, leading to the release of I F 2. Th e result is a TUS initiation complex. When the 70S initiation complex has been form ed, the ribosome is ready for the elongation phase of protein synthesis . The fMet-tRNA r molecule occupies the P site on the ribosome. The other two sites for tRNA mole· cules, the A site and the E site, are empty. Formylmethionyl-tRNA r is po· sition ed so that its anticodon pairs with the initiating AU G (or GUG or UUG ) codon on mRNA . This interaction sets the readin g frame for the translation of t he entire mRNA. .
EF·T"
Elongation Factors Deliver Am inoacyl-tRNA to the Ribosome
Guanine
nucleotide -
The second phase of protein synthesis is the elongation cycle. This phase begins with the insertion of an aminoacyl-tRNA into the empty A site on the ribosome. The particular species in serted depend s on the mRNA codon in the A site. The cognate aminoacyl -tRNA does not simply leave the synthetase and diffu se to the A site. Rather, it is d elivered to the A site in asso· ciati on with a 43- kd protein called elongation fa ctor Tu (EF -Tu ). Elongation factor Tu, another member of the G -protein family, req uires G TP to bind aminoacyl -tRNA (Figure 30. 23) amI to bind the ribosome. The binding of Amlnoacyl· tRNA ---,.~ Figure 30.23 Structure of elongation factor Tu. The structure of a complex between
elongation fa ct or Tu (EF-Tu) and an aminoacyl-tRNA. Notice the P-Ioop NTPase domain (purple shad ing) at t he amino-termina l end of EF-Tu. Thi s NTPase domain is similar to those in other G proteins. [Draw n fro m lB23.pdb.]
876
EF -Tu to aminoacyl -tRNA serves two functions . First, EF -Tu protects the delicate ester linkage in aminoacyl-tRNA from hydrolysis . Second, the GT P in EF -Tu is hydrolyzed to GOP when an appropriate complex between the EF -Tu- aminoacyl-tRNA complex and the ribosome has formed. If the anticodon is not properly paired with the codon, hydrolysis does not take place and the aminoacyl-tRNA is not transferred to the ribosome. This mechanism allows the free energy of GTP hydrolysis to contribute to the fidelity of protein synthesis. GTP hydrolysis also releases EF -Tu from the ribosome. EF -T u in the GOP form must be reset to the G TP form to bind another aminoacyl-tRNA. Elongation factor T s, a second elongation factor, joins the EF-Tu complex and induces the dissociation of GOP. Finally, GTP binds to EF -Tu, and EF -T s is concomitantly released. It is noteworthy that EFTu does not interact with fMet-tRNAr- Hence, this initiator tRNA is not delivered to the A site. In contrast, Met-tRNA m , like all other aminoacyl tRNAs, does bind to EF -Tu . These findings account for the fact that interned AUG codons are not read by the initiator tRNA. Conversely, IF2 rec ognizes fMet-tRNA f but no other tRNA.
-y
This GTP- GDP cycle of EF -Tu is reminiscent of those of the heterotrimeric G proteins in signal transduction (p. 387) and the Ras proteins in growth control (p. 398). This similarity is due to their shared evolutionary heritage, seen in the homology of the amino -terminal domain of EF -T u to the P -loop N TPase domains in the other G proteins. The other two domains of the tripartite EF -Tu are distinctive; they mediate interactions between aminoacyl-tRNA and the ribosome. In all these related en zymes, the change in conformation between the GTP and the GOP forms leads to a change in interaction partners. A further similarity is the require ment that an additional protein catalyzes the exchange of GTP for GOP; ET- T s catalyzes the exchange for ET-Tu, just as an activated receptor does for a heterotrimeric G protein. The Formation of a Peptide Bond Is Followed Translocation of tRNAs and mRNA
by the GTP-Driven
After the correct aminoacyl-tRNA has been placed in the A site, the transfer of the polypeptid e chain from the tRNA in the P site is a thermodynamically spontaneous process, driven by the formation of the stronger peptide bond in place of the ester linkage. However, protein synthesis cannot continue without the translocation of the mRNA and the tRNAs within the ribosome. The mRNA must move by a distance of three nucleotides so that the next codon is positioned in the A site for interaction with the incoming aminoacyl -tRNA. At the same time, the deacylated tRNA moves out of the P site into the E site on the 30S subunit and the peptidyl -tRNA moves out of the A site into the P site on the 30S subunit. The movement of the peptidyltRNA into the P site shifts the mRNA by one codon, exposing the next codon to be translated in the A site. The three-dimensional structure of the ribosome undergoes significant change during translocation, and evidence suggests that translocation may result from properti es of the ribosome itself. However, protein factors accelerate the process. Translocation is enhanced by elongatiunJactor G (EF -G, also call ed translocase). A possible mechanism for accelerating the translocation process in shown in Figure 30.24. First, EF -G in the GTP form binds to the ribosome near the A site, interacting with the 23S r RNA of the 50S subunit. The binding of EF -G to the ribosome stimulates the GTPase activity of EF -G. On GTP hydrolysis, EF -G undergoes a conformational change that displaces the peptidyl-tRNA in the A site to the P site, carrying
877 30.4 Protein Factors
EF-G
p.,
o
Figure 30.24 Translocation mechanism. In the GTP form , EF-G binds t o the EF-Tu-bi nding site on the 50S subunit Thi s stimulates GTP hydrolysis, inducing a conformatio nal change in EF-G that forces the tRNA s and mRNA t o move through th e ribosome by a dist ance corresponding to one co don.
the m RNA and the deacylated tRNA with it. The dissociation ofEF -G leaves the ribosome ready to accept the next aminoacyl -tRNA into the A site.
,adenine
Protein Synthesis Is Terminated by Release Factors That Read Stop Codons
\.OH
The final phase of translation is termination . How does the synthesis of a polypeptide chain come to an end when a stop codon is encountered? Aminoacyl-tRNA does not normally bind to the A site of a ribosome if the codon is UAA, UClA, or UAG, because normal cells do not contain tRNAs with anticodons complementary to these stop signals. Instead, these stop cuduns are recognized by release factors (RF s), which are proteins that promote the release of the completed protein from the last tRNA. One of these release factors, RF1 , recognizes UAA or UAG. A second factor, RF2, recognizes UAA or UGA. A third factor, RF3, mediates interactions between RFl or RF2 and the ribosome. RF3 is another G protein homologous to EF -Ill. RF1 and RF2 are compact proteins that in eukaryotes resemble a tRNA molecule. When bound to the ribosome, the proteins unfold to bridge the gap between the stop codon on the mRNA and the peptidyl transferase center on the 50S subunit. Although the precise mechanism of release is not known, the release factor may promote, assisted by the peptidyl transferase , a water m olecule's attack on the ester linkage, freeing the polypeptide chain . The detached polypeptide leaves the ribosome. Transfer RNA and messenger RNA remain briefly attached to the 70S ribosome until the entire complex is dissociated in a GTP -dependent fashion in response to the binding of EF -G and another factor, called the ribosome re lease factor (RRF ) (Figure 30.25)
""'0
R
NH polypeptide/
tRNA \
o o ,adenine
\. HO
OH
+ H
H -'"
R
NH polypeptide/
RFl Peptide cleaved from tRNA )
UAA
UAA
UM
Figure 30.25 Termination of protein synthesis. A release factor recogn izes a stop codon in the A site and st imu lates the release o f th e comp leted p ro t ei n fro m th e tRNA in t he P site.
878
30.S
879
Eukaryotic Protein Synthesis Differs from Prokaryotic Protein Synthesis Primarily in Translation Initiation
30.5 Eukaryotic Protein Synthesis
The basic plan of protein synthesis in eukaryotes and archaea is simil ar to that in bacteria. The major structural and mechan istic themes recur in all domains of life. However, eukaryotic protein synthesis entail s more protein components than does prokaryotic protein sy nthesis, and some steps are more intricate. Som e noteworthy sim ilarities and differences are as follows : 1. I~ ibosomes . Eukaryotic ribosomes are larger. They consist of a 60S large subunit and a 40S small subunit, which come together to form an 80S particle having a mass of 4200 kd, compared with 2700 kd for the prokaryotic iOS ribosome. The 40S subunit contains an 18S RNA that is homologous to the prokaryotic 16S R NA. The 60S subunit contains three RNA s: t he 5S RNA, which is homologous to the prokaryotic 5S rRNA; the 28S RNA. which is homologo us to the prokaryotic 23S molecul es; and the 5.8S RNA, which is homologous to the 5' end of the 23 S RNA of prokaryotes. 2. Initiator tRNA. In eukaryotes, the initiating amino acid is methionine rather than N-formylmethionine. However, as in prokaryotes, a special tRNA participates in initiation. This aminoacyl-tR NA is called MettRNA i or Met-tRNA r (the subscript "i" stands for initiation , and Hf" indi cates that it can be formylated in vitro).
3. lnitiation. T he initiating codon in eukaryotes is always AUG. Eukaryotes, in contrast with prokaryotes, do not have a specific purine- ri ch sequence on the 5' side to distinguish initiator AUGs from internal ones . Instead, the AUG nearest the 5' end of mR A is usually selected as the start site. A 40S ribosome, with a bo und Met-tRNA i , attaches to the cap at the 5' end of eukaryotic mRNA (p. 846) and searches for an AUG codon by moving step-by-step in the 3' direction (Figure 30.26) . T his scanning process is catalyzed by helicases that move along the mRNA powered by ATP hydrolysis. Pairing of the anticodon of Met-tRNA i with the AUG codon of mR A signals that the target has been found . In almost all cases, eukaryotic mR A has only one start site and hence is the tem plate for a single protein . In contrast, a prokaryotic mRNA can have multiple Shine Dalgarno sequences and, hence, start sites, and it can serve as a tem plate for the synthesis of several proteins. Eukaryotes utilize many more initiation factors than do prokaryotes, and their interplay is much more intricate. The prefix el F denotes a eukaryotic initiation factor. For example, eI F -4E is a protein that binds directly to the 7·methylguanosine cap (p. 846). whereas eIF-2. in association with GTP. delivers the met-tRNA i to the ribosome. T he difference in initiation mechanism between prokaryotes and eukaryotes is, in part. a consequence of the difference in RNA processing. The 5' end of mRNA is readily available to ribosomes immediately after transcription in prokaryotes. In contrast. premRNA must be processed and transported to the cytoplasm in eukaryotes before translation is initiated. The 5' cap provides an easil y recogni zable starting point. In addition, the complexity of eukaryotic translation initia tion provides another mechanism fo r regulation of gene expression that we shall explore further in Chapter 31 . 4. The Structure of mRNA. Eukaryoti c mRNA is circular. T he elF -4E protein that binds to the mR A cap structure also binds to the poly(A ) tail through two protein intermediaries. The protein binds first to the
mRNA
Cap Initiation factors + GlP ~ MeHRNA; 405 subunit
Met
405 subunit
with initation
n AlP
components
n AD? + Pi Met
60S sub unit
Initldtion
fadors
Met
80S initiation complex
Figure 30.26 Eukaryotic translation initiation. In eukaryotes, translation initiation starts with the assembly o f a complex on the 5' cap that includes the 405 subunit and M et - tRNA,. Driven by ATP hydrol YS iS, this complex scans th e mRNA until th e first AUG is reached . The 60S subunit is then added to fo rm the 80S initiation complex.
5'
elF -4G protein , which in turn binds to a protein associated with the poly(A ) tail, the poly(A)-bindin g protein (PABPI ; Figure 30 .27). Cap and tail are thu s b ro ught together to form a circle of mRNA. The circul ar structure m ay facilitate the rebinding of the riboso mes fo ll owing protein- synth esis termination .
eIF-4 G
m
PASPI
PASPI
3'
80S
Figure 30.27 Protein interactions circularize eukaryotic mRNA. [After H. Lodish et aI., Molecular Cell Biology. 5th ed. (w. H. Freeman and Company, 2004). Fig. 4.31.]
S. Elongation and Termination. Eukaryotic elongation factors EFI ex and EFl i3'Y are the counterparts of prokary· otic EF -Tu and EF -T s. The GTP form of EFl CI. deli vers aminoacyl-tRNA to the A site of the ribosome, and EFI i3'Y catalyzes the exchan ge of GTP for bound GOP. Eukaryotic EF2 mediates GTP -driven translocation in much the sam.e way as does prokaryotic EF -G. T ermination in eukaryotes is carried out by a single release factor, eRFl , compared with two in prokaryotes . Finally, elF -3, like its prokaryotic counterpart IF3, prevents the reassociation of ribosomal subunits in the absence of an initiation compl ex .
30.6
Figure 30.28 Ribosomes are bound to the endoplasmic reticulum. In t his electron micrograph, ribosomes appear as small black dots bindi ng t o the cytoplasmic side of the endoplasmic reticulum to give a rough appearance. In contrast , the smooth endoplasmic reticulum is devoid of ribosomes. [From G. K. Voletz, M. M. Rol ls, and T. A. Rapoport, EMBO Rep. 3(2002): 944 -950.]
Ribosomes Bound to the Endoplasmic Reticulum Manufacture Secretory and Membrane Proteins
A newly synthesized protein in E. coli can stay in the cytoplasm or it can be sent to the plasma membrane, the outer m embrane, the space between them, or th e extracellular m edium . Eukaryotic cells can direct proteins to internal sites such as lysosomes, mitochondria, chloroplasts, and the nu cleus. How is sorting accomplished ? In eukaryotes, a key choice is made soon after the synthesis of a protein begins. The ultimate destination of a protein d epends broadly on the location of the ribosom e on which it is being synthesized . In eukaryotic cells, a ribosom e remains free in the cyto plasm unl ess it is directed to the endoplasmic reticulum (ER), the extensive m embrane system that comprises about half the total m embran e of a cell. The region that binds riboso m es is called the rough ER because of its studded appearance, in co ntrast with the smooth ER , which is devoid of ribosomes (Figure 30. 28). Free ribosomes synthesize proteins that remain within the cell , either within the cytoplasm or directed to organelles bounded b y a double membrane, such as the nucleus, mitochondria and chloroplasts. Ribosomes bound to the ER usually synthesize proteins d estined to leave the cell or to at least contact the cell exterior from a position in the cell membrane. These proteins fall into three m ajor classes: secretory proteins (proteins ex· ported by the cell), lysosomal proteins, and proteins spanning the plasma membrane. Virtually all integral m embrane proteins of the cell , except those located in the membranes of mitochondri a and chloropl asts, are form ed by ribosomes bound to the ER. A vari ety of strategies are used to send proteins synthesized by free ribosom es to the nucleus, peroxisomes, mitochondria, and chloroplasts of eukaryotic cells. However, in this section, we will focus on the targeting of proteins p roduced by ribosomes bound to the endoplasmic reticulum.
Signal Sequences Mark Proteins for Trans location Across t he Endoplasm ic Ret iculum Membrane The synthesis of proteins destined to leave the cell or become embedded in the plasma membrane begins on a free ribosome but, shortly after synthesis begins, it is halted until the ribosome is directed to the cytoplasmic side of 880
the endoplasmic reticulum . When the ribosome docks with th e membrane, protein sy nthesis begins again. As th e newly forming peptide chain exits the ribosome, it is transported, cotranslationally, through the membrane into the lumen of the endoplasmic reticulum. Free ribosom es that are synthesizing proteins for use in the cell are identical with those attached to the ER. What is the process that directs the ri bosome synthesizing a protein destined to enter the ER to bind to the ER? The translocation consists of four components. Cleavage site Human growth hormone
MA TGS
Human proinsulin Bovine proalbumin
TSLLLAFGLLCLPWLQEGSA
FPT
MALWM R LLPLLALLALWGPDPAAA
FVN
M
WVTFISLLLFSSAYS
RGV
VLSLLYLLTAIPHIMS
DVQ
5 L L I L V L C F L P K L AA L G
KVF
F L V N V A L V F MV V Y I S Y I Y A
APE
L L VVAVIACMLIGFADPASG
CKD
I F C LIM L L G L SA 5 AA T A
5 I F
•
Mouse antibody H chain
M
Chicken lysozyme
M
Bee promellitin
Drosophila glue protein lea maize protein 19 Yeast invertase Human influenza virus A
M M
M A A
MLLOAFLFLLAGFAA
ISA
SMT
M A L LV L L Y A F V A G
DQ I
1. The Signal Sequence. The signal sequence is a sequence of 9 to 12 hydrophobic amino acid residues, sometimes containing positively charged amino acids (Figure 30.29). This sequence is usually near the amino terminus of the nascent polypeptide chain . The presence of the signal sequence identifies the nascent peptide as one that must cross the ER membrane. Some signal sequences are maintained in the mature protein, whereas others are cleaved by a signal peptidase on the lumenal side of the ER membrane (see Figure 30.29). 2. The Signal-Recognition Particle (51
~
o
..c
0-
e 1:;
-
OJ OJ
-
o c .-o
-1.0
1:; .-~ Cl
-
0.2
90 ~ 2 __________~___
CHAPTER 31 The Control of Gene Expression
gene -regulatory processes. Other aspects of eukaryotic gene regulation are quite different from those in prokaryotes. They relate primarily to the role of DNA packaging in eukaryotic genomes.
Multiple Transcription Factors Interact with Eukaryotic Regulatory Sites The basal transcription complex described in Chapter 29 initiates transcription at a low frequency. Recall that several general transcription factors (the preinitiation complex) join with RNA polymerase II to form the basal transcription complex. Additional transcription factors must bind to other sites for a gene to achieve a high rate of mRNA synthesis. In contrast with the regulators of prokaryotic transcription, few eukaryotic transcription factors have any effect on transcription on their own. Instead, each factor recruits other proteins to build up large complexes that interact with the transcriptional machinery to activate transcription. A major advantage of this mode of regulation is that a given regulatory protein can have different effects, depending on what other proteins are present in the same cell. This phenomenon, called combinatorial control, is crucial to multicellular organisms that have many different cell types. Even in unicellular eukaryotes such as yeast, combinatorial control allows th, generation of distinct cell types.
Eukaryotic Transcription Factors Are Modular Transcription factors usually consist of several domains. The DNA-binding domain identifies and binds regulatory sequences that can either be adjacenl to the promoter or at some distance from it. Some activators also incl ude a regulatory domain , which prevents DNA binding under certain conditions. After a transcription factor has bound to the DNA, the activation domain initiates transcription through interactions with RNA polymerase II or its associated proteins. The DNA -binding domain is essential for determining which genes are transcribed. A transcription factor is activated in response to a stimulus and is then responsible for activating the transcription of a set of genes . For exam· pie, the transcription factor NF -KB is activated in response to injury, and it activates the transcription of genes that produce an immune response, help· ing to fight infection. The DNA-binding domain recognizes and binds to a short conserved recognition sequence in the promoter region of each gene or in a more distant enhancer. Often, to increase specificity, the recognition sequence is repeated at regular intervals, and the activators must dimerize before binding to the repeated recognition sequences. Transcription factors can be grouped into families on the basis of the structure of their sequence· specific DNA-binding domains. The helix-tum-helix, homeodomain, bZip, and zinc-finger domains introduced in Section 31.1 are examples of common DNA-binding domains. Transcription factors can often act even if their binding sites lie at a can· siderable distance from the promoter. These distant regulatory sites are called enhancers (p. 838). The intervening DNA can form loops that bring the enhancer-bound activator to the promoter site, where it can act on other transcription factors or on RNA polymerase.
Activation Domains Interact with Other Proteins The activation domains of transcription factors generally recruit other pro· teins that promote transcription. Some of these activation domains interact directly with RNA polymerase II. In other cases, an activation domain may
have multiple interaction partners. These activation domains act through intermediary proteins, which bridge between the transcription factors and the polymerase. An important target of activators is mediator, a complex of 25 to 30 subunits that is part of the preinitiation complex. Mediator acts as a bridge between enhancer-bound activators and promoter-bound l{NA polymerase II (Figure 31.1 8). Activation domains are less conserved than DNA-binding domains. In fact, very little sequence similarity has been found. For example, they may be acidic, hydrophobic, glutamine rich, or proline ri ch. However, certain feat ures are common to activation domains. First, they are redundant. That is, a part of the activation domain can be deleted without loss of function. Second, as described earlier, they are modular and can activate transcription when paired with a variety of DNA-binding domains. Third, activation domains act synergistically : two activation domains acting together create a much stronger effect than either acting separately. We have been addressing the case in which gene control' requires the ex pression of a gene. In many cases, the expression of a gene must be halted by ceasing gene transcription. The agents in such cases are transcriptional repressors. In contrast with activators, repressors bind proteins that block the association of RNA polymerase II with the DNA.
903 31.3 Eukaryotic Regulation of Transcription
Mediator
DNA
Transcription factor
RNA polymerase II
Figure 31.18 Mediator. Med iator. a large complex of pro tein subunits. act s as a bridge between transcription fa ct ors beari ng activation domains and RNA po lymerase II. The se interactions help recru it and stabilize RNA polymerase II near spec ific genes that are then transcribed.
Nucleosomes Are Complexes of DNA and Histones
The control of eukaryotic gene transcription is complicated by the fact that DNA in eukaryotic chromosomes is not bare. Instead, eukaryotic DNA is tightly bound to a group of small basic proteins called histones. In fact, histones constitute half the mass of a eukaryotic chromosome. The entire complex of a cell's DNA and associated protein is called chromatin. Five major histones are present in chromatin: four histones, called H2A, H2B, H 3, and H4, associate with one another; th e other histone is called H1. Histones have strikingly basic properties because a quarter of the residues in each histone are either arginine or lysine. Chromatin is made up of repeating units, each containing 200 bp of DNA and two copies each of H2A, H2B, H 3, and H4 , called the histone octamer. These repeating units are known as nucleosomes. Strong support for this model comes from the results of a variety of experiments, including observations of appropriately prepared samples of chromatin viewed by electron microscopy (Figure 31. 19). C hromatin viewed with the electron microscope has the appearance of beads on a string; each bead has a diameter of approximately 100 A. Partial digestion of chromatin with DNase yields the
I
I
100 nm
Figure 31.19 Chromatin structure. An electron mi crograph of chro matin shOWing its "beads on a st ri ng" character. [Courtesy of Dr. Ad a O lins and Dr Dona ld Olins.]
Amino-terminal tail
(C)
(8)
(A)
~ Figure 31 .20 Nucleosome core particle. The structure consists of a core of eight
histone proteins surrounded by DNA. (A) A view show ing the DNA wrapping around the histone core. (B) A view related to that in part A by a 90-degree rotation. Not ice that the DNA forms a left-handed superhel ix as it wraps around the core. (e) A schematic view. [Drawn from 1AOI.pdb.]
isolated beads . These particles consist of fragments of DNA about 200 bp in length bound to the eight histones. More-extensive digestion yields a shorter DNA fragment of 145 bp bound to the histone octamer. The smaller complex formed by the histone octamer and the 145 -bp DNA fragment is the nucleosome core particle. The DNA connecting core particles in undigested chromatin is called linker DNA. Histone Hl binds, in part, to the linker DNA. Eukaryotic DNA Is Wrapped Around Histones to Form Nucleosomes
The overall structure of the nucleosome was revealed through electron microscopic and x -ray crystallographic studies pioneered by Aaron Klug and his colleagues. More recently, the three-dimensional structure of a reconstituted nucleosome core (Figu re 31.20) was determined to higher resolution b y x-ray diffraction methods. As was shown by Evangelos Moudrianakis, the four types of histone that make up the protein core are homologous and similar in structure (Figure 31.21). The eight histones in the core are arranged into a (H3 h(H4)2 tetramer and a pair of H2A- H 2B dimers. The tetramer and H2A
H28
H3
H4
~ Figure 31.21 Homologous histones. Histones H2A, H2B, H3, and H4 adopt a similar
three-dimensiona l structure as a consequence of common ancestry. Some parts of the tails at the termini of the proteins are not shown. [Drawn from 1AOl.pdb.]
904
dimers come together to form a left -handed superhelical ramp around which the DNA wraps. In addition, each histone has an amino-terminal tail that extends out from the co re structure. These tails are flexible and contain a number of lysine and arg inine residues. As we shall see, covalent modifications of these tails play an essential role in modulating the affinity of the histones for DNA and other properties. The DNA forms a left-handed superhelix as it wraps around the outside of the histone octamer. The protein core forms contacts with the inner surface of the DNA superhelix at many points, particularly along the phosphodiester backbone and the minor groove. N ucleosomes will form on almost all DNA sites, although some sequences are preferred because the dinucleotide steps are properly spaced to favor bending around the histone core. A histone with a different structure from that of the others, called histone HI , seals off the nucleosome at the location at which the linker DNA enters and leaves. The amino acid sequences of histones, including their amino-terminal tails, are remarkably conserved from yeastthrough human beings. The winding of DNA around the nucleosome core contributes to the packing of DNA by decreasing its linear extent. An extended 200-bp stretch of DNA would have a length of about 6RO A. Wrapping this DNA around the histone octamer reduces the length to approximately 100 A along the long dimension of the nucleosome. Thus the DNA is compacted by a factor of 7. However, human chromosomes in metaphase, which are highly con 4 densed , are compacted by a factor of 10 Clearly, the nucleosome is just the first step in DNA compaction. What is the next step? Theo nucleosomes themselves are arranged in a helical array approximately 360 A across, forming a series of stacked layers approximately 110 A apart (Figure 31.22). The folding of these fibers of nucleosomes into loops further compacts DNA. The wrapping of DNA around the histone core as a left-handed helix also stores negative supercoils; if the DNA in a nucleosome is straightened out, the DNA will be underwound (p. 789 ). T his underwinding is exactly what is needed to separate the two DNA strands during replication and transcnptlOn. •
905 31.3 Eukaryotic Regulation of Transcription
•
The Control of Gene Expression Can Require Chromatin Remodeling Does chromatin structure playa role in the control of gene expression? Early observations suggested that it does indeed . DNA that is densely packaged into chromatin is less susceptible to cleavage by the nonspecific DNA-cleaving enzyme DNase 1. Regions adj acent to genes that are being transcribed are more sensitive to cleavage than are other sites in the genome, suggesting that the DNA in these regions is less compacted than it is elsewhere and more accessible to proteins. In addition, some sites, usuall y within 1 kb of the start site of an active gene, are exquisitely sensitive to DNase I and other nucleases. These hypersensitive sites correspond to regions that have few nucleosomes or contain nucleosomes in an altered conformational state. Hypersensitive sites are cell-type specific and developmentally reg ulated. For example, globin genes in the precursors of erythroid cells from 20- hour- old chicken embryos are insensitive to DNase 1. However, when hemoglobin synthesis begins at 35 hours, regions adjacent to these genes become highly su sceptible to digestion . In tissues such as the brain that produce no hemoglobin, the globin genes remain resistant to DNase I through out development and into ad ulth ood. These studies suggest that a prerequisite for gene expression is a relaxing of the chromatin structure. Recent experiments even more clearly revealed the role of chromatin structure in regulating access to DNA binding sites. Genes required for
lloA
( 11 nm)
I
360 A (36 nm)
Figure 31.22 Higher-order chromatin structure. A pro posed model fo r chromatin arranged in a helical array co nsist ing o f six nucleosomes per t urn of helix. The DNA double heli x (shown in red) is wo und around each histone octamer (shown in blue). [After J. T. Finch and A. Klug. Proc. Na tl. Acad. Sci. U. S. A
73(1976):1897-1901.]
906 CHAPTER 31 Expression
The Control of Gene Z n ~=
'1b
Figure 31.23 GAL4 binding sites. The yeast transcription fa ctor GAL4 binds to DNA sequences of the form 5'-CGG(N)l1CCG-3'. Two zin c-based domains are present in the DNA-bind ing region of this protein. Notice that these domains contact the 5' -CGG-3' sequences, leaving the cent er of t he si t e uncontacted. [Drawn from lD66.pdb.]
Start site
TATA
CAGOG
Enhancer • region
Z n ~==
galactose utilization in yeast are activated b y a DNA-binding protein called GAL4, which recognizes DNA binding sites with two 5' -CGG -3' sequ ences separated by 11 base pairs (Figure 31. 23) . Approximately 4000 potential GAL4 binding sites of the form 5' -CGG( N)II CCG-3' are present in the yeast genome, but only 10 of them regulate genes necessary for galactose metabolism. What fraction of the potential binding sites are actually bound by GA L4? This question is addressed through the use of a technique called chromatin immunoprecipitation (ChIP). GAL4 is first cross-linked to its DNA binding sites in chromatin. The DNA is then fragmented into small pieces, and antibodies to GAL4 are used to isolate the chromatin fragments containing GAL4. The cross-linking is reversed , and the DNA is isolated and characterized. The results of these studies reveal that only approxi· mately 10 of the 4000 potential GAL4 sites are occupied by GAL4 when th e cells are growing on galactose; more than 99% of the sites appear to be blocked. Thus, whereas in prokaryotes all sites appear to be equally accessi· ble, chromatin structure shields a large number of the potential binding sites in eukaryotic cells. GAL4 is thereby prevented from binding to sites that are unimportant in galactose metabolism. These lines of evidence and others reveal that chromatin structure is al· tered in active genes compared with inactive ones. How is chrom atin struc· ture modified? As we shall see later (p. 9 10), specific covalent m odifications of histone proteins are crucial. In addition, the binding of specific proteins to enhancers at specific sites in the genome plays a role.
(AGOG
Enhancers Can Stimulate Transcription in Specific Cell Types
CAGCTG
W e now return to the action of enhan cers (p . 902). Recall that these DNA sequences, although they have no promoter activity of their own, greatly increase the activities of many promoters in eukaryotes, even when the enhan cers are located at a distance of several thousand base pairs from the gene being expressed . Enhancers function by serving as binding sites for specific regulatory proteins (Figure 31.24). An enhancer is effective only in the specific cell types in which appropriate regulatory proteins are expressed . In many cases, these DNA-binding proteins influence transcription initiation by perturbing the local chromatin structure to expose a gene or its regulatory sites rather than by direct interactions with RNA polymerase. This mechanism accounts for the ability of enhancers to act at a distance.
nATAAnAA (CATGTAAGG
Figure 31 .24 Enhancer binding sites. A schematic structure for the region 1 kb upstream of the start site for the muscle creatine kinase gene. One binding site of the form 5' -CAG CTG-3' is present near the TATA box. The enhancer region farther upstream contains two binding sites for the same protein and two additional binding sites for other proteins.
Sets
muscle
cells expressing p-galactosidase
Figure 31.25 An experimental demonstration of enhancer function. A promoter for muscle creatine kinase artificially drives the transcripti on of ~ - galactos i dase in a zebrafish embryo. Only specific sets of muscle cells produce ~-galactosidase , as vi sualized by t he formatio n of the blue product on treatment of the em bryo with X-Gal. [From F. Muller, D. W. Williamson, J. Kobolak, L. Gauvry, G. Goldspink, L. Orban, and N. Maclean. Mol.
907 31.3 Eukaryotic Regulation of Transcription
Reprod. Dev. 47(1997):404- 412.]
T he properties of enhancers are illustrated by studies of the enhancer controlling the muscle isoform of creatine kinase (p. 416) . The results of mutagenesis and other studies revealed the presence of an enhancer located between 1350 and 1050 base pairs upstream of the start site of the gene for this enzyme. Experimentall y inserting this enhancer near' a gene not normally expressed in muscle cells is sufficient to cause the gene to be expressed at high levels in muscle cells but not in other cells (Figure 31.25).
The Methylation of DNA Can Alter Patterns of Gene Expression The degree of methylation of DNA provides another mechanism, in addition to packaging with histones, for inhibiting gene expression inappropriate to a specific cell type. Carbon 5 of cytosine can be methylated by specific methyltransferases. About 70% of the 5' -CpG -3' sequences in mammalian genomes are methylated. However, the distribution of these methylated cytosines varies, depending on the cell type. Consider the l3-globin gene. In cells that are actively expressing hemoglobin, the region from approximately 1 kb upstream of the start site to approximately 100 bp downstream of the start site is less methylated than the corresponding region in cells that do not express this gene. The relative absence of 5-methylcytosines near the start site is referred to as hypomethylation. The methyl group of 5-methylcytosine protrud es into the major groove where it could easily interfere with the binding of proteins that stimulate transcription .
H
N
deoxyribose 5-Methylcytosine
The distribution of CpG sequences in mammalian genomes is not T uniform . Many C pG sequences have been converted into TpG through mutation by the deamination of 5-methylcytosine to thymine. However, sites near the 5' ends of genes have been maintained because of their role in gene expression. Thus, most genes are found in CpG islands, regions of the genome that contain approximately four times as many C pG sequences as does the remainder of the genome. .>(J)'
Steroids and Related Hydrophobic Molecules Pass Through Membranes and Bind to DNA-Binding Receptors We next look at an example that illustrates how transcription factors can stimu late changes in chromatin structure that affect transcription. We will consider in some detail the system that detects and responds to estrogens. Synthesized and released by the ovaries, estrogens, such as estradiol, are cholesterol -derived, steroid hormones (p. 753) . They are required for the development of female secondary sex characteristics and, along with progesterone, participate in the ovarian cycle. Because they are hydrophobic molecules, estrogens easily diffuse across cell membranes. When inside a cell , estrogens bind to highly specific, solu ble receptor proteins. Estrogen receptors are members of a large family of
CH 3 ,/
H
Estradiol (an estrogen)
OH
908 CHAPTER 31 Expression
The Contro l of Gene
proteins that act as receptors for a wide range of hydrophobic molecules, in· cluding other steroid hormones, thyroid hormones, and retinoids. I
o /'
o
I - OOC
I Thyroxine (L-3,5,3 ', 5' -Tetraiodothyronine) (a thyroid hormone)
All-trans-retinoic acid
(a retinoid)
The human genome encodes approximately 50 members of this family, often referred to as nuclear hormone receptors. The genomes of other multicellular eukaryotes encode similar numbers of nuclear hormone receptors, although they are absent in yeast. All these receptors have a similar mode of action . On binding of the signal molecule (called, generically, a ligand), the ligand- receptor complex modifies the expression of specific genes by binding to control elements in the DNA. Estrogen receptors bind to specific DNA sites (referred to as estrogen response elements or E R Es) that contain the consensus sequence S' -AGGTCANNNTGACCT -3'. As expected from the symmetry of this sequence, an estrogen receptor binds to such sites as a dimer. A comparison of the amino acid sequences of members of this famil y reveals two highly conserved domains: a DNA-binding domain and a ligandbinding domain (Figure 31. 26). The DNA -binding domain lies toward the center of the molecule and consists of a set of zinc-based domains different from the Cys2His2 zinc-finger proteins introduced near the beginning of the chapter. These zinc-based domains bind to specific D NA sequences by virtue of an 01 helix that lies in the major groove in the specific DNA com· plexes formed by estrogen receptors.
