Progress in Nucleic Acid Research and Molecular Biology, Volume 32

PROGRESS IN Nucleic Acid Research and Molecular Biology Volume 32 This Page Intentionally Left Blank PROGRESS IN ...

Author: Waldo E. Cohn

8 downloads 729 Views 17MB Size Report

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!
Report copyright / DMCA form

DOWNLOAD PDF

PROGRESS IN

Nucleic Acid Research and Molecular Biology Volume 32

This Page Intentionally Left Blank

PROGRESS IN

Nucleic Acid Research and Molecular Biology edited by

WALDO E. COHN

KlVlE MOLDAVE

Biology Division Oak Ridge National Laboratory Oak Ridge, Tennessee

University of Calijornia Santa Cruz, California

Volume 32

7985 ACADEMIC PRESS, INC Harcourt Brace Jovanooich, Publishers

Oriondo Sun Diego New York Austin London Montreal Sydney Tokyo Toronto

COPYRIGHT @ 1985 BY ACADEMIC PRESS, hIC. ALL RIGHTS RESERVED. NO PART OF THIS PUBLICATION MAY BE REPRODUCED OR TRANSMITTED IN ANY FORM OR BY ANY MEANS, ELECTRONIC OR MECHANICAL, INCLUDING PHOMCOPY, RECORDING, OR ANY INFORMATION STORAGE AND RETRIEVAL SYSTEM, WITHOUT PERMISSION IN WRITING FROM THE PUBLISHER.

ACADEMIC PRESS,

INC.

&lMdo. Florida 32887

United Kin dom Edition ublished by

ACADWfC PRESS & I.

(LONDON)LTD.

24-28 Oval Road, London NWI 7DX

LIBRARY OF CONGRESS CATALOG CARD NUMBER: 63-1 5841 ISBN 0-12-540032-2 PRlNTED IN M E UNITED STATES OF AMERICA

85 86 81 88

9 8 7 6 5 4 3 2 I

Contents CONTRIBUTORS . .. . . . . . . . . . . . . . . . . . , . . . . . . .. . . . . . . . . . , . . . . . . . .

.. . .., .. ,

,

...

ix

ABBREVIATIONS AND SYMBOLS . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . , . . . . . . . . . .

xi

SOMEARTICLESPLANNED FOR FUTURE VOLUMES.. , ., . , ...... .... .. ... .... .

xv

Gene Conversion in Trypanosome Antigenic Variation Etienne Pays I. Summa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. Antigenic Variation Is due to Differential Gene Expression . . 111. Antigen Gene Expression Is Often Linked to DNA Rearrange IV. Evidence for Gene Conversion as a Mechanism for Antigenic Variation V. The Extent of Gene Conversion Is Variable, Depending on the Degree of Homology between the Recombinant Sequences . . . . . . . . . . . . . . . . . . ...................... VI. Gene Conversion Endpoints .......... VJI. Gene Conversion Frequency. . . . . . . . . . . . . . . . . . VIII. Orientation of the Gene Conversion Mechanism; .............. Transcription . . . . . . . . . . . . . . . . ire Evolution . . . . . . . . . . . . . . . . . IX. Gene Conversion and Antigen R X. Sexual Conjugation Further Leads to Evolution of the Antigen Gene .................................. Repertoire.. . . . .

XI.

.................

.......... References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

7 14

15 15 17 21 21 23

Hypermodified Nucleosides of tRNA: Synthesis, Chemistry, and Structural Features of Biological Interest Ryszard W. Adamiak and Piotr G6rnicki I. Hypermodified Nucleosides of tRNA: A Bioorganic Chemist’s View. . . . 11. Synthesis and Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. Structural Features of Hypermodified Nucleosides and

Codon-Anticodon Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

27 35 53 67

Ribosomal Translocation: Facts and Models Alexander S. Spirin I. Definition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. Experimental Tests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V

75 77

vi

CONTENTS

I11. IV. V. VI.

VII . VIII .

IX.

Two-tFtNA-Site Model for the Ribosomal Elongation Cycle . . . . . . . . . . . Main Facts Concerning Translocation ............................. Sequence of Events in Translocation Promoted by EF-G . . . . . . . . . . . . . Energetics of Translocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Kinematics ofTranslocation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Are Conformational Movements of the Ribosome Required for Translocation? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

79 81 84 86 93 105 108 109

Chemical Changes Induced in DNA by Ionizing Radiation Franklin Hutchinson I. The Mechanisms by which Ionizing Radiations Act on DNA . . . . . . . . . . I1 . Indirect Action: The Effects of Reactive Species Formed from Water on DNA in Dilute Solution ......................................... 111. Effects of Irradiation in the SoIid State ............................. IV. Irradiation of DNA in Cells ...................................... V. Quantitative Measurements That Should Be Made on Irradiated DNA . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

116 117 131 138 148 149

Comparative Anatomy of 16-S-like Ribosomal RNA Robin R . Gutell. Bryn Weiser. Carl R . Woese. and Harry F . Noller I . Comparative Anatomy of 16-S-like Ribosomal RNA . . . . . . . . . . . . . . . . . . 156 I1. A Computer-Assisted Search for Coordinated Base Changes 199 in 16-S rRNA .................................................. References

....................................................

214

SV40 Promoters and Their Regulation Gokul C . Das. Salil K . Niyogi. and Norman P . Salzman I . Regulatory Region of SV40....................................... I1. DNA Binding Property of T Antigen .............................. 111. Regulation of Transcription ...................................... IV. Conclusions ................................................... References ....................................................

218 227 229 232 232

vii

CONTENTS

The Role of the Anticodon in Regulation of tRNA by Aminoacyl-tRNA Synthetases Lev L. Kisselev I. Concise Background of the Problem. . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . 11. The Role of the Anticodon in Acceptor Function . . . . . . . . . , . . . . . . . . . . 111. General Remarks.. . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

239 243 258 263

Properties and Spatial Arrangement of Components in Preinitiation Complexes of Eukaryotic Protein Synthesis Heinz Bielka I. Arrangement of Proteins in Small Ribosomal Subunits . . . . . . . . . . . . . . . 11. Function and Arrangement of Components in Preinitiation Complexes.. 111. Summary and Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. References.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

268 274 281 287

Complementary-Addressed (Sequence-Specific) Modification of Nucleic Acids Dmitri G. Knorre an d Valentin V. Vlassov I. Synthesis of Complementary-Addressed Reagents . . . . . . . . . . . . . . . . . . . 11. Complementary-Addressed Modification of Model Oligonucleotides and Polynucleotides ............................................ 111. Biochemical Applications of Complementary-Addressed Modification. . . IV. Concluding Remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

.

INDEX. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . CONTENTS OF PREVIOUS VOLUMES.

. .. . ........ .. . . . .. . . .. . .

292 299 309 315 316 323 329


Contributors Numbers in parentheses indicate the pages on which the authors’ contributions begin.

RYSZARD W. ADAMIAK (27), lnstitute of Bioorganic Chemistry, Polish Academy of Sciences, Poznan, Poland HEINZBIELKA(267), Central lnstitute of Molecular Biology, Department of Cell Physiology, Academy of Sciences of the GDR, 1115 Berlin-Buch, Gemnan Democratic Republic GOKULC. D A S(217), ~ Laboratory of Biology of Viruses, NIAZD, National Znstitutes of Health, Bethesda, Maryland 20205 PIOTRG~RNICKI (27), Institute of Bioorganic Chemistry, Polish Acade m y of Sciences, Poznan, Poland ROBINR. GUT ELL^ (155), Thimann Laboratories, University of California, Santa Cruz, California 95064 FRANKLINHUTCHINSON (115), Department of Molecular Biophysics and Biochemistry, Yale University, New Haven, Connecticut 0651 1 LEV L. KISSELEV (237), lnstitute of Molecular Biology, The U S S R Academy of Sciences, Moscow 117984, U S S R DMITRI G. KNORRE(291),Novosibirsk lnstitute of Bioorganic Chemistry, 630090 Novosibirsk, U S S R SALILK. NIYOCI(217), Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831 HARRY F . NOLLER(155),Thimann Laboratories, University of California, Santa Cruz, California 95064 ETIENNEPAYS(l),Department of Molecular Biology, University of Brussels, B1640 Rhode Saint Gen&se,Belgium NORMANP. SALZMAN (217),Laboratory of Biology of Viruses, NZAID, National lnstitutes of Health, Bethesda, Maryland 20205 ALEXANDER S . SPIRIN(75), Institute of Protein Research, Academy of Sciences of the USSR , Pushchino, Moscow Region, U S S R VALENTINV . VLASSOV(2911, Novosibirsk Institute of Bioorganic Chemistry, 630090 Novosibirsk, U S S R BRYNWEISER(155), Thimann Laboratories, University of California, Santa Cruz, California 95064 CARLR. WOESE(155),Department of Genetics and Development, University of Illinois, Urbana, Illinois 61801 Present address: Division of Molecular Biology and Biophysics, School of Basic Life Sciences, University of Missouri-Kansas City, Kansas City, Missouri 64110. Present address: Department of Genetics and Development, Mom11 Hall, University of Illinois, Urbana, Illinois 61801. ix


Abbreviations and Symbols All contributors to this Series are asked to use the terminology (abbreviations and symbols) recommended by the IUPAC-IUB Commission on Biochemical Nomenclature (CBN) and approved by IUPAC and IUB, and the Editor endeavors to assure conformity. These Recommendations have been published in many journals (1. 2) and compendia (3) in four languages and are available in reprint form from the Office of Biochemical Nomenclature (OBN), as stated in each publicatioir, and are therefore considered to be generally known. Those used in nucleic acid work, originally set out in section 5 of the first Recommendations ( 1 ) and subsequently revised and expanded ( 2 , 3 ) ,are given in condensed form (I-V) below for the convenience of the reader. Authors may use them without definition, when necessary. 1. Eases, Nucleosides, Mononucleotides

1 . Buses (in tables, figures, equations. or chromatograms) are symbolized by Ade, Gua, Hyp, Xan, Cyt, Thy, Oro, Ura; Pur = any purir.e, Pyr = any pyrimidine, Base = any base. The prefixes S-, H,, F-, Br, Me, etc., may be used for modifications of these. 2. Ribonucleosides (in tables, figures. equations. or chromatograms) are symbolized, in the same order, by Ado, Guo, Ino. Xao. Cyd. Thd, Ord, Urd (Yrd), Puo, Pyd, Nuc. Modifications may be expressed as indicated in (I) above. Sugar residues may be specified by the prefixes r (optional), d (=deoxyribo), a, x, 1, etc., to these, or by two three-letter symbols, as in Ara-Cyt (for aCyd) or dRib-Ade (for dAdo). 3. M o n o - . di-. und triphosphotcs of irtrck,osit/c,s (5‘)are designated by NMP, NDP. NTP. The N (for “nucleoside”) may be replaced hy any one of the nucleoside symbols given in 11-1 below. 2’-, 3‘-, and 5’- are uwd as prcfixrs when necessary. The prefix d signifies “deoxy.” [Alternativelv, nucleotides may be expressed by attaching P to the symbols in (2) above. Thus: P-Ado = AMP; Ado-P = 3’-AMP] cNMP = cyclic 3’:5’-NMP; Bt,cAMP = dibutyryl CAMP. etc. II. Oligonucleotides and Polynucleotides 1. Ribonucleoside Residues

(a) Common: A, 6, I, X, C, T, 0. U. 4‘.’ H. Y, N (in the order of 1-2 above). (b) Base-modified: sI or M for thioinosine = 6-mercaptopurinr ribonucleoside; sU o r S for thiouridine; brU or B for 5-bromouridine; h U or D for 5,6-dihydrouridine; i for isopentenyl; f for ( ~ u s ( ~ (in conformyl. Other modifications are similarly indirated by appropriate / o ~ c ~ ~ r - prefixes trast to 1-1 above) ( 2 . 3). (c) Sugar-modified: prefixes are d, a, x. or I as in 1-2 above; alternatively, by italics or boldface type (with definition) unless the tmtirr chain is specified by an appropriate prefix. The 2’-O-methyl group is indicated by suffix in (t..g.,-Am- for 2’-()-methyladenosine, but -mA- for

6-methy ladenosine). (d) Locants and multipliers, when iierc.ssary, are indicated by superscripts and subscripts, respectively, e . g . , -m!A- = 6-dimethyladrnosint.; -s‘U- or -‘S- = 4-thiouridinr; -ac‘” p is cytidine 2’:3’-cyclic phosphate ( 1 , 2 . 3 ) : p < A is adenosine 3’:5‘-cyclic phosphate.. xi

xii

ABBREVIATIONS AND SYMBOLS

(b) Internal: hyphen (for known sequence), comma (for unknown sequence); unknown sequences are enclosed in parentheses. E.g., PA-G-A-C(C,,A,U)A-U-G-C > p is a sequence with a (5’)phosphate at one end, a 2’:3’-cyclic phosphate at the other, and a tetranucleotide of unknown sequence in the middle. (Only codon triplets s h o d be written without some punctuation separating the residues.) 3. Polarity, or Direction of Chain The symbol for the phosphodiester group (whether hyphen or comma or parentheses, as in 2b) represents a 3 -5 link (i.e., a 5’. . . 3‘ chain) unless otherwise indicated by appropriate numbers. ”Reverse polarity” (a chain proceeding from a 3’ terminus at left to a 5‘ terminus at right) may be shown by numerals or by right-to-left arrows. Polarity in any direction, as in a two-dimensional array, may be shown by appropriate rotation of the (capital) letters so that 5’ is at left, 3’ at right when the letter is viewed right-side-up. 4. Synthetic Polymers

The complete name or the appropriate group of symbols (see 11-1 above) of the repeating unit, enclosed in parentheses if complex or a symbol, is either (a) preceded by “poly,” or (b) followed by a subscript “n” or appropriate number. No space follows “poly” (2, 5). The conventions of 11-2b are used to specify known or unknown (random) sequence, e.g., polyadenylate = poly(A) or A,,, a simple homopolymer; poly(3 adenylate, 2 cytidylate) = poly(A&) or (A3.CJn. an irrcpdur copolymer of A and C in 3 :2 proportions; polfldeoxyadenylate-deoxythymidylate) = poly[d(A-T)] or poly(dA-dT) or (dA-dT), or d(A-T)., an alternating copolymer of dA and dT; poly(adenylate.guanylate,cytidylate,uridylate) = poly(A,C,C,U) or (A.G,C;, U),,, a random assortment of A, C , C. and U residues, proportions unspecified. The prefix cwpoly or oligo may replace poly, if desired. The subscript ”n” may be replaced by numerals indicating actual size, e.g., A, .dT,t.,8. 111. Association of Polynucleotide Chains 1. Assor.iutrd (e.p.. H-bonded) chains, or bases within chains, are indicated by a rmfrr & ~ t (not a hyphen or a plus sign) separating the coiiipletr names or symbols, e . g . : p d y ( A ) . pdy(U) or A. ’ U, poly(A) . 2 poly(U) or A. . 2U, or (dA-dC), . (dG-dT),. p)ly(dA-dC) . poly(dC-dT)

2. .Vonossocicited chains are separated by the plus sign, c . g . : or

-

2[poly(A) . poly(U)l + poly(A) 2 polyfU) 21& . Urn] . 2U, + A,,.

+ p)ly(A)

3. Unspecified or unknown association is expressed by a comma (again meaning “unknown”) between the completely specified chains. N(J~c In: all cases. each chain is completely specified in one or the other of the two systems described in 11-4 above.

IV. Natural Nucleic Acids

RNA DNA mRNA; rRNA; nRNA hnRNA D-RNA; cRNA

ribonucleic acid or ribonucleati. deoxyribonucleic acid o r deoxyrib”~ucleate messenger RNA; ribosomal RNA; nuclear RNA heterogeneous nuclear RNA “DNA-like.” RNA; complementary RNA

xiii

ABBREVIATIONS AND SYMBOLS mtDNA tRNA

mitochondria1 DNA transfer (or acceptor or amino-acid-accepting) RNA; replaces sRNA, which is not to b e used for any purpose aminoacyl-tRN A “charged” tRNA (i.e., tRNA’s carrying aminoacyl residues); may be abbreviated to AA-tRNA alanine t R k A or tRNA normally rapable of accepting alanine. to form tRNAAIa,etc. danyl-tRNA. etc. alanyl-tRNA or The same. with alanyl residue covalently attached. alanyl-t RN AALa ( N o t r : N e t = formylmethionyl; hence tRNAme‘. identical with tRNAp‘] Isoacceptors are indicated by appropriate subscripts. i.e., tRNA:Ia. tRNA:la, etc.

V. Miscellaneous Abbreviations P,. pp, inorganic orthophosphate. pyrophosphate RNase, DNase ribonuclease, deoxyribonuclease melting temperature (“C) I, (not T , ) Others listed in Table I1 of Reference 1 may alsci be used without definition. No others, with or without definition, are used unless. in the opinion of the editor. they increase the ease of reading.

Enzymes In naming enzymes, the 1978 recommrndations of the I U B Commission on Biochemical Nomenclature ( 4 ) are followed as far as possible. At first mention, each enzyme is described rithrr by its systematic name o r by the ccluation for the reaction catalyzed or by the recommended trivial name, followed b y its E C number i n parentheses. Thereafter, a trivial name may be used. Enzyme names are not to be aht)reviatt.d except when the substrate has an approved abhreviation (e.g.. ATPase, hut not L D H , is acceptable).

REFERENCES 1 . J B C 241, 527 (1966); Brhcnt 5, 144.5 (1%): BJ 101, 1 (1966); A B B 115, I ( 1 W ) . 129, 1 (1969); and rlsewhere. t 2. F:/B 15, 203 (1970);J B C 245, 5171 (1970). J M B 55, 299 (1971); and cisewhere.t 3. “Handbook of Biochemistry” ( C . Fasman, e d . ) . 3rd ed. Chemical Rubber C o . , Cleveland, Ohio, 1970, 1975, Nucleic Acids. Vols. I and 11. pp. 3-59. 4. “Enzyme Nomenclature” [Recommendations ( 1978) of the Nomenclature Committee of the IUB]. Academic Press, New York. 1979. 5. “Nomenclature of Synthetic Polypt,ptides.” J B C 247, 323 (1972); Biopo/yttwr,s 1 1 , 321 (1972); and elsewhere. t

Abbreviations of Journal Titles Jorcrnd~

Annu. Rev. Biochem. Annu. Rev. Genet. Arch. Biochem. Biophys. Biochem. Biophys. Res. Commun. Biochemistry Biochem. J. Biochim. Biophys. Acta Cold S p r i n g Harbor

A hhrooia t io ns i c s d

ARB ARGen

ABB BBRC Bchern BJ BBA CSH

xiv Cold Spring Harbor Lab. Cold Spring Harbor Symp. Quant. Biol. Eur. J. Biochem. Fed. Proc. Hoppe-Seyler’s Z. physiol. Chem. J. Amer. Chem. SOC. J. Bacteriol. J. Biol. Chem. J. Chem. SOC. J. Mol. Biol. J. Nat. Cancer Inst. Mol. Cell. Biol. Mol. Cell. Biochem. Mol. Gen. Genet. Nature, New Biology Nucleic Acid Research Proc. Nat. Acad. Sci. U.S. Proc. SOC.Exp. Biol. Med. Progr. Nucl. Acid. Res. Mol. Biol.

ABBREVIATIONS AND SYMBOLS

CSHLab CSHSQB EJB FP ZpChem JACS J. Bact. JBC JCS JMB JNCI MCBiol MCBchem MGG Nature NB NARes PNAS PSEBM This Series

Some Articles Planned for Future Volumes Dynamics of Nucleosome Structure in Transcriptional Control

VINCENTALLFREY ATP-Ubiquitin-DependentDegradation of lntrocellulor Proteins

AVRAMHEHSHKO lnteroctive Three-Dimension01Computer Graphics in Protein and Genetic Engineering

ROBEHT LANGRIDGE Recent Studies on DNA Polymerose

LAWRENCE A. LOEB Small Phages os Models for Vorious Biochemical Processes

PETERMODEL DNA Repair in the Radiation-Resistant Deinococci

R. €3. MOSELEY The Recombination-likeActivities of E. cofi r e d Protein

CHARLESRADDINC Site-specific Mutogenesis

U . L. RAJBHANDARY Genetic Recombination in Bacteriophage

FRANKLIN W. STAHL Arrangement of Genes in Chloroplost DNA of Higher Plants

K. K. TEWARI DNA Methylation: The Primaly DNA Sequence Determines h viho Methylation by Mammalian DNA Methyl Transkrases

ARTHURWEISSBACH, CHEHYL WARD,ARTHURBOLDEN,AND CARLONATIN

XV


Gene Conversion in Trypunosome Antigenic Variation ETIENNE PAYS Department of Molecular Biology Unioersity of Brussels Rhode Saint Genise, Belgium

I. Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. Antigenic Variation Is due to Diffcrential Gene Expression. . . . . . . 111. Antigen Gene Expression Is Often Linked to DNA Rearrangements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Evidence for Gene Conversion as a Mechanism for Antigenic Variation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . V. The Extent of Gene Conversion Is Variable, Depending on the Degree of Homology between the Recombinant Seque VI. Gene Conversion Endpoints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Gene Conversion Frequency. . . . . . . ...... VIII. Orientation of the Gene Conversion Mechanism; Relationship with Transcription. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX. Gene Conversion and Antigen Repertoire Evolution . . . . . . . . . . . . . X. Sexual Conjugation Further Leads to Evolution of the Antigen Gene Repertoire . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . XI. Problems and Outlooks References . . . . . . . . . . . . . . . . . . . . . . . . .

1 2

3

4 7 14 15 15 17 21 21 23

1. Summary Antigenic variation allows African trypanosomes’ to escape the immune defenses of their hosts. This occurs through differential surface antigen gene activation, with only one antigen gene being expressed at any one time among a large repertoire of different se-

’

Abbreviations: The nomenclattcre of the trypanosome antigenic types fdlows the system recommended by Lumsden [W. H. R. Lumsden, System. Parasitol. 4, 373 (1982)l.It is based on the name of the place where the imniunological characterization of the antigen repertoire was performed. For instance, AnTat 1.10 stems from Aritwerp Trypanozoon antigenic type number 10, lroin the repertoire number 1. BC = Basic Copy of the antigen gene: it is the template used for the synthesis of the additional, expression-linked gene copy. The BC is the donor in the gene conversion process. ELC = Expression-Linked Copy of the antigen gene: it is the copy synthesized in the clone where the gene is activated. kb = Kilobase pairs. 1 Progress i n Nucleic Acid Research and Molecular Biology, Vol. 32

Copvright 0 1985 hy Academic Press, Inc.

All rights of reproduction in any form reserved.

2

ETIENNE PAYS

quences. The differential gene activation can be achieved by gene conversion between antigen-specific sequences, taking place in a te-i Iomeric gene expression site. The extent of the gene conversion event may vary considerably (from 1 to more than 40 kb), although it is generally about 3 kb. The gene conversion size seems to depend on the extent of homology between the donor and target sequences involved in the recombinational process. Analysis of several gene conversion endpoints, and comparison with the mechanism for mating type interconversion in yeast, suggests that the process could be triggered by a cut upstream from the antigen gene, then resolved by a crossing-over in a region of homology downstream from this point. This crossing-over could take place anywhere within the regions of homology, but only the recombinations leading to the successful generation of new antigen-coding sequences would be selected. Based on this gene conversion model, it is possible to explain the generation of “ mini-chromosomes” harboring antigen-specific sequences. Those could represent residual “targets” carrying a replication origin. Translocation of such haploid telomeric sequences to the other end of other chromosomes would lead to internalization of formerly telomeric genes, and a clustering of haploid antigen-specific sequences near the chromosome termini. I suggest that the orientation of the conversion could be determined by the chromatin structure around the antigen gene. Actively transcribed chromatin indeed exhibits an “open” configuration that may be more susceptible to cutting b y the putative sitespecific endonuclease catalyzing the gene conversion event. This hypothesis implies that, among all antigen-specific sequences, the potentially transcribable ones could be preferential targets for gene conversion. The existence of such preferred target sequences has been postulated to be essential for the generation of polymorphism by gene conversion in multigene families (1,2).In combination with another gene activation mechanism that does not depend on gene duplication, gene conversion leads to rapid modifications of the trypanosome surface antigen repertoire, by gain, loss and alterations of different sequences, and by changes in their activation rate. Sexual conjugation between trypanosomes containing different copy numbers of antigen-specific sequences further leads to evolution of the antigen gene repertoire.

II. Antigenic Variation Is due to Differential Gene Expression African trypanosomes are parasitic flagellates that grow alternately in two successive hosts: first Glossina flies and then various mammals,

GENE CONVERSION

3

including man. At the end of their development in the fly salivary glands, the trypanosomes completely surround themselves with a thick layer of a single, densely packed glycoprotein of about 65,000 Da. Following insect bite, the trypanosomes invade the bloodstream of a mammal, where the protein part of their surface molecule is recognized as an antigen by the immune system of the host. The antibody response leads to a rapid destruction of most cells in this first parasite population, but antigenic variation allows a few individuals (less than 0.1%) to survive. Some of them can undergo proliferation into new parasitemia waves, characterized by different antigenic specificities. The interplay between antigenic variation and immune response gives rise to the chronic infection characteristic of the African trypanosomiases (for reviews on trypanosome antigenic variation and antigen structure, see 3-7). The repertoire of different antigenic types in each cell consists of at least 100 antigens ( 8 ) ,and it has been established that to this large antigen repertoire corresponds a similar collection of different antigen genes (9-12). Only a single antigen is synthesized by a given cell, due to transcriptional control of gene expression (13-16). Chronic infection by African trypanosomes is thus maintained by the differential and successive expression of a large number of different antigen genes. The expression of these genes is not random, since some antigenic types preferentially appear early in the chronic infection (predominant types), whereas others are only rarely observed (late types) (1721). Moreover, in the fly salivary glands, only a small, characteristic fraction of the total antigen repertoire is expressed (metacyclic types) (22,23).These observations point to a possible programming, or induction of antigen gene expression. However, at least in bloodstream trypanosomes, no inducer for antigenic variation has so far been found; in particular, antibodies do not seem to be involved in this mechanism, since antigenic variation has been reported to occur in immunosuppressed mice ( 2 4 ) ,or in uitro (25).

111. Antigen Gene Expression Is Often Linked to DNA Rearrangements The use of specific cDNA probes (10, 15, 16,26) has allowed the study of the expression of several variant-specific antigen genes. In different cases, genome rearrangements were clearly involved in antigen gene expression: it was indeed found that some of the specific antigen mRNAs are synthesized on an additional copy (ELC, for expression-linked copy) of the gene (BC, for basic copy), after transposi-

4

ETIENNE PAYS

tion of the ELC into an expression site distinct from the BC environment (12,14,27-35). However in other cases, no DNA rearrangement could specifically be linked to the antigen gene expression (26, 3639).The trypanosome variant-specific antigen genes can thus be activated by at least two different mechanisms (39-44). Initial reports for genome rearrangements unlinked to antigen gene expression (26) were in fact due to the telomeric location of many of these genes or related sequences (28,29,34,45-48).Indeed many antigen-specific sequences are found near chromosome ends and the terminal size of the telomeres is continuously changing. A constant length increase, probably linked to the mechanism of DNA replication in chromosome ends (49, 50) is corrected by occasional shortenings whose extent and frequency differ between transcriptionally active and inactive telomeres (49, 51 ). These DNA alterations do not seem to be involved in antigenic variation, although telomere terminal shortenings have often, but not always, been observed in connection with the telomere involvement in antigen gene expression (51). It is worth stressing here that the trypanosome antigen genes are always transcribed in a telomere, whatever their activation mechanism may be. In particular, the genes activated without apparent DNA rearrangement have always been found to be telomeric (37,38,39,43, 49). Moreover, non-telomeric antigen-specific sequences have never been found to be transcribed. The meaning of these observations is completely unknown; at least they point to the need of a telomeric location for antigen genes to be expressed.

IV. Evidence for Gene Conversion as a Mechanism for Antigenic Variation The additional copy (ELC) of variant-specific antigen genes can be transposed into an expression site that seems identical in different trypanosome clones from the same stock (28,29,31,34).This observation suggests that at each antigenic switch, the successive ELCs are replacing each other in the same site, driving out the formerly expressed gene copy. This hypothesis found support in the observation that a sequence at the 3’ end of the gene is shared by different antigen coding sequences (52-55); thus this sequence might be considered as one of the recognition sequences involved in the recombinational processes of ELC replacement (9, 28, 30, 31, 56). Similarly, there is indirect evidence that a repeat located upstream from the antigen gene may be required for ELC transposition, at least in some cases (9,

28,31, 57, 58).

5

GENE CONVERSION

-

dA

9-

probe :

1

FIG.1. T h e .’2 brucei AnTat 1.1 gene family. T h e five AnTat 1.1-specific seqnences were separated by electrophoresis in 0.85% agarose gel after PstI digestion; as compared with the genomic DNA of the clone expressing the AnTat 1.3 antigenic type (first lane), the DNAs from the clones expressing the AnTiit 1.1, 1.10 and 1.1B antigenic types contain an additional AnTat 1. l-specific sequence (ELC, designated by an arrowhead), which is the one transcribed (27, -59).S q t i c n c e s A and t3 are involved in the ELC modifications from AnTat 1.1 to 1.1B ( w e trxt); they are represented in Fig. 2 by shaded and open boxes, respectively.

A proof for this ELC replacement hypothesis was provided by the analysis of the antigen gene expressed in the T . b. brucei AnTat 1.1 clone (AnTat for Antwerp Trypanozoon antigenic type), as well as in two clones successively derived from AnTat 1.1, namely AnTat 1.10 and 1.1B (59).The BCs for the AnTat 1.1and 1.10 ELCs were found to be two members of the same multigene family (Fig. 1). Interestingly, both are located in telomeric regions (59).The way in which the ELCs synthesized on these templates replace each other in the expression site during AnTat 1.1to 1.1B clone switching is summarized in Fig. 2.

6

ETIENNE PAYS

E LC

Trypanosome Clone Derivation AnTat l . X

I a

AnTat 1.1

I

---I

b

A n T a t 1.10

I

I

Cd

AnTat 1.16

I

I

f

AnTat 1.1C

9

7 sequence

c- -i 1

kb

FIG.2. Segmental antigen gene conversions in different hypanosome clones. The successive transformations of the antigen gene are schematically represented as follows: the black box in AnTat l.X represents the putative ELC of the clone from which AnTat 1.1 was derived (“companion” sequence: 29); the shaded and open areas are copies of the A and B sequences from the AnTat 1.1 gene family, respectively (see Fig. l), except in AnTat l.lC, where the shaded areas correspond to the A sequence itself (43). The small area in the middle of the AnTat 1.1B coding sequence is of unknown origin (see legend of Fig. 5). Letters on junctions between different areas refer to the DNA sequences presented in Fig. 5. Junctions (a), (b), and (0 are located respectively at about 480, 30, and 160 b p upstream from the gene initiation codon, whereas junction (e) is about 40 bp downstream. The maximum extent of the AnTat 1.1C conversion domain is 1115 bp. Clone AnTat 1.1C was indirectly derived from AnTat 1.1 (see derivation scheme in Figs. 4 and 6 ) .

GENE CONVERSION

7

As a result, the AnTat 1.1B coding sequence appears arranged as a chimeric gene, composed of two sequences copied on two different AnTat 1.1 gene family members. During the DNA rearrangements leading to this AnTat 1.1B expression, the two templates involved were not altered. Antigenic variation in this clone derivation was thus clearly achieved by replacements of some antigen-specific sequences by copies of others. Such a nonreciprocal transfer of information from one DNA duplex to another can be described as a gene conversion event (60). According to present models (60,61),gene conversion could result from a double-strand break in the target DNA, followed by the invasion of a homologous region of another DNA duplex (the donor) by one of the free DNA ends of the target. The donor would then be copied by the invading strands up to the next region of homology with the target sequence. The converted zone of the target may either be digested by an exonuclease, up to a region where the second free DNA end may be able to invade the homologous sequence in the donor, or it could be conserved and recombine with the donor in the homology region (61).A scheme of the second hypothesis, applied to a trypanosome antigenic switch, is presented in Fig. 3 (see below for details). The yeast mating type interconversion, described as a “cassette” model for cell type-specific sequences expression (62), is believed to occur by such a mechanism (61,63-65). In that system too, copies of “basic” genes (HML and HMR) replace each other in an expression site (the MAT locus), where they are transcribed. There are, however, notable differences between the trypanosome and yeast “cassette” models. The yeast mating type interconversion involves only two silent gene copies and utilizes long stretches of homologous DNA, whereas there may be up to as many as 1000 silent antigen genes in trypanosomes (9), whose homologies to the expression site seem highly variable (see paragraph below). Gene conversion has also been invoked as a possible process in the evolution of several eukaryotic multigene families: immunoglobulins (66-71 ), globins (72), major histocompatibility antigens (73-76), yeast transposable elements (77).

V. The Extent of Gene Conversion Is Variable, Depending on the Degree of Homology between the Recombinant Sequences The extent of the converted region often exceeds the size of the antigen coding sequence. Generally (28, 29, 31, 33, 34; also see

8

ETIENNE PAYS

P

: :2(3)

-8-

0.

u.

415-(

a

ELC B

0

FIG.3. Hypothetical scheme for the gene conversion leading to the switch from A to B antigen gene expression. This scheme is a modified version of one of the gene conversions models by Strathern et ul. (61); see the text for explanations. Each line represents a double-stranrled DNA seqnence. The dark and open I)oxes represent the antigen-specific A and B sequences, flanked by S (0)and 3’ (0)homology I h c k s whose location respective to the gene can vary considera1)ly. The length of the different elements is not drawn to scale. The wavy arrow represents the transcription, and P the putative transcription promoter. In ( l ) , the actively. transcrihed sequence A could he either an ELC, or a telonieric gene activated without duplication (43, 44); gene B could also be telomeric (34,43,.59).Thc possible presence o l a centromere is indicated by (8). In (I), a double-strand cut ( t ) triggers the gene conversion process. It could occur in, or near, “CACA”-type stretches located upstream from the gene (Fig. 5) and downstream from a homology block. The lutter can be the 5’ repeat, located generally about 1.5 kb upstream from the gene (9, 28, 58, 79), but sometimes more than 40 kb upstream (43); it could also be within the gene, if the target and the donor share sequence similarities (Fig. 2). In (2),one of the free DNA ends of the target invades the donor, perhaps after some exonucleolytic digestion. In ( 3 ) ,the donor is copied by the invading strands, up to the 3’ repeat; if the donor is telomeric, the invasion could extend further downstream in the telomere (78), possibly up to the DNA end, although this supposition is only speculative. In this figure, recombination is shown to take place between the copy and the target sequence in the 3’ repeat; alternately (not shown), the converted stretch of the target could be first completely digested by an exonuclease, and the second free DNA end of the target could invade the donor in the 3’ homology region (60,61).In (4), the A sequence could be conserved in a “minichromosome” (45), provided that a centromere is located between the gene and the 5’ repeat; otherwise, this sequence could be hydrolyzed or diluted during the cell divisions. Notice that the trypanosome antigen gene conversion is always telomeric, since the target sequence (either the ELC, or a telomeric gene) is always located in a chromosome end.

GENE CONVERSION

9

Fig. 4),a stretch of 1.5 to 2 kb is cotransposed in front of the gene copy, to reach an ELC length of 3 to 3.5 kb. However in several cases, the extent of this conversion is lower than 3 kb, as observed for the AnTat 1.1, 1.10, and 1.1B clones, where the ELC length varies from about 1 to 2 kb (Fig. 2). In these three variants, a sequence of at least 1 kb, called the “companion” sequence, is present in front of the ELC. We found strong evidence that this sequence is a relic from the element contransposed in front of the gene in a previous ELC (29; see Fig. 2). However, in other AnTat 1.1 expressors, where the same BC is used to generate the ELC (AnTat 1.1D or l . l E ) ,this “companion” sequence is not present in front of the ELC. In the latter examples, the ELC is of a more usual size, at least 2.5 kb long (29). It follows that the 5’ recombination locus between two ELCs is variable. Such a variability has also been reported for the 3’ recombinational point in other clones (56). In particular, a “companion”-like sequence has been observed downstream from a MiTat 1.5 ELC (MiTat for Molteno Institute Trypanozoon antigenic type) (78),suggesting that the 3’ gene conversion endpoint can in some cases be more than 1 kb downstream from the gene. The variability in the ELC size most probably depends on the extent of homology between the target and donor sequences. This is suggested by several observations, discussed below. Usually, the homology between the two recombinant partners is restricted to blocks about 3 kb apart, that flank the antigen-specific sequences. One of these blocks is present within the 3’ end of the antigen gene, and differs from the other one, which is located 1.5 to 2 kb upstream from the gene (9,12,28,31,58, 79). Both repeats seem to be conserved among different antigen genes, and could thus be used as initiation points for gene conversions leading to antigenic variation. The extent of gene conversion is therefore most frequently about 3 kb. Some antigen-specific sequences are not flanked by both of the usual blocks of homology. Indeed, these blocks are not always found in pairs in the genome (9). In one such case, this characteristic was related to the inability for a gene to be expressed: in the T . gambiense LiTat 1.6 gene family, we have noted the presence of a member apparently unable to be expressed because of the lack of the 5‘ block of homology thought to be generally required for gene conversion (57). One can conceive that such antigen-specific sequences could only be expressed by partial gene conversion with some other genes that would share with them “unusual” sequence homology. According to this view, some genes could be expressed late because their duplicative transposition would require the presence, in the expression site, of sequences shared by a limited number of other genes. Interest-

?

'f

..

11

GENE CONVERSION

ingly, partial gene conversions (about 2 kb long) have so far only been observed for late antigenic types (AnTat 1.1 and 1.13: 29, 102). The differential rate of antigen gene activation, leading to a semi-ordered succession of antigenic types during chronic infection ( 1 7-21 ), could thus be related to the nature and extent of homology between each gene and the expression site. If the two recombinant genes display extensive similarity, as in the AnTat 1.1to 1.10, or 1.10 to 1.1B switches, the converted stretch can be smaller than the coding sequence, provided that the rearrangement leads to enough antigenic variation to ensure its selection. The smallest ELC would accordingly only represent the sequence(s) coding for the surface-exposed antigen epitope(s). It is clear that only the 5’ half of the gene codes for the surface-exposed epitope, since the replacement of the 5’ half only leads to antigenic variation in the AnTat 1.1B clone (Fig. 2). However some kind of additional constraint seems to preside over the generation of successful ELCs by gene conversion between closely related sequences, since the 3’ gene conversion endpoint appears to be the same in two independent AnTat 1.l-expressor clones, AnTat 1.1B and 1.1C (43; also see Figs. 2 and 5). Moreover, in an AnTat 1.10B clone, derived from AnTat l.lB, the gene rearrangement is very similar to that of AnTat 1.10 (E. Pays, S. Houard, A. Pays, S. Van Assel and M. Steinert, unpublished; see Fig. 2), suggesting that the DNA recombinations are not effected at random, but that the total number of possible rearrangements is limited. A possible interpretation is that only recombinations occurring outside the sequences specifying the different protein domains could be successful for the synthesis of antigens. As a matter of fact, in AnTat 1.1B and 1.1C genes, the gene conversion ends in a DNA region roughly corresponding to a site hypersensitive to trypsin in native antigens ( 4 ) ,such as a hinge between two protein domains; this has been confirmed by conformaFIG.4. Restriction maps of different tt~lomel-escarrying sequences involved in antigen gene expression, in a series of trypanosome clones derived from AnTat 1.1 (-. = switching of antigenic type; = trypanosome cloning). Clones of successive variants, all grown in mice, were set up from heterotypes arising in the diversifying clone of the previous variant. Symbols used for restriction endonuclease sites are: B=BgII; Ra=BamHI; Bg=BglIl; C=CloI; E=EcoRI; H=HindIII; Hi=HinfI; K=KpnI; M=MspI; P=PstI; Pv=PouII; S=SalI; Sp=SphI; Ss=SstI, T=TuyI. Sites labelled with a dot seem conserved in several maps. The maps have arbitrarily been aligned on a SphI site (arrow).The bars under RC maps represent the known extent of the ELCs; in AnTat 1.1 map, the ECL/a is the one found in variant AnTat 1.1, whereas ELC/b is the transposed copy in AnTat 1.1C (black box in AnTat 1.1C map). The boxes represent the known extent of the cDNAs. Also indicated under some maps are the cDNA fragments used as hybridization probes for the construction of the restriction maps.

-

12

ETIENNE PAYS

5’ JUNCTIONS

(a)

...AGCGGACGTAAGGAAACGA t .

-

CAGGCAGCAACACAAGACATTTTCAGCGACC...

( b ) ..GAAGCCAAAGAGGGAGAGC AACTCATTTTCACCCCTAGTGCCAACAATGGTCGC.

t

(e)

...AAAATTGTAATGTTAGTCGt GTGCCGCACTGACACTACACCAACAACAAGCT...

(f)

...CAAAAGAATTGACTCATATt GCTGTGGCATGCTATAGAGGAAAAAAAT...

(MiTat 1.4 E X )

(MiTat 1.5 BC)

..

...TAATAATAATAGAt AGAGTGTTGTGAGTGTGTGTA...

...TAATAATAAGAGTGTGTTGTGAGTGTGTATA...

3’ JUNCTIONS

c(c)

...GCTAAACTGGCAACTGTAA t

(d)

...GCCCKGGACTTCGATGCC t CACATCAAAAAAGTG...

(g)

...GCCTGAGGACTTCGACGCC tCACATCAAAAAAGTG...

(MiTat 1.4a)

CAGCGGCACGACCTACAAACAAG...

...CCTCTATTCTAGTAACCAAtGAAATTCGCC...

(MiTat 1.4b)

...TGGGA

(MiTat 1.5a)

...AACAAGCAAACAATTCGCCt TTCAGCCTGGTI’...

(MiTat 1.5~)

...AATTTTTGCTACTTGAAAAt ACTTTTGATATATTT...

(MiTat 1.5d)

...ATTCGCCCTAATGGTTTCTt TCTGCATTT...

t

GAATAATAAT..

a

FIG.5. Endpoints for gene conversion. This figure shows the DNA sequence at the borders ofsome converted domains ofgenes AnTat 1.1, 1.10, l . l B , l.lOB, and 1.1C (Fig. 2: junctions a to g), as well as the gene conversion limits in MiTat 1.4 and 1.5 hypanosome clones. Sequence (a) is from Michiels et al. (97);sequences (b) to (g)are from Pays

GENE CONVERSION

13

tional analysis of the AnTat 1.1B protein, based on the gene sequence, which indicates that the 3‘ gene conversion endpoint hinges two different domains (P. Degand, personal communication). In AnTat 1. l B and l . l C , as well as in AnTat 1.10 and 1.10B, the other gene conversion endpoint is located close to the 5’ extremity of the antigen coding sequence (Fig. 2), while in different MiTat 1.4 and 1.5clones (30,33, 56),the 3’ gene conversion endpoint is found in the last codons of the mature protein, or close downstream. The extent of the gene conversion may thus be related to the coding specificity of the DNA sequence. If, in contrast to the AnTat 1.1, 1.10 and 1.1B cases, the target sequence for the gene conversion is very different from the donor one, the converted stretch can be very long, as illustrated in the AnTat 1.1C to 1.3B switch (43; also see Fig. 4).The restriction map of the AnTat 1.1C gene-containing telomere is indeed different from that of the telomere harboring the AnTat 1.3B gene and from the other known antigen gene expression sites (Fig. 4).In particular, the AnTat 1.1C gene-containing telomere does not share the SphIIBglIIPuuIIIPstI/ BgZIIIHindIII restriction sites cluster, which seems conserved in several telomere maps in the AnTAR 1 repertoire (Fig. 4),and which could be present, at least partially, in the antigen gene expression site in other repertoires (31,3 3 , 4 9 ) . Moreover, the AnTat 1.1C telomere, as well as, for instance, the telomere carrying the MiTat 1.2 gene in another repertoire ( 4 4 ) ,is not characterized by the presence of a large “barren” region directly upstream from the antigen gene domain, as is the case for most telomere maps (Fig. 4).These particularities of the AnTat 1.1C telomere could explain why the ELC of the variant directly derived from AnTat l.lC, namely AnTat 1.3B, is very large, at least 40 kb long (43; also see Fig. 4): it probably extends to a region of homology very distant from the antigen-specific sequence. Such large telomeric conversions could be responsible for the similarities observed between different telomeres (Fig. 4). et al. (43, and unpublished). The 3‘ gene conversion endpoints of MiTat 1.4 and 1.5 genes are from Michels et al. (33).The 5’ limit of MiTat 1.4ELC is from Campbell et ul. (79), while the 5‘ homology block of the MiTat 1.5 BC is from Liu et u1. (58).Arrows below the junction sequences refer to the putative conversion endpoints. In all cases except two (d and f), this point cannot be precisely determined, due to some extent of sequence identity between the donor and the target; the endpoint is then considered as the minimum limit of conversion. The direction of conversion is indicated by a horizontal arrow. In (c), it is hypothesized, although other possibilities are not ruled out, that a first conversion, upstream from the junction, was followed by a second, downstream, conversion involving an unknown donor. Sequences of the “CACA” type (see text) are underlined.