Ligandbinding _ pocket
DNA-binding domain
Ligand-binding domain
~ Figure 31.26 Structure of two nuclear hormone receptor doma ins. Nucl ear ho rmone
recepto rs con t ain t wo cruc ial conserved doma ins: (1) a DNA-binding domain toward the center of the sequence and (2) a ligand-binding do main t o ward the carbo xy l term inus. The structu re of a dimer of t he DNA-bind ing domain bound to DNA is sho wn, as is o ne mono mer of the normally d imeric ligand-binding domain. [Drawn from 1HCQ and 1LBD.pdb.]
CH OH 3
Estradiol
\
~
)
~ Figure 31 .27 ligand binding to
nuclear hormone receptor. The ligand lies completely surrounded within a pocket in the liga nd-binding domain. Notice that the last c< helix, helix 12 (shown in purple), folds into a groove on the side o f the structure on ligand bind ing. [Drawn from 1LDB and 1ERE.pdb.]
Helix 12
•
Nuclear Hormone Receptors Regu late Transcription by Recruiting Coactivators to the Transcription Complex
The second highly conserved domain of the nuclear receptor proteins lies near the carboxyl terminus and is the ligand-binding site. This domain folds into a structure that consists almost entirely of a. helices, arranged in three layers. The ligand binds in a hydrophobic pocket that lies in the center of this array of helices (Figure 31.27). This domain changes conformation when it binds estrogen . How does ligand binding lead to changes in gene expression? The simplest model wou ld have the binding of ligand alter the DNA-binding properties of the receptor, analogously to the lac repressor in prokaryotes. H owever, experiments with purified nuclear hormone recep tors revealed that ligand binding does not significantly alter DNA -binding affinity and specificity. Another mechanism is operative. Because ligand binding does not alter the ability of nuclear hormone receptors to bind DNA, investigators sought to determine whether specific proteins might bind to the nuclear hormone receptors only in the presence of ligand. Such searches led to the identification of several related proteins called coactivators, such as SRC-l (steroid receptor coactivator-i ), GRIP-l (glucocorticoid receptor interacting protein -i), and NcoA-l (nuclear hormone receptor coactivator-1). T hese coactivators are referred to as the p160 family because of their size. T he binding of ligand to the receptor induces a conformational change that allows the recruitment of a coactivator (Figure 31. 28). In many cases, these coactivators are enzymes that catalyze reactions that lead to the modification of chromatin structure.
Estrogen
(ligand)
-\
Coactivator )
\
)
a helix
Figure 31.28 Coactivator recruitment. The binding of ligand to a nuclear hormone receptor induces a conformationa l change in the ligand-binding domain. This change in conformation generat es favorab le sites for the binding of a coactivator.
909
Steroid-Hormone Receptors Are Targets for Drugs
910 CHAPTER 31 Expression
The Control of Gene
....
Molecules such as estradiol that bind to a receptor and trigger signaling pathways are called agonists. Athletes sometimes take natural and synthetic agonists of the androgen receptor, a member of the famil y of nuclear hormone receptors, because their binding to the androgen receptor stimulates the expression of genes that en hance the development of lean muscle mass. CH,
CH, 0
OH II''''
/'
\
CH,
o
o
·CH,
# Dianabol (methandrostenolonej (a synthetic androgen)
Androstendione
(a natural androgen)
Referred to as anabolic steroids, such compounds used in excess are nol without side effects. In men, excessive use leads to a decrease in the secretion of testosterone, to testicular atrophy, and sometimes to breast enlargement (gynecomastia) if some of the excess androgen is converted into estrogen. In women, excess testosterone causes a decrease in ovulation and estrogen secretion ; it also causes breast regression and growth of facial hair. Other molecules bind to nuclear hormone receptors but do not effectively trigger signaling pathways . Such compounds are called antagonists and are, in many ways, like competitive inhibitors of enzymes. Some important drugs are antagonists that target the estrogen receptor. For example, tamoxiJen and raloxifene are used in the treatment and prevention of breast cancer, because some breast tumors rely on estrogen-mediated pathways for growth . These compounds are sometimes called selective estrogen receptor modulaton (SERMs). OH
HO-
\
~ O----'/~-IN Tamoxifen
Helix 12
Tamoxifen
"l:l
Figure 31.29 Estrogen receptor- tamoxifen complex. Tam ox ifen binds in the pocke t normally occupi ed by estrogen. Ho wever, notice t hat part o f the tam ox ifen structure extends fro m this pocket, and so he li x 12 cannot pack in its usual posi t ion. Instead. this he lix blo cks the coacti vato r-binding site. [Drawn from 3ERT.pdb.]
Raloxifene
The determination of the structures of complexes between the estrogen receptor and these drugs revealed the basis for their antagonist effect (Figure 31 .29). Tamoxifen binds to the same site as estradiol does. However, tamoxifen has a group that extends out of the normal ligand-binding pocket, as do other antagonists . These groups block the normal conformational changes induced by estrogen. Tamoxifen blocks the binding of coactivators and thus inhibits the activation of gene expression. Chromatin Structure Is Modulated Through Covalent Modifications of Histone Tails
We have seen that nuclear receptors respond to signal molecules by recruit ing coactivators. Now we can ask, How do coactivators modulate transcriptional activity? These proteins act to loosen the histone complex from the DNA, exposing additional DNA regions to the transcription machinery.
911 - - - - - - - - - - -- 31.3 Eukaryotic Regulation of Transcription Histone H3 tail
Coenzyme A
~ Figure 31.30 Structure of histone acetyltransferase. The amino -terminal tail of
histone H3 extends into a pocket in wh ich a lysine side chain c an accept an acety l group from acetyl e o A bound in an adjacent si te. [Drawn from 1QSN.pdb.]
Much of the effectiveness of coactivators appears to result from their ability to covalently modify the amino-terminal tails of histones as well as regions on other proteins. Some of the p160 coactivators and the proteins that they recruit catalyze the transfer of acetyl groups from acetyl CoA to specific lysine residues in these amino-terminal tails.
o o
H
•
+
~
""'N H
CoA------.
S
o
Lysine in histone tail
- ---.,
""
N H
H
+
•
CoA
SH
+
H+
N H
o
Acetyl eoA
Enzymes that catalyze such reactions are called histone acetyltransferases (HATs) . The histone tails are readily extended ; so they can fit into the HAT active site and become acetylated (Figure 31.30). What are the consequences of histone acetylation? Lysine bears a positively charged ammonium group at neutral pH. The addition of an acetyl group generates an uncharged amide group. This change dramatically reduces the affinity of the tail for DNA and modestly decreases the affinity of the entire histone complex for DNA, loosening the histone complex from the DNA. In addition, the acetylated lysine residues interact with a specific acetyllysine-binding domain that is present in many proteins that regulate eukaryotic transcription. This domain, termed a bromodomain, comprises approximately 110 amino acids that form a four-helix bundle containing a peptide-binding site at one end (Figure 31.31). Bromodomain -containing proteins are components of two large complexes essential for transcription. One is a complex of more than 10 polypeptides that binds to the TATA-box-binding protein. Recall that the TATA -box-binding protein is an essential transcription factor for many genes (p . 837). Proteins that bind to the TATA -box-binding protein are called TAPs (for TATA-box-binding protein associated factors ). In partic ular, T AF1 contains a pair of bromodomains near its carboxyl terminus. The two domains are oriented such that each can bind one of two acetyllysine residues at positions 5 and 12 in the histone H 4 tail. Thus, acetylation
Histone H4 tail
Acetyllysine
~ Figure 31.31 Struct ure of a
bromodomain. Th is fo ur-hel ixbundle doma in binds pep t ides contai ni ng acetyllysi ne. An acetylated pept ide o f histo ne H4 is shown bound in th e stru ct ure. [Drawn f rom 1EGl. pdb. ]
CD
)
(3)
)
o
)
8)
)
®
)
Transcription factor
Exposed site
Coactivator Acetylated lysine residues
RNA polymerase II
Remodeling • engine
Figure 31.32 Chromatin remodeling. Eukaryoti c gene regu lation begins with an activated t ranscription fa ctor bound to a specific site on DNA . One scheme for the initiation of tran scripti on by RNA polymerase II requires fi ve steps: (1) rec ruitm ent of a coactivator, (2) acetyl ation o f lysine residues in the histone tails, (3) binding of a remode lingengine complex t o the acetylated lysi ne residues, (4) Al P-d ependent remode ling of the chro matin structure to expose a bindi ng site for RNA polymerase or for other factors, and (5) recru itment of RNA polymerase. Only two subunits are shown fo r each complex, although the actual complexes are much larger. Other schemes are possible.
of the histone tails provides a mechanism for recruiting other components of the transcriptional machinery. Bromod omains are also present in some components of large complexes known as chromatin-remodeling engines. These complexes, which also contain domains homologous to those of helicases, utilize the free energy of ATP hydrolysis to shift the positions of nucleosomes along the DNA and to induce other conformational changes in chromatin (Figure 31.32). Histone acetyla· tion can lead to a reorganization of the chromatin structure, potentiallyex. posing binding sites for other factors. Thus, histone acetylation can activate transcription through a combination of three mechanisms: by reducing the affin· ity of the his tones for DNA, by recruiting other components of the transcriptional machinery, and by initiating the remodeling of the chromatin structure. N uclear horm one receptors also include regions that interact with com· ponents of the mediator complex. Thus, two m echani sm s of gene regulation can work in concert. Modification of his tones and chromatin remodeling can open up regions of chromatin into which the transcription complex can be recruited through protein protein interactions. Histone Deacetylases Contribute to Transcriptional Repression
Just as in prokaryotes, some changes in a cell's environment lead to the reo pression of genes that had been active. The modification of histone tails again plays an important role. H owever, in repression , a key reaction appears to be the de acetylation of acetylated lysine, catalyzed by specific histone deacetylase enzym es. Tn many ways, the acetylation and deacetylation of lysine residues in histone tail s (and, likely, in other proteins) is analogous to the phosphoryla· tion and dephosphorylation of serine, threonine, and tyrosine residues in other stages of signaling processes. Like the addition of phosphoryl groups, the addition of acetyl groups can induce conformational changes and gener· ate novel binding sites. Without a m eans of removing these groups, how· ever, these signaling switches will become stuck in one position and lose their effectiveness. Like phosphatases, deacetylases help reset the switches. Acetylation is not the only m odifi cation of his tones and other proteins in gene- regul ation processes. T he m ethylation of specific lysine and arginine residues also can be important. The elucidation of the roles of these processes is a very active area of research at present. 912
31.4
913
Gene Expression Can Be Controlled at Posttranscriptional Levels
31.4 Posttranscriptional Gene Regulation
The modulation of the rate of transcriptional initiation is the most common mechanism of gene regul ation. However, other stages of transcription also are targets for regulation in some cases. In addition, the process of translation provides other points of intervention for regulating the level of a protein produced in a cell. These mechanisms are quite distinct in prokaryotic and eukaryotic cells because prokaryotes and eukaryotes differ greatly in how transcription and translation are coupled and in how translation is initiated. We will consider two important examples of posttranscriptional regulation: one from prokaryotes and the other from eukaryotes. In both examples, regulation depends on the formation of distinct secondary struc tures in mRNA. •
Attenuation Is a Prokaryotic Mechanism for Regulating Transcription Through the Modulation of Nascent RNA Secondary Structure A new means for regulating transcription in bacteria was discovered by Charles Yanofsky and his colleagues as a result of their studies of the tryp tophan operon. This operon encodes five enzymes that convert chorismate into tryptophan (p . 694). Analysis of the 5' end of trp mRNA revealed the presence of a leader sequence of 162 nucleotides before the initiation codon of the first enzyme. The next striking observation was that bacteria produced a transcript consisting of only the first 130 nucleotides when the tryptophan level was high, but they produced a 7000 -nucleotide trp mRNA, including the entire leader sequence, when tryptophan was scarce. Thus, when trytophan is plentiful and the biosynthetic enzymes are not needed , transcription is abruptly broken off before any coding mRNA for the enzymes is produced. The site of termination is call ed the attenuator, and this mode of regulation is called attenuation. Attenuation depends on features at the 5' end of the mRNA product (Figure 31.33). The first part of the leader sequence encodes a 14-amino-acid leader peptide. Foll owing the open reading frame for the peptide is a region of RN A representing the attenuator, which is capable of forming several al ternative structures . Recall that transcription and translation are tightly coupled in bacteria. Thus, the translation of the trp mRNA begins soon after the ribosom e-binding site has been synthesized. How does the level of tryptophan alter transcription of the trp operon? An important clue was the finding that the 14-amino-acid leader peptide includes two adjacent tryptophan residues. A ribosome is able to translate the leader region of the mRNA product only in the presence of adequate concentrations of tryptophan. When enough tryptophan is present, a stem-loop structure
(A)
Figure 31.33 Leader region of trp mRNA . (A) The nucleotide sequence of the 5' end of trp mRNA includes a short open reading frame that encodes a peptide comprisi ng 14 amino acids; the leader encodes two t ryptophan residues and has an untranslated attenuator region (blue and red nucleotid es). (8 and C) The attenuator regi on can adopt t wo distinct stem-loop structures.
Attenuator M et - Lys - Ala - lie - Phe - Val- Leu - Lys - Gly - Trp - Trp - Arg - Thr - Ser - Stop , ' , 5'- ... AUG AAA GCA AUU UUC GUA CUG AAA GGU UGG UGG CGC ACU UCC UGA(N)4,CAGCCCGCCUAAUGAGCGGGCU UUU UUUUGAACAAAAU. .. 3 '
(S) AAU
U G A CC G G'C CG C' G C'G G·C -CA ' UUUU UUUUGAACAAAAU-
(C)
AA G' C U'A U'A U'A U·A -CAGCCCGCCUAAUGAGCGGGCU UUU U-
(B)
(A)
ww c •••••••• : I r:
••
Ribosome
Terminates transcription
Alternative structure No termination
trp mRNA RNA polymerase
forms in the attenuator region, which leads to the release of RNA polymerase from the DNA (Figure 31.34). However, when tryptophan is scarce, transcription is terminated less frequently. Little tryptophanyl-tRNA is present, and so the ribosome stalls at the tandem UGG codons encoding trypto· phan. This delay leaves the adjacent region of the mRNA exposed as transcription continues. An alternative RNA structure that does not fun ction as a terminator is formed , and transcription continues into and through the coding regions for the enzymes. Thus, attenuation provides an elegant means of sensing the supply of tryptophan required for protein synthesis. Several other operons for the biosynthesis of amino acids in E. coli also are regulated by attenuator sites. The leader peptide of each contains an abundance of the amino acid residues of the type synthesized by the operon (Figure 31.35). For example, the leader peptide for the phenylalanine operon includes 7 phenylalanine residues among 15 residues . The threonine operon encodes enzymes required for the synthesis of both threonine and isoleucine; the leader peptide contains 8 threonine and 4 isoleucine residues in a 16-residue sequence. The leader peptide for the histidine operon in· cludes 7 histidine residues in a row. In each case, low levels of the corre· sponding charged tRNA causes the ribosome to stall, trapping the nascent mRNA in a state that can form a structure that allows RNA polymerase to read through the attenuator site.
Figure 31 .34 Attenuation. (A) In the presence of adequate concentrations of tryptophan (and, hence, Trp-tRNA), translation proceeds rapidly and an RNA structure forms that terminates transcript ion. (B) At low concentrations o f trypt ophan, translation stalls w hile awaiting Trp-tRN A, giving t ime for an alternati ve RNA structure to form that does not terminate transcripti on efficiently.
Figure 31.35 Leader peptide sequences. Amino acid sequences and the co rresponding mRNA nucleotide sequences of the (A) threo nine operon, (B) phenylalan ine o peron, and (C) histid ine operon. In each case. an abundance of one amino acid in the leader peptide sequence leads t o attenuation.
Met - Lys - Arg - lie - Ser - Thr - Thr - lie - Thr - Thr - Thr - lie - Thr - lie - Thr - Thr .
(A) 5'
AUG AM CGC AUU AGC ACC ACC AUU ACC ACC ACC AUC ACC AUU ACC ACA
3'
Met - Lys - His - lie - Pro - Phe - Phe - Phe - Ala - Phe - Phe - Phe - Thr - Phe - Pro - Stop
(B) 5'
AUG AM CAC AUA CCG UUU UUC UUC GCA UUC UUU UUU ACC UUC CCC UGA
3'
Met - Thr - Arg - Val. - Gin - Phe - Lys - His - His - His - His - His - His - His - Pro - Asp -
(C) 5'
AUG ACA CGC GUU CM UUU AM CAC CAC CAU CAU CAC CAU CAU CCU GAC
3'
Genes Associated with Iron Metabolism Are Translationally Regulated in Animals RNA secondary structure plays a role in the regulation of iron metabolism in eukaryotes. Iron is an essential nutrient, required for the synthesis of hemoglobin, cytochromes, and many other proteins. However, excess iron can be quite harmful because, untamed by a suitable protein environment, iron can initiate a range of free -radical reactions that damage proteins, lipids, and nucleic acids. Animals have evolved sophisticated systems for the accumula· tion of iron in times of scarcity and for the safe storage of excess iron for later use. Key proteins include transferrin , a transport protein that carries iron \n the serum, transferrin receptor, a membrane protein that binds iron-loaded transferrin and initiates its entry into cells, and f erritin, an impressively 914
efficient iron-storage protein found primarily in the liver and kidneys. Twenty-four ferritin polypeptides form a nearly spherical shell that encloses as many as 2400 iron atoms, a ratio of one iron atom per amino acid (Figure 31.36). Ferritin and transferrin-receptor expression levels are reciprocally related in their responses to changes in iron levels. When iron is scarce, the amount of transferrin receptor increases and little or no new ferritin is synthesized. Interestingly, the extent of mRNA synthesis for these proteins does not change correspondingly. Instead, regulation takes place at the level of translation. Consider ferritin first . Ferritin mRNA includes a stem-loop structure termed an iron-response element (IRE) in its 5' un translated region (Figure 31.37). This stem-loop binds a 90-kd protein, called an IREbinding protein (IRP), that blocks the initiation of translation. When the iron level increases, the IRP binds iron as a 4Fe-4S cluster. The IRP bound to iron cannot bind RNA, because the binding sites' for iron and RNA substantially overlap. Thus, in the presence of iron, ferritin mRNA is released from the IRP and translated to produce ferritin, which sequesters the excess iron . An examination of the nucleotide sequence of transferrin-receptor mRNA reveals the presence of several IRE-like regions . However, these regions are located in the 3' untranslated region rather than in the 5' untranslated region (Figure 31.38). Under low -iron conditions, IRP binds to these IREs. However, given the location of these binding sites, the transferrin-receptor mRNA can still be translated. What happens when the iron level increases and the IRP no longer binds transferrin -receptor mRNA? Freed from the IRP, transferrin -receptor mRNA is rapidly degraded. Thus, an increase in the cellular iron level leads to the de struction of transferrin-receptor mRNA and, hence, a reduction in the production of transferrin-receptor protein .
(A)
Iron
oxide-hydroxide core
(8) ~ Figure 31.36 Structure of ferritin.
G
(A) Twenty-four ferritin po lypept ides form a nearl y spheri cal shell. (B) A cutaway view reveal s th e core that stores iron as an iron OXide- hydrox ide complex. [Draw n from lIES.pdb.]
A· U C' G A .U C
A . U Iron-response C.G element
U· G U· A C·C
G' C C'G UG.C
A· U G' C G' C
5'- - -
' ' --
Coding region
- -_ _ _ _ _ __
_ _ -3'
Figure 31.37 Iron-response element. Ferritin mRNA includes a stem-loop stru cture, t ermed an iron-response element (IRE), in its 5' untranslated region. The IRE binds a specific protein that blocks the translation of this mRNA under low iron conditions.
Iron-response elements
Coding region - - -3'
Figure 31.38 Transferrin-receptor mRNA. This mRNA has a set of iron-respon se elements (IREs) in its 3' untranslated region. The bind ing of th e IRE-bindi ng protein t o these elements stabi lizes the mRNA but does not interfere w ith translation.
915
916 CHAPTER 31 Expression
(A)
(B)
The Control of Gene
4Fe-4S cluster -
High-iron conditions
~ Figure 3139 The IRE-BP is an
aconitase. (A) Aconitase cont ains an unstable 4Fe-4S cluster at it s center. (B) Under cond it ions of low iro n, the 4Fe-4S cl ust er d issociates and appro priate RNA molecules can bind in it s place. [Drawn fro m 1C96.pdb.)
Low-iron conditions
T he purif ication of the IRP and the clon in g of its cDNA were source~ of truly remarkable insight into evolution . T he IRP was found to b, approximately 30% identical in amino acid sequence with the citric acid cycle enzyme aconitase from mi toch ondria. F urther an alysis revealed that the IRP is, in fact , an active aconitase enzym e; it is a cytoplasmic aconitase t hat h ad been known for a long tim e, bu t its function was n ot well understood (Figure 31.39). T he iron- sulfur center at the active site of the IRP is rath er unstable, and loss of the iron tri ggers significant changes in protein conform ation. Thus, this protein can serve as an iron- sensing factor. O th er m RNAs, including those taking part in heme syn thesis, have been found to contain I REs. Thus, genes en coding proteins requ ired for iron m etabolism acquired sequences that, when transcribed , p rovided binding sites for the iron- sensing p rotein. An environmental signal the concentration of iron controls the translation of proteins requi red for the m etabolism of this m etal. Thus, mutations in the untranslated region of mRNAs have been selected for beneficial regulation by iron levels.
Summary 31.1
Many DNA-Binding Proteins Recognize Specific DNA Sequences T he regulation of gene expression d epends on the interplay between specific sequences within the genom e and proteins th at bind specifically to these sites. Specific DNA- binding proteins recognize regul atory sites that usually lie adj acent to the genes whose transcription is regul ated by these proteins. M any families of such DNA-binding proteins have been iden tified . In prokaryotes, the proteins of the largest family contain a helix -turn -helix m otif. T he first helix of this motif inserts into the major groove of DNA and makes specific hydrogen -bonding and other contacts with the ed ges of the base pairs. In eukaryotes, important classes of DNA-binding proteins include t he homeodom ains, the basic-leucine zipper (bZip) proteins, and Cys2H is2 zinc-fin ger proteins. Each of these classes of p roteins uses an ex helix to make specific contacts with DNA. Although th e use of ex helices in DNA recognition is m ost common, some p roteins use other structural elements.
31.2 Prokaryotic DNA-Binding Proteins Bind Specifically to Regulatory Sites in Operons In prokaryotes, many genes are cl ustered into operons, wh ich are units of coordinated genetic expression . An operon consists of control sites (an operator and a promoter) and a set of structural genes. In addition,
regulator genes encode proteins that interact with the operator and promoter sites to stimulate or inhibit transcription. The treatment of E. coli with lactose induces an increase in the production of [3-galactosidase and two additional proteins that are encoded in the lactose operon. In the absence oflactose or a similar galactoside ind ucer, the lac repressor protein binds to an operator site on the DNA and blocks transcription. The binding of allolactose, a derivative oflactose, to the lac repressor induces a conformational change that leads to dissociation from DNA. RNA polymerase can then move through the operator to transcribe the lac operon. Some proteins activate transcription by directly contacting RNA polymerase. For example, cyclic AMP, a hunger signal, stimulates the transcription of many catabolic operons by binding to the catabolite activator protein. The binding of the cAMP- CAP complex to a specific site in the promoter region of an inducible catabolic operon enhances the binding of RNA polymerase and the initiation of transcription. 31.3 The Greater Complexity of Eukaryotic Genomes Requires Elaborate Mechanisms for Gene Regulation
Eukaryotic genomes are larger and more complex than those of prokaryotes. Some regulatory mechanisms used in eukaryotes are similar to those used in prokaryotes. In particular, most eukaryotic genes are not expressed unl ess they are activated by the binding of specific proteins, called transcription factors, to sites on the DNA . These specific DNAbinding proteins interact directly or indirectly with RNA polymerases or their associated proteins. Eukaryotic transcription factors are modular: they consist of separate DNA-binding and activation do mains. Activation domains interact with RNA polymerases or their associated factors or with other protein complexes such as mediator. Enhancers are DNA elements that can modulate gene expression from more than 1000 bp away from the start site of transcription. Enhancers are often specific for certain cell types, depending on which DNAbinding proteins are present. E ukaryoti c DNA is tightly bound to basic protein s called histones; the combination is called chromatin. DNA wraps around an octamer of core histones to form a nucleosome, blocking access to many potential DNA binding sites. Changes in chromatin structure playa major role in regulating gene expression. Steroids such as estrogens bind to eukaryotic transcription factors called nuclear hormone receptors. These proteins are capable of binding DNA whether or not ligands are bound . The binding of ligands induces a conformational change that allows the recruitment of additional proteins called coactivators. Among the most important functions of coactivators is to catalyze the additi on of acetyl groups to lysine residues in the tails of histone proteins. Histone acetylation decreases the affinity of the histones for DNA, making additional genes accessible for transcription. In addition, acetylated histones are targets for proteins containing specific binding units called bromodomains. Bromodomains are components of two classes of large complexes : (1 ) chromatin-remodeling engines and (2 ) factors associated with RNA polymerase II. These complexes open up sites on chromatin and initiate transcription. 31.4 Gene Expression Can Be Controlled at Posttranscriptional Levels
Gene expression can also be regulated at the level of translation. In prokaryotes, many operons important in amino acid biosynthesis are regulated by attenuation, a process that depends on the formation of alternative structures in mRNA, one of which favors the termination
917 Summary
918 CHAPTER 31 The Control of Gene Expression
of transcription. Attenuation is mediated by the translation of a leader region of mRNA. A ribosome stalled by the absence of an aminoacyltR A needed to translate the leader mRNA alters the structure of mRNA, allowing RNA polymerase to transcribe the operon beyond the attenuator site. In eukaryotes, genes encoding proteins that transport and store iron are regulated at the translational level. Iron-response elements, structures that are present in certain mRNAs, are bound by an IRE-binding protein when this protein is not binding iron. Whether the expression of a gene is stimulated or inhibited in response to changes in th e iron status of a cell depends on the location of the I RE within the mRNA.
Key Terms heli x-turn - heli x m otif (p . 89 5)
cell type (p . 901)
anabolic st eroid (p . 9 10)
hom eod omain (p . 895)
combinatorial control (p . 9 0 2)
antagonist (p _9 10)
basic-leucine zipper (bZip) protein (p . 895)
enhancer (p .90 2) mediator (p . 9 03)
selective estrogen modulato r (SERM) (p . 910 )
CyszHis 2 zinc-finger domain (p . 895)
histo ne ( p . 903)
histone acetyl transferase (HAT) (p. 91 1:
J3 -galactosidase (p . 896)
chromatin (p . 9 03 )
acetyllysine -binding domain (p . 911)
operon mo d el (p . 89 7)
nucleosome (p . 903)
bromodomain (p . 9 1 1)
repressor (p . 89 7)
nucl eosome core particle (p . 904)
lac repressor (p . 89 7) lac o pera to r (p . 898 )
hypersensitive site (p. 90S)
TATA - box -binding protein associated factor (TAF) (p . 9 11 )
inducer (p . 898)
chro matin immunoprecipitation (ChIP ) (p . 906)
chromatin -remodeling engine (p. 912) histone d eacetylase (p . 9 12)
isopropyl t hiogalactoside (IPTG ) (p . 898)
hy po m ethylation (p . 9 0 7)
attenuatio n (p . 9 13)
C pG is land (p . 90 7)
transferrin ( p _914 )
pur repressor (p . 899)
nuclear h ormone receptor (p . 908)
transferrin receptor (p . 9 14)
corepressor (p . 900)
estrogen response element (ERE ) (p . 908)
ferritin (p . 9 14)
coactivator (p . 909 )
IRE-binding protein (IRP) (p. 915)
catabolite repressio n (p . 900) catabo lite acti vator protein (CAP) (p . 900)
iron- response element (IRE) (p. 9 15)
agoni st (p. 9 10)
Selected Readings Where to Start Pabo, C . 0 ., and Sauer, R. T 1984. Protein- D N A recognition Annu. Rev. IJiocltem. 53 :293- 32 1. Slruhl. K. 1989 . ~I el ix - lurn - hel i x , zinc-finger, and leucine-zipper moti fs for eukaryoti c transcri ptional regulatory proteins . Irends Biochem. Sci. 14:137- 140. Struhl. K. 1999. Fundamentally different logic of gene regulation in eukaryotes and proka ryotes. Ce ll 98 :1- 4. Korzus, E., Torchia, ]" Rose, D . W ., Xu, L. , Kurokawa, R .. Mclnerney, E. M ., Mullen, T. M ., G lass, C. K., and Rosenfeld, M . G . 1998 . Transcription factor-specific requirements for coactivators and their acetyltransferase fun ct.ions. Science 279: 703- 707 . Aalfs, j. D., and Kingston, R. E. 2000. What does "chromatin remodeling" mean? Trends Biochem. Sci. 25 :54 8- 555.
Wolffe, A . 1992 . Chromatin S tructure and Function. Academic Press. Lodish, H ., Berk , A ., M atsudaira, P., Kai ser, C . A ., Krieger, M.. Scott, M . P., Zipursky, S. L. , and D arnell, ] ., 2004. M olecular Cell Biology (5th ed .). W . H . Freeman and Company.
Books Ptashne, M. 2004 . A Genetic Switch: Phage A Revisited (3d ed .). Cold Spring Harbor Laboralory Press. McKnight, S. L., and Yamamoto, K. R. (Eds.). 1992. Transcriptional Regulation (vols. 1 and 2). Cold Spring Harbor Laboratory Press. Larchman , D. S. 2004 . Eukaryotic Tmnscription Factors (4th cd .). Academic Press.
DNA and inducer. Science 271 :1247- 1254. N iu , W ., Kim , Y. , Tau , G ., Heyduk, T, and Ebri ght, R. H. 1996.
Prokaryotic Gene Regulation Balaerr. A ., Mahadevan, L. and Schul ten, K. 2004. Structural basis for cooperalive D NA binding by CAP and lac repressor. Structure 12: 123- 132 . nell , C . E., and Lewis, M. 200 1. The Lac repressor: A second generation of structural and functional studies. Curro Opin. Struct. BioI. 11 :19- 25. Lewis, M .. Chang , G .. Horton, N . c., Kercher. M . A., Pace. H. C.. Schumacher. M . A., Brennan, R. G ., and Lu , P. 1996. Crystal stru cture of th e lactose operon re pressor and its complexes with
Transcription acti vation at class TT C AP -dependent promoters:
Two interactions hetween C AP and RNA polymerase. Cell 87:1 123- 1134. Schultz, S. c., Shields, G . C ., and Steitz, T A . 1991. Crystal struclure of a CAP-D NA complex: The DNA is benl by 90 degrees. Science 253:1001- 1007.
Selected Readings 919 Parkinson, G., Wil son, c., Gunasekera, A ., Ebright, Y. W ., Ebright, R. E., and Berman, H . M . 1996.Structureofthe C AP-DNA complex at 2. 5 A resolution : A complete picture of the protein- 0 NA interface. j. Mol. Bioi. 260 :39 5- 408. Busby,S., and Ebright, R. H . 1999. Transcription activation by catabolite activator protein (CAP ). j. Mol. Bioi. 293: 199- 213. Somers, W. 5., and Phillips, S. E. 1992. C rystal structure of the met repressor-operator complex at 2.8 A resolution reveals DNA recognition by [3 -strands. Nature 359:387- 393 .
Eukaryotic Gene Regulation Green, M . R. 2005. Eukaryotic transcription activation: Right on target . Mol. Cell 18:399- 402. Kornberg, R. D . 2005. Mediator and the mechanism of transcriptional activation. Trends Biochem. Sci. 30:235- 239. Luger, K., Mader, A . W .. Richmond . R. K., Sargent, D . F , and Richmond. 1'. j . 1997. C rystal structure of the nucleosome core particle at 2.8 Aresolution . Nature 389:25 1- 260. Arents, G ., and Moudrianakis, E. N . 1995 . The histone fold : A ubiqui tous architectural motif utilized in DNA compaction and protein dimerization . Proc. Natl. Acad. Sci. U. S. A. 92: 111 70- 111 74. Baxevanis, A . D ., Arents, G., Moudrianakis, E . N. , and Landsman, D . 1995. A variety of DNA-binding and multimeric proteins contain the histone fold motif. Nucleic Acids Res. 23:2 685- 269 1. Clements, A., Rojas, j . R., Trievel, R. C., Wang, L., Berger, S. L. , and Marmorstein, R. 1999. C rystal structure of the histone acetyltransferase domain of the human PCAF transcriptional regulator bound to coenzyme A . EMBO J. 18:3 521 - 3532. Deckert. J ., and Struhl, K. 200 1. Histone acetylation at promoters is differentially affected by specific activators and repressors. Mol. Cell. BioI. 21:272 6 2735. Dutnall . R . N ., Tafrov. S. 1'., Sternglanz . R. . and Ramakrishnan . V. 1998. Structure of the histone acetyltransferase Hatl : A paradigm for the G CN5- related N-acetyltransfera se superfamily. Cell 9 4 : 4 27-4 ~R .
Finnin, M . S., Oonigian , j. R., Cohen, A ., Richon, V. M ., Rifkind, R . A., Marks, P. A., Breslow, R. , and Pavletich, N . P. 1999 . Stru ctures of a histone deacetylase homologue bound to the T SA and SA H A inhibitors. Nature 40 1:188- 193. Finnin. M . S., D onigian, j. R., and Pavletich. N. P. 200 1. Structure of the histone deacetylase SI R2 . Na t. Struct. Bioi. 8:621 - 625. Jacobson, R. H ., Ladurner, A . G ., King. D . S .. and 1Jian, R. 2000. Structure and function of a human TAFII250 double bromodomain module. Science 288: 1422 1425. Rojas, J . R .. Trievel. R . C. Zhou. J., Mo, Y. Li, X., Berger. S. L., Allis, C. D ., and Marmorstein . R . 1999 . Stru cture of Tetrahymena GCN 5 bound to coenzyme A and a histone H 3 peptide. Nature 40 1:93- 98.
Nuclear Hormone Receptors Downes. M ., Verdecia. M . A., Roecker, A . J .. Hughes, R ., Hogenesch, J. B., Kast-W oelbern. H . R., Bowman, M . E .. Ferrer, J . L., Anisfeld. A . M .. Edwards. P. A .. Rosenfeld. J . M ., Alvarez, J. G .. Noel, j. P., Nicolaou. K . C , and Evans. R.M . 20U3 . A chemical, genetic, and structural anaJysis of the nuclear bile acid receptor FXR. Mol. Cell 11 :1079- 109 2. Evans, R. M . 2005 . Th e nuclear receptor superfamily: A Rosetta stone for physiology. Mol . Endocrinn/. 1 9: 14 29- 14 ~R. Xu, W ., Cho, H ., Kadam, S., flanayo, E . M ., Anderson, S., Yates, J . R. , ~d, Emerson, B. M ., and Evans, R. M . 2004. A methylation-mediator complex in hormone signaling. Genes Dev. 18:144- 15 6. Evans, R. M . 1988. The steroid and thyroid hormone receptor super fami ly. Science 240 :889- 895. Yamamoto. K . R. 1985. Steroid receptor regulated transcription of spe cific genes and ge ne networks. Amlll . Rev. Genet. 19:2U9- 252. Tanenbaum, D . M .. W ang. Y., Williams. S. p .. and Sigler. P. B. 1998. S rystallographic comparison of the estrogen and progesterone
receptor 's ligand binding domains. Proc. Nat/. Acad. Sci. U. S . A. 95: 5998- 6003. Schwabe. j. W ., Chapman, L.. Finch, ). 1'., and Rhodes, D. 1993. The crystal structure of the estrogen receptor DNA -binding domain bound to DNA : How receptors discriminate between their response elements . Cell 75:567- 578. Shiau. A . K., Barstad, D ., Loria. P. l-.iJ.. C heng, L., Kushner, P. J ., Agard, D . A .. and Greene, G. L. 1998 . The structural basis of estrogen receptor/ coactivator recognition and the antagonism of this interaction by tamoxifen . Ce ll 95:927- 937 . Collingwood. 1'. N ., Urnov. F. D., and Wolffe, A . P. 1999 . Nuclear receptors: Coactivators, corepressors and chromatin remodeling in the control of transcription . }. Mnl. Endncrino/. 2 ~ :2 55-2Ti .
Chromatin and Chromatin Remodeling Elgin , S. C . 1981. D NAase I-hypersensitive sites of chromatin . Cell 27:4B-41S. Weintraub, H ., Larsen , A ., a.nd Groudine, M . 19R1. a-G lobin -gene switching durin g the development of chicken embryos: Expression and chromosome structure. Ce1l 24:~~~-~44 . Ren, B., Robert, F., W yri ck, j . j., Aparicio, 0., Jennings, E. G ., 5imon, I., Zeitlinger, J ., Schreiber, J., Hannett, N ., Kanin, E., Volkert, T. L., Wilson , C. j., Bell , S. P., and Young. R. A . 2000. Genome-wiJe location and function of DNA -binding proteins. Science 290:2306- 2309 . Goodrich, J. A., and 1 J ian. R. 1994 . TBP -TAF complexes: ~e1ecti vi ty factors for eukaryotic transcription . Curro Opin. Ce ll. Bioi. 6:403 409 . Bird . A . P.. and Wolffe. A . P. 1999. Methylation -induced repression : Belts, braces, and chromatin . Cell 99 :451- 454 . Cairns, fl. R . 199R. Chromatin remodeling machines: Similar motors, ulteri or motives. Trends flin chem. Sci. 23:20- 25. Albright, S. R., and Tjian , R. 2000. TAFs revisited : More data reveal new twists and confirm old ideas. Gene 242: 1- 13. Urnov, F D., and Wolffe, A . P. 2001 . C hromatin remodeling anJ tran scriptional activation : The cast (in order of appearance). Oncogene 20:2991 - 3006. Posttranscriptional Regulation Kolter, R., and Yanofsky, C. 1982. Attenuation in amino acid biosyn · thetic operons. Annu. Rev. Genet. 16: 113- 134. Yanofsky, C. 1981. Attenuation in the control of expression of bacteri al operons. Nature 289:7 51 - 758. Rouault, T. A., Stout, C . D ., Kaptain, S .. Harford, J. 13., and Klausner, R. D . 199 1. Structural relationship between an iron -regulated RNA -binding protein (lRE-BP) and aconitase: Functional impli· cations. Ce ll 64 :881 - 883. Kl ausner. R. 0 .. Rouault, 1'. A ., and Harford , ). B. 1993. Regulati ng the fate of mR NA: The control of cellular iron metaboli sm. Cell 72: 19- 2R. Gruer, M . J., Artymiuk , P. J ., and Guest, J. R. 1997. The aconitase family : Three structural variations on a common theme. Trends Bi()chem. Sci. 22:3- 6. Theil, E. C. 1994. Iron regulatory elements (IREs): A family of mRNA non-coding sequences. Biuchem. }. 304:1 - 11 .
Historical Aspects Lewis, M . 2005 . The lac repressor. C. R. Bioi. 328:52 1-548. Jacob. F., and M onad, j. 1961 . Genetic regulatory mechanisms in the synthesis of proteins. }. Mol . fliol. ~:] 1R-~56 . Ptashne, M ., and Gilbert, W . 1970. Geneti c repressors. Sci. Am. 222(6) :36- 44 . Lwoff, A., and Ullmann, A. (Eds .). 1979. Origins oj Molecular Biology : A Tribute to Jacques Monod. Academic Press. Judson , H . 1996 . The Eighth Day oj Creation : Makers oj the Revolution in Biology. Cold Spring Harbor Laboratory Press.
,
920
CHAPTER 31 The Control of Gene Expression
Problems 1. M issing genes. Predi ct th e effects of del etin g the following regions of D N A : (a) T he gene encoding lac repressor (b) T he lac operator (c) T he gene encoding CAP 2. Minimal concentration. C alculate the concentration of lac repressor, assumin g lhal one molecule is present per cell. A ssume th at each E. coli cell has a volume of 10- 12 cm ] Would you expect the single molecule to be free or bound to D N A '
3. Counting sites. Calcul ate the expected number of times that a given 8-base- pair DNA site should be present in the E. co li genome. A ssume that all four bases are equall y probable. Repeat for a IO- base- pair site and a 12- base- pair site. 4. Charge neutralization. G iven the histone amino acid sequences illustrated below, estimate the charge of a histone oclamer at pH 7. A ssume that histidine residues are uncharged at this pH. How does thi s charge compare with the charge on 150 base pairs o f D NA?
7. A new domain, A protein d omain that recognizes S- methyl · cytosine in the context of double-stranded DNA has been characterized . What role might proteins containing such a do· main play in regulating gene expression ? Where on a double· stranded DNA molecule would you expect such a domain to bind ? 8. Th e same but not the same. The lac repressor and the pur repressor are homologous proteins with very similar three-dimensional structures, yet they have different effects on gene expression. Describe two important ways in whi ch the gene-regulatory properties of these proteins differ.
9. The opposite direction. Some compo unds called anti-inducers bind to repressors such as the lac repressor and inhibit the action of inducers; that is. transcription is repressed and higher concentrations of inducer are required to induce transcription. Propose a mechanism of action for anti-inducers. 10 . Inverted repeats. Suppose that a nearly perfect inverted repeat
is observed in a D N A sequ ence over 20 base pairs. Provide two possible explanations.
Histone H2A MSGRGKQGGKARAKAKTRSSRAGlQFPVGRVHRllRKGNYSERVGAGAPVYlAAVlEYlTAEILELAGNA
Mechanism Problem
ARONKKTRl lPRHl QLAIRNDEElNKllGRVTIAQGGVlPNIQAVllPKKTESHHKAKGK
11 . A cetyltransf erases. Propose a mechanism for the transfer of an acetyl group from acetyl CoA to the amino group of lysine.
Histone H1B MPEPAKSAPAPKKGSKKAVTKAQKKDGKKRKRSRKESYSVYVYKVlKQVHPDTGISSKAMGIMNSFVNOI
FERIAGEASRLAHYNKRSTITSREIQTAVRl LlPGELAKHAV5EGTKAVTKYT55K
Data Interpretation Problem
Histone H3
12. Limited res triction. Th e restriction enzyme Hpall is a
MARTKQTARKSTGGKAPRKQLATKAARKSAPSTGGVKKPHRYRPGTVALREIRRYQKSTELlIRKLPFQR
powerful tool for analyzing D N A methylation . This enzyme cleaves sites of the form 5' -CCGG -3 ' b u t will not cleave such sites if th e DNA is m ethylated on any of the cytosine residues. Genomic DNA from different organi sm s is treated with HpaII and the results are analyzed b y gel electrophoresis (see the adjoining patterns) . Provide an explanation for the observed pattern s.
LVREIAQOFKTDlRFQSAAIGALQEASEAYlVGlFEDTNlCAIHAKRVTIMPKDIQLARRIRGERA
Histone H4 MSGRGKGGKGlGKGGAKRHRKVLRDN IQCITKPAIRRLARRGGVKRISGUYEETRGVLKVFlENV1RDA
VMEHAKRKMAMDVVYAl KRQGRTlYGFGG
5. Chromatin immunoprecipitation. You have used the technique of chromatin immunoprecipitation to isolate DNA fragments containing a D N A -bind ing protein of in terest. Suppose that you wish to know whether a particular known DNA fragment is present in the isolated mixture. How might you detect its presence? How many different fragments would you expect if you used antibodies to the lac repressor to perform a chromatin immunoprecipitation experiment in E. coli ? If you used antibodies to the pur repressor?
Mouse
> 50 kb
6. Nitrogen substitution. Growth of mammalian cells in the presence of 5-azacytidine results in the activation of some normally inactive genes. Propose an explanation . NH2
100 bp
N
N
deoxyribose 5-Azacytidine
H
Drosophila
E. coli
Chapter
Sensory Systems
Color perception requires specific photo receptors. The p hotorecept or rh o dops in (right), which absorbs light in the process of vision, cons ists of the protein opsin and a bound vitamin A derivative, retinal. The amino acids (shown in red) that su rround the retinal determine the color of light that is most efficiently absorbed. Individual lacking a lightabsorb ing photoreceptor for the co lor green will see a colorful fruit stand (left ) as mostly yellows (m iddle). [(Left and middle) From L. T. Sharpe, A. St ockman, H. )agle, and ). Nathans, Opsin genes, cone photopigments, color vision, and color bl indness. In Color Vision: from Genes to Perception, K. Gegenfurtner and L. T. Sharpe, Eds. (Cambrid ge University Press, 1999), pp. 3-51]
ur senses provide us with means for detecting a diverse set of external signals, often with incredible sensitivity and specificity. For example, when fully adapted to a darkened room, our eyes allow us to sense very low levels of light, down to a limit of less than 10 photons. With more light, we are able to distinguish millions of colors. Through our senses of smell and taste, we are able to detect thousands of chemicals in our environment and sort them into categories: pleasant or unpleasant? healthful or toxic? Finally, we can perceive mechanical stimuli in the air and around us through our senses of hearin g and touch. How do our sensory systems work? How are the initial stimuli detected? How are these initial biochemical events transformed into perceptions and experiences? W e have already encountered systems that sense and respond to chemical signals namely, receptors that bind to growth factors and hormones . Our knowledge of these receptors and their associated signaltransduction pathways provides us with concepts and tools for unraveling some of the workings of sensory system s. For example, 7TM receptors (seven-transmembrane receptors, Section 14 .1) play key roles in olfaction, taste, and vision. Ion channels that are sensitive to mechanical stress are essential for hearin g and touch.
Outline 32.1 A Wide Variety of Organic Compounds Are Detected by Olfaction 32.2 Taste Is a Combination of Senses That Function by Different Mechanisms 32.3 Photoreceptor Molecules in the Eye Detect Visible Light 32.4 Hearing Depends on the Speedy Detection of Mechanical Stimuli 32.5 Touch Includes the Sensing of Pressure, Temperature, and Other Factors
92 1
In this chapter, we focus on the five major sensory systems found in human beings and other mammals : olfaction (the sense of smell i.e., the detection of small molecules in the air), taste, or gustation (the detection of selected organic compound s and ions by the tongue), vision (the detection of light), hearing (the detection of sound, or pressure waves in the air), and touch (the detection of changes in pressure, temperature, and other factors by the skin). Each of these primary sensory systems contains specialized sensory neurons that transmit nerve impulses to the central nervous system (Figure 32.1). In the central nervous system, these signals are processed and combined with other information to yield a perception that may trigger a change in behavior. By these means, our senses al low us to detect changes in our environments and to adjust our behavior appropriately.
J
(
Vision
, Taste
32.1 A Wide Variety of Organic Compounds Are Detected by Olfaction Touch
Human beings can detect and distinguish thou sands of different compounds by smell , often w ith considerable sensitivity and specificity. Most odorants are small organic Figure 32.1 Sensory connections to the brain. Sensory nerves compounds with sufficient volatility that they can be carconnec t sensory o rgan s t o t he brain and spinal cord. ried as vapors into the nose. For example, a major component responsible for the smell of almonds is the simple aromatic compound benzaldehyde, whereas the sulfhydryl compound 3methylbutane-1-thiol is a major component of the smell of skunks.
o OH
Benzaldehyde (Almond)
o
H \'"
r---- CH,
H2 C
922
o
.
H,C~
H
CH 2
R-Carvone
S-Carvone
(Spearmint)
(Caraway)
3-Methylbutane-l-thiol (Skunk)
Geraniol
(Rose)
Zingiberene (Cinger)
What properties of these molecules are responsible for their smells' First, the shape of the molecule rather than its other physical properties is crucial. We can most clearly see the importance of shape by comparing molecules such as those responsible for the smells of spearmint and caraway. These compound s are identical in essentially all physical properties such as hydrophobicity because they are exact mirror images of one another. Thus, the smell produced by an odorant depends not on a physical property but on the compound's interaction with a specific binding surface, most likely a protein receptor. Second, some human beings (and other animals) suffer from specific anosmias; that is, they are incapable of smelling specific compounds even though their olfactory systems are otherwise normal. Such anosmias are often inherited . These observations suggest that mutations in individual receptor genes lead to the loss of the ability to detect a small subset of compounds.
To olfactory
bulb
923 32.1 Olfaction
Olfactory bulb Main olfactory epithelium Nasal cavity ---t Volatile - - - - -odorant compound
Sensory --+-'--+ neuron Cilia Mucous lining -t=-__________~
Figure 32.2 The main nasal epithelium . Th is reg ion of the nose. which lies at the top of the nasal cavi t y, contain s approximately 1 million sensory neurons. N erve impulses generated by odorant molecules binding to receptors on the cilia travel from the sensory neurons to th e olfactory bulb.
olfaction Is Mediated by an Enormous Family of Seven-Transmembrane-Helix Receptors
Odorants are detected in a specific region of the nose, called the main olfactory epithelium, that lies at the top of the nasal cavity (Figure 32.2). Approximately 1 million sensory neurons line the surface of this region. Cilia containing the odorant- binding protein receptors project from these neurons into the mucous lining of the nasal cavity. Biochemical studies in the late 1980s examined isolated cilia from rat olfactory epithelium that had been treated with odorants. Exposure to the odorants increased the cellular level of cyclic AMP, and this increase was observed only in the presence of GTP. On the basis of what was known about signal-transduction systems, the participation of cAMP and GTP strongly suggested the involvement of a G protein and, hence, TTM receptors. Indeed, Randall Reed purified and cloned a G-protein ex subunit, termed G(nl[J' which is uniquely expressed in olfactory cilia. The involvement of 7TM receptors suggested a strategy for identifying the olfactory receptors themselves. Complementary DNAs were sought that (1 ) were expressed primarily in the sensory neurons lining the nasal epithelium, (2) encoded members of the 7TM-receptor family, and (3) were present as a large and diverse fam ily to account for the range of odorants . Through the use of these criteria, cDNAs for odorant receptors from rats were identified in 1991 by Richard Axel and Linda Buck.
Y
The odorant receptor (hereafter, OR) family is even larger than expected: more than 1000 OR genes are present in the mouse and the rat, whereas the human genome encodes approximately 350 ORs. In addition, the human genome includes approximately 500 OR pseudogenes con taining mutations that prevent the generation of a fulllength , proper odorant receptor. The OR family is thus one of the largest gene families in human beings. Further analysis of primate OR genes reveals that the fraction of pseudo genes is greater in species more closely related to human beings (Figure 32.3). T hus, we may have a glimpse at the evolutionary loss of acuity in the sense of smell as higher mammals presumably became less dependent on
Figure 32.3 Evolution of odorant receptors. Odorant receptors appear to have lost function t hrough conversion into pseudogenes in the course of pr imate evolution. The percentage of OR genes that appea r to be functional for each species is shown in parentheses.
this sense for survival. For rodents that are highly dependent on their sense of smell, essentially all OR genes encode N functional proteins. The OR proteins are typically 20% identical in sequence with the f3-adrenergic receptor (Section 14.1) and from 30% to 60% identical with one another. Several specific sequence features are present in most or all OR family members (Figure 32.4). The central region, particularly transmembrane helices 4 and 5, is highly variable, suggesting that this region is the site of odorant binding. That site must be different in odorant receptors that bind distinct odorant molecules. What is the relation between OR gene expression and C the individual neuron? Interestingly, each olfactory neuron expresses only a single OR gene, among hundreds available. Figure 32.4 Conserved and variant regions in odorant receptors. Apparently, the precise OR gene expressed is determined Odorant receptors are members of t he 7TM -receptor fa mily. The green cyli nders represent th e seven presumed t ransmembrane largely at random_ After one OR gene is expressed and a helices. Stro ngly conse rve d residues charact erist ic o f th is protein functional OR protein is produced, t he expression of all fam ily are shown in blue, whereas highly va ria ble re sid ues are other OR genes is suppressed by a feedback mechanism shown in red. that remains to be fully elucidated. The binding of an odorant to an OR on the neuronal surface initiates a signal-transd uction cascade that results in an action potential (Figure 32.5). The ligand-bound OR activates G (elf ), the specific G protein mentioned earl ier. G (elf) is initially in its GDP-bound form. When activated, it releases GDP, binds GTP, and releases its associated f3'Y subunits. T he Ct subunit then activates a specifi c adenylate cyclase, increasi ng t he intracellu lar concentration of cAMP. The rise in the intracellular concentration of cAMP activates a nonspecific cation channel that allows calcium and other cations into the cell. The flow of cations through the channel depolarizes the neuronal membrane and initiates an action potential. This action potential, combined with those from other olfactory neurons, leads to the perception of a specific odor. odorants Are Decoded by a Combinatorial Mechanism
An obvious challenge presented to the investigator by the large size of the OR family is to match each OR with the one or m ore odorant molecules to which it binds _ Exciting progress has been made in this regard. rnitiall y, an OR was matched with odorants by overexpressing a single, specific OR gene in rats. This OR responded to straight -chain aldehydes, most favorably to n-octanal and less strongly to n-heptanal and n-hexanal. Moredramatic progress was made by taking advantage of our knowledge of the OR signal-transduction pathway and the power ofPCR (p _140). A section
Odorant
0 ...
Adenylate cyclase
Receptor Figure 32.5 The olfactory signal transduction cascade. The binding o f odorant to the ol factory recept or acti vates a Signaling pathway sim ilar to those init iated in response to th e bind ing of some hormones to th eir receptors. The f inal result is the o pen ing of cAMP-gated ion cha nnels and the init iati o n of an action pot ential.
924
I "\;) GTP GDP
ATP
Na ' Ca 2 +
cAMP-gated ion channel
o
925 32.1 Olfaction OH
carboxylic acids (i ~ 2- 7)
Alcohols (i ~ 4- 8)
o Br •
J
OH
Bromocarboxylic acids (i ~ 3-7)
HO
OH
Dicarboxylic acids (i ~ 4-7)
Figure 32.6 Four series of odorants tested for olfactory-receptor activation. Receptor
of nasal epithelium from a mouse was loaded with the calI 234 5 6 789 III 4 cium-sensitive dye Fura-2 (p . 389). The tissue was then treated with different odorants, one at a time, at a specific concentration. If the odorant had bound to an OR and activated it, that neuron could be detected und er a microscope by the change in fluorescence caused by the influx of calcium that occurs as part of the signal -transduction process. To determ ine which OR was responsible for the response, cDNA was generated from mRNA that had been isolated from single identified neurons. T he cDNA c was then subjected to PCR with the use of primers that are '" o effective in amplifying most or all OR genes . The se'"o quence of the PCR product from each neuron was then determined and analyzed. Using this approach, investigators analyzed the responses of neurons to a series of compounds having varyi.ng chain lengths and terminal functional groups (Figure 32.6) . The results of these experiments appear surprising HOOC-C4 -COOH HOOC- -+-+-+-+-+- - - --t-+----to at first glance (Figure 32.7). importantly, there is not a si mple 1:1 correspondence between odorants and receptors. Almost every odorant activates a number of receptors (usually to different extents) and almost every receptor is Figure 32.7 Patterns of olfactory-receptor activation. Fourteen activated by more than one odorant. Note, however, that different receptors were tested for responsiveness to the each odorant activates a unique combination of receptors. compounds shown in Figure 32.6. A colored box indicates that In principle, this combinatorial mechanism allows even a the receptor at the top responded to the compound at the left. small array of receptors to distinguish a vast number of Darker colors indicate that the receptor was activated at a lower concentration of odorant. odorants. How is the information about wh ich receptors have been activated transm itted to the brain? Recall that each neuron expresses only one OR and that the pattern of expression appears to be largely random. A substantial clue to the connections between receptors and the brain has been provided by the creation of mice that express a gene for an easily detectable colored marker in conjunction with a specific OR gene. Olfactory neurons that express the OR- marker-protein combination were traced to their destination in the brain, a structure called the olfactory bulb (Figure 32 .8). The processes from neurons that express the same OR gene were found to connect to the same location in the olfactory bulb. Moreover, this pattern of neuronal connection was found to be identical in
-
Figure 32.8 Con verging olfactory neurons. This section of the nasa l cavity is stained to reveal processes from sensory neurons expressing the same olfactory receptor. The processes converge to a single location in the olfactory bulb. [From P. Mombaerts, F. Wang, C. Dulac. S. K. Chao. A. Nemes. M . Mendelsohn. J. Edmondson. and R. Axel. Cell
87(1996):675-689.]
all mice examined. Thus, neurons that express specific ORs are linked to spe· cific sites in the brain. This property creates a spatial map of odorant· responsive neuronal activity within the olfactory bulb. Can such a combinatorial mechanism truly distinguish many different odorants? An electronic "nose" that functions by the same principles provides compelling evidence that it can (Figure 32.9). The receptors for the electronic nose are polymers that bind a range of small molecules. Each polymer binds every odorant, but to varying degrees. Importantly, the electrical properties of these polymers change on odorant binding. A set of 32 of these polymer sensors, wired together so that the pattern of responses can be evaluated, is capable of distinguishing individual compounds such as n-pentane and n-hexane as well as complex mixtures such as the odors of fresh and spoiled fruit. Functional Magnetic Resonance Imaging Reveals Regions of the Brain Processing Sensory Information
Figure 32.9 The Cyranose 320. The electronic nose may f ind uses in the food industry. animal husbandry. law enforcement, and medicine. [Courtesy o f Cyrano Sciences.]
Figure 32.10 Brain response to odorants. A functi o nal magneti c resonance image reveal s brain response to odorants. The light spots indicate regio ns of the brain activated by odorants. [Fro m N. Sobel et al..}. Neurophysiol. 83(2000):537- 551 ; courtesy of Dr. Noam SobeL]
Can we extend our understanding of how odorants are perceived to events in the brain? Biochemistry has provided the basis for powerful methods for examining responses within the brain. One method, functional magnetic reso· nance imaging (fMRI), takes advantage of two key observations. The first is that, when a specific part of the brain is active, blood vessels relax to allow more blood flow to the active region. Thus, a more active region of the brain will be richer in oxyhemoglobin. The second observation is that the iron center in hemoglobin undergoes substantial structural changes on binding oxygen (p. 185). These changes are associated with a rearrangement of electrons such that the iron in deoxyhemoglobin acts as a strong magnet, whereas the iron in oxyhemoglobin does not. The difference between the magnetic properties of these two forms of hemoglobin can be used to image brain activity. Nuclear magnetic resonance techniques (p. 98) detect signals that originate primarily from the protons in water molecules but are altered by the magnetic properties of hemoglobin. With the use of appropriate techniques, images can be generated that reveal differences in the relative amounts of deoxy- and oxyhemoglobin and thus the relative activity of various parts of the brain. These noninvasive methods reveal areas of the brain that process sensory information. For example, subjects have been imaged while breathing air that either does or does not contain odorants. When odorants are present, the fMRI technique detects an increase in the level of hemoglobin oxygenation (and, hence, brain activity) in several regions of the brain (Figure 32 .10). Such regions include those in the primary olfactory cortex as well as other regions in which secondary processing of olfactory signals presumably takes place. Further analysis reveals the time course of activation of particular regions and other features. Functional MRI shows tremendous potential for mapping regions and pathways engaged in processing sensory information obtained from all the senses. Thus, a seemingly incidental aspect of the biochemistry of hemoglobin has yielded the basis for observing the brain in action.
32.2
Taste Is a Combination of Senses That Function Different Mechanisms
by
The inability to taste food is a common complaint when nasal congestion reo duces the sense of smell. Thus, smell greatly augments our sense of taste (also known as gustation), and taste is, in many ways, the sister sense to 926
927 -
0'
32,2 Taste
,0
.;/
" -'c
H
H' ~~
HO
OH ..-
HO/ HO
,.'
0
"'\' OH Glucose
(sweel)
+ H] N
OH Na+
Sodium ion (salty)
.H
N
';
0 /..0 C
H3 C/
Ii0
Glutamate (umami)
Quinine
Hydrogen ion
(bitter)
(sour)
Figure 32.11 Examples of tastant molecules, Tastants fall into five groups: sweet, sa lty, umami , bitter, and sour,
olfaction , Nevertheless, the two senses differ from each other in several important ways, First, we are able to sense several classes of compounds by taste that we are unable to detect by smell; salt and sugar have very little odor, yet they are primary stimuli of the gustatory system, Second, whereas we are able to discriminate thousands of odorants, discrimination by taste is much more modest, Five primary tastes are perceived: bitter, sweet, sour, salty, and umami (the taste of glutamate and aspartate from the Japanese word for "deliciousness"), These five tastes serve to classify compounds into potentially nutritive and beneficial (sweet, salty, umami) or potentially harmful or toxic (bitter, sour), Tastants (the molecules sensed by taste) are quite distinct for the different groups (Figure 32 ,1 1), The simplest tastant, the hydrogen ion, is perceived as SOUL Other simple ions, particularly sodium ion, are perceived as salty, The taste called umami is evoked by the amino acids glutamate and aspartate, the former often encountered as the flavor enhancer monosodium glutamate (MSG), In contrast, tastants perceived as sweet and, particularly, bitter are extremely diverse, Many bitter compounds are alkaloids or other plant products of which many are toxic. However, they do not have any common structural elements or other common properties, Carbohydrates such as glucose and sucrose are perceived as sweet, as are other compounds including some simple peptide derivatives, such as aspartame, and even some proteins, These differences in specificity among the five tastes are due to differences in their underlying biochemical mechanisms, The sense of taste is, in fact, a number of independent senses all utilizing the same organ, the tongue, for their expression, Tastants are detected by specialized structures called taste buds, which contain approximately 150 cells, including sensory neurons (Figure 32,12) , Fingerlike projections called microvilli, which are rich in taste receptors, project from one end of each sensory neuron to the surface of the tongue, Nerve fibers at the opposite end of each neuron carry electrical impulses to the brain in response to stimulation by tastants, Structures called taste papil lae contain numerous taste buds,
Sequencing of the Human Genome Led to the Discovery of a Large Family of 7TM Bitter Receptors
Just as in olfaction, a number of clues pointed to the involvement of G proteins and, hence, 7TM receptors in the detection of bitter and sweet tastes, The evidence included the isolation of a specific G -protein ex subunit
Sensory neuron
...- Microvilli
containing receptors
Nerve fiber
Figure 32.12 A taste bud, Each taste bud contains sensory neuro ns that extend microvilli to the surface of the tongue, where they interact with tastants,
(A)
928
(8)
CHAPTER 32 Sensory Systems
Figure 32.13 Expression of gustducin in the tongue. (A) A secti on o f t ongue stained with a flu orescent antibo dy reveals th e positio n o f t he taste buds. (B) The same region stained with an ant ibo dy directed against gustducin reveal s that thi s G pro tein is expressed in t aste buds. [Courtesy of Dr. Charles S. Zuker.)
o HN 5
N H
6-n-Propyl-2-thiouraci I (PROP)
termed gustducin, which is expressed primarily in taste buds (Figure 32.13). How could the 7TM receptors be identified? The ability to detect some compounds depends on specific genetic loci in both human beings and mice. For instance, the ability to taste the bitter compound 6-n-propyl -2thiouracil ( PROP) was mapped to a region on human chromosome 5 by comparing DNA markers of persons who vary in sensitivity to this compound. This observation suggested that thi s region might encode a 7TM receptor that responded to PROP. Approximately 450 kilobases in this region had been sequenced early in the human genome project. This sequence was searched by computer for potential 7TM -receptor genes, and, indeed, one was detected and named T2Rl . Additional database searches detected approximately 30 sequences similar to T2Rl in the human genome. The encoded proteins are between 30 and 70% identical with T2Rl (Figure 32 .1 4). Are these proteins, in fact, bitter receptors? Several lines of evidence suggest that they are. First, their genes are expressed in taste-sensitive cells in fact, in many of the same cells that express gustducin. Second, cells that express individual members of this family respond to specific bitter compounds. For example, cells that express a specific mouse receptor (mT2RS ) responded when exposed specifically to cycloheximide. Third, mice that had been found unresponsive to cycloheximide were found to have point mutations in the gene encoding mT2R5 . Finally, cycloheximide
Figure 32.14 Conserved and variant regions in bitter receptors. The bitter receptors are members o f th e 7TM-receptor family. Stro ngly conserved residues characteri st ic of t his protein family are sho wn in blue, and highly variable residues are shown in red.
specifically stimulates the binding ofGTP analogs to gustducin in the presence ofthe mT2R5 protein (Figure 32 .15). Tmportantly, each taste-receptor cell expresses many different members of the T2R family. This pattern of expression stands in sharp contrast to the pattern of one receptor type per cell that characterizes the olfactory system (Figure 32.16). The difference in expression patterns accounts for the much greater specificity of our perceptions of smells compared with tastes. We are able to distinguish among subtly different odors because each odorant stimulates a unique pattern of neurons. In contrast, many tastants stimulate the same neu rons. Thus, we perceive only "bitter" without t he ability to discriminate cycloheximide from qui nine. OLFACTION
o
c:
::l
o E ..:
TASTE (bitter)
Figure 32.15 Evidence that T2R proteins are bitter taste receptors. Cyclo heximide uniquely stimulates the binding of the GTP analog GTP,,/S to gustducin in the presence of the mT2R protein. [After J. Chandrashekar. K. L. Mueller. M . A. Hoon, E. Adler, L. Feng, W. Guo, C. S. Zuker, and N. J. Ryba. Cell 100(2000):703- 711.]
•• ••• •
Sensory neurons
--
Brain
Sensory neurons
Brain
Figure 32.16 Differing gene expression and connection patterns in olfactory and bitter taste receptors. In olfaction, each neuron expresses a si ngle OR gene. and the neurons expressing the same OR converge to specific sites in the brain. enabling specific perception of different odorants. In gustation. each neuron expresses many bitter receptor genes, and so the identity of the tastant is lost in transmission.
A Heterodimeric 7TM Receptor Responds to Sweet Compounds Most sweet compounds are carbohydrates, energy rich and easil y digestible. Some noncarbohydrate compounds such as saccharin and aspartame also taste sweet. Members of a second family of 7T M receptors are expressed in taste-receptor cells sensitive to sweetness. T he three members of this family, referred to as TIR1, Tl R2, and Tl R3, are distinguished by their large extracellular domains compared with those of the bitter receptors. Studies in knockout mice have revealed that T1 R2 and T1R3 are expressed simultaneously in mice able to taste carbohydrates (Figure 32 .17). Thus, it ap pears t hat T1 R2 and T1 R3 form a specific heterodimeric receptor that is responsible for mediating the response to sugars. This heterodimeric receptor also responds to artificial sweeteners and to sweet-tasting proteins and therefore appears to be the receptor responsible for responses to all sweet tastants . Note thatT1R2 and T1R3 do respond to sweet tastants individually, but only at very high concentrations of tastant. The requirement for an oligomeric 7TM receptor for a fully functional response is surprising, considering our previous understanding of 7TM re ceptors. This discovery has at least two possible explanations. First, the sweet receptor could be a member of a small su bset of the 7TM-receptor family t hat functions well only as oligomers. Alternatively, many 7TM 929
930
.
Tl R2. Tl R3
. .... I .. .'
TlRl . TlR3
CHAPTER 32 Sensory Systems
• .' ' •
.-
• • • • •
"'C
~
•• ••
'"E
0..
,8, -
••
• • •
10
••
.--
Q)
--'"
> .-
..
.' • •
. -.- 6. _____
Q)
'"
'
.' .' •
o
• • •• ••
••••• • • • • • •
.'
,
• ••• ..-- -
::::::::::: . ::::::::::: ___________ 30
100
300
T1 Rl . TlR2 T1Rlonly
1000
[Sucrosel. mM Figure 32.17 Evidence for a heterodimeric sweet receptor. The sensitivity to sweetness of mice with genes for either T1R1 . T1R2, T1R3, or both T1R2 and T1R3 were determined by observing the relative rates at which they licked solutions containing various amount of sucrose. These studies revealed that both T 1R2 and T1R3 were required for a full response to sucrose. Mice with a disrupted T1R1 gene were indistinguishable from wild-type mice in this assay (not shown). [After G. Q . Zhao, Y. Zhang, M . A. Hoon. J. Chandrashekar. I. Erlenbach, N. J. P. Ryba, and C. S. Zuker. Cell 115(2003):255- 266.]
receptors may function as oligomers. but this notion is not clear. because these oligomers contain only one type of 7TM-receptor subunit. Further studies will be required to determine which of these explanations is correct.
Umami. t he Taste of Glutamate and Aspartate. Is Mediated by a Heterod imeric Receptor Related to t he Sweet Recept or The family of receptors responsible for detecting sweetness is also responsible for detecting amino acids. In human beings. only glutamate and aspartate elicit a taste response. Studies similar to those for the sweet receptor revealed that the umami receptor consists ofT1Rl and T1R3. Thus. this receptor has one subunit (T1R3) in common with the sweet receptor but has an additional subunit (T1Rl) that does not participate in the sweet response. This observation is supported by the observation that mice in which the gene for Tl R 1 is disrupted do not respond to aspartate but do respond normally to sweet tastants; mice having disrupted genes for both T1Rl and T1R3 respond poorly to both umami and sweet tastants.