14

ETIENNE PAYS

The variability in the size of the converted sequences apparently differentiates antigenic variation in trypanosomes from the mechanism of mating type interconversion in yeast: in the latter, gene conversion seems to involve sequences of the same size, namely between two large and conserved blocks of homology that flank the involved sequences (64,65,80-83). But in fact, the actual extent of this conversion is variable. Although the process is clearly triggered at a specific locus downstream from the MAT locus, the terminal recombination can take place in several locations within the 5‘ homology block (J. Strathem, personal communication).

VI. Gene Conversion Endpoints Figure 5 summarizes the information collected about the DNA sequence at the junction between converted stretches and their flanking sequences. While the conversion limits cannot be characterized by a rigorously conserved sequence, it is interesting to note the presence of several C-A or G-T dinucleotides, sometimes in reverse orientation (such as CAAC or GTTG), near all 5’ conversion endpoints analyzed so far. These sequences belong to the “CACA” category described by Rogers (84). It has been observed that, in numerous systems, CACA sequences behave as hotspots for recombination, in particular in y-globin gene conversion (72) or yeast mating type interconversion (CAACA: 85).It has been hypothesized (84,86)that these similarities may reflect some common basic requirements for the association, cutting and resealing of the DNA strands involved in recombination. In the yeast mating type interconversion system, the sequence CAACA is specifically cleaved by an endonuclease, which triggers the gene conversion process (61,87).By analogy, it can be speculated that an endonuclease digestion in, or near ACAAUTGTTG sequences, upstream from the antigen gene, could lead to the conversion associated with antigenic variation in trypanosomes (Fig. 3). Since no evident conserved sequence can be identified in 3‘ conversion endpoints, it is possible that the process could be resolved by a crossing-over downstream from the initial cut, in a region of homology between the two DNA partners (Fig. 3).However only some successful recombinations could subsequently be selected, since the 3’ gene conversion limit does not appear to be randomly distributed when present within the antigen-coding sequence itself (AnTat 1.1B and 1.1C: see above, and Fig. 5), whereas it is highly variable when present at the end or downstream of the gene (33; Fig. 5).

15

GENE CONVERSION

VII. Gene Conversion Frequency Contrary to the high frequency of mating type interconversion in yeast (88), the rate of antigenic switching in trypanosomes is low, (3, 19, 20, 25). This compares with the freprobably around quency of mitotic gene conversion in yeast and mouse cells (122,123). However, this rate should be considered as a minimal estimate, since only DNA rearrangements leading to the appearance of a different and not yet expressed antigen type can be selected in the bloodstream. Moreover, every possible recombination could not be equally successful in the production of surface antigens (see above). At least in bloodstream trypanosomes, gene conversions seem to occur spontaneously, probably during cell division. Being an intrinsic property of the cell, this mechanism could be controlled by a low amount of a site-specific recombination enzyme ( 5 ) ,but the presence of such an enzyme has never been demonstrated. It may well be that the seemingly specialized rearrangements underlying antigenic variation are performed by the general recombination machinery, but operating in “hotspots” sites, as in immunoglobulin heavy chain class switching (124, 125). Although not induced by external agents, trypanosome antigenic variation does not, however, occur totally at random, since some antigens are produced more often than others ( 8 , 1 9 , 2 1 ) .Obviously, it is advantageous for the parasite to change its surface antigen infrequently and not randomly, as a rapid and random antigenic switching would entail the risk of expressing the whole antigen repertoire in the first parasitemia waves. By comparison, the mating type interconversion mechanism of yeast appears to be a highly specialized, programmed gene converThe frequent and regular induction of mating type switchsion (64). ing in yeast indeed depends on the controlled activation of a gene (HO) coding for a specific endonuclease, at a precise moment of the cell life-cycle, in connection with the onset of DNA replication (61,

85, 87).

VIII. Orientation of the Gene Conversion Mechanism; Relationship with Transcription Gene conversion is, by definition, an oriented process in which a target sequence is replaced b y a copy of a donor sequence. Models accounting for the evolution of multigene families, which involve not only homogenization, but also polymorphism of the different mem-

16

ETIENNE PAYS

bers, depend on the need for a directed recombination mechanism, with preferred targets for DNA sequence transposition (I, 2,89). The single assumption involving the existence of some preferential target sequences is indeed sufficient to explain how multigene families, such as the class I histocompatibility antigen gene family, can develop extensive allelic polymorphism while maintaining specific family characters. AnaIysis of some examples of gene conversion in trypanosome antigenic variation provides hints as to how this orientation could be achieved. We have observed, without exception, that the converted sequences are the transcribed ones (43, 59). A relevant example is provided by the analysis of the expression of sequence A from the AnTat 1.1gene family (see Fig. 1):this sequence can be either a donor or a target in the gene conversion process, but it is a target only when used for transcription (Fig. 2). Since the transcribed antigen-specific sequences are characterized by a chromatin configuration highly sensitive to DNase I (27, 29, 33, 34, 43, 59), it would appear that an open” chromatin structure allows a sequence to behave as preferential target for the site-specific endonuclease that triggers the gene conversion, especially if the amount of enzyme is low. However, this observation is seriously limited by the fact that gene conversions in silent genes could not be detected, since they are not selected for. In this respect, however, an interesting comparison can again be made between antigenic switching in trypanosomes and mating type interconversion in yeast. In the latter, when the silent HML and HMR loci are allowed to express, they become effectively converted into the opposite mating type, a process that occurs normally only at the “active” MAT locus (90). In other words, the HML and HMR sequences behave as targets for gene conversion once they are transcribed, probably because only active chromatin allows the double-strand cut that triggers the conversion (91).By analogy it is tempting to speculate that in trypanosomes, the transcriptional activity of an antigen gene could preferentially enable it to be converted. Accordingly, only functional antigen genes, but not pseudogenes (such as the “2.15 kb” and “4 kb” AnTat 1.1gene family members, which carry a stop codon within the antigen coding sequence: 59, and unpublished data) could play the role of preferred target sites for gene conversion. Moreover, since apparently only telomeric antigen-specific sequences are allowed to be transcribed, the chromosome-internal antigen genes could perhaps only behave as potential donors (BCs) for gene conversion. If transcription of a gene may condition its conversion, could conversely a DNA rearrangement be a prerequisite for transcription? In “

17

GENE CONVERSION

the trypanosome antigenic variation system, this question could be resolved by the complete DNA sequencing of every antigen gene that becomes activated without duplication, and comparison of these sequences with those of their silent equivalents in nonexpressor clones. So far, this has only been done for the AnTat 1.1C gene (43), where a partial gene conversion has clearly been found in connection with gene activation. Interestingly, in a similar case (MiTat 1.2 gene: 92), restriction map alterations have been noticed in the vicinity of the activated antigen gene: they could perhaps be due to gene conversion. In the mating type interconversion in yeast (93, 94), or phase variation in Salmonella (86),the gene rearrangement triggers the onset of transcription, by a so-called “position effect.” In these cases, the transcription promoter is indeed derepressed in transposed copies. I n the trypanosome antigenic variation system, some observations suggest that the promoter implicated in expression may be absent from the converted element (28, 58, 95-99). However, examples of gene activation by very large gene conversion, without detectable alteration over a considerable distance (43), are difficult to reconcile with this view.

IX. Gene Conversion and Antigen Repertoire Evolution Gene conversion can partially or completely transform antigen gene copies (ELCs), or even genes themselves, as schematized in Fig. 2. If, in a series of different trypanosome variants, gene conversion is effected in alternation with a mechanism of antigen gene activation that does not involve duplication of the gene, some converted sequences can be conserved, at the expense of others (39,43; see two examples in Fig. 6). The conserved sequences could theoretically be either ELCs identical to their template, chimeric ELCs, or even the genes themselves, possibly after conversion. Each of these combinations would contribute to a rapid evolution of the antigen gene repertoire, as suggested by the following considerations. The conservation of an ELC identical to its template obviously adds a new member to its gene family. This is illustrated for AnTat 1.3 and 1.16 gene families in Fig. 6. In different cases, conserved ELCs seem to be preferentially reexpressed, by either nonduplicative activation (100,101) or duplicative transposition (43,102). The conserved ELCs can thus be used efficiently as template for new ELCs, and so are not likely to be lost rapidly. It is possible that these additional

3

probes :

52-

16

-

2.7 1 2.-

probes :

dGENE6

6

U-

1c

FIG.6.Gain and loss of different antigen genes during antigenic variation. The top panels illustrate the gain of sequences by ELC conservation (AnTat 1.3 or l.lbspecific, for left and right panel, respectively), whereas the bottom panels show the loss of the AnTat 1.6 and 1.1C genes, due to a conversion by the ensuing ELC. The DNAs are from AnTat 1.1, 1.3, 1.6, 1.16, 1.1C, and 1.3B clones, successively derived in this order as shown in Fig. 4; probes are AnTat 1.3, 1.16, 1.6,and 1.1C-specific,as indicated. Digestions are by SphI, HindIII, EcoRI+BglII, and PstI, from first to last panel, respectively. In this clone series, only the antigen genes of the AnTat 1.6 and 1.1C variants (dots in bottom panels) are not activated by duplication. The AnTat 1.3B ELC is surimposed to the conserved AnTat 1.3 ELC (14.9 kb fragment in last lane of first panel) (43),and the AnTat 1.1 ELC is in the 2 kb PstI fragment in first lane of last panel (12).

GENE CONVERSION

19

sequences further evolve into new antigen genes, since in the AnTat 1.1gene family, it is thought that sequence B (see Fig. 1)has evolved from a copy of sequence A, to eventually encode for a surface antigen of a different serological specificity (AnTat 1.1 instead of 1.10) (59). Similarly, Young et al. (103) concluded that antigen gene families evolve by mutations in conserved ELCs. Evolution of antigen-specific sequences, such as conserved ELCs, could be speeded up by hypermutagenesis (104),this process probably being restricted to telomeric sequences (104,105).The rapid evolution of conserved ELCs would explain the generation of unique telomeric antigen genes such as AnTat 1.3 (34),as well as the limited distribution of such genes in Such a fast evolution of trypanosome stocks taken up in the field (104). antigen-specific telomeric sequences has actually been noted for the BC of the “companion” sequence (29). Partial gene conversions in ELCs can create chimeric structures such as the AnTat 1.1B transcribed element, which is made up of at least four recombined sequences (Fig. 2). Conservation of these rearranged ELCs directly provides the cell with a new gene family member, different from the others. Conservation of chimeric ELCs has actually been observed in the AnTat 1.1and 1.13 gene families (101, 102). The generation of such chimeric genes could be a way for the parasite to expand its antigenic potential late in chronic infection, when the repertoire is largely expressed. If successive variants each utilize the nonduplicative mode of gene activation, only translocation of the corresponding antigen genes could possibly be expected. Since these genes can be altered when they are transcribed (see AnTat 1.1C gene in Fig. 2 ), their conservation also leads to a modification of the antigen repertoire. Furthermore, the location of the antigen genes seems to influence their ability to be expressed early or late in chronic infection (102).Translocation of antigen genes could therefore modulate the expression of the repertoire. If the gene conversion applies to a gene activated without duplication, all or only a fraction of this gene can be lost from the genome and replaced by a new ELC. If the replaced sequence was one determining the antigenic specificity of the surface-exposed epitope, inevitably this antigenic type will be lost from the repertoire unless other sequences are present that code for the same antigenic specificity. This has been observed in the AnTat 1.1C clone, where the sequence converted (sequence A of the AnTat 1.1 gene family) was coded for the AnTat 1.10 antigenic type (59). Accordingly the AnTat 1.10 coding sequence was lost from the AnTat 1.1C DNA, and, as predicted, no

20

ETIENNE PAYS

variant expressing AnTat 1.10 could yet be derived from the AnTat 1.1C or ensuing clones (N. Van Meirvenne, unpublished). Clear examples of complete loss of antigen genes are presented in Fig. 6. Such gene losses are due to replacement by the next ELC (39). Expansion of the antigen gene repertoire could be achieved by the conservation of the target, when gene conversion applies to whole telomeres. This hypothesis is based on the analysis of the “minichromosome” observed in the IlTat 1.3 gene family (45).This “minichromosome,” of about 80 kb, indeed looks like the relic of an antigen gene expression site, actually harboring an inactive ELC (103).One could speculate that, in a telomeric sequence, the initial doublestrand cut triggering the conversion can take place far upstream from the gene, releasing a large piece of DNA that could not be completely degraded by exonucleolytic digestion as in shorter target sequences (60),but on the contrary could be conserved in the genome provided that it contains a centromere (Fig. 3). This hypothesis provides an explanation for the large number of antigen-specific telomeric sequences observed in the Trypanosoma brucei genome (51,59). As examplified with the sequence A of the AnTat 1.1 gene family, the same antigen gene can be activated by either gene conversion or non-duplicative activation (43). Similarly, the MiTat 1.2 antigen gene can also be activated by both mechanisms (44),and this might indeed be frequent for telomeric genes (41,106,107).Therefore, it is possible that the same gene could be amplified in some repertoires or lost from others, depending on the way the two activation mechanisms alternate. However, there are cases, as for the AnTat 1.6 antigen, where the gene activation proceeds only, or at least mainly, by the nonduplicative mechanism (101). It is therefore possible to predict both the gain and loss of different genes when the clone derivation is selected to occur through AnTat 1.6. In this way, it has yet been possible to observe the amplification of the AnTat 1.3 gene family from one to three members (39).Conversely, clones lacking both AnTat 1.6 and 1.10 antigenic types have been obtained (43). Since the AnTat 1.6 gene is one of the few to be activated in the metacyclic trypanosomes of the fly salivary glands, it is interesting to note that, following infection of flies with clones lacking the AnTat 1.6 gene, AnTat 1.6 disappeared from the metacyclic repertoire. These are only a few examples of directed transformations of trypanosome antigen repertoires; for instance, we could transform a non-telomeric (and late expressed) antigen gene into a telomeric (and early expressed) one, by conservation of the ELC through AnTat 1.6 derivation (102).

21

GENE CONVERSION

X. Sexual Conjugation Further Leads to Evolution of the Antigen Gene Repertoire At least several antigen-specific sequences appear to be haploid (44,101,108, 109),although being in the context of a diploid genome ( 1 1 0 , 1 1 1 ) . In particular, it is clear that the ELC is haploid (12,14,44, 109). Recent evidence suggests that sexual conjugation takes place during the parasite development in the fly (L. Jenni, personal communication). Such a conjugation occurring between cells within a clone would obviously make it possible for at least some haploid sequences to become diploid. This phenomenon, together with the conservation of ELCs, or the loss of genes (see preceding paragraph), may lead to changes in the copy number of the antigen-specific sequences. This is indeed reflected in the comparative analysis of different antigen gene families in several T . brucei stocks (57,104). If sexual conjugation takes place between stocks harboring such differences in antigen gene families, it is expected to lead to further changes in the antigen gene repertoire, due to the random fusion of gametes containing different subsets of the gene families. This has been directly observed in the AnTat-1.8-specific pattern of populations obtained following cyclical transmission of mixed T . brucei clones from different stocks (P. Paindavoine, E. Pays, L. Jenni and M . Steinert, unpublished). The interplay between gene conversion and sexual conjugation therefore also contributes to shaping the antigen repertoire of the parasite.

XI. Problems and Outlooks Some major problems are still unsolved in the analysis of the gene control during antigenic variation. Extensive studies are now being conducted in order to elucidate the nature of the nonduplicative mechanism of gene activation. In addition, it is not yet understood how the choice of the gene activation mechanism is determined, or how transcription of the antigen gene is controlled. It has been proposed (34,43, 92) that the nonduplicative gene activation could be a telomeric reciprocal recombination. However, except in one case (112),attempts to locate the hypothetical crossing-over point have so far failed (39,43,44,113). Any other model would imply the complex control of different possible telomeric antigen gene expression sites; since the trypanosome genome is probably made up of a large collection of chromosomes (51, 59, 113), this problem might be difficult to

22

ETIENNE PAYS

solve. It is conceivable that an unique piece of mobile DNA could function as telomere “activator” by transposition into any one of the possible telomeric expression sites. In at least some instances, this hypothesis is unlikely. When telomeric antigen genes are activated by very large gene conversions (43),no sequence alteration seems linked to the gene activation within more than 40 kb upstream. A means for the selective telomere activation may be provided by the existence of a complex and unusual telomeric DNA modification system (101,108, 109). This allows us to differentiate active from inactive telomeres, since the modification is never observed in actively transcribed genes. The search for the antigen gene transcription promoter has led to the discovery of a 35 nucleotide-“mini-exon” transcript placed in front of antigen mRNAs (44,95-99,114-116). It seem likely now that this small RNA can be found as common spliced leader for a large number of different (possibly all) trypanosome mRNAs, either after discontinuous transcription initiated b y the synthesis of a 137 nucleotide“mini-exon” transcript (117-120), or due to a posttranscriptional RNA rearrangement analogous to polyadenylation (121). This original transcription procedure has probably no specific connection with the mechanism of antigenic variation by itself. The study of antigenic variation in trypanosomes has revealed original ways for eukaryotic genes to be differentially activated. It has shown how rapidly antigen gene families can evolve and diversify. This great variation potential obviously raises the question of the feasability of a vaccine against trypanosomiases. To tackle this problem, the best approach, for two reasons, is to analyze the metacyclic antigenic types: first, these antigens seem to represent only a limited fraction of the whole repertoire; secondly, they are the first ones to be observed on invasion of the mammalian bloodstream, and are therefore the obligatory pathway for antigenic variation. in T. b. gambiense, responsible for the sleeping sickness disease in man, antigenic variation seems to be much more limited than in other trypanosome subspecies (57; A. B. Eldirdiri, H. S. Abdalla, E. Magnus, N. Van Meirvenne and D. Le Ray, unpublished), perhaps due to a reduced number of telomeric antigen genes. There is thus some hope that a vaccine against T. b. gambiense metacyclic antigens can be developed.

ACKNOWLEDGMENTS I thank Prof. M. Steinert, Prof. P. Borst, and Dr. N. Murphy for suggestions and comments on the manuscript, M. Laurent, M. Guyaux, S. Van Assel, N. Van Meirvenne, and D. Le Ray for helpful discussions and aid. Investigations reported herein received support from the Commission of the European Communities (TSD-M-023-B),from the FRSM (Brussels), from the ILRAD/Belgian Research Centres Agreement for Collabora-

23

GENE CONVERSION

tive Research (Nairobi), and from the Trypanosomiases component of the UNDPiWorld BanklWHO Special Programme for Research and Training in Tropical Diseases (Geneva).

REFERENCES I. F. Bregegere, Biochimie 65, 229 (1983). 2. 3. 4. 5. 6. 7. 8. 9.

T. Otha, Genetics 106,517 (1984). K. Vickerman, Nature 273,613 (1978). G. A. M. Cross, Proc. R. Soc. Lond. B 202, 55 (1978). P. Borst and G. A. M. Cross, Cell 29, 291 (1982). P. T. Englund, S. L. Hajduk and J. C. Marini, ARB 51, 695 (1982).

M. J. Turner, Ado. Parasitol. 21, 69 (1982). A. Capbern, C. Giroud, T. Baltz and P. Mattern, Exp. Parasitol. 42, 6 (1977). L. H. T. Van der Ploeg, D. Valerio, T. De Lange, A. Bernards, P. Borst and F. G. Grosveld, NARes 10,5905 (1982). 10. J . H. J . Hoeijmakers, P. Borst, J. Van den Burg, C. Weissmann and G. A. M. Cross, Gene 8,391 (1980). 11. P. Borst, A. C. C. Frasch, A. Bernards, L. H. T. Van der Ploeg, J. H. J. Hoeijmakers, A. C. Arnberg and G. A. M. Cross, CSHSQB 45,935 (1981). 12. E. Pays, N. Van Meirvenne, D. Le Ray and M. Steinert, PNAS 78,2673 (1981). 13. M. Lheiireux, M. Lheureux, T. Vervoort, N. Van Meirvenne and M. Steinert, NARes 7, 595 (1979). 14. J. H. J. Hoeijmakers, A. C. C. Frasch, A. Bernards, P. Borst and G. A. M. Cross, Nature 284, 78 (1980). 15. E. Pays, M. Delronche, M. Lheureux, T. Vervoort, J. Bloch, F. Gannon and M. Steinert, NARes 8,5965 (1980). 16. M. Milhausen, R. G. Nelson, M. Parsons, G. Newport, K. Stuart and N. Agabian, Mol. Biochem. Parasitol. 9, 241 (1983). 17. A. R. Gray,]. Gen. Microbiol. 41, 195 (1965). 18. G. J. C. McNeillage, W. H. Herbert and W. H. R. Lumsden, E x p . Parasitol. 25, 1 (1969). 19. N. Van Meirvenne, P. G. Janssens and E. Magnus, Ann. SOC. Belge Mid.Trop. 55, 1 (1975). 20. E. N. Miller and M. J. Turner, Parasitology 82,63 (1981). 21. S . L. Hajduk and K. Vickerman, Parasitology 83,609 (1981). 22. S . L. Hajduk, C. R. Cameron, J. D. Barry and K. Vickerman, Parasitology 83, 595 ( 1981). 23. J. S. Crowe, J. D. Barry, A. G. Luckins, C. A. Ross and K. Vickerman, Nature 306, 389 (1983). 24. M. F. Shirazi, M. Holman, K. M. Hudson, G. G. B. Klaus and R. J. Terry, Parasite Inzmunol. 2, 155 (1980). 25. J. J. Doyle, H. Hirumi, K. Hirumi, E. N. Lupton and G. A. M. Cross, Parasitology 80, 359 (1980). 26. R. 0 .Williams, J. R. Young and P. A. 0. Majiwa, Nature 282,847 (1979). 27. E. Pays, M. Lheureux and M. Steinert, Nature 292,265 (1981). 28. E. Pays, M. Lheureux and M. Steinert, NARes 10,3149 (1982). 25. E. Pays, S . Van Assel, M. Laurent, B. Dero, F. Michiels, P. Kronenberger, G. Matthyssens, N. Van Meirvenne, D. Le Ray and M. Steinert, Cell 34,359 (1983). 30. A. Bernards, L. H. T. Van der Ploeg, A. C. C. Frasch, P. Borst, J. C. Boothroyd, S. Coleman and G. A. M. Cross, Cell 27, 497 (1981). 31. L. H. T. Van der Ploeg, A. Bernards, F. A. M.Rijsewijk and P. Borst, NAAes 10,593 (1982).

24

ETIENNE PAYS

32. S. Longacre, U. Hibner, A. Raibaud, H. Eisen, T. Baltz, C. Giroud and D. Baltaz, MCBiol. 3,399 (1983). 33. P. A. M. Michels, A. Y. C. Liu, A. Bernards, P. Sloof, M. M. W. Van der Bijl, A. H. Schinkel, H. H. Menke, P. Borst, G. H. Veeneman, M. C. Tromp and J. H. Van Boom, J M B 166,537 (1983). 34. M. Laurent, E. Pays, E. Magnus, N. Van Meirvenne, G . Matthyssens, R. 0. Williams and M. Steinert, Nature 302, 263 (1983). 35. E. Pays and M. Steinert in “Hormones and Cell Regulation” (3. E. Dumont and J. Nunez, eds.), Vol. 8, p. 289. Elsevier, Amsterdam, 1984. 36. J. R. Young, J. E. Donelson, P. A. 0. Majiwa, S. Z. Shapiro and R. 0. Williams, NARes 10,803 (1982). 37. J. E. Donelson, J. R. Young, D. Dorfman, P. A. 0. Majiwa and R. 0. Williams, NARes 10,6581 (1982). 38. N. A. Penncavage, M. A. Julius and K. B. Marcu, NARes 11,8343 (1983). 39. M. Laurent, E. Pays, K. Delinte, E. Magnus, N. Van Meirvenne and M. Steinert, Nature 308,370 (1984). 40. P. A. 0. Majiwa, J. R. Young, P. T. Englund, S. Z. Shapiro and R. 0. Williams, Nature 297, 514 (1982). 41. J. R. Young, E. N. Miller, R. 0.Williams and M. J. Turner, Nature 306,196 (1983). 42. M. Parsons, R. G. Nelson, G . Newport, M. Milhausen, K. Stuart and N. Agabian, Mol. Biochern. Parasitol. 9, 255 (1983). 43. E. Pays, M. F. Delauw, S. Van Assel, M. Laurent, T. Vervoort, N. Van Meirvenne and M. Steinert, Cell 35, 721 (1983). 44. A. Bernards, T. De Lange, P. A. M. Michels, A. Y. C. Liu, M. J. Huisman and P. Borst, Cell 36, 163 (1984). 45. R. 0. Williams, J. R. Young and P. A. 0. Majiwa, Nature 299, (1982). 46. T. De Lange and P. Borst, Nature 299,451 (1982). 47. A. Raibaud, C . Gaillard, S. Longacre, U. Hibner, G. Buck, G. Bernardi and H. Eisen, PNAS 80,4306 (1983). 48. E. H. Blackburn and P. B. Challoner, Cell 36,447 (1984). 49. A. Bernards, P. A. M. Michels, C. R. Lincke and P. Borst, Nature 303,592 (1983). 50. L. H. T. Van der Ploeg, A. Y. C. Liu and P. Borst, Cell 36,459 (1984). 51. E. Pays, M. Laurent, K. Delinte, N. Van Meirvenne and M. Steinert, NARes 11, 8137 (1983). 52. G . Matthyssens, F. Michiels, R. Hamers, E. Pays and M. Steinert, Nature 293,230 (1981). 53. E. Pays, M. Lheureux and M. Steinert, NARes 9,4225 (1981). 54. M. Majumder, J. C. Boothroyd and H. Weber, NARes 9,4745 (1981). 55. A. C. Rice-Ficht, K. K. Chen and J. E. Donelson, Nature 294,53 (1981). 56. P. A. M. Michels, A. Bernards, L. H. T. Van der Ploegand P. Borst, NARes 10,2353 (1982). 57. E. Pays, P. Dekerck, S. Van Assel, A. B. Eldirdiri, D. Le Ray, N. Van Meirvenne and M. Steinert, Molec. Biochem. Parasitol. 7, 63 (1983). 58. A. Y. C. Liu, L. H. T. Van der Ploeg, F. A. M. Rijsewijk and P. Borst,/MB 167,57 (1983). 59. E. Pays, S. Van Assel, M. Laurent, M. Darville, T. Vervoort, N. Van Meirvenne and M. Steinert, Cell 34,371 (1983). 60. J. W. Szostak, T. L. Orr-Weaver, R. J. Rothstein and F. W. Stahl, Cell 33,25 (1983). 61. J. N. Strathem, A. J. S. Klar, J. B. Hicks, J. A. Abraham, J. M. Ivy, K. A. Nasmyth and C. McGill, Cell 31, 183 (1982). 62. J. B. Hicks, J. N. Strathern and I. Herkowitz, in “DNA Insertion Elements, Plasmids and Episomes” (A. I. Bukhari, J. Shapiro and S. Adhya, eds.), p. 452 Cold Spring Harbor, New York, 1977.

GENE CONVERSION

25

A. J. S. Klar, J. McIndoo, J. N. Strathern and J. B. Hicks, Cell 22,291 (1980). J. E. Haber, D. T. Rogers and J. H. McCusker, Cell 22,277 (1980). J. E. Haber and D. T. Rogers, Nature 296,768 (1982). A. L. M. Bothwell, M. Paskind, M. Roth, T. Imaniski-Kari, K. Rajewsky and D. Baltimore, Cell 24,625 (1981). 67. P. H. Schreier, A. L. M. Bothwell, B. Mueller-Hill and D. Baltimore, PNAS 78, 4495 (1981). 68. M. Bruggemann, A. Radbruch and K. Rajewsky, EMBO]. 1,629 (1982). 69. R. Dildrop, M. Bruggemann, A. Radbruch, K. Rajewsky and K. Beyreuther, EMBO J . 1,635 (1982). 70. D. L. Bentley and T. H. Rabbitts, Cell 32, 181 (1983). 71. R. 0110 and F. Rougeon, Cell 32, 515 (1983). 72. J. L. Sligthom, A. E. Blechl and 0. Smithies, Cell 21,627 (1980). 73. L. R. Pease, D. H. Schulze, G. M. Phffenbach and S. G. Nathenson, PNAS 80,242 (1983). 74. E. H. Weiss, A. Mellor, L. Golden, K. Fahrner, E. Simpson, J. Hurst and R. A. Flavell, Nature 301, 671 (1983). 75. D. H. Schulze, L. R. Pease, S. S . Geier, A. A. Reyes, L. A. Sarmiento, R. B. Wallace and S. G. Nathenson, PNAS 80,2007 (1983). 76. A. L. Mellor, E. H. Weiss, K. Ranlachandran and R. A. Flavell, Nature 306, 792 (1983). 77. G. S . Roeder and G. R. Fink, PNAS 79,5621 (1982). 78. T. De Lange, J. M. Kooter, P. A. M. Michels and P. Borst, NARes 11,8149 (1983). 79. D. A. Campbell, M. P. Van Bree and J. C. Boothroyd, NARes 12,2759 (1984). 80. J. Hicks, J. N. Strathem and A. J. S. Klar, Nature 282, 478 (1979). 81. K. A. Nasmyth and K. Tatchell, Cell 19, 753 (1980). 82. J. N. Strathem, E. Spatola, C. McGill and J. B. Hicks, PNAS 77, 2839 (1980). 83. C. Astell, L. Ahlstrom-Jonasson, M. Smith, K. Tatchell, K. A. Nasmyth and 8.D. Hall, Cell 27, 15 (1981). 84. J. Rogers, Nature 305, 101 (1983). 85. R. Kostriken, J. N. Strathem, A. J. S . Klar, J. B. Hicks and F. Heffron, Cell 35, 167 (1983). 86. M. Simon, J. Zieg, M. Silverman, G. Mandel and R. Doolittle, Science 209, 1370 (1980). 87. K. Nasmyth, Nature 302,670 (1983). 88. I. Herkowitz and Y. Oshima, i n “The Molecular Biology of the Yeast Saccharomyces cereuisiae: Mating Type and Mating-Type Interconversion” (J. N. Strathern, E. Jones and J. Broach, eds.), Vol. 1, p. 181. Cold Spring Harbor, New York, 1981. 89. D. Baltimore, Cell 24, 592 (1981). 90. A. J. S. Klar, J. N. Strathern and J. B. Hicks, Cell 25, 517 (1981). 91. K. Nasmyth, Cell 30,567 (1982). 92. P. Borst, A. Bernards, L. H. T. Van der Ploeg, P. A. M. Michels, A. Y. C. Liu, T. De Lange and P. Sloof, in “Proc. Cetus-UCLA Meeting on Molecular Biology of HostParasite Interactions” (N. Agabian and H. Eisen, eds.). Park City, Utah, in press. 93. A. J. S. Klar, J. N. Strathem, J . R. Broach and J. B. Hicks, Nature 289,239 (1981). 94. K. A. Nasmyth, K. Tatchell, B. D. Hall, C. Astell and M. Smith, Nature 289, 244 ( 1981). 95. L. H. T. Van der Ploeg, A. Y. C. Liu, P. A. M. Michels, T. De Lange, P. Borst, H. K. Majumber, H. Weber, G. H. Veeneman and J. Van Boom, NARes 10,3591 (1982). 96. J. C. Boothroyd and G. A. M. Cross, Gene 20,281 (1982). 97. F. Michiels, G. Matthyssens, P. Kronenberger, E. Pays, B. Dero, S. Van Assel, M. Darville, A. Cravador, M. Steinert and R. Hamers, E M B O J . 2, 1185 (1983).

63. 64. 65. 66.

26

ETIENNE PAYS

98. T. D e Lange, A. Y. C. Liu, L. H. T. Van der Ploeg, P. Borst, M. C. Tromp and J. H. Van Boom, Cell 34,891 (1983). 99. R. G. Nelson, M. Parsons, P. J. Barr, K. Stuart, M. Selkirk and N. Agabian, Cell 34, 901 (1983). 100. P. A. M. Michels, L. H. T. Van der Ploeg, A. Y. C. Liu and P. Borst, EMBOJ. 3, 1345 (1984). 101. E. Pays, M. Laurent, M. F. Delauw and M. Steinert, in “Genome Rearrangement. UCLA Symposia on Molecular and Cellular Biology, New Series, Vol. 20” (1. Herskowitz and M. Simon, eds.). Alan R. Liss, New York, 1985. 102. M. Laurent, E. Pays, A. Van der Werf, D. Aerts, E. Magnus, N. Van Meirvenne and M. Steinert, NARes 12, 8319 (1984). 103. J. R. Young, J. S. Shah, G. Matthyssens and R. 0. Williams, Cell 32, 1149 (1983). 104. A. C. C. Frasch, P. Borst and J. Van den Burg, Gene 17, 197 (1982). 105. E. Pays, M. Lheureux, T. Vervoort and M. Steinert, MoZ. Biochem. Parasitol. 4,349 (1981). 106. G. A. Buck, S. Longacre, A. Raibaud, U. Hibner, C . Giroud, T. Baltz, D. Baltz and H. Eisen, Nature 307, 563 (1984). 107. P. Myler, R. G. Nelson, N. Agabian and K. Stuart, Nature 309,282 (1984). 108. A. Bernards, N. Van Harten-Loosbroek and P. Borst, NARes 12,4153 (1984). 109. E. Pays, M. F. Delauw, M. Laurent and M. Steinert, NARes 12,5235 (1984). 110. A. Tait, Nature 287,536 (1980). 1 1 1 . P. Borst, M. Van der Ploeg, J. F. M. Van Hoek, J. F. M. Tas and J. James, M o l . Biochem. Parasitol. 6, 13 (1982). 112. E. Pays, M. Guyaux, D. Aerts, N. Van Meirvenne and M. Steinert, Nature, in press. 113. L. H. T.Van der Ploeg, D. C. Schwartz, C. R. Cantor and P. Borst, Cell 37, 77 (1984). 114. R. G. Nelson, M. Parsons, M. Selkirk, G. Newport, P. J. Barr and N. Agabian, Nature 308, 665 (1984). 115. T. D e Lange, P. A. M. Michels, H. J. G. Veerman, A. W. C. A. Cornelissen and P. Borst, NARes 12,3777 (1984). 116. T. D e Lange, T. M. Berkvens, H. J. G. Veerman, A. C. C. Frasch, J. D. Barry and P. Borst, NARes 12,4431 (1984). 117. M. Parsons, R. G. Nelson, K. P. Watkins and N. Agabian, Cell 38,309 (1984). 118. D. A. Campbell, D. A. Thornton and J. C. Boothroyd, Nature 311,350 (1984). 119. J. M. Kooter, T. D e Lange and P. Borst, EMBOJ. 3,2387 (1984). 120. M. Milhausen, R. G. Nelson, S. Sather, M. Selkirk and N. Agabian, Cell 38, 721 (1984). 121. S . M. Berget, Nature 309, 179 (1984). 122. J. A. Jackson and G. R. Fink, Nature 292, 306 (1981). 123. R. M. Liskay and J. L. Stachelek, Cell 35, 157 (1983). 124. T. Kataoka, S. I. Takeda and T. Honjo, PNAS 80,2666 (1983). 125. A. Rosen and G. Klein, Nature 306, 189 (1983). Addendum: Translocation of such haploid telomeric sequences to the end of other chromosomes, as observed in one instance (Van der Ploeg et al., Cell 39, 213, 1984), would lead to internalization of formerly telomeric genes. Repetition of this process would cluster the antigen genes in tandem arrays near the ends of chromosomes, in accordance with findings that chromosome-internal antigen genes are clustered (9)and flanked by sequences normally found at telomeres (48).This hypothesis would readily explain why antigen genes, even located within large diploid chromosomes, are haploid

(5,44,101,110,111).

Hypermodified Nucleosides of

tRNA: Synthesis, Chemi

"r.

and Stnrctural Features o Biological Interest

RYSZARDW. ADAMIAK AND PIOTR G6RNICKI

Institute of Bioorganic Chemistry Polish Academy of Sciences Poznan, Poland

I. Hypermodified Nucleosides of tRNA: A Bioorganic Chemist's View . 11. Synthesis and Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Hypermodified Uridines ................................... B. Hypermodified Adenosines. . . . . . . . C. Wyosine and Derivatives . . . . . . . . . 111. Sh-uctural Features of Hypemiodifiecl Nucleosides and Codon-Anticodon Interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Unique Features Imposed by Hypermodification ...... B. Hypermodified N d e o s i d e s within the Anticodon Loop Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . C. Involvement in Codon-Aitticodon Interaction. . . ...... References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

27 35 35

53 54 59 64 67

1. Hypermodified Nucleosides of tRNA: A Bioorganic Chemist's View T h e biological function of modified nucleosides' is one of the Zeitmotifs of the very extensive studies of transfer ribonucleic acids (tRNA) that have been carried out in the last 20 years. Despite many attempts and much experimental data, little is yet known about the biological rationale for developing rather complicated biosynthetic

The numbering system of nucleosides in tRNA and abbreviations of modified nucleosides are those of Gauss and Sprinzl ( I For the abbreviations of hypermodified nucleosides, see legends to Fig. 1, 2, and 3. The numbering system of Gauss and Sprinzl (NARes 12, rl, 1984) differs in only minor detail from that of Singhal et ol. (in this series, 23, 227, 1979, and 28,211, 1983). Residues 20:1,20:2,47:1,47:2,etc., ofthe earlier compendia become 20:A, 20:B, 47:A, 47:B, etc. in the latter. All numbers not so subdivided are the same in both, running from 1 to 76. Almost 300 sequences are listed in the compilations of Singhal et a / . Both compilations are available in computer language. (Eds.) 27 Progress in Nucleic Acid Research and Molecular Biology, Vol. 32

Copyright 0 1985 by Academic Press, Inr. All rights of reproduction in anv form reserved.

28

RYSZARD W. ADAMIAK AND PIOTR G6RNICKI

machinery for posttranscriptional nucleoside modification by an organism, but even less is known about the molecular mechanism of the function of modified nucleosides. Many of them are simple methyl derivatives of the common nucleosides, but 5,6-dihydrouridineY pseudouridine, thio-analogs of uridine, cytidine, and ribothymidine, N4-acetylcytidine, inosine, and a group of more than 20 rather complex nucleosides, called hypermodified, have also been found in tRNA (Figs. 1, 2, and 3 ) . Closer inspection of more than 280 known tRNA sequences (1;see also 9) reveals that the modified nucleosides are located in certain positions of tRNA molecules, and that the anticodon loop is especially rich in modification (modified nucleosides constitute about 25% of an average anticodon loop). An interesting feature is that all of the hypermodified nucleosides are located in this region: hypermodified uridines and queuosine (and derivatives) in the first (“wobble”) position of the anticodon, and hypermodified adenosines and wyosine (and derivatives) in the position adjacent to its 3’ end (Table I). Moreover, the structure of hypermodified nucleosides is correlated with the structure of the neighboring nucleoside (Table I) (see Section 111). The only exception is hypermodified uridine acp3U (X)which has been found in other regions of tRNA, in the variable loop at position 47 and in the dihydrouridine loop at position 20:A.l3 The problem of hypermodified nucleosides has been the subject of many review papers concerning their isolation and identification (2-6), occurrence and location in tRNA (2-5, 7-11), and function (2-5, 7,8,10-18a). A bibliography compiled by Agris (19) covers the studies from 1970 to 1979. Those studies concerning hypermodified nucleosides can be classified, somewhat arbitrarily, into two main streams, biochemical and bioorganic. However complementary and extensively interconnected, they differ in the methods and techniques applied, but what is even more important, in our opinion, is that they search for answers to two different questions: what is the biological function of hypermodified nucleosides and how do they act on the molecular level? Before entering the discussion on the results of the bioorganic approach we would like to comment on some general problems. There are several factors that make drawing conclusions about hypermodified nucleosides function especially difficult. First, tRNAs are involved in many activities and they interact with other RNAs and a variety of proteins during their lifetimes (7,20-22). Thus, functional studies should include studies of different phenomena, from tRNA transcription and processing to nucleolytic degradation, and should

HYPERMODIFIED NUCLEOSIDES OF

29

tRNA

R

R

R

(I1

(111

(1111

I

R (XI

FIG.1. Hypermodified uridines found in tRNA. (R = p-D-ribosyl): (I) S-methoxyuridine (mo5U);(11) riridine-5-oxyacetic acid (o"U, ciiio"U, or V); (111) nritline-5-oxyacetic (mainssszU acid methyl ester (mc1nn5Uor mV); (IV) 5-mt~thylaminoniethyl-2-thiouridine or ninni5sW); (V) 5-carboxymethylaniinonicthyluridine (cninnv5U); (VI) 5-carboxymethylaminomethyl-2-thionridine (ciiinmsss"U);(VII) 5-methoxycarhonylniethyluridine (mcmsU);(VIII) 5-methoxycarbonylmcthy l-2-thionridine (mcmsss"U);(IX)5-carboxyhydroxyniethylnridine (chmW); (X) 3-(3-aniin0-3-carboxypropyl)nridine (acpW or X). Other abbreviations: 5-carboxynietliyliiritlirle (cm5U); 5-carbainoylinethylii~idine (ncm5U);5-carboxymethyl-2-thiniiridiiie(cm5s2U).

be carried out at different levels: cellular, ribosomal, and at the level of interaction with other RNAs and individual proteins. In general, such studies have been based on the comparison of the biological properties of tRNAs (in vitro or in vivo) in which a hypermodified

30

RYSZARD W. ADAMIAK AND PIOTR CdRNICKl

R

R

IXII)

XI)

R ( XVI)

0 FOOH II HNCNHCHCHCH3

R (XVII

0 FONHC(CH20H)3 I HNkNHCHCHCH3

Ql$Y 1

OH

R (XIX)

FIG.2. Hypermodified adenosines found in tRNA. (R = p-D-ribosyl):(XI)N6-(2isopenteny1)adenosine (PA); (XII)2-methylthio-N6-(2-isopenteny1)adenosine( ms2i6A); (XIII) cis-ribosylzeatin(io6A);(XIV) 2-methylthio-cis-ribosylzeatin (ms2i06A);(XV)W (threoninocarbony1)adenosine(PA);(XVI) NG-methyl-W-(threoninocarbony1)adenosine (mt6A);( M I )2-methylthio-N6-(threoninocarbony1)adenosine(ms2PA);(XVIII)P-(glycinocarbony1)adenosine ($A); (XIX) N-{N-[(9-~-~-ribofuranosylpurin-6-yl)carbamoyl] threonyl}-2-amino-2-hydroxymethyIpropane-1,3-diol. Other abbreviations: N6-(2-isopenteny1)adenine (i6Ade); 2-methylthio-N6-(2-isopenteny1)adenine (ms2i6Ade); W(threoninocarbony1)adenine(t6Ade);W-(glycinocarbony1)adenine (86Ade).