Salt y Tast es Are Detected Primarily by t he Passage of Sodium Ions Through Channels 0
CI ~
H2N
NH2 , , ,•
N
•
N H
"'"
N
+
NH2 Amiloride
•
-NH 2
Salty tastants are not detected by 7TM receptors. Rather. they are detected directly by their passage through ion channels expressed on the surface of cells in the tongue. Evidence for the role of these ion channels comes from examining known properties ofNa + channels characterized in other biological contexts. One class of channels. characterized first for their role in salt reabsorption. are thought to be important in the detection of salty tastes because they are sensitive to the compound amiloride. which mutes the taste of salt and significantly lowers sensory-neuron activation in response to sodium. An amiloride-sensitive Na + channel comprises four subunits that may be either identical or d istinct but in any case are homologous. An individual su bunit ranges in length from 500 to 1000 amino acids and includes two
presumed membrane-spanning helices as well as a large extracellular domain in between them (Figure 32.18). The extracellular region includes two (or, sometimes, three) distinct regions rich in cysteine residues (and, presumably, disulfide bonds). A region just ahead of the second membrane-spanning helix appears to form part of the pore in a manner analogous to the structurally characterized potassium channel (p. 364). The members of the amiloridesensitive Na + -channel family are numerous and diverse in their biological roles. We shall encounter them again in the context of the sense of touch. Sodium ions passing through these channels produce a significant transmembrane current. Amiloride blocks this current, accounting for its effect on taste. However, about 20% of the response to sodium remains even in the presence of amiloride, suggesting that other ion channels also contribute to salt detection.
Sour Tastes Arise from t he Effects of Hydrogen Ions (Acids) on Channels Like salty tastes, sour tastes are detected by direct interactions with ion channels, but the incoming ions are hydrogen ions (in high concentrations) rather than sodium ions. For example, in th e absence of high concentrations of sodium, hydrogen ion flow can induce substantial transmembrane currents through amiloride-sensitive Na + channels. However, hydrogen ions are also sensed by mechanisms other than their direct passage through membranes. Binding by hydrogen ions blocks some potassium ion channels and activates other types of channels. Together, these mechanisms lead to changes in membrane polarization in sensory neurons that produce the sensation of sour taste. We shall consider an additional receptor related to taste, one responsible for the "hot" taste of spicy food, when we examine mechanisms of touch perception.
32.3
Cysteine-ri ch region 2
Cysteine-rich region 1
_ ~
o a.
N
Membrane• spannong helices
c
Figure 32.18 Schematic structure of the amiloride-sensitive sodium channel. Only one of the four subunits that constitute the functional channel is illustrated. The amiloride-sensitive sodium channel belongs to a superfamily having common structural features, including two hydrophobic membrane-spanning regions, intracellular amino and carboxyl term ini; and a large, extracellular region with conserved cysteine-rich domains.
Photoreceptor Molecules in the Eye Detect Visible Light
Vision is based on the absorption of light by photoreceptor cells in the eye. These cells are sensitive to light in a narrow region of the electromagnetic spectrum, the region with wavelengths between 300 and 850 nm (Figure 32.19). Vertebrates have two kinds of photoreceptor cells, called rods and cones because of their distinctive shapes. Cones function in bright light and are responsible for color vision, whereas rods function in dim light but do not perceive color. A human retina contains about 3 million cones and 100 million rods. Remarkably, a rod cell can respond to a single photon, and the brain requires fewer than 10 such responses to register the sensation of a flash of light.
I
I
I
I:!! ~
I
I
I
I
I
I
I
I
10- 10
10- 7 10-"
102
X-rays
Visible light
Radio waves
Wavelength (m) Figure 32.19 The electromagnetic spectrum. Visible light has wavelengths between 300 and 850 nm.
931
932
Rhodopsin, a Specialized 7TM Receptor, Absorbs Visible Light
CHAPTER 32 Sensory Systems
Rod s are slender, elongated structures; th e outer segment is speciali zed for photoreception (Figure 32 .20). It contains a stack of about 1000 discs, whi ch are membrane-enclosed sacs densely packed with photoreceptor molecul es. The photosensitive molecule is often call ed a visual pigment because it is hi ghl y colored owing to its ability to absorb light. T he photoreceptor molecul e in rod s is rhodopsin (Section 14.1), which consists of the protein opsin linked to 77 -cis-retinal , a prosthetic group .
.'
l1 -cis-Retinal
H
==
-
..-I- Discs
Outer segment
Figure 32.20 The rod cell . (Left) Scanning electron m icrograph o f retinal rod cells. (Right) Schematic representati on o f a ro d cel l. [ Pho t ograph courtesy of Dr. Deric Bo wnds.]
~
,
40,000
500
E u
-!;;
30,000
E
.-'u"
it: o u
20,000
'"
g .w c:
10,000 ~
.-
:E
~o""'" o - - :4.1. 50""'0"- - :6,-'00,,----= -..J 700 00=----=Wavelength (nm)
Figure 32.21 Rhodopsin absorption spectrum. Almost all photons wi th wavelengths near 500 nm that st ri ke a rhod opSin mo lecule are absorbed .
Rhodopsin absorbs light very effi cientl y in the middle of the visible spectrum, its absorption being centered on 500 nm , which nicely matches the solar output (Figure 32.2 1). A rhodopsin molecul e will absorb a high percentage of the photons of the correct wavelength that strike it, as indicated by the extinction coefficient of 40 ,000 M - 1cm - I at 500 nm . The extin ction coefficient for rhodopsin is more than an order of magni tude greater th an that 'for tryptophan , the most effi cient absorber in proteins that lack prostheti c groups. Opsin , the protein component of rhod opsin, is a member of the 7TMreceptor family. Indeed, rhodopsin was the first member of this famil y to be purified, its gene was the first to be cloned and sequenced , and its threedimensional structure was the first to be determined. The color of rhodopsin and its responsiveness to light depend on the presence of the light-absorbing group (chromophore) 11-cis-retinal. This compound is a powerful absorber of light because it is a polyene; its six alternating single and double bonds constitute a long, unsaturated electron network. Recall that alternating single and doubl e bonds account for the chromophoric properties of chlorophyll (Section 19 .2). The aldehyde group of II -cis-retinal forms a Schiff base (Figure 32.22) with th e e -amino group ofly sine residue 296, which lies in the center of the seventh transmembrane helix. F ree retinal absorbs maximally at 370 nm, and its unprotonated Schiff-base adduct absorbs at 380 nm, whereas the protonated Schiff base absorbs at 440 nm or longer wave-
Schiff base
Protonated Schiff base
f
.'
\
)
H (l1-cis-Retina l)
~~/~/~'-..---/
~
lysine
Figure 32 .22 Retinal- lysine linkage. Retinal is linked to lysine 296 in opsin by a Schiff-base linkage. In the resting state of rhodopsin, this Schiff base is protonated.
lengths. Thus, the 500-nm absorption maximum for rhodopsin strongly suggests that the Schiff base is protonated; additional interactions with opsin shift the absorption maximum farther toward the red. The positive charge of the protonated Schiff base is compensated by the negative charge of glu tamate 113 located in helix 2; the glutamate residue closely approaches the lysine- retinal linkage in the three-dimensional structure of rhodopsin. Light Absorption Induces a Specific Isomerization of Bound ll-cis-Retinal
How does the absorption oflight by the retinal Schiff base generate a signal? George Wald and his coworkers discovered that light absurptiun results in the isomerization of the ll -cis-retinal group uf rhudopsin to its a ll-trans form (Figure 32.23). This isomerization causes the Schiff-base nitrogen atom to move approximately 5 A, assuming that the cyclohexane ring of the retinal group remains fixed . In essence, the light energy of a photon is converted into atomic mution. The change in atomic positions, like the binding of a ligand to other 7TM receptors, sets in train a series of events that lead to the closing of ion channels and the generation of a nerve impulse. The isomerization of the retinal Schiff base takes place within a few picoseconds of a photon being absorbed . The initial product, termed bathorhodopsin, contains a strained all-trans -retinal group. Within approximately 1 ms, this intermediate is converted through several additional intermediates into metarhodupsin II. In metarhodopsin II, the Schiff base is deprotonated and the opsin protein has undergone significant reorganization . Metarhodopsin II (also referred to as R*) is analogous to the ligand-bound state of 7TM receptors such as the f3 r adrenergic receptor (Section 14.1) and
Light
lys
)
sA l1 -cis-Retinal
AII-trons-retinal
• Figure 32.23 Atomic motion in retinal. The Schiff-base nitrogen atom moves 5 A as a consequence of the light-induced iso merization of ll-cis-retinal to all-trans-ret inal by rotation about the bo nd shown in red.
933
o
\.
)
Ligand-bound 7TM receptor
the odorant and tastant receptors discussed previously (Figure 32.24). Like these receptors, this form of rhodopsin activates a heterotrimeric G protein that propagates the signal. The G protein associated with rhodopsin is called transducin . Metarhodopsin II triggers the exchange of GDP for GTP by the ex subunit of transducin (Figure 32. 25). On the binding ofGTP, the J3-y subunits of transducin are released and the ex subunit switches on a cGMP phosphodiesterase by binding to an inhibitory subunit and removing it. The activated phosphodiesterase is a potent enzyme that rapidly hydrolyzes cGMP to GMP. The reduction in cGMP concentration causes cGMP-gated ion channels to close, leading to the hyperpolarization of the membrane and neuronal signaling. At each step in this process, the initial signal the absorption of a single photon is amplified so that it leads to sufficient membrane
hyperpolarization to result in signaling.
Light-Induced Lowering of the Calcium Level Coordinates Recovery
Light )
Metarhodopsin II Figure 32.24 Analogous 7TM receptors. The conversion of rhodopsin into metarhodopsin II activates a signaltransduction pathway analogously to the activation induced by the binding of other lTM receptors to appropriate ligands.
As we have seen, the visual system responds to changes in light and color within a few milliseconds, quickly enough that we are able to perceive continuous motion at nearly 1000 frames per second. To achieve a rapid response, the signal must also be terminated rapidly and the system must be returned to its initial state. First, activated rhodopsin must be blocked from continuing to activate transducin. Rhodopsin kinase catalyzes the phosphorylation of the carboxyl terminus of R- at multiple serine and threonine residues. AYTestin, an inhibitory protein (p. 388), then binds phosphorylated R ' and prevents additional interaction with transducin. Second, the ex subunit of transducin must be returned to its inactive state to prevent further signaling. Like other G proteins, the ex subunit possesses built-in GTPase activity that hydrolyzes bound GTP to GDP. Hydrolysis takes place in less than a second when transducin is bound to the phosphodiesterase. The GDP form of transducin then leaves the phosphodiesterase and reassociates with the J3-y subunits, and the phosphodiesterase returns to its inactive state. Third, the level of cGMP must be raised to reopen the cGMP-gated ion channels. The action of guanylate cyclase accomplishes this
third step by synthesizing cGMP from GTP. Calcium ion plays an essenti al role in controlling guanylate cyclase because it markedly inhibits the activity of the enzyme. In the dark, Ca2+ as well as Na + enter the rod outer segment through the cGMP-gated channels. Calcium ion influx is balanced by its efflux through an exchanger, a
Light
Phosphodiesterase
Rhodopsin
Transducin
GTP
cGMP-gated ion channel
GDP
GMP
cGMP
Figure 32.25 Visual signal transduction. The light-induced activation of rhodo psin leads to the hydrolysis of cGMP, which in turn leads t o ion-channel cl osing and the in itiati on of an action potential.
934
935
transport system that uses the thermodynamically favorable flow of four 2 Na + ions into the cell and one K + ion out of the cell to extrude one Ca + ion. 2 After illumination, the entry of Ca + through the cGMP-gated channels stops, but its export through the exchanger continues. Thus, the cytoplas2 mic Ca + level drops from 500 nM to SO nM after illumination. This drop markedly stimulates guanylate cyclase, rapidly restoring the concentration of cGMP to reopen the cGMP-gated channels. Activation
32.3 Vision
Recovery
Ion Guanylate [cGM PJ-.l.. ~) channels ~) [Ca2+}J- ~) cyclase ~) [cGMP]t • • activIty closed increased By controlling the rate of cGMP synthesis, Ca2 + levels govern the speed with which the system is restored to its initial state.
Color Vision Is Mediated by Three Cone Receptors That Are Homologs of Rhodopsin Cone cells, like rod cells, contain visual pigments. Like rhodopsin, these photoreceptor proteins are members of the 7TM-receptor family and use II -cis-retinal as their chromophore. In human cone cells, there are three distinct photoreceptor proteins with absorption maxima at 426, 530, and - 560 nm (Figure 32 .2 6). These absorbances correspond to (in fact, define) the blue, green, and red regions of the spectrum. Recall that the absorption maximum for rhodopsin is 500 nm. The amino acid sequences of the cone photoreceptors have been compared with one another and with rhodopsin. The result is striking. Each of the cone photoreceptors is approximately 40% identical in sequence with rhodopsin. Similarly, the blue photoreceptor is 40% identical with each of the green and red photoreceptors. The green and red photoreceptors, however, are > 95 % identical with one another, differing in only 15 of 364 positions (Figure 32. 27).
N
c Figure 32.27 Comparison of the amino acid sequences of the green and red photoreceptors. Open circles correspond to identical residues, whereas colored circles mark residues that are different. The differences in the three black positions are respo nSible for most of the difference in their absorption spectra.
426
530 560
t
300
400
SOO
600
700
800
Wavelength (nm) Figure 32.26 Cone-pigment absorption spectra. The abso rpti o n spectra o f the cone visual pigment responsibl e for color •
•
VISio n .
936 CHAPTER 32
Chicken
Hum"n being
Mo use
Sensory Systems
600
Red Red
550
Green Figure 32 .28 Evolutionary relationships among visual pigments. Visual pigments have evolved by gene duplication along different branches of the animal evolutionary tree. The branch lengths of th e "trees" correspo nd to the percentage of amino acid divergence. [After J. Nathans. Neuron 24(1999):299- 312; by permission of Cell Press. ]
Green Rhodopsin I
Rhodopsin
~
Green Rhodopsin
500
" -"
~
-"
DO
Pinopsin Blue Violet
E
450
-'" ~
~
Blue 400
Blue
350
These observations are sources of insight into photoreceptor evolution. First, the green and red photoreceptors are clear! y products of a recent evolutionary event (Figure 32.28). The green and red pigments appear to have diverged in the primate lineage approximately 35 million years ago. Mammals, such as dogs and mice, that diverged from primates earlier have only two cone photoreceptors, blue and green. They are not sensitive to light as far toward the infrared region as we are, and they do not di scriminate colors as well. In contrast, birds such as chickens have a total of six pigments: rhodopsin, four cone pigments, and a pineal visual pigment called pinopsin. Birds have highly acute color perception. Second, the high level of similarity between the green and red pigments has made it possible to identify the specific amino acid residues that are responsible for spectral tuning. T hree residues (at positions 180, 277, and 285) are responsible for most of the difference between the green and the red pigments. In the green pigment, t hese residues are alanine, phenylalanine, and alanine, respectively; in the red pigment, they are serine, tyrosine, and threonine. A hydroxyl grou p has been added to each am ino acid in the red pigment. The hydroxyl grou ps can interact with the photoexcited state of retinal and lower its energy, leading to a shift toward the lower-energy (red) region of the spectrum. Rearrangements in the Genes for the Green and Red Pigments Lead to "Color Blindness" Homologous recombination
The exchange of DNA segments at equivalent pOSition s between chromosomes wi th
substantial sequence similarity.
The genes for the green and red pigments lie adjacent to each other on the human X chromosome. These genes are more than 98% identical in nucleotide sequence, including introns and untranslated regions as well as the protein-coding region. Regions with such high similarity are very susceptible to unequal homologous recombination. Recombination can take place either between or within transcribed regions of the gene (Figure 32.29). If recombination takes place between transcribed regions, the product chromosomes will differ in the number of pigment genes that they carry. One chromosome will lose a gene and thus may lack the gene for, say, the green pigment; the other chromosome will gain a gene. Consistent with this scenario, approximately 2% of human X chromosomes carry only a single color pigment gene, approx imately 20% carry two, 50% carry three, 20% carry four, and 5% carry five or more. A person lacking the gene for the green pigment will have trouble distinguishing red and green color, characteristic of the most common form of color blindness. Approximately 5% of males have this form of color blindness. Recombination can also take place within the transcription units, resulting in genes that encode hybrids of the green and red photoreceptors. The absorption maximum
(A) Recombination between genes
)
(8) Recombination within genes
Greenlike
)
Redlike hybrid
Figure 32.29 Recombination pathways leading to color blindness. Rearrangements in the course of DNA replication may lead to (A) the loss of visual pigment genes or (B) the formation o f hybrid pigment genes that encode photoreceptors with anomolous absorption spectra. Because the amino acids most important for determining absorption spectra are in the carboxyl- terminal half o f each photoreceptor protein, t he part of the gene that encodes this region most strongly affects the absorption characteristics of hy brid receptors. [After J. Nathans. Neuron 24(1999):299-312; by permission of Cell Press.]
of such a hybrid lies between that of th e red and green pigments. A person with such hybrid genes who also lacks either a functional red or a functional green pigment gene does not discriminate color well.
32.4
Hair cell
Hearing Depends on the Speedy Detection of Mechanical Stimuli
Hearing and touch are based on the detection of m echani cal stimuli. Although the proteins of these senses have not been as well characterized as those of the senses already di scussed, anatomical, physiological, and biophysical studies have elucidated the fundamental processes. A majur clue to the mechanism of hearing is its speed. We hear frequencies ranging from 200 to 20,000 Hz (cycles per second), corresponding to times of 5 to 0.05 ms. Furthermore, our ability to locate sound sources, one of the m ost important funct ions of hearing, depends on the ability to detect the time delay between the arrival of a sound at one ear and its arrival at the other. G iven the separation of our ears and the speed of so und , we must be able to accurately sense time differences of 0.7 m s. In fact, human beings can locate sound sources associated with temporal delays as short as 0.02 ms. T his high time resolution implies that hearing must employ direct transduction mechaIlisms that do not depend on second messengers. Recall that, in vision, for which speed also is important, the signal -transduction processes take place in milliseconds.
Figure 32.30 Hair cells, the sensory neurons crucial for hearing. These speCialized neurons are capped with hairlike projection called stereocilia that are responsib le for detecting very subtle vibrations. [After A. J. Hudspeth. Nature
341(1 989):397- 404.]
Hair Cells Use a Connected Bundle of Stereocilia to Detect Tiny Motions
Sound waves are detected inside th e cochl ea of the inner ear. The cuchlea is a fl uid -filled, membranous sac that is coiled like a snail shell. The primary detection is accomplished by specialized neurons inside the cochlea called hair cells (Figure 32 .30). Each cochlea contains approximately 16,000 hair cell s, and each hair cell contains a hexagonally shaped bundle of20 to 300 hairlike projections called stereocilia (Figure 32 .31). These stereocilia are graded in length across the bundle . Mechanical deflection of the hair bundle, as occurs
Figure 32.31 An electron micrograph of a hair bundle. [Courtesy o f Dr. A. Jacobs and Dr. A. J. Hudspeth.]
937
open channels close, and the membrane hyperpolarizes. Thus, the mechanical motion of the hair bundle is directly converted into current flow across the hair-cell membrane.
939 32.5 Touch
Mechanosensory Channels Have Been Identified in Drosophila and Vertebrates
The search for ion channels that respond to mechanical impulses has been pursued in a variety of organisms. Drosophila have sensory bristles used for detecting small air currents. These bristles respond to mechanical displacement in ways similar to those of hair cells; displacement of a bristle in one direction leads to substantial transmembrane current. Strains of mutant fruit flies that show uncoordinated motion and clumsiness have been examined for their electrophysiological responses to displacement of the sensory bristles. In one set of strains, transmembrane currents were dramatically reduced . The mutated gene in these strains was found to encode a protein of 1619 amino acids, called NompC for no mechanoreceptor potential. The carboxyl-terminal 469 amino acids of NompC resemble a class of ion channel proteins called TRP (transient receptor potential) channels. This region includes six putative transmembrane helices with a porelike region between the fifth and sixth helices. The amino-terminal 1150 amino acids consist almost exclusively of 29 ankyrin repeats (Figure 32.35). Ankyrin repeats are structural motifs consisting of a hairpin loop followed by a helix-tum -helix. Importantly, in other proteins, regions with tandem arrays of these motifs mediate protein- protein interactions, suggesting that these arrays couple the motions of other proteins to the activity of the N ompC channel. Figure 32.35 Ankyrin repeat structure. One ankyrin do main is Recently, a strong candidate for at least one composhown in red in thi s series of four ankyrin repeat s. Notice the nent of the mechanosensory channel involved in hearhairpin loop followed by a helix-turn-helix moti f in the redcolo red ankyrin unit. Ankyri n do mains interact w ith other pro teins, ing has been identified. The protein, TRPA1, is also a pri marily through their loops. [Drawn f ro m lAWC.pdb.] member of the TRP channel family. The sequence of TRPA1 also includes 17 ankyrin repeats. TRPAl is expressed in hair cells, particularly near their tips . Based on these and other studies, it appears very likely that TRPAl represents at least one component of the mechanosensory channel that is central to hearing. Further studies are under way to confirm and extend this exciting discovery.
32.5
Touch Includes the Sensing of Pressure, Temperature, and Other Factors
Like taste, touch is a combination of sensory systems that are expressed in a common organ in this case, the skin. The detection of pressure and the detection of temperature are two key components. Amiloride-sensitive Na + channels, homologous to those of taste, appear to playa role . Other systems are responsible for detecting painful stimuli such as high temperature, acid, or certain specific chemicals. Although our understanding of this sensory system is not as advanced as that of the other sensory systems, recent work has revealed a fascinating relation between pain and taste sensation, a relation well known to anyone who has eaten "spicy" food .
940 CHAPTER 32 Sensory Systems
Stud ies of Capsaicin Reveal a Recept or for Sensing High Temperatures and Other Painfu l Stimuli Our sense of touch is intimately connected with the sensation of pain. Specialized neurons, termed nociceptors, transmit signals from skin to painprocessing centers in the spinal cord and brain in response to the onset of tissue damage. What is the molecular basis for the sensation of pain? An intriguing clue came from the realization that capsaicin, the chemical responsible for the "hot" taste of spicy food, activates nociceptors.
F
o
H
o
3
-;:Y~'-y--/,,- N /' H
capsaicin
N
Figure 32.36 The membrane topo logy deduced for VR1, the capsaiCin receptor. The proposed site of the membrane pore is indicated in red, and the three ankyrin (A) repeats are shown in o range. The active receptor comprises four of these subunits. [After M. J., Caterina, M . A., Schumacher, M . To minaga, A. Rosen, J. D. Levine, and D. Jul ius. Nature 389(1997):816- 824.]
Figure 32.37 Response of the capsaicin receptor to pH and temperature. The abil ity o f thi s receptor to respond to acid and to increased temperature helps detect potentially noxious situations. [A fter M. Tominaga, M. J. Caterina, A. B. Malmberg, T. A. Rosen, H. Gi lbert. K. Skinner, B. E. Raumann, A. I. Basbaum, and D. Julius,
Neuron 21(1998):531- 543.]
Early research suggested that capsaicin would act by opening ion channels that are expressed in nociceptors . Thus, a cell that expresses the capsaicin receptor should take up calcium on treatment with the molecule. This insight led to the isolation of the capsaicin receptor with the use of cDNA from cells expressi ng th is receptor. Such cells had been detected by their fluorescence when loaded with the calcium-sensitive compound Fura-2 and then treated with capsaicin or related molecules. Cells expressing the capsaicin receptor, which is called VRl (for vanilloid receptor 1), respond to capsaicin below a concentration of 1 fJ.M . The deduced 838 -residue sequence of VR1 revealed it to be a member of the TRP channel family (Figure 32 .36). The amino -terminal region ofVR1 includes three ankyrin repeats. C urrents through VR1 are also induced by temperatures above 40 °C and by exposure to dilute acid, with a midpoint for activation at pH 5.4 (Figure 32 .37). Temperatures and acidity in these ranges are associated with infection and cell inj u ry. T he responses to capsaicin, temperature, and acidity are not independent . T he response to heat is greater at lower pH, for example. Thus, VRl acts to integrate several noxious stimuli. We feel these responses as pain and act to prevent the potentially destructive conditions that cause the u npleasant sensation. Mice that do not express VR 1 suggest that this is the case; such mice do not mind food containing high concentrations of capsaicin and are, indeed, less responsive than control mice to normally noxi ous heat. Plants such as chili peppers presumably gained the ability to synthesize capsaicin and other "hot" compounds to protect themselves from being consu med by mammals. Birds, which play the beneficial role of spreading pepper seeds into new territory, do not appear to respond to • • capSalCll1.
c:
-
~
:::J
U
8
7
6
5 pH
4
3
20
30
40
Temperature (0e)
50
W
Because of its ability to simulate VR1, capsaicin is used in pain ~ management for arthritis, neuralgia, and other neuropathies. How can a compound that induces pain assist in its alleviation? Chronic exposure to capsaicin overstimulates pain -transmitting neurons, leading to their desensitization. More Sensory Systems Remain to Be Studied There may exist other subtle senses that are able to detect environmental signals that then influence our behavior. The biochemical basis of th ese senses is now under investigation. One such sense is our ability to respond, often without our awareness, to chemical signals called pheromones, released by other persons. Another is our sense of time, manifested in our daily (circadian) rhythms of activity and restfulness. Daily changes in light exposure strongly influence these rhythms. The foundations for these senses have been uncovered in other organisms; future studies should reveal to what extent these mechanisms apply to human beings as well.
Summary Smell, taste, vision, hearing, and touch are based on signal-transduction pathways activated by signals from the environment. These sensory systems function similarly to the signal-transduction pathways for many hormones. These intercellular signaling pathways appear to have been appropriated and modified to process environmental information.
32.1 A Wide Variety of Organ ic Compounds Are Detected by Olfaction The sense of smell, or olfaction, is remarkable in its specificity; it can, for example, discern stereoisomers of small organic compounds as distinct aromas. The 7TM receptors that detect these odorants operate in conjunction with G (olf), a G protein that activates a cAMP cascade resulting in the opening of an ion channel and the generation of a nerve impulse. An outstanding feature of the olfactory system is its ability to detect a vast array of odorants. Each olfactory neuron expresses only one type of receptor and connects to a particular region of the olfactory bulb. Odors are decoded by a combinatorial mechanism: each odorant activates a number of receptors, each to a different extent, and most re ceptors are activated by more than one odorant. 32.2 Taste Is a Combination of Senses That Function by Different Mechanisms We can detect only five tastes: bitter, sweet, salt, sour, and umami. The transduction pathways that detect taste are, however, diverse. Bitter, sweet, and umami tastants are experienced through 7TM receptors acting through a special G protein called gustducin. Salty and sour tastants act directly through membrane channels . Salty tastants are detected by passage though Na + channels, whereas sour taste re sults from the effects of hydrogen ions on a number of types of channels. The end point is the same in all cases membrane polarization that results in the transmission of a nerve impulse. 32.3 Photoreceptor Molecules in the Eye Detect Visible Light Vision is perhaps the best understood of the senses. Two classes of photoreceptor cells exist: cones, which respond to bright lights and colors, and rods, which respond only to dim light. The photoreceptor in rods is rhodopsin, a 7TM receptor that is a complex of the protein opsin and the chromophore ll-cis-retinal. The absorption of light by
941 Summary
942 CHAPTER 32 Sensory Systems
ll -cis-retinal changes its structure into that of all -trans-retinal , setting in motion a signal-transduction pathway that leads to the breakdown of cGMP, to membrane hyperpolarization, and to a subsequent nerve impulse. Color vision is mediated by three distinct 7TM photoreceptors that employ ll -cis-retinal as a chromophore and absorb light in the blue, green, and red parts of th e spectrum. 32.4 Hearing Depends on the Speedy Detection of Mechanical Stimuli
The immediate receptors for hearing are found in the hair cells of the cochleae, which contain bundles of stereocilia. When the stereocilia move in response to sound waves, cation channels will open or close, depending on the direction of movement. The mechanical motion of the cilia is converted into current flow and then into a nerve impulse. 32.5 Touch Includes the Sensing of Pressure, Temperature, and Other Factors
Touch, detected by the skin , senses pressure, temperature, and pain. Specialized nerve cells called nociceptors transmit signals that are interpreted in the brain as pain. A receptor responsible for the perception of pain has been isolated on the basis of its ability to bind capsaicin , the molecule responsible for the hot taste of spicy food. The capsaicin receptor, also called VRl, functions as a cation channel that initiates a nerve impulse.
Key Terms main olfactory epithelium (p. 923)
G(olf) (p. 923) functional magnetic resonance imaging (fMRJ ) (p . 926) gustducin (p. 926) amiloride-sensitive a + channel (p . 930) rod (p . 93 1) cone (p . 93 1)
rhodopsin (p . 932) opsin (p. 932)
arrestin (p . 934)
retinal (p . 932)
hair cell (p . 937) stereocilium (p . 937)
guanylate cyclase (p. 934)
chromophore (p. 932) transducin (p . 93 4)
tip link (p. 938)
cGMP phosphodiesterase (p . 934) cGMP-gated Ca2 + channel (p . 934) rhodopsi n kinase (p . 934)
nociceptor (p . 940) capsaicin receptor (VR1 receptor) (p. 940)
Selected Readings Where to Start Axel, R. 1995. The molecular logic of sroell . Sci. Am. 273(4): 154- 159. Dulac, C . 2000. The physiology of taste, vintage 2000. CellI 00:607- 61 o. Zhao, G . Q., Zhang. Y., Hoon, M . A., C handrashekar, J., Erlenbach , I., Ryba, . J. P., and Zuker, C. S. 2003. The receptors for mammalian sweet and umami taste. Cell 11 5:2 55 266. Stryer, L. 1996. Vision : From photon to perception . Proc. Na tl. Acad. Sc;. U. S. A. 91 :557- 559. Hudspeth, A. J. 1989 . How the ear's works work. Nature 341 :397-404.
Olfaction Buck, L., and Axel, R.1991. A novel multigene family may encode odorant receptors: A molecular basis for odor recognition. Cell 65:175 187. Malnic, B., Hirono, J., Sato, T ., and Buck, L. B. 1999 . Combinatorial receptor codes for odors. Ce ll 96:713- 723. Mombaerts, P., Wang, F., D ulac, C., C hao, S. K. , Nemes, A., Mendelsohn , M ., Edmondson, J.. and Axel, R. 1996. Visualizing an olfactory sensory map. Cell 87:675 686. Buck, L. 2005 . Unraveling the sense of smell (Nobel lecture). A ngew. Chern. Int. Ed . Eng!. 44 : 6128- 6140. l:lclluscio, L., Gold, G. H. , Nemes, A., and Axel , R. 1998 . Mice defi cient in G(olf) are anosmic. Neuron 20 :69- 81 .
Vosshall , L. l:l., Wong, A . M ., and Axel , R. 2000 . An olfactory sensory map in the fly brain . Cell I 02:147- 159. Lewcock, J. W., and Reed, R. R. 2003. A feedback mechanism regulates monoallelic odorant receptor expression . Proc. Nat l. Acad. Sci. U. S. A101:1069- 1074. Reed , R . R . 2004 . After the holy grail: Establishing a molecu lar mecha· nism for mammalian olfaction . C.1l11 6:329- 336.
Taste Herness, M . S., and G ilbertson , T. A. 1999. Cellular mechanisms of taste transduction . Annu. Rev. Physiol. 61:873 900 . Adler, E ., H oon , M . A., Mueller, K . L., C handrashekar, J.. Ryba, N.J.. and Z uker, C . S, 2000. A novel fami ly of mammalian taste recep· tors. CellI 00:693- 702. Chandrashekar, J., Mueller, K . L. , Hoon , M . A., Adler, E .. Feng, L.. Guo, W., Zuker, C . S., and Ryba, N . j. 2000. T 2 Rs function as bitter taste receptors. Cell 100:703- 711. Mano, I., and Driscoll , M . 1999. DEG / ENaC channels: A touchy suo perfamily that watches its salt. Bioessays 21:568-578. Ilenos, D . J., and Stanton , B. A . 1999 . Functional domains within the degenerin / epithelial sodium channel (Oeg/ ENaC) superfamily of ion channels. j. Physiol. (Lond.) 520(part 3): 631 - 644 .
Problems 943 McLaughlin, S. K., McKinnon, P. j., and Margobkee, K. F. 1992. Gustducin is a taste-cell -specific G protein clusely related to the transducins. Nature 357 :5 63- 569. Nelson, G., Hoon, M. A., Chandrashekar, J., Zhang, y, Ryba, N . j ., and Zuker, C . S. 200 I. Mammalian sweet taste receptors. Cell 106:381- 390.
Vision Strycr, L. 1988. Molecular basis of visual excitation. Co ld Spring Harbor Syrnp. Quant. Bioi . 53: 283- 294 . Wald, G . 1968. The molecular basis of visual excitatiun. Nature 219:800 807 . Ames, J. B., Dizhoor, A. M ., Jkura, M., Palczewski, K., and Stryer, L. 1999. Three-dimensional structure of guanylyl cyclase activating protein -2 , a calcium-sensitive modulator of photoreceptor guanylyl cyclases.}. BioI. Chern. 274:19329- 19337 . athans, j. 1994. In the eye of the beholder: Visual pigments and in herited variation in human vision. Cell 78:357 360. Nathans, ] . 1999. The evolution and physiology of human color vision: Insights from molecular genetic studies of visual pigment!o;, Neuron
24:299- 312. Palczewski, K., Kumasaka, T, Hori, T, Behnke, C. A., Motoshima, H., Fox, 13. A., LeTrong, I. , Teller, D . c., Okada, T , Stenkamp, R. E., Yamamoto, M ., and Miyano, M . 2000. C rystal structure of rhodopsin : A G prote.in -coupled receptor. Science 289 :739- 745. Filipek, S, Teller, D. c., Palczewski , K., and Stemkamp, R. 2003. The crystallographic model of rhodopsin and its use in studies of other
G protein -coupled receptors. Annu. Rev. Biuploys. Biumol. StTUCt. 32:175- 197.
Hearing Hudspeth, A. j . 1997. How hearing happens. Neuron 19 :947- 950 . Pickles,]. 0 ., and Corey, D. P. 1992. Mechanoelectrical transduction by hair cells. Trends Neurosci . 15:2 54- 259. Walker, ]{. G ., Willingham, A. T, and Zuker, C . S. 2000 . A Drosophila mechanosensory transduction channel. Science 287:2229- 2234. ):-Iudspeth, A. ] ., C hoe, Y, Mehta, A. D., and Martin, P. 2000. Putting ion channels to work : Mechanoelectrical transduction, adaptation, and amplification by hair cell s. Proc. Natl. Acad. Sci. U. S. A.
97: 11 765- 11 772.
Touch and Pain Reception Franco-Obregon, A., and Clapham, D. E. 1998. Touch channels sense blood pressure. Neuron 2 1:1224 1226. Caterina, M. j ., Schumacher, M. A., Tominaga, M., Rosen, T A., Levine, J. D., and julius, D. 1997. The capsaicin receptor: A heatactivated ion ehalUlel in the pain pathway. Nature 389 :R1 6-ll24. Tuminaga, M., Caterina, M. ] ., Malmberg, A. B., Rosen, T A., Gilbert, li ., Skinner, K., Raumann, B. E., Basbaum , A. I., and j ulius, D. 1998. T he cloned capsaicin receptor integrates multiple pain -producing stimuli . Neuron 21:531- 543. Caterina , M . j. , and julius, D. 1999. Sense and specificity: A molecular identity for nociceptors. Curro Opin. Neurobiul. 9:525-530 . Clapham , D. E. 2001 . T RP channels as cellular sensors . Nature 426:51 7- 52 4.