31

tRNA HNCOOCH3 I

C H 300CyH ?42

HyCOOCH3 C H 3 0 0 C FH

HNCOOCH3 I

CH300CFH

HC-OH

HC-OOH

I

CH2 0 C H 3 4 3I - 3 9 CH3 R

CH3

I xxrrr1

IXXIII

R’= H

/3- D -mannosy1 p - 0 -galactosyl

lXXIVI

FIG.3. Hypermodified nucleosides related to guanosine found in tRNA. (R = p-Dribosyl): (XX) wyosine (W); (XXI) wybutosine (yW); (XXII) wybutoxosine (ozyW); (XXIII) hydroxywybutosine (oyW); (XXIV)R’ =H, queuosine (Q), R‘ = /3-D-mannoSyl, P-D-mannosylqueuosine (manQ), R’ = P-D-galactosyl, /3-D-galactosylqueuosine (galQ).

nucleoside is missing or specifically altered; if the effect of the hypermodification is not drastic, it may easily escape detection. Analysis of tRNA sequences, the approach used for many years, from the early proposal by Nishimura (4) to the evolutionary analysis by Cedergren et al. ( I ] ) , shows that in many cases the modification

TABLE I MODIFICATIONPAITERNOF THE ANTICODONLOOPDEDUCED FROM tRNA SEQUENCES BY GAUSSAND SPRINZL (1)" Position in tRNA ~~

35

36

37

Q

U U U U U U U

U U U U U

PA mt6A ms2PA A m'A,m6A,m'G

U

OW moiU

U

U

A A A A

32

33

U,Y,C,Cm U,Y,C,Cm,A C

u,c u,c

\Y,c

C,Nb C,Cm U,Um,C,Cm

C C U,*,C U,Um,\V,\Ym, C,Cm,G

U U

U U U,N U

U,C

34

G mam5szU mcm5s2U cmnm5sW U C,s2U,N

U,C,Cm,m5C,ac4C,C,Gm,I,N

C C C u,v,c U,Um,l,C,Cm, m3C,s2C,A

U U U

mcmsU cmnm5U chmsU

U U

U

U,Um,Cm u,c U,\Y,C,m3C U,Um,\Y,C,Cm, m3C,A,N

U U U U

C

U

U,C,Cm,G,I,A,N

C C C C C

G G G G G

i6A ms2i6A A m'G,m2A,N

*

G

m7G

38

YW,OZYW

A,* A A A,C A

U,C,A,mfiAA,m2A,m1G,m'I ifiA/ms2i6A FA

U,q,C,mSC,A,G,N A A

A,m2A,mRA,G,m*G,N

m'G

A

-

0 Positions 32-35: correlations between the structure of "wobble iucleosides and their neighbors. Positions 36-38: correlation between the structure of nucleosides adjacent to the 3'-end of the anticodon and their neighbors. N, unidentified nucleoside.


tHNA

33

pattern is species-dependent (eukaryotic vs prokaryotic tRNAs, cytoplasmic vs organelle tRNAs etc.) and/or sequence-specific (the structure of the modified nucleoside depends on the preceding and following nucleosides). For example, the phenylalanine tRNAs (anticodon GAA or GmAA) from mycoplasma, E . coli, and plants contain at position 37 m'G, ms2i6A(XII), and ozyW (XXII), respectively ( I ) . Thus the structure of the nucleoside next to essentially the same anticodon changes from the simple methyl to the most complex guanosine analog containing (in its side chain) the very unusual and controversial hydroxyperoxy group (23-27). In addition, many new sequences are reported every year and our knowledge about tRNA modification patterns is changing; for example, PA (XV) once believed to occur only in a sequence U-FA, has recently been found also in a sequence C-t6A, in mosquito mitochondrial tRNA@& (28). Second, the number and structural variety of hypermodified nucleosides is rather large and many still unidentified nucleosides have been found in sequenced tRNAs (comparing to tRNA sequencing techniques, structural elucidation of an unknown nucleoside requires substantial amounts of material). Moreover the discussion is concentrated on those hypermodified nucleosides localized in tRNA sequence, but several structurally related nucleosides have been found in unfractionated tRNAs from different organisms. This group includes ribosylzeatin (XIII) (29-32) and 2-methylthioribosylzeatin (XIV) (30, 31,33), nucleosides found in plant tRNA and showing cytokinin activity (34),g6A (XVIII) found in yeast tRNA (3,5),and W (XX)found in T . u t i l i s tRNA (36).The group includes also analogs of mcm5U (VII): 5-carboxymethyluridine (cm5U) (38) and 5carbamoylmethyluridine (ncmW) (39, 40), and e'-O-methyl derivatives of all three nucleosides ( 4 1 , 42). It has been suggested that the nucleosides are biosynthetically related (39) and that in native, mature tRNA the carboxyl group is blocked by esterification (43,44).The detection of the respective methyl transferase activity seems to support the hypothesis that the free carboxylic acid is a precursor of mcmW (VII) (44, 45). Recently, selenium-containing nucleosides have been found in some bacterial (45b,c, d , e,f )and probably also in mammalian tRNAs (45g).The presence of selenium in Clostridiurn sticklandii tRNAG'I1is essential for aminoacylation ( 4 5 4 . One such nucleoside, 5-methylaminomethyl-2-selenouridine, has been identified and its synthesis has been described (45f, h ) . On the other hand, identification of a t6A analog (XIX) in E . coli tRNA (45a)illustrates the way in which the structure of a hypermodi-

34

RYSZARD W. ADAMIAK AND PlOTR G6RNICKI

fied nucleoside can be altered by conditions (in this case, by growing bacteria in the presence of Tris buffer) employed for tRNA isolation, degradation, etc. And finally, at least in some cases, the level of hypermodification is lower (hypomodification) in tumors and rapidly proliferating cells and tissues, and it changes dramatically during starvation of certain nutrients and during metamorphosis. The meaning of such alterations, relatively well documented for wyosine and queuosine and its derivatives (15-18a), remains unclear. In general, it has been postulated that hypermodification does not change the biological properties of tRNAs drastically, but rather serves for the fine-tuning of a process in which tRNA is involved allowing, for example, smooth and in-phase translation by adjusting the codon-anticodon association energy as proposed by Grosjean and Chantrenne ( 1 2 ) .It may mean that hypermodification is not absolutely required but may confer some selective advantage on an organism. The aim of this article is to give a brief review of the achievements and drawbacks of the bioorganic approach including synthetic, chemical, and structural studies of hypermodified nucleosides carried out at different levels of complexity, from nucleoside to tRNA. Both the applications of chemical properties of hypermodified nucleosides for tRNA studies as well as the ability of this approach to explore their functions are discussed. However, the specific profile of this review and its length impose certain limitations. The results of numerous biochemical studies dealing with the problem, and results of the tRNA studies based on the application of the unique properties (chemical and physicochemical) of hypermodified nucleosides, are not included. Neither are the problems concerning queuosine and its derivatives, which have been recently reviewed by several authors (15-1 8 ) . What is the status of the bioorganic approach? Chemical modification methods, despite earlier expectations, have failed to give definite answers, mostly because of the analytical problems. and interpretation difficulties. In many cases, the effect of the modification was either very small or absent, and in some other cases bulky and hydrophobic groups were introduced into tRNA, making it rather difficult to attribute the effect of the modification merely to its influence on hypermodified nucleoside function. The studies, however, yielded several methods of introducing different labels (fluorescent, photoreactive, antigenic, etc.) to specific sites in tRNA molecules. Such tRNA derivatives have been useful in


tRNA

35

studies on tRNA interactions with elongation factors, ribosomes, etc.; in this case, lack of any greater influence of the chemical modification becomes an advantage. On the other hand, there is a large gap in the structural studies between crystal structures, solved with enough accuracy only for two tRNAs, and rather incomplete studies on nucleoside and oligoribonucleotide conformation, their ability for hydrogen bonding and stacking. X-Ray crystallographic analysis and high resolution NMR spectroscopy are the only techniques available today that can give such detailed information. Synthetic methods developed in different laboratories enable preparation of most hypermodified nucleosides and also oligoribonucleotides containing some of them. Many low-yield and multistep synthetic routes need improvements. Large quantities of the oligonucleotides are necessary for physicochemical studies, and also for the preparation of “chimeric” tRNAs by chemical and enzymatic methods (46-48). The latter approach, enabling introduction of specific alterations to the anticodon loop at both positions commonly occupied by hypermodified nucleosides b y introducing their precursors or analogs, seems very promising. It has recently been applied in studies on biosynthesis of some hypermodified nucleosides at the “wobble” position of tRNA (48-50).

II. Synthesis and Chemistry A. Hypermodified Uridines 1. SYNTHESIS

The nature of the chemical bond joining the pyrimidine ring at carbon atom 5 and a side-chain dictates the mode of synthesis of modified uridines, either by chemical transformations at the C-5 of uridine and its derivatives, or by total synthesis from ribose and modified uracils. I n the case of mo5U (I) and OW(11), having an ether linkage at the junction, methylation or carboxymethylation of 5-hydroxyuridine obtained from uridine (51) was chosen, respectively (52, 53). The synthesis of mo5U (I) from methoxyuracil has also been described (54). In the case of mam5s2U (IV) mcm5U (VII), and mcm5s2U (VIII), having a carbon-carbon bond at the junction, condensation of 2’,3‘,5’0-protected-1-halogen- or acetyl-ribose with modified uracils has been applied. This method is especially effective for 2-thio analogs

36

RYSZARI) W. ADAMlAK AND PIOTR G6RNICKI

CZH~OCOCHCH NCH3

2l

BOC

H

(XXV)

I Si(CH313

(XXVI)

(XXVII)

i

HO OH ( IVl

BzoY$ BzO

OBz

(XXIX)

FIG.4. Total synthesis of mam5s2U (IV) (58, 59).

due to the higher reactivity of thiourea compared to urea in the pyrimidine ring-closure reactions. Baczynskyj et al. (55)described the synthesis of mcm5s2U(VIII) from activated ribose and modified bases prepared from thiourea and the dimethyl ester of formylsuccinic acid. In this case, and also in the case of mcm5U (VII) and cm5U, the application of the mercury salt method for nucleoside bond formation gave rather low final yields (56, 57). The synthesis of mam5s2U (IV) (58, 59) is an example of a more efficient approach (Fig. 4). The inteimediate (XXV) prepared by the formylation of ethyl 3-methylaminopropionate with ethyl formate in the presence of sodium hydride, after protection of its secondary amino group with a t-butoxycarbonyl group (Boc), was cyclized with thiourea to a 5-substituted 2-thiouracil (XXVI) with a 60%yield. The silylation of (XXVI)with hexamethyldisilizane to (XXVII) followed by the condensation with activated ribose (XXVIII) in the presence of SnC12 or AgC104 as a catalyst led to the O-protected uridine (XXIX) with a 70% yield with the simultaneous removal of the Boc protecting group (59).(IV) was obtained with an 80% yield after removal of the

X~

~

c

H

z

o

37

tRNA


H X

I

(XXXI

I

CH3NH2

X

I

~

..

R’

x.0.s

R’:

(1x1

Rd

I R’

]

IMeOH

HCI

x=o R’= ribose

0 0

X

k (IV) R=

HO OH

x=s

R

(VIII (VIII)

x:o x=s

FIG. 5. Synthesis of mam5s2U (IV), mcm5U (VII), and mcm5s2U (VIII) using 5chloromethyluridine as key intermediate (61).

benzoyl groups with methanolic ammonia. A similar method of nucleoside synthesis, in which 2,2,2-trichloroethoxycarbonyl or trifluoroacetyl group was applied for the side chain protection, has been reported (60). Ikeda et al. (61) described the synthesis of mam5s2U(IV), mcm5U (VII), and mcm5s2U(VIII) based on transformations of 2’,3’-isopropylidene-5-chloromethyluridineor its 2-thio analog (XXX) as a key intermediate (Fig. 5).Aminolysis of the intermediate (XXX) with an appropriate glycine ester was also employed in preparation of cmam5U (V) and cmam5s2U(VI) (62).The two nucleosides can also be prepared in good yield from 2 ,3 -isopropylidene-5-pyrrolidinomethyluridineor %thiouridine, an intermediate capable of reacting with the t-butyl ester of glycine, when transformed into methiodide (63).

38


Malkiewicz and Sochacka (60, 6 4 ) recently accomplished the chemical synthesis of the anticodon triplets mam5s2U-U-U and mcm5s2U-U-Uby the phosphotriester method. The synthesis of acp3U (X)based on the alkylation of 2’,3‘-isopropylideneuridine sodium salt with ethyl 2-benzamido-4-bromobutyrate was accomplished during its structure elucidation (64a);acp3U was obtained in moderate yield after the removal of the protecting groups with hydrochloric acid. 2. CHEMISTRY Rao and Cherayil (65) described the desulfurization of thionucleosides in 35S-labeled E. coli tRNA by various chemical reagents. Cyanogen bromide, hydrogen peroxide, and sodium periodate, being the most reactive, led to complete desulfurization, whereas hydroxylamine, iodine, nitrous acid, potassium permanganate, and sodium bisulfite caused only partial desulfurization. Cyanogen bromide (66-68) under mild conditions (pH 8.9, room temp.) transforms 2-thiouridines into the respective isothiocyanates (XXXI) and under more drastic conditions (pH 8.5, 100°C) into 5-substituted uridines (69).This re-

(XXXII

agent was used in the structure elucidation of cmam5s2U(VI) (70)and for the modification of mam5s2U (IV) and mcm5s2U (VIII) in various tRNAs (68, 71-76). Detailed analysis of the modified tRNAs showed little or no effect of the replacement of thioketo by keto groups on their aminoacylation (71-73) but a decreased affinity toward the respective aminoacid-tRNA ligases of the modified tRNAs bearing an -SCN group (71,73,76). Cyanogen bromide-treated (under both conditions) tRNAs containing s2U derivatives at the “wobble” position retained the ability to form complexes with elongation factor Tu and GTP (72), but their ribosome binding was decreased (72) or even abolished (71). Both unmodified and CNBr-modified yeast tRNAkys containing mcm5s2U (VIII) or mcm5U (VII) at the “wobble” position could recognize AAA much better than AAG. This suggested that the 2-thio group was not responsible for the observed “wobble” properties of mcm5s2U(VIII) (72).The results were confirmed by studies on the coding properties of yeast tRNAgg containing mcm5U (VII) (77).


tRNA

39

Of the other desulfurization reagents, iodine (78) and hydrogen peroxide (79) have also been used for tRNA modification. The latter reagent was employed by Watanabe (79) for studies on mam5s2U(1V)modified E . coli tRNA$’“. The reaction was followed by CD measurements (the thioketo group exhibits a characteristic band above 300 nm). A lower reaction rate observed on the tRNA level, compared with that on the nucleoside level, was interpreted as an indication of mam5s2U(IV) involvement in the anticodon structure (stacking). The highly specific reaction with cyanogen bromide was not used because the thiocyanate derivative of uridine (XXXI) also shows a CD band above 300 nm (80). Yang and So11 (81)reported the attachment of a fluorescent probe for measurements of distances to mam5s2U (IV) in E . coli tRNAGIL1 between different parts of tRNA by an energy transfer technique. tRNA was selectively modified with 4-bromomethyl-7-methoxy-2-oxo2H-benzopyran adsorbed on Cellite. Under heterogeneous conditions described as nondenaturating, pseudouridines present in this tRNA and capable of reacting were not accessible to the reagent. Alkylation of both thioketo and secondary amino groups of the side-chain of mam5s2U (IV) was considered, but the exact structure of the product was not determined. Attachment of a hydrophobic group to mam5s2U (IV) by its modification with the reagent mentioned above (81)or with benzoic acid anhydride (82) abolished most aminoacylation ability. Caron and Dugas (83,84) introduced spin labels to E . coli tRNAG’” with the aim of studying its structure by EPR spectroscopy, by modification of mam5s2U(IV) with reagents of the anhydride and carbodiimide type capable of reacting with the secondary amino group, and bromoacetamide-type reagents capable of reacting with the thioketo group. Ofengand et al. (85-86) showed that tRNAs containing mo5U (I) or 05U (11) at the “wobble” position, when bound to the ribosomal P-site, can be very efficiently crosslinked to the RNA of the small ribosomal subunit by UV irradiation (over 300 nm). Chemical, photochemical, and sequence studies revealed that the crosslink has a cyclobutane structure (XXXII) that involves the modified uridine in the “wobble” position in tRNA and cytidine in rRNA (87-90). Further analysis of the phenomenon yielded important information about codon-anticodon interactions and ribosome architecture (87, 91 ). The unique chemical reactivity of the carboxyl group of uridine-5oxyacetic acid (11) was employed for attachment of the dinitrophenyl antigen and photoreactive azidonitrophenyl group to the anticodon of E. coli tRNAyd. In the first step, the carboxyl group was condensed

40


t UNA

r RNA

R = H , COOH

h vl >30 0n m)

I

tRNA

I

hv I 2 5 L rim)

I

rRNA

( XXXII )

with ethylenediamine using water-soluble carbodiimide as the condensing agent. In the second step, the primary amino group introduced was acylated with N-hydroxysuccinimide ester. The modified tRNAs retained their ability for aminoacylation and ribosome binding. The dinitrophenyl-modified tRNA (XXXIII)was used for the localization of the codon-anticodon recognition site on E. coli ribosome by immunoelectron microscopy (92),and the azidonitrophenyl-modified tRNA (XXXIV) for the probing of the ribosome decoding site by an affinity-labeling technique (91).

I

tRNA

R = --NO2 -Ng

(XXXIII) (XXXIV)

The presence of an a-amino-acid side-chain and the location of acp3U (X)in tRNA create several possibilities for modification as a means of introducing different reporter groups (Fig. 6). The structure of the phenoxyacetyl derivative of acp3U (XXXV) was described inde-


41

tRNA

(XXXVI

IXXXVII I

IX X X V I 1

H

I

t RNA

IX L I I I I OH

FIG.6. tRNA derivatives prepared by chemical modification of the acpW (X) residue for structure-function studies.

pendently by Ohasi et al. (64a) and Friedman et al. (93, 94). The compound was isolated from unfractionated E . coli tRNA modified with the N-hydroxysuccinimide ester of phenoxyacetic acid. The aaminoacid residue of the nucleoside can be detected by the ninhydrin test ( M a ) . In most cases, N-hydroxysuccinimide esters were used as acylating agents. The antigenic-determining dinitrophenyl group (XXXVI) (95),

42

RYSZARD

W. ADAMIAK AND PIOTR G6RNICKI

photoreactive groups (XXXVII) (96,97) and (XXXVIII) (98), paramagnetic labels (XXXIX) (83,84,96,87), and fluorescent probes (XL) and (XLI) (99)were introduced into E. coli tRNAPheat position 47. Fluorescent tRNA derivatives were also prepared by modification with reagents specific for primary aliphatic amino groups such as fluorescamine (XLII) (100)and fluorescein isothiocyanate (XLIII) (101).The large excess of the modifying reagents and the addition of organic solvents such as dimethyl sulfoxide or acetonitrile necessary to carry out the reaction raised a question about the biological activity of the modified tRNAs. Despite some earlier controversies (84, 94, 96, 97, 99,102), it is believed now that the modification of acp3U at position 47 has no effect on the aminoacylation of E. coli tRNAPh"(100, 101, 103). For some of the modified tRNAs, other biological properties were also tested. They were shown to be able to support polyphenylalanine synthesis (99b, 100, IOl),to form the ternary complex with elongation factor Tu and GTP (100, 101, 103), and to bind to both A and P ribosomal sites (100, 103). Absorbtion and fluorescence of the fluorescein-modified tRNA (XLIII) responded markedly to the addition of Mg2+.The label appeared to be a sensitive probe of tRNA structure (101). E. coli tRNAPhe bearing an azidonitrophenyl group attached to acp3U47was successfully used for studies on the interaction of tRNA with elongation factor Tu and on ribosome topography (104,105). For the same purpose, the amino group of acp3U located in the dihydrouridine loop at position 20:A, in yellow lupine tRNA,M"' was modified with a photoreactive azidonitrophenyl reagent (106).

8. Hypennodified Adenosines 1. SYNTHESIS N6-Isopentenyladenosine (XI), the first hypermodified nucleoside of tRNA to be identified (107,108), received much attention in the late 1960s. The earlier studies, including synthesis and chemistry of i6A (XI), ms2i6A (XII), and ribosylzeatin (XIII) have been reviewed by Hall (109). Until then, two general methods of i6A synthesis were known, one by aminolysis of 6-chloropurine-ribonucleosidewith isopentenylamine (110), the other by alkylation of adenosine with isopentenyl bromide followed by a Dimroth rearrangement (111).A synthesis of ms2i6A (XII) by the aminolysis of 2,6-bismethylthiopurine with isopentenylamine followed by a coupling to ribose has been reported (112). truns-Ribosylzeatin (113),its 2-methylthio analog (32),


I‘

( XLVI

1

43

tRNA

xL1v

( XIV

1

FIG.7. Synthesis of ms2i06A(XIV)(33).

and cis-ribosylzeatin (XIII) (11 4 ) were also obtained by the aminolyand its cis sis method using hydroxy-3-methyl-trans-2-butenylamine isomer, respectively. For the synthesis of 2-methylthio-cis-ribosylzeatin (XIV), 2,6-dichloropurine was chosen as a substrate because of the higher reactivity of C(6)-C1 than C(6)-SCH3 toward nucleophiles. After its fusion with tetra-0-acetyl-P-D-ribose (115),the mixture of a and 0 anomers of XLV was reacted with cis-hydroxyamine (XLIV) prepared according to the scheme depicted in Fig. 7. Removal of acetyl groups with ammonia led to the 2-chloro derivative (XLVI) with a 70% yieId. Reaction of the intermediate with sodium methyl mercaptide followed by chromatography on Sephadex LH-20 gave the final product (6-anomer), XIV, in 18%yield (33) (Fig. 7). A new efficient method of i6A synthesis, especially useful for the synthesis of PA-containing oligoribonucleotides, based on aminolysis of a 6-methylsulfonyl purine system, formed by oxidation of its 6methylthio precursor built into a protected oligoribonucleotide, has been reported recently (116).

44

RYSZARD W. ADAMIAK A N D PIOTR G6RNICKI

0

oc3:ib

i XLVIII 1

I ac3;ib

IX L V I I 1

ac3tib

I XLIX 1 FIG.8. Two routes of PA (XV)synthesis (117,118).

Ureidonucleosides belong to a rare class of natural products containing an N,N’-disubstituted urea system. Chheda and Hong (117) proposed two methods for synthesis of FA (XV) and g6A (XVIII) (Fig. 8). In the first method, the exo-amino group of O-protected adenosine (XLVII) was transformed with ethyl chloroformate into an ethoxycarbony1 derivative (ethyl carbamate) (XLVIII) that, after aminolysis with L-threonine and removal of acetyl groups, gave t6A (XV) with a 25% overall yield. The second method required the preparation of a dibenzyl derivative of L-threonine isocyanate. Its reaction with an 0protected adenosine (XLVII) gave the intermediate XLIX with a 48% yield. Benzyl groups were removed with sodium in liquid ammonia (117) or in an improved method by hydrogenolysis (118).Both methods were quite satisfactory for the synthesis of some artificial ureido nucleosides (118-120), but the yields of t6A, &A, and the respective 5’-phosphates were rather low (121 ). The first method, employing ethyl carbamate (XLVIII) aminolysis, was further improved (122). Amino-acid esters more reactive in aminolysis were also used for the preparation of PA (123,124)and its 3’,5’-cyclic phosphate (125).mtGA (XVI) can be obtained by the isocyanate method from N6-methyladenosine (119)or by methylation of t6A (XV) (124).


45

tRNA

0

-

0

COOCH:

NH 1 2

0

II

H,

0 II

COOH

I

CI O O C H G N 0 2

HNCNHCHCHCH3 OH

-

I

I

ILIVJ

- protected oligoribonucleotide

0- unprotected

oligoribonucleotide

FIG.9. Chemical introduction of the threonylcarbonyl side chain of FA (XV)to an adenosine residue in a protected oligoribonucleotide (126, 127).

Adamiak and Stawiliski (126)described the method of synthesis by the N,N’-disubstituted urea system, designed for the introduction of the threonylcarbonyl moiety into an adenosine unit located in a protected oligoribonucleotide (Fig. 9.). The free exo-amino group of the adenosine unit (L) was transformed into a phenoxycarbonyl derivative (phenyl carbamate) (LI)by reaction with phenoxycarbonyltetrazole, a reagent unreactive toward other sites of the protected oligoribonucleotide. The phenyl carbamate (LI) was then subjected to aminolysis with L-threonine p-nitrobenzyl ester. Both reactions were quantitative, carried out under very mild conditions and enabling a “one flask synthesis” (127).The free hydroxyl group of the threonine residue was then protected b y the tetrahydropyranyl group (LIII) to prevent hydrolysis of the t6A unit (to adenosine) during deblocking of the oligoribonucleotide under basic conditions. This method was used for the synthesis of the oligoribonucleotide C-C-C-A-U-@A-A of type (LIV), which sequence is identical with that of the anticodon loop of tRNAyetfrom yellow lupine (127).It appears to be more efficient than

46

RYSZARD W. ADAMIAK A N D PIOTR C6RNICKI

the oligoribonucleotide synthesis based on the use of a t6A monomer building block, the approach applied for the preparation of U-PA

(123).

2. CHEMISTRY The presence of a reactive double bond of the allylic type decides the unique properties of i6A (XI) and its derivatives, and has stimulated relatively extensive chemical studies. i6A, during prolongated storage, undergoes transformation to fluorescent derivatives LV and LVI (128). Most likely an intramolecular cyclization involving the endonitrogen atom N( 1)of the purine residue and the double bond is followed by air oxidation to form LV which transforms into ethenoadenosine (LVI) with the release of an acetone molecule. There is no evidence so far that the reaction takes place at tRNA level, the method once suggested for the introduction of the fluorescent residue to the anticodon loop (128).

ci b

(LV)

l LVI I

The properties of i6A, such as instability in acid solutions, reactivity with iodine, potassium permanganate, and sodium periodate were reviewed in 1970 (109).The specificity of the reaction with iodine was further investigated; in the absence of thionucleosides in tRNA, only i6A is modified (129).Based on this observation is the method developed by Weygand-Durasevic et al. (130)for the introduction of a paramagnetic label into yeast tRNATyrwith the aim of monitoring conformational changes in free tRNA (130),in its complexes with E. coli tRNAVal(complementary anticodons) and with elongation factor Tu and GTP (131),and for studies on the anticodon loop interactions within ribosome by EPR (133). In this method, an iodo-derivative (LVII) obtained by the procedure described earlier (134),was further reacted with 4-amino-2,2,6,6-tetramethylpiperidine-l-oxyl to give the spin-labeled tRNA (LVIII). Oxidation of the i6A double bond with potassium permanganate is not specific enough to be applied on tRNA level (129) and even more complex transformations of ms2i6A(XII) can be expected because reagents like potassium permanganate, sodium periodate, and iodine are potential desulfurization agents (65).On the other hand, the selec-


(LVII1

47

tRNA

( LVIII

1

tive desulfurization of ms2ifiAto i6A on the tRNA level was achieved by the catalytical hydrogenolysis with Raney nickel under conditions suppressing hydrogenation of the allylic double bond. The method was developed in a search for the biosynthetic precursor of ms2i6A (135). The reaction of tRNA containing i6A with sodium bisulfite at neutral pH leads to the formation of adducts to the 5,6-double bond of uridines and to the double bond of i6A, but the uridine modification can easily be reversed by the subsequent incubation of the modified tRNA at a higher pH (9.0).The treatment does not affect the i6Amodification. This reaction has been studied on yeast tRNATy*(136)and E . coli tRNAPhe(137,138).The selective modification of the isopentenyl side chain of ms2i6Ain E . coli tRNAPhedoes not affect tRNA charging nor its recognition by the cognate Phe-tRNA ligase (138). Midden et al. (139)described a heavy atom tRNA labeling method for X-ray crystalographic analysis in which the double bond of i6A in yeast tRNATyrwas treated with osmium tetraoxide. The product, further stabilized with 2,2’-bipyridyl (L) was proposed to have a cyclic ester structure (LIX). The selective modification of i6A was achieved due to a reaction rate difference between i6A and pyrimidines with reactive 5,6-double bonds of more than three orders of magnitude.

Studies on the chemical stability of t6A (XV)carried out during its structure elucidation (140, 141 ) showed that the nucleoside treated with 0.1 N HCI at 100°C gave ribose and W(threoninocarbony1) adenine that, under more drastic conditions, was further degraded to

48

RYSZARD W. ADAMlAK AND PIOTR CdRNICKI

adenine and threonine. On the other hand incubation of t6A in 0.2 N ammonium hydroxide ( 100°C) gave adenine and the cyclic threonine carbamate (LX)formed by an intramolecular attack of the threonine hydroxyl group on the ureido carbonyl group. In a stronger base (0.1 N NaOH, 100°C),PA decomposes to threonine. mPA (XVI) is even more alkali-labile (119,142), but decomposes in a similar way to N6-methyladenosine. A partial decomposition of PA observed under milder conditions (pH over 8.5) (143)and of the mt6A at elevated temperatures in water (119)have to be considered in chemical and synthetic studies of both nucleosides. It also raises the question about the origin of N6-methyladenosine at position 37 of certain tRNAs (142).t6A is stable under conditions used for nucleoside detection at the picomole level in tRNA. The test includes oxidation with sodium periodate and reduction with tritium-labeled sodium borohydride at pH 8.5 (143). In pyridine in the presence of dicyclohexylcarbodiimide, the side chain of PA undergoes an intramoIecular condensation yielding the hydantoin derivative (LXI) (144). A different type of t6A reactivity toward carbodiimide was observed in water in the presence of amines (145).The carboxyl group of FA can be quantitatively condensed with amines (aniline, ethylene diamine, glycine ethyl ester) in the presence of a water-soluble carbodiimide, yielding the respective amides. Under conditions used for PA modification (pH about 4, ZOOC), the four common nucleosides and the internucleotide linkage of U-A are unreactive and 5'-AMP is transformed to the respective phosphoramidate at a rate less than onehundredth the P A modification rate. In general, the method can be applied for the modification of the hypermodified nucleosides bearing a free carboxyl group, and for their detection in tRNA by modification with tritium-labeled aniline (106).It has been applied to the modification of PA in tRNAyet, mt6A and acp3U in tRNA,M"' (from yellow lupine) (106), as well as for the modification of 05U (11)in E . coli tRNAy' (Section II,A,2). In another method, the carboxyl group of PA has been alkylated with the fluorescent reagent 4-bromomethyl-7-


tHNA

49

methoxy-2-oxo-2H-benzopyran in the presence of an anion-activating quaternary n-butylammonium salt (146).

C. Wyosine and Derivatives 1. SYNTHESIS The complex structure of the tricyclic nucleosides wyosine (XX), wybutosine (XXI), and wybutoxosine (XXII) presents a great synthetic challenge. Many problems concerning their synthesis were studied on the base level, i.e. for wybutine (LXII) [the base derived from wybutosine (XXI)] and for wyeine [the base derived from wyosine (XX)]in the early 1970s (147, 148) as indicated by Frihart (149) and described later in detail (150, 151). The alkylation-cyclization reaction of 3-methylguanine (LXIV) with the appropriate a-bromoketone (LXV) or bromoacetone was the crucial step in the first reported synthesis of wybutine (LXII) (147) and wyeine (LXIII) (148), respectively (Fig. 10). Such factors as the multistep synthesis of 3-methylguanine (LXIV) (151 ) and its low solubility, nonspecific alkylation, and the thermal lability of a-bromoketone (LXV) caused rather low 2 and 27% yields reported for wybutine (LXII) (147) and wyeine (LXrespectively. Protection of the N(7) site with a benzyl group 111) (150), removable by hydrogenolysis increased solubility of the substrate (LXVI) and eliminated side reactions. Yields of the desired products were increased to 20 and 82% for LXVII and LXVIII, respectively. Model studies presented b y Kasai et al. (150)showed that ribosylation of wyeine (LXIII) took place on the imidazole nitrogen, not that in natural wyosine (XX), and suggested for this group of nucleosides a synthetic route from a substrate with the proper nucleoside linkage. Two such routes were reported for wyosine (XX), but the synthesis of its more complex analogs has not been yet achieved. I n the first route, 3-methylguanosine (LXXII) was used as the key intermediate for the alkylation-condensation reaction with bromoacetone. Nakatsuka et al. (152) reported a multistep synthesis of LXXII from ~-amino-4-cyano-l-(tri-O-acety~-~-~-ribofuranosy~)imidazole via derivatives LXIX and LXX with an overall 10% yield (Fig. 11).Independently, the same S-methylisothiourea system was used for the formation of the pyrimidine ring during the synthesis of 3,9-dimethylguanine, the substrate for preparation of the methyl analog of wyosine (153). The route via 3-methylguanosine was further improved by Itaya and Ogava (154) and then simplified by the application of an N-cyanomethylamino derivative (LXXI) as the intermediate See Addendum, p. 321.

50


kH3 ( LXIIII

(

LXVlII

I

FIG. 10. Synthesis of hypermodified bases wybutine (LXII) and wyeine (LXIII) from 3-methylguanine (147-151).

in a relatively efficient synthesis of 3-methylguanosine (WXII) (155). Reaction with bromoacetone gave wyosine (XX)with 52% yield. The second route (Fig. 11) proposed by Golankiewicz and Folkman (156) was based on the observation that 2’,3‘,5’-tri-O-acetyldesmethylwyosine (LXXIII), obtained in the reaction of guanosine with bromoacetone followed by 0-acetylation, can be methylated


51

tRNA

0

(LXIX 1

i

( LXXI

1

( LXXII

I

(XX)

FIG. 11. Synthesis of wyosine (XX) via 3-methylguanosine (152, 154, 155) or by direct methylation of the tricyclic ribonucleoside (LXXIII) (156).

with diazomethane at the desired nitrogen atom leading to O-acetylated wyosine with a 3%overall yield. Wyosine was obtained by the subsequent deacetylation of the product with methanolic ammonia. The efficiency of this method is comparable to that described above and reported by Nakatsuka et al. (152).The methylation procedure was also applied for the synthesis of wyosine 5'-phosphate from the analogous N-desmethyl intermediate bearing a 5'-bis-(2-~yanoethyl)phosphate group (157). 2. CHEMISTRY Incubation of yeast tRNAPhcat pH 2.9 leads to the excision of wybutine (LXII) leaving the RNA chain uncleaved. (158).Analogous conditions have been used for the excision of wyeine (LXIII) (148, 150). The following half-life times of the N-glycosyl bond hydrolysis in dilute hydrochloric acid at 37°C have been reported recently for synthetic wyosine (XX)(155):0.1 N HCl, 42 seconds; pH 2.1, 7 minutes; pH 2.9,41 minutes [55 minutes was reported in (256)I.l Wyosine is stable in the pH range of 7.0-8.5, but decomposes rapidly to unknown products in 0.1 N NaOH (half-life of 63 minutes at 37"C).' Thiebe et al. (159)studied chemical transformations of the side chain of wybutine (LXII) isolated from unfractionated yeast tRNA under

52

RYSZARD

R-CHNHR'

R = -COOCH3

I

-COOH

w. ADAMIAK

AND PIOTR C ~ R N I C K I

R'= -COOCH3

I LXII I

-COOCH3

(LXXIV I

, CH3

conditions leaving the fluorescent aromatic core intact. Treatment with NaOH (0.1 N, 37°C) led to the complete hydrolysis ofthe ester to LXXIV, and under more drastic conditions (1 N NaOH, 50OC) to LXXV. In 25% ammonia at 37"C, both hydrolysis to LXXIV and aminolysis to LXXVI took place. The specific acid-promoted excision of the tricyclic bases from tRNAPhehas been employed in many studies. 1. The functional and physicochemical properties of tRNA with an excised base were compared with those of native tRNA (160-162). 2. The weakly fluorescent natural base was replaced with better fluorophores such as proflavine, aminoanthracene, or ethidium by Schiff base formation between the amino group of the reagents and the ribose aldehyde group formed after depurination (163-166). The insertion of an artificial fluorescent probe was shown not to impair the activities of tRNAPhe(164,167-169). Fluorescence of both the natural (170-1 77) and artificial (166,173,176,178-184) probes has been used in structural and functional studies on tRNA. 3. Depurination followed by the specific aniline hydrochloride cleavage of the tRNA chain leads to tRNA half-molecules that, in the multistep enzymatic procedure developed by Bruce and Uhlenbeck (46),can be altered and ligated together yielding chimeric tRNA molecules in which the tetranucleotide of the anticodon loop of tRNAPhe (positions 34-37) is replaced by an oligonucleotide of a preprogrammed length and sequence. The photochemical properties of synthetic wyeine have been described (185). It appears that UV irradiation of the compound in an oxygen-saturated solution leads to the cleavage of the tricyclic ring structure and a decrease in fluorescence. Two products, LXIV and LXXVII, identified in the reaction mixture, were proposed to be formed via oxygen photo adducts (Fig. 12). Crosslinking of tRNAPhe to Phe-tRNA ligase (186) and crosslinking of the A-site-bound tRNAPheto poly(U) in E . coli ribosomes [poly(U)programmed] by UV irradiation (over 300 nm) can be explained by the photoreactivity of wybutine and wyeine, but their

HYPERMODlFIED NUCLEOSIDES OF

53

tRNA

CH3

CH3

CH3 ( LXXVII)

FIG.12. Photochemical transfonnations of wyeine (LXII) (185).

involvement in crosslink formation was not proved unequivocally (187). The presence of the unusually stable hydroxyperoxy group in wybutoxosine is very interesting from both chemical and biochemical points of view. It can be detected by a microcolor test sensitive to peroxides employing ferrous isothiocyanate (25).

111. Structural Features of Hypermodified Nucleosides and Codon-Anticodon Interaction In the model proposed by Fuller and Hodgson (188), the codonanticodon recognition is realized via minihelix formation between mRNA and tRNA in which the anticodon loop is in its 3’-stacked conformation (Fig. 13) closely resembling that found in the crystal structure of yeast tRNAPhe(189-194). The 3‘-stacked conformation is the general feature of tRNA both in crystal (195-197) and in solution (198). In this context the location of hypermodified nucleosides, and a rather strong sequence correlation (Table I) suggest their involvement in the recognition process, presumably because of their inter- and intrastrand interactions. The real picture is certainly much more complex d u e to simultaneous action of many elements of the protein biosynthetic apparatus. In this section, we discuss the unique features of hypermodified nucleosides that seem to play a significant role in high fidelity codon-

54

RYSZARD W. ADAMIAK AND PIOTR C6RNICKI

FIG.13. Fuller-Hodgson model of codon-anticodon interaction (188).

anticodon interaction. Crystal structure analysis (199), supported in some cases by NMR solution studies, is the prime source of detailed information concerning the conformation of hypermodified nucleosides, their ability for hydrogen bonding, stacking and cation binding. Some CD, UV, fluorescence, and NMR studies on short oligoribonucleotides have also been reported. It is believed that some of the structural features observed at the nucleoside level not only resist the strain but in fact are to some extent responsible for the tRNA architecture. Although clear-cut evidence is still missing, some predictions about such structural phenomena can be made.

A. Unique Features Imposed by Hypermodification 1. “WOBBLE”URIDINES Extensive studies have been performed to correlate the molecular structure of hypermodified uridines with their restricted or extended wobble properties. The most important structural information obtained from X-ray crystallographic analysis is presented in Table 11. 3’-Endo puckering of the ribose ring (except for cm5U)and anti-orientation around the glycosyl bond are the common features, as is the intramolecular interaction C(6)-H ...O(5’)stabilizing the anti confor-


55

tKNA

TABLE I 1 SELECTED STRUCTURAL FEATURES OF HYPEHMOI)IFIED URIIIINES (FROM X-RAY CRYSTALLOGRAPHIC STUDIES)

Sugar pucker

Glycosvlic dihedral angle x (")

Base-side chain orientation

Relerences

nicmo"U (111)

3'-endo 3'-endo 3'-endo

2 5 4 anti 23.1 anti 34.3 anti

C(S)-0-C type Coplanar Coplanar Coplanar

202 20.5 21J.3, 204

inam"s"U (IVY cm5U ncn9U cm5s2U mcm"s"U (VIII)

3'-endo 2'-endo 3'-endo 3'-endo 3'-endo

15.0 anti 49.4 anti 5.2 anti 3 anti 16 anti

C(5)-C-C type Skew Skew Skew Skew Skew

206 208 208 209 209

Nucleoside

m0"u (I)

Studied also as the acetonide (207).

mation and gauche-gauche conformation of the ribose exocyclic group found for some of the nucleosides studied (mo5U, mcm5s2U, mam5s2U). T h e glycosyl dihedral angle for the thionucleosides is within the range of 0-20", i.e., typical for 2-thiopyrimidine nucleosides (200, 201). The hypermodified uridines can be classified into two groups according to the conformation of the side-chain, which can be approximately coplanar with the pyrimidine ring due to the conjugation effect [C(5)-0-C type side-chain] or skew to the pyrimidine ring [C(5)-CC type side-chain]. It was suggested on the basis of the bond-length analysis that the presence of the side-chain of mcmo5U (203,204)(Fig. 14) and mo5U(202) changes hydrogen-bond accepting ability or shifts keto-enol equilibria toward the enol form of 2-keto and 4-keto groups, respectively. This effect was originally proposed to be responsible for

R

FIG.14. Base pairing patterns proposed for mcmo5U (111) (R' (203).

=

OCH2COOCH3)

56

RYSZARD W. ADAMIAK A N D PIOTR G6RNICKI

the wobble properties of the nucleosides. However, some recent more accurate measurements have shown that the differences in bond length, at least between mo5U and uridine, are insignificant and do not justify such conclusion (205).In addition, in view of the more general discussion of Topal and Fresco (210),base-pairing of nucleosides in their rare tautomeric forms seems unlikely; for the common nucleosides it was precluded. C(5)-substituents, especially those with strong electron-withdrawing properties, can alter not only the electron distribution in the pyrimidine ring but also the conformation of the ribose moiety (211). Thus they could be considered as potential effectors of codon-anticodon interaction that do not interfere directly with base pairing. However, NMR studies of mo5U (I) (205,211)showed that its conformation is very similar (almost identical) to that of uridine, and that the 3'endo(gg)anti conformer, found in the solid state, prevails in solution. Substituents of the other type can be expected to have even less effect; preliminary NMR analysis showed that the syn-anti equilibrium is not affected by C(5)-substituents such as in mcm5U, cm5U, and ncm5U (212, 213). However, CD studies suggest that the syn-anti equilibrium in mcm5U is more restricted than in uridine (213a). Substitution of oxygen at the 2-position of the pyrimidine ring b y sulfur has important consequences. The reduced ability of the sulfur atom to hydrogen bond was suspected to be responsible for the wobble restriction of 2-thiouridines by reducing their ability for hydrogen bonding with the N(1)-H of guanosine (71, 214). However, further studies revealed that the C(5)-substituent rather than the thioketo group is responsible for the effect. (72, 77).The hydrogen bond with sulfur as the acceptor was observed in the secondary structure of polynucleotides containing 2-thiopyrimidine nucleosides (215,216).A hydrogen bond involving sulfur is 0.4-0.8 A longer than its oxygen counterpart (227). Lengthening of the hydrogen bond can weaken the other hydrogen bonds typical for uridine base-pairs. Polynucleotides containing 2-thiopyrimidine nucleosides form very stable secondary structures (215,217).The effect was originally ascribed to the strong stacking interaction between the sulfur S(2)and nitrogen N ( l ) of the adjacent base (218).It was not observed for 4-thioanalogs (217 ) .Conformational analysis of mam5s2U(IV) and mcm5s2U (VIII) based on systematic 'H-NMR studies led Yokoyama et al. (219) to a different hypothesis. The sulfur atom at the 2-position (but not at the 4-position) of the pyrimidine ring was responsible for the shift of the conformational equilibrium toward the 3'-endo(gg)anti conformer-the same found in the crystal structure (Table 11). The same


57

tHNA

TABLE I11 STRUCTURAL FEATURES OF THE SIDE-CHAINS OF HYPERMODIFIED ADENOSINES (FROM X-RAY CRYSTALLOGRAPHIC STUDIES) Base or nucleoside i6AAde

ms"ihAde

PA& (K+ or Rb'salt) ghAde (K' salt)

ghAde (dihydrochloride)

P A (XV).

ghA (XVIII)'l

I'

'I

Side chain orientation Trans to iniidazole ring, isopentenyl chain planar and skew to the base plane Trans to imidazolc ring, isopentenyl chain planar and perpendicular to the base plane Trans to iniidazole ring, ureidopurine system coplanar Trans to iniidazole ring, nreidopnrine system coplanar Trans to i 111i dazo1e ring, ureidopurinr system coplanar Trans to iniidazole ring, ureidopurine system coplanar Trans to iniidazolt. ring, ureidopurine system coplanar

Nucleoside conformation: 2'-endo, anti Nucleoside conformation: 3'-endo, anti

Intramolecular hydrogen bonding

References

220

221.222

N ( l )...H-N (threonine) bifurcated to -OH of threonine moiety N(1)...H-N (glycine)

224, 225

N( I)-H ...O= (ureido)

226

N( 1)...H-N(threonine) bifurcated to -OH of threonine moiety N(1)...H-N(glycine) bifurcated to C=O of glycine moiety

223

227,228

227.228

( x = :33.1°). ( x = 4.0").

conclusion was drawn from CD studies on mcm5s3U ( 2 1 3 ~ A ) . 3'-endo(gg)anti conformation is known to stabilize the A structure of RNA and to enhance the base-stacking interactions. Thus the fixation of the conformation observed for the nucleoside unit can introduce some rigidity to the structure of the wobble region of the anticodon loop. Steric effects that may explain different wobble properties of the hypermodified uridines are discussed in Section III,B,3.

2. HYPERMODIFIED ADENOSINES ADJACENTTO THE ANTICODON3'-END Most of the structural features of hypermodified adenosines and related purines are revealed by X-ray analysis (Table 111). For both

58

RYSZARD

w. ADAMIAK


FIG.15. Molecular structure of ifiAde-top right (220) and FA (XV)-bottom left (228).

types of hypermodification (i6- and t6-), the N(6)-substituent is trans (distal) to the imidazole ring. Intramolecular hydrogen bonding involving purine N ( l ) and threonine N(a)-H (-2.7 A) is typical for ureido-purines and related nucleosides (223,224,226-228). Both features are presented in Fig. 15. The ureido system is nearly planar with the purine ring due to the extensive eIectron delocalization as revealed by the bond-length analysis. The six-membered ring structure formed by the hydrogen bond, further bifurcated to the oxygen atom of the threonine hydroxyl group (PA) or the glycine carboxyl group ($A), contributes to the rigid conformation of the ureidonucleoside side-chain. The presence of the intramolecular hydrogen bond in solution is also suggested by 'H NMR (122,141)and CD (122) studies. The conformation of N(6)-substituted adenosines if present at the tRNA level would have important functional implications. The sidechain would block both the N(6)-H and N ( l ) sites of adenine, which are involved in the formation of Watson-Crick base pairs (but still allowing Hoogsteen or reverse Hoogsteen base pairing), setting the correct frame for the reading process. Possible interactions of the @A side chain are discussed in Section III,B,3. It must be pointed out that in both ureidonucleoside structures discussed above, the carboxyl group was un-ionized, which is not too likely to be the situation in tRNA under physiological conditions. The ionized carboxyl group of hypermodified ureidonucleosides is the potential ligand for cation binding. Preliminary experiments using


tRNA

59

FIG.16. Proposed structure of pPA.Mn2+( 1 : 1) complex (231).

13C NMR (229)and potentionietric titration (230)indicate that the carboxyl group, N(6) and N ( l ) or N(7) sites are involved in Mg2+ and paramagnetic Mn2+ binding ( Mn2+ properties mimic those of Mg2+). Further studies (231, 232) extended to t6A 5’-phosphate and m@A (XVI) showed a stronger Mn2+binding to the mononucleotide and a weaker one to the N(6)-methyl analog. Binding of Mn2+(Mg2+)to pPA was 5 orders of magnitude stronger than to 5’-AMP (231).A model of the ptGA.Mn2+complex (1: 1)has been proposed (Fig. 16). It is interesting to note that the rigid conformation of the ureido group discussed above and observed also in crystal structures of monovalent cation salts (see Table 111) is disrupted. The lower stability of the mt6A-Mn2+complex was explained by the elimination of one metal coordination site N(6)-H (pK, 8.55) by the methylation. These observations are interesting in view of the discussion on the mode of Mg2+ binding to the anticodon loop (see Section 111,B). PA shows the preference for the 2’-endo pucker of the ribose ring. It exists in the crystal structure (227,228)and it also prevails in solution as indicated by IH NMR measurements (231).It is in contrast to the @A molecule for which the 3’-endo pucker was found in the crystal state (228).

B.

Hypermodified Nucleosides within the Anticodon Loop Architecture

ANTICODONLOOPFRAGMENTS Our knowledge of conformation and the interactions of hypermodified adenosines at the oligonucleotide level is very limited. Dinu1. STUDIESON

THE

60

RYSZARD

w. ADAMIAK


cleoside monophosphates such as A-ifiA,A-ms2i6A,and U-FA, typical sequences in which these nucleosides occur in tRNA (Table I), have been studied (233) by means of UV, CD, and ‘H NMR techniques. The two types of hypermodification (i6-and tb-) had opposite effects on stacking interactions which for A-ifiA and A-ms2i6A were slightly weaker and for U-PA was stronger than in the unmodified counterparts. It was suggested that hypermodification in tRNA could adjust stacking interactions between the last anticodon nucleoside and 3’neighbor in such a way that the difference in stacking observed for U-A and A-A was largely diminished. In fact, the strength of the stacking interaction was very similar for U-t6A, A-PA, C-A, and A-EA (EA is 1,W-ethenoadenosine, used as a model for wyosine). Thus the role of hypermodification could be keeping the flexibility in this region of the anticodon loop constant, independent of its sequence. It was also suggested that enhanced interaction between U36 and FA37 could prevent a 3’-wobble, i.e., unprecise reading of the first letter of the codon (233). NMR studies indicate that side-chains of the hypermodified nucleosides are not folded between base residues but rather protrude into solution. There was no Mg2+binding to FA in U-PA detected by CD measurements. It contradicts the results discussed in Section III,A,2. Oligoribonucleotides containing wyosine and wybutosine obtained by excision from T. utilis and yeast tRNAPhe,respectively, have been investigated more extensively. Fluorescence, NMR, UV, and C D studies on Gm-A-A-yW-A-Vp, Gm-A-A--A-qp (oligonucleotide with wybutine excised by acid treatment, Section II,C,2) and related shorter oligoribonucleotides containing wybutosine (XXI) (234, 235, 236) as well as NMR studies of Gm-A-A-W-A-V (237) revealed that the tricyclic nucleosides are in a unique way involved in the strongly stacked hexanucleotide conformation. The properties of the hexanucleotide containing wybutosine (yW) suggested that its structure is relatively stable to thermal perturbation and in many respects resembles that of the respective fragment of the anticodon loop in tRNAPhe,that yW interacts stronger with its 3’neighbor and forms a “zig-zag” stacking pattern within the A-yW-A fragment. It was then proposed that the role of this hypermodified nucleoside is to disrupt the regularity of stacking between nucleosides in position 36 and 37 without disturbing the anticodon triplet stacked conformation and thus to hold the anticodon in an exposed position proper for codon-anticodon interaction. NMR studies of the hexanucleotide containing wyosine (W) led to a quite different conclusion. Increased temperature caused a disrup-


tRNA

61

k tion of the stacked oligonucleotide structure with the wyosine residue being the “focal point” (237) of the conformational change. The base residue underwent destacking and the conformational equilibrium of the ribose residue was shifted toward the more flexible 2’-endo conformation. Two adenosines next to wyosine and located originally in the anticodon were relatively less affected. From these studies it was suggested that in general the hypermodified nucleoside adjacent to the 3’-end of the anticodon might serve as a “hinge point” in the anticodon loop structure (237). The authors found this interpretation especially attractive in view of Lake’s hypothesis (238) of functional changes in the anticodon loop conformation. It is interesting to point out that the two conclusions were drawn from studies of oligoribonucleotides that differ only by the presence of the side chain of yW and the 3’-terminal phosphate group. IN tRNAPheANTICODON 2. WYBUTOSINE LOOPSTRUCTURE

The high resolution analysis of the crystal structure of yeast tRNAPhesheds some light on the involvement of wybutosine in the anticodon loop architecture. The refined structure of this tRNA (239) reveals that the 3’-stack with the U-turn at its 5’-end, the central feature of the anticodon loop structure, held by the interaction of uridine 33 with phosphate 36 [N(3)-H ...O-P hydrogen bond] and with phosphate 35 [base-phosphate stacking] is further stabilized by the lockedin hydrated magnesium ion coordinated directly to phosphate 37 and indirectly (via the water molecules of its hydration shell) to q 3 9 , A38, yW37, Cm32, and Ad1 (Fig. 17).The base residue of yW37 is well stacked between adjacent adenine residues within the 3’-stack and its sidechain does not interact with other parts of the anticodon loop. An analogous arrangement was deduced from ‘H NMR studies in solution: the side chain was in essentially the same environment in the

62


FIG. 17. The model of the anticodon loop structure based on the refined yeast tRNAPhe X-ray analysis. Reproduced with permission from (239).

free base as in tRNAPhe(240,241 ). The fluorescence life-time of yeast tRNAPhein crystals and in solution also indicated a similar wybutine environment in both states (242).The structure of the tRNA molecule can be expected to contribute to the fixation of the anticodon loop conformation; however the main features of the structure around the hypermodified nucleoside seem to exist even on the hexanucleotide level (Section III,B,l). On the other hand, excision of wybutine from tRNA (Section II,C,2) results in structural changes, some of which, due to long range effects, can be transmitted to distal parts of the molecule (160) and induce functional changes, including codon recognition. In this context, the sensitivity of wybutine fluorescence to tRNA conformational changes induced by temperature jump, magnesium, or

HYPERMODIFIED NUCLEOSIUES OF

tHNA

63

codon binding is not surprising. Various fluorescent techniques based on the intrinsic properties of wybutine (243-245) or artificial fluorophores replacing it (163, 165) and allowing kinetic measurements of fast transitions as well as NMR spectroscopy (246-248), have been employed in the investigations of conformational changes of the anticodon loop. Such changes were crucial elements of several hypotheses concerning various aspects of the mechanism of protein biosynthesis (188, 238, 249, 250). The observed changes in wybutine fluorescence have been ascribed either to the conformational transition between 3'- and 5'stacked conformers or between multiple conformations of the anticodon loop, with magnesium ion selecting a specific one (179,251-255). The fluorescence studies, which potentially could give information about both the environment of wybutosine and its involvement in such anticodon loop conformational changes, have not found unequivocal interpretation so far. Urbanke and Maass (251)measured temperature-induced quenching of wybutine fluorescence by the temperature-jump technique in yeast tRNAPh'. They concluded that one of the observed relaxation processes reflected the transition from the 3'- to the 5'- stacked conformer, with the 3'-stacked conformer resembling that found in the crystal structure and the 5'-stacked conformer exposing the fluorescent base and allowing its more extensive interaction with the solvent. On the contrary, Wells (254) concluded that, at low Mg2+ concentration, the anticodon loop is in a rather flexible conformation in which wybutine is tilted or twisted with respect to its neighbors and thus is more accessible to the solvent, whereas at high Mg2+ concentration, bases are lined-up i n parallel in a rigid conformation. 'H NMR data (241) seem to favor this interpretation (254). Studies on tRNAPheand oligoribonucleotides show that there is some conformational freedom within the wybutosine (wybutine)unit. It remains to be seen whether this effect, detected by fluorescence, which results only from specific properties of the tricyclic nucleosides, is characteristic for all hypermodified nucleosides found in this position, reflecting their function, or is a general feature of the anticodon loop architecture.

3. HYPERMODIFIED NUCLEOSIDES IN tRNAPhe ANTICODONLOOPSTRUCTURE: SIMULATION STUDIES The preceding sections deal with direct effects of hypermodification altering nucleoside base-pairing and stacking properties as a result of inherent electronic and conformational changes. Even a brief

64

RYSZAFID

w. ADAMIAK AND

PIOTR G ~ R N I C K I

inspection of the anticodon loop structure of tRNAPhesuggests the possibility of some indirect effects resulting from interactions with the anticodon loop backbone, such as steric repulsion of bulk substituents, electrostatic repulsion of ionized groups (carboxyl and phosphodiester), hydrogen bonding and Mg2+binding. Lack of accurate information about the tertiary structures of different tRNAs makes evaluation of such indirect effects rather speculative; some information can be gleaned from simulation studies in which hypermodified nucleosides in their crystal structures (Tables I1 and 111) are “builtinto” the yeast tRNAPheanticodon loop at the positions of GmJI(206, 208,209) and of yW37 (228),with the assumption that at least the main structural features are preserved in all tRNAs. It must be pointed out that structural information available for hypermodified uridines do not explain per se their wobble properties in view of the geometrical rules of the Crick hypothesis (256).Therefore, Hillen et al. performed simulation studies for mam5s2U (IV) (206),mcm5s2U(VIII) (209), and mo5U (I) (206).They postulated that interactions of bulky side-chains skew to the base plane can restrict the flexibility of the anticodon loop and make the distortion required for U.G base-pairing impossible, thus leading to the restricted wobble recognition observed for this group of hypennodified uridines [C(5)C-C type side chain]. In addition, this type of interaction can prevent stacking between wobble uridine and its 5‘-neighbor. Such “helixterminator” (209)effects could be especially important in the case of 2-thiouridine analog showing enchanced stacking ability. On the contrary, mo5U [C(5)-C-C type of side chain], which shows amplified wobble recognition, seems to retain its full flexibility. The presence of the magnesium binding site within the anticodon loop of yeast tRNAPhe(Fig. 17), on one hand, and the ability of t6A for magnesium binding (Section III,A,2) on the other, suggest that one of the functions of the carboxyl group of t6A and other hypermodified nucleosides could be cation binding. In view of the influence of the @A carboxyl group ionization and Mg2+binding on its side chain conformation, the simulation experiments performed for PA -carboxylic acid need reconsideration. The possible Mg2+ binding to the base residue of i6A (257) is controversial (233).

C. Involvement in Codon-Anticodon Interaction The studies of hypermodified nucleosides carried out at different levels of structural complexity and undertaken with the aim of exploring their functions has led to several proposals explaining observed


tRNA

65

sequence correlations (Table I), tRNA properties in the classical in vitro ribosome binding, and in vivo suppression tests, or simply correlating some unique features with the general model of the decoding process. The bioorganic approach also offers simple systems for studies based on ribosome-free association of tRNA with complementary oligoribonucleotides (anticodon-codon association) (2.58-261) or with another tRNA with complementary anticodon (anticodon-anticodon association) (12, 262, 263) measured by equilibrium dialysis (258-261, 264), fluorescence and UV-detected temperature jump relaxation method (172, 255, 262, 263, 265) or NMR spectroscopy (266-268). Two tRNA molecules associating via their complementary anticodons form complexes whose structures mimic the tRNA-codon interaction of the type proposed originally by Fuller and Hodgson (Fig. 13). They are six orders of magnitude more stable than the complex of the corresponding free trinucleotides (262, 263). On the other hand, temperature-jump experiments carried out for over 60 pairs of tRNA with complementary or partially complementary anticodons revealed that the life-time of the complementary complex is similar irrespective of the number of A.U and G C base-pairs involved (263). These observations show that the problem of codon-anticodon interaction must be discussed in terms of its thermodynamics, as proposed by Ninio (269,270). It seems that the inherent properties of the anticodon loop architecture are the main source of stabilization energy involved in codon-anticodon interaction. However, contributions of the other major participants in polypeptide synthesis (ribosomes and mRNA), suggested for example by observations on the codon context effect (272-274), cannot be neglected. How d o hypermodified nucleosides, as building units of the tRNA anticodon loop, contribute to the overall energy of this interaction? The results of the anticodon-anticodon association experiments (263) indicate that they tend to stabilize the A*Uor U.A base-pairs formed, bringing their energy level closer to that of the G C base-pair. In this respect, two sets of more direct observations concerning yW and t6A are of interest. It has been found (262) that yeast tRNAPhe(anticodon Gm-A-A) containing wybutosine at position 37 associates with yeast tRNAE'" (anticodon mam5s2U-U-C)with a AH for the complex formation of -24 kcal M-'. The strength of the complex is largely decreased (AH = - 17 kcal M-') by excision of wybutine. The association of tRNAPhefrom Mycoplasma containing m'G instead of yW with yeast tRNA,G1llis also weaker (AH = -20 kcal M-').

66

RYSZARD

w. ADAMIAK AND PIOTR G ~ R N I C K I

Equilibrium dialysis shows that yeast initiator tRNAMet(anticodon CAU) containing PA binds AUG and, unexpectedly, GUG triplets twice as efficiently as E. coli initiator tRNAMetlacking the hypermodification (275). Studies of the stabilization effect of PA on the formation of A*U and G-U (noncomplementary) with the application of two species of yeast tRNA,,i\’g (anticodon mcm5UCU)-one containing FA and second containing unmodified A-indicate that tRNA,,f’g (PA) binds the AGA triplet with an association constant twice that of tRNA,,i\’g (A) (276). Both species of tRNA,,;Q’g associate also with E. coli tRNAv (anticodon GGA). The presence of PA stabilizes the complex involving the G-U pair by a factor of four. However, such a pair is still one-tenth as stable as the G C pair formed during association ofE. coli tRNAg’y (anticodon U’CC) and E. coli tRNAper.The results on the stabilization of the G.U pair (275,276) contradict the proposition that one of the roles of FA in eukaryotic initiator tRNA is to prevent mispairing in the first position of the codon (277) (see Section III,B,l). Stabilization of the codon-anticodon interaction by hypermodified purines results most likely from enhanced stacking (11,12,263,276). It is well-recognized that nonpaired bases at the end of the RNA double-helix have a stabilizing effect (278-281 ), often called the “dangling-base” effect, but the origin of the stacking forces involved is not fully understood. In one view of the involvement of hypermodified nucleosides, the observation that the dangling 3’-adenosine might contribute as much as 65-90% of the energy of a full base-pair, compared to 30-70% for G and U, and 10-20% for C (280), is of interest. In the case of anticodon-anticodon interaction, this effect can be amplified by the ordered structure of the “dangling ends” and modulated by nucleoside modification (12). Experiments concerning yW and t6A indicate that their contribution to the stabilization energy of the codon-anticodon complex is rather small, but pronounced enough to have the “tuning” effect on this process (12,276). The involvement of hypermodified uridines in the anticodon-anticodon association is difficult to evaluate directly due to the lack of appropriate models. Some speculations on the implications of their steric and stacking properties have been given in Sections III,A,l and 111,B,3. Studies of codon-anticodon association and the ribosome binding, together with comparisons of tRNA sequences, have led to the proposal (11) of an evolutionary model of codonanticodon recognition in which hypermodified nucleosides play a key role. Weissenbach and Grosjean (276) concluded that the role of hypermodified nucleosides is “to adjust the association energy of a comple-


tRNA

67

mentary (or nearly complementary) codon-anticodon pair so as to bring the residue time of all tRNAs in the decoding site of the mRNAribosome-peptidyl tRNA complex within an optimal range of time, irrespective of codon composition, thus allowing a smooth in-phase translation.”

ACKNOWLEDGMENTS We wish to thank M. Wiewi6rowski for interest and encduragement, and A.R. Trim for information about the studies of the late D. Dunn. We also thank E. Pasternak and J. Bartoszewicz for the art work. This work was supported by grants from Polish Academy of Sciences.

REFERENCES 1. D. H. Gauss and M. Sprinzl, NARes 12, r 1 (1984). 2. D. B. Dunn and R. H. Hall, in “Handbook of Biochemistry and Molecular Biology,” 3rd ed., Nucleic Acids, Vol. 1(G. Fasman, ed.), p. 65. CRC Press, Cleveland, ( 1975). 3. R. H. Hall, “The Modified Nucleosides in Nucleic Acids.” Columbia University Press, New York, (1971). 4 . S. Nishimura, This series 12,49 (1972). 5. J. A. McCloskey and S. Nishimura, Ace. Chern. Res. 10,403 (1977). 6. C. W. Gehrke, K. C. Kuo and R. W. Zumwalt, in “The Modified Nucleosides of Transfer RNA 11” (P. F. Agris and R. A. Kopper, eds.), p. 59. Liss, New York, (1983). 7. J. P. Goddard, Prog. Biophys. M o l . B i d . 32, 233 (1977). 8. G. Dirheimer, G. Keith, A. P. Sibler and R. P. Martin, in “Transfer RNA: Structure, Properties and Recognition” (P. R. Schimmel, D. Sol1 and J. N. Abelson, eds.), p. 19. Cold Spring Harbor Laboratory, New York, (1979). 9. R. P. Singhal and P. A. M. Fallis, This series 23,227 (1979). See also R. P. Singhal, E. F. Roberts and V. N. Vakharia, This series 28,211 (1983). 10. M. Y. Feldman, Prog. Biophys. Mol. B i d 32, 83 (1977). 11. R. J. Cedergren, D. Sankoff, B. LaRue and H. Grosjean, CRC Crit. Rel;. Biochem. 9, 35 (1981). 12. H. Crosjean and H. Chantrenne, Mol. B i d . Biochem. Biophys. 32,348 (1980). 13. S. Nishimura, in “Transfer RNA: Structure, Properties and Recognition” (P. R. Schimmel, D. So11 and J. N. Abelson, rds.), p. 59. Cold Spring Harbor Laboratory, New York, (1979). 14. P. F. Agris and D. So11, in “Nucleic Acids and Protein Recognition” (P. Vogel, ed.), p. 321. Academic Press, New York, (1977). 15. S. Nishimura, This series 28,50 (1983). 16. R. P. Singhal, This series 28, 75 (1983). 17. H. Kersten, This series 31, 59 (1984). 18. W. R. Farkas, Nucleosides Nucleotides 2, 1 (1983). 18a. D. Crunberger, R. G. Pergolizzi, Y. Kuchino, J. F. Muchinsky and S. Nishimura, Recent Res. Cuncer Res. 84, 133 (1983). 19. P. F. Agris, “The Modified Nucleosides of Transfer RNA: A Bibliography of Biochemical and Biophysical Studies from 1970 to 1979.” Liss, New York, (1980).

68

RYSZARD

w. ADAMIAK AND PIOTR

G~RNICKI

20. B. F. C. Clark, in “Nonsense Mutations and tRNA Suppresors (J. E. Cellis and J. D. Smith eds.), p. 1. Academic Press, London, (1979). 21. J. Ofengand, in “Molecular Mechanism of Protein Biosynthesis” (H. Weissbach and S. Pestka, eds.), p. 8. Academic Press, New York, (1979). 22. A. Rich and U. L. RajBhandary, ARB 45, 805 (1976). 23. K. Nakanishi, S. Blobstein, M. Funamizu, N. Furutach, G. van Leer, D. Grunberger, K. W. Lanks and I. B. Weinstein, Nature 234, 107 (1971). 24. S. Blobstein, D. Grunberger, I. B. Weinstein and K. Nakanishi, Bchem 12, 188 (1973). 25. A. M. Feinberg, K. Nakanishi, J. Barciszewski, A. J. Rafalski, H. Augustyniak and M. Wiewibrowski, JACS 96,7799 (1974). 26. M. Barciszewska, M. Kaminek, J. Barciszewski and M. Wiewibrowski, Plant Sci. Cell. 20,387 (1981). 27. H. Kasai, Z. Yamaizami, Y. Kuchino and S. Nishimura, NARes 6,992 (1979). 28. C . C. HsuChen, G . R. Cleaves and D. T. Dubin, Plasmid 10,55 (1983). 29. R. H. Hall, L. Csonka, H. David and B. McLennon, Science 156,69 (1967). 30. W. J. Burrobs, D. J. Armstrong, M. Kaminek, F. Skoog, R. M. Bock, S. M. Hecht, L. G. Dammann, N. J. Leonard and J. Occolowitz, Bchem 9, 1867 (1970). 31. H. J . Vreman, F. Skoog, C. R. Frihart and N. J. Leonard, Plant Physiol. 49, 848 (1972). 32. W. H. Dyssen and R. H. Hall, Plant Physiol. 50, 616 (1972). 33. H. J. Vreman, R. Y. Schmitz, F. Skoog, A. J. Playtig, C. R. Frihartand N. J. Leonard, Phytochemistry 13,31 (1974). 34. F. Skoog and D. Armstrong, Annu. Rev. Plant. Physiol. 21, 359 (1970). 35. M. P. Schweitzer, K. McGrath and L. Baczynskyj, BBRC 40, 1046 (1970). 36. S. Takemura, H. Kasai and M. Goto,]. Biochem. 75, 1163 (1974). 38. M. W. Gray and B. G. Lane, Bchem 7,3441 (1968). 39. D. B. Dunn and M. D. M. Trigg, Biochem. Soc. Trans. 3,656 (1975). 40. D. B. Dunn and M. D. M. Trigg.John Znnes Inst. Rep. 11 (1975). 41. M. W. Gray, Bchem 15,3046 (1976). 42. M. W. Gray, Can. J . Biochem 53, 735 (1975). 43. K. Baczynskyj, K. Biemann and R. H. Hall, Science 159, 1481 (1968). 44. P. Bronskil, T. D. Kennedy and B. G. Lane, BBA 262,275 (1972). 45. T. D. Kennedy and B. G. Lane, Can. J. Biochem 53, 1 (1975). 45a. H. Kasai, K. Murao, S. Nishimura, J. G. Liehr, P. F. Crin and J. A. McCloskey, EJB 69,435 (1976). 4%. M. Hartmanis and T. C. Stadtman, PNAS 79,4912 (1982). 4%. C. Chen and T. C. Stadtman, PNAS 77, 1403 (1980). 45d. W.-M. Ching and T. C. Stadtman, PNAS 79,374 (1982). 45e. A. J. Wittwer, JBC 258,8637 (1983). 45f. W.-M. Ching, A. J. Wittwer, L. Tsai and T. C. Stadtman, PNAS 81, 57 (1984). 45g. W.-M. Ching, PNAS 81, 3010 (1984). 45h. A. J. Wittwer, L. Tsai and T. C. Stadtman, Bchem 23,4650 (1984). 46. A. G. Bruce and 0. C. Uhlenbeck, Bchem 21,855 (1982). 47. L. H. Schulman, H. Pelka and M. Susani, NARes 11, 1439 (1983). 48. P. Carbon, E. Haumont, S. De Hanau, G. Keith and H. Grosjean, NARes 10,3715 (1982). 49. M. Founier, E. Haumont, S. De Hanau, J. Gangloff and H. Grosjean, NARes 11,707 (1983). 50. P. Carbon, E. Haumont, M. Founier, S. D e Hanau and H. Grosjean, EMBOJ. 2, 1093 (1983).


tRNA

69

T. Ueda, Chem. Pharm. Bull (Tokyo) 5,455 (1960). K. Murao, T. Hasegawa and H. Ishikura, NARes 3,2852 (1976). K. Murao, M. Saneyoshi, F. Harada and S. Nishimura, BBRC 38,657 (1970). M. G. Stout and R. K. Robins, J . Heterocycl. Chem 9,544 (1972). L. Baczynskyj, K. Biemann, M. H. Fleysher and R. H. Hall, Can. J . Biochem. 47, 1202 (1969). 56. I. D. Fissekis and F. Sweet, Bchem 9, 3136 (1970). 57. G. A. Ivanovics, R. J. Rousseau and R. K. Robins, Physiol. Chem. Phys. 3, 489 (1971). 58. H. Vorbriiggen and K. Krblikiewicz, Angew. Chem. Int. Ed. 14, 255 (1975). 59. H. Vorbriiggen and K. Kdikiewicz, Liebigs Ann. Chem. 9, 1438 (1980). 60. A. Malkiewicz and E. Sochacka, Tetrahedron Lett. 24, 5387 (1983). 61. K. Ikeda, S. Tanaka and Y. Mizuno, Chem. Pharm. BulE. (Tokyo)23,2958 (1975). 62. A. Malkiewicz, E. Sochacka, A. F. SayedAhmed and S. Yassin, Tetrahedron Lett. 24, 5395 (1983). 63. C. B. Reese and Y. S. Sanghvi, J . Chem. S O C . Chem. Commun. 1,62 (1984). 64. A. Malkiewicz and E. Sochacka, Tetrahedron Lett. 24,5391 (1983). 64a. Z. Ohashi, M. Maeda, J. A. McCloskey and S. Nishimura, Bchem 13, 2620 (1974). 65. Y. S. Rao and I. D. Cherayil, Biochem J . 143, 285 (1974). 66. M. Saneyoshi and S. Nishimura, BBA 145,208 (1967). 67. M. Saneyoshi and S. Nishimura, BBA 204,389 (1970). 68. M. Saneyoshi and S. Nishimura, BBA 246, 123 (1971). 69. R. T. Walker and U. L. RajBhandary, BBRC 38,907 (1970). 70. Y. Yamada, K. Murao and H. Ishikura, NARes 9, 1933 (1981). 71. P. F. Agris, D. Soll and T. Seno, Bchem 12,4331 (1973). 72. G. C. Sen and H. P. Ghosh, NARes 3,523 (1976). 73. T. Seno, P. F. Agris and D. Sol], BBA 349,328 (1974). 74. R. P. Singhal, Bchem 13,2924 (1974). 75. Y. S. PrasadoRao and J. D. Cherayil, BJ 143,285 (1974). 76. D. Kern and L. Lapointe, Bchem 18,5819 (1979). 77. J. Weissenbach and G. Dirheimer, BBA 518,530 (1978). 78. E. Rudolf and K. Hike, ZpChem 356, 1359 (1975). 79. K. Watanabe, Bchem 19, 5542 (1980). 80. M. Saneyoshi, T. Anawi, S. Nishimura and T. Samejima, ABB 152, 677 (1972). 81. C. H. Yang and D. Soll, Bchem 13,3615 (1974). 82. R. J. Cedergren, N. Beaucheim and J. Toupin, Bchem 12, 4566 (1973). 83. M. Caron and H. Dugas, NARes 3, 19 (1976). 84. H. Dugas, Acc. Chem. Res. 10, 47 (1977). 85. I. Schwartz and J. Ofengand, Bchem 17,2524 (1978). 86. J. Ofengand, R. Liou, J. Kohout 111, I. Schwartz and R. A. Zimmerniann, Bchem 18, 4322 (1979). 87. J. Ofengand and R. Liou, Bchem 19,4814 (1980). 88. J. Ofengand and R. Liou, Bchem 20,552 (1981). 89. C. Ehresmann, B, Ehresmann, B. Millon, J. P. Ebel and K. Nurse, Bchem 23,429 (1984). 90. J. Ofengand, P. Gbrnicki, K. Chakraburtty and K. Nurse, PNAS 79, 2817 (1982). 91. J. Ofengand, P. Gbmicki, K. Nurse and M. Boublik, Alf7ed Benzoti Symp. 19,293 (1984). 92. P. Gbmicki, K. Nurse, W. Hellman, M. Boublik and J. Ofengand,JBC 259, 10493 (1984). 51. 52. 53. 54. 55.

70

RYSZARD W. ADAMIAK AND PIOTR C6RNICKl

93. S. Friedman, Nature NB 244, 18 (1973). 94. S. Friedman, H. J. Li, K. Nakanishi and G. Von Leer, Bchem 13,2932 (1974). 95. F. Seela, F. Hansske, K. Watanabe and F. Cramer, NARes 4,711 (1977). 96. F. Hansske, K. Watanabe and F. Cramer, ZpChem 359, 1659 (1978). 97. F. Hansske, K. Watanabe, F. Seela, F. Cramer, Methods Enzymol. 59, 166 (1979). 98. J. Ofengand and R. Liou, NARes 5, 1325 (1978). 99a. P. W. Schiller and A. N. Schechter, NARes 4,2161 (1977). 99b. S. Friedman, JBC 254, 7111 (1979). 100. M. Sprinzl and H. G . Faulhammer, NARes 5,4837 (1978). 101. J. A. Plumbridge, H. G. Bauniert, M. Ehrenberg and R. Rigler, NARes 8, 827 (1980). 102. V. Nauheimer and C. Hedgcoth, ABB 160,631 (1974). 103. I. Schwartz and J. Ofengand, BBA 697,330 (1982). 104. J. Ofengand, F.-L. Lin, L. Hsu, M. Keren-Zur and M. Boublik, Ann. N . Y. Acad. Sci. 346, 324 (1980). 105. T.-H. Kao, D. L. Miller, M. Abo and J. Ofengand,JMB 166,393 (1983). 106. P. Ghnicki, A. Wolanski, M. Judek, K. Gulewicz and W. Krzyzosiak, 14th IUPAC Sym3osium on Chemistry of Natural Products, Poznan, (1984). 107. R. H. Hall, M. J. Robins, L. Stesink and R. Theaford,JACS 88,2614 (1966). 108. K. Biemann, S. Tsunakawa, K. Sonnenbichler, H. Feldman, D. Dutting and H. G . Zachau, Angew. Chem. 7,600 (1966). 109. R. H. Hall, This series 10,57 (1970). 110. R. H. Hall and M. J. Robins, in “Synthetic Procedures in Nucleic Acid Chemistry” (W. W. Zorbach and R. S. Tipson, eds.), p. 210. Wiley, New York, (1968). 111. W. A. H. Grimm, T. Fujii and N. J. Leonard, ibid p. 212. 112. W. J. Burrows, D. J. Armstrong, F. Skoog, S. M. Hecht, I. T. A. Boyle, N. J. Leonard and J. Occolowitz, Science 161,681 (1978). 113. G . Shaw, B. M. Smallwood and D. V. Wilson,JCS C, 921 (1966). 114. A. J. Playtis and N. J. Leonard, BBRC 45, 1 (1971). 115. T. Sato, in “Synthetic Procedures in Nucleic Acid Chemistry” (W. W. Zorbach and R. S. Tipson, eds.), p. 264, Wiley, New York, (1968). 116. R. W. Adamiak, E. Biala and P. Swiderski, 14th IUPAC Symphosium on Chemistry of Natural Products, Poznan, (1984). 117. G . B. Chheda and C. I. Hang,]. Med. Chem. 14,748 (1971). 118. C. I. Hong, G. B. Chheda, S. P. Dutta, A. O’Grady-Curtis and G. T. Tritsch, J . Med. Chem. 16, 139 (1973). 119. S. P. Dutta, C. I. Hong, G . P. Murphy, A. Mittelmann and G. B. Chheda,J. Med. Chem. 18,3144 (1975). 120. C. I. Hong, A. Mittelman and G. B. Chheda, Pharm. Sci. 67, K 569 (1978). 121. C. I. Hong and G . B. Chheda, J . Med. Chem. 16,957 (1973). 122. R. W. Adamiak and M. Wiewi&owski, Bull. Acad. Polon. Sci. Ser. Sci. Chim. 23, 241 (1975). 123. J. Boryski and B. Golankiewicz, Bull. Acad. Polon. Sci. Ser. Chim.26, 21 (1978). 124. J. Boryski and B. Golankiewicz, J . Carbohydr. Nucleosides Nucleotides 6, 497 (1978). 125. C. I. Hong, G. L. Tritsch, A. Mittelman, P. Hebborn and G. B. Chheda,J. Med. Chem. 18,465 (1975). 126. R. W. Adamiak and J. Stawiliski, Tetrahedron Lett. 1935 (1977). 127. R. W. Adamiak, E. Biala, K., Grzeskowiak, R. Kierzek, A. Kraszewski W. T. Markiewicz, J. Okupniak, J. Stawinski and M. Wiewibrowski, NARes 5, 1889 (1978).


tRNA

71

128. G. B. Chheda, S. P. Dutta, A. Mittelman and L. Baczynskyj, Tetrahedron Lett. 5, 433 (1974). 129. D. Horvath and G. Denes, BBA 474, 188 (1977). 130. I. Weygand-Durasevic, V. Nothing-Laslo, J. N. Herek and Z. Kucan, BBA 479,332 (1977). 131. I. Weygand-Durasevic, T. A. Krause and B. F. C. Clark, EJB 116, 59 (1981). 132. V. Nothing-Laslo, I. Weygand-Durasevic, T. Zivkovic and Z. Kucan, EJB 117,263 (1981). 133. I. Weygand-Durasevic, V. Nothing-Laslo and Z. Kucan, EJB 139, 541 (1984). 134. F. Fittler and R. H. Hall, BBRC 25,441 (1966). 135. S. M. Hecht, L. Kirkegaard and R. M. Bock, PNAS 68,48 (1971). 136. Y. Furuichi, Y.Wataya, H. Hayatsu and T. Ukita, BBRC 41, 1185 (1970). 137. M . Lowdon and J. P. Goddard, NARes 3, 3383 (1976). 138. J. P. Goddard and M. Lowdon, F E E S Lett. 130,221 (1981). 139. W. R. ,Midden and E. J. Behrman, FEBS Lett. 103,301 (1979). 140. G. B. Chheda, R. H. Hall, D. I. Magrath, J. Mozeiko, M. P. Schweizer L. Stasiuk and P. R. Taylor, Bchem 8,3278 (1969). 141. M. P. Schweizer, G. B. Chheda, L. Baczynskyj and R. H. Hall, Bchem 8, 3283 ( 1969). 142. F. Kimura-Harada, D. L. von Minden, J. A. McCloskey and S. Nishimura, Bchem 11, 3910 (1972). 143. J. P. Miller and M. P. Schweizer, A d Biochem. 50, 327 (1972). 144. C. I. Hong and G. B. Chheda,J. Med. Chem. 18,79 (1975). 145. W. J. Krzyzosiak, J. Biernat, J. Ciesiolka, P. G6rnicki and M.Wiewihowski, NARes 7, 1663 (1979). 146. R. W. Adamiak, P. Gbrnicki and ?(I. Wiewi6rowski, NARes Sp. Publ. 4,215 (1978). 147. M. Funamizu, A. Terehara, A. M. Feinberg and K. Nakanishi, JACS 93, 6707 (1971). 148. H. Kasai, M. Goto, S. Takemura, T. Goto and S. Matsuura, Tetrahedron Lett. 2725 (1971). 149. C. R. Frihart, in “Proc. Int. Conf. Recent Developments in Oligonucleotide Synthesis and Chemistry of Minor Bases of tRNA” (Z. Paryzek, ed.), p. 261. A. Mickiewicz Univ. Press, Poznari, (1974). 150. H. Kasai, M. Goto, K. Ikeda, M. Zama, Y. Mizuno, S. Takemura, S. Matsuura, T. Sugimoto and T. Goto, Bchem 15, 898 (1976). 1.51. C. R. Frihart, A. M. Feinberg and K. Nakanishi,]. Org. Chem. 43, 1644 (1978). 152. S. Nakatsuka, T. Ohgi and T. Goto, Tetrahedron Lett. 2579 (1978). 153. K. Ienaga and W. Pfleiderer, Tetruhedron Lett. 1477 (1978). 154. T. Itaya and K. Ogawa, Tetrahedron Lett. 2907 (1978). 155. T. Itaya, T. Watanabe and H. Masumoto,]CS Chem. Commun. 1158 (1980). 156. B. Golankiewicz and W. Folkman, NARes 11, 5243 (1983). 157. B. Golankiewicz, W. Folkman, J. Boryski and E. Zielonacka-Lis, NARes Synip. Ser. 9, 127 (1981). 158. R. Thiebe and H. G. Zachau, EJB 5,546 (1968). 159. R. Thiebe, H. G. Zachau, L. Baczynskyj, K. Biemann and J. Sonnenbichler, BBA 240, 163 (1971). 160. W. J. Krzyzosiak and J. Ciesiolka, NARes 11, 6913 (1983). 161. G. Krauss, F. Peters and G. Maass, NARes 3, 631 (1976). 162. G. Krauss, R. Roemer, D. Riesner and G. Maass, FEBS Lett. 30, 1 (1973). 163. W. Wintermeyer and H. G. Zachau, FEBS Lett. 18,214 (1971). 164. W. Wintermeyer and H. G. Zachau, Mol. B i d . 9, 49 (1975).

72

RYSZARD W. ADAMIAK AND PIOTR G6FWICKI

165. 0. W. Odom, B. B. Craig and B. A. Hardesty, Biopolymers 17,2909 (1979). 166. W. Wintermeyer and H. G. Zachau, EJB 98,465 (1979). 167. W. Wintermeyer and H. G. Zachau, FEBS Lett. 18,807 (1971). 168. 0. W. Odom, D. Hardesty, W. Wintermeyer and H. G. Zachau, BBA 378, 159 (1975). 169. J. M. Robertson, M. Khan, W. Wintermeyer and H. G. Zachau, EJB 72,117 (1975). 170. D. Labuda and J. Augustyniak, EJB 79,303 (1977). 171. D. Labuda, T. Hertle and J. Augustyniak, EJB 79,293 (1977). 172. D. Labuda and D. Porschke, Bchem 19,3799 (1980). 173. B. B. Craig, 0. W. Odom, D. Foyt, J. M. White and B. Hardesty, in “Biomolecular Structure and Function” (P. F. Agris, ed.), p. 545. Academic Press, New York, (1978). 174. R. H. Fairclough and C. R. Cantor, JMB 132,575 (1979). 175. H. Paulsen, J. M. Robertson and W. Wintermeyer, NARes 10,2651 (1982). 176. H. Paulsen, J. M. Robertson and W. Wintermeyer, JMB 167,411 (1983). 177. K. Nagamatsu and Y. Miyzawa, /. Biochem 94, 1967 (1983). 178. J. M . Robertson, M. Kahan, M. Wintermeyer and H. G. Zachau, EJB, 72, 117 (1977). 179. R. Rigler, M. Ehrenberg and W. Wintermeyer, Springer Ser. M o l . Biochem. Biophys. 24, 219 (1977). 180. M. Ehrenberg, R. Rigler and W. Wintermeyer, Bchem 17,4588 (1979). 181. J. Robertson and W. Wintermeyer, J M B 151,57 (1981). 182. D. Robbins, 0. W. Odom, J. Lynch, G. Kramer, B. Hardesty, R. Liou and J. Ofengand, Bchem 20, 5301 (1981). 183. W. Wintermeyer and J. M. Robertson, Bchem 21,2246 (1982). 184. W. Wintermeyer and C. Gualerzi, Bchem 22,690 (1983). 185. S. Paszyc and M. Rafalska, NARes 6,385 (1979). 186. M. Baltzinger, F. Fasiolo and P. Remy, EJB 97,481 (1979). 187. A. J. M. Matzke, A. Barta and E. Kuechler, EJB, 112, 169 (1980). 188. W. Fuller and A. Hodgson, Nature 215,817 (1967). 189. A. Jack, J. E. Ladner and A. Klug,JMB 108,619 (1976). 190. A. Jack, J. E. Ladner, D. Rhodes, R. S. Brown and A. Klug,JMB, 111,315 (1977). 191. S. R. Holbrook, J. L. Sussman, R. W. Warrant, G. M. Church and S.-H. Kim, NARes 4, 2811 (1977). 192. J. L. Sussman, S. R. Holbrook, R. W. Warrant, G. M. Church and S,-H. Kim,JMB 123,607 (1978). 193. G. J. Quigley, M. M. Teeter and A. Rich, PNAS 75, 64 (1978). 194. C. D. Stout, H. Mizuno, S. T. Rao, P. Swaminathan, J. Rubin, T. Brennan and M. Sundaralingam, Acta Crystallogr. B34, 1529 (1978). 195. R. W. Schevitz, A. D. Podjamy, N. Krishnamachari, J. J. Hughes, P. B. Sigler and J. L. Sussman, Nature NB 278, 188 (1979). 196. D. Moras, M. B. Comarmond, J. Fischer, R. Weiss, J. C. Thierry, J. P. Ebel and R. Giege, Nature NB 288,669 (1980). 197. N. H. Woo, B. A. Roe and A. Rich, Nature NB 286,346 (1980). 198. P. R. Schimell and A. G. Redfield, Annu. Reu. Biophys. Bioeng. 9, 181 (1980). 199. W. Saenger, “Principles of Nucleic Acids Structure” (C. R. Cantor Ed.), SpringerVerlag, Berlin, (1983). 200. G. H.-Y. Lin and M. Sundaralingam, Acta Crystallogr. B27,961 (1971). 201. S. W. Hawkinson, Acta Crystallogr. B33,80 (1977). 202. W. Hillen, E. Egert, H. J. Lindner, H. G. Gassen and H. Vorbruggen,J. Carbohydr. Nucleosides. Nucleot. 5,23 (1978).


tHNA

73

203. K. Morikawa, K. Torri, Y. Iitaka, M. Tsuboi and S. Nishimura, FEBS. Lett. 48,279 (1974). 204. K. Morikawa, K. Torii, Y. Iitaka and M. Tsuboi, Acta Crystallogr. B31,1004 (1975). 205. G . I. Birnhaum, W. J. P. Blonski and F. E. Hruska, Can.J. Chem. 61,2299 (1983). 206. W. Hillen, E. Egert, H. J. Lindner and H. G. Gassen, FEBS. Lett. 94,361 (1978). 207. H. Kasai, S. Nishimura, H. Vorbrtiggen and Y. Iitaka, F E B S . Lett. 103,270 (1979). 208. H. M. Berman, D. Marcu, H. Narayanan, J. D. Fissekis and R. L. Lipnick, NARes 5, 893 (1978). 209. W. Hillen, E. Egert, H. J. Lindner and H. G. Gassen, Bchem 17,5314 (1978). 210. M. D. Topal and J. R. Fresco, Nature N B , 263,289 (1976). 211. W. Uhl, J. Reiner and H. G. Gassen, NARes 11, 1167 (1983). 212. R. L. Lipnick and J. R. Fissekis, Can. J . Biochem. 58, 1355 (1980). 213. R. L. Lipnick and J. R. Fissekis, J. Heterocycl. Chem. 17, 195 (1980). 213a. A. Galat, P. Serafinowski and J. Koput, BBA 801,40 (1984). 214. M. Yoshida, K. Takeishi and T. Ukita, BBA, 228, 153 (1971). 215. W. Bahr, P. Faerber and K.-H. Scheit, EJB 33, 535 (1973). 216. P. Faerber, FEBS Lett. 44, 111 (1974). 217. W. Saenger, S. K. Mazumdar, D. Suck and P. C. Manor in “Structure and Conformation of Nucleic Acids and Protein-Nucleic Acids Interactions” (M. Sundaralingam and S. T. Rao Eds.), p. 537. Univ. Park Press, Baltimore, (1974). 218. S. K. Mazurndar, W. Saenger and K.-H. Scheit,JMB 85,213 (1974). 219. S. Yokoyama, Z. Yamaizumi, S. Nishimura and T. Miyazawa, NARes 6, 2611 (1979). 220. C . E. Bugg and V. Thewalt, BBRC 46,779 (1972). 221. R. K. McMullan and M. Sundaralingam, BBRC 43, 1158 (1971). 222. R. K. McMullan and M. Sundaralingam, JACS 93, 7050 (1971). 223. R. Parthasarathy, J. M. Ohrt and G. B. Chheda, BBRC 57, 649 (1974). 224. R. Parthasarathy, J. M. Ohrt and G. B. Chheda, JACS 96,8087 (1974). 225. D. A. Adamiak, T. L. Blundell, I. J. Tickle and Z. Kosturkiewicz, Acta Crystallogr. B31,1242 (1975). 226. R. Parthasarathy, M. Soriano-Garcia and G. B. Chheda, Nature N B 260,807 (1976). 227. R. Parthasarathy, J. M. Ohrt and G. B. Chheda, BBRC 60, 211 (1974). 228. R. Parthasarathy, J. M. Ohrt and G . B. Chheda, Bchern 16,4999 (1977). 229. M. P. Schweizer and M. P. Hamill, Jr., BBRC 85, 1367 (1978). 230. P. R. Reddy, M. P. Schweizer and G. B. Chheda, F E B S Lett, 106,63 (1979). 231. P. R. Reddy, W. D. Hamill Jr., G. B. Chheda and M. P. Schweizer, Bchern. 20,4979 (1981). 232. M. P. Schweizer, N. De, M. Pulsipher, M. Brown, P. R. Reddy, C. R. Petrie, 111, and G. B. Cheda, BBA 802,352 (1984). 233. M. T. Watts and I. Tinoco Jr., Bchem 17,2455 (1978). 234. L. S. Kan, P. 0. P. Ts’o, F. von der Haar, M. Sprinzl and F. Cramer. Bchem 14, 3278 (1975). 235. A. Maelicke, F. von der Haar, M. Sprinzl and F. Cramer, Biopolyners 14, 155 (1975). 236. A. Maelicke, F. von der Haar and F. Cramer, Biopolyrner.9 12, 27 (1973). 237. P. Dea, M. Aka, S. Patt and M.P. Schweizer, NARes 5, 307 (1978). 238. J. A. Lake, PNAS 74, 1903 (1977).(See also this series 30, 163.) 239. M. M. Teeter, G. J. Quigley and A. Rich, in “Nucleic Acid Metal Ion Interactions” (T. G. Spiro, ed.), p. 147 Wiley, New York, (1980). 240. L. S. Kan, P. 0. P. Ts’o, F. von der Haar, M. Sprinzl and F. Cramer, BBRC 59,22 (1974).