Problems 1. Mice and rats. As noted on page 924, one of the first odorant receptors to be matched with its ligand was a rat receptor that responded best to n-octanal. The sequence of the corresponding mouse receptor d iffered from the rat receptor at 15 positions. Surprisingly, the mouse receptor was found to respond best to n heptanal rather than n -octanal. The substitution of isoleucine at position 206 in the mouse for valine at this position in t he rat receptor was found to be important in determining the specificity for n-heptanal. Propose an explanation .
2. Olfactiun in
Unl ike the olfactory ne u rons in the mammali an systems discussed herein, olfactory neurons in the nematode C. elegam express multiple olfactory receptors . In particular, one neuron (called AWA ) expresses receptors for compounds to which the nematode is attracted, whereas a different n euron (called AWB) expresses receptors for com pounds that the nematode avoids . Su ppose that a tTansgenic nematode is gen erated s uch that one of the receptors for an attractant is expressed in AWB rather than A WA . What behavior wou ld you expect in the presence of the corresponding attractant? WOmts.
3. Odorant matching. A mixture of two of the compound s illustrated in Figure 32 .6 is applied to a section of olfactory epithelium . Only receptors 3,5,9, 12, and 13 are activated, according to Figure 32. 7. Identify t he likely compounds in the mixture. 4. Timing . Compare the aspects of taste (bitter, sweet, salty, sour) in regard to their potential for rapid time resolution.
5. Two ears. O ur abili ty to determine the direction from which a sound is coming is partly based on the d ifference in time at
which our two ears detect the sound . Given the speed of sound (350 m s - I) and the separation between our ears (0 .1 5 m ), what difference is expected in the times at which a sound arrives at our two ears? How does this difference compare with the time resolution of th e human hearing system? Would a sensory system that utilized 7TM receptors and G proteins be capable of adequate time resolution?
6. Constitutive mutants . What effect within the olfactory system would yo u expect for a mutant in which adenyl ate cyclase is al ways fully active? What effect within the visual system would you expect for a mutant in which guanylate cyclase is always fully active?
7. Bottle chuice. A widely used method for quantitatively moni toring rodent behavior with regard to taste is the bottle-choice assay. An animal is placed in a cage with two water bottles, one of which contains a potential tastant. After a fixed period of time (24 48 hours) , the amount of water remaining in each bottle is measured. Suppose that much less water remains in the bottle containing the tastant after 48 hours . Do you suspect t he tastant to be sweet or bitter?
8. It's better to be bitter. Som e nontoxic plants taste very bitter to us. Suggest one or m ore explanations.
9. Of mice and men. In human beings, the umami taste is triggered only by glutamate and aspartate. Jn contrast, mice respond to many more amino acids. Design an experiment to test which of the subunits (T1Rl or TIR3 ) determines the specificity of this response. Assume that all desired m ouse strains can be read ily produced .
944
CHA PTER 32 Sensory Syst e ms
Chapter Integration Problem
Mechanism Problem
10 . Energy and inforTrUltion. The transmission of sensory information requires the input of free energy. For each sensory system (olfactio n, gustation, vision, hearing, and touch), identify mechanisms for the input of free energy that allow the transmission of sensory in formation .
11 . Schiff-base Jonnation . Propose a mechanism for the reaction between opsin and 11 -cis -retinal.
Chapter
The Immune System
Antibody
Influenza hemagglutinin
Just as medieval defenders used their weapons and the castle wall s t o defend their city, the immune syst em constantly battles against foreign invaders such as viruses, bacteria, and parasites to defend the organism. Antibody molecules pro vide a key element in the immune system's defensive arsenal. For example, specific antibodies can bind to molecules on the surfaces o f viruses and prevent the viruses from infecting ce lls. Above right, an antibody binds t o o ne subunit on hemagglutinin fro m the surface o f influenza virus. [(Left) The Granger Collecti o n.]
e are constantly exposed to an incredible diversity of bacteria, viruses, and parasites, many of which would flourish in our cells or extracellular fluids were it not for our immune system. How does the immune system protect us? The human body has two lines of defense: an innate immune system that responds rapidly to features present in many pathogens, and an adaptive immune system that responds to specific features present only in a given pathogen. Both the innate and the adaptive immune systems first identify features on disease-causing organisms and then work to eliminate or neutralize those organisms. This chapter focuses on the mechanisms of pathogen identification. The immune system must meet two tremendous challenges in the identification of pathogens: (1) to produce a system of receptors diverse enough to recognize the diversity of potential pathogens and (2) to distinguish invaders and their disease-causing products from the body and its own prod ucts (i.e., self- versus non-self-recognition). To meet these challenges, the innate immune system evolved the ability to recognize structural elements, such as specific glycolipids or forms of nucleic acid, that are well conserved in pathogens but absent in the host organism. The repertoire of such elements is limited, however, and so some pathogens have strategies to escape detection. The adaptive immune system has the remarkable 12 8 ability to produce more than 10 distinct antibodies and more than 10 T-cell receptors (TCRs), each of which presents a different surface with
I Outlin e l 33.1 Antibodies Possess Distinct AntigenBinding and Effector Units 33.2 The Immunoglobulin Fold Consists of a Beta-Sandwich Framework with Hypervariable Loops 33.3 Antibodies Bind Specific Molecules Through Their Hypervariable Loops 33.4 Diversity Is Generated by Gene Rearrangements 33.5 Major-Histocompatibility-Complex Proteins Present Peptide Antigens on Cell Surfaces for Recognition by T-Cell Receptors 33.6 Immune Responses Against Self-Antigens Are Suppressed
945
946 CHAPTER 33 The Immune System
Leucine-rich repeat
Cysteine-rich domain
TRI domain (signal transduction)
Figure 33.1 Toll -like receptor. Each receptor comprises a set of 18 or more leucine-rich repeat sequences, followed by a cysteine-ri ch domain, a single transmembrane helix. and a TIR (Toll- interleukin 1 receptor) domain that functions in signal transduction.
Figure 33.2 Lipopolysaccharide structure. Lipopolysaccharide, a potent activator of t he innate immune system, is found on the surfaces of Gram-negative bacteria. The structure is built around lipid A. a specialized lipid that has four fatty acyl chains linked to two N -acetylglucosamine residues. Lipid A is linked to a polysaccharide chain consisting of a core and a more variable region termed the O -specific chain.
the potential to specifically bind a structure from a foreign organism. In producing this vast range of defensive molecules. however. the adaptive immune system has the potential to create antibodies and T -cells that recognize and attack cells or molecules normall y present in our bodies a situation that can result in autoimmune diseases. This chapter will examine these challenges. focusing first on the structures of proteins that recognize foreign organisms and then on the mechanisms for protecting us from a specific pathogen once it has been recognized. The chapter will closely examine the modular construction of the proteins of the immune system identifying structural motifs and considering how spectacular diversity can arise from modular construction . Innate Immunity Is an Evolutionarily Ancient Defense System
Innate immunity is an evolutionarily ancient defense system found. at least in some form. in all multicellular plants and animals. The genes for its key molecules are expressed without substantial modification. unlike genes for key components of the adaptive immune system. which un dergo significant rearrangement. Through many millions of years of evolution. proteins expressed by these genes have gained the ability to recognize specific features present in most pathogens and yet not respond to materials normally present in the host. The most important and best- understood receptors in the innate immune system are the Toll-like recepturs (TLRs). At least 10 TLRs have been identified in human beings. although only a single such receptor is present in C. elegans, for example. The name "toll-like" is derived from a receptor known as Toll encoded in the Drosophila genome; Toll was first identified in a screen for genes important for Drosuphila development and was subsequently discovered to also playa key role in the innate immune system later in development. The TLRs have a common structure (Figure 33.1 ). Each receptor consists of a large domain built primarily from repeated amino acid sequences termed leucine-rich repeats (LRRs) because each repeat includes six residues that are usually leucine. The human TLRs have from 18 to 27 LRR repeats. These repeats are followed by a sequence forming a single transmembrane hel ix and then by a signaling domain common to the TLRs as well as to a small number of other receptors. This signal ing domain is not a protein kinase but acts as a docking site for other proteins. A protein that docks to a TLR initiates a signal transduction pathway that ultimately leads to the activation of specific transcription factors . Most TLRs are expressed in the cell membrane for the detection of extracellular pathogens such as fungi and bacteria. Other TL Rs are located in the membranes of internal compartments for the detection of intracellular pathogens such as viruses and some bacteria. Each TLR is targeted to a specific molecular characteristic. often called a pathogen-associated molecular pattern (PAMP). found primarily on invading organisms. One particularly important PAMP is lipopolysaccharide (LPS). a specific class of glycolipids found in the cell walls of Gram-negative bacteria such as E. coli (Figure 33.2). LPS is built around a specialized lipid. called lipid A, that contains two linked N-acetylglucosamine residues and four fatty acyl chains. Lipid A is connected to a polysaccharide chain consisting of a core structure and a more variable region referred to as the O-specific chain. LPS is also known as endotoxin. The response of the innate immune system to LPS can be easily demonstrated. Injection of less than
Lipid A
- - il Inner core
Outer core 1 1 --
O-specific chain
947
One repeat unit
The Immune System
Jl strand
~
Leu
Asn
Leu
~
strand
Leu
Leu
(A)
(8)
1 mg of LPS into a human being produces a fever and other signs of inflammation even though no living organisms are introduced . LPS is recognized primarily by TLR-4, whereas other TLRs recognize other classes of PAMP. For example, TLR -S recognizes the protein fla gellin, found in flagellated bacteria, and TLR-3 recognizes double-stranded RNA. Note that, in each case, the target of the TLR is a key component of the pathogen, and so mutations cannot easily block recognition by the TLR and, hence, escape detection by the innate immune system. In some cases, TLRs appear to form heterodimers that either enhance or inhibit PA M P • • recogmtIOn. How do TLRs recognize PAMPs? The leucine-rich repeat domain from human TL R-1 has a remarkable structure (Figure 33.3). Each of its LRR units contributes a single [3 strand to a large parallel [3 sheet that lines the inside of a concave structure. This hooklike structure immediately suggests a model for how TLRs bind PAMPs namely, that the PAMP lies on the inside of the "hook." This model is likely accurate for some TLRs. However, for other TLRs, the PAMP-binding site appears to lie on one side of the structure, and the central hole is blocked by host carbohydrates linked to the structure. Regardless of the details of the interaction, PAMP binding appears to lead to the formation of a specific dimer of the TLR. The cytoplasmic side of this dimer is a signaling domain that initiates the signal-transduction pathway. Because the TLRs and other components of the innate immune system are always expressed, ready to target conserved structures from pathogens, they provide the host organism with a rapid response system to resist attack by pathogens. We now tu rn to the adaptive immune system, which, remarkably, is able to target specific pathogens, even those that it has never encountered in the course of evolution. The Adaptive Immune System Responds by Using the Principles of Evolution
The adaptive immune system comprises two parallel but interrelated systems: humoral and cellular immune responses. In the humoral immune response, soluble proteins called antibodies (immunoglobulins) function as recognition elements that bind to foreign molecules and serve as markers
Figure 33.3 PAMP-recognition unit of the Toll -like receptor. (A) The structure of the leucine-rich repeat (LRR) domain from human TLR-3. Notice that the LRR units come together to form a centra l parallel J3 sheet that curls to form a concave structure. (B) The structure of a single LRR showing the positions of the residues that are generally approximately conserved. Notice that the leucine residues come together to form a hydrophobic core with the single J3 strand along on one side. [Drawn from lZIW.pdb].
Figure 33.4 Immunoglobulin production. An electron micrograph of a plasma cell sho ws the highly develo ped rough endoplasmic reti culum necessary fo r antibody secreti on. [Courtesy o f Lynne Mercer.]
948
signaling foreign invasion (Figure 33.4). Antibodies are secreted by plasma cells, which are derived from B lymphocytes (B cells). A foreign macromolecule that binds selectively to an antibody is called an antigen. In a physiological context, if the binding of the foreign molecule stimulates an immune response, that molecule is called an immunogen. The specific affinity of an antibody is not for the entire macromolecular antigen but for a particular site on the antigen called the epitope or antigenic determinant. In the cellular immune response, cells called cytotoxic T lymphocytes (also commonly called killer T cells) kill cells that have been invaded by a pathogen. Because intracellular pathogens do not leave markings on the exteriors of infected cells, vertebrates have evolved a mechanism to mark the exterior of cells with a sample of the interior contents, both self and foreign. Some of the internal proteins are broken into peptides, which are then bound to a complex of integral membrane proteins encoded by the major histocompatbility complex (MHC). T cells continually scan the bound peptides (pMHCs) to find and kill cells that display foreign motifs on their surfaces. Another class of T cell s called helper T lymphocytes contributes to both the humoral and the cellular immune responses by stimulating the differentiation and proliferation of appropriate B cells and cytotoxic T cells. The celluar immune response is mediated by specific receptors that are expressed on the surfaces of the T cells. The remarkable ability of the immune system to adapt to an essentially limitless set of potential pathogens requires a powerful system for transforming the immune cells and molecules present in our systems in response to the presence of pathogens. This adaptive system operates through the principles of evolution, including reproduction with variation followed by selection of the most well suited members of a population. If the human genome contains, by the latest estimates, only 25,000 8 genes, how can the immune system generate more than 10 different anti12 body proteins and 10 T -cell receptors? The answer is found in a novel mechanism for generating a highly diverse set of genes from a limited set of genetic building blocks. Linking different sets of DNA regions in a combinatorial manner produces many distinct protein -encoding genes that are not present in the genome. A rigorous selection process then leaves for proliferation only cells that synthesize proteins determined to be useful in the immune response. The subsequent reproduction of these cells without additional recombination serves to enrich the cell population with members expressing particular protein species. Critical to the development of the immune response is the selection process, which determines which cells will reproduce. The process comprises several stages. In the early stages of the development of an immune response, cells expressing molecul es that bind tightly to self-molecules are destroyed or silenced, whereas cells expressing molecules that do not bind strongly to self-molecules and that have the potential for binding strongly to foreign molecules are preserved. The appearance of an immunogenic invader at a later time will stimulate cells expressing antibodies or T-cell receptors that bind specifically to elements of that pathogen to reproduce in evolutionary terms, such cells are selected for. Thus, the immune response is based on the selection of cells expressing molecules that are specifically effective against a particular invader; the response evolves from a population with wide-ranging specificities to a more-focused collection of cells and molecules that are well suited to defend the host when confronted with that particular challenge. Not only are antibodies and T -cell receptors a result of genetic diversity and recombination, but antibodies have highly diverse structures as well. Antibodies require many different structural solutions for binding many
949
different antigens, each of which has a different form. T-cell receptors, in contrast, are not structurally diverse, because they have coevolved with the MHC. The docking mode of a T-cell receptor to the peptide bound to MHC is similar for all structures. As a consequence of this coevolution, every T-cell receptor has an inherent reactivity with every MHC. The coevolution ensures that all T -cell receptors can scan all peptide~MHC complexes on all 12 tissues . The genetic diversity of the 10 different T -cell receptors is con centrated in a highly diverse set of residues in the center of the MHC groove. This localized diversity allows the T -cell receptor to recognize the many different foreign peptides bound to the MHC. T -cell receptors must survey many different MHC~peptide complexes with rapid turnover. Therefore, the binding affinities between T -cell receptors and the MHC are weaker than those between antibody and antigen.
33.1
Antibodies Possess Distinct Antigen-Binding and Effector Units
Antibodies are central molecular players in the immune response, and we examine them first. A fruitful approach in studying proteins as large as an tibodies is to split the protein into fragments that retain activity. In 1959, Rodney Porter showed that immunoglobulin G (lgG), the major antibody in serum, can be cleaved into three 50-kd fragments by the limited proteolytic action of papain. Two of these fragments bind antigen. They are called Fah (F stands for fragment, ab for antigen binding). The other fragment, called Fcbecause it crystallizes readily, does not bind antigen, but it has other important biological activities, including the mediation of responses termed effector functions. These functions include the initiation of the complement cascade, a process that leads to the lysis of target cells. Although such effector functions are crucial to the functioning of the immune system, they will not be considered further here. How do these fragments relate to the three-dimensional structure of whole IgG molecules? Immunoglobulin G consists of two kinds of polypep tide chains, a 25-kd light (L) chain and a 50-kd heavy (H) chain (Figure 33. 5). The subunit composition is L 2 H 2 . Each L chain is linked to an H chain by a disulfide bond, and the H chains are linked to each other by at least one disulfide bond. Examination of the amino acid sequences and three-dimensional structures of IgG molecules reveals that each L chain comprises two homologous domains, termed immunoglobulin domains, to be
(8)
(A)
light chain N ~--=,
N""" Heavy chain
~ Figure 33.5 Immunoglobul in G
structure. (A) The threedimensio nal stru cture o f an IgG mo lecule showing the light chains in yellow and the heavy chains in blue. (B) A schematic view of an IgG molecule indi cating the positions of the interchain disulfide bonds. Abbreviations: N, amino terminus; C, carboxyl terminu s. [Drawn from lIGTpdb.]
Interchain disulfide bonds
/,- " C
•
33.1 Antibody Units
950
TABLE 33.1 Properties of immunoglobulin classes
CHAPTER 33 The Immune System Class IgG IgA IgM IgD IgE
Serum concentration (mgml - ') 12
3 1 0.1 0 .001
Mass (kd)
Sedimentation
Light
Heavy
Chain
coefficient(s)
chains
chains
structu re
150 180- 500 950 175 200
7 7. 10, 13 18- 20 7 8
or A K or A..
"Y
K,}"y,}
Kor A
'I'-"
(""' ,)n or (X ,a,~ ("1'-,).; or (X,I',),
K
or X.
&
K 20 2
or
K
or X.
E
K2E2
or A2EZ
K
or
>" 2"17
A28 2
Note: n = 1, 2, or 3. ISM and oligomers of IgA also contai n J chains that connect immunoglobulin mo lecules. IgA in
secretions has an additional component
Papain cleavage
? 3~ , i
2, ~ ,$
Figure 33.6 Immunoglobulin G cleavage. Treatment of intact IgG mo lecules with the protease papain resu lts in the f o rmation of three large fragments: two F,b fragments that retain antigen-bind ing capability and o ne Fe fragment that does not.
Antigen ..........
described in detail in Section 33,2, Each H chain has four immunoglobulin domains. Overall, the molecule adopts a conformation that resembles the letter Y, in which the stem, corresponding to the Fe fragment obtained by cleavage with papain, consists of the two carboxyl -terminal immunoglobulin domains of each H chain and in which the two arms of the Y, corresponding to the two Fab fragments, are formed by the two amino-terminal domains of each H chain and the two amino- terminal domains of each L chain. The linkers between the stem and the two arms consist of extended polypeptide regions within the H chains and are quite flexible . Papain cleaves the H chains on the carboxyl -terminal side of the disulfide bond that links each Land H chain (Figure 33.6). Thus, each Fab consists of an entire L chain and the amino-terminal half of an H chain, whereas Fe consists of the carboxyl -terminal halves of both H chains. Each F:,h contains a single antigen -binding site. Because an intact IgG molecule contains two F:,b components and therefore has two binding sites, it can cross-link multiple antigens (Figure 33.7). Furthermore, the Fe and the two Fab units of the intact IgG are joined by flexible polypeptide regions that allow facile variation in the angle between the Fah units through a wide range (Figure 33.8). This kind of mobility, called segmental flexibility, can enhance the formation of an antibody- antigen complex by enabling both combining sites on an antibody to bind an antigen that possesses multiple binding sites, such as a viral coat composed of repeating identical monomers or a bacterial cell surface. The combining sites at the tips of the F;,b units simply move to match the distance between specific determinants on the antigen. Immunoglobulin G is the antibody present in highest concentration in the serum, but other classes of immunoglobulin also are present (Table 33.1). Each class includes an L chain (either K or A) and a distinct H chain (Figure 33.9). The heavy chains in IgG are called 'Y chains, whereas those in immunoglobulins A, M, D, and E are called 0', f.L , 1>, and E, respectively.
Antigen-binding sites
Hinge
Figure 33.7 Antigen cross-linking. Because IgG molecules include two antigen-binding sites, antibo dies can cross -link multivalent antigens such as vi ral surfaces.
Figure 33.8 Segmental flexibility. The linkages between the Fab and the Fe regi ons of an IgG molecule are flexible, allOWing the two antigen-binding si t es t o adopt a range of orientations with respect to one another. This fle Xi bility allows effective interactions with a multivalent antigen without requiring that the epitopes o n the target be a precise distance apart.
IgA (dimer)
IgG
IgM (pentamer)
IgO
o chain
y chain f
IgE
l::
chain
chain Jl chain
Immunoglobulin M (IgM) is the first class of antibody to appear in the serum after exposure to an antigen. The presence of 10 combining sites enables IgM to bind especially tightly to antigens containing multiple identical epitopes. The strength of an interaction comprising multiple independent binding interactions between partners is termed avidity rather than affinity, which denotes the binding strength of a single combining site. Immunoglobulin A (IgA) is the major class of antibody in external secretions, such as saliva, tears, bronchial mucus, and intestinal mucus . Thus, IgA serves as a first line of defense against bacterial and viral antigens. The role of immunoglobulin D (IgD ) is not yet known. Immunoglobulin E (IgE) is important in conferring protection against parasites, but IgE also participates in allergic reactions. Ig E- antigen complexes form cross-links with receptors on the surfaces of mast cells to trigger a cascade that leads to the release of granules containing pharmacologically active molecules. Histamine, one of the agents released, induces smooth-muscle contraction and stimu lates the secretion of mucus. A comparison of the amino acid sequences of different IgG antibodies from human beings or mice shows that the carboxyl-terminal half of the L chains and the carboxyl-terminal three-quarters of the H chains are very similar in all of the antibodies. Importantly, the amino-terminal domain of each chain is more variable, including three stretches of approximately 7 to 12 amino acids within each chain that are hypervariable, as shown for the H chain in Figure 33.10. The amino-terminal immunglobulin domain of each
Figure 33.9 Classes of immunoglobulin. Each of five cla sses of immunoglobulin has the same light chain (shown in yellow) combined with a different heavy chain b, Ct., fL, &, or E). Disulfide bonds are indicated by green lines. The IgA dimer and the IgM pentamer have a small polypeptide chain in additio n to the light and heavy chains.
150
100
..-
-
.0
."' 50
o o
20
40
60
Residue
80
100
120
Figure 33.10 Immunoglobulin sequence diversity. A plot of sequence variability as a function of position along the sequence of the amino-terminal immuno globulin domain of the H chain of human IgG molecules. Three regions (in red) show remarkably high levels of variability. These hypervariable regions correspond to three loops in the immunoglobulin domain structure. [After R. A. Goldsby, T. J. Kindt, and B. A. Osborne, Kuby Immunology, 4th ed. (w. H. Freeman and Company, 2000), p. 9l.]
951
chain is thus referred to as the variable region, whereas the remaining immunoglobulin domains are much more similar in all antibodies and are referred to as constant regions (Figure 33. 11 ).
33_2
Figure 33.11 Variable and constant regions. Each Land H chain includes one immunoglobulin domain at its amino terminus that is quite variable from one antibo dy to another. These domains are referred to as Vl and VH . The remaining doma ins are more constant from one antibo dy to another and are referred to as constant domains (Cll. CH1. CH2. and CH3).
The Immunoglobulin Fold Consists of a BetaSandwich Framework with Hypervariable Loops
An IgG molecule consists of a total of 12 immunoglobulin domains. These domains have many sequence features in common and adopt a common structure, the immunoglobulin fold (Figure 33.12). Remarkably, this same structural domain is found in many other proteins that play key roles in the immune system and in nonimmune functions . The immunoglobulin fold consists of a pair of f3 sheets, each built of antiparallel f3 strands, that surrou nd a central hydrophobic core. A single disulfide bond bridges the two sheets. Two aspects of this structure are particularly important for its function. First, three loops present at one end of the structure form a potential binding surface. These loops contain the hypervariable sequences present in antibodies and in T -ceLl receptors (see Section 33.3 and p. 963). Variation of the amino acid sequences ofthese loops provides the major mechanism for the generation of the vastly diverse set of antibodies and T-cell receptors expressed by the immune system . These loops are referred to as hypervariable loops or complementarity-determining regions (CDRs). Second, the amino terminus and the carboxyl terminus are at opposite ends of the structure, which allows structural domains to be strung together to form chains, as in the Land H chains of antibodies. Such chains are present in several other key molecules in the immune system. The immunoglobulin fold is one of the most prevalent domains encoded by the human genom e: more than 750 genes encode proteins with at least one immunoglobulin fold recognizable at the level of amino acid sequence. Such domains are also common in other multicellular arumals such as flies and nematodes. However, from inspection of amino acid seq uence alone, immunoglobulin-fold domains do not appear to be present
N terminus
_ - -- Hypervariable loops ------.
Disulfide bond
~ Figure 33.12 Immunoglobulin fold.
An immunoglobulin domain consists of a pair of 13 sheets linked by a disulfide bond and hydrophobic interactions. Notice that three hy pervariable loops lie at one end of the structure. [Drawn from lDQJ.pbd.]
952
C terminus
Front view
Side vie¥'J.·
in yeast or plants, although these organisms possess other structurally similar domains, including the key photosynthetic electron-transport protein plastocyanin in plants (p. 551). Thus, the immunoglobulin -fold family appears to have expanded greatly along evolutionary branches leading to animals particularly, vertebrates.
33.3
953 33.3 Antibody Binding
Antibodies Bind Specific Molecules Through Their Hypervariable Loops
For each class of antibody, the amino-terminal immunoglobin domains of the Land H chains (the variable domains, designated V L and V H) come together at the ends of the arms extending from the structure. The positions of the complementarity -determining regions are striking. These hypervariable sequences, present in three loops of each domain, come together so that all six loops form a single surface at the end of each arm (Figure 33.13). Because virtually any V L can pair with any V H, a very large number of different binding sites can be constructed by their combinatorial association.
~ Figure 33.13 Variable domains. Two
Side view
End-on view
X-ray Analyses Have Revealed How Antibodies Bind Antigens
The results of x-ray crystallographic studies of several hundred large and small antigens bound to F ab molecules have been sources of much insight into the structural basis of antibody specificity. The binding of antigens to antibodies is governed by the same principles that govern the binding of substrates to enzymes. The apposition of complementary shapes results in numerous contacts between amino acids at the binding surfaces of both molecules. Many hydrogen bonds, electrostatic interactions, and van der Waals interactions, reinforced by hydrophobic interactions, combine to give specific and strong binding. A few aspects of antibody binding merit specific attention, inasmuch as they relate directly to the structure of immunoglobulins. The binding site on the antibody has been found to incorporate some or all of the CDRs in the variable domains of the antibody. Small molecules are likely to make contact with fewer CD Rs, with perhaps 15 residues of the antibody participating in the binding interaction. Macromolecules often make more extensive contact, sometimes interacting with all six CDRs and 20 or more
views of the variable domains of the L chai n (yellow) and the H chain (blue); the complementarity-determining regi ons (CDRs) are shown in red. Notice on the left that the six CDRs come together to form a binding surface. The specificity of the surface is determined by the sequences and structures of the CDRs. [Drawn from lDQJ .pdb.]
954 CHAPTER 33 The Immune System
residues of the antibody. Small molecules often bind in a cleft of the antigen· binding region. Macromolecules, such as globular proteins, tend to interact across larger, fairly flat apposed surfaces bearing complementary protru· sions and depressions. The search for an HIV vaccine has recently extended our understanding of antibodies and the way that they bind small molecules. The persistent problem in HJV vaccine design has been the lack of a neutralizing antibody response. 1n other words, most human antibodies do not recognize the HIV virus. A few rare antibodies isolated from asymptomatic, HIV-infected people show the neutralizing response. One of these antibodies, b1 2, gives an example of an antigen -binding surface that is not flat. Instead, b 12 has a very long CD R3100p that forms a "fingerlike" projection that can probe the canyons and valleys on the virus's surface. Another of these rare HIV-reactive antibodies, called 2G 12, also has an unusual form; instead of the normal "Y" shape of the IgG molecule, 2G 12 has its two arms pointing vertically and adjacent to one another. The two Fab "arms" form a tightly packed dimer because their V H domains are swapped . A well -studied case of small-molecule binding is seen in an example of phosphorylcholine bound to Fab. Crystallographic analysis revealed phos· phorylcholine bound to a cavity lined by residues from five CDRs two from the L chain and three from the H chain (Figure 33.14). The positively charged trimethylammonium group of phosphorylcholine is buried inside the wedge-shaped cavity, where it interacts electrostatically with two negatively charged glutamate residues. The negatively charged phosphoryl group of phosphorylcholine binds to the positively charged guanidinium group of an arginine residue at the mouth of the crevice and to a nearby lysine residue. The phosphoryl group is also hydrogen bonded to the hydroxyl group of a tyrosine residue and to the guanidinium group of the arginine side chain. Numerous van der Waals interactions, such as those made by a tryptophan side chain, also stabi li ze this complex.
Asp 197
Phosphorylcholine Trp Hl07
~ Figure 33.14 Binding of a small
antigen. The structure of a complex between an Fab fragment of an antibody and its target- in this case. phosphorylcholine. Residues from t he antibody interact with phosphorylcholine through hydrogen bonding and electrostati c and van der Waals interactions. [Drawn from 2MCP.pdb.]
Asn Hl0l
Tyrll00 Tyr H33
Residues from five CDRs participate in the binding of phosphorylcholine to human Fah . This binding does not significantly change the struc· ture of the antibody, yet induced fit plays a role in the formation of many antibody- antigen complexes. A malleable binding site can accommodate many more kinds ofligands than can a rigid one. Thus, induced fit increases the repertoire of antibody specificities. Large Antigens Bind Antibodies with Numerous Interactions
How do large antigens interact with antibodies? A large collection of antibod· ies raised against hen egg-white lysozyme has been structurally characterized in great detail (Figure 33.15). Each different antibody binds to a distinct
~ Figure 33.15 Antibodies against
lysozyme. (A) The structures of three complexes (i, ii, iii) between F. b fragments (blue and yellow) and hen eggwhite lysozyme (red) shown with lysozyme in the same orientation in each case. The three antibodies recogn ize completely different epitopes on the lysozyme molecule. (B) The F. b fragments from part A (corresponding from left to right to i, ii, and iii) with points of contact highlighted as space-filling models. Notice the different shapes of the antigen-binding sites. [Drawn from 3HFL, lDQJ, and lFDL.pdb.]
(ii)
(iii)
(A)
(8)
surface oflysozyme. Let us examine the interactions in one of these complexes (complex ii in Figure 33.15A) in detail. This antibody binds two polypeptide segments that are widely separated in the primary structure, residues 18 through 27 and 116 through 129 (Figure 33.16). All six CDRs of the antibody make contact with this epitope. The region of contact is quite extensive (about 30 X 20 A). The apposed surfaces are rather flat . The only exception is the side chain of glutamine 121 of lysozyme, which penetrates deeply into the antibody's binding site, where it forms a hydrogen bond with a main -chain carbonyl oxygen atom and is surrounded by three aromatic side chains. The formation of 12 hydrogen bonds and numerous van der Waals interactions contributes to the high affinity (Kd = 20 nM) of this antibody- antigen interaction. Examination of the Fab 955
--.". Figure 33.16 Antibody- protein '
11 . Crystallization. T he proteolyti c digestion of a population of IgG molecules isolated from human serum results in the generation ofF"" and F e fragments. Why do Fe fragments crystallize more easily than F ab fragments generated from such a population? 12 . Presentation. T he amino acid sequence of a small protein is
0.8
• ••
• Antibody A • Antibody B
•
•
• • • -9
-8
-7
-6
-5
log [antibody] (M) A fter repeated immunizations with the same protein, another cell line is d erived that expresses a different antibody (antibody B). The results of an alyzin g the binding of antibody B to the protein also are shown. From these data, estimate (a) the dissociation constant (K d ) for the complex between the protein and antibody A. (b) the dissociation consta nt for the complex between the protein and an tibody H. Co mparison of the amino acid sequences of a ntibod y A and antibod y B reveals them to be id en t ical except for a single amino acid . What d oes thi s findin g suggest about the m echanism by which the gene e ncodin g antibody B was generated?
Chapt e r
4
Molecular Motors
The horse, like all animals, is powered by the molecu lar-motor protein myosin. A part o f myosin moves dramatical ly (as shown above) in response t o ATP bind ing, hydrolysis, and product release, pro pel ling myosin along an actin f ilament. This molecular movement is t ranslated into movement of the entire animal, vividly depicted in da Vinc i's rearing horse. [(Left) Leonardo da Vinci 's study of a rearing horse for the Battle of Anghiari (ca. 1504) from The Royal Collection © Her Royal Majesty Queen Elizabeth 11.]
rganism s, from human beings to bacteria, move to adapt to changes in ~ their environments, navigating toward food and away fro m danger. Cell s, th emselves, are not static but are bustling assemblies of moving proteins, nucleic acids, and organelles (Figure :14. 1). Remarkably, the fundamental biochemical mechanisms that produce contractions in our muscles are the same as those that propel organell es al ong defined paths inside cells. In fact, many of the proteins th at play key roles in converting chemical energy in the form of ATP into kin etic energy, the energy of motion, are members of the same protein family, the P -loop N TPases. These molecular motors are homologous to proteins that we have encountered in other con texts, including the G proteins in protein sy nthesis, signaling, and other processes. Once again, we see the economy of evolution in adapting an existin g protein to perform new fun ctions. Molecular motors operate by small increments, converting changes in protein conformation into directed motion . O rderl y motion across distances requires a track that steers the moti on of the motor assembly. Indeed , we have previously enco untered a class of molecular motors that utilize mech anisms that we will examin e here namely, the helicases that m ove along DNA and RNA tracks (Section 28.2). T he proteins on which we will focus in this chapter move along actin and microtubules protein filaments co mposed of repeatin g subunits . T he motor proteins cycle between
[ Outlin e 34.l
Most Molecular-Motor Proteins Are Members of the P-Loop NTPase Superfamily
34.2 Myosins Move Along Actin Filaments 34.3 Kin esin and Dynein Move Along Microtubules 34.4 A Rotary Motor Drives Bacterial Motion
977
'":~'''1''~'''
~,.~'~")'
\
.,
. ... .
,
-
•
Figure 34.1 Motion within celis. Thi s high-
form s having high or low affinity for the filament tracks in response to ATP binding and hydrol ysis, enabling a bind, pull, and release mechanism that generates motion . W e will also consider a completel y different strategy for generating motion , one used by bacteria such as E. coli. A set of fl agella act as propellers, rotated by a motor in the bacterial cell membrane. T his rotary motor is d riven by a proton gradient across the membrane, rather than by AT P hy· d rolysis. The mechanism for coupling the proton gradi ent to rotatory mo· tion is analogous to that used by the F 0 subunit of AT P synthase (p. 522). Thus, both of the m ajor modes for storing biochemi cal energy namely, ATP and ion gradients have been h arnessed by evolution to drive organized molecul ar motion .
voltage electron micrograp h sho ws the mit o tic apparatus in a metaphase mammalian cell. The large cylindrical obiect s are chromosomes, and th e threadlike structures stret ched ac ross the center
34.1
Most Molecular-Motor Proteins Are Members of the P-Loop NTPase Superfamily
are microt ubules- t rac ks for the mo lecular mot ors th at move chromosomes. M any processes, incl ud ing chro mosome segregat ion in mitosis. depend on t he action o f molecular-motor p ro teins. [Courtesy of Dr. J. R. M Clntosh.l
,
, . ~':':- ' ..;~...'-,';. j '.• . :~• ~ I~:.ea::r ••, •. ' ',...• " , . ~.~. .{." ·· t'," . ... ...
,"' " ............. .... ... ..... .. ..,. .."..
.~.'
;
...···. ".'1""'.',.. ......... . ...... ... .. '(
.,
.
~,'
.. " \ ' ,-
'f "
. '. ~' ,.:~.: .... :;. , >",:. "' .... .> "''t .... ._ . . ,'_ .,' / , .... . . . " - " ' , ' - ' "....... , ... (,
•
• ••
" ''; ' ' '"
.. ,.
.
~.
"; ' , .
• .. , · .... f ... .
"'.~ ' ,~" ~.'
I . .... ... - ..,. . ... . ,. ' . . ' ./.'. ~··' .· ·i I ' ",.-"............ ,~... ·.".' · . .'0'-.'. ........ """ ' 0",, ' J ,"-,,' """ '" \ . .... . . ....., · . . ..,.,..•......;. .. .,_.. , ", ,. .
,,~
. . ... .. . . .. ........ .... ," ................ ,.. · ., .. . ..','. -" ,
~,
.,~
~
'.",.~
'
" .~ ,: .
~'
.
'
..· ....
•••. ,."' A ..... ,,' ,.-" . .. ... , , . , . , . " "' ......... - "
~.'-
'
,t!. '.'• •~. , ,
~
.. .
,,,,", ,
-"
't",,~
~
•
!,o ••' .••• ; .' 0;'...... ,; •• "
,
. . , ''''' ' ''
,- .. ~ . . . :':,': t ." " "" " · .. ,·t .. ·.,· . , .." ••. ' .(' "". ' :'·\· ."of " " I ..... " .(,f> L · ~
:'_ '. t'; .:., .....;:-".•'" ." ,,~' I~";" '{.~.' ,,,'. , .' . , ...· • 1 ...... ··:' .!"",:. . ' . . ' 'I .; ; '/ ''''-' ','" ~ \
..
.. .. ... J . . .... . ' . . . . .. . ..
;. .:",; { ::,~ "~ :!-:;f'.~~J"~~~' ~"'."-
A Motor Protein Consists of an ATPase Core and an Extended Structure
·' 0 ' · ,"-'.} '''1'
~
Figure 34.2 Myosin structure at low resolution. Electron m icrographs o f myosin mo lec ules reveal a two-headed st ructure w it h a long, t hi n tai l. [Courtesy of Dr. Pau la Flicker, Dr. Theo Wa lii man, and Dr. Peter Vi bert .]
978
Eukaryotic cells contain three major families of motor p roteins: myosins, kinesins, and dyneins. A t first glance, these protein families appear to be quite d ifferent from one another. Myosin, first characteri zed on the basis of its role in muscle, moves along fil aments of the protein actin . Muscle myosin consists of two copies each of a heavy chain with a molecular mass of 87 kd, an essential light chain, and a regulatory light chain. The human genome appears to encode more th an 40 distinct myosins; some function in muscle contraction, and others participate in a variety of other processes. Kinesins, which have roles in protein, mR NA, and vesicle transport as well as construction of the mitotic spindle and chromosome segregation, are generally dimers of two polypeptides. The human genome encodes more than 40 kinesins. Dyneins power the m otion of cilia and fl agella, and a general cyto· plasmic d ynein contributes to a variety of moti ons in all cells incl uding vesicle transport and various transport events in mitosis. Dyneins are enormous, with heavy chains of molecular mass greater than 500 kd. T he hu man genome appears to encode approximately 10 d yneins. Comparison of the amino acid sequences of myosins, kinesins, and dyneins did not reveal significant relationships between these protein fam· ilies but, after their three-dim ensional stru ctures were determined, memo bers of the myosin and kinesin families were found to have remarkable similarities. In particular, both myosin and kinesin contain P-loop NTPase cores homologous to those found in G proteins. Sequence analysis of the dynein heavy chain reveals it to be a member of the AM subfamily of P-Ioo p N TPases that we encountered in the context of the 19S proteasome (p. 653). Dynein has six sequences encoding such P -loop N T Pase domains arrayed along its length , although only four actually bind a nu cleotide. T hus, we can d raw on our knowledge of G proteins and other P- Ioop NTPases as we analyze the mechanisms of action of these motor proteins.
Let us first consider the structure of myosin. T he results of electron microscopic studi es of skeletal- muscle m yosin show it to be a two-headed structure linked to a long stalk (Figure 34.2). A s we saw in C hapter 33, limited proteolysis can be a powerful tool in probing the acti vity of large proteins. The treatment of myosin with trypsin and papain results in the formation of four fragments: two 51 fr agments; an 52 fragment, also called heavy
979 Trypsin
51
v. V.OW,," '
",l'''''''M\,
110 ....-_'"
S ,W'l:'f jiii
t_
52
,
•
Light meromyosin (LMM) _ _ _ __ _ __ L
_ _ _ _ _,
Y
Heavy meromyosin (HMM)
Figure 34.3 Myosin dissection. Treatment of muscle myosin with proteases forms stable fragme nts, including subfragments Sl and S2 and light meromyosin. Each Sl fragment inc ludes a head (shown in yellow o r purple) from the heavy chain and one copy of each light chain (shown in blue and o range).
meromyosin (HMM ); and a fragment called light meromyosin (LMM; Figure 34.3). Each S 1 f ragment corresponds to one of the heads from the intact stru cture and includes 850 amino-terminal amino acids from one of the two heavy chains as well as one copy of each of the light chains. Examination of the structure of an Sl fragment at high resolution reveals th e presence of a P-Ioop NTPase-domain core that is the site of ATP binding and hydrolysis (Figure 34.4). Essential light chain Regulatory light chain P-Ioop
~
Actinbinding
Figure 34.4 Myosin structure at high resolution. The stru cture of t he Sl fragment from muscle myosin revea ls the presence of a P-Ioop NTPase domain (shaded in purple). Notice that an a helix that extends from thi s domain is the binding site for the two light chains. [Drawn from 1DFL.pdb.]
Nucleotidebinding site
Extending away from this structure is a long a helix from the heavy chain. This helix is the binding site for the two light chains. The light chains are members of the EF -hand family, simil ar to calmodulin, although most of the EF hands in ligh t chains do not bind metal ions (Figure 34.5) Like calmodulin, these proteins wrap around an a helix, serving to thicken and stiffen it. The remaining fragments of myosin S2 and light meromyosinare largely a helical, forming two -stranded coiled coils created by the
Essential light chain
Regulatory light chain
Calmodulin
Mg2+
-,)",.. Figure 34.5 Myosin light chains. "0 The structures of the essential and regulatory light chains of muscle myosin are compared wi th the structure of calmodulin. Each o f these homologous protei ns binds an a hel ix (not shown) by w rapping around it . [Drawn from 1DFL.pdb and lCM1 .pdb.]
-~ Figure 34.6 Myosin two-stranded
coiled coil. The two Ct helices form left-handed supercoi led structures that sp iral aro und each other. Such stru ctu res are stabilized by hyd rophob ic residues at t he contact points between the t wo hel ices. [Draw n from 2TMA .pdb.]
Nucleotidebinding site
P-Ioop
remaining lengths of the two heavy chains wrapping around each olher (Figure 34.6). These structures, together extending approximately 1700 A, link the myosin heads to other structures . In muscle myosin , several LMM domains come together to form higher-order bundles. Cunventional kinesin (kin esin 1), the first kinesin discovered, has a structure having several features in common with myosin. The dimeric protein has two heads, connected by an extended stru cture. The size of the head domain is approximately one-third of that of myosin. D etermination of the three-dimensional structure of a kinesin frag ment revealed that the head domain also is built around a P- loop NTPase core (Figure 34 .7). The myosin domain is so much larger than that of kinesin because of two large insertions in the myosin domain thal bind to actin filaments. For conventional kinesin, a region of approximately 500 amino acids extends from the head domain . Like the corresponding region in myosin, the extended part of kinesin form s an a -helical coil ed coil. Conventional kinesin also has light chains, but, unlike those of myosin , these light chains bind near the carboxyl terminus of the heavy chain and are thought to link the motor to intracellular cargo. Dynein has a rather different structure. As noted earlier, the dynein heavy chain includes six regions that are homologous to the AAA subfamily of ATPase domains. Although no crystallographic data are yet available, the results of electron microscopic studies and comparison with known structures of ot her AAA ATPases have formed the basis for th e construction of a model of the dynein head str ucture (Figure 34. 8) . T he head domain is appended to a region of approximately 1300 amino acids that forms an extended structure that links dynein units together to form oligomers and interacts with other proteins. ATP Binding and Hydrolysis Induce Changes in the Conformation and Binding Affinity of Moto r Proteins
A key feature of P -Ioop N TPases such as G proteins is that they undergo structural changes ind uced by NTP binding and hydrolysis . Moreover, these structural changes alter their affinities for binding partners. Thus, it is not surprising that the N TPase domains of motor proteins display analogo us res ponses to nucleotide binding. The Sl fragment of myosin from
4
5
~ Figure 34.7 Structure of head
domain of kinesin at high resolution. Not ice that the head domain of kinesin has th e st ructure o f a P-Ioop NTPase core (indicated by purple shading). [Drawn fro m 11 6I.pdb.]
2
6 ;;::--- ATP 1 ~ Figure 34.S Dynein head-domain m ode l. ATP is bo und in the first o f six P-Ioop
NTPase domains (numbered) in this mo del for the head domain of d ynein. The model is based on elect ron micrographs and the stru ctures of other members o f the AAA ATPase family. The precise role o f the six si t es is no t fully understood. [Drawn fro m lHN 5.pdb.]
980
Myosin-ADP- V0 4 1 - complex
Myosi n- AOP co mplex
981 34.1 Molecular-Motor Protein ,
Lever arm
Relay helix -'---- P-Ioop : - . Switch I and switch II
'"1
Figure 34.9 Lever-arm motion. Two forms of the Sl fragment of scallop-muscle ~ myosin. Notice the dramati c conformational changes when the identity of the bound nucleotide changes from ADP-VO,' to ADP or vice versa. including a nearly 90-degree reorientatio n o f the lever arm. [Drawn f rom lDFL.pdb and 1B7T.pdb.]
scallop muscle provides a striking example of the chan ges observed (Figure 34.9 ). The stru cture of the Sl frag ment has been determin ed for Sl bound 3 to a complex fo rmed of A DP and vanadate (V0 4 - ) , whi ch is an analog of ATP, or, more precisely, the ATP-hydrolysis transition state. In the pres 3 ence of the ADP- V0 4 - complex, the long helix that binds the light chains (hereafter referred to as the lever arm ) protrudes outward from the head do3 main. In the presence of ADP without V0 4 - , the lever arm has rotated by • 3 nearly 90 degrees relative to its position in the A DP- V0 4 - complex. How does the identity of the species in the nucl eotide-binding site ca use th is dramatic transition ? Two regions around the nu cleotid ebinding site conform closely to the group in the position Position of of the ", -phosphoryl gro up of AT P and adopt a looser lever arm wh en ADP is bound conformation wh en such a grou p is absent (Figure 34 .10). Thi s co nfo rm ational change all ows a long a helix (termed the relay helix ) to adjust it s position. T he carboxyl-terminal end of the relay helix interacts with structures at the base of the lever arm, and so a change in the position of the relay helix leads to a reorientation of the Position of lever arm . lever arm wh en The binding of AT P significantly decreases the affinADP- VO.'- is bound ity of the m yosin head for actin filamen ts. No structures of myosin- actin complexes have yet been determ ined at Relay helix high resolution , so the mechanistic basis fo r this change P-Ioop remains to be elucidated. H owever, the amin o-termin al end of the relay helix interacts with the domains of myosin Switch II ~ that bind to actin , suggestin g a cl ear pathway for the cou pling of nucleotide binding to changes in actin affinity. I The importance of the changes in actin -binding affinity Figure 34.10 Relay helix. A superp os it ion of key element, in two will be clear later when we examine the role of myosin in fo rms o f scallo p myosi n revea ls th e st ruct ural changes t hat are generating directed motion (Section 34.2). tran smitted by the rel ay helix f rom th e switch I and switch II loops Analogous conformational changes take place in ki to the base o f the lever ann. The sw it ch I and swit ch II loops nesin. The kinesins also have a relay helix that can adopt interact with VO/- in th e posit ion that wo uld be occupied by different configu ratio ns when kin esin binds different nu the 'Y- pho' phory l gro up o f ATP. Th e struct ure o f th e myos in- ADP- V0 4 ' - complex is show n in lighter colors. cleotid es. Kinesin lacks an a -helical lever arm , however.
Kinesin-ATP complex
,
.--'
,,
Relay helix
.-. .. '
, Neck linker ,
Kinesin-ADP complex
., .... ... ........ ' .. ••
•
"
'
•
•• •
P-Ioop Switch I and switch II ~ Figure 34.11 Neck linker. A
comparison of the stru ctu res of a kinesin bound to ADP and bound to an ATP analog. Notice that th e neck linker (orange), whi ch connects the head domain to the remainder of the kinesin molecule. is bound to the head domai n in the presence of t he ATP analog but is free in the presence of ADP on ly. [Drawn from 1161.pdb and 1I5S.pdb.]
Instead, a relatively short segment termed the neck linker changes conformation in response to nucleotide binding (Figure 34.11). The neck linker binds to the head domain of kinesin when ATP is bound but is released when the nucleotide-binding site is vacant or occupied by ADP. Kinesin differs from myosin in that the binding of ATP to kinesin increases the affinity between kinesin and its binding partner, microtubules. Before turning to a discussion of how these properties are used to convert chemical energy into motion, we must consider the properties of the tracks along which these motors move.
34.2
Myosins Move Along Actin Filaments
Myosins, kinesins, and dyneins move by cycling between states with different affinities for the long, polymeric macrom olecules that serve as their tracks. for myosin, the molecular track is a polymeric form of actin, a 42-kd protein that is one of the most abundant proteins in eukaryoti c cells, typi cally accounting for as much as 10% of the total protein . Actin polymers are continually being assembled and disassembled in cells in a highly dynamic manner, accompanied by the hydrolysis of AT P. On the microscopic scale, actin filaments participate in the dynamic reshaping of the cytoskeleton and the cell itself and in other motility mechanisms that do not include myosin. In muscle, myosin and actin together are the key components responsible for muscle contraction.
Muscle Is a Complex of Myosin and Actin Vertebrate muscle that is under voluntary control has a banded (striated) appearance when examined under a light microscope. It consists of multinucl eated cells that are bounded by an electrically excitable plasma membrane. A muscle cell contains many parallel myofibrils, each about 1 j-Lm in diameter. The functional unit, called a sarcomere, typically repeats every 2.3 I-Lm (23, 000 A ) along the fibril axis in relaxed muscle (Figure 34.1 2). A dark A band and a light I band alternate regularly. The central region of the A band, termed the H zone, is less dense that the rest of the band . The I band is bisected by a very dense, narrow Z line. The underlying molecular plan of a sarcomere is revealed by cross sections of a myofibril. These cross sections show the presence of two kinds of intera ctin~ protein filaments. The thick filam ents have diameters of about 15 nm (150 A ) and consist primarily of m yosin . The thin filaments have 982
(A)
983 34.2 Myosin and Actin Si ngle muscle fiber (cell) --.. Nucleus Plasma membrane
Myofibrils
Single myofibril
Sarcomere
-
I band - - -.. , ,- - - - - - A
band ------~,···--- I
-.-.- -H zone---->.
Z line ~
band Z line
~
(8)
•
•
• ::- ..
•
(C)
•
• •
• • • • •• • • • •• • • • • • • • • • • • • • •• • • • •• •• • • • • • • 1
~Im
• • •
•
•• •• •• •• • • • • • • • •• •• •• • •• • • • •••• •• • •
• •
•
• • • •
•
•
•
•
•
• • • • • • • • • • • • • • • • • Thick filaments only
•
• • • • • • • • • • • • • • • • •
• • • • •• • • • • • • • •
• • • • • • • •• •• . •• . •• . • .• ••• • • • • • • •
•
· . .• •
• •
•
• •
Thick and thin filaments
•
•
•
'
•
'. • •
•
• • •• •
Thin filaments only
Figure 34.12 Sarcomere. (A) Structure of muscle cell and myofibril containing sarcomeres. (B) Electron micrograph of a longitudinal section of a skeletal-muscle myofibril, showing a single sa rcomere. (C) Schemati c representations of cross sections correspond to the regions in the micrograph . [Courtesy of Dr. Hugh Huxley.]
diameters of approximately 8 nm (80 A) and consist of actin as well as tropomyosin and the troponin comp lex. Muscle contraction is achieved through the sliding of the thin filaments along the length of the thick filaments, driven by the hydrolysis of ATP (Figure 34.13). To form the thick fi laments, myosin molecules self-assemble into thick bipolar structures with the myosin heads protruding at both ends of a bare region in the center (Figure 34.14A). Approximately 500 head domains line the surface of each thick filament . Each head-rich region associates with two
Figure 34 .13 Sliding-filament model. Muscle contraction depends on the motion of thin filaments (blue) relative to thick filaments (red). [After H. E. Huxley. The mechanism of muscular contraction. Copyright © 1965 by Scientific American, Inc. All rights reserved.]
984 CHAPTER 34
(A) Molecular Motors
(8)
,---- - - -- -- --
Z line
Thin
-
Sarcomere - - - --
Thick filament
-
- - -- -
Z line
Figure 34.14 Thick filament. (A) An electron micro graph of a reco nstituted th ick filament reveals t he presence of myosin head domains at each end and a relatively narrow cen tral region. A schematic view below shows how myosin molecules come together t o form the thick filament. (B) A diagram showing the interaction of thick and thin filaments in skeletal-muscle cont raction. [(A, top) Courtesy of Dr. Hugh Huxley.]
actin filaments, one on each side of the myosin molecules (Figure 34.14B). The interaction of individual myosi n heads with actin units creates the slid· ing force that gives ri se to muscle contraction. Tropomyosin and the troponin complex regul ate this sliding in response to nerve impulses. U nder resting conditions, tropomyosin blocks the inti· mate interaction between myosin and actin. A nerve impulse leads to an in· crease in calcium ion concentration within the muscle cell. A component of 2 the troponin complex senses the increase in Ca -, and, in response, relieves the inhibition of m yosin- actin interactions by tropomyosin. Although myosin was discovered through its role in muscle, other types of myosin play crucial roles in a number of physiological can· texts. Some defects in hearing in both mice and human beings have been linked to mutations in particular myosin homologs that are present in cells of the ear. For example, Usher syndrome in human beings and the shaker mutation in mice have been linked to myosin VIla, expressed in hair cells (Section 32.4). The mutation of this m yosin results in the formation of splayed stereocilia that do not function well. Myosin VIla differs from mus· cle myosin in that its tail region possesses a number of amino acid sequences that correspond to domains known to m ediate specific protein- protein in· teractions . Instead of assembling into fibers as muscle myosin does, myosin VIla functions as a dimer.
985
Actin Is a Polar, Self-Assembling, Dynamic Polymer
34.2 Myosin and Acti n
The structure of the actin monomer was determined to atomic resolution by x-ray crystallography and has been used to interpret the structure of actin filaments, already somewhat understood through electron microscopy studies at lower resolution . Each actin monomer comprises four domains (Figure 34.15 ). These domains come together to surround a bound nu cleotide, either ATP or ADP. The ATP form can be converted into the ADP form by hydrolysis. Nucleotidebinding site
--a.
Figure 34.15 Actin structure. (Left) '<J Schematic view of acti n monomers (one in blue) of an actin f ilament (Right) The domains in the four-d omain structure of an act in mo no mer are identified by different shades of blue. [Drawn from
lJ6Z.pdb.]
Barbed end
Actin monomers (often called G-actin for globular) come together to form actin filaments (often called F-actin; see Figure 34.15). F -actin has a helical structure; each monomer is related to the preceding one by a translation of 27.5 A and a rotation of 166 degrees around the helical axis. Because the rotation is nearly 1 RO degrees, F -actin resembles a two stranded cable. Note that each actin monomer is oriented in the same di rection along the F -actin filament, and so the structure is polar, with dis cernibly different ends. One end is called the barbed (plus) end, and the other is called the pointed (minus) end . The names "barbed" and "pointed" refer to the appearance of an actin filament when myosin S1 fragments are bound to it. How are actin filaments formed? Like many biological structures, actin filaments self-assemble; that is, under appropriate conditions, actin monomers will come together to form well -structured, polar filaments . The aggregation of the first two or three monomers to form a filament is highly unfavorable. Thus, specialized protein complexes, including one called Arp2l3, serve as nuclei for actin assembly in cel ls. Once such a filament nu cleus exists, the addition of subunits is more favorable. Let us consider the polymerization reaction in more detail. We designate an actin filament with n subunits An- This filament can bind an additional actin monomer, A, to form A,,+l.
•
+ A
986 CHAPTER 34 Molecular Motors
The dissociation constant, K d , for this reaction, defines the monomer concentrations at which the polymerization reaction will take place, because the concentration of polymers of length n + 1 will be essentially equal to that for polymers of length n. Thus, [A" HA J [An J - [An+1J andKd = [A J - [AJ 11 + 1
In other words, the polymerization reaction will proceed until the monomer concentration is reduced to the value of K d . If the monomer concentration is below the value of K d , the polymerization reaction will not proceed at all; indeed, existing filaments will depolymerize until the monomer concentration reaches the value of K d . Because of these phenomena, Kd is referred to as the critical concentration for the polymer. Recall that actin contains a nucleotide-binding site that can contain either ATP or ADP. The critical concentration for the actin- ATP complex is approximately 20 -fold lower than that for the actin- ADP complex; actin- ATP polymerizes more readily than does actin- ADP. Actin filaments inside cells are highly dynamic structures that are continually gaining and losing monomers. Nucleation by complexes such as Arp2/3 can initiate the polymerization of actin- ATP. In contrast, the hydrolysis of bound ATP to ADP favors actin depolymerization. This reaction acts as a timer to make actin filaments kinetically unstable. Proteins that bind actin monomers or promote the severing of actin filaments also play roles. Polymerization reactions can exert force, pushing or pulling on cell membranes. Regulated actin polymerization is central to the changes in cell shape associated with cell motility in amebas as well as in human cells such as macrophages.
~ Figure 34.16 Actin and hexokinase.
A comparison of actin (blue) and hexokinase from yeast (red) reveals structural similarities indi cat ive o f homology. Notice that both proteins have a deep cleft in whi ch nucleoti des bind.
A well-defined actin cytoskeleton is unique to eukaryotes; prokaryotes lack such structures. How did filamentous actin evolve? Comparisons of the three-dimensional structure of G-actin with other proteins revealed remarkable similarity to several other proteins, including sugar kinases such as hexokinase (Figure 34.16; see also p . 437). Notably, the nucleotide-binding site in actin corresponds to the ATP-binding site in hexokinase. Thus, actin evolved from an enzyme that utilized ATP as a substrate. More recently, a closer prokaryotic homolog of actin was characterized. This protein, called MreB, plays an important role in determining cell shape in rod-shaped, filamentous, and helical bacteria. The internal structures formed by MreB are suggestive of the actin cytoskeleton of eukaryotic cells, although they are far less extensive. Even though this protein is only approximately 15% identical in sequence with actin, MreB folds into a very similar three-dimensional structure. It also polymerizes into structures that are similar to F -actin in a number of ways, including the alignment of the component monomers. Motions of Single Motor Proteins Can Be Directly Observed
Muscle contraction is complex, requiring the action of many different myosin molecules. Studies of single myosin molecules moving relative to actin filaments have been sources of deep insight into the mechanisms underlying muscle contraction and other complex processes. A powerful tool for these studies, called an optical trap, relies on highly focused laser beams (Figure 34.17). Small beads can be caught in these traps and held in place in solution.
•
Laser beam
120 ~
E
Thin filament
c:
Bead
80
~
OJ
u
c:
40
-
.-c'"'"
o -40 ' --'--'-- ' -- -o 1
(A)
Glass slide
(B)
The position of the beads can be monitored with nanometer precision. James Spudich and coworkers designed an experimental arrangement consisting of an actin filament that had a bead attached to each end. Each bead could be caught in an optical trap (one at each end of the filament) and the actin filament could be pulled taut over a microscope slide containing other beads that had been coated with fragments of myosin such as the heavy meromyosin fragment (see Figure 34.17). On the addition of ATP, transient displacements of the actin filament were observed along its long axis. The size of the displacement steps was fa irly uniform with an average size of 11 nm (110 A ). The results of these studies, performed in the presence of varying concentrations of ATP, are interpreted as showi ng that individual myosin heads bind the actin filament and undergo a conformational change (the power stroke) that p ulls the actin filament, leading to the displacement of the beads . After a period of time, the myosin head releases the actin, which then snaps back into place.
2
---"-- - - -------''-----' 3 4 5
Time (s) Figure 34.17 Watching a single motor protein in action. (A) An act in f i lament (blue) is placed above a heavy meromyosin (HMM) fragment (yellow) that projects from a bead on a glass slide. A bead attached t o each end o f the actin f ilament is held in an optical trap pro duced by a f ocused, intense infrared laser beam (orange). The position o f these beads can be measured with nano meter precision. (B) Recording of the displacement of an actin filament due to a myosin derivative attached to a bead. influenced by the addition of ATP. Note the fairly uniform st ep sizes that are observed. [(A) After J. T. Finer, R. M . Simmo ns, and J. A. Spudi ch. Nature 368(1994):113- 119; (B) From R. S. Rock, M . Rief, A. D. M etra. and J. A. Spudich Methods 22(2000):378-381.]
Phosphate Release Triggers the Myosi n Po we r Stroke How does ATP hydrolysis drive the power stroke? A key observation is that the addition of ATP to a complex of myosin and actin results in the dissociation of the complex. T hus, ATP binding and hydrolysis cannot be directly responsible for the power stroke. We can co mbine this fact with the structural observations described earli er to construct a mechanism for the motion of myosin along actin (Figure 34.18). Let us begin with myosin- ADP bound to actin. The release of ADP and the binding of ATP to actin result in the dissociation of myosin from actin. As we saw earlier, the binding of ATP by its -y-phosphoryl group to the myosin head leads to a significant conformational change, amplified by the lever arm. This conformational change moves the myosin head along the actin filament by approximately 110 A. The ATP in the myosin is then hydrolyzed to ADP and Pi, which remain bound to myosin. The myosin head can then bind to the surface of actin, resulting in the dissociation of Pi from the myosin. Phosphate release, in turn, leads to a conform ational change that increases the affinity of the myosin head for actin and allows the lever arm to move back to its initial position. The conformational change associated with phosphate release corresponds to the power stroke. After the release ofP j , the myosin remains tightly bound to the actin and the cycle can begin again. How does this cycle apply to muscle contraction? Recall that hundreds of head domains project from the ends of each thick filament . The head domains are paired in myosin dimers, but the two heads within each dimer act independently. Actin filaments associate with each head-rich region, with the barbed ends of acti n toward the Z line. In the presence of normal levels of ATP, most of t he myosin heads are detached from actin. Each head can 987
Myosin
ATP
ATP
ADP ATP
ADP
\.. -{
----
)
8
Actin
(2)
o
H20
(
Figure 34.18 Myosin motion along actin. A myosin head (yellow) in the ADP form is bound t o an actin fi lament (blue). The exchange of ADP for ATP results in (1) the release of myosin from actin and (2) substantial reorientation of the lever arm of myosin. The hydrolysis of ATP (3) allows the myosin head to rebind at a site displaced along the actin filament (4). The release of P, (5) accompanying this binding increases the strength of th e interaction between myosin and actin and resets the orientation o f the lever arm.
independently hydrolyze ATP, bind to actin, release Pi, and undergo its power stroke. Because few other heads are attached, the actin filament is relatively free to slide. Each head cycles approximately five times per second with a movement of 110 A per cycle. However, because hundreds of heads are interacting with the same actin fj.laplent. the overall rate of movement of myosin relative to the actin filament may reach 80,000 A per second. allowing a sacromere to contract from its fully relaxed to its fully contracted form rapidly. Having many myosin heads briefly and independently attaching and moving an actin filament allows for much greater speed than could be achieved by a single motor protein. ,
4
-
•
~
I
V>
•
3
Wild type
E
::L
0
~
•
OIl
c .." -.-
1
•
Vl
o L-_--'--_----'_ _ - - - ' - - - - - - ' o 1 2 3 4 Number of light-chain binding sites Figure 34.19 Myosin lever-arm length. Examination of the rates of actin movement supported by a set of myosin mutants with different numbers of lightchain binding sites revea led a linear relation; the greater the number o f lightcha in binding sites (and, hence, the longer the lever arm), the fa ster the sliding velocity. [After T. Q . P. Uyeda, P. D. Abramson, and J. A. Spud ich. Proe. Natl.
Acad Sci. U.S.A. 93(1996):4459- 4464.]
The Length of the Lever Arm Determines M ot o r Velocity
A key feature of myosin motors is the role of the lever arm as an amplifier. The lever arm amplifieso small structural changes at the nucleotide-binding site to achieve the 11 O-A movement along the actin filament that takes place in each ATP hydrolysis cycle. A strong prediction of the mechani sm proposed for the movement of myosin along actin is that the length traveled per cycle should depend on the length of this lever arm. Thus. the length of the lever arm should influence the overall rate at which actin moves relative to a collection of myosin heads. This prediction was tested with the use of mutated forms of myosin with lever arms of different lengths. The lever arm in muscle myosin includes binding sites for two light chains (Section 34.1). Thus investigators shortened the lever arm by deleting the sequences that correspond to one or both of these binding sites. They then examined the rates at which acti n fil aments were transported along collections of these mutated myosi ns (Figure 34.19). As predicted. the rate decreased as th e lever arm was shortened. A
988 •
mutated form of myosin with an unusually long lever arm was generated by inserting 23 amino acids corresponding to the binding site for an additional regulatory light chain. Remarkably, this form was found to support actin movement that was Jaster than that oj the wild-type protein. These results strongly support the proposed role of the lever arm in contributing to myosin motor activity.
34.3
Kinesin and Dynein Move Along Microtubules
In add ition to actin, the cytoskel eton includes other components, notably intermediate filaments and mi crotubules. Microtubules serve as tracks for two classes of motor proteins namely, kinesins and dyneins. Kinesins moving along microtubules usually carry cargo such as organelles and vesicles from the center of a cell to its periphery. Dyneins are important in sliding microtubules relative to one other during the beating of cilia and flagella on the surfaces of some eukaryotic cells. Additionally, dynein carries cargos from the cell periphery to the cell center.
W
Some members of the kinesin family are crucial to the transport of ~ organelles and other cargo to nerve endings at the peripheries of neurons. It is not surprising, then, that mutations in th ese kin esins can lead to nervous system disorders. For example, mutations in a kinesin called KlF1B13 can lead to the most common peripheral neuropathy (weakness and pain in the hands and feet ), C harcot-Marie-Tooth disease, which affects 1 in 2500 people. A glutamine-to -Ieucine mutation in the P-Ioop of the motor domain of this kinesin has been found in some affected persons. Knockout mice with a disruption of the orthologous gene have been generated. Mice heterozygous for the disruption show symptoms similar to those observed in human beings; homozygotes die shortly after birth. Mutations in other kinesin genes have been linked to human spastic paraplegia. In these disord ers, defects in kinesin-linked transport may impair nerve func tion directly, and the decrease in the activity of specific neurons may lead to other degenerative processes. Microtubules Are Hollow Cylindrical Polymers Micro tubules are built from two kinds of homologous 50 -kd subunits,
and l3-tubulin, which assemble in a helical array of alternating tubulin types to form the wall of a hollow cylinder (Figure 34.20). Alternatively, a microtubule can be regarded as 13 protofilaments that run parallel to its
0. -
(A)
u -Tubulin
(6)
~-Tubulin
Figure 34.20 Microtubule structure. Schematic views o f t he helical structure o f a microtubule. a-Tubulin is sho wn in dark red and j3-tubulin in light red. (A) Top view. (B) Side v iew.
300
A (30 nm)
989 34.3 Kinesin and Dynein
Figure 34.21 Microtubule arrangement. Electron micrograph of a cross section of a fl agellar axoneme shows nine microtubule doublets su rrounding two Singlet s. [Courtesy of Dr. Joel Rosenbaum.]
long axis. The outer diameter of a microtubule is 30 nm, much larger than that of actin (8 nm). Like actin, microtubules are polar structures. The minus end of a microtubule is anchored near the center of a cell, whereas the plus end extends toward the cell surface. Microtubules are also key components of cilia and flagella present on some eukaryotic cells. For example, sperm propel themselves through the motion of flagella containing microtubules. The microtubules present in these structures adopt a common architecture (Figure 34.21). A bundle of mi crotubules called an axoneme is surrounded by a membrane contiguous with the plasma membrane. The axoneme is composed of a peripheral group of nine microtubule pairs surrounding two singlet microtubules. This reo curring motif is often called a 9 + 2 array. Dynein drives the motion of one member of each outer pair relative to the other, causing the overall structure to bend. Microtubules are important in determining the shapes of cell s and in separating daughter chromosomes in mitosis. They are highly dynamic structures that grow through the addition of Ct- and (3 -tubulin to the ends of existing structures. Like actin, tubulins bind and hydrolyze nucleoside triphosphates, although for tubulin the nucleotide is G TP rather than ATP. The critical concentration for the polymerization of the GTP forms of tubu· lin is lower than that for the GDP forms . Thus, a newly formed microtubule consists primarily ofGTP-tubulins. Through time, the GTP is hydrolyzed to CDP. The GDP-tubulin subunits in the interior length of a microtubule remain stably polymerized, whereas GOP subunits exposed at an end have a strong tendency to dissociate. Marc Kirschner and Tim Mitchison found that some microtubules in a population lengthen while others simultaneously shorten. This property, called dynamic instability, arises from random fluctuations in the number of GTP- or CDP-tubulin subunits at the plus end of the polymer. The dynamic character of microtubules is crucial for processes such as mitosis, which require the assembly and disassembly of elaborate microtu bule- based structures .
a -Tubulin
Minus end
Inner surface
Outer surface
Plus end
J3-Tubulin
The structure of tubulin was determined at high resolution by electron crystallographic methods (Figure 34.22). As expected from their 40% sequence identity, Ct - and (3-tubulin have very similar three-dimensional structures. Further analysis revealed that the tubulins are members of the P -Ioop NTPase family and contain a nucleotide-binding site adjacent to the P-Ioop. Tubulins are present only in eukaryotes, although a prokaryotic homolog has been found. Sequence analysis identified a prokaryotic protein called FtsZ (forjilamen· tous temperature-sensitive mutant Z) that is quite simi· lar to the tubulins. The homology was confirmed when the structure was determined by x-ray crystallography. Interestingly, this protein participates in bacterial cell di· vision, forming ring-shaped structures at the constriction that arises when a cell divides. These observations suggest that tubulins may have evolved from an ancient cell-di vision protein.