74

RYSZARD

w. ADAMIAK AND PIOTR

C~RNICKI

241. L. S. Kan, P. 0. P. Ts’o. M. Sprinzl, F. von der Haar and F. Cramer, Bchem 16, 3143 (1977). 242. R. Langlois, S.-H. Kim and C. R. Cantor, Bchem 14,2554 (1975). 243. K. Nakanishi, N. Furutachi, M. Funamizu, D. Grunberger and I. B. Weinstein, JACS 92,7616 (1970). 244. K. Beardesley, T. Tao and C. R. Cantor, Bchem 9,3524 (1970). 245. J. Eisinger, B. Feuer and T. Yamane, PNAS 65,638 (1970). 246. D. R. Kearns, This series 18,91 (1976). 247. P. H. Bolton and D. R. Kearns, EBA 477, 10 (1977). 248. D. J. Patel, Annu. Reu. Phys. Chem. 29, 337 (1978). 249. C. Woese, Nature NB 226,817 (1970). 250. F. H. C. Crick, S. Brenner, A. Klug and G. Pieczenik, Origins o f l i f e 7,389 (1976). 251. C. Urbanke and G. Maass, NARes 5, 1551 (1978). 252. R. Ehrlich, j. F. Lefevre and P. E. Reniy, EJB 103, 145 (1980). 253. W. E. Blumberg, R. E. Dale, J. Eisinger and D. M. Zuckerman, Biopolymers 13, 1607 (1974). 254. B. D. Wells, NARes 12, 2157 (1984). 255. D. Labuda and D. Porschke, Bchem 21,49 (1982). 256. F. H. C. Crick,JMB 19,548 (1966). 257. R. Thedford and D. B. Straus, Bchem 13,535 (1974). 258. G. Hogenauer, EJB 12,527 (1970). 259. 0. C. Uhlenbeck, J. Baller and P. Doty, Nature NB 225,508 (1970). 260. 0. C. Uhlenbeck, ] M E 65,25 (1972). 261. J. Eisinger and P. F. Spahr, ] M E 73, 131 (1973). 262. H. Grosjean, D. G. Sol1 and D. M. Crothers,JMB 103,499 (1976). 263. H. Grosjean, S. de Henau and D. M. Crothers. PNAS 75, 610 (1978). 264. A. Moller, U. Schwarz, R. Lipecky and H. G. Gassen, FEBS Lett. 89,263 (1978). 265. R. Rigler, C . R. Rabl and T. Jovin, Rev. Sci. Znst. 45,550 (1974). 266. P. Davanloo, M. Sprinzl and F. Cramer, Bchem 18,3189 (1979). 267. H. A. M. Geerdes, J. H. van Boom and C. W. Hilbers,JMB 142, 195 (1980). 268. H. A. M. Geerdes, J. H. van Boom and C. W. Hilbers, JMB 142,219 (1980). 269. J. Ninio, ] M E 56, 63 (1971). 270. J. Ninio, This series 13,301 (1973). 271. E. Akoboshi, M. Inouye and A. Tsugita, MGG 149, 1 (1976). 272. S. I. Feinstein and S. Alhnan, Genetics 88,201 (1978). 273. E. J . Murgola, F. T. Pagel and K. A. Hijazi, ] M E 174, 19 (1984). 274. M. J. Caner and R. W. Buckingham,JME 174,29 (1984). 275. S. M. Freier and I. Tinoco, Jr., Bchem 14,3310 (1975). 276. J. Wiessenbach and H. Grosjean, EJB 116,207 (1981). 277. S. K. Dube, K. A. Marcker, B. F. C. Clark and S. Cory, Nature N E 218,232 (1968). 278. F. M. Martin, 0. C. Uhlenbeck and P. Doty,JMB 57,201 (1971). 279. P. J. Romaniuk, D. W. Hughes, R. J. Gregoirre, T. Neilson and R. A. Bel1,lACS 100,3971 (1978). 280. S. M. Freier, B. J. Burger, D. Alkema, T. Neilson and D. H. Turner, Bchem 22, 6198 (1983). 281. S. M. Freier, M. Petersheim, D. R. Hickey and D. H. Turner,]. Bionzol. Stmct. Dynam. 1, 1229 (1984).

Ribosomal Translocation: Facts and Models

I

ALEXANDERS. SPIRIN Institute of Protein Research Academy of Sciences of the USSR Pushchino, Moscow Region, USSR

I. Definition.. . . . . . . . . . . . Experimental Tests. . . . . Two-tRNA-Site Model for the Ribosomal Elongation Cycle . . . . . . . Main Facts Concerning Translocation . . . . Sequence of Events in Translocation Prom VI. Energetics of Translocation A. General Considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Catalysis of Transloca C. Translocation Models with GTP (Translocation Driven by GTP) . . . . . . . . . . . . . . . . . . . D. Translocation Models Implying All Displacements to Be Thermodynamically Spontaneous EF-G) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . VII. Kinematics of Translocation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . A. Mutual Orientation of the Two tRNAs B. Helical Displacement of Peptidyl C. Movements of tRNAs on the Ribosome ...... VIII. Are Conformational Movements of the Ribosome Required for Translocation? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IX. Concluding Remarks. . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . .... ................... 11. 111. IV. V.

75 77 79 81 84

86 86 88 90

91 93 93 97 101

105 108 109

1. Definition Ribosomes synthesize the polypeptide chain of a protein successively from the NHz-end to the COOH-end. The elongation of the polypeptide by one amino-acid residue proceeds as a result of a cycle in which the ribosome binds the substrate, attaches it to the nascent chain, and prepares to accept the next substrate. In each elongation cycle (Fig. 1)the amino-acid residue enters the ribosome as aminoacyl-tRNA in accordance with the specificity of the template codon located at the given moment in a defined (acceptor) site of the ribosome. The nascent peptide chain is retained in the ribosome so that its COOH-end remains covalently bound to the tRNA residue that delivered the preceding amino-acid residue. The peptide bond is formed as a result of the transpeptidation reaction between 75 Progress in Nucleic Acid Research and Molecular Biology, Vol. 32

Copyright 0 1985 hy Acadcmic Preys, Inr. All rights of reproduction in any form reserved.

76

ALEXANDER S. SPIRIN

Translocatio

tRNA’ binding

Transpeptidat ion FIG.1. Scheme of‘the elongation cycle performed by the ribosome.

the aminoacyl-tRNA that has entered the ribosome and the peptidyltRNA retained on it: Pept (n)-tRNA’ + Aa-tRNA”-

tRNA’ + Pept (n+l)-tRNA

(1)

Thus the substrates of the ribosome in this reaction are peptidyl-tRNA (donor of the peptidyl) and aminoacyl-tRNA (acceptor of the peptidyl), and its products are an elongated peptidyl-tRNA and a deacylated tRNA. However, ribosomal transpeptidation does not result in the release of the products, tRNA’ and Pept(n+ 1)-tRNA”,into solution. In this posttranspeptidation state, the ribosome is incapable of binding the next substrate (aminoacyl-tRNA”’)because the substrate-binding (acceptor) site is occupied by the product (peptidyl-tRNA). On the other hand, the peptidyl-tRNA cannot be a substrate in the next transpeptidation reaction because it occupies the position of the product. In order for the ribosome to be competent for further elongation, the peptidyl-tRNA must be displaced from its product position to the position the substrate peptidyl-tRNA‘ occupied before transpeptidation. Translocation is the term used to denote the intraribosomal transition of the elongated peptidyl-tRNA from the position of the reaction product, where it is incapable of participating in transpeptidation as a

RIBOSOMAL TRANSLOCATION

77

FIG.2. Scheme of the sequence of events in the ribosome-catalyzed puromycin (PM) reaction.

peptidyl donor, to the position of the substrate where it can again react with the other substrate, the acceptor. This transition is coupled with the release of the second product of the reaction, the deacylated tRNA. Translocation is accompanied by a shift of the template (messenger) polynucleotide relative to the ribosome by one codon and the appearance of competence for binding the second substrate, the aminoacyl-tRNA.

II. Experimental Tests In accordance with the abovementioned, there are four ways for measuring translocation based on the following criteria: (1)transpeptidation with the low-molecular-weight acceptor substrate, puromycin; (2) binding of aminoacyl-tRNA and its incorporation into the polypeptide chain; (3)release of deacylated tRNA; (4)shift of the template polynucleotide. 1. The puromycin reaction is the simplest and the most widely used test. Puromycin is a low-molecular-weight analog of aminoacyltRNAs serving as an acceptor for the peptidyl-transferase center of the ribosome: the amino group of its aminoacyl residue attacks the ester group of the donor substrate, peptidyl-tRNA (or its analog); as a result, transpeptidation between them occurs and the peptidyl-puromycin falls out of the ribosome (Fig. 2). The peptidyl-tRNA of the posttranslocation ribosome reacts with puromycin, whereas that of the pretranslocation ribosome does not (1-10; see also reviews 1 1 , 12). Thus, the release of the peptide from ribosomes in response to the addition of puromycin is a quantitative measure of the posttranslocation state in the ribosome preparation under investigation. Conversely, noncompetence for puromycin indicates a pretranslocation state of the particles. 2. Aminoacyl-tRNA binding can also serve as a quantitative measure of translocation (5-8, 10, 13-1 6). Codon-dependent aminoacyltRNA binding with the translating ribosome is impossible immedi-

78

ALEXANDER S. SPIRIN

3'

r I

Translocation

5' tRNA'

5’ FIG.3. Scheme of the 5'-3'-end-directed movement of mRNA relative to the ribosome by one triplet as a result of translocation.

ately after the transpeptidation reaction; the substrate will be bound only after translocation. The binding of aminoacyl-tRNA is usually followed by its immediate incorporation into the growing peptide, so that the posttranslocation state of the ribosome can be judged also by the appearance of a new amino-acid residue in the product. 3. Release of deacylated tRNA from the translating ribosome is coupled with translocation (17-25). If the tRNA residue of the peptidyl-tRNA molecule retained in the ribosome is tracked, it will appear in the deacylated form after the transpeptidation reaction [see Eq. (l)],but will still be bound to the ribosome (see Fig. 1, bottom left), and only translocation will release the deacylated tRNA. 4. Shift of the template polynucIeotide as a translocation test is technically the most complicated. It can be either indirect when based on the appearance of competence for the binding and incorporation of the aminoacyl-tRNA specific to the codon following that previously fixed in the ribosome (9, 10, 15, 16, 26, 27), or direct if the change of the ribosome-protected section of the template polynucleotide is analyzed. The direct test has shown that the shift of the template polynucleotide relative to the ribosome by one nucleotide triplet (Fig. 3) accompanies the appearance of competence for puromycin and for amino-acyl-tRNA binding (28,29).


79

111. Two-tRNA-Site Model for the Ribosomal Elongation Cycle According to the classical model, the elongation of a polypeptide on the ribosome proceeds by means of repeating cycles each consisting of three successive steps: aminoacyl-tRNA binding (I), transpeptidation (11), and translocation (111) ( 3 0 , 3 1 ) .Two tRNA binding sites on the ribosome, the so-called A and P (or D) sites, are postulated. In Step I, an aminoacyl-tRNA binds to the A site containing a vacant codon of a template polynucleotide; the binding is catalyzed by the elongation factor Tu (EF-Tu) with GTP. A peptidyl-tRNA, i.e., another tRNA molecule coupled to the nascent peptide, is situated at this moment in the P site. In Step 11, the aminoacyl-tRNA of the A site reacts with the peptidyl-tRNA of the P site, so that transpeptidation takes place (Eq. 1); the reaction is catalyzed by the peptidyltransferase (EC 2.3.2.12)of the ribosome itself. Now the tRNA residue of the elongated peptidyl-tRNA occupies the A site, and the deacylated tRNA is in the P site. This is the pretranslocation state. In Step 111, the tRNA residue of the peptidyl-tRNA together with the template codon is displaced from the A site to the P site, and the deacylated tRNA is released from the P site; the displacements are catalyzed by elongation factor G (EF-G) with GTP. As a result, the posttranslocation state is again achieved where a peptidyl-tRNA is in the P site and the A site, with a new vacant codon, is ready to accept the next aminoacyl-tRNA. These notions are represented schematically in Fig. 4. The above model is consistent with most experimental data obtained in the last 15 years. At present there are no well-established facts contradicting this basic scheme of the elongation cycle, with two nonoverlapping tRNA-binding sites on the ribosome. At the same time, a multi-center character of tRNA binding, a transient participation of protein factors, and a certain conformational flexibility of both the tRNA and the ribosome suggest that intermediate binding states on the way to the final position of tRNA in its site can exist. This has given rise to the concept of additional binding sites overlapping either the A or the P site. For example, a “recognition (R) site” was proposed as a preliminary stage of aminoacyl-tRNA binding to the A site (32,33).Binding of tRNA to the A site through intennediate stages was recently demonstrated by kinetic techniques (34).Naturally, either the real existence of the intermediate site overlapping the A site or simply a functional substate of tRNA binding with the same A site are equally plausible alternatives here. In any case, these are consistent with the two-tRNA-site model for the elongation cycle.

80

ALEXANDER S . SPIRIN

t RNA EF-G.GDP

IRNA. EF-TU.GTP

4

GDP

5

FIG.4. Scheme of the elongation cycle in terms of the two-tRNA-sites model demonstrating the sequence of events and the participation of elongation factors with GTP, as well as the tentative positions of the ligands. The ribosome is shown by a solid-line contour as built u p of two subunits, the upper being a 30 S and the lower a 50 S; tRNAs are depicted as L-shaped wedged figures, mRNA as a segmented ribbon, amino-acid residues as small open circles. A and P are ribosomal sites for tRNAs.

On the other hand, the possible existence of a nonoverlapping “exit” (E) site, in addition to A and P sites, has also been discussed (35-40). This means that translocation displaces deacylated tRNA not immediately into solution but first into a third site on the ribosome. If the “E site” is just a transient position from which the deacylated tRNA spontaneously dissociates into solution independently of the subsequent step of the elongation cycle (41a), this does not undermine the two-tRNA-site mode1 for the elongation cycle. In this case, the “E site” can be considered simply as a transient “presolution” state of the released tRNA. However, if the deacylated tRNA is found to be firmly bound to the “E site” and released from the ribosome only in response to the next step of the elongation cycle (37),the concept of the two-tRNA model should be revised. There are experimental data contradicting a three-tRNA-site model of the elongation cycle (see, e.g., 41, 42).


81

Below, I use the concepts ofthe A and P sites, assuming, however, that some substates of the binding of tRNA to the ribosome can exist.

IV. Main Facts Concerning Translocation 1. Translocation is catalyzed by elongation factor G (EF-G) with GTP (3-5,14,15,26,43) or, in the case of eukaryotic ribosomes, by its . (or, correspondingly, EF-2) is a analog EF-2 with GTP ( 6 , 7 , 1 3 ) EF-G rather large protein that interacts with GTP and with the ribosome. This interaction induces GTPase activity, and GTP is cleaved to GDP and orthophosphate. Upon interaction (complex formation) of EF-G + GTP with the pretranslocation state ribosome, rapid translocation takes place, whereupon EF-G, GDP and orthophosphate are released from the ribosome. 2. Catalysis of translocation results from the attachment of EF-G with GTP to the ribosome. This fact has been demonstrated in experiments where translocation is effectively induced by EF-G with a noncleavable analog of GTP (guanylyl methylene diphosphonate or guanylyl imidodiphosphate) instead of GTP ( 2 1 , 23, 44-47). The attachment of EF-G with the GTP analog to the pretranslocation state ribosome allows translocation to proceed rapidly, as judged by the puromycin reaction and by the release of deacylated tRNA. However, the attached EF-G prevents the binding of the next aminoacyl-tRNA; the removal of EF-G from the post-translocated ribosome permits aminoacyl-tRNA binding ( 23, 47) .Thus, GTP (or its analog) is considered to be an allosteric effector of the protein EF-G, inducing its affinity for the ribosome. Formation of the complex of ribosome and EF-G catalyzes translocation. T h e subsequent cleavage of GTP is necessary for EF-G to lose its affinity for the ribosome and so to be released into solution. 3. Translocation can also proceed without EF-G with GTP (4855). Such a “nonenzymic translocation” proceeds much slower than that catalyzed b y (EF-G)*GTP,but, nonetheless, leads to the normal posttranslocation state of the ribosome capable of continuing elongation. Consequently, the translocation process can be thermodynamically spontaneous, and the translocation mechanism is principally inherent to the ribosome itself (and is not introduced by EF-G) (56). 4 . Translocation is inhibited by high magnesium concentrations and stimulated by low concentrations of magnesium ions (22,24, 50, 51,5557).This pertains both to the EF-G-catalyzed and to the nonenzymic translocation. The pretranslocation state is practically frozen at

82

ALEXANDER S. SPIRIN

30 mM Mg2+even in the presence of EF-G with GTP (22).On the contrary, at 3 mM Mg2+ spontaneous nonenzymic translocation proceeds so rapidly that EF-G with GTP practically makes no additional contribution to the reaction rate (22). The posttranslocation state resulting from the effect of low Mg2+concentrations is normal in the sense that ribosomes can accept the next amino-acyl-tRNA and continue elongation. Thus, a decrease of the Mg2+concentration can imitate the effect of EF-G with GTP. 5. The ribosome-bound tRNAs interact with codons of the polynucleotide template in both the A and the P site (24,34,58-66).Moreover, if the anticodon loop of peptidyl-tRNA is photocross-linked with the mRNA in the A site, it moves into the P site during translocation, thus demonstrating that the tRNA and mRNA (anticodon and codon) are transported together remaining bound to each other (67). 6. Shift of the template during translocation is passive and is driven by tRNA. The best proof of this fact are experiments where the mutant tRNA, having a nucleotide quadruplet instead of a triplet as the anticodon, suppresses (+ 1)frame-shift mutation, i.e., moves the mRNA in the ribosome by four (and not by three) nucleotide residues (68). 7. The presence of the template polynucleotide and, consequently, the interaction of tRNA with the codon is not obligatory for translocation. Deacylated tRNA, amino-acyl-tRNA, and peptidyl-tRNA can bind with the ribosome and be retained on it in the absence of the template polynucleotide, though with a lower affinity than in template-directed binding (59, 61, 63, 65, 69-77). Some aminoacyltRNAs, such as Lys-tRNA, Ser-tRNA, Thr-tRNA, and Asp-tRNA, can serve as substrates for template-free synthesis of the corresponding polypeptides on the ribosome in the presence of both the elongation factors (EF-Tu and EF-G) and GTP (78-80). Without EF-G, only dipeptides are formed, indicating that the A site and the P site are occupied by aminoacyl-tRNA and that transpeptidation takes place between them, but without the subsequent translocation. The addition of EF-G with GTP induces translocation in the absence of the template and, consequently, a further elongation. Translocation in the absence of a template is inhibited by the same specific drugs known to act on the translocation process in conventional template-dependent systems. 8. Of the two ribosomal tRNA sites (A and P), the P site has a preference for tRNA binding. The main fact is that the P site in the vacant ribosome is the first to be settled, while the A site can be stably occupied only after the occupation of the P site (34,58,62,81).At low Mg2+concentrations, the binding of tRNA or its acylated derivatives


83

with the template-programmed vacant ribosome results in the occupation of only the P site, and tRNA can be retained in the A site as well only at increased Mg2+concentrations (18,19, 74,82). Later quantitative studies on the binding of deacylated tRNA (38, 63, 83) or peptidyl-tRNA and its analogs (25, 59, 61, 84-86) with the vacant 7 0 4 ribosome, both in the presence and absence of the template polynucleotide, have shown that in all cases the affinity constant for the P site (with the A site vacant) is higher than that for the A site (with the P site occupied). 9. There are, apparently, two states of the ligand (amino-acyl-tRNA or peptidyl-tRNA) bound in the A site of the ribosome: the reversible and the occluded (34, 81, 87). In the occluded state, the tRNA is incapable of dissociating or exchanging with the extraneous tRNA, or does so very slowly. If the aminoacyl-tRNA binding to the A site takes place in the complex with EF-Tu and GTP, then its transition into the occluded state is observed after the GTP cleavage but before the peptidyltransferase reaction. It is likely that the transpeptidation reaction proceeds in this state. After the transpeptidation reaction, the resulting peptidyl-tRNA in the A site appears also to be temporarily occluded. 10. Reduced retention of peptidyl-tRNA in the A site of the pretranslocation ribosome correlates with the high probability of translocation. A striking example is the observation that Ac-Phe-Tyr-tRNATYr is unstable in the A site of the pretranslocation poly(U)-programmed ribosome, and is either released from the ribosome or moved to the P site (thus displacing the cognate deacylated tRNAPhe)without the participation of EF-G (88). The antibiotic viomycin inhibits both this spontaneous translocation and the release of peptidyl-tRNA from the A site (88,89). Similar observations have been reported with the antibiotic hygromycin B (90). Inhibition of translocation by some other antibiotics, such as neomycin, kanamycin, and gentamycin (91), also correlate with stabilization of tRNA retention in the A site (92, 93). 11. Occupation of the A site is prerequisite for the release of deacylated tRNA from the P site under translocation conditions. In other words, if the P site contains deacylated tRNA and the A site is vacant, the addition of EF-G with GTP effects no release of the deacylated tRNA from the ribosome (17-20,24,94). For the release of the tRNA from the P site to take place under conditions of translocation, the A site can be occupied by any tRNA derivative, either by peptidyl-tRNA (normal situation during elongation) ( 1 7), or by aminoacyl-tRNA (25, 84), or by deacylated tRNA (18, 19, 24). 12. The peptidyl residue coupled to the tRNA in the A site does not

84

ALEXANDER S. SPIRIN

seem to be obligatory for the principal translocation mechanism. Indeed, if both the A site and the P site of the ribosome are occupied by deacylated tRNAs, the addition of EF-G with GTP results in the release of the tRNA from the P site and, seemingly, the transport of another tRNA from the A to the P site (28, 1 9 , 2 4 ) .

V. Sequence of Events in Translocation Promoted by EF-G Some of the experimental facts, such as those listed above in Section IV as Facts 1and 2 as well as the results of studies on the action of specific inhibitors, suggest a sequence of events during EF-G-promoted translocation that is schematically given in Fig. 5. The first event is the interaction of the pretransIocation state ribosome with EF-G and GTP. This event can be inhibited by thiostrepton, a specific antibiotic capable of binding to the 50-S ribosomal subunit in the region of the EF-G binding site (94a).The result of this event should be the formation of the complex between the pretranslocation state ribosome and (EF-G).GTP or (EF-G).GMPPCP. However, no stable (long-lived) complex involving a pretranslocation state ribosome can be observed under experimental conditions. It is likely that the complex is very unstable and short-lived, so that it either quickly dissociates or immediately transforms into the next (posttranslocation) state. Anyhow, the interaction of the pretranslocation ribosome with (EF-G)*GTPor (EF-G)*GMPPCPseems to induce quick translocation. This is the second event in the scheme under consideration. The translocation event proper can be inhibited either by decreased temperature or by high Mg2+concentration; for example, at 4"C, or in 30 mM Mg2+at a physiological temperature (30-37°C) the translocation is virtually blocked (22). Some specific agents such as viomycin also inhibit this event, i.e., translocation in the strict sense of the word (89). The third event in the considered scheme is the hydrolysis of GTP and the release of orthophosphate. At this stage, the posttranslocation ribosome firmly retains EF-G until GTP hydrolysis occurs. Of course, if GTP is replaced by a nonhydrolyzable analog such as GMPPCP, this event does not take place, and the posttranslocation ribosome will continue to be in a complex with (EF-G).GMPPCP. After GTP hydrolysis has taken place, (EF-G).GDP is found in a complex with the posttranslocation ribosome. This complex is unstable and dissociates-that is, EF-G and GDP are released from the

85


( 9 ) EF-G I N T E R A C T I O N

SHORT-LIVED

(21 T R I N S L O C A T I O N

n

-

tRNA

(31 GTP

(4) EF-G

EF-G. GDP

FIG.5. The sequence of events in the process of the (EF-G),GTP-promotedtranslocation. Designations and symbols are the same as in Fig. 4. The big shaded or open circle inside the ribosome contour is an EF-G molecule hound with GTP or GDP, respectively.

86

ALEXANDER S . SPIEUN

ribosome. This is the fourth event in the scheme. Fusidic acid specifically inhibits precisely this stage by fixing the complex and preventing the release of (EF-G).GDP from the ribosome (94a, 95).

VI. Energetics of Translocation A. General Considerations The question can be posed: What does the utilization of GTP energy for translocation mean? It is evident that this can mean either the performance of “useful work” against thermodynamic potential (“uphill” process), or just the overcoming of barriers in the spontaneous (“downhill”) process with no accumulated result. The useful work done in the process of translation is the formation of a template-determined polypeptide. It includes the formation of peptide bonds between amino-acid residues and their non-random, predetermined arrangement in the chain. Correspondingly, the useful work of one elongation cycle consists of a codon-directed selection of an aminoacyl-tRNA and its incorporation into polypeptide. The possibility of carrying out precise translation, though a slow one, in the absence of the elongation factors and GTP (see 55, 56) suggests that all the useful work for synthesis of a template-determined polypeptide from aminoacyl-tRNA can be done only at the expense of energy accumulated during tRNA aminoacylation. Indeed, in ribosome-catalyzed transpeptidation, the ester bond in Pept(n)-tRNA’ is replaced by the amide bond in Pept(n+ 1)-tRNA” The standard free energy of hydrolysis of the ester [see reaction (l)]. bond between the aminoacyl residue and the tRNA residue is about -7 to -8 kcal/mol, whereas the standard free energy of hydrolysis of a peptide (amide) bond is about -0.5 kcal/mol [see Chapter 5 in (96)l; hence, the formation of peptide bonds during elongation is a “downhill” process and is accompanied by the release of free energy of about -750.5 kcal per each bond (under standard conditions). The entropy loss owing to the ordered arrangement of the amino-acid residues along the polypeptide chain can be estimated to be no more than 2.5 kcal/mol of amino acid. Thus, the elongation process is fully provided with energy at the expense of the transpeptidation reaction, i.e., the useful work in polypeptide synthesis requires no additional energy sources. Hence, if the cleavage of the nucleoside triph‘osphates (GTP) takes place in addition to the transpeptidation reaction, the energy released here cannot be utilized in any way other than to be dissipated as heat as a result of every elongation cycle. This means


87

that the energy of GTP hydrolysis is expended only to overcome energy barriers. Thus, both general considerations and the fact of factor-free translation contradict the concept that the energy of GTP cleavage is utilized in aminoacyl-tRNA binding and translocation for performing useful work for translation, i.e., the idea of an energy (thermodynamic) contribution of both EF-Tu with GTP and EF-G with GTP to protein synthesis should be rejected. Some comments should be made here to avoid misunderstanding. It is often claimed that an elongation factor with GTP carries out “work” at a definite step of the elongation cycle, such as unwinding a template polynucleotide, carrying over aminoacyl-tRNA to its site or its “accomodation,” expulsion of deacylated tRNA from the pretranslocation ribosome, or translocation as a whole (see above), etc. In reality, in all these cases the matter under consideration is only the overcoming of energy barriers, but in no way the performance of useful work. It is evident that if the given step of the cycle is thermodynamically allowed (AGIO for the given step) but kinetically inhibited or retarded, the additional energy expended for its real accomplishment or acceleration cannot be spent for doing useful work. Moreover, if the net energy balance of the complete cycle boils down to the decrease of free energy (AGO for the given step) at the expense of the whole cycle, and, consequently, the performance of additional work here cannot take place either. Thus, while energy is expended during GTP cleavage at the steps of aminoacyl-tRNA binding and translocation, the result is not performance of useful work, but acceleration of the cycle (56). Consequently, the elongation factors with GTP introduce purely kinetic (catalytic) contributions into translation. The following question can be considered. What could be the meaning of GTP-driven “active transport” or “active expulsion” of the ribosomal ligands without the performance of useful work? It could be thought that induction of electrostatic repulsion or mechanical shift of some rigid molecular structure in the region of a binding site could ensure such an active step (in accordance with the models considered below, in Section V1,C). However, as already mentioned, in any case this would mean only the energy-dependent overcoming of the barrier in the thermodynamically spontaneous process, i.e., a kinetic contribution. Alternatively, diffusion movements could be considered with equal success, without “active” pulling or pushing, provided an energy-dependent structural change of the ligand-binding site results

88

ALEXANDER S. SPIRIN

in either a temporary decrease of affinity for the ligand or removal of steric restrictions for diffusion in the required direction. At the molecular level, the utilization of heat movement (see below, in Section V1,D) seems more likely than either rigid kinematics and mechanical movements inherent to macromachines or electrostatic repulsion of large ligands in a counter-ion-saturated medium. Thus, in summarizing, it can be asserted that at a certain level of aminoacyl-tRNA in the medium, the elongation cycle is free-energyensured in the absence of GTP; as a result of each cycle the free energy of the system decreases. Hence, all the steps of the cycle, including translocation, can take place without the participation of additional GTP energy, independently of whether the given separate step proceeds “downhill” or “uphill.” It is evident that the decrease of the free energy level of the system is conditioned by the transformation of aminoacyl-tRNA into deacylated tRNA in the course of elongation. In natural conditions, a permanent regeneration of aminoacyltRNA from the released deacylated tRNA takes place at the expense of ATP energy with the participation of aminoacyl-tRNA synthetases. This means that the aminoacylation cycle is the energy motor of the elongation cycle, and, consequently, of the translocation.

B. Catalysis of Translocation Under usual conditions, the dissociation of deacylated tRNA from the P site of the pretranslocation ribosome and the translocation of peptidyl-tRNA from the A site to the P site in the absence of EF-G with GTP are kinetically retarded, i.e., factor-free translocation is slow. EF-G with GTP strongly increases the rate of translocation. The fundamental starting point for understanding the role of GTP in the catalysis of translocation is the fact that EF-G with a non-cleavable GTP analog (e.g., guanylyl methylene diphosphonate) also speeds up translocation (see Fact 2 in Section IV). It turned out that without GTP or its noncleavable analog EF-G has no significant affinity for the ribosome. The formation of the complex of EF-G with GTP or its noncleavable analog induces its affinity for the ribosome. The following attachment of the (EF-G)*GTPor the (EF-G)*GMPPCPcomplex to the ribosome is sufficient for translocation to proceed rapidly. GTP hydrolysis is not required for the acceleration of translocation. Hence, GTP plays the role of an allosteric effector inducing the affinity of EFG for the ribosome. Thus, translocation is catalyzed by the formation of a complex between the ribosome and EF-G. A number of observations suggest that


89

this catalysis is of an entropic nature (56, 97). This means that the presence of EF-G results not in a new translocation pathway through intermediate steps to by-pass a high activation barrier, but better spatial conditions in the ribosome for the same translocation pathway. One of the ways to do this might be a simple fixation of one of the thermally fluctuating substates of the ribosome which would be favorable for translocation (see below). Such a fixation or orientation effect of the attachment of EF-G as a large additional ribosomal ligand seems likely. Nonetheless, in the course of the elongation cycle, GTP is cleaved with the participation of EF-G. GTP cleavage occurs directly after translocation (see Fact 2 in Section IV). What is the role of GTP hydrolysis? If the noncleavable GTP analog is used, translocation takes place, but the further course of the elongation cycle proves to be blocked. This is due to the fact that the EF-G with the noncleavable GTP analog remains bound to the posttranslocation ribosome and blocks the following step of the cycle, namely aminoacyl-tRNA binding (23,47).Thus the role of GTP hydrolysis is that it results in the destruction of GTP as an allosteric effector of EF-G, and, consequently, to the elimination of EF-G affinity for the ribosome. The question arises why the catalyst of translocation hinders the next step of the elongation cycle and must be necessarily removed after completion of the catalyzed step? It can be thought that this is directly connected with the entropic nature of the translocation catalysis. Indeed, according to the concepts considered earlier (97),an entropic catalyst selects and fixes a strictly defined conformational situation in the ribosome, uniquely favorable for the given step of the process. However, remaining bound, it continues to fix this conformation which is not suitable for another step requiring other spatial conditions. That is why it must be removed immediately after completion of the catalyzed step. Just this may be the case of EF-G as a translocation catalyst. Thus, in summarizing, the process of translocation in the ribosome can be fully described as channelled diffusionary displacements of the tRNA ligands. Correspondingly, all the displacements are just heat motions and proceed “downhill.” Hence, the energy of GTP cleavage is not required to perform useful work here. GTP serves as a cleavable allosteric effector for EF-G. The (EF-G)*GTPcomplex is a catalyst of the translocation process. The catalysis seems to be of an entropic nature, and so, in order to abolish the structural restrictions, the catalyst (EF-G) must be removed (by GTP cleavage) prior to the next step of the elongation cycle.

90

ALEXANDER S. SPIRIN

C. Translocation Models Im lying Energy Contribution of EF-G wit GTP (Translocation Driven by GTP)

K

Of all the above-mentioned facts (Section IV), the participation of GTP in the process of translocation was the first to be known. This created a powerful psychological stimulus for speculating on special energetics problems in the translocation process that were to be solved at the expense of the energy of GTP cleavage. Practically all the translocation models proposed since then implied that the energy of EF-G-mediated GTP-cleavage is utilized, in one way or another, for the active displacement or at least for the active “expulsion” of the ribosomal ligands from their binding sites. Below, the models of this group are classified according to a conceivable point of primary application of the GTP-fed force of EF-G. 1. Utilizing the energy of GTP, EF-G shifts mRNA relative to the ribosome by one triplet from the 5’-end toward the 3‘-end. As a consequence, the peptidyl-tRNA fixed by the anticodon in the A site is displaced and driven into the P site by its codon, while the deacylated tRNA in the P site is dragged out by its codon from the P site to the exterior. According to this model, translocation cannot proceed in the absence of the template polynucleotide and codon-anticodon interaction [see, e.g., (98)]. Models of such a kind where the driving role belongs to the displacement of mRNA contradict the facts expounded in Section IV, such as Facts 6 (the driving role of the anticodon and the passive one of the template), 7 (translocation in the absence of the template), and 10 (increased translocation in the case of a weakened codon-anticodon interaction in the A site); they also meet difficulties in explanation of Facts 3 (factor-free translocation) and 11 (necessity for the A site to be occupied). At present the models where mRNA is the application point of the force developed by EF-G with GTP are not popular. 2. GTP energy through EF-G is applied to peptidyl-tRNA occupying the A site so that the developed force shifts it together with its codon towards the P site [see, e.g., (19,99,100). As a result the mRNA, carried along (driven) by peptidyl-tRNA, is also displaced thus helping to oust the deacylated tRNA from the P site. Strictly speaking, there are no facts whatsoever contradicting this class of models. Moreover, this is the only model of the given group that explains well Facts 10 (increased translocation at a weakened retention of tRNA in the A site) and 11 (necessity for the A site to be occupied for the translocational removal of tRNA from the P site). The greatest difficulty arises in interpreting Fact 3 (factor-free translocation) mentioned in Section


91

IV. However, this fact (slow factor-free translocation) can be explained by the spontaneous passage of the peptidyl-tRNA from the A site to the P site in accordance with the higher affinity for the latter, provided there is a slow spontaneous dissociation of the deacylated tRNA from the P site into solution. 3. GTP energy is realized by EF-G primarily for the removal of deacylated tRNA from the P site. Then peptidyl-tRNA easily passes from the A site to the vacant P site for which it has a higher affinity carrying over with itself its cognate codon and thus moving the mRNA along the ribosome by one triplet. This model seems to be an elegant one, as it creates the impression of being the simplest explanation of the translocational events proceeding from the thermodynamic parameters of peptidyl-tRNA binding with the ribosomal sites. Perhaps, for this reason it is now the most popular ( 1 1 , 20, 84, 101). However, the simplicity here is illusionary, as the mechanism of conjugation of the EF-G-mediated GTP cleavage with the removal of deacylated tRNA from the P site is just as unclear as that in the case of the (EF-G)*GTP-driventransport of peptidyl-tRNA from the A site to the P site. Besides, this model does not explain Fact 11 (necessity of occupation of the A site for the deacylated tRNA to be pushed out from the P site). 4.The force developed by EF-G at the expense of GTP energy is applied to parts of the ribosome itself to shift them relative to each other, and this mechanical change of the ribosome is, in its turn, the driving mechanism for the ligand displacements. Such an idea in a general form was first mentioned by Lipmann and co-workers (43, 102), and the mechanistic model was later proposed by the author of this survey (103-105). A combination of this idea with any of the above-mentioned classes of models is possible, depending on the point thought to be directly conjugated with the ribosomal change, such as the shift of the mRNA, the transport of peptidyl-tRNA, or the explusion of deacylated tRNA. In the original version of my model it was presumed that some drawing apart of the ribosomal subunits drives the displacement of peptidyl-tRNA, which, in its turn, drags the template and thus pushes out the deacylated tRNA (103-105).

D. Translocation Models Implying All Displacements to Be Thermodynamically Spontaneous (Translocation Catalyzed by EF-G) In the translocation models described above, it is assumed that the role of EF-G with GTP is, in one way or another, active and consists in the direct “pulling” or “pushing” of tRNA ligands of the ribosome.

92

ALEXANDER S. SPIRlN

However, since the contribution of EF-G with GTP in promoting translocation must be considered only as a catalytic one, the models where thermal motion is used for ligand displacements seem more likely, provided EF-G with GTP creates special conditions for such diffusion. Two following models of the catalytic contribution of EF-G with GTP to translocation (which do not necessarily exclude each other and may be complementary) can be considered. 1. Attachment of EF-G with GTP to the ribosomal complex decreases the affinity of the tRNA ligands for their binding sites and, in particular, the affinity of peptidyl-tRNA for the A site (whereas EF-Tu with GTP increases the affinity of aminoacyl-tRNA for the A site). Such a change of affinity should be considered as catalysis since it is temporary and takes place only in such an intermediate complex as the ribosome.(EF-G)*GTP,which rapidly breaks down upon GTP hydrolysis. It is evident that diffusion displacements of tRNA ligands during the existence of this intermediate complex with GTP will be facilitated. The decrease of tRNA affinity for the ribosome by EF-G (and the increase of tRNA affinity for the ribosome by EF-Tu) was claimed by Holschuh and Gassen (84). Such a model is also consistent with the fact that the decrease of Mg2+concentration, known to lower the tRNA affinity for both sites of the ribosome, promotes translocation and can even mimic the effect of EF-G with GTP (see Fact 4 in Section IV). In addition, it is in accord with the fact that translocation is stimulated by a modification of the 3 0 4 ribosomal subunit by pchloromercuribenzoate (IOS), which lowers the affinity constant of tRNA binding to the 30-S subunit by an order of magnitude (107). 2. Attachment of EF-G with GTP fixes a certain “unlocked” state of the ribosomal complex where a greater freedom is allowed for diffusion movements of tRNA ligands, within the limits assigned by the construction of the ribosome. In essence, this model represents just a minor modification of the model of locking-unlocking subunits proposed earlier (103-105); here the thermal unlocking with a fixation of the unlocked state by (EF-G)-GTPreplaces the previous (EF-G)*GTPdriven unlocking. This is the only model explaining such facts as “rearrangement” of the fast-reversible state of tRNA binding into the slowly-reversible one without a change of affinity (see 34), or the occluded (“irreversible”) state of tRNA binding on the ribosome [Fact 9 in Section IV; see also (81,87)].According to this model, an equilibrium exists between the locked and unlocked states of the ribosome; the ribosome fluctuates between these two states. (EF-Tu)*GTPmust fix the unlocked state of the ribosome introducing the aminoacyltRNA to the A site (reversible binding, selection stage, “R site” bind-


93

ing state); after GTP cleavage the affinity of EF-Tu for the ribosome and the aminoacyl-tRNA changes, providing the latter with one more chance to dissociate into the medium from the unlocked state of the ribosome (reversible binding, “correction” stage); only after the release of (EF-Tu).GDP is the unlocked state of the ribosome not fixed and it passes into the predominantly locked state where the aminoacyl-tRNA is occluded (final A-site binding state). Transpeptidation proceeds in the locked state of the ribosome. The pretranslocation state of the ribosome is also mainly locked, but some fraction of the unlocked state provides chances for factor-free translocation. (EFG).GTP, as already mentioned, again fixes the unlocked state where diffusion displacements of the ligands, called translocation, take place. After the release of EF-G with GDP, the ribosome again fluctuates between the two states. It is reasonable that in the case of one tRNA ligand per ribosome (posttranslocation state) the equilibrium between the locked and unlocked states is shifted more towards the unlocked one than in the case of two tRNA ligands per ribosome (pretranslocation state).

VII, Kinematics of Translocation

A. Mutual Orientation of the Two tRNAs Two tRNA molecules bound at the A and P sites of the ribosome are interacting with two adjacent codons of mRNA (Fact 5 in Section IV). This means that the anticodons of the two tRNAs are immediate neighbors in the pretranslocation-state ribosome. At the same time, the orientation of two tRNA molecules in the ribosome should allow their acceptor ends to be brought into close proximity in order to provide for transpeptidation reaction. Immediately after transpeptidation, in the pretranslocation ribosome, the free 3‘-end hydroxyl of the deacylated tRNA (P site) should be in the vicinity of the newly formed (tRNA-proximal) peptide bond of the peptidyl-tRNA (A site). Stereochemical considerations, taking the above requirements into account, urge one to exclude the side-by-side parallel orientation of the two tRNAs. (If the parallel orientation of two L-shaped tRNAs were the case, it would be impossible to bring the anticodons into properly close contact, no more than the acceptor ends, without serious distortions of the tRNA conformation known from crystallography.) Hence, the two tRNA molecules in their L-shaped conformation should be positioned in such a way that the planes containing both

94

ALEXANDER S. SPIRIN

FIG.6. Plausible mutual orientation of two ribosome-bound tRNA molecules represented as ribbon-drawn models. The anticodons (up) are immediate neighbors on the mRNA chain, the acceptor ends (down) are brought together as well, while the corners are apart.

limbs of the molecule form an angle with each other, while the apexes of both limbs, i.e., the anticodons and the acceptor ends, are brought into immediate proximity (Fig. 6) (see 108, 109). Messenger RNA joins the anticodons of the two tRNAs together. As the messenger is read out from the 5’ to the 3’ end and the anticodons are paired with codons in an antiparallel fashion, the very top nucleotide residue (the first anticodon nucleotide) of the P site tRNA is adjacent to the third anticodon nucleotide of the A site tRNA (Fig. 7 ) . In other words, the anticodon of tRNA in the P site (deacylated tRNA in the pretranslocation ribosome) should be positioned under the anticodon of tRNA in the A site (peptidyl-tRNA in the pretranslocation ribosome) when the tRNAs are viewed sideways with the anticodons up and the acceptor ends down (see Figs. 6 and 7). Recent fluorescent-energy-transfer measurements have shown that the distance between the anticodon loop of the A site tRNA and the D loop of the P site tRNA is somewhat longer than the distance between the anticodon loop of the P site tRNA and the D loop of the A site tRNA (110). This was interpreted as an indication of an asymmetric


95

FIG.7. Schematic representation of some deductions concerning the mutual arrangement of two ribosome-bound tRNA molecules: position of the P site tRNA anticodon under the A site tRNA anticodon (anticodon loops are squared), position of the acceptor end of the A site tRNA above the acceptor end of the P site tRNA (acceptor ends are indicated by arrows), approximately right angle between the planes of the two tRNAs (the corner of the A site tRNA faces the viewer whereas that of the P site tRNA is directed to the left), clockwise position of the P site tRNA relative to the A site tRNA (when viewed from the anticodons).

orientation of the two tRNAs, shifting the anticodon loops against each other. Indeed, this result is fully consistent with the deduction made above. It suggests that the P site tRNA as a whole is shifted down (if the anticodons are viewed up) relative to the A site tRNA along the line lying approximately through the anticodons and the acceptor ends. The shift of the tRNAs mentioned above also fulfils the requirement of the closest approach to the attacking amino group of the aminoacyl-tRNA to the attacked ester group of the peptidyl-tRNA prior to transpeptidation and, correspondingly, a resultant proximity of the newly formed peptide group of the peptidyl-tRNA to the free 3' hy-

96

ALEXANDER S. SPIRIN

droxyl of the deacylated tRNA after transpeptidation. When the tRNA anticodons are viewed up, the A site tRNA acceptor end is found to be somewhat above the P site tRNA acceptor end (Fig. 7). As rotations around five internucleoside bonds between two codons are possible in principle, there are no stereochemical limitations concerning the angle between the two tRNA planes, within the range from about 40” to 320” (0+40 is not allowed because of strong steric overlapping of the atoms of the tRNA molecules). From different considerations and indirect experimental data, some variants were proposed as more likely than others. Rich (108)was the first to discuss the right angle between the A site tRNA and the P site tRNA (+90” when counted clockwise from the A site to the P site, anticodons to the viewer). However, Sundaralingam et al. (109)applied a more detailed stereochemical analysis and arrived at the conclusions that the most favorable angles between the planes of the tRNA molecules are in the region of either +240 or +300” (when counted clockwise, as above); this is also close to the right angle between the tRNA planes but with the opposite sign (-90?30”, i.e., counterclockwise). The coplanar orientation of the two tRNAs (the angle between tRNAs being of about 180”) was excluded by experiments of Johnson et al. (111), who showed that the distance between the tRNA corners (between s4U8conjugated fluorescent dyes) is of about 30 A or shorter. The results of Wintermeyer’s group (110) evaluating the distance between the D loops of the two tRNAs of about 35 A were in full accordance with the above and also excluded both coplanar and side-by-side orientations of the molecules. Ofengand et al. (112-114) obtained results interpreted as consistent with both the “clockwise” and “counterclockwise” angle arrangements of the two tRNAs on the ribosome. However, determination of the handedness (chirality) of the asymmetric 30-S ribosomal particle urged them to incline to the “counterclockwise” position of the P site tRNA relative to the A site tRNA. On the other hand, comprehensive stereochemical analysis of the transpeptidation reaction made recently by Lim and Spirin (115) has led to the conclusion that the attacking amino-acid residue of the aminoacyl-tRNA (A site) should be positioned relative to the attacked ester group of the peptidyl-tRNA (P site) in such a way that a-helical orientation of two COOH-terminal peptide groups results from transpeptidation. The axis of the a-helix generated in the peptidyltransferase center of the ribosome has been shown to lie approximately on the line connecting the anticodons with the acceptor ends. Since the a-helix is a right-handed conformation, its formation by addition of a C-terminal amino-acid residue can be imagined only if the right-


97

handed mutual orientation of the A site and the P site tRNAs is assumed. This implies the “clockwise” position of the P site tRNA relative to the A site tRNA (Fig. 7). Moreover, the above suggests the angle between the planes of the two tRNAs to be close to +loo” (in accordance with the axial angle between amino-acid residues in the

a-helix). An additional argument in favor of the “clockwise” arrangement of the A and P site tRNAs can be drawn from stereochemical consideration of the rotational movements of the tRNA molecule during translocation (see below). The point is that the clockwise rotation of the tRNA molecule around the axis lying from the anticodon region to the acceptor end area by the force applied, say, to the corner will operate with a rigid molecule conformation, whereas the opposite (counterclockwise) way of rotation will contribute to unwinding the molecule and its right-handed helices (Lim, unpublished). The movement of tRNA as a rigid body seems to be a necessary requirement for translocational displacements.

B. Helical Displacement of Peptidyl-tRNA Translocation results in the displacement of the peptidyl-tRNA from the A site to the P site. Since the anticodons as well as the acceptor ends of the two tRNAs are together and the corners are apart, i.e., the plane of the P site tRNA is at a certain angle relative to the A site tRNA plane, the displacement should include a rotational operation (108). Evidently, the rotation should proceed around the axis lying approximately on the line connecting the anticodon region with the acceptor end. At the same time, as the P site tRNA is shifted relative to the A site tRNA along the same axis, the displacement should also include a translational operation. Hence, the translocation of tRNA from the A site to the P site looks like a helical displacement. (Of course, the real trajectory of the movement may deviate from the helical one but in any case the pretranslocation and the posttranslocation positions of the peptidyl-tRNA are related to each other by a helical displacement operation.) The translocation involves simultaneous movements of the tRNA residue, its bound mRNA codon, and the peptidyl residue. It seems unlikely that different parts of this (peptidyl-tRNA)*codon complex are forced and moved independently; they are rather to be pushed or pulled as a rigid whole, with just one main point or force application (say, the corner of tRNA might be such a point). Pushing or pulling of a nonrigid body would create a lot of mechanical problems. It is important to repeat once again that a tRNA molecule, including the stacked

98

ALEXANDER S. SPIRIN

4 FIG.8. Ball-and-rod drawing of the 3'-terminal ribose ring and COOH-terminal ahelical turn of the peptide residue of the peptidyl-tRNA. The ribbon helix is a schematic representation of an imaginary more distal part of the a-helix. The encircled A is the adenine of the 3'-teminal residue of the tRNA. Black balls are carbons, the white ones are nitrogens, and the shaded ones are oxygens; hydrogens are omitted. Pointed rods adjacent to Ca atoms indicate directions of Ca-Ca bonds of the peptide.

right-handed CC end, behaves as a rigid body when pushed clockwise (the anticodon to the viewer) and tends to unwinding when forced counter-clockwise around the axis connecting the anticodon and acceptor end regions (Lim, unpublished). If the (peptidyl-tRNA) . codon complex is helically displaced as a rigid body, all the parts of it in the pretranslocation and posttranslocation states should be related by the same helix parameters, i.e., the same helix axis, turning angle, and pitch. Assuming the a-helical conformation of the newly linked amino-acid residues adjacent to tRNA as a result of transpeptidation (115)(Fig. 8), the translocation should imply moving of the COOH-terminal end of the peptide in such a way that the situation for the a-helical orientation of the next amino acid residue is repeated in the following elongation cycle. Hence, the COOH-terminal end of the peptide should be moved by one aminoacid residue in the coordinates of the a-helix. This means that it will be rotated around the a-helix axis by 100" clockwise and translated along the same axis by 1.5A. The a-helix axis of the COOH-end of the nascent peptide lies on the line passing near the 3'-terminal ribose


99

ring of the tRNA to the anticodon region. If a rigid body movement is assumed, the displacement of the tRNA residue and the bound mRNA codon from the A site to the P site should proceed by 100” clockwise turn around the same axis with 1.5 A forward translation, when viewed from the anticodon. It is interesting to consider the translocational movement of an mRNA codon with the paired anticodon in a little more detail. The codon-anticodon pair seems to be a duplex with the parameters of the A form of double-stranded RNA. The length of the duplex along the RNA helix axis is about 7 to 8 A. Because of the incompatibility of this length with the distance of the translation shift of the amino-acid end of the tRNA (1.5 A), it is clear that the axis of the translocational movement cannot coincide with the codon-anticodon helix axis. Indeed, the direction of the anticodon helix axis in tRNA is at an angle to the line connecting the anticodon and acceptor end. As a result of translocation, the codon-anticodon duplex of the peptidyl-tRNA is displaced to the new position (P site), which, in terms of the model under consideration, should be related to the subsequent codon (A site) by a helical displacement operation, the rotation being 100” and the translation 1.5 A along the axis directed to the acceptor end. In such a case, from evident geometrical considerations, a kink should exist between the two codons; in other words, the two codon-anticodon duplexes cannot be coaxial and the helix axis of one is at an angle to the axis of the other. The translocation axis will pass through the point near the intercodon kink (Fig. 9). Setting the peptidyl-tRNA into the P site requires the preliminary or simultaneous evacuation of the deacylated tRNA. (From Fact 11 given in Section IV, it is reasonable to assume that the evacuation of the deacylated tRNA from the P site is initiated by the beginning of the (peptidyl-tRNA)*codoncomplex movement.) The release of the deacylated tRNA from the P site should be accompanied by tearing it away from its codon. The trajectory of the deacylated tRNA release from the P site is unknown but stereochemical considerations suggest that moving of its anticodon away from the mRNA codon should most probably proceed in the plane of the tRNA molecule (Lim, unpublished). Often, when considering the translocation of the peptidyl-tRNA from the A site to the P site, it is assumed that the deacylated tRNA is preliminarily evacuated from the P site, either spontaneously as a result of unblocking its exit (“opening the door”) or by some active force. In such a case, it would be very easy to imagine the fast spontaneous redistribution of the peptidyl-tRNA from the A site to the P site,

100

ALEXANDER S. SPIFUN

5’ anticodon Asite

translocot ion oxis

FIG.9. Schematic representation of the intercodon kink with the paired anticodons of the two tRNAs: view in the direction of the translocation axis (indicated by the encircled point).

in accordance with its greater affinity to the latter (see Fact 8 in Section IV). However, two facts, namely the necessity of the A site to be occupied for the release of tRNA from the P site (Fact 11, Section IV) and the acceleration of translocation as a result of weakened retention of tRNA in the A site (Fact 10, Section IV) suggest that the mechanism is more complicated. Anyhow, the conjugated displacement of the two tRNAs during translocation seems likely. It is not excluded that the two tRNA molecules in the ribosome interact with each other not only through mRNA but also by their own parts directly [see, e.g., (132)l.It can be speculated that the translocation starts with a simultaneous movement of the whole complex consisting of the two tRNAs and the template hexanucleotide, for example, by some turning clockwise; then the peptidyl-tRNA will continue the helical trajectory while the deacylated tRNA will dissociate on the trajectory lying in the plane of the molecule.


101

SIDE

FIG.10. Contour drawing of the 70-S ribosome and its t w o subunits in the overlap projection. The 3 0 4 subunit with its head up and the side bulge (“platform”)on the left is on the 50-S subunit with its central protuberance up and the rod-like appendage to the right.

C. Movements of tRNAs on the Ribosome Translocation is performed on the ribosome and by the ribosome. Therefore it is very important to localize the ribosomal sites where the events take place and to coordinate the models of translocation with structural characteristics of the ribosome. Unfortunately, however, direct experimental localization of tRNAs on the functioning ribosome has not yet been made. At the same time, the morphology of the ribosome has been studied in detail by electron microscopy, and now the agreement between different groups has been achieved concerning the shape, the asymmetry and the hand of the ribosomal subunits, as well as their mutual orientation in the 70-S ribosome (116-125). In Fig. 10 the contours of the two subunits of the 70-S ribosome in the overlap projection are shown. The linear dimensions of the two tRNAs complex are from one quarter to one third of those of the ribosome, so that it does not seem quite senseless to look for a noticeable pocket or cavity where the tRNAs could be placed. It is seen that such a pocket or cavity is formed by the concave edge of the 3 0 4 subunit and the uncovered surface of the 50-S subunit at the base of the rod-like appendage (L71 L12 stalk), i.e., on the right of the 3 0 4 subunit in the overlap projection (Fig. 10).This room can be nicely filled with two tRNAs (126). In Fig. 11 the contour of the side projection of the 70-S ribosome with

102

ALEXANDER S. SPIRIN

FIG.11. Contour drawing of the 7 0 4 ribosome in the side projection (view from the rod-like appendage) with two tRNA molecules tentatively positioned in the pocket at the base of the rod-like appendage. The A site is supposed to be on the right and the P site on the left.

two tRNAs placed in the cavity under consideration is depicted. The orientation of the two tRNAs relative to the ribosomal subunits is done in accordance with the well-known facts that the peptidyltransferase that should make contact with the acceptor ends is a property of the 50-S subunit (lower) whereas the mRNA and anticodon binding sites are located on the 30-S subunit (upper). If the translocation involves a clockwise displacement of peptidyl-tRNA (viewed from the anticodon), the A site tRNA should be on the right and adjoin the heads of the 30-S and 50-S subunits, whereas the P site tRNA will be on the

left. The location of the tRNAs on the ribosome proposed above is consistent with a number of experimental facts [for review see (I26)l. First of all, proteins L7/L12 constituting the rod-like appendage, as well as the adjacent area of the 50-S subunit surface are known to take part in binding and functioning of the elongation factors, EF-Tu and EF-G, which are directly involved in tRNA binding and translocation, respectively. Secondly, the 30-S subunit proteins, S3, S4, S5, S10, S14, S19 and some others localized on the concave edge, opposite to the side bulge [see also (125, 127, 128) have been reported to be in the vicinity of the ribosome-bound tRNAs and important for the binding. EF-G has been directly localized by immunoelectron microscopy on the uncovered surface of the 50-S subunit, at the base of the rodlike appendage (L7/L12 stalk) (129),and at the same time EF-G is found close to and interacting with the D-loop of the P site tRNA as recently shown by fluorescent technique (130). There is some evidence that the A site tRNA makes contact with the ribosome on its right side, when viewed from the corner, antico-


103

don up (112,113,131,132),while the left side of the L-shaped molecule is accessible for the interaction with EF-Tu (133, 134). At the same time, the P site tRNA seems to adjoin the ribosome by its left side (112, 113, 130, 132). This is consistent only with the clockwise way of translocation on the ribosome, from the right to the left tRNA position (Fig. 11). On the other hand, localization of tRNA at the other side of the 30S subunit, in the region of the side bulge or platform (on the left of the 30-S subunit in the overlap projection of the 70-S ribosome, Fig. 10) has been also argued (32, 112-114, 125, 135, 136). Thus, there are several observations giving evidence that codon-anticodon interaction occurs near the cleft between the side bulge (platform) and the head of the 30-S subunit. The strongest is that that anticodon of the P site tRNA can be directly cross-linked with the base in position 1400 of the 16-S RNA (137,138),which seems to be not far from the 3' end (position 1542) and from the two dimethylated adenines (positions 1518-1519) localized by immunoelectron microscopy on the top of the side bulge (platform) or between the side bulge and the head of the 30-S subunit (133-241 ), The immunoelectron microscopy of the hapten-carrying tRNA cross-linked through its anticodon suggested the same localization of the codon-anticodon site on the 30-S subunit (142).Besides, the 3'-end sequence of 16-S RNA is known to interact with the polypurine sequence of mRNA in close proximity (about 10 nucleotide residues upstream) to the initiation codon [see ( 1 4 3 ) ] .In addition, some proteins of the side bulge such as S6, S11 and S18 have been reported to be concerned in mRNA or tRNA binding [for review see (125)l.It should be mentioned, however, that all these facts can fit the alternative model as well, because the anticodon arms of the tRNAs can extend from the concave edge over the back of the 30-S subunit and reach the cleft under consideration. The most striking observation is that the A-site-bound tRNA crosslinked with the 30-S subunit through its corner (central part) is detectable by immunoelectron microscopy predominantly in the region of the 30-S subunit head adjacent to the side bulge (platform) and, to a lesser extent, in the region of the cleft between the head and the side bulge (114).The latter seems to be consistent only with the model of the side-bulge localization of the tRNA binding site. The most recent version of the model where the A site is on the left and the P site is on the right (114)implies, however, counterclockwise translocation of the peptidyl-tRNA (viewed from its anticodon) which looks less likely than the clockwise helical displacement (see the considerations above). Besides, studies with fluorescent derivatives of tRNA (132)

104

ALEXANDER S. SPIRIN

have shown that the fluorescence of the D-loop of the P site tRNA is affected by the tRNA binding to the A site; this rather suggests the Dloop of the P site tRNA to face the A site tRNA, i.e., the clockwise way of translocation. Of course, the information about the exact site of the peptidyltransferase on the 50-S subunit would be very useful for choosing between the one or the other model of the tRNA binding site position. The experiments with affinity labeling of the ribosome by puromycin derivatives demonstrated their cross-linking with the protein L23 (136, 144,145),which was previously localized by immunoelectron microscopy on the left of the central protuberance (head) of the 50-S subunit (viewed from the interface, as in Fig. 10) (146).However, more recently the protein L23 was localized in a quite opposite region, near the bottom (if viewed as in Fig. lo), on the external side of the subunit (147).Nevertheless, the immunoelectron microscopists still claim that the positions of puromycin which is a substrate of the peptidyltransferase, and chloramphenicol, which is a specific inhibitor of the enzymatic center, to be on the left side of the 50-S subunit (viewed as in Fig. lo), either directly on the left protuberance (L1 ridge) in the case of puromycin, or in the region between the central and the left protuberances in the case of chloramphenicol(123,124,136).On the other hand, several mutations affecting the peptidyltransferase center have been identified as changes in the phylogenetically conserved region of 23-S RNA corresponding to nucleotide positions from 2447 to 2504 of the E . coli RNA (see (148)). Direct cross-linking of the photoactivated label-carrying acceptor end of tRNA was achieved with position 2584 (149), which is in the immediate neighborhood with the above The above region is region on the secondary structure map (see (148)). in the vicinity of the ribosomal protein L6 cross-linkable with the 23-S RNA positions 2473-2481 (150). As the protein L6 is located at the base of the rod-like appendage (151) and, correspondingly, neighbors with the proteins L7/L12 (152)as well as with the ribosome-bound EF-G (153),the position of the peptidyltransferase center should be somewhere in the area under consideration, i.e., near to the rod-like appendage base of the 50-S subunit and, hence, under the concave edge of the 30-S subunit. The fluorescent energy-transfer measurements of the distances between the 3' end of 16-S RNA and the A site and the P site bound tRNAs (154)have demonstrated that the anticodons are removed from the top of the side bulge of the 30-S subunit by more than 60 A, and the central part of the A site tRNA from the same region of the ribosome by about 70 to 90 A, the distance to the derivativized acp3U47


105

[acpW = 3-(3-amino-3-carboxypropyl)uridine]on the left side of tRNA (if viewed from the corner, anticodon up) being longer than to the inner part of the corner. The distances measured seem to be too big for the model with the side bulge localized tRNAs and more consistent with the model where the tRNAs are at the concave edge of the 30-S subunit.

VIII. Are Conformational Movements of the Ribosome Required for Translocation? As long as only thermodynamic problems of translocation are considered it is not necessary to draw attention to any kind of dynamic features of ribosome structure. Indeed, all the translocational movements could be explained as ribosome-restricted (chanelled) diffusionary displacements of the ligands. However, such problems as fast association and dissociation of the large ligands involving multicenter binding, asymmetry of the displacements and entropic catalysis of the ligand displacements within the ribosome require knowledge both of active center conformations and their possibilities to change during functioning. Large-block mobility of the ribosome could play an important role in resolving the dynamic problems mentioned. (Largeblock mobility is understood here as thermal relative shifts of the ribosomal subunits and their major domains in a limited range of both distances and directions of movement. They can be very small.) 1.When such large molecules as tRNA interact with spread ribosomal sites having several binding points (multicenter binding), high kinetic barriers are inevitable during their association and dissociation. The utilization of thermal motion of large ribosomal blocks sharing a binding site would help to overcome these barriers. Indeed, on the one hand, the Brownian movement of large blocks can provide a pathway for step-wise rupture (and formation) of the numerous contacts of the ligand with its site. On the other hand, as the blocks under consideration are not independent, but coupled with each other in a definite manner, the fluctuations along some degrees of freedom will be limited, which can channel the movements along the rest of the degrees of freedom in favor of a given step (say, translocation). 2. I t is likely that the displacement of the tRNA ligand from one site to the other does not proceed by its complete breakaway from one site and subsequent reassociation with the other, but rather a “handto-hand” transition. If binding is of a multi-center nature such a transition is difficult to imagine without a structural flexibility of the corresponding ribosomal region.

106

ALEXANDER S. SPIRIN

3. Fixation and defixation of the bound ligand without a change of affinity seems to be possible only if a mechanism of occlusion of the ligand between the ribosomal blocks or some kind of “valve mechanism” exists. In both cases, relative shifts, even if small, of parts of the ribosomal structure seem to be required. 4.The mechanism of entropic catalysis by the elongation factors could be imagined as the fixation of a defined confonnational substate of the ribosome attained as a result of thermal motion, or as a specific limitation and channelling of the intraribosomal large-block Brownian movement. The following commonly known facts concerning the construction of the ribosome and its functional sites should be recalled: 1. The ribosome consists of two large separable subparticles, or subunits (the 3 0 4 and 5 0 3 for prokaryotic ribosomes, or 40-S and 60S for eukaryotic ribosomes). 2. The 30-S (or 40-S) subunit carries mRNA and, consequently, the binding sites for both anticodons of the t w o tRNA molecules. 3. The 50-S (or 60-S) subunit carries the peptidyltransferase center, and, consequently, the binding sites for both reacting groups of the two substrate tRNAs, namely peptidyl-tRNA and aminoacyl-tRNA. The question arises: For what purpose has nature placed one part of each substrate (and product) molecule on one subunit and the other part on the other subunit? Why is the boundary between the large ribosomal blocks the place where the events of the elongation cycle, including translocation, are played out? Simple logic implies that the relative movement of the two subunits (and, perhaps, of the large blocks of each subunit) is somehow required for function. In 1968, I proposed the idea of the relative movement of the two In the proposed model a slight swingribosomal subunits (103-105). ing type movement of one subunit relative to the other provided, on the one hand, a final fixation in their sites and the bringing into close contact with each other of the two substrate ligands (upon drawing together, or locking) and, on the other, their coming out of contact with each other and release from their binding sites (upon drawing apart, or unlocking); it is just the drawing apart of the subunits that promoted translocation. In the same year, Bretscher (155)proposed the idea of a slipping type of movement of the two ribosomal subunits in order to explain the displacement of tRNA from the A site to the P site, not through a break off-reassociation but via an intermediate A-P “hybrid site.” Attempts experimentally to detect a change in the size and shape of the functioning ribosome have been made, mainly by comparing


107

the sedimentation coefficients of the pretranslocation and posttranslocation state ribosomes (29, 105, 156-159). In all cases, the pretranslocation state was characterized by a slightly greater sedimentation coefficient than that of the posttranslocation state, as was predicted by the model [drawing apart during translocation, see (l05)].However, later experiments have shown that the results of sedimentation experiments cannot be interpreted unambiguously, since sedimentation induces a number of artifacts due to shift of equilibrium under conditions of component separation and high hydrostatic pressure (159-163). Besides, there is also an inevitable difficulty in interpreting any slight difyerences observed by physical methods: the pretranslocation ribosome contains two tRNA molecules while the posttranslocation ribosome has one, and this can contribute to the measured physical parameters of the ribosome, irrespective of changes in its compactness. Recently the latter difficulty was overcome by use of the neutron scattering technique in various mixtures of light and heavy water; in a proper mixture the nucleic acid component of the ribosome can be made “invisible” for neutrons, thus allowing the compactness of the particle only from its protein component to register. It turned out that the radius of gyration of the protein component of the posttranslocation ribosome is slightly, but reliably, greater than that of the pretranslocation ribosome, i.e., translocation really makes the ribosome slightIy less compact (164).The next problem to be solved is whether the decrease in the ribosome compactness upon translocation reflects an intersubunit or intrasubunit change. In the first case, it would mean that there is a physical unlocking (swinging apart) of the two ribosomal subunits. If an intrasubunit change is the case, some interdomain movements within one or the other subunit will be suspected. In addition to the possibility of some movement of the two ribosomal subunits relative to each other, mobility of the rod-like appendage (L71L12 stalk) ofthe 50-S subunit also deserves attention in relation to translocation. It has been demonstrated by NMR that this part is not fixed, as are all others, in the entire ribosomal structure, but seems to be very flexibly connected with the rest of the ribosome (165).The only moment when it becomes fixed is the state of the ribosome with the EF-G (plus noncleavable GTP analog) attached (166).I would like to speculate that the mobility of the rod-like appendage (L71L12 stalk) relative to the ribosome is destined to facilitate the release of deacylated tRNA from the P site during translocation: in the model drawn schematically in Fig. 11 the rod-like appendage (directed to the viewer) looks like a movable guide for sliding the P site tRNA in the

108

ALEXANDER S. SPIRIN

plane of the molecule (as suggested from stereochemical considerations).

IX. Concluding Remarks The problem of ribosomal translocation is very difficult and complicated because it is a part of a more general problem of macromolecular movements (dynamics) where direct approaches are practically absent. There are just a few direct facts, much more indirect data, and a lot of models. Here I have tried to classify the models concerning different aspects of ribosomal translocation and to discuss them in the light of today's factual grounds. First of all, the two-tRNA-binding-sites (A and P) model for the ribosomal elongation cycle is briefly considered and stated to be the most substantiated basic scheme of the sequence of events. Intermediate and additional tRNA binding positions may exist without detriment to the main scheme. A three-tRNA-binding-site model where the sequence of events differs from the above is also mentioned. The energetics models of ribosomal translocation deserved special attention because of great confusion in this field. They are classified into two groups. One group includes the models where ribosomal ligands, in particular tRNAs, are actively pushed or pulled, mechanically or electrostatically, from one position to another at the expense of the energy of the coupled GTP hydrolysis (GTP-driven translocation). The other group of models implies that the translocational displacements are thermodynamically spontaneous and proceed as ribosome-channeled diffusional movements of the ligands, the role of EF-G being considered as an entropic catalyst and GTP as a cleavable effector of EF-G (EF-G-catalyzed translocation). The preference is given to the second group of models. The kinematic and topographical models of ribosomal translocation are also discussed. As the A site and P site tRNAs should have their anticodons as well as their acceptor ends together, the first question is about the angle between the planes of the two L-shaped tRNAs. Most evidence and general considerations are in favor of an approximately right angle. However, since translocation of tRNA from the A site to the P site is a helical displacement operation, the turn could be either clockwise or counter-clockwise (if viewed from the anticodon); correspondingly, there are two groups of models. The clockwise turn by about 90-100" from the A site to the P site seems more likely. The third question concerns the localization of the tRNA binding sites and

109


tRNAs on the ribosome. Agains there are two major proposals: either the A site and P site tRNAs are positioned near the side bulge (platform) of the 30-S subunit and the left protuberance (L1 ridge) of the the 50-S subunit, or they are localized on the other side, in the cavity formed by the right concave edge of the 30-S subunit and the base of the rod-like appendage (L7/L12 stalk) of the 50-S subunit (the 70-S ribosome is considered here as in Fig. 10 , in the overlap projection, when the 30-S subunit is lying on the 50-S subunit). I believe that the second one is better grounded. Finally, the question about conformational movements (largeblock mobility) of the ribosome itself during translocation is raised. The possibility of a slight intersubunit movement of the swingingapart type is considered. Attention is also paid to the mobility of the rod-like appendage (L7/L12 stalk) of the 50-S subunit. I hope that the survey will contribute to further thinking and experimentation in studies of molecular mechanisms of protein synthesis.

ACKNOWLEDGMENTS I would like to express my deep gratitude to Alexander Chetverin, Alexander Girshovich, Valery Lim, Oleg Ptitsyn, and Evgeny Shakhnovich for detailed discussions.

REFERENCES 1. R. R. Traut and R. E . Monro,]MB 1 0 , 6 3 (1964). 2. R. Heintz, H. McAllister, R. Arlinghaiis and R. Schweet, CSHSQB 31,633 (1966). 3. N. Tanaka, T. Kinoshita and H. Masukawa, BBRC 30,278 (1968). 4. N. Brot, R. Ertel and H. Weissbach, BBRC 31,563 (1968). 5. A.-L. Haenni and J. Lucas-Lenard, PNAS 61, 1363 (1968). 6. L. Skogerson and K. Moldave, ABB 125,497 (1968). 7. L. Skogerson and K. Moldave,]BC 243,5361 (1968). 8. K. Moldave, W. Galasinski, P. Rao and J. Siler, CSHSQB 34,347 (1969). 9. R. W. Erbe, M. M. Nau and P. Leder,/MB 39,441 (1969). .to. P. Leder, A. Bernardi, D. Livingston, B. Loyd, D. Roufa and L. Skagerson, CSHSQB 34,411 (1969). [ I . P. Leder, Ado. Protein Chem. 27,213 (1973). 12. N. Brot, in “Molecular Mechanisms of Protein Biosynthesis” (H. Weisshach and S . Pestka, eds.), p. 375. Academic Press, New York, 1977. 13. L. Skogerson and K. Moldave,]BC 243,5354 (1968). 14. S. Pestka, PNAS 61, 726 (1968). 15. P. Leder, L. E. Skogerson and D. J. Roufa, PNAS 62,928 (1969). 16. L. Skogerson, D. Roufa and P. Leder, PNAS 68,276 (1971). 17. J. Lucas-Lenard and A.-L. Haenni, PNAS 63,93 (1969). 18. A. Kaji, K. Igarashi and H. Ishitsuka, CSHSQB 34, 167 (1969). 19. H. Ishitsuka, Y. Kuriki and A. Kaji,]BC 245, 3346 (1970).

110

ALEXANDER S. SPIRIN

D. J. Roufa, L. Skogerson and P. Leder, Nature 227,567 (1970). N. Inoue-Yokosawa, C . Ishikawa and Y. Kaziro,JBC 249, 4321 (1974). N. V. Belitsina and A. S. Spirin, EJB 94, 315 (1979). N. V. Belitsina, M. A. Glukhova and A. S. Spirin, Methods Enzymoi. 60,761 (1979). K. Holschuh, J. Bonin and H. G. Gassen, Bchem 19, 5857 (1980). K. Holschuh, D. Riesner and H. G. Gassen, Nature 293,675 (1981). R. W. Erbe and P. Leder, BBRC 31, 798 (1968). D. Roufa, B. P. Doctor and P. Leder, BBRC 39,231 (1970). S. S. Thach and R. E. Thach, PNAS 68, 1791 (1971). S. L. Gupta, J. Waterson, M. L. Sopori, S. M. Weissman and P. Lengyel, Bchen 10, 4410 (1971). 30. J. D. Watson, Bull. SOC. Chin. Bid. 46, 1399 (1964). 31. F. Lipmann, Science 164, 1024 (1969). 32. J. A. Lake, PNAS 74, 1903 (1977). 33. A. E. Johnson, R. H. Fairclough and C. R. Cantor, in “Nucleic Acid-Protein Recognition” (H. J. Vogel, ed.), p. 469. Academic Press, New York, 1977. 34. W. Wintermeyer and J. M. Robertson, Bchem 21,2246 (1982). 35. F. 0. Wettstein and H. Noll.JMB 11, 35 (1965). 36. H.-J. Rheinberger and K. H. Nierhaus, Biochem. Int. 1,297 (1980). 37. H.-J. Rheinberger and K. H. Nierhaus, PNAS 80,4213 (1983). 38. H.-J. Rheinberger, H. Sternbach and K. H. Nierhaus, €“AS 78,5310 (1981). 39. R. A. Grajevskaja, Y. V. Ivanov and E. M. Saminsky, EJB 128,47 (1982). 40. S. V. Kirillov, E. M. Makarov and Y. P. Semenkov, FEBS Lett. 157, 91 (1983). 41. J. M. Robertson, R. Lill and W. Wintermeyer, in “Proc. Vth Symp. on Metabolism and Enzymology of Nucleic Acids” (J. Zelinka and J. Balan, eds.), Slovak Academy of Sciences, Bratislava, 1984. 41a. W. Wintermeyer, R. Lill, and J. M. Robertson, in “EMBO Workshop on Function, Structure and Genetics of Prokaryotic and Eukaryotic Elongation Factors, ThiverVal-Gringnon (France), July 16-20, 1984.” Abstracts, p. 47. 42. A. S. Spirin, FEBS Lett. 165, 280 (1984). 43. Y. Nishizuka and F. Lipmann, ABB 116,344 (1966). 44. N. V. Belitsina, M. A. Glukhova and A. S. Spirin, FEBS Lett. 54, 35 (1975). 45. N. V. Belitsina, M. A. Glukhova and A. S. Spirin, JMB 108,609 (1976). 46. J. Modolell, T. Girb6s and D. VBzquez, FEBS Lett. 60, 109 (1975). 47. T. GirbBs, D. Vhquez and J. Modolell, EJB 67,257 (1976). 48. J. Gordon and F. Lipmann, JMB 23,23 (1967). 49. S. Pestka, JBC 243,2810 (1968). 50. S. Pestka, JBC 244, 1533 (1969). 51. S. Pestka, CSHSQB 34, 395 (1969). 52. L. P. Gavrilova and V. V. Smolyaninov, Mol. Bid. (USSR) 5, 883 (1971). 53. L. P. Gavrilova and A. S. Spirin, FEBS Lett. 17, 324 (1971). 54. E. Hamel, M. Koka and T. Nakamoto, JBC 247,805 (1972). 55. L. P. Gavrilova, 0. E. Kostiashkina, V. E. Koteliansky, N. M. Rutkevitch and A. S. Spirin, J M B 101,537 (1976). 56. A. S. Spirin, This Series 21,39 (1978). 57. N. V. Belitsina, L. P. Gavrilova, and A. S. Spirin, Dokl. Akad. Nauk SSSR 224,1205 (1975). 58. N. D e Groot, A. Panet and Y. Lapidot, EJB 23,523 (1971). 59. V. B. Odinkov and S. V. Kirillov, NARes 5,3871 (1978). 60. P. Wurmbach and K. H. Nierhaus, PNAS 76,2143 (1979). 20. 21. 22. 23. 24. 25. 26. 27. 28. 29.


111

61 M. Peters and M. Yarus,JMB 134,471 (1979). 62. R. Luhrmann, H. Eckhardt and G. Stoffler, Nature 280,423 (1979). 63. R. H. Fairclough, C. R. Cantor, W. Wintermeyer and H. G. Zachau, J M B 132,557 (1979). 64. R. H. Fairclough and C. R. Cantor, JMB 132,575 (1979). 65. S. V. Kirillov, K. Sh Kemkhadze, E. M. Makarov, V. I. Makhno, V. B. Odintsov and Yu. P. Semenkov, FEBS Lett. 120,221 (1980). 66. J. Ofengand and R. Liou, Bcheni 20, 552 (1981). 67. A. J . M. Matzke, A. Barta and E. Kuechler, PNAS 77, 5110 (1980). 68. D. L. Riddle and J. Carbon, Noture N B 242, 230 (1973). 69. M. Cannon, R. Krug and W. Gilbert, J M B 7,360 (1963). 70. I. Rychlik, BBA 114,425 (1966). 71. I. Rychlik, Coll. Czech. Chem. Commutz. 31, 2583 (1966). 72. M. Cannon, BJ 104,934 (1967). 73. Y. Takeda, I. Suzuka and A. Kaji,JBC 243, 1075 (1968). 74. N. D e Groot, A. Panet and Y. Lapidot, EJB 15,215 (1970). 75. R. A. Grajevskaja, E. M. Saminsky and S. E. Bresler, BBRC 46, 1106 (1972). 76. S. V. Kirillov, V. I. Makhno, V. B. Odinzov and Yu. P. Semenkov, EJB 89, 305 (1978). 77. Y. V. Ivanov, R. A. Grajevskaja and E. M. Saminsky, EJB 106,449 (1980). 78. N. V. Belitsina, G. Zh. Tnalina and A. S . Spirin, FEBS Lett. 131, 289 (1981). 78. N . V. Belitsina, G. Zh. Tnalina a n d A. S. Spirin, Biosystems 15, 233 (1982). 80. G. Zh. Tnalina, N. V. Belitsina and A. S. Spirin, Dokl. Akad. Nauk S S S R 266, 741 (1982). 81. Yu. P. Semenkov, E. M. Makarov, V. I. Makhno and S. V. Kirillov, FEBS Lett. 144, 125 (1982). 82. K. Igarashi and A. Kaji, EJB 14, 41 (1970). 8,3. R. Lill, J. M. Robertson and W. Wintermeyer, 7th EMBO Atinu. Symp., Ribosome Structure and Function, Poster Abstr., 1981. 81. K. Holschuh and H. G. Gassen, JBC 257, 1987 (1982). 8.5. M. Schmitt, U. Neugebauer, C. Bergmann, H. G. Gassen and D. Reisner, EJB 127, 525 (1982). 86. S. V. Kirillov and Yu. P. Semenkov, F E B S Lett. 148, 235 (1982). 87. R. C. Thompson, D. B. Dix, R. B. Gerson and A. M. Karim,JBC 256,81 (1981). 88. M. J . Cabaxias and J. Modolell, Bchem 19, 5411 (1980). 89. J . Modolell and D. Vazquez, EJB 81,491 (1977). 90. M. J. Cabarias, D. VBzquez and J. Modolell, EJB 87, 21 (1978). 91. M. J. Cabaxias, D. VBzquez and J. Modolell, BBRC 83,991 (1978). 92. S. Campuzano, M. J. Cabahas and J . Modolell, EJB 100, 133 (1979). 93. Yu. P. Semenkov, V. I. Katunin, E. M. Makarov and S. V. Kirillov, FEBS Lett, 144, 121 (1982). 94. J. Modolell, B. Cabrer and D. Vizquez, €“AS 70,3561 (1973). 94a. E. F. Gale, E. Cundliffe, P. E. Reynolds, M. H. Richmond and M. J. Waring, i n “The Molecular Basis of Antibiotic Action,” 2nd ed., pp. 402-547 . Wiley-Interscience, New York, 1981. 95. J. W. Bodley, F. J. Zieve and L. Lin, JBC 245,5662 (1970). 56. J. D. Watson, “Molecular Biology of the Gene,” 3rd ed. Benjamin, Menlo Park, California, 1976. 57. A. B. Chetverin and A. S. Spirin, BBA 683, 153 (1982). 98. H. Weissbach, B. Redfield and N. Brot, ABB 127, 705 (1968).

112

ALEXANDER S. SPIRIN

99. B. Hardesty, W. Culp and W. McKeehan, CSHSQB 34, 331 (1969). 100. J. W. Bodley and L. Lin, Nature 227, 60 (1970).

101. K. Holschuh and H. G. Gassen, FEBS Lett. 110, 169 (1980). 102. T. W. Conway and F. Lipmann, PNAS 52, 1462 (1964). 103. A. S. Spirin, Dokl. Akad. Nauk SSSR 179, 1467 (1968). 104. A. S. Spirin, Curr. Mod. B i d . 2, 115 (1968). 105. A. S. Spirin, CSHSQB 34, 197 (1969). 106. L. P. Gavrilova and A. S. Spirin, FEBS Lett. 22, 91 (1972). 107. N. V. Belitsina, M. A. Glukhova and A. S. Spirin, Dokl. Akad. Nuuk S S S R 216,925 (1974). 108. A. Rich, in “Ribosomes” (M. Nomura, A. TissiCres and P. Lengyel, eds.), pp. 871884. Cold Spring Harhor Laboratory, New York, 1974. 109. M. Sundaralingam, T. Brennan, N. Yathindra and T. Ichikawa, in “Structure and Conformation of Nucleic Acids and Protein-Nucleic Acid Interactions” (M. Sundaralingam and S. T. Rao, eds.), pp. 101-115. University Park Press, Baltimore, 1975. 110. H. Paulsen, J. M. Robertson and W. Wintermeyer,JMB 167,411 (1983). I l l . A. E. Johnson, H. J. Adkins, E. A. Matthews and C. R. Cantor,]MB 156,113 (1982). 112. J. Ofengand, in “Ribosomes, Structure, Function and Genetics” (G. Chambliss, G. R. Craven, J. Davies, K. Davis, L. Kahan and M. Nomura, eds.), pp. 497-529. University Park Press, Baltimore, 1980. 113. J. Ofengand, F.-L. Lin, L. Hsu and M. Boublik, in “Molecular Approaches to Gene Expression and Protein Structure” (M. A. Q. Siddiqui, M. Krauskopf and H. Weissbach, eds.), pp. 1-31. Academic Press, New York, 1981. 114. F.-L. Lin, M. Boublik, and J. Ofengand,JMB 172, 41 (1984). 115. V. I. Lim and A. S. Spirin, Dokl. Akad. Nauk S S S R 280,235 (1985). 116. V. D. Vasiliev, Acta Bid.Med. Gem. 33, 779 (1974). 117. J. A. Lake,JBM 105, 131 (1976). 118. J. A. Lake,]MB 129, 155 (1979). 119. J. A. Lake,]ME 161,89 (1982). 120. M. Boublik and W. Hellmann, PNAS 75,2829 (1978). 121. V. D. Vasiliev, 0. M. Selivanova, V. I. Baranov and A. S. Spirin, FEBS Lett. 155, 167 (1983). 122. V. D. Vasiliev, 0. M. Selivanova and S. N. Ryazantsev,JMB 171,561 (1983). 123. G. Stofller and m. Stofller-Meilicke. in “Modem Methods in Protein Chemistry” (H. Tschesche, ed.), pp. 409-457. Walter d e Gruyter Verlag, Berlin and New York, 1983. 124. G. StiifRer and M. StofRer-Meilicke, Annu. Rev. Biophys. Bioeng. 13,303 (1984). 125. J. A. Lake, This Series 30, 163 (1983). 126. A. S. Spirin, FEBS Lett. 156,217 (1983). 127. L. Kahan, D. A. Winkelman and J. A. Lake, J M B 145, 193 (1981). 128. V. Ramakrishnan, M. Capel, M. Kjeldgaard, D. M. Engelman, and P. B. Moore, ] M B 174,265 (1984). 129. A. S. Girshovich, T. V. Kurtskhalia, Yu. A. Ovchinnikov and V. D. Vasiliev, FEBS Lett. 130,54 (1981). 130. J. M. Robertson, in “EMBO Workshop on Function, Structure and Genetics of Prokaryotic and Eukaryotic Elongation Factors, Thiverval-Grignon (France), July 1 6 2 0 , 1984.” Abstracts, p. 63. 131. L. M. Hsu, F.-L. Lin, K. Nurse and J. Ofengand, ] M B 172,57 (1984).