~ Figure 34.22 Tubulin. Microtubul es
can be viewed as an assembly o f a -tubulin-l3-tubulin dimers. The structures o f a -tubulin and l3-tubulin are quite similar. Notice t hat each includes a P-Ioo p NTPase doma in (purple shading) and a bound guanine nucl eotide. [Drawn from 1JFF.pdb.]
990
The continual lengthening and shortening of rnicrotubules is essential to their role in cell division. Taxol , a compound isolated from the bark of the Pacific yew tree, was discovered through its ability to interfere with cell proliferation. Taxol binds to microtubules and stabilizes the polymerized form.
o
991
/
34.3 Kinesin and Dynein
---0
o
HO
•• •
~,
HO
o
o
H
0
0 ,=
Taxol
Taxol and its derivatives have been developed as anticancer agents because they preferentially affect rapidly dividing cells, such as those in tumors.
Kinesin Motio n Is Highly Processive Kinesins are motor proteins that move along microtubules . We have seen that myosin moves along actin filaments by a process in which actin is released in each cycle; a myosin head group acting independently dissociates from actin after every power stroke. In contrast, when a kinesin molecule moves along a microtubule, the two head groups of the kinesin molecule operate in tandem: one binds, and then the next one does. A kinesin molecule may take many steps before both head groups are dissociated at the same time. In other words, the motion of kinesin is highly processive. Singlemolecule measurements allow processive motion to be observed (Figure 34.23). A single kinesin molecule will typically take 100 or more steps toward the plus end of a microtubule in a period of seconds before the molecule
-
Bead or vesicle
Kinesin
Plus end Minus end
Microtubule
Glass slide
(A)
100 ~
E 80
c
~ ~
c
:>
• • • • • • • • • • •• • • • • •• •• ••
"
•
•
,.•,
•"
':." ...
,
.
,'... .
" ,J,.:' '
"
,,......' :.. ,•••
," • •
-~.----
• • •
.... · • .. " •• • •
Bacterial Chemotaxis Depends on Reversa l of the Direction of Flagellar Rotation
Many species of bacteria respond to changes in their environments by adjusting their swimming behavior. Examination of the paths taken is highly revealing (Figure 34.30). The bacteria swim in one direction for some length oftime (typically about a second ), tumble briefly, and then set off in a new direction. The tumbling is caused by a brief reversal in the direction of the flagellar motor. When the flagella rotate counterclockwise, the helical filaments form a coherent bundle favored by the intrinsic shape of each fil ament, and the bacterium swims smoothly. When the rotation reverses, the bundle fl ies apart because the screw sense of the helical flagella does not match the direction of rotation (Figure 34.31). Each flagellum then pulls in a different direction and the cell tumbles. In the presence of a gradient of certain substances such as glucose, bacteria swim preferentially toward the direction of the higher concentration of the substance. Such compounds are referred to as chemoattractants. Bacteria also swim preferentially away from potentially harmful com pounds such as phenol , a chemorepellant. The process of moving in specific directions in respon se to environmental cues is called chemotaxis. In the presence of a gradient of a chemoattractant, bacteria swim for longer periods of time without tumbling when moving toward higher concentration s of the chemoattractant. In contrast, they tumble more frequently when moving toward lower concentrations of the chemoattractant. This behavior
•
'
'
..
. •
.' •
• •• • •• • • • • • •
-...
.. " •
.' "
•
·• ·,•," ·....
,
. .t'
•,•
I J
••
.... ,
•
•• . • :.,
• •
• • • • • • • • • •
• • • •
....•...
~'.
•
~"
)· ..
• • •••• •• •
,•
•• • •
.. ,
• • •
• • • •,
••
••
•
•••
, . • • ."
••
.'
. y•
'.'
• •• • •
50
~m
Figure 34.30 Charting a course. This projection of the track of an E. coli bacterium was obtained with a microscope that automatically follow s bacterial motion in three dimensions. The points show the locations of the bacterium at 80-ms intervals. [After H. C. Berg. Nature 254(1975):389-392.]
Figure 34.31 Changing direction. Tumbling is caused by an abrupt reversa l of the flagellar motor, which disperses the flagellar bundle. A second reversal of the motor restores smooth swimmi ng, almost always in a different direction. [Aft er a drawing kindly provided by Dr. Daniel Koshland , Jr.]
995
996 CHAPTER 34
Molecular Motors
Unoccupied
)
Clockwise rotation (tumbling)
)
Attractant
- I~
o
CheY
Counterclockwise rotation (smooth swimming)
Repellant
)
)
Clockwise rotation (tumbling)
Figure 34.32 Chemota xi s signal ing pathway. Receptors in the plasma membrane initiate a signaling pathway leading to the phosphorylation of the CheY protein. Phosphorylated CheY binds to the flagellar motor and favors clockwise rotation. When an attractant binds to the receptor, this pathway is blocked, and counterclockwise f lagellar rotation and, hence, smooth swimming result. When a repellant binds, the pathway is stimulated, leading to an increased concentration of phosphorylated CheY and, hence, more- frequent clockwi se rotation and tumbling.
is reversed for chemorepellants. T he result of these actions is a biased random walk that facilitates net motion toward conditions more favorabl e to the bacterium. Chemotaxis depends on a signaling pathway that terminates at the flagellar motor. The signaling pathway begins with the binding of molecules to receptors in the plasma membrane (Figure 34.32). In their unoccupied forms, these receptors initiate a pathway leading eventually to the phosphorylation of a specific aspartate residue on a soluble protein called Che Y. In its phosphorylated form , C he Y binds to the base on the flagellar motor. When bound to phosphorylated C heY, the fl agellar motor rotates in a clockwise rather than a counterclockwise direction , causing tumbling. The binding of a chemoattractant to a surface receptor blocks the signaling pathway leading to CheY phosphorylation. Phosphorylated CheY spontaneously hydrolyzes and releases its phosphate group in a process accelerated by another protein, CheZ . The concentration of phosphorylated C he Y drops, and the flagella are less likel y to rotate in a clockwise direction. U nder these conditions, bacteria swim smoothly without tumbling. Thus, the reversible rotary flagell ar motor and a phosphorylation-based signaling pathway work together to generate an effective means for responding to environmental conditions. Bacteria sense spatial gradients of chemoattractants by measurements separated in time. A bacterium sets off in a random direction and, if the concentration of the chemoattractant has increased after the bacterium has been swimming for a period of time, the likelihood of tumbling decreases and the bacterium continues in roughly the same direction . If the concentration has decreased, the tumbling frequency increases and the bacterium tests other random directions. The success of this mechanism once again reveals the power of evolutionary problem solving : many possible solutions are tried at random, and those that are beneficial are selected and exploited .
997 Summary 34.1 Most Molecular-Motor Proteins Are Members of the P-Loop NTPase Superfamily
Eukaryotic cells contain three families of molecular-motor proteins: myosins, kinesins, and dyneins. These proteins move along tracks defined by the actin and microtubule cytoskeletons of eukaryotic cells, contributing to cell and organismal movement and to the intracellular transport of proteins, vesicles, and organelles. Despite considerable differences in size and a lack of similarity detectable at the level of amino acid sequence, these proteins are homologous, containing core structures of the P -Ioop NTPase family. The ability of these core structures to change conformations in response to nucleoside triphosphate binding and hydrolysis is key to molecular-motor func tion . Motor proteins consist of motor domains attached to extended structures that serve to amplify the conformational changes in the core domains and to link the core domains to one another or to other structures. 34.2 Myosins Move Along Actin Filaments
The motile structure of muscle consists of a complex of myosin and actin, along with accessory proteins. Actin, a highly abundant 42 -kd protein, polymerizes to form long filaments. Each actin monomer can bind either ATP or ADP. Muscle contraction entails the rapid sliding of thin filaments, based on actin, relative to thick filaments, composed of myosin . A myosin motor domain moves along actin filaments in a cyclic manner: (1) myosin complexed to ADP and Pi binds actin; (2) Pi is released; (3) a conformational change leads to a large motion of a lever arm that extends from the motor domain, moving the actin relative to myosin; (4) ATP replaces ADP, resetting the position of the lever arm and releasing actin; and (5) the hydrolysis of ATP returns the motor domain to its initial state. The length of the lever arm determines the size of the step taken along actin in each cycle. The ability to monitor single molecular-motor proteins has provided key tests for hypotheses concerning motor function. 34.3 Kinesin and Dynein Move Along Microtubules
Kinesin and dynein move along microtubules rather than actin. Microtubules are polymeric structures composed of C/.- and l3 -tubulin, two very similar guanine- nucleotide-binding proteins . Each micro tubul e comprises 13 protofilaments with alternating C/. - and l3 -tubulin subunits. Kinesins move along microtubules by a mechanism quite similar to that used by myosin to move along actin, but with several important differences. First, ATP binding to kinesin favors motordomain binding rather than dissociation. Second, the power stroke is triggered by the binding of ATP rather than the release of Pi. Finally, kinesin motion is processive. The two heads of a kinesin dimer work together, taking turns binding and rel easing the microtubule, and many steps are taken along a microtubule before both heads dissociate. Most kinesins move toward the plus end of microtubules. 34.4 A Rotary Motor Drives Bacterial Motion
Many motile bacteria use rotating flagella to propel themselves. When rotating counterclockwise, multiple flagella on the surface of a bacterium come together to form a bundle that effectively propels the bacterium through solution . When rotating clockwise, the flagella fly apart and the bacterium tumbles . In a homogeneous environment, bacteria
Summary
998 CHAPTER 34 Molecular Motors
swim smoothly for approximately 1 s and then reorient themselves by tumbling. Bacteria swim preferentially toward chemoattractants in a process called chemotaxis. When bacteria are swimming in the direction of an increasing concentration of a chemoattractant, clockwisE flagellar motion predominates and tumbling is suppressed, leading to a biased random walk in the direction of increasing chemoattractant concentration. A proton gradient across the plasma membrane, rather than ATP hydrolysis, powers the flagellar motor. The mechanism for coupling transmembrane proton transport to macromolecular rotation appears to be similar to that used by ATP synthase.
Key Terms myosi n (p. 978) kinesin (p. 978)
sarcomere (p. 982) tropomyosin (p. 983)
tubulin (p. 990) dynamic instability (p. 990)
dynein (p . 978) SI fragment (p. 979)
troponin complex (p. 983) G -actin (p . 985)
flagellin (p. 993) MotA- MotB pair (p . 994)
conventional kinesin (p. 980)
F -actin (p. 985)
FliG (p. 994)
lever arm (p . 981) relay helix (p. 981)
critical concentration (p. 986)
chemoattractant (p . 995) chemorepellant (p . 995)
neck linker (p. 982 ) actin (p. 982 )
optical trap (p . 986) power stroke (p . 987) microtubul e (p. 989)
chemotaxis (p . 995) Che Y (p . 996)
myofibril (p. 982 )
Selected Reading Where to Start Vale, R. D. 2003. The molecular motor toolbox for intracellular trans port. Cell 11 2:467-4RO. Vale, R. D ., and M illigan, R. A . 2000 . The way th ings move: Looking under the hood of molecular motor proteins. Science 288: 88- 95. Vale, R. D. 1996. Switches, latches, and amplifiers: Common themes of G proteins and molecular motors. f. Cell Bioi. 135:291 - 302. Mehta, A D, Rief. M ., Spudich, J. A, Smith , D. A. , and Simmons, R. M. 1999. Single-molecule biomechanics with optical methods. Science 283: 1689 1695. Schuster, S. C., and Khan, S. I 994.The bacteriaillagellar motor. Annu. Rev. Biophys. Biomol. Struct. 23:509 539.
Books !-Iuwanl, ]. 200 1. M echanics of Motor Proteins and the Cytosketon.
Sinauer. Squire, J. M. 1986. Muscle Design, Diversity, and Disease. Benjamin Cummings. Pollack, G . H ., and Sugi, I-I. (Eds. ). 1984. Contractile Mechanism., in Muscle. Plenum.
Myosin and Actin Fischer , S., Windshugel, B., Horak, D ., H olmes, K. C., and Smith, j. C. 2005. Structural mechanism of the recovery stroke in the myosin molecular motor . Proc. Natl. Acad. Sci . US.A 102:6873--6878. Holmes, K. C., Angert, L, Kull, F J ., Jah n, W., and Schroder, R. R. 2003. E lectron cryo- microscopy shows how strong binding of myosin to actin releases nucleotide. Nature 425:423- 427. H olmes, K. C, Schroder, R. R., Sweeney, H. L., and Houdusse, A 2004. The structure of the ri gor complex and its impli cations for the power stroke. Philos. Trans. R. Soc. Lond. B BioI. Sci. 359: 1819 1828. Purcell , T J ., Morris, C, Spudich, J . A., and Sweeney, I-l . L. 2002. Role of the lever arm in the processive stepping of myosin V. Proc. Na tl. Acad. Sci . US.A. 99: 141 59- 14164.
Purcell, T.]., Sweeney, H . L., and Spudich, J. A 2005. A force-dependent state controls the coordination of processive myosin V. Proc. Nat/. Acad. Sci. U.S.A. 102: 13R73- 13878. Holmes, K . C . 1997. The swinging lever-arm hypothesis of muscle can· traction. Curro Bioi. 7:RI12- RI1 8. Berg, J. S., Powell, B. C ., and C heney, R. E. 2001. A millennial myosin cen sus. Mol. BioI. Cell 12:780- 794. Houdusse, A., Kalabokis, V. N., Himmel, D. , Szent-Gyorgyi, A. G., and Cohen , C. 1999. Atomic structure of scallop myosin subfrag. ment SI complexed with MgADP: A novel conformation of the myosin head . Cell 97:459-470. H oudusse, A., Szent-Gyorgyi, A. G., and Cohen, C. 2000. Three con· fo rmational states of scallop myosin S I. Proc. Na tl. Acad. Sci. U.S.A 97: 11238- 11 243 . Uyeda, T. Q., Abramson, P. D. , and Spudich, J. A 1996. The nL'Ck re· gion of the ,myosin motor domain acts as a lever arm to generate movement . Proc. Natl. Acad. Sci. US.A 93:4459-4464. Mehta, A. D. , R ock, R. S., R ief, M., Spudich , J A ., Mooseker, M. S , and C heney, R. E . 1999. Myosin-Vis a processive actin· based motor. Nature 400:590- 593. Otterbein, L. R., Graceffa, P., and Dominguez, R. 2001 . The crystal stru cture of uncomplexed actin in the ADP state. Science 293:708-7 11. Holmes, K. C., Popp, D., Gebhard, W., and Kabsch, W . 1990. Atomic model of the actin filament. Na ture 347:44- 49. Sclhutt, C . E., Myslik, J. C , Rozycki, M. D., Gooneseker., N. C., and Lindberg, U 1993. The structure of crystalline profilin-f1 -actin. Nat ure 365:81 0- 816. van den Ent, E, Amos, L. A ., and Lowe, J . 2001. Prokaryotic origin of the actin cytoskeleton . Nature 413 :39- 44 . Schutt, C. E. , and Lindberg, U 1998. Muscle contraction as a Markov process l: Energetics of th e process. Acta Physiol. Seand. 163:307- 323.
Problems 9 9 9 Rief. M ., Rock, R. S., M ehta, A. D ., Mooseker, M. S., C heney, R. E., and Spudich, J. A. 2000 . Myosin -V stepping kinetics: A molecular model for processivity. Proc. Na tl . Acad. Sci. U.S.A. 97:9482- 9486. Friedman, T B., Sellers, J. R., and Avraharn, K. B. 1999 . Uncon ventional myosins and the genetics of hearing loss. Am. f. Med. Genet. 89:147- 157.
Kinesin, Dynein, and Microtubules Yildiz, A., Tomishige, M., Vale, R. D, and Selvin , P. R . 2004 . Ki nesin walks hand-over-hand . Science 303:676-678. Rogers, G . C ., Rogers, S. L. , Schwimmer, T A., Ems-McClung, S. C., Walczak, C. E., Vale, R. D ., Scholey, J. M ., and Sharp, D . J. 2004. Two mitotic kinesins cooperate to drive sister chromatid separation during anaphase. Nature 427: 364-370 . Vale, R. D ., and Fletterick , R. J. 1997 . The design plan ofkinesin motors. Annu. Rev. Cell. Dev. BioI. 13:745- 777. Kull, F. J., Sablin, E. P., Lau, R. , Fletterick, R. J., and Vale, R. O . 1996. Crystal structure of the kinesin motor domain reveals a structural similarity to myosin. N ature 380:550- 555. Kikkawa, M., Sablin, E. P., Okada, Y., Yajima, H ., Fletterick, R. J., and Hira kawa, N. 2001. Switch -based mechanism of kinesin motors. Nature 41 1 :439-445. Wade, R. H ., and Kozielski, F. 2000 . Structural links to kinesin directionality and movement. Nat. Struct. Bioi. 7:456 460. Yun, M ., Zhang, X., Park, C. G ., Park, H . W ., and E ndow, S. A. 2001. A structural pathway for activation of the kinesin motor ATPase. EMB O f. 20 :2611 - 2618. Kozielski, F., De Boni s, S., Burmeister, W . P., Cohen -Addad, C., and Wade, R . H . 1999. The crystal structure of the minus-end-directed microtubule motor protein ned reveals variable climer conformations. S tructure Fold Des. 7:1407- 141 6. Lowe, J., Li, H ., Downing, K. H ., and Nogales, E. 2001. Refined structureofo.l3 -tubulin at3.5 A resolution.}. Mol . BioI. 313: 1045 1057 . Nogales, E., Downing, K. H ., Amos, L. A., and Lowe, J. 1998. Tubulin and FtsZ form a distinct family of GTPases. Nat. Struct. Bioi. 5:451-458. Zhao, C., Takita, J., Tanaka, Y., Setou, M ., Nakagawa, T , Takeda, S., Yang, H . W ., Terada, S., Nakata, T, Takei, Y., Saito, M ., T suji, S., Hayashi, Y., and Hirokawa, N. 2001 . C harcot -Marie-Tooth disease
type 2A caused by mutation in a microtubule motor KII'IBI3. Cell 105:587-597. Asai, D . J ., and Koonce, M . P. 2001 . 'fhe dynein heavy chain : Structure, mechanics and evolution . Trends Cell Bioi. II : 196- 202 . Mocz, G., and G ibbons, I. R. 2001. Model for the motor component of dynein heavy chain based on homology to the AAA family of oligomeric ATPases. Structure 9:93-103.
Bacterial Motion and Chemotaxis Sowa, Y, Rowe, A. D. , Leake, M . C., Yakushi , T., Homma, M. , Ishijima, A., and Berry, R. M . 2005. Direct observation of steps in rotation of the bacterial flagellar motor. Na ture 437:916-919. Berg, H . C. 2000. Constraints on models for the flagellar rotary motor. Philos. Trans. R. Soc. Lond. B Bio/. Sci. 355: 491 - 501. DeRosier, D . J. 1998. The turn of the screw : The bacterial flagellar motor. Cel/9 3: 17- 20. Ryu, W . S., Berry, R. M. , and Ucrg, H . C. 2000. Torque-generatin g uni ts of the fl agellar motor of Escherichia coli have a high duty ratio. Nature 403: 444-447 . Lloyd , S. A., Whitby, F. G ., Blair, D . F., and Hill , C. P. 1999. Stnlcture of the C -terminal domain of FliG, a component of the rotor in the bacterial flagellar motor. Nature 400:472-475. Purcell, E . M . 1977. Life at low Reynolds number. Am. j. Physiol. 45:3- 11. Macnab, R. M ., and Parkinson , J. S. 1991. Genetic anal ysis of the bacterial flagellum . Trends Genet. 7: 196- 200.
Historical Aspects Huxley, H. E. 1965. The mechanism of muscular contraction. Sci. Am. 213(6):1 8- 27. Summers, K. E., and Gibbons,!. R. 1971. ATP-induced sliding of tubules in trypsin-treated flagella of sea-urchin sperm . Proc. Natl. A cad. Sci. U.S.A. 68 :3092 3096. Macnab, R. M ., and Koshland, D . E., Jr. 1972. The gradient-sensing mechanism in bacterial chemotaxis. Proc. Nat!. A cad. Sci. US.A. 69 :2509 25 12. Taylor, E. W . 2001. 1999 E. R. Wi lwn lecture: The cell as molecular machine. Mol. Bioi. Cel/1 2:25 1- 254.
Problems I. Diverse motors. Ske letal muscle, e ukaryotic cilia, and bacte rial flagella use diffe rent strategies for the conversion of free e n ergy into coherent m o tio n . Compare and contrast these m o tility system s with resp ect to (a ) the free-en e rgy source and (b) the number of essential com pon e nts and their ide nti ty.
2. You call that slow? At maximum speed , a kinesin m olecule moves at a rate of 6400 A p e r second . G iven the dimensions of the motor region of a kinesin dimer of approximately SO A, calculate its sp eed in "body length s" per second . To what speed does this b ody -len gth sp eed correspond for an automobile 10 feet long?
3. Heavy lifting. A single myosin motor domain can gene rate a force of approximately 4 picon ewton s (4 pN). H ow many times its "body weight" can a myosin motor domain lift? Note that 1 newton = 0 .22 pounds (100 gm s). Assume a molecu lar mass of 100 kd for the motor domain .
4. Rigor mortis. Why does the body stiffen after d eath ?
5. Now you see it, now you don 't. U nder certain stable con cen tration condition s, actin m on o m ers in the ir ATP form will p o ly -
m erize to form fi laments that disperse again into free actin monomers over time. Explai n .
6. Helicases as motors. Heli eases suc h as P c rA (p . 797) can use single-stranded DNA as trac ks. rn each cycle, the helicase moves one base in the 3' ~5' directio n . Given that PcrA can hyd rolyze A TP at a rate of SO mol ecules per second in the pres ence of a single -stranded DNA template, calcul ate the velocity of the helicase in m icrometers p er second . H ow does this velocity compare with that of kinesin ?
7. New moves. When bacte ria su ch as E. coli a re starved to a sufficient extent, they become n onmotile. H owever, when su ch bacteria are placed in an acidic solution, they resume sw imming . Explain . S. llauling a load. Con sider the action of a single kinesin m o lec u le in m oving a vesicle along a microtubul e track . The force required to drag a sphe rical particle of radius a at a velocity v in a m edium having a v iscosity 7) is
F = 6'IT7)av
1 000
CHAPTER 34 Molecular Motors
Suppose thata2-~m diameter bead is carried ata velocity of 0.6 ~m s - I inan aqueous medium ('1/ = 0.0 1 poise = 0.0 1 gcm - I S- I).
The rate of ATP hydrolysis by myosin has been examined as a fun ction of ATP concentration, as shown in graph A .
(al What is the magnitude of the force exerted by the kinesin molecule? Exp ress the value in dynes (1 dyne = 1 g cm 5- 2 ). (b 1 How much work is performed in 1 s? Express the value in ergs (1 erg = 1 dyne em). (c) A kinesin motor hydrolyzes approximately 80 m olecules of ATP per second . What is the energy associated with the hydrol ysis of this much ATP in ergs? Compare this value with the actual work performed.
- '"
9. Un usual strides. A p ublication describes a kinesin molecule
'"
that is claimed to move along mi crotubules with a step size of 6 nm. You are skeptical. Why? 10. The sound oj one hand clapping. K I F1A is a motor protein that moves toward the plus end of microtubules as a monomer. KIFl A has on ly a sin gle motor domain . What add itional structural elements wou ld you expect to find in the KIF1A structure?
14 12
.'"
'" 0 '1
'"
10
>-c: ..c: ._
8
'">-
6
>-~
~ "O~
0- 0
!;;: E 0
-''""
~
•
•
• •
• •
•
a. 4 2
a
50
100
[ATP],
(A)
150
200
~M
(al Estimate t he valu es of keat and KM for ATP. With the use of optical-trap measurements, the motion of si ngle myosin V dimers could be followed, as shown in graph B.
Mechanism Problem
11. Backward rotation. O n the basis of the proposed structure in Figure :H.30 for the bacterial flagellar motor, suggest a pathway for transmembrane proton flow when the flagellar motor is rotating clockwise rather than counterclockwise.
400 350
E ..s
-
300
c:
Chapter Integration Problem
12. Smooth muscle. Smooth muscle, in contrast with skeletal muscle, is not regulated by a tropom yosin- troponin mechanism. I nstead, vertebrate smooth-muscle contraction is controlled by the degree of phosphorylation of its light chains. Phosphorylation in duces contraction, and dephosphorylation leads to relaxation . Like that of skeletal muscle, smooth-muscle contraction is tri ggered by an increase in the cytoplasmic calcium ion level. Propose a mechanism for this action of calcium ion on the basis of your knowledge of other signal-transduction processes.
'" '" Ci. '"
E 250 u
.-Cl'"
200
150 100
(8)
o
0 .2
0.4
0 .6
O.B
1
Time (5)
[Based on M. Rief, R. S. Rock, A. D. Mehta, M. s. Mooseker, R. E. Cheney, and J. A. Spudich. Proc. Natl. Acad. Sci. u.s.A. 97(2000):9482- 9486.]
Data Interpretation Problem
13. Myosin V An abundant myosin -family member, myosin V is isolated from brain tissue. T hi s myosin has a number of unusual properties. First, on the basis of its amino acid sequence, each heavy chain has six tandem binding sites for calmodulin -like li ght chains . Second , it forms dimers but n ot higher -order oligomers. Finally, unlike almost all other myosin-family mem bers, myosin V is highly processive.
(b ) Estimate the step size for myosin V. The rate of ADP release from myosin V is found to be approximately 13 molecul es s - I. . (cl Combine the observations abou t the amino acid sequence of myosin, the observed step size, and the kinetics results to propose a mechanism for the processive motion of myosin V.
Chapter
Drug Development COOH
o
Many drugs are based on natural products. Aspi rin (above) is a chemical derivative of a compound isolated from willow bark (near left). Extracts of willow bark had been long known to have medicinal properties. The active compo und was isolated, modified, and, beginning in 1899, packaged for consumers (far left). [For left: Used with permission of Bayer Corporation. Near left: Image Ideas/ Picture Quest.]
he development of drugs represents one of the most important interfaces between biochemistry and medicine. In most cases, drugs act by binding to specific receptors or enzymes and inhibiting, or otherwise modulating, their activities. Thu s, knowledge of th ese molecules and the pathways in which they participate is crucial to drug development. An effective drug is much more than a potent modulator of its target, however. Drugs must be readily administered to patients, preferably as small tablets taken orally, and must survive within the body long enough to reach their targets . Furthermore, to prevent unwanted physiological effects, drugs must not modulate the properties of biomolecules other than the target molecules. These requirements tremendously limit the number of compounds that have the potential to be clinically useful drugs. Drugs have been discovered by two, fundamentally opposite, approaches (Figure 35.1). The first approach identifies a substance that has a desirable physiological consequence when administered to a human being, to an appropriate animal, or to cells. Such substances can be discovered by serendipity, by the fractionation of plants or other materials known to have medicinal properties, or by screening natural products or other "libraries" of compounds. In this approach, a biological effect is known before the molecular target is identified . The mode of action of the substance is only later
Outline 35.1 The Development of Drugs Presents Huge Challenges 35.2 Drug Candidates Can Be Discovered by Serendipity, Screening, or Design 35.3 The Analysis of Genomes Holds Great Promise for Drug Discovery 35.4 The Development of Drugs Proceeds Through Several Stages
1001
1002
(A)
CHAPTER 35 Drug Development Compound
)
Molecular target
)
Physiological effect
Molecular target
----------~
(8) Compound
Physiological effect
)
Figure 35.1 Two paths to drug discovery. (A) A compound is discovered t o have a desirable physiolo gical effect . The molecular target can be identified in a separate step as needed. (B) A mo lecular target is selected f irst . Drug candidates that bind to the target are identified and then examined for their physiological effects.
-
Pharmacology The science that deals with the discovery. chemistry. composition, identification. biologi cal and physio logical effect s, uses, and manufacture of drugs.
.
-
identified after substantial additional work. The second approach begins with a known molecular target. Compounds are sought, either by screening or by designing molecules with desired properties, that bind to the target molecule and modulate its properties. Once such compounds are available, scientists can explore their effects on appropriate cells or organisms. Many unexpected results may be encountered in this process as the complexity of biological systems reveals itself. In this chapter, we explore the science of pharmacology. We examine a number of case hi stories that illustrate drug development including many of its concepts, methods, and challenges. We then see how the concepts and tools from genomics are influencing approaches to drug development. We conclude the chapter with a summary of the stages along the way to developing a drug.
35.1
The Development of Drugs Presents Huge Challenges
Many compounds have significant effects when taken into the body, but only a very small fraction of them have the potential to be useful drugs. A foreign compound, not adapted to its role in the cell through long evolution, must have a range of special properties to function effectively without causing serious harm. We next review some of the challenges faced by drug developers.
1.0
-
c
.- -
o~
-",
Drug Candidates Must Be Potent Modulators of Their Targets
~
:::l
+
r o_
~ci 0.5 ro-
c~
0;:::::'
. - ...J
~
u:
es
I I I I
i/ I
[Ligand] = Kd
0 ---""---- - - - -- [Ligand] Figure 35.2 Ligand binding. The titration of a recepto r, R, w ith a ligand. L. results in t he forma t ion of th e compl ex RL. In uncompl icated cases, the binding reacti o n f ollows a simple saturatio n curve. Half o f the receptors are bound to ligand wh en the ligand concentrat ion equals the dissoc iati on constant, Kd , fo r the RL complex.
Most drugs bind to specific proteins, usually receptors or enzymes, within the body. To be effective, a drug needs to bind a sufficient number of its target proteins when taken at a reasonable dose. O ne factor in determining drug effectiveness is the strength of binding, often governed by the principles of binding, related to the Michaelis-Menten model introduced in Chapter 8. A molecul e that binds to some target molecule is often referred to. as a ligand. A ligand-binding curve is shown in Figure 35.2 . Ligand molecules occupy progressively more target binding sites as ligand concentration increases until essentially all of the available sites are occupied. The tendency of a ligand to bind to its target is measured by the dissociation constant, Kd , defined by the expression •
Kd = [R][L]/[RL]
where [R) is the concentration of the receptor, [L) is the concentration of the ligand, and [RL) is the concentration of the receptor- ligand complex. The dissociation constant is a measure of the strength of the interaction between the drug candidate and the target; the lower the value, the stronger the interaction. The concentration of free ligand at which one-half of the binding sites are occupied equals the dissociation constant, as long as the concentration of binding sites is substantiall y less than the dissociation constant. Many complicating factors are present under physiological conditions. Many drug targets also bind ligands normally present in tissues; these ligands and the drug candidate compete for binding sites on the target. We en countered this situation when we considered competitive inhibitors in Chapter 8. Suppose that the drug target is an enzyme and the drug candi date is a competitive inhibitor. The concentration of the drug candidate necessary to inhibit the enzyme effectively will depend on the physiological concentration.of the enzyme's normal substrate (Figure 35.3). The higher the concentration of the endogenous substrate, the hi gher the concentration of drug candidate needed to inhibit the enzyme to a given extent. This effect of substrate concentration is expressed by the apparent dissociation constant, KdPP The apparent dissociation constant is given by the expression
where [S) is the concentration of substrate and KM is the Michaelis constant for the substrate. Note that, for an enzyme inhibitor, the dissociation con stant, K d , is often referred to as the inhibition constant, K i . In many cases, more complicated bio[ogical assays (rather than direct enzyme or binding assays) are used to examine the potency of drug candidates. For example, the fraction of bacteria killed might indicate the potency of a potential antibiotic. In these cases, values such as EC so are used. EC so is the concentration of drug candidate required to elicit 50% of the maximal biological response (Figure 35.4). Similarly, EC 90 is the concentration required to achieve 90% of the maximal response. In the example of an antibiotic, EC 90 would be the concentration required to kill 90% of bacteria exposed to the drug. For inhibitors, the corresponding terms IC so and IC 90 are often used to describe the concentrations of the inhibitor required to reduce a re sponse to 50% or 90% of its value in the absence of inhibitor, respectively. These values are measures of the potency of a drug candidate in modu lating the activity of the desired biological target. To prevent unwanted effects, often called side effects, ideal drug candidates should not bind biomolecules other than the target to any appreciable extent. Developing such a drug can be quite challenging, particularly if the drug target is a member of a large family of evolutionarily related proteins. The degree of specificity can be described in terms of the ratio of the Kd values for the binding of the drug candidate to any other molecules to the Kd value for the binding of the drug candidate to the desired target. Drugs Must Have Suitable Properties to Reach Their Targets
Thus far, we have focused on the ability of molecules to act on specific target molecules. However, an effective drug must also have other characteristics. Tt must be easily administered and must reach its target at sufficient concen tration to be effective. A drug molecule encounters a variety of obstacles on its way to its target, related to its absorption, distribution, metabolism, and excretion after it has entered the body. These processes are interrelated to one another as summarized in Figure 35.5. Taken together, a drug's ease of absorption, distribution, metaboli sm, and excretion are often referred to as ADME (pronounced "add-me") properties.
1.0
No substrate
c
-'" --'"'"
.-0
Low substrate
~
::J
0.5
-
Higher substrate
'"C0
I I I
.-
U
:
'" ~
u..
[Ligand] = K~PP
:/
0
[Ligand]
•
Figure 35.3 Inhibitors compete with substrates for binding sites. These binding curves give results for an inhibitor binding to a target enzyme in the absence of substrate and in the presence of increasing concentrations of substrate.
~
~
~
100 90
-- -------------------- I I I I
'"'"c
o0..
'"~
-:g .-M -oo
.-eD
I
50
I
-
I I
:
[Ligand] : [Ligand] = EC so : = Ee go
:/
:/
o --~----~~---------[Ligand)
,
Figure 35.4 Effective concentration s. The concentrati on of a ligand required t o elicit a biological response can be quantified in terms of EC sQ, the concentration required to give 50% of the maximum response, and EC90 , the concentration required to give 90% o f the maximum response.