113

132. J. M. Robertson, H. Paulsen and W. Wintermeyer, in “Proc. IVth Symp. on Metabolism and Enzymology of Nucleic Acids” (J. Zelinka and J. Balan, eds.), pp. 349360. Slovak Academy of Sciences, Bratislava, 1982. 133. F. P. Wikman, G. E. Siboska, H. U. Peterson and B. F. C. Clark, EMBOJ 1, 1095 (1982). 134. T.-H. Kao, D. L. Miller, M. Abo and J. Ofengand,JMB 166,383 (1983). 135. J. A. Lake, in “Ribosomes, Structure, Function and Genetics” (G. Chambliss, G. R. Craven, J. Davies, K. Davis, L. Kahan and M. Nomura, eds.), pp. 207-236. University Park Press, Baltimore, 1980. 136. H. M. Olson, P. G. Grant, B. S. Cooperman and D. G. Glitz,JBC 257,2649 (1982). 137. J. B. Prince, B. H. Taylor, D. L. Thurlow, J . Ofengand and R. A. Zimmennann, PNAS 79,5450 (1982). 138. C. Ehresmann, B. Ehresmann, R. Millon, J.-P. Ebel, K. Nurse and J. Ofengand, Bchem 23,429 (1984). 139. I. N. Shatsky, L. V. Mochalova, M. S. Kojouharova, A. A. Bogdanov and V. D. Vasiliev, JMB 133,501 (1979). 140. H. M. Olson and D. G. Glitz, PNAS 76,3769 (1979). 141. S. M. Politz and D. G. Glitz, PNAS 74, 1468 (1977). 142. M. Keren-Zur, M. Boublik and J. Ofengand, PNAS 76, 1054 (1979). 143. J. A. Steitz, in “Ribosomes, Structure, Function and Genetics” (G. Chamhliss, G. R. Craven, J . Davies, K. Davis. L. Kahan and M. Noninra, eds.), pp. 479-495. University Park Press, Baltimore, 1980. 144. P. G. Grant, W. A. Strycharz, E. N . Jaynrs, Jr., and B. S. Cooperman, Bchem 18, 2149 (1979). 145. A. W. Nicholson, C. C. Hall, W. A. Strycharz and B. S. Cooperman, Bchem 21,3797 (1982). 146. G. Stoffler, R. Bald, B. Kastner, R. Liihrmann, M. Stoffler-Meilicke and G. Tischendorf, in “Ribosomes, Structure, Function and Genetics” ( G . Chambliss, G . R. Craven, J. Davies, K. Davis, L. Kahan and M. Nomura, eds.), pp. 171-205. University Park Press, Baltimore, 1980. 147. M. Stoffler-Meilicke, M. Noah and G . Stoffler, PNAS 80, 6780 (1983). 148. H. F. Noller, J. Kop, V. Wheaton, J . Brosius, R. R. Gritell, A. M. Kopylov, F. Dohme, W. Herr, D. A. Stahl and R. Gupta, NARe.7 9,6167 (1981). 149. A. Barta, G. Steiner, J. Brosius, H. F. Noller and E. Kuechler, PNAS 81, 3607 (1984). 150. I. Wower, J. Wower, M. Meinke and R. Brimacornbe, NARes 9, 4285 (1981). 151. M. Stoffler-Meilicke, B. Epe, K. G. Steinhauser, P. Woolley and G . Stoffler, FEBS Lett. 163, 94 (1983). 152. R. R. Traut, J. M. Lambert and J. W. Kenny,JBC 258, 14592 (1983). 153. S.-E. Skdld, EJB 127,255 (1982). 154. D. Robbins and B. Hardesty, Bchen 22,5675 (1983). 155. M. S. Bretscher, Nature 218,675 (1968). 156. M. H. Schreier and H. Noll, PNAS 68,805 (1971). 157. D.-M. Chuang, H. A. Silberstein and M. V. Simpson, ABB 144,778 (1971). 158. D.-M. Chuang and M. V. Simpson, PNAS 68, 1474 (1971). 159. J. Waterson, M. L. Sopori, S. L. Gupta and P. Lengyel, Bchem 11, 1377 (1972). 160. A. S . Spirin, FEBS Lett. 14, 349 (1971). 161. A. S. Spirin, N. V. Belitsina and E. B. Lishnevskaya, FEBS Lett. 24, 219 (1972). 162. A. A. Infante and M. Krauss, BBA 246,81 (1971).

114

ALEXANDER

S. SPIRIN

163. A. A. Infante and R. Baierlein, PNAS 68, 1780 (1971). 164. A. S. Spirin, V. I. Baranov, I. N. Serdyuk and R. May, Dokl. Akad. Nauk S S S R 274, 1260 (1984). 165. A. T. Gudkov, G. M. Gongadze, V. N. Bushuev and M. S. Okon, FEBS Lett. 138, 229 (1982). 166. G. M. Gongadze, A. T. Gudkov, V. N. Bushev and N. P. Sepetov, Dokl. Acad. Nauk S S S R 279,230 (1984).

Chemical Changes Induced in DNA by ionizing Radiation FRANKLIN HUTCHINSON Depurtment of Molecular Biophysics and Biochemistry Yale Unioer.sity N e w Haoen, Connecticut

I. The Mechanisms by Which Ionizing Radiations Act on DNA.

......

116

11. Indirect Action: The Effects of Reactive Species Formed from Water

.................... 111. Effects of Irradiation in the Solid State. . . . .

117 118 122 123 123 129 130 131 132 132

B. Yields of Chemical Changes i n Molecules from Irradiation . . . . . . adiated in the C. Products in DNA and in DNA Compon 134 Solid State.. . . . . . . . . . . . . . . . . . . . . . . 138 IV. Irradiation of DNA in Cells . 140 A. Single-Strand Breaks and 143 B. Base D a m a g e . . . . . . . . . . 146 C. Clusters of Damage Sites D. Radiation-Induced DNA Adducts. . . . . . . . . . . . . . . . 147 V. Quantitative Measurements That Should Be Made 148 on Irradiated DNA. . . . . . . . . 149 References . . . . . . . . . . . . . . .

This article gives my present understanding of the effects of ionizing radiation on DNA, mainly the effects produced by irradiation in the presence of oxygen. The literature in this field is immense, full of contradictions, and with many serious gaps. Thus, any attempt to portray a coherent picture requires choices to be made between different views in some areas, and extrapolations from inadequate data in others. The picture presented is intended to be reliable enough so that molecular biologists can use it as a guide to interpret their results. It is also hoped that radiobiologists will be inspired, and provoked, to fill in some of the gaps and to remove the inconsistencies in this presentation. 115 Progress in Nucleic Acid Research and Molecular Biology, Vol. 32

Copyright D 1985 by Academic Press. Inc. All rights of reproduction in any form reserved.

116

FRANKLIN HUTCHINSON

Over the past decade, several reviews have both put in order much of the available information, and critically examined some of the contradictions. The most comprehensive is a 383-page multiauthor book, “Effects of Ionizing Radiations on DNA” ( I ) , which covers most work before 1978. There are also especially useful reviews that critically examine the information in particular areas: on radiation effects on sugars and bases in dilute aqueous solution ( 2 , 2 a ) ,on sugar damage ( 3 , 4 ) ,and on the results of electron paramagnetic resonance measurements on DNA and DNA components irradiated in the solid state (5). The major difficulty in studying the effects of ionizing radiation on DNA is the large number of products. For example, 24 different products have been identified for thymine irradiated in dilute oxygenated solution ( 6 ) , with other products not yet identified. Another set of products is found for irradiation in dilute anoxic solution ( 6 ) ,and still a third set when radiation induces ionization directly in solid thymine. A similar diversity of products occurs for cytosine (6, 7) and deoxyribose ( 3 , 4 ) ,and possibly for the purines. Thus, there are hundreds of kinds of radiation-induced products in DNA. A second difficulty is that many of the initial products are unstable, converting into other products over periods of hours or days. Fortunately, in most cases the yields of most of these products are extremely low. Many of the products reported in the literature are for molecules that have been acted upon two or more times because the experimenters used high doses to get measurable yields. This review concentrates on the initial products formed in bases and sugars in the presence of oxygen. The number of kinds of products in DNA is still daunting, but is the order of dozens, not hundreds. One caveat about this statement must be made: there is no good evidence for either the number or the kinds of radiation-induced crosslinks between DNA and other molecules

1. The Mechanisms by Which Ionizing Radiations Act on DNA The radiations whose effects are covered in this review-X rays, y rays, and high energy electrons-interact with matter by transferring energy to the electrons in the irradiated sample. If the energy transferred to one electron is less than the effective ionization potential, the molecule is excited to a higher energy state. If the energy transferred is well above that amount, the electron leaves the molecule, leaving a positive molecular ion. The ejected electron moves through the surrounding matter, losing energy by forming more excited molecules and positive ions. A slowly moving electron is frequently captured by

CHANGES IN

DNA BY

IONIZING RADIATION

117

a positive ion in the medium, leading to a highly excited molecule. Alternatively, the slow electron may b e trapped b y a structure and form a semistable negative ion that can take part in ion-molecule reactions and eventually recombine with a positive ion. The excited molecules thus produced are frequently. in triplet states andlor at high levels of excitation; the fraction of these that undergo chemical change is of order of magnitude unity. Molecules that absorb ultraviolet light (e.g., wavelength 254 nm) are excited to singlet states at lower levels, and the fraction undergoing chemical change is typically very much lower. Excitation of tightly bound inner electrons in atoms of carbon or heavier elements leads to a qualitatively different situation, in which energy many times that of chemical bonding is concentrated in one molecule. The consequences of such high energy concentration are poorly understood, but typically include localized clusters of chemical changes. These events are quite rare in irradiated organic molecules (Section 111). Ionizing radiations differ from most other physical or chemical agents in that they act indiscriminately on all molecules in a treated sample. To a reasonable degree of approximation, the primary effect on a particular molecule is proportional to the number of electrons. For organic molecules made up of relatively light elements, the number of electrons is approximately proportional to the mass. This indiscriminate action of ionizing radiations leads to a bewildering variety of reactions occurring in irradiated cells. Usually, radiation effects on particular molecules, such as DNA, are ascribed to the sum of two processes. By Direct Action is meant the effects of energy deposited directly in the target molecule. By Indirect Action is meant the effects of reactive species formed in the surroundings that diffuse to the target molecule and react with it.

II. Indirect Action: The Effects of Reactive Species Formed from water on DNA in Dilute Solution For DNA in dilute aqueous solution, the effects of irradiation are caused by the products formed by the action of ionizing radiations on water: the OH radical, the hydrated electron caqr the H atom, HzOz, and Hz. The yields' are given in Table I. The yields in Table I and other places in this article are given as G values, the unit commonly used in radiation chemistry: G = the number of product molecules formed per 100 e V of energy absorbed in the sample.

118

FRANKLIN HUTCHINSON

TABLE I YIELDS OF PHODUCTS IN THE ACTIONOF SPARSELY IONIZING RADIATIONS ON

WATER(8)

G value

Product

2.7 2.7 2.7

0.55 0.45 0.7

-

In aerated solution, reactions (1)and (2) occur very rapidly. e-aq + Oz

H

+02

02-

Hop

pK = 4.9

H'

+ 09-

(1) (2)

The superoxide radical 0 2 - is relatively unreactive with DNA, and at the concentration produced in most radiation experiments, it is responsible for a very small fraction of the products in DNA. HzOz is also produced in concentration too low to play an important role. Thus, the major effective species in oxygenated solution is the OH radical. This reacts chiefly with organic molecules either by adding to a double bond, or by extracting an H atom from a C-H bond to form HzO and a carbon radical. In anoxic solutions, H atoms can also add to double bonds, or extract an H atom from a C-H bond. The hydrated electron can form a negative ion by adding to a conjugated system, such as the aromatic rings in the DNA bases. The OH radical reacts essentially at a diffusion-controlled rate with DNA and DNA components (Table 11). The hydrated electron reacts quickly with the bases, but not with the sugars (Table 11). H atoms react more sIowly (Table 11).

A. Deoxyribose Products in Oxygenated Solution OH radicals are the only reactive species in oxygenated solutions that react with any efficiency with the sugar. H atoms attached to each of the five carbon atoms can be extracted by OH (4,17). In oxygenated solution, oxygen reacts rapidly and irreversibly with each of the carbon radicals so formed. Figure 1 shows the structures of sugar products that have been identified after irradiation of oxygenated DNA solutions. Extraction of an H atom from the 4'-C or the 5'42 of the sugar, then

CHANGES IN

DNA

119

BY IONIZING RADIATION

TABLE I1 THERATESAT WHICHVARIOUSRADICALSREACTWITH POLYNUCLEOTIDES AND THEIR I N NEUTRAL SOLUTION (8) COMPONENTS

k ( x loy M-' sec-I) Solute Bases Thy CYt Ura Ade

OH

Ref.

H

18 13.2 15 9

9 10 11 10

0.38

Ref.

e-aq

5.6 4.7 5.2 4.4

9

1.6" 1.3h

13 13

C > 1.3);moderately conserved (1.3> C > 0.3);and variable (C < 0.3).In the schematic diagram, based on the E. coli secondary structure, conservation values are indicated by large letters (universally conserved); small letters (highly conserved); large dots (moderately conserved); smoll dots (variable).

.-. .... . . ....

. . . . . . . . . * I _ . .

:......... 1III.II. ,.

I

11.11 ‘.“A&

*.

.... ......f

III1.l G

. .-.. : ..-. . ..

FIG.28. Primary structure conservation in 16-S-like rRNA. Conservation values were computed and indicated as for Fig. 27, except that mitochondria1 rRNAs were excluded from the sequence alignment. Functionally relevant sites are indicated as follows: K,kethoxal-reactive (and therefore single-stranded and exposed) in 3 0 4 ribosomal subunits (69);0, kethoxal-reactive sites protected by 50-S subunits (70); e, sites that show enhanced kethoxal reactivity in 7 0 3 ribosomes; M, kethoxaI-reactive sites whose modification interferes with binding of 5 0 3 subunits (71); A, kethoxal-reactive sites protected in polysomes but reactive in vacant 7 0 4 couples (72).Sites of mutation affecting antibiotic sensitivity or resistance are indicated by: Spc, spectinomycin ( 7 3 ) ; Pmr, paromomycin (38, 7 4 ) ; Hmr, hygromycin (74); Ksg, kasugamycin (75); tRNA, site of photochemical crosslinking of the S nucleotide of the anticodons of certain tRNAs to 16-S rRNA (76);mRNA, region involved in iiiRNA selection in bacteria (77, 78);colE3, site of cleavage by colicin E3 (79, 80).

TABLE III 16-S-LIm rRNA SEQUENCES USED IN COMPUTER-ASSISTED COMPARATIVE ANALYSIS'

EUBACTERIA

*

Escherichia coli P r o t e u s q z

Mycoprasma ca ricolum Anacystis n i u ans

a+--

CHLOROPLAST

*

Tobacco Maize Chlamydomonas Euglena

MI TOCHONDRI A

I I I I I I I I plan"

Maize Wheat

lower euk.

*

Aspe r g i 11 us Yeast Paramec i um SJ. I Paramec i um 4

..=.

mamma 1 s

*

Human Bov i ne Mouse Rat

Ii

* I

*

Dictyostelium discoideum thermophila 7 Yeast

7Tetrahymena

1

E i ~ ~ ulaevis s

Sources of sequence data are listed in Table 11. Nucleotides at the bottom show the observed phylogenetic variation at position 1293 of 1 6 4 rRNA for the sequences, i.e., the list of organisms at the right has been twisted 90 degrees, so that the first nucleotide at the left (C) is that found in E . coli 1 6 3 rRNA at position 1293; the second nucleotide (U) is that found in Proteus vulgaris, etc.. The different sequences are separated into five phylogenetic groupings, as shown, to facilitate reading the Tables. The number pattern at the bottom shows the principle of the computer algorithm used to detect coordinated base changes. The first nucleotide at the bottom left is encoded by the number 1. The second nucleotide is also called 1 if identical to the first

197

~ ~ - S - L I RIBOSOMAL KE RNA TABLE IV

DISTRIBUTION OF UNIVERSALLY CONSERVED BASESBETWEEN HELICAL AND NONHELICAL REGIONS“ ~~

Base

Total number

A G C U

389 487 352 314

Total 1542 Percentage

A G C U

Total Percentage

119 112 64 65 360

Percentage

Basepaired

All E. coli positions 25.2 147 31.6 337 22.8 249 21 I 20.4

944 61.2

~

~

Unpaired

Percentage unpaired

242 150 103 103

62.2 30.8 29.3 32.8

598 38.8

Universally conserved positions only 33.1 22 97 31.1 60 52 17.8 30 34 18.1 36 29

148 41.1

81.5 46.4 53.1 44.6

212 58.9

” Universally conserved bases are defined here as those that are present at the same position in all organisms listed in Table 111, excluding mitochondria.

which makes hydrophobic interactions with proteins an attractive possibility.

E. Functional Sites and rRNA Conservation Several investigators have contributed evidence bearing on the functional involvement of various specific regions of 16-S rRNA in protein synthesis. It is useful to review this evidence briefly in the context of primary and secondary structure conservation. In Fig. 28, we show the positions of several sites of functional interest in relation to phylogenetically conserved structural features. Much of the evidence bearing on the various proposed functional sites has been reviewed previously (1)and is summarized only briefly here. one; ifdifferent, it is encoded as 2. The third nucleotide, ifdifferent from the first two,is called 3; and so on. The example at the bottom of the Table shows that the first eight sequences have C,U,G,G,G,G,A,A at the position corresponding to residue 1293 of E. coli 16-S rRNA. Thus, all C’s are encoded as 1, U’s as 2, G’s as 3, and A’s as 4, according to the order of their appearance, from left to right. This gives the pattern 12333344 for the first eight positions, as shown at the bottom of the Table. Sequence positions that show identical number patterns are then identified by the computer, as described in the text.

198

ROBIN R. GUTELL ET AL.

1. mRNA Selection. The proposal (77)that the 3‘-terminal CCUCC sequence (positions 1535-1539) is involved in mRNA selection via base pairing to sequences about 10 nucleotides upstream from the start codon is strongly supported by experimental evidence (78, 83, 84). This sequence is present in eubacteria and archaebacteria, but not in eukaryotes, nor in mitochondria (except plant). 2. Colicin E3. The bacteriocin Colicin E3 inactivates E. coli ribosomes by means of a single phosphodiester bond cleavage after position 1493 (79, 80). 3. Antibiotic Resistance. Kasugamycin is believed to inhibit translational initiation. Resistance to kasugamycin is conferred when methylation of dimethyladenosines 1518 and 1519 is blocked (75).Resistance to spectinomycin, thought to interfere with translocation, is conferred by a C-to-U transition at position 1192 (73).Hygromycin B appears to affect translocation and translational fidelity in sensitive cells; resistance to this antibiotic is conferred b y a U-to-C transition at position 1495 in Tetrahymena (74).Resistance to paromomycin, an antibiotic known to interfere with translational fidelity (reviewed in 85), appears to be the result of creating a mispair at positions C-1409/ G-1491. Thus, paromomycin resistance has been observed in yeast mitochondria where the C at 1409 is changed to G (38),and in Tetruhymena, where the G at 1491 is changed to A (74).There is a canonical base-pair at these positions in all of the eubacteria (including chloroplast), plant and fungal mitochondria, archaebacteria, and some eukaryotes (Dictyostelium, Tetrahymena, and maize). Other eukaryotes (yeast, Xenopus, and rat) have an A at position 1491 while maintaining the C at 1409, suggesting that these eukaryotes are naturally resistant to this antibiotic, while those species that have a G at position 1491 are naturally sensitive. It is interesting to note that the first two positions of the penultimate helix in mammalian mitochondria are unpaired (Figs. 11 and 12). 4. tRNA Binding. Direct crosslinking of the 5’ (“wobble”) base of the tRNA anticodon to position 1400 places the site of codon-anticodon interaction close to one of the longest universally conserved sequences in 16-S rRNAs (76). Protection of positions 530, 693, 966, 1338, and 1517 from kethoxal in polysomes, but not in vacant 70-S ribosomes, suggests these sites as possible tRNA and/or mRNA contacts (72). The 3’ terminus of peptidyl-tRNA is proximal to positions 2584 and 2585 of 23-S rRNA (86). 5. 50-S subunit contacts. Protection of specific sites in 16-S rRNA from kethoxal(70) or cobra venom RNase (87)in 70-S ribosomes (but not in 30-S subunits) identifies positions 337, 674,703,705, 773,791, 803, 818, 1064, 1405, 1408, 1409, 1490, 1497, and 1516 as potential

16-S-LIKE RIBOSOMAL

RNA

199

50-S subunit contacts. Damage/selection experiments using kethoxal indicate that modification of any of the positions 674, 703, 70.5, 791, 818, 1064, 1497, or 1516 interferes with 50-S binding (71).

II. A Computer-Assisted Search for Coordinated Base Changes in 1 6 4 rRNA At this stage in the development of our understanding of 16-S ribosomal RNA structure, analysis of secondary structure, at least for eubacteria, seems virtually complete. Models have been tested by comparative analysis using sequences from all of the major phylogenetic groups, and by a wide range of experimental approaches. There is presently little doubt that the proposed secondary structure is essentially correct, although a number of details remain to be resolved. Meanwhile, important steps are being taken toward the folding of this molecule into its three-dimensional structure (reviewed in 1 , 3 , 5 , 1 3 ) . In this section, we briefly discuss experimental methods being used to study rRNA tertiary structure, and then present the results of a computer-assisted search for potential base-base tertiary interactions. For the most part, it is premature to conclude that these potential interactions in fact represent tertiary contacts. However, this information will be of value, for example, in attempting to extrapolate results from crosslinking and other approaches to the level of detailed base-base interactions. Experimental determination of rRNA tertiary structure currently involves several strategies. The direct approach of X-ray crystallography remains only a remote possibility, due to the difficulty of preparing high quality crystals of ribosomes or their subunits. For the foreseeable future, three-dimensional features of rRNA will have to be deduced using more indirect methods. Proximal regions of the rRNA can be physically associated by means of crosslinking, either chemical or UV-induced. The majority of these crosslinks tend simply to confirm secondary structure features (reviewed in 1).Those that are not consistent with secondary interactions are possible tertiary candidates. These include the following positions or regions:' 1.680, 10.1180, 490980, 610-1320, 930.1540, 1320.1360 (88), 360.1330, 620.1420, and 960-1510 (89). Noncovalent association of non-basepaired fragments of 16-S rRNA in the absence of ribosomal proteins suggests possible tertiary contacts between sites contained in these fragments (90-93). See footnote 2.

200


Another general approach to deducing tertiary structure involves mapping the locations of landmark features of the rRNA onto the surface of the three-dimensional eIectron microscopy model of the 30-S subunit from data obtained by immuno-electron microscopy (reviewed in 13)and/or neutron scattering (94). These features include posttranscriptionally modified nucleotides, the 5' and 3' ends, and ribonucleoproteins whose rRNA binding sites have been established (reviewed, in 1, 13,95,105).Unfortunately, none of these approaches is capable of identifying the specific nature of the tertiary interaction, but they have begun to suggest certain general features of the threedimensional folding pattern of the 16-S rRNA.

A. Tertiary Interactions Constrain tRNA Primary Structure Underlying the comparative search for secondary structure is the assumption that structure is likely to be conserved in 1 6 4 rRNA from phylogenetically diverse sources. A similar assumption can be made concerning tertiary structure: ribosomes are likely to share (for the most part) common tertiary structure features, even though the primary structures of the different 16-S-like rRNAs vary. Making this assumption, and encouraged by the success this approach has yielded in deducing 16-S rRNA secondary structure, we have begun a search for conserved base-base tertiary interactions. We do not embark on this search without precedent. By 1969, the primary structures for 14 transfer RNAs had been determined. In addition, a secondary structure common to these nucleotide sequences had been established. At that time, Levitt (14) identified two sets of coordinated base changes involving unpaired nucleotides in tRNA and predicted that they are involved in tertiary contacts. The first set involved the D loop (position 15) and the variable loop (position 48). Whenever the nucleotide at position 15 was a G, position 48 was a C , and whenever 15 was an A, 48 was a U (G15C48 or A15-U48).6The second interaction involves positions 9 (located between the A and D stems) and 23 (located within the D stem, and base-paired to position 12). One of two nucleotide patterns usually existed for these three positions: (G12.C23).G9 or (U12.A23).A9. Both of these proposed tertiary interactions were confirmed when the complete tRNA threedimensional folding pattern was established by X-ray crystallography (15, 16).In addition, Levitt proposed a number of other interactions involving invariant nucleotides, based on experimental crosslink inSee footnote 2.

16-S-LIKE RIBOSOMAL

RNA

201

formation and other structural constraints, some of which were also verified by the crystal structures. The specific constraints on some of the tRNA tertiary features merit further discussion. Some tertiary contacts appear to be absolutely conserved in nature, while others are less stringent. The two interactions involving positions U8-Al4 and T54-A58 are constrained due to reverse Hoogsteen pairing, which can only occur between A and U. Another tertiary contact, involving the invariant nucleotides G18 and U55 can only be made between G and U. The tertiary interaction between invariant nucleotides G19 and C56 is the only tertiary Watson-Crick (orthodox) base pair. The reason why coordinated base changes are not found within this interaction is not obvious. On the other hand, the contact involving positions 15 and 48 is less constrained, involving RY-constrained coordinated base changes that maintain canonical pairing, as is the triple tertiary interaction involving positions 46,22, and 13. Finally, it should be noted that exceptions are found for even the “strict” coordinated base changes noted here. Consideration of these tRNA tertiary interactions serves to exemplify how sequence constraints can be dictated by nucleic acid geometry. The principles that can be extracted from tRNA are important to bear in mind in studies on the higher order structure of 16-S rRNA, although we anticipate finding classes of tertiary interaction in 1 6 4 rRNA that are not found in tRNA.

B. Searching for Coordinated Base Changes in 164 rRNA Any systematic search for coordinated base changes in a molecule the size of 16-S rRNA requires the use of computer methods. By definition, w e wish to identify groups of two or more positions in an RNA chain whose phylogenetic variation is mutually constrained by some set of rules. This might be a Watson-Crick (orthodox) relationship, or any other set of rules, as long as the pairing relationships are unambiguous (at least at the outset). In our approach to this problem, these requirements are met by a simple and general mathematical strategy implemented by means of a computer program. An algorithm was devised (H.F.N., R.R.G., and T. Goldstein, unpublished) in which the pattern of nucleotide variation at a given position in the RNA chain is generalized to a pattern of numbers; the computer then sorts all number patterns into similar groups. Nucleotide positions belonging to a given group thus represent bases that always vary phylogenetically in a concerted fashion. Pairs of positions within each group may then be

202


further classified according to whether their coordinated relationship follows Watson-Crick or other rules. The actual procedure is best illustrated by an example. Consider the observed variation at position 1293 for the 16-S-like rRNA sequences listed in Table 111. The first nucleotide (corresponding to the E. coli position) is assigned the number 1. If the second nucleotide is identified to the first, it is also assigned the number 1;if different, it is called 2. The next nucleotide that differs from the preceding ones is called 3, and so on. Deletions are always encoded as 5. Thus, the nucleotide pattern CUGGGGAA... is converted to the number pattern 12333344.... After sorting, all positions having this general pattern are identified. Using our present sequence compilation, the only other position in 16-S rRNA with this number pattern is position 1244, which has the nucleotide pattern GACCCCUU.... Comparison of the two nucleotide patterns reveals in this case a strict WatsonCrick complementarity between the two positions, which turn out to be the fourth base-pair in the (1241-1247).( 1290-1296) helix. It should be noted that positions that exhibit any type of strictly coordinated variation will be identified by this approach. Although secondary as well as tertiary interactions are expected to be identified by this procedure, an important point is that any helical base-pair whose variations include at least one G-U or other unorthodox basepair will be excluded from any such list. Thus, this type of search will identify all helices that contain base pairs that obey strict WatsonCrick pairing rules. For this reason, our search serves as a kind of systematic, though not exhaustive test of the secondary structure model. In fact, only one small helix [(725-726)*(731-732)I that had escaped detection by more informal searches was identified. This result implies that few, if any, helices have been overlooked in the present version of our model. It also implies that alternative, mutually exclusive helices (helical “switches”) are unlikely to exist within 16-S rRNA. The results of a search for stringently coordinated base substitutions from an alignment of 28 different complete 16-S-like rRNA sequences are presented in Table V. Groups of nucleotide positions (indicated by E. coli position number at the left in Table V) are sorted according to their common patterns, as described above, in order of increasing complexity. Because of space limitations, universally conserved nucleotides and pattern groups that contain only a single nucleotide position have been eliminated from the Table. As expected, many pairs of positions correspond to previously identified base-pairs that maintain strict Watson-Crick complementarity (denoted by an x at