1003
1004
Target compartment
CHAPTER 35 Drug Development
Other compartments Bound
Bound
1
Free
Free
DISTRIBUTION Bloodstream ABSORPTION
Figure 35.s Absorption, distribution, m etabolism, and excretion (ADM E), The concentration of a com pound at its target site (yellow) is affected by the extent s and rates of absorption, dist ribution, metabolism, and excretion.
Free ,
>
Bound
EXCRETION
Metabolites METABOLI SM Transformation
Administration and Absorption. Ideally, a drug can be taken orally as a small tablet. An orally administered active compound must be able to survive the acidic conditions in the gut and then be absorbed through the intestinal epithelium. Thus, the compound must be able to pass through cell membranes at an appreciable rate. Larger molecules such as proteins cannot be administered orally, because they often cannot survive the acidic conditions in the stomach and, if they do, are not readily absorbed. Even many small molecules are not absorbed well, because, for example, if they are too polar they do not pass through cell membranes readily. The ability to be absorbed is often quantified in terms of the oral bioavailability. This quan tity is defined as the ratio of the peak concentration of a compound given orally to the peak concentration of the same dose injected directly into the bloodstream. Bioavailability can vary considerably from species to species so results from animal studies may be difficult to translate to human beings. Despite this variability, some useful generalizations have been made. One powerful set is Lipinski's rules. Lipinski's rules tell us that poor absorption is likely when
H
Two hydrogen0 bond donors
H
N' H
Four hydrogenbond acceptors
H
Molecular weight = 285
log(P) = 1.27 Figure 35.6 Lipinski's rules applied to morphine. Morphine satisfies all of Lipinski's rules and has an oral bioavailability in human beings of 33%.
1.
the molecular weight is greater than 500 .
2.
the number of hydrogen-bond donors is greater than 5.
3.
the number of hydrogen-bond acceptors is greater than 10.
4.
the partition coefficient [measured as 10g(P )] is greater than S.
The partition coefficient is a way to measure the tendency of a molecule to dissolve in membranes, which correlates with its ability to dissolve in organic solvents. It is determined by allowing a compound to equilibrate between water and an organic phase, n-octanol. The 10g(P) value is defined as logl o of the ratio of the concentration of a compound in n-octanol to the concentration of the compound in water. For example, if the concentration of the compound in the n-octanol phase is 100 times that in the aqueous phase, then 10g(P ) is 2. Morphine, for example, satisfies all of Lipinski's rul es and has moderate bioavailability (Figure 35.6). A drug that violates one or more of these rules may still have satisfactory bioavailability. Nonetheless, these rules serve as guiding principles for evaluating new drug candidates.
1005 35.1 Drug Development Challenges
~ Figure 35.7 Structure of
the drug carrier human serum albumin . Seven hydrophobic molecules (in red) are shown bound to the molecule. [Drawn from 1BKE.pdb.]
Distribution. Compo unds taken up by intestinal epithelial cells can pass into the bloodstream. However, hydrophobic compounds and many others do not freely dissolve in the bloodstream. These compounds bind to proteins, such as albumin (Figure 35.7), that are abundant in the blood serum and by this means are carried everywhere that the bloodstream goes. When a compound has reached the bloodstream, it is distributed to difrerent fluids and tissues, which are often referred to as compartments. Some compounds are highly concentrated in their target compartments, either by binding to the target molecules themselves or by other mechanisms. Other compounds are distributed more widely (Figure 35.8). An effective drug will reach the target compartment in sufficient quantity; the concentration of the compound in the target compartment is reduced whenever the compound is distributed into other compartments. Some target compartments are particularly hard to reach . Many compounds are excluded from the central nervous system by the blood brain
F
F Fluconazole
Figure 35.8 Distribution of the drug fluconazole. O nce taken in, compounds distribute themselves to various organs within the body. The distribution of the antifunga l agent f luconazole has been monit ored thro ugh t he use of positron emission tomography (PET) scanning. These images were taken of a healthy human volunteer 90 minutes after 1 injection of a dose of 5 mg kg - of flu conazole containing trace amounts of flu conazole labeled with the positronem itt ing isotope 1BF. [From A. J. Fisch man et al.. Antimicrob. Agents Chemother. 37(1993): 1270-1277.]
•
1006 CHAPTER 35 Drug Development
barrier, the tight junctions between endothelial cells that line blood vessels within the brain and spinal cord.
Figure 35.9 P450 conversion of ibuprofen. Cytochrome P450 isozymes, primarily in the liver, catalyze xenobiotic metabolic reactio ns such as hydroxylation. The rea ctio n introduces an oxygen atom derived from molecular oxygen.
Metabolism and Excretion. A final challenge to a potential drug molecule is to evade the body's defenses against foreign compounds. Such compounds (often called xenobiotic compounds) are often released from the body in the urine or stool, often after having been metabolized somehowdegraded or modified to aid in excretion. This drug metabolism poses a considerable th.reat to drug effectiveness because the concentration of the desired compound decreases as it is metabolized. Thus, a rapidly metabolized compound must be administered more frequently or at higher doses. Two of the most common pathways in xenobiotic metabolism are oxidation and conjugation. Oxidation reactions can aid excretion in at least two ways: by increasing water solubility, and thus ease of transport, and by introducing functional groups that participate in subsequent metabolic steps. These reactions are often promoted by cytochrome P450 enzymes in the liver (p. 750). T he human genome encodes more than 50 different P450 isozymes, many of which participate in xenobiotic metabolism. A typical reaction catalyzed by a P450 isozyme is the hydroxylation of ibuprofen (Figure 35.9).
NADPH + H+ + 0 , +
H
COOH - - - +) NADP+ + H 20
-.........
+
HO COOH
Ibuprofen
Conjugation is the addition of particular groups to the xenobiotic compound, Common groups added are glutathione (p. 586), glucuronic acid, and sulfate (Figure 35.10). The addition often increases water solubility and provides labels that can be recognized to target excretion, Examples of conjugation include the addit ion of glutathione to the anticancer drug cyclophosphamide, the addition of glucuronidate to the analgesic morphine, and the addition of a sulfate group to the hair-growth stimulator minoxidil. CI
0
COO
II / N
o ...--- P"
S NH
0
H -
HN ~
~
0
COO
H2N CH 3
I
HO OH
0
NH
N H
'" '" ,
NH3+
0 -0
H OOC
N N
N
/
"'::- s
0 / , \• 0
HN
HO
H
Cyclophosphamide-glutathione conjugate
Morphine glucuronidate
Minoxidil sulfate
Interestingly, the sulfation of minox idil produces a compound that is more active in stimulating hair growth than is the unmodified compound . Thus, the metabolic products of a drug, though usually less active than the drug, can sometimes be more active, Note that an oxidation reaction often precedes conjugation because the oxidation reaction can generate hydroxyl and other groups to which groups such as glucuronic acid can be added, The oxidation reactions of xenobiotic compounds are often referred to as phase I transformations, and the conjugation reactions are referred to as phase II transformations, These reactions take
SR
SH
o •
'.
H
0
H
,
OOC
H
,
.1 \
H
o
H '" N "-../COO
•
Figure 35.23 The effect of anti-HIV drug development. Deat h rat es fro m HIV infection (A IDS) revea l th e tremendous effect o f HIV protease inhi bit ors and th ei r use in combinat ion w ith inhibito rs o f HIV reve rse t ranscript ase, These are death rates from the lead ing causes of death among persons 24 to 44 years o ld in the Un ited States, [From Centers fo r Di sease Contro L]
fr 0.
30
g
25
o
~-
20
- - Heart disease
---- --- -- -- --- -- - ---
---- - --- ~
~--,.~_ _ _- - - - - - ; - - -_ _
~
'"
0. 15
------
-5'"
------ - ------- ---- ----
'" 10
'"
C
5
-------------------
-------- ---------
- - - Suicide - - Homicide - - - Chronic liver disease - - Cerebrovascular diseases - - - Diabetes
------- - - -------------- - -----------o '--- --'--'---'--"----'---'-- --:::';:::--'-:-; :::- -;:;! 1988
1990
1992
1994
1996
1998
2000
treat AIDS with much more encouraging results than had been obtained previously (Figure 35 ,23), Aspirin targets the cyclooxygenase site in prostaglandin H2 synthase, as discussed earlier. Animal studies suggested that mammals contain not one but two distinct cyclooxygenase enzymes, both of which are targeted by aspirin, The more recently discovered enzyme, cyclooxygenase 2 (COX2), is expressed primarily as part of the inflammatory response, whereas cyclooxygenase 1 (COX1) is expressed more generally, These observations suggested that a cyclooxygenase inhibitor that was specific for COX2 might be able to reduce inflammation in conditions such as arthritis without producing the gastric and other side effects associated with aspirin, The amino acid sequences of COXl and COX2 were deduced from cDNA cloning studies, These sequences are more than 60% identical, clearly indicating that the enzymes have the same overall structure, Nevertheless, there are some differences in the residues around the aspirin-binding site, X -ray crystallography revealed that an extension of the binding pocket was present in COX2, but absent in COX 1 , This structural difference suggested a strategy for constructing COX2-specific inhibitors namely, to synthesize compounds that had a protuberance that would fit into the pocket in the COX2 enzyme, Such compounds were designed and synthesized and then further refined to produce effective drugs familiar as Celebrex and Vioxx (Figure 35 ,24), Vioxx was subsequently withdrawn from the market because some individuals experienced adverse events, These effects appear to be due to the inhibition of COX2, the intended target, Thus, although the development of these drugs is a triumph for structure-based drug design, these outcomes highlight the fact that the inhibition of important enzymes can lead to complex physiological responses,
o
o Figure 35,24 COX2-specific inhibitors, These co mpounds have protuberances (shown in red) t hat fit into a pocket in the COX2 isozyme but steri cally clash with the COXl isozyme,
/,/
o Celecoxib (Celebrex)
Rofecoxib (Vioxx)
"\
0
35.3
1017 35.3 The Promise of Genome Analysis
The Analysis of Genomes Holds Great Promise for Drug Discovery
The completion of the sequencing of the human and other genomes is a potentially powerful driving force for the development of new drugs . Genomic sequencing and analysis projects have vastly increased our knowledge of the proteins encoded by the human genome. This new source of knowledge may greatly accelerate early stages of the drug -development process or even allow drugs to be tailored to the individual patient. Potential Targets Can Be Identified in the Human Proteome
•
The human genome encodes approximately 25 ,000 proteins, not countin g the variation produced by alternative mRNA splicing and posttranslational modifications. Many of these proteins are potential drug targets, in particular those that are enzymes or receptors and have significant biological effects when activated or inhibited . Several large protein families are particularly ri ch sources of targets. For example, the human genome includes genes for more than 500 protein kinases that can be recognized by comparing the deduced am ino acid sequences. One of them, Bcr-Abl kinase, is known to contribute to leukemias and is the target of the drug imatinib mesylate (G leevec; p . 401) . Some of t he other protein kinases undoubtedly play central roles in particular cancers as well. Similarly, the human genome encodes approximately 800 7TM receptors (p. 383) of which approximately 350 are odorant receptors . Many of the remaining 7TM receptors are potential drug targets. Some of them are already targets for drugs, such as the ~-b l ocker atenolol, which targets the j3-adrenergic receptor, and the antiulcer medication ranitidine (Zantac). The latter compound is an antagonist of the histamine H2 receptor, a 7TM receptor that participates in the control of gastric acid secretion .
o
N H H
OH
H N NH2
o
N0 2 Atenolol
Ranitidine
Novel proteins that are not part oflarge families already supplying drug targets can be more readi ly identified through the use of genomic information. T here are a number of ways to identify proteins that could serve as targets of dru g-development programs . O ne way is to look for changes in expression patterns, protein localization , or posttranslational modifications in cells from disease-afflicted organisms. Another is to perform studies of tissues or cell types in which particular genes are expressed. Analysis of the human genome should increase the number of actively pursued drug targets by a factor of an estimated two or more.
1018 CHAPTER 35 Drug Development
Animal Models Can Be Developed to Test the Validity of Potential Drug Targets The genomes of a number of model organisms have now been sequenced. The most important of these genomes for drug development is that of the mouse. Remarkably, the mouse and human genomes are approximately 85% identical in sequence, and more than 98% of all human genes have recognizable mouse counterparts. Mouse studies provide drug developers with a powerful tool the ability to disrupt ("knock out") specific genes in the mouse (p. 155). If disruption of a gene has a desirable effect, then the product of this gene is a promising drug target. The utility of this approach has been demonstrated retrospectively. For example, disruption of the gene for the ex subunit of the H+ -K+ ATPase, the key protein for secreting acid into the stomach, produces mice with less acid in their stomachs. The stomach pH of such mice is 6.9 in circumstances that produce a stomach pH of 3.2 in their wild-type counterparts. This protein is the target of the drugs omeprazole (Prilosec) and lansoprazole (Prevacid and Takepron), used for treating gastric-esophageal reflux disease. H
N
§
H3C~
0
N
o
II
\
0
H3C
' ~
H
CH 3
CH,
N
~./
§ N
o
II
H3C
\ ~
N Omeprazole
0\ CFJ
\
N
;)
Lansoprazole
Several large-scale efforts are underway to generate hundreds or thousands of mouse strains, each having a different gene disrupted. The phenotypes of these mice are a good indication of whether the protein encoded by a disrupted gene is a promising drug target. This approach allows drug developers to evaluate potential targets without any preconceived notions regarding physiological function.
Potential Targets Can Be Identified in the Genomes of Pathogens Human proteins are not the only important drug targets. Drugs such as penicillin and HIV protease inhibitors act by targeting proteins within a pathogen. The genomes of hundreds of pathogens have now been sequenced, and these genome sequences can be mined for potential targets. New antibiotics are needed to combat bacteria that are resistant to many existing antibiotics. One approach seeks proteins essential for cell survival that are conserved in a wide range of bacteria. Drugs that inactivate such proteins are expected to be broad-spectrum antibiotics, useful for treating infections from any of a range of different bacteria. One such protein is peptide deformylase, the enzyme that removes formyl groups that are present at the amino termini of bacterial proteins immediately after translation (p. 871). Alternatively, a drug may be needed against a specific pathogen. A recent example of such a pathogen is the organism responsible for severe acute respiratory syndrome (SARS). Within one month of the recognition of this emerging disease, investigators had isolated the virus that causes the syndrome, and, within weeks, its 29,751- base genome had been completely sequenced. This sequence revealed the presence of a gene encoding a viral protease, known to be essential for viral replication from studies of other members of the coronavirus family to which the SARS virus belongs. Drug developers are already at work seeking specific inhibitors of this protease (Figure 35.25).
1018 CHAPTER 35 Drug Development
Ani ma l Models Can Be Developed to Test the Val idity of Potential Drug Targets The genomes of a number of model organisms have now been sequenced. The most important of these genom es for drug development is that of the mouse. Remarkably, the mouse and human genomes are approximately 85% identical in sequence, and more than 98% of all human genes have recognizable mouse counterparts. Mouse studies provide drug developers with a powerful tool the ability to disrupt ("knock out" ) specific genes in the mouse (p . 155). If disruption of a gene has a desirable effect, then the product of this gene is a promising drug target . The utility of this approach has been demonstrated retrospectively. For example, disruption of the gene for the Cl subunit ofthe H + -K + ATPase, the key protein for secreting acid into th e stomach , produces mi ce with less acid in their stomachs. The stomach pH of such mice is 6.9 in circumstances that produce a stomach pH of 3.2 in their wild-type counterparts. This protein is the target of the drugs omeprazole (Prilosec) and lansoprazole (Prevacid and Takepron), used for treating gastric-esophageal reflux disease. H
N
II
H, C ~
0
o
I; 5
0
H,C
N
H
CH,
CH,
N
~
./
II
o
H, C
CF,
N
N
Ome pral ole
0\
I; 5
N
Lansopralole
Several large-scale efforts are underway to generate hundreds or thousands of mouse strains, each having a different gene disrupted . The phenotypes of these mice are a good indication of whether the protein encoded by a disrupted gene is a promising drug target. This approach allows drug developers to evaluate potential targets without any preconceived notions regarding physiological fu nction.
Potential Targets Can Be Identified in the Genomes of Pathogens Human proteins are not the only important drug targets. Drugs such as penicillin and H IV protease inhibitors act by targeting proteins within a pathogen . T he genomes of hundreds of pathogens have now been seq uenced, and these genom e sequences can be mined for potential targets. New antibiotics are needed to combat bacteria that are resistant to many existing antibiotics. One approach seeks proteins essential for cell survival that are con served in a wide range of bacteria. Drugs that inactivate such proteins are expected to be broad-spectrum antibiotics, useful for treating infections from any of a range of different bacteria. One such protein is peptide deformylase, the enzyme that removes formyl groups that are present at the amino termini of bacterial proteins immediately after translation (p . 871 ). Alternatively, a drug may be needed against a specific pathogen . A recent example of such a pathogen is the organism responsible for severe acute respiratory syndrome (SARS). Within one month of the recognition of this emerging disease, investigators had isolated the virus that causes the syndrome, and, within weeks, its 29,751 - base genome had been completely seq uenced. This sequence revealed the presence of a gene encoding a viral protease, known to be essential for viral replication from studies of other members of the coronavirus family to which the SARS virus belongs. Drug developers are already at work seeking specific inhibitors of this protease (Figure 35. 25).
1019 35.3 The Prom ise of Genom e Analysis
~ Figure 35.25 Emerging drug target. The structure of a protease from the coronavirus
that causes SARS (severe acute respiratory syndrome) is shown bound to an inhibitor. This structure was determined less than a year after the identification of the virus. [Drawn from lP9S.bdb.]
Genetic Differences Influence Individual Responses to Drugs
Many drugs are not effective in everyone, often because of genetic differences between people. Nonresponding persons may have slight differences in either a drug's target molecule or proteins taking part in drug transport and metabolism. The goal of the emerging fields of pharmacogenetics and pharmacogenomics is to design drugs that either act more consistently from person to person or are tailored to individuals with particular genotypes. Drugs such as metoprolol that target the [31 -adrenergic receptor are popular treatments for hypertension. H
0
/ H,C
f
0
\
OH
H N H
CH, CH,
Metoprolol
But some people do not respond well. Two variants of the gene coding for the [31 -adrenergic receptor are common in the American population . The most common allele has serine in position 49 and arginine in position 389. In some persons, however, glycine replaces one or the other of these residues. In studies, participants with two copies of the most common allele responded well to metoprolol: their daytime diastolic blood pressure was reduced by 14.7 :t: 2.9 mm Hg on average. In contrast, participants with one variant allele showed a smaller reduction in blood pressure, and the drug had no significant effect on participants with two variant alleles (Figure 35.26). These observations suggest the potential utility of genotyping individuals at these positions. One could then predict whether or not treatment with metoprolol or other [3-blockers is likely to be effective. Given the importance of ADME and toxicity properties in determining drug efficacy, it is not surprising that variations in proteins participating in drug transport and metabolism can alter a drug's effectiveness. An important example is the use of thiopurine drugs such as 6-thioguanine, 6mercaptopurine, and azothioprine to treat diseases including leukemia, immune d isorders, and inflammatory bowel disease.
"be I
i5E
.2E ..0
....
~
'"
:.=c: 0:.=
o -2
-
-4 -
-6
I-
~ ~
-8 .."..0 - 10 .5 E - 12 l00- - 14 Ic: '" ..c: :O - 16 u~ - 18
'"
'" e '" -
0..
SRISR SRI GR SRISG GRISG
Figure 35.26 Phenotype-genotype correlation. Average changes in diastolic blood pressure o n treatment with metoprolol. Persons with two copies of the most common (S49R389) allele showed significant decreases in blood pressure. Those with one variant allele (GR or SG) showed more modest decreases. and those with two variant alleles (GR /SG ) showed no decrease. [From J. A. Johnson et al.. Clin. Pharmacal. Ther. 74(2003):
44- 52.]
1020 CHAPTER 35
/
N0 2
Drug Development
SH
SH
5
H N
H
H2N 6-Thioguanine
N
H N
j
j
N
N
6-Mercaptopurine
Azathioprine
A minority of patients who are treated with these drugs show signs of toxicity at doses that are well tolerated by most patients . These differences between patients are due to rare variations in the gene encoding the xenobioticmetabolizing enzyme thiopurine methyl transferase, which adds a methyl group to sulfur atoms. SH H ____ N
;;
+ S-adenosylmethionine
Thiopurine methyltransferase
,-~~===='
H N
+ S-adenosylhomocysteine + H+
N 6-Mercaptopurine
The variant enzyme is less stable. Patients with these variant enzymes can build up toxic levels of the drugs if appropriate care is not taken. Thus, genetic variability in an enzyme participating in drug metabolism plays a large role in determining the variation in the tolerance of different persons to particular drug levels. Many other drug-metabolism enzymes and drugtransport proteins have been implicated in controlling individual reactions to specific drugs. The identification of the genetic factors will allow a deeper understanding of why some drugs work well in some persons but poorly in others. In the future, doctors may examine a patient's genes to help plan drug-therapy programs.
35.4
The Development of Drugs Proceeds Through Several Stages
In the United States, the FDA requires that drug candidates be demonstrated to be effective and safe before they may be used in human beings on a large scale. This requirement is particularly true for drug candidates that are to be taken by people who are relatively healthy. More side effects are acceptable for drug candidates intended to treat significantly ill patients such as those with serious forms of cancer, where there are clear, unfavorable consequences for not having an effective treatment.
Clinical Trials Are Time Consuming and Expensive Clinical trials test the effectiveness and potential side effects of a candidate drug before it is approved by the FDA for general use. These trials proceed in at least three phases (Figure 35.27). In phase 1, a small number (usually from 10 to 100) of healthy volunteers take the drug for an initial study of safety. These volunteers are given a range of doses and are monitored for signs of toxicity. The efficacy of the drug candidate is not specifically evaluated .
phase 1
Phase 2
1021
Phase 3
Preclinical drug discovery
Clinical use Safety
Safety Efficacy Dosage
35.4 Stages of Drug Development
•• •
Safety Efficacy Side effects
Figure 35.27 Clinical-trial phases. Clinical trials proceed in phases examining safety and efficacy in increasi ngly large groups.
In phase 2, the effi cacy of the drug candidate is tested in a small number of persons who might benefit from the drug. Further data regarding the drug's safety are obtained. Such trials are often controlled and double-blinded. In a controlled study, subj ects are divided randomly into two groups. Subjects in the treatment group are given the treatment under investigation. Subjects in the control group are given either a placebo that is, a treatment such as sugar pills known to not have intrinsic value or the best standard treatment avai lable, if withholding treatment altogether would be unethical. [n a doubleblinded study, neither the subj ects nor the researchers know which subjects are in the treatment group and which are in the control group. A double- blinded study prevents bias in the course of the trial. When the trial has been completed, the assignments of the subjects into treatment and control groups are unsealed and the results for the two groups are compared. A variety of doses are often investigated in phase 2 trials to determine which doses appear to be free of serious side effects and which doses appear to be effective. One should not underestimate the power of the placebo effect that is, the tendency to perceive impro vem ent in a subject who believes that he or she is receiving a potentially beneficial treatment. In a study of arthroscopic surgical treatment for knee pain, for exampl e, subj ects who were led to be lieve that they had received surgery through th e use of videotapes and other means showed the sam e level of improvement, on average, as subject s who were actually operated on. In phase 3, similar studi es are performed on a larger population . This phase is intended to more firmly establish the efficacy of the drug candidate and to d etect side effects that may develop in a small percentage of the subjects who receive treatment. T housands of subj ects may participate in a typical phase 3 study. C linical trials can be extremely costly. Hundred s or thousands of patients must be recruited and monitored for the duration of th e trial. Many physicians, nurses, clinical pharmacologists, statisticians, and others participate in the d esign and execution of the trial. Costs can run from tens of millions to hundred s of millions of dollars . Extensive record s must be kept, including documentation of any adverse reactions. These data are compiled and submitted to th e FDA. The full cost of developing a drug is currently estimated to be from $400 million to $800 million. Even aft er a drug has been approved and is in use, difficulties can arise. As mentioned earlier, rofecoxib (Vioxx), for example, was withdrawn from the market after significant cardiac side effects were d etected in additional clinical trials. Such events highlight the necessity for users of any drug to balance beneficial effects against potential risks.
The Evolution of Drug Resistance Can Limit the Utility of Drugs for Infectious Agents and Cancer Many d rugs are used for long periods of time without any loss of effective ness. However, in some cases, particularly for the treatment of infectious di seases or of cancer, drug treatments that were initially effective beco me
less effective. In other words, the disease becomes resistant to the drug therapy. Why does this occur? Infectious diseases and cancer have a common feature namely, that an affected person contains many cells (or viruses) that can mutate and reproduce. These conditions are necessary for evolution to take place. Thus, an individual microorganism or cancer cell may by chance have a genetic variation that makes it more suitable for growth and reproduction in the presence of the drug than is the population of microorganisms or cancer cells at large. These microorganisms or cells are more fit than others in their popu lation, and they will tend to take over the population. As the selective pressure due to the dru g is continually applied, the population of microorganisms or cancer cells will tend to become more and more resistant to the presence of the drug . Note that resista nce can develop by a number of mechanisms. The HIV protease inhibitors discussed earlier provide an important example of the evolution of drug resistance. Retroviruses are very well suited to this sort of evolution because reverse transcriptase carries out replication without a proofreading mechanism. In a genome of approximately 9750 bases, each possible single point mutation is estimated to appear in a virus particle more than 1000 times per day in each infected person. Many multiple mutations also occur. Most of these mutations either have no effect or are detrimental to the virus. However, a few of the mutant virus particles encode proteases that are less susceptible to inhibition by the drug. In the presence of an HIV protease inhibitor, these virus particl es will tend to replicate more effectively than the population at large. Over time, the less susceptible viruses will come to dominate the population and the virus population will become resistant to the drug. Pathogens may become resistant to antibiotics by completely different mechanisms. Some pathogens contain enzymes that inactivate or degrade specific antibiotics. For example, many organisms are resistant to [3 -lactams such as penici llin because they contain [3-lactamase enzymes. These enzymes hydrolyze the [3-lactam ring and rend er the drugs inactive.
1022 CHAPTER 35 Drug Development
o
o R-
R-
-'(
H
o
.~ s
HN ,
HN'r_ ~ S -N~
H ~
+ H2 0
·····""' CH
-
o
3
fj-l actama se >
0=
R, CTG --> CGG Karachi: residue 5, A --> P, GCC --> CCC Swan River: residue 6, D ) G, GAC --> GGC
Chapter 6 1. There are 26 identities and two gaps for a score of 210. The two sequences are approximately 26% identical. This level of homology is likely to be statistically significant. 2. They are likely related by divergent evolution, because three-dimensional structure is more conserved than is sequence identity. 3. (1) Identity score = 225; Blosum score = 7; (2) identity score = 15; Blosum score = 210.
4. U
o
NI--'
I
.0.-· . ...• ·H---
u 40
co ")4
•
5. There are 4 ,or 1.2 X 10" ,dlfferent molecules. Each - ')0 molecule has a mass of 2.2 X 1 0 ~ , because 1 mol of polymer
Answers to Problems
C5
,,
hasamassof330gmol- 1 X 40,andthereare6.02 X 1023 molecules per mole. Therefore, 26,4 kg of RNA would be required. 6. Because three-dimensional structure is much more closely associated with function than is sequence, tertiary structure is more evolutionarily conserved than is primary structure. In other words, protein function is the most important characteristic, and protein function is determined by structure. Thus, the structure must be conserved, but not necessarily a specific amino acid sequence, 7, Alignment score is 6 X 10 = 60. Many answers are possible, depending on the randomly reordered sequence. A possible result is Shuffled sequence: TKADKAGEYL Alignment: (1) ASNFLDKAGK TKADKAGEYL Alignment score is 4 X 10
40.
=
1.0
__ 0.8
c
o i!!
.'" .a
'" -'"
0.6
~
..g
0.4
~~..
8. (a) Almost certainly diverged from a common ancestor. (b) Almost certainly diverged from a common ancestor. (c) May have diverged from a common ancestor, but the sequence alignment may not provide supporting evidence. (d) May have diverged from a common ancestor, but the sequence alignment is unlikely to provide supporting evidence. 9. Protein A is clearly homologous to protein B, given 65% sequence identity, and so A and B are expected to have quite similar three-dimensional structures. Likewise, proteins Band C are clearly homologous, given 55% sequence identity, and so Band C are expected to have quite similar three-dimensional structures. Thus, proteins A and C are likely to have similar three-dimensional structures, even though they are only 15% identical in sequence. 10. The likely secondary structure is /N-N,
G
'c
attached (in similar fashion to the conformational change in hemoglobin). 5. The modified hemoglobin should not show cooperativity. Although the imidazole in solution will bind to the heme iron (in place of histidine) and will facilitate oxygen binding, the imidazole lacks the crucial connection to the particular a helix that must move so as to transmit the change in conformation. 6. Inositol pentaphosphate in part c 7. (a)
A
G/
>- 0.2
o o
20
40
60
80
100
p02 (torr)
(b) 1.0
-... 0.8
c .-o
~
.a
0.6
:Jl
-c
'".S! U
0.4
tt. >- 0.2 o'----_ _L -_ _-'--_ _--L-_ _
o
~ ~
N-N-N/
50
100
150
200
p02 (torr)
'N-N
11. 107 or 108 identities (depending on which annotated human sequence is chosen)
Chapter 7 1. The whale swims long distances between breaths. A high concentration of myoglobin in the whale muscle maintains a ready supply of oxygen for the muscle between breathing episodes. 2. 62.7% oxygen-carrying capacity 3. A higher concentration of BPG would shift the oxygenbinding curve to the right, causing an increase in Pso. The larger value ofP so would promote dissociation of oxygen in the tissues and would thereby increase the percentage of oxygen delivered to the tissues. 4. Oxygen binding appears to cause the copper ions and their associated histidine ligands to move closer to one another, thereby also moving the helices to which the histidines are
8. Release of acid will lower the pH. A lower pH promotes oxygen dissociation in the tissues. However, the enhanced release of oxygen in the tissues will increase the concentration of deoxy- Hb, thereby increasing the likelihood that the cells will sickle. 9. The "picket fence" provides a pocket for the reversible binding of oxygen to Fe while preventing Fe-O-Fe dimerization. The methyl imidazole ligand protects Fe from the other side and prevents dimerization. 10, (a)Y = 0.5whenp02 = 10 torr, The plot ofYversus p02 appears to indicate little or no cooperativity. (b) The Hill plot shows slight cooperativity with n = 1.3 in the central region. (c) Deoxy dimers oflamprey hemoglobin could have lower affinity for oxygen than do the monomers. If the binding of the first oxygen atom to a dimer causes dissociation of the dimer to give two monomers, then the process would be cooperative. In
C6
ANSWERS TO PROBLEMS
9. The mutation slows the reaction by a factor of 100 because the activation free energy is increased by 53 .22 kJ mol - 1 (12.72 kcal mol - '). Strong binding of the substrate relative to the transition state slows catalysis. 10. 11 fLmo l minute - 1 11 . If the total amount of enzyme (E T ) is increased, Vm ax will increase, because Vmax = " 2[E.,]. But KM = (L , + "2)lk , ; that is , it is independent of substrate concentration. The middle graph describes this situation. 12 . (a)
this mechanism , oxygen binding to each monomer would be easier than bind ing the fir st oxygen atom to a deoxy dimer.
Chapter 8 1. (a) 31.1 fLmol; (b) 0.05 fLmol; (c) 622 S- I 2. (a) Yes , KM = 5.2 X 10 - 6 M ; (b ) Vn " " = 6.8 X 10 - 10 mol minute - I; (c)337 S - l 3. P enicillinase, like glycopeptide transpeptidase, forms an acyl -enzyme intermediate with its substrate but transfers the intermediate to water rather than to the terminal glycine residue of the pentaglycine bridge. 4. (a) In the absen ce of inhibitor, V max is 47 .6 fLmol minute and KM is 1.1 X 10 - 5 M. In the presence of inhibitor, V m ax is the same and the apparent K M is 3 .1 X 10 - 5 M. (b ) Competitive (c) 1. 1 X 10 - 3 M (d )I Es is 0.243, andiEI is 0.488. (e) IEs is 0.73 in the absence of ip..hibito r and 0.49 in the 3 presence of 2 X 10 - M inhibitor. The rati o of these values, 1.49, is the same as the ratio of the reaction velocities under these conditions. 5. (a) V max is9.5 fLmolminute - 1 K M is 1.1 x 10 - 5 M,the same as without inhibitor. (b ) Noncompetitive (c) 2.5 X 10 - ' M (d ) fEs = 0.73, in the presence or absence of this noncompetitive inhibitor. 6. (a) V = Vm " - (V/[SJ) KM (b) Slope = 2K M, Y intercept = Vrna" X intercept = V maJ K YI. (c) An Eadie-Hofstee plot:
1
1 No inhibitor
2 Competitive inhibitor 3 Noncompetitive inhibitor
v
VI [5] 7. Potential hydrogen-bond donors at pH 7 are the side chai ns of the following residues: arginine, asparagine, glutamine, histidine, lysine, serine, threonine, tryptophan, and tyrosine. 8. The rates of utilization of substrates A and B are given by
k2 KM
1/Vo
-
\. 1/[5]
(b) This behavior is substrate inhibition: at hi g h concentrations, the substrate forms unproductive complexes at the active site. The adjoining drawing shows what might h appen. Substrate normally binds in a defined orientation, shown in the drawing as red to red and blue to blue . At high concentrations, the substrate may bind at the active site such that the proper orientation is met for each end of the molecule, but two different substrate molecules are binding . Enzyme active site
Enzyme active site
Normal substrate binding at the active site. Substrate will be cleaved to red and blue balls. II II
Substrate inhibition
13 . The first step will be the rate-limiting step. Enzymes Ell and Ec are operating at Y, V max> whereas the KM for enzyme E" is greater than the sub strate concentration . EA would be operating at approximately 10 - 2 Vmax' 14. (a) When [S+] is muc h greater t han the value of KM , pH will have a negligible effect on the enzyme because S + will interact with E - as soon as the enzyme becomes available. Vmax ~_______________
[E ][A] A
and 2
[E] [B]
4
6
8
10
pH
D
Hence, the ratio of these rates is
V A/V B =
~:
A[A ]I ~:I B[B]
Thu s, an enzyme discriminates between competing substrates on the basis of their values of k21 KM rather than of KM alone .
(b) When [5 +1 is much less than the value of K M, the plot of Vo versus pH becomes essentially a titration curve for the ioni zable groups, with enzyme activity being the titration marker. At low pH, the high concentration of H + will keep the enzyme in the EH form and inactive. As the pH rises, more and more of the
Answers to Problems
enzyme will be in the E - form and active. At high pH (low H +), all of the enzyme is E -.
13. (a) Cysteine protease: The same as Figure 9.8, except that cysteine replaces serine in the active site (b) Aspartyl protease:
o:
H
.
\
/~