203

~ ~ - S - L I RIBOSOMAL KE RNA TABLE V POSITlONS IN ~ ~ - S - L I K rRNA E SHOWING COORDINATED

PATTERNS

OF PHYLOGENETIC VARIATION' ~~

~~~

~

~

985 1220

CCCCCCCC CCCCCC CCCC CCCC CGCCCC X - 1 2 2 0 GGGGGGGG GGGGGG GGGG GGGG GCGGGG X - 9 0 5

54 5 500 1516 797 785 364

CCCCCCCC GGGGGGGG GGGGGGGG CCCCCCCC GGGGGGGG GGGGGGGG

5 10 728 692 957 1065 58 354 5 18 275

AAAAAAAA AAAAAA hAAA AAAG AhAAAA AAAAAAAA AAAAAA AAhA AAAG AAAAAA

312 115

1110 1229 952 314 113 22 393 368 1411 1489 864

uuuuuuuu uuuuuuuu uuuuuuuu cccccccc

CCCCCC GGGGGG GGGGGG CCCCCC GGGGGG GGGGGG

CCCC GGGG GGGG CCCC GGGG GGGG

CCCC GGGG GGGG CCCC GGGG GGGG

uuuuuu UUUU uuuc uuuuuu uuuu uuuc uuuuuu uu3u uuuc cccccc cccc CCCG

AAAAAA X - 5 0 0 UUUUUU X - 5 4 5

UUUUUU GGGGGG X - 7 8 5 CCCCCC X - 7 9 7 CCCCCC 379

uuuuuu uuuuuu

uuuuuu

1191 x-354 GGGGGGGG GGGGGG GGGG GGGC GGGGGG X - 5 8

cccccc

cccccccc cccccc cccc cccu cccccc

GGGGGGGG GGGGGG GGGG GGGA GGGGGG

CCCCCCCC CCCCCC CCCC GGGC GGGGGG X - 1 1 5 GGGGGGGG GGGGGG GGGG CCCG CCCCCC X - 3 1 2 AAAAAAAA AAAAAAAA UUUUUUUU CCCCCCCC GGGGGGGG GGGGGGGG

AAAAAA AAAA GGGG AAAAAA AAAP. GGGG LWUUUU UUUU CCCC CCCCCC CCCC GGGG GGGGGG GGGG CCCC GGGGGG GGGG CCCC

GGGGGG GGGGGG CCCCCC GGGGGG CCCCCC CCCCCC

288 118 532 804 16 1204

X-952 X-1229 X-113 X-314 12

AAAAAAAA AAAAAA AAAA GGGC AAAAAA X - 3 6 8 UUUUUUUU UUUUUU UUUU CCCG UUUUUU X - 3 9 3 CCCCCCCC CCCCCC CCCC GUCC CUACAA X - 1 4 8 9 GGGGGGGG GGGGGG GGGG CAGG GAUGUU X - 1 4 1 1 AAAAAAAA AAAAAA a a a u AAAA AAAAAA

911 UUUUUUUU UUUUUU UUUA UUUU UUUUUU 36 548 1370

249

889

cccccccc cccccc

CCUU cccc cccccc x - 5 4 8 GGGGGGGG GGGGGG GGAA GGGG GGGGGG X - 3 6 GGGGGGGG GGGGGG GGAA GGGG GGGGGG 1352 AAAAAAAA AAAAAA AUUU AAAA aaaaaa UUUUUUUU UUUUUU uaaa UUUU UUUUUU AAAAAAAA UUUUUUUU AAAAAAAA AAAAAAAA

AAAAAA CCCC AAAA AAAAAA UUUUUU GGGG UUUU UUUUUU X - 7 7 5 RAAAAA GGGG AAAA AAAAAA AAAAAA GGGG AAAA AAAAAA (continued)

204

ROBIN R. CUTELL ET AL.

TABLE V (Continued) 1363 1330 62 899 803 1353 1054 1230 524 725 775 32 552

AAAAAAAA AAAAAA GGGG AAAA AAAAAA

uuuuuuuu uuuuuuuu

UUUUUU cccc UUUU UUUUUU cccc uuuu CCCCCCCC CCCCCC AAAA CCCC GGGGGGGG GGGGGG UUUU GGGG GGGGGGGG GGGGGG UUUU GGGG

uuuuuu UUUUUU CCCCCC GGGGGG GGGGGG

cccccccc cccccc uuuu cccc cccccc

CCCCCCCC CCCCCC UUUU CCCC CCCCCC

1369

951

GGGGGGGG GGGGGG AAAA GGGG GGGGGG GGGGGGGG GGGGGG AAAA GGGG GGGGGG 732 GGGGGGGG GGGGGG AAAA GGGG GGGGGG X - 8 0 4 AAAAAAAA AAAAAA UUUU AAAA UUUUUU X - 5 5 2 UUUUUUUU UUUUUU AAAA UUUU AAAAAA X - 3 2

1340 943

AAAAAAAA AAAAAA UUUU UUAA AAAAAA x - 9 4 3 UUUUUUUU UUUUUU AAAA AAUU UUUUUU X - 1 3 4 0

1197 1060

UUUUUUUU UUUUUU CCCC CCCC CCCCCC X - 1 1 9 7

507 9

AAAAAAAA AAAAAA GGGG GGGG GGGGGG X - 1 0 6 0

CCCCCCCC CCCCCC UUUU GGGG GGGGGG GGGGGGGG GGGGGG AAAA CCCC CCCCCC

25

10 AAAAAAAA AAAAAA GGGG CCCC UUUUUU X - 2 4 2 4 UUUUUUUU UUUUUU CCCC GGGG AAAAAA X - 1 0 1209 1048

308 654 1074 1083 282 676 1213 56 1205 792 788 355 791

CCCCCCCC CCCCCC GAGG CCCC CCCCCC x - 1 0 4 8 GGGGGGGG GGGGGG CUCC GGGG GGGGGG X - 1 2 0 9

.... GGGG CCCC UUUUUU 292 .... aaaaaA GGGGGGGG G G gggg .... AAAG GGGGGG x - 1 0 8 3 UUUUUUUU UUUUUU .... CCCU UUUUUU X - 1 0 7 4 CCCCCCCC CCCCCC

GGGGGGGG GGGGGG

AAAAAAAA AAAAGG AAAA AAAA AAAAAA AAAAAAAA AAAAGG aaaa AAAA AAAAAA AAAAAAAA AAAAGG AAAA AAAA AAAAAA

244

UUUUUUUU UUUUCC UUUU UUUU UUUUUU U U U U U U W UUUUCC U W U UUUU UUUUUU

356 1053

AAAAAAAA AAAAUU AAAA AAAA AAAAAA

UUUUUUUU UUUUAA UUUU UUUU UUUUUU CCCCCCCC CCCCAA CCCC CCCC CCCCCC

57

GGGGGGGG GGGGUU GGGG GGGG GGGGGG

CCCCCCCC CCCCUU CCCC CCCC UUUUUU X-809

770 809

GGGGGGGG GGGGAA GGGG GGGG AAAAAA X - 7 7 0

883 567

CCCCCCCC CCCCUU CCCC CCCC GGGGGG x - 5 6 7 GGGGGGGG GGGGAA GGGG GGGG CCCCCC X - 8 8 3

779

CCCCCCCC CCCCUU AAAA CCCC CCCCCC

205

~ ~ - S - L IRIBOSOMAL KE RNA TABLE V (Continued) 1050

GGGGGGGG GGGGAA UUUU GGGG GGGGGG

1344 939

CCCCCCCC CCCCUU GGGG GGGC CCCCCC X - 9 3 9 GGGGGGGG GGGGAA CCCC CCCG GGGGGG X - 1 3 4 4

175 147 23 726 810 936 980 1237 1399 11

51 571 731 769 1337 1504 805 774 1320 1487 311 289 732 1053 1187 739 667 1328 1309 651 588 1496 1405

CCCCCCCC CCCCAA GGGGGGGG GGGGUU

.... GCCC .... CGGG

1208

CCCCCC X - 1 4 7 GGGGGG X - 1 7 5

cccccccc cccucc cccc cccc cccccc

x-11

CCCCCCCC CCCUCC CCCC CCCC CCCCCC X - 7 3 1 CCCCCCCC CCCUCC CCCC CCCC CCCCCC X - 7 6 9 cccccccc cccucc c c c c cccc c c c c c c

cccccccc cccucc cccc cccc cccccc cccccccc cccucc cccc cccc cccccc cccccccc cccucc cccc cccc cccccc

GGGGGGGG GGGGGGGG GGGGGGGG GGGGGGGG GGGGGGGG GGGGGGGG GGGGGGGG

GGGAGG GGGAGG GGGAGG GGGAGG GGGAGG GGGAGG GGGAGG

GGGG GGGG GGGG GGGG GGGG GGGG GGGG

GGGG GGGG GGGG GGGG GGGG GGGG GGGG

GGGGGG X - 2 3 355 GGGGGG 164 GGGGGG GGGGGG X - 7 2 6 GGGGGG X - 8 1 0 GGGGGG GGGGGG

cccccccc cccucc cccc cccc ccccuu

x-774

GGGGGGGG GGGAGG GGGG GGGG GGGGAA X - 8 0 5

cccccccc cccucc cccc cccc uuuuuu GGGGGGGG GGGAGG GGGG GGGG AAAAAA

1413

CCCCCCCC CCCUCC CCCC GGGC GGGGGG X - 2 8 9 GGGGGGGG GGGAGG GGGG CCCG CCCCCC X - 3 1 1 CCCCCCCC CCCUCC UUUU CCCC CCCCCC GGGGGGGG GGGAGG AAAA GGGG GGGGGG GGGGGGGG GGGAgg AAAA gggg gggggg

725 1205 1113

CCCCCCCC CCCUCC aaaa AGCC CCCCCC x - 6 6 7 GGGGGGGG GGGAGG uuuu UCGG GGGGGG X - 7 3 9 CCCCCCCC CCCUCC GAGG CCCC CCCCCC X - 1 3 0 9 GGGGGGGG GGGAGG CUCC GGGG GGGGGG X - 1 3 2 8 CCCCCCCC CCCUCC GGGGGGGG GGGAGG

.... CCCC ...... X - 5 8 8 .... GGGG ...... X - 6 5 1

cccccccc cccuuu cccc cccc cccccc GGGGGGGG GGGAAA GGGG GGGG GGGGGG

386 377

CCCCCCCC CCCUGG UUUU GGGG GGGGGG X - 3 7 7 GGGGGGGG GGGACC AAAA CCCC CCCCCC X - 3 8 6

40 402

CCCCCCCC CCCAUU UUAA UCUC gggggg X - 4 0 2 GGGGGGGG GGGUAA AAUU AGAG c c c c c c X - 4 0 (continued)

206


TABLE V (Continued)

768 1235 694 813 1240

AAAAAAAA UUUUUUUU AAAAAAAA UUUUUUUU UUUUUUUU

94 0 1343

CCCCCCCC CCUCCC UUUU CCGG GAGGGG X - 1 3 4 3 GGGGGGGG GGAGGG A A A A GGCC CUCCCC X - 9 4 0

271 255 1396 921 18 917 525 1403 1501 506 1401

AAGAAA UUCUUU AAUAAA UUAUUU UUAUUU

CCCCCCCC CCACUU GGGGGGGG GGUGAA AAAAAAAA UUUUUUUU CCCCCCCC GGGGGGGG

AAUUAA UUAAUU CCGGCC GGCCGG

AAAA UUUU AAAA UUUU UUUU

AAAA UUUU AAAA UUUU UUUU

AAAAAA UUUUUU AAAAAA UUUUUU UUUUUU

946 1297

.... CCCC CCCCCC X - 2 5 5 .... GGGG GGGGGG X - 2 7 1 AAAA UUUU CCCC GGGG

AAAA UUUU CCCC GGGG

AAAAAA UTJUUUU CCCCCC X - 9 1 7 GGGGGG X - 1 8

cccccccc ccuucc cccc cccc cccccc cccccccc ccuucc cccc cccc cccccc cccccccc ccuucc cccc cccc cccccc

GGGGGGGG GGAAGG GGGG GGGG GGGGGG GGGGGGGG GGAAGG GGGG GGGG GGGGGG

569 881

CCCCCCCC CCUUCC CCCC CCCC GGGGGG X - 8 8 1 GGGGGGGG GGAAGG GGGG GGGG CCCCCC X - 5 6 9

866 570

CCCCCCCC CCAACC c c c c GGGG A A A A A A GGGGGGGG GGUUGG GGGG CCCC UUUUUU

880

1 3 4 2 CCCCCCCC CCUUCC UUUU CCCC UUUGUU X - 9 4 1 9 3 1 GGGGGGGG GGAAGG A A A A GGGG AAACAA X - 1 3 4 2 1226 954

CCCCCCCC CCUUCC GGGG CCCC CCCCCC X - 9 5 4 GGGGGGGG GGAAGG CCCC GGGG GGGGGG X - 1 2 2 6

968 1498

A A A A A A A A AAUUUU A A A A A A A A A A A A A A UUUUUULJU U U A A A A UUUU UUUU UUUUUU

622 439

AAAAAAAA A A u u u u UUUUUUUU U U A a a a

316 337

CCCCCCCC CCUUUU GGGGGGGG GGAAAA

342 347

cccccccc ccuu..

931 1386 403 39

GGGGGGGG GGAA..

.... AAAA ......

.... UUUU ...... 4 9 5 .... CCUU GAGGGG X - 3 3 7 .... GGAA CUCCCC X - 3 1 6 cccc cccc cccccc

x-347

GGGG GGGG GGGGGG X - 3 4 2 .

CCCCCCCC CCUAUU UUUU GGGG GGGGGG X - 1 3 8 6 GGGGGGGG GGAUAA A A A A CCCC CCCCCC X - 9 3 1 CCCCCCCC CCAUUU AACC A A A k aaaaaa x - 3 9 GGGGGGGG GGUAAA UUGG UUUU uuuuuu X - 4 0 3

TABLE V (Corititturd)

1084 1073

.... GGGG GGGGGG .... UUUU UUUUUU .... A A A A ...... .... UUUU ......

GGGGGGGG G G C ~ U U UUUUUUUU UUaggg

X-1073 x-1084

640 598

AAAhAAAA AAG.aa UUUUUUUU UUC.UU

x-598 X-640

248 276

CCCCCCCC GGGGGG GGGG GGGG CCCCCC X - 2 7 6 GGGGGGGG CCCCCC CCCC CCCC GGGGGG X - 2 4 8

370 391

CCCCCCCC GGGGGG AAAA UUCC CGCCCC X - 3 9 1 GGGGGGGG CCCCCC UUUU AAGG GCGGGG X - 3 7 0

153 168

CCCCCCCC GGGGGGGG

1149 1124

097 952

...... .... CCCC ...... .... GGGG CCUaUU .... UUCC

CCCCCCCU GGGGGGGA GGAuAA

.... hAGG

CCCCCC X - 1 6 8 GGGGGG X - 1 5 3 UUUUUU x - 1 1 2 4 aaaaaa x - 1 1 4 9

CCCCCCCU CCGACC CCCC CCCC CCCCUU X - 9 0 2 GGGGGGGA GGCUGG GGGG GGGG GGGGXA X - 8 9 7

1225 955

AAAXAAUA AAAAAA A A A A AkAG GGGGGG X - 9 5 5 UUUUUUAU UUUUUU UUUU UUUC CCCCCC X - 1 2 2 5

1509 1526

CCCCCCGC CCGCCC GGGG CCCC CCCCCC X - 1 5 2 6 GGGGGGCG GGCGGG CCCC GGGG GGGGGG X - 1 5 0 9

1510 1525

CCCCCCGC CCUCGG UCCC CCCC CUCCCC X - 1 5 2 5 GGGGGGCG GGAGCC AGGG GGGG GAGGGG X-1510

.... GGCG ...... X - 6 0 3 .... CCGC ...... X - 6 3 5

635 633

AAAAGGAA UU..aa UUUUCCUU AA..uu

681 709

AAAACCAC GGUAUU aaaa CCAC UAAAGG x - 7 0 9 UUUUGGUG CCAUAA uuuu GGUG AUUUCC X - 6 8 1

1082 1075

AAAAGGGA AAagaa UUUUCCCU U U U C U U

.... GGGA

.... CCCU

UUUUUU x - 1 0 7 5 GGGGGG X - 1 0 8 2

514 537

CCCCUUUC CCCUUU CCCC GGGG GGGGGG X - 5 3 7 GGGGAAAG GGGAAA GGGG CCCC CCCCCC X - 5 1 4

613 621

CCCUCCCC GG.... GGGAGGGG CC....

990 1215

.... ...... X - 6 2 7 .... GAUC CUAG ...... X - 6 1 3

CCCUCCUU CCUUUU CCCC CCGG UCCCCC X - 1 2 1 5 GGGAGGAA GGAAAA GGGG GGCC AGGGGG X - 9 9 0

784 798

AAAGGGAA AAACAA GGGG CCCG AAAAAA X - 7 9 8 UUUCCCUU UUUGUU CCCC GGGC UUUUUU X - 7 8 4

614 626

CCCGAAAA CC.... GGGClNUU GG....

.... CCCC ...... X - 6 2 6 .... GGGG ...... X - 6 1 4 (contitiued)

208


TABLE V (Continued) 52 359

CCUCCCCC CCCCCC CCCC GGGG GCGGGG X - 3 5 9 GGAGGGGG GGGGGG GGGG CCCC CGCCCC X - 5 2

309 291

AAGAAAAA AAGGAA UUCUUUUU UUCCUU

1484 1416

.... CCCCCC .... CCCC GGGG GGGGGG

X-291 X-309

CCUCCCCC CCUUUU GGGG GGGG GAGGGG X - 1 4 1 6 GGAGGGGG GGAAAA CCCC CCCC CUCCCC X - 1 4 8 4

5 13 538

CCACCCCC CCAUAA UUUU CCCU UUUUUU X - 5 3 8 GGUGGGGG GGUAUU AAAA GGGA AAAAAA X - 5 1 3

736 670

CCUCCCCU UUCAUU u c u u UUUC CUUUUU X - 6 7 0 GGAGGGGA AAGUAA agaa AAAG GAAAAA x - 7 3 6

553 30

AAGAAAGA CCAAAA UCUU GAAC UUUUUU X - 3 0 UUCUUUCU GGUUUU AGAA CUUG AAAAAA X - 5 5 3

502 54 3

AACCAAAA CCCCUU GGGG GGGG GGGGGG X - 5 4 3 UUGGUUUU GGGGAA CCCC CCCC CCCCCC X - 5 0 2

178 144

CCGGCCCC AAUAUU GGCCGGGG UUAUAA

1081 1076

AAGGGGAG G G g a a a UUCCCCUC C C c u u u

371 390 1368 1354

.... AGGGGG .... CGCG GCGC UCCCCC

X-144 X-178

.... UUUUUU x - 1 0 7 6 .... GGGA CCCU AAAAAA X - 1 0 8 1

AAGGGGGG AAUAGG GGGG AAGC UCCCCC X - 3 9 0 UUCCCCCC UUAUCC CCCC UUCG AGGGGG X - 3 7 1 AAGGGGGG GGAAGG UUUU GGGG AGGGGG X - 1 3 5 4 UUCCCCCC CCUUCC AAAA CCCC UCCCCC X - 1 3 6 8

253 273

AAUUUUUU UUAACC CCUU CCUC UUUUUU X - 2 7 3 UUAAAAAA AAUUGG GGAA GGAG AAAAAA X - 2 5 3

19 9 16

AACCCCCC CCAUUU CCCC CCCC CCCCCC X - 9 1 6 UUGGGGGG GGUAAA GGGG GGGG GGGGGG X-19

1508 1527

AAGGGGGG GGCACC AAAA GGGG UUUUUU X - 1 5 2 7 WCCCCCC CCGUGG UUUU CCCC AAAAAA X - 1 5 0 8

286 240

CCGGGGGG UUGAUU UCCC GGAG GGGGGG X - 2 4 0 GGCCCCCC AACUAA AGGG CCUC CCCCCC x-286

876 824

CCAGAACA AAAAAA UUUU GGUC CCCCCC X - 8 2 4 GGUCUUGU UUUUUU AAAA CCAG GGGGGG X - 8 7 6

503 542

CCAUUUCU CCUCAA UUUU CCCG GGGGGG X - 5 4 2 GGUAAAGA GGAGUU AAAA GGGC CCCCCC X - 5 0 3

823

CCGAAAAA GGAAGG CUUU UUGG CCCCCC X - 8 7 7

16-S-LIKE RIBOSOMAL

209

RNA TABLE V (Cotitittued)

877

GGCUUUUU CCUUCC G k A A A A C C GGGGGG X - 8 2 3

624 616

GAGGGGGk A A . .

124 237

CAGGGGGG GGUAUU C C C C GGGG GGGGGG X - 2 3 7 GUCCCCCC CCAUAA GGGG C C C C C C C C C C X - 1 2 4

1293 1244 456 476

cuccccc“ u u . .

.. .... U G C C ...... X - 6 1 6 .. .... ACGG ...... X - 6 2 4

CUGGGGAk UUGUUU AAAA GGAU c a c c g a x - 1 2 4 4 GACCCCLJU AACAAA UUUU C C U A guggcu X - 1 2 9 3 AGS.... UCA.....

.

..aauu ..uuaa

. . . . . . . . . . . . . . x-476 . . . . . . . . . . . . . . x-456

Base variation for each position (using E . coli 164 rRNA numbering) is shown, ordered from left to right according to the scheme shown in Talde 111. Positions are sorted, using a computer program described in the text, according to their patterns of phylogenetic variation. Position nunibers are indicated at the left; positions that are shown as base paired in Figure 1 are denoted by an x at the right, followed by the position of their respective pairing partners. Croups are listed in order of increasing pattern complexity.

the right of Table V, followed by the position of the corresponding pairing partner). The more complex patterns are dominated by secondary structure interactions, as anticipated from consideration of observed sequence variation in tRNA. One must exercise caution in making any extrapolation concerning tertiary structure from tRNA to rRNA, because of fundamental differences in their respective modes of molecular organization. While tRNA exists in uiuo for significant lengths of time as a free RNA species, rRNA functions only as a ribonucleoprotein complex. Thus, much of the three-dimensional folding of 164 rRNA may be directed and/or stabilized by quaternary interactions. At present, the question of whether “naked” 16-S rRNA is able to attain something approaching its ribosomal conformation is a matter of controversy (96-99). If tRNA can be used as a guide, we may expect tertiary interactions to involve .more highly conserved positions. If true, identification of such relationships will generally be more difficult, because conserved positions show simpler patterns of variation (by definition) and so are more likely to arise by chance. Additional criteria are thus required to identify the more attractive possibilities from among the many examples listed in Table V. Examples of the kinds of criteria that might be invoked include (1) mutual proximity in the primary or secondary structure; (2) clusters of positions in two different regions of the struc-

210


ture that maintain coordinated patterns of variation; (3) interactions that are strongly supported by experimental observations; and (4)patterns that show differences within, rather than between, phylogenetic groups. The latter criterion derives from the fact that independent variation at two positions is generally less common within a group, making the coordinated event more statistically significant. On this basis, we consider several examples of potential tertiary interactions selected from the list in Table V. Figure 29 shows their respective relationships to the 16-S rRNA secondary structure model. 1. U1180A288~ (Fig. 29,). The pattern of coordinated base changes between positions 118 and 288 is interesting since it involves only A-U pairing, suggesting a specific conformational constraint. While the pattern of variation is not complex, we note that (1)one of the four mammalian mitochondria differs from the other three at both positions, (2) this pattern occurs only twice, (3) the two occurrences are complementary, and (4) these two regions must be in physical proximity due to constraints imposed by the secondary structure. Taken together, these facts make this coordinated base-change appealing. We also note the occurrence of a universal U at position 287, and a nearuniversal A at position 119. 2. G9C507 (Fig. 29b). The pattern linking positions 9 and 507 involves a more distinctive pattern, with three different pairing possibilities. But since the differences in the pattern fall along phylogenetic lines, and not within them, the correlation is not as strong as it might at first appear. If the (9-13).(21-25) helix is made simultaneously, this would be an example of a potential base triple. 3. G570C866; U571.A865 (Fig. 29c). An unusually strong correlation is found between positions 570 and 866. All four pairing types and orientations exist, giving rise to a quite distinctive pattern. While no differences are seen within the phylogenetic groups in Table V, three additional correlated changes within phylogenetic groups have recently been detected (100). This potential interaction can be extended in an antiparallel orientation by pairing between the universal positions A865 and U571, and might be augmented by another potential pairing between positions 576 and 858. Although the latter pattern is relaxed, it has a number of compensatory changes as shown below.

576 CCCCCCCC GGGGAA .... CCCC AAAAAA 858 GGGGGGGG CCUUUU .... GGGG UUUUUU ' S e e footnote 2.

~ ~ - S - L IRIBOSOMAL KE RNA

211

FIG.29. Schematic diagram of the E . coli 1 6 3 rRNA secondary structure (Fig. l), showing potential tertiary base-base interactions identified by computer analysis of coordinated base changes. (a) U1118.A288; (b)GSC507; (c) G570C866; (d) A673.717; ( e )U921.AI396; (f)A1398.Ul406; (g) C1399C1405; (h) C1399C1504; (i) G1401C1501; (i)G1405C1496.

Comparative evidence contradicting the unproven base-pairs 570-880 and 576-765(unpublished results) would free positions 570 and 576 to participate in tertiary interactions as suggested above. 4. U961.Al499; A968.mU1498;C980-1504.These three sets ofcoordinated base changes associate the 970 and 1500 regions of 16-S

212


rRNA. Although the three number patterns are not complex, and the compensatory base substitutions occur only in mitochondria, the fact that three independent sets of strict correlations are observed linking these two regions is very suggestive. In addition, these two regions have been associated by psoralen crosslinking ( 5 ) .Note that any interaction between A968 and mu1498 would have to be non-WatsonCrick in nature, due to base methylation of U1498. 5. C1399.Gl504; G1401.C 1501; G 1405.C1496; A1398.U 1406; C1399eG1405 (Fig. 29f,g,h,ij). These coordinated base changes involve the decoding site of the ribosome and imply association of two of the most conserved regions of the 16-S rRNA. Three of these sets of patterns, although simple (with the compensatory changes occurring in the mitochondria), link the 1400 and 1500 regions in an antiparallel fashion. The positions associated with these patterns (1399.1504; 1401.1501; 1405.1496) if combined with four potential universal canonical pairs (1404-1497; 1407.1494; 1394-1506; 1395.1505) could form an extended antiparallel interaction enclosed on one side by the (1409-1430).(1470-1491) helix and on the other side by the (923933).(1384-1393) and (1506-1515)*(1520-1529) helices. There are, in addition, two sets of coordinated base changes within the 1400 region, involving positions 1398.1406 and 1399.1405,from changes that occur in mosquito mitochondria (67), yeast mitochondria (38, 39), and trypanosome kinetoplast (101) 16-S-like rRNA sequences. However negative evidence does exist in the Paramecium mitochondria sequence. Furthermore, note that there are nucleotides in common between the potential interactions 1399.1405, 1399.1504, and 1405.1496, and SO some of these may be mutually exclusive. The approach used here will find only those positions whose interactions (either secondary or tertiary) involve strictly coordinated replacements; that is, a one-to-one correspondence between two positions, in which canonical pairing is always maintained as a result of nucleotide substitutions. Using this strict criterion, only a few sets of potential tertiary contacts have been identified. Many other candidates are not in serious contention, for various reasons, but usually the strongest reason for not considering those coordinate base changes is the lack of significance of the pattern of nucleotide changes. In addition, many tertiary contacts are likely to exist between positions that are universally conserved. Such contacts cannot be identified by our approach. Since the number of potential tertiary interactions listed here is less than what would be expected based on comparison with tRNA, our criteria need to be reconsidered.

~ ~ - S - L I RIBOSOMAL KE RNA

213

As noted earlier, there are tertiary interactions in tRNA that involve a strict pattern of coordinate base changes (e.g.,15/48),but there are other examples from tRNA that show canonical as well as non(102).A simultaneous canonical (e.g., G to U) pairing (G46*G22)*U13 search for G to C and G to U pairings would require use of a different, more complex algorithm. But as the stringency of the search is relaxed, allowing tertiary interactions to surface, the number of false correlations will also greatly increase. On the other hand, as more sequences become available, the likelihood of chance occurrence of identical patterns should decrease. At some point, exclusion of mitochondrial rRNAs, which may contain some anomalous structural features, could bring other possibilities to light. In the present analysis, we have considered only Watson-Crick relationships; since bases may interact according to other kinds of rules, the data in Table V may be interpreted in a variety of ways that have not been considered here. If sequencing or sequence alignment errors exist, as is likely, eliminating these will improve the quality of our analysis. Finally, different kinds of search algorithms (103),may help to extend the list of possibilities. A more exciting possibility is to subject these potential tertiary interactions to experimental test. Besides conventional crosslinking and crystallographic approaches, recent developments in comparative analysis and in vitro genetics offer hope of establishing the identity of specific base-base tertiary contacts in 16-S rRNA. Intensifying the comparative approach, Woese and co-workers (100)have developed a strategy based on their library of RNase T1 oligonucleotide catalogs of 16-S-like rRNAs from over 400 species. Starting with a potential tertiary base-base interaction, such as those listed in Table V, a computer search identifies phylogenetically independent examples of variation at one or the other of the two positions in question. The proposed interaction is then tested by determining the identity of the base at the other position by direct sequencing of the rRNA using reverse transcriptase. Another potential method, yet to be implemented for rRNA, involves oligonucleotide-directed mutagenesis (104)of a position involved in a putative tertiary contact. If the interaction is authentic, mutagenesis of the second site to restore complementarity should restore the structural integrity of the RNA. Use of selectable genetic markers in the rRNA gene, such as spetinomycin resistance (73) would facilitate such an approach. Finally, it should be cautioned that identification of stringently obeyed relationships between two positions in an RNA molecule does

214


not necessarily imply that they physically interact. For example, interaction with another molecule, or other sites in the same molecule could dictate similar constraints. In this connection, the site-directed8 mutagenesis approach could distinguish between such possibilities.

ACKNOWLEDGMENTS We are grateful to the following people who shared their unpublished sequences during the course of these studies: E. Blackbiirn, A. Bock, Y-L. Chan, D. Cummings, I. Eperon, H. Leffers, R. Garrett, M. Gray, J . Messing, N. Pace, R. Sederoff, J. Seilhamer, L. Simpson, P. Sloof, M. Sogin, B. Spangler, M. Sugiura, and I. Wool. We thank T. Goldstein for writing computer programs for sequence editing and analysis of coordinated base changes. Finally, we are indebted to M. Waterman for derivation of the conservation expression, and for numerous stimulating discussions.

REFERENCES 1. H. F. Noller, ARB 53, 119 (1984). 2. C. R. Woese, R. R. Gutell, R. Gupta and H. F. Noller, Microbiol. Reo. 47, 621

(1983). 3. R. Brimacombe, P. Maly and C . Zwieb, This series 28, 1 (1983). 4. J. P. Ebel, C. Branlant, P. Carbon, B. Ehresmann, C. Ehresmann, A. Krol, and P. Stiegler, in “Structure Dynamics, Interactions and Evolution of Biological Macromolecules” (C. Helene, ed.), pp. 177-193. Reidel, Boston, 1983. 5. J. F. Thompson and J. E. Hearst, Cell 33, 19 (1983). 6. H. F. Noller and P. H. van Knippenberg, in “Horizons in Biochemistry and Biophysics’’ (F. Palmieri, ed.), Vol. 7, p. 71. (1983). Wiley, New York. 7. H. F. Noller and C. R. Woese, Science 212,403 (1981). 8. A. A. Bogdanov, A. M. Kopylov and I. N. Shatsky, Subcell. Biochem. 7,81 (1980). 9. H. F. Noller, in “Ribosomes: Structure, Function and Genetics” (G. Chambliss, G. R.Craven, J. Davies, K. Davis, L. Kahan, and M. Nomura, eds.), pp. 3-22. Univ. Park Press, Baltimore, 1979. 10. R. A. Zimmennann, see ref. 9, pp. 135-170. 11. C. R. Woese, see ref. 9, pp. 357-373. 12. R. Brimacombe, TZBS 9,273 (1984). 13. H. F. Noller and J . A. Lake, in “Membrane Structure and Function” (E. E. Bittar, ed.), Vol. 6, p. 217. Wiley, New York, 1984. Also, J. A. Lake in Vol. 30 of this series. 14. M. Levitt, Nature 224,759 (1969). 15. A. Rich and U. L. RajBhandary, ARB 45,805 (1976). 16. A. Jack, J. E. Ladner and A. H u g , J M B 108,619 (1976). 17. J. D. Watson and F. H. C. Crick, Nature 171, 737 (1953). 18. E. Chargaff, Erperientia 6,201 (1950). 19. G. E. Fox and C. R. Woese, Nature 256,505 (1975). 20. K. Nishikawa and S. Takemura, FEBS Lett. 40, 106 (1974). 21. C. R. Woese, Sci. Am. 244,98 (1981). 22. L. B. Zablen, M. S. Kissel, C. R. Woese and D. E. Buetow, PNAS 72,2418 (1975). “Complementary-addressed” in Vlassev and Knorre in this volume. [Eds.]

~ ~ - S - L IRIBOSOMAL KE RNA

215

23. C. R. Woese and G. E. Fox, PNAS 74, 5088 (1977). 24. J. Brosius, M. L. Palmer, P. J. Kennedy and H. F. Noller, FNAS 75,4801 (1978). 25. J. Brosius, T. J. Dull, D. D. Sleeter and H. F. Noller,JMB 148, 107 (1981). 26. P. Carbon, C. Ehresmann, B. Ehresmann and J.-P. Ebel, EJB 100,399 (1979). 27. P. Carbon, J.-P. Ebel and C. Ehresmann, NARes 9,2325 (1981). 28. M. Iwami, A. Muto, F. Yamao and S. Osawa, Mol. G e n . Genet. 196,317 (1984). 29. N. Tomioka and M. Sugiura, Mol. Gen. Genet. 191, 46 (1983). 30. Z. Schwarz and H. Kossel, Nature 283, 739 (1980). 31. N. Tohdoh and M. Sugiura, Gene 17,213 (1982). 32. L. Graf, E. Roux, E. Stutz and H. Kossel, NARes 10, 6369 (1982). 33. M. Dron, M. Rahire, J.-D. Rochaix, NARes 10, 7609 (1982). 34. S. Chao, R. R. Sederoff and C. S. Levings, 111, NARes 12, 6629 (1984). 35. D. F. Spencer, M. N. Schnare and M. W. Gray, FNAS 81,493 (1984). 36. J. J. Seilhamer, G. J. Olsen and D. J. Cummings,JBC 259,5167 (1984). 37. H. G. Kochel and H. Kuntzel, NARes 9, 5689 (1981). 38. M. Li, A. Tzagoloff, K. Underbrink-Lyon and N. C. Martin,/BC 257,5921 (1982). 39. F. Sor and H. Fukuhara, C.R. Acacl. Sci. Furis t . 291,933 (1980). 40. I. C. Eperon, S. Anderson and D. P. Nierlich, Nature 286,460 (1980). 41. S. Anderson, M. H. L. D e Bruijn, A. R. Coulson, I. C. Eperon, F. Sanger and I. G. Young, J M B 156,683 (1982). 42. R. A. Van Etten, M. W. Walberg and D. A. Clayton, Cell 22, 157 (1980). 43. M. Kobayashi, T. Seki, K. Yaginuma and K. Koike, Gene 16,297 (1981). 44. I. C. Eperon, J. W. G. Janssen, J. H. J. Hoeijmakers and P. Borst, NARes 11, 105 (1983). 45. R. Gupta, J. M. Lanter and C. R. Woese, Science 221,656 (1983). 46. H. Leffers and R. A. Garrett EMROJ. 3, 161.3 (1984). 47. R. McCarroll, G. J. Olsen, Y. B. Stahl, C. R. Woese and M. L. Sogin, Echem 22, 5858 (1983). 48. P. M. Rubtsov, M. M. Musakhanov, V. M. Zakharyev, A. S. Krayev, K. G. Skryabin and A. A. Bayev, NARes 8,5779 (1980). 49. J. Messing, J. Carlson, G . Hagen, I. Rubenstein and A. Oleson, DNA 3,31 (1980). 50. F. Takaiwa, K. Oono and M. Sugiura, NARes 12,5441 (1984). 51. M. Salim and E. H. Maden, Nature, 291, 205 (1981). 52. Y-L. Chan, R. Gutell, H. Noller and I. G. Wool,JBC 259, 224 (1984). 53. R. Torczynski, A. P. Bollon and M. Fuke, NARes 11,4879 (1983). 54. J. F. Connaughton, A. Rairkar, R. E. Lockard and A. Kumar, NARes 12,4731 (1984). 55. J-A. Kop, A. M. Kopylov, L. Magrum, R. Siege], R. Gupta, C. R. Woese and H. F. Noller, in press (1984). 56. J. J. Hogan, R. R. Gutell and H. F. Noller, Bchem 23, 3322 (1984). 57. F. Sor and H. Fukuhara, NARes 10, 1625 (1982). 58. R. Muller, R. A. Garrett and H. F. Noller,JBC 254,3873 (1979). 59. R. A. Zimmermann and K. Singh-Bergmann, BEA 563,422 (1979). 60. P. Stiegler, P. Carbon, J. P. Ebel and C. Ehresmann, EJB 120,487 (1981). 61. R. Brimacombe, Biochem. int. 1, 162 (1980). 62. C. Zwieb, C. Glob and R. Brimacombe, NARes 9,3621 (1981). 63. A. S. Mankin, A. M. Kopylov and A. A. Bogdanov, F E E S Lett. 134, 11 (1981). 64. G. J. Olsen, R. McCarroII and M. L. Sogin, NARes 11,8037 (1983). 65. J. Atmadja, R. Brimacornbe and B. E. H. Maden, NARes 12, 2649 (1984). 66. M. D. Cole, Ph.D. Thesis, Johns Hopkins University, 1978. 67. D. T. Dubin and C. C. HsuChen, Plusmid 9,307 (1983).

216


68. C. Shannon and W. Weaver, “The Mathematical Theory of Communication,” 1962. 69. H. F. Noller, Bchem 13,4694 (1974). 70. N. M. Chapman and H. F. Noller,JMB 109, 131 (1977). 71. W. Herr, N. M. Chapman and H. F. Noller, J M B 130,433 (1979). 72. D. A. Brow and H. F. Noller,JMB 163,27 (1983). 73. C. D. Sigmund, M. Ettayebi and E. A. Morgan, NARes 12,4653 (1984). 74. B. Spangler and E. Blackburn, submitted (1984). 75. T. L. Helser, J. E. Davies, J. E. Dahlberg, Nature NB 235, 6 (1972). 76. J. B. Prince, B. H. Taylor, D. L. Thurlow, J. Ofengand and R. A. Zimmerman, PNAS 79,5450 (1982). 77. J. Shine and L. Dalgarno, PNAS 71, 1342 (1974). 78. J. A. Steitz, see ref. 9, pp. 479-495. 79. C. M. Bowman, J. E. Dahlberg, T. Ikemura, J. Konisky and M. Nomura, PNAS 68, 964 (1971). 80. B. W. Senior and 1. B. Holland, PNAS 68,959 (1971). 81. K. Rietveld, K. Linschooten, C. W. A. Pleij and L. Bosch, EMBO]. 3, in press (1984). 82. D. A. Peattie, S. Douthwaite, R. A. Garrett and H. F. Noller, PNAS 78,7331 (1981). 83. L. Gold, D. Pribnow, T. Schneider, S. Shinedling, B. Singer and G. Stormo, Annu. Rev. Microbiol. 35, 365 (1981). 84. M. Grunberg-Manago, see ref. 9, pp. 445-477. 85. E. Cundliffe, in “The Molecular Basis of Antibiotic Action” (E. F. Gale, E. Cundliff, P. E. Reynolds, and M. H. Richmond, M. J. Waring, eds.), pp. 402-547. John Wiley, New York, 1982. 86. A. Barta, G. Steiner, J. Brosius, H. F. Noller and E. Kuechler, PNAS 81, 3607 (1984). 87. S. K. Vassilenko, P. Carbon, J.-P. Ebel and C. Ehresmann, J M B 152,699 (1981). 88. P. L. Wollenzien and C. R. Cantor,JMB 159, 151 (1982). 89. J. F. Thompson and J. E. Hearst, Cell 32, 1355 (1983). 90.A. Ross and R. Brimacombe, Nature 281,271 (1979). 91. C. Glotz and R. Brimacombe, NARes 8,2377 (1980). 92. P. Spitnik-Elson, D. Elson, S. Avital and R. Abramowitz, NARes 10, 1995 (1982). 93. P. Spitnik-Elson, D. Elson, S. Avital and R. Abramowitz, A’ARes 10, 4483 (1982). 94. V. Ramakrishnan, M. Capel, M. Kjeldgaard, D. M. Engelman and P. B. MooreJMB 174, 265 (1984). 95. J. B. Prince, R. R. Gutell and R. A. Garrett, TlBS 8,359 (1983). 96. V. D. Vasiliev, 0. M. Selivanova, and V. E. Koteliansky, FEBS Lett. 95,273 (1978). 97. M. Boublik, N. Robakis and W. Hellmann, Eur. J . Cell. B i d . 27, 177 (1982). 98. S. H. Allen and K.-P. Wong, Bchem 253,8759 (1978). 99. M. F. Tam, J. A. Dodd and W. E. Hil1,JBC 256,6430 (1981). 100. R. R. Gutell, W. Weisburg, H. Oyaizu, Y. Oyaizu, H. F. Noller and C. R. Woese, manuscript in preparation (1985). 101. I. C. Eperon, J. W. G. Janssen, J. H. J. Hoeijmakers and P. Borst, NARes 11, 105 (1983). 102. A. mug, J. Ladner and J. D. De Robertus,JMB 89,511 (1974). 103. G . Olsen, Ph.D Thesis, University of Colorado, 1983. 104. M. J. Zoller and M. Smith, NARes 10,6487 (1982). 105. R. A. Garrett, B. Vester, H. Leffers, P. M. Sorensen, J. Kjems, S. 0. Olesen, A. Christensen, J. Christensen and S. Douthwaite, in “Gene Expression” (B. F. C. Clark and H. U. Petersen, eds.). Alfred Benzon Symposium 19, Copenhagen, 1984.

SV40 Promoters and Their Regulation GOKULC. DAS,I* SALILK. NIYOCI?AND NORMAN P. SALZMAN~ I

I ~ b o r u t o r yof Biology (If Vinrses, N IA 1D , Nut i o nu1 Institutes of Heulth, Bethesdu, Maryland Biology Divisiori, Oak Ridge National Lnborutclnjt, O u k Ridge, Tennessee

I Regulatory Region of SV40 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. DNA Binding Property of T Antigen . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. Regulation of Transcription . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IV. Conclusions. ............................................... References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

218 227 229 232 232

Simian Virus 40 (SV40) is a small oncogenic virus of the papovavirus group that transforms nonpermissive rodent cells in animals or in cell cultures to a tumorigenic state (for general reviews, see 1). It undergoes a lytic cycle of replication in “permissive” monkey kidney cells. The virus contains a small double-stranded circular DNA genome of 5243 base-pairs (bp),whose sequence is completely known (2,3).The viral genome is complexed with four cellular histones H2A, H2B, H3, and H4, and appears in the form of a minichromosome inThe five or six genes in the viral genome are side the infected cell (4). transcribed by cellular RNA polymerase I1 (5, 6). The posttranscriptional processing of SV40 transcripts is similar to that of cellular RNAs and involves “capping,” polyadenylation, splicing and internal methylation (7-11). Many of the studies with SV40 have focused on an understanding of the mechanism of cellular transformation. SV40 has also served as a model to understand some of the basic mechanisms of gene expression in eukaryotic systems. * Present address: Division of Molecular Biology and Biophysics, School of Basic Life Sciences, University of Missouri-Kansas City, Kansas City, Missouri 64110. t Operated by Martin Marietta Energy Systems, Inc. under contract DE-ACOS840R21400 with the U.S. Department of Energy. 217 Progress in Nucleic Acid Research and Molecular Biology, Vol. 32

Copyright 0 1985 by Academic Press, Inc. All rights of reproduction in any form reserved.

218

GOKUL C . DAS ET AL.

The lytic cycle of SV40 in permissive cells is expressed in two temporally regulated phases (1). “Early” gene expression begins shortly after infection and gives rise to two tumor antigens called T and t (“large T” and “small t”). Synthesis of these proteins continues in a regulated fashion throughout the lytic cycle. High levels of expression of the “late” genes is generally delayed until the onset of viral DNA replication at about 12-16 hours postinfection, and the synthesis of “large T” is required for viral DNA replication. Lateregion mRNA, which is abundant during the late phase, is needed to produce the viral capsid proteins VPl, VP2, and VP3 that are used to form the progeny virions ( 12) .In addition to viral capsid proteins, a minor protein, called “agnoprotein,” is also produced at late times (13, 1 4 ) . One of the early gene products, large T, interacts with the origin region and plays a regulatory role in the developmental cycle of SV40, including autoregulation of early transcription, initiation of DNA replication, and stimulation of late transcription (I5-22). The temporal regulation of the early and late genes allows for the developmental expression of the genes needed first for establishing the viral infection and DNA replication, and then for synthesis of structural proteins needed to encapsidate the newly synthesized viral DNA. A number of general reviews of papovaviruses, which include data on SV40, appeared a few years age (I, 23, 2 4 ) . Since then, we have made tremendous progress toward understanding the general organization of the regulatory region and in identifying the promoter elements and their roles in the regulation of transcription from these promoters. The purpose of this article is to document the current status of our understanding of the complex regulatory region of SV40. It may serve as a model of the regulatory mechanisms that operate in many other eukaryotic systems.

1. Regulatory Region of SV40 The regulatory region of SV40 (Fig. 1) is 420 bp long and is contained in the HindIII-HpaII fragment located between nucleotide positions 5171 and 346 ( 25) .This DNA is positioned between the coding sequences of the two sets of genes that are transcribed divergently at early and late times in infection. Most of this region inside the infected cells is generally devoid of nucleosomes (2 6 ).This region contains sequentially a perfect 27-bp palindrome, a 17-bp (A + T)-rich sequence, three copies of a (G + C)-rich 21-bp repeat, and two copies of a 72-bp repeat at the extreme upstream segment (2, 3 ) . The segment containing the “core” origin of replication (ori) is characterized by the 27-bp palindromic structure flanked on either side by

219

SV40 PROMOTERS Late Transcriptional Control a) Lus b) L,,

Early Transcriptional Control a) EarlyEarly (E-E) b) Lale-Early (L-E)

I

I

+?I

I

5' Ends of Late RNAs

Llffl

432.5

5' Ends of Early RNAs

T-antigen Binding Sites Replication Origin (Ori)

4-1

I I

-I1

I11

I

FIG.1. Regulatory region of the SV40 genome (HindIII-HpaII fragment between nucleotide positions 5171 and 346). Constituents of this region are 27-bp palindrome (position 5231 to 14), 17 b p (A + T)-rich region containing the TATA box (m), one imperfect and two perfect 21-bp repeats (0) containing two of the (G + C)-rich motifs (m), and the two 72-bp repeats (El). The 5' ends of the early RNAs at early (E-E) and late (L-E) infection (three arrows) and two major late RNAs initiating at positions 185 (L185) and 325 (L3B) (single right arrows) are marked. Regions involved in DNA replication or in controlling the efficiency of early and late promoters are marked by bold and thin lines, where the former represents the principal and the latter the auxilliary sequences. The boundaries are approximate and the major uncertainties are marked by question marks (?). ( 0 )indicates a TATA-box-like region that controls the efficiency of initiation of L3s RNAs. Approximate boundaries of the three T-antigen binding sites (I, 11, and 111) are marked, and the degree of boldness reflects the binding affinity to that region.

short segments (16,27,28). Sequences corresponding to the 5' ends of early and late mRNAs are located within this region, as are the promoters for early and late transcription (29, 30). Three closely spaced binding sites (I, 11, and 111) for large T have been identified within this region and are thought to be involved in the initiation of viral DNA replication and in the regulation of gene expression (1.5-22). T h e SV40 early promoter is a prototype RNA polymerase I1 promoter, the organization and control of which are better understood than those of the late promoter (31-39). T h e latter, on the other hand, does not have the prototype promoter-like elements in the segment of DNA that contains the entire late promoter (40-45). The regulatory region of SV40, linked in either orientation, can drive other heterologous genes (21,46). Recently, this property has been utilized in mutational analysis of the different promoter elements within this region by fusing it to marker genes that can readily be assayed in transient in

220


uiuo assays. Large T has been purified and is found to bind to the origin region and regulate early transcription in uiuo (15,31,47-49). These experimental observations have facilitated studies of the regulation of two different classes of promoters that share some common elements, and they are discussed in the following section.

A. Organization of the Early Promoter The SV40 early promoter contains at least three spatially distinct elements: (1)two 72-bp repeats comprising the enhancer sequence; (2) three 21-bp repeats, each containing two (G + C)-rich motifs; and (3) the TATA box or Goldberg-Hogness box. The functional significance of each element is discussed below.

1. “ENHANCER” SEQUENCE The working definition of an enhancer is a cis-acting short stretch of DNA, sometimes in the form of repeats, that can potentiate the transcription of diverse genes independent of its position and orientation with respect to the RNA initiation site. It was first identified in SV40 as a 144-bp segment made up of two 72-bp repeats located between positions 107 and 251 (34,35). Deletion of only one copy of the two repeats does not abolish its activity, whereas deletion of both leads to a drastic decrease in early transcription as measured in transient assays (34, 35). Transcription is largely restored by reinserting the 72-bp repeats in either orientation at a variety of positions as far as a few kb from the RNA initiation site (50, 51). This element also increases the level of transcription of different cellular genes when at least one 72-bp sequence is present in the same plasmid (52-55). Initial in uitro experiments failed to reproduce the effects of the enhancer sequence that are observed in uiuo. It was later demonstrated that the element, at least partially, functions in uitro both on homologous and heterologous promoters (56-58). Since one of the two 72-bp sequences of SV40 is sufficient for full augmentation of early transcription, it is not understood why this element is duplicated. The enhancer regions of several other viruses also contain repeats, indicating that duplication of a sequence may be important for the functioning of this element. The SV40 enhancer appears to function most efficiently as an intact unit, but fine-structure mapping indicates that a portion of the sequence may have a more critical role (32, 35, 59). Besides SV40, enhancer sequences are also present in other viral systems such as polyoma virus (60-62), JC virus (63), bovine papilloma virus (64), adenovirus (65,66), BK virus (67) and also within the

22 1

sv40 PROMOTERS

long-terminal repeats (LTRs) of the RNA viruses or retroviruses (68, 69). [A comprehensive review of all the enhancers is contained in a recent symposium (70).]This element still functions when enhancers are interchanged; e.g., replacement of the 72-bp segments of SV40 by the 72173-bp repeat of Moloney murine sarcoma virus generates a viable virus (68).Enhancer sequences are also present in many cellular genes, e.g., the immunoglobulin heavy-chain gene (71-73), in mouse and human genomic libraries (67), and even sometimes within introns (71, 73). By comparing sequence homology between the enhancers of SV40 and those of other DNA and RNA tumor viruses, the following “consensus” enhancer sequence has been derived (59): (G)TGGAAA/TTT(G).This sequence alone is not enough for the enhancing activity, but may serve as a core that interacts with regulatory factors. Presently, little is understood about the molecular basis of enhancer function, although several models have been proposed. Since this element functions in a manner independent of position and orientation, it is not a promoter element in a strict definition of the term. It may provide a bidirectional entry site for the RNA polymerase I1 (50, 60, 74) or may activate promoters by controlling the chromatin structure (32, 75). Support for the latter idea comes from the observation that the 72-bp repeat sequences are sensitive to various nucleases (76, 77) and are devoid of nucleosomes (26). Lastly, enhancer sequences may create a microenvironment for the DNA structure that is recognized by RNA polymerase 11. It has been shown recently that negatively supercoiled SV40 DNA contains Z-DNA segments within the transcriptional enhancer sequences that might be related to transcriptional activation (78). It is interesting to note that the functional activities of different enhancers vary depending upon the host (69)and/or on the tissue (72, 73, 79, 80), into which they are introduced. This host and tissue dependence could either be a function of the primary sequence of the enhancers or it could involve specific cellular factors which recognize the enhancer elements, or both. Recent studies demonstrate the presence of cellular factors that are necessary for enhancer function (81).

2.

UPSTREAM

(G

+ C)-RICH Murws

There are three 21-bp repeats each containing two of the (G + C)rich motifs of the structure GGGCGGRR (where R is purine) between np 40 and 103. The C-rich strand of the sequence occurs in the polarity of early mRNA whereas the G-rich strand occurs in that of the late mRNA. This element, located about 50 bp upstream from the major

222


RNA initiation sites early in infection, consitutes the second important element of the early promoter (31-33,37, G. C. Das, unpublished observation). But studies of the various sequences contained in this region have led to inconsistent conclusions. It was concluded from a study of deletion mutants that at least two of the GGGCGG-repeats are needed for maximal early promoter function and that any two of These results differ from that obthe six are equally effective (8.2~). tained using a different set of mutants and assay procedures where SV40 early promoter activity was found to decrease by at least 3- to 4fold when four of the six (G + C)-rich motifs are deleted (37, G. C . Das, unpublished). Moreover, studies with point mutants in the (G + C)-rich motifs demonstrate that a single C + T transition in any of these repeats is sufficient to impair promoter function of the 21-bp repeat region in d u o . However, the effect is not reproduced in uitro (37,57). Another interesting characteristic of this (G + C)-rich element is that it can function bidirectionally like the 72-bp repeat (37,43,57). The early promoter activity of the complete SV40 origin region is reduced only to one-half when the 21-bp repeat is reversed in its orientation (37, 57). A direct demonstration that this element is an important part of the early promoter comes from the identification and purification of a SV40 promoter-specific factor Spl, which binds specifically to this region and is required for early transcription in a reconstituted in uitro system (82,83). DNA sequences located upstream of the TATA box have been analyzed in various viral and cellular promoters (84).No obvious sequence homology is apparent, but it appears that these regions are very rich in G C pairs. This region is thought to be equivalent to the recognition sequences (-35 bp region) of the bacterial promoter (85, 8%). It remains to be determined whether the recognition of this region by RNA polymerase is mediated by factor(s) such as Spl. The binding of such a factor to this region might cause the conversion of SV40 chromatin into an active conformation, capable of initiating transcription in both orientations. Such a structural change in the template is recognized b y the creation of DNase I hypersensitive sites (8%). The bidirectionality of this unit also seems to be necessary for its involvement in late transcription, as discussed later. The effect of the 21-bp repeat region on late transcription decreased strikingly with increasing distance from the start sites, although it is still detectable over a distance of 220 bp (57).It is also remarkable that the SV40 21bp repeat region can activate transcription in uiuo and in uitro from heterologous TATA box elements (86, R. Hen and N. Miyamoto, unpublished). Thus, the 21-bp repeat region shares with the enhancer

sv40 PROMOTERS

223

element three characteristic features: bidirectionality, action at a distance, and stimulation of heterologous promoters (37,43, 57,86, 87).

3. THETATA Box OR GOLDBERG-HOGNESS Box The third element of the SV40 early promoter is the TATA sequence or the Goldberg-Hogness box, TATA(A/T)A(T/A),between positions 16 and 21. Unlike other TATA boxes, this element is embedded in a 17-bp (A + T)-rich region. This element is located at about 25-35 nucleotides upstream from the RNA initiation sites not only in the SV40 early promoter but also in many other eukaryotic promoters. It functions in the following way. (1)It serves as one element in fixing the 5 end of the RNA both in uizjo and in uitro. Deletion of this sequence makes heterogeneous the start sites of RNA synthesis, and removal of sequences downstream from the TATA box generates a new start site located about 25 nucleotides downstream from the TATA sequence (35, 36, 81a, 88, 89). (2) It controls the efficiency of transcription, as point mutations within the TATA box of SV40 drastically reduce the level of transcription (90). A similar inhibition of transcription was also observed with the TATA boxes of ovalbumin, conalbumin, silk fibroin, and with the adenovirus major late promoter (91-94). However, this depression in the level of transcription does not necessarily create a new transcription start site. These are clear indications that the transcription machinery probably interacts with the TATA box region. This proposition seems to be justified since a factor, present in HeLa cells, interacts with the TATA box region and is essential for the in vitro initiation of transcription (9.5). Analysis of deletion mutants that include the TATA box has, in many instances, led to the conclusion that the TATA box does not control the efficiency of transcription in uiuo and in citro ( 3 2 , 4 3 ,5 1 ) . Observations on deletion mutants should be interpreted with much caution, as a large deletion not only may remove other regulatory sequences together with the TATA box but also may bring some other sequences together. These novel junctions may create other functional elements having different regulatory roles. We think that, regarding the role of the TATA box, the data obtained with point mutations or micro deletions within the TATA box are more meaningful. Thus, the TATA box serves a similar function, at least partly, as the “Pribnow” box of the bacterial promoter, though the latter is located somewhat closer to the RNA initiation sites. Another inconsistency in the literature is observed when in uitro and in uiuo results are compared. For most promoters, including the SV40 early promoter, alterations of DNA sequences in the vicinity of

224

GOKUL C. DAS ET AL.

the TATA box have a pronounced effect on transcription in uitro (96loo),but the important sequences controlling the efficiency in uiuo are located further upstream (35,36,101-103). These inconsistencies can be explained by the following consideration. Although the template in the in vitro system may form a nucleoprotein complex with the primary DNA sequences the proteins present in the extract (104), probably play a more dominant role in the recognition process. But in uiuo, either in transient or long term assays, the template DNA is in a more ordered nucleoprotein structure than the DNA that is transcribed in a whole cell extract. The SV40 origin region contained in plasmid pBR322 assumes a nucleosomal conformation in COS cells shortly after transfection that is similar to that in SV40 chromatin prepared late after infection of BSC-1 cells, as judged by the presence of similar DNase I hypersensitive sites (105).Late proteins do not modify the chromatin structure of this region (105). In prokaryotic systems, the essential promoter elements, such as the “Pribnow box” (-10 bp region) and the “-35 bp region,” are located within 50 bp upstream from the RNA initiation site. Mutational analysis shows that E. coli RNA polymerase makes simultaneous contact with these two regions (85),and a spacer length of 17 bp fits very well with the “consensus” length for maximum efficiency of gene expression (106).The eukaryotic TATA box is structurally very similar to the prokaryotic Pribnow box and is located slightly farther upstream from the RNA initiation site. On the other hand, the six (G + C)-rich sequences may act as recognition sequences for RNA polymerase I1 and/or other factors and are thought to be analogous to the -35 bp region of the bacterial promoter (32,33).Thus, the region between the TATA box and the (G + C)-rich sequences might be termed “spacer” by analogy with the prokaryotic promoters. The molecular details of the interaction of RNA polymerase I1 with this region are still poorly understood. From mutational analysis, it appears that a narrow region between the TATA box and the (G + C)-rich region from position 38 to 41 is an important domain of the early promoter in uiuo and in uitro (107). Analysis of deletion and insertion mutants in this region indicates that RNA polymerase I1 or other transcription factors interact simultaneously with the TATA box and the (G + C)rich region (107).Furthermore, this study suggests an optimal spacing requirement between these two domains, as observed in prokaryotic systems, which is accurately reflected only in uiuo (107).

4. OVERLAPPING EARLY PROMOTERS Late in lytic infection, there is a shift in the initiation sites used for early transcription (108). Before the onset of viral DNA replication,

SV40 PROMOTERS

225

transcripts appear with the 5’ ends mapping downstream from the 27bp inverted repeat at the origin region (ori) from positions 5231 to 14. The positions of the 5‘ termini are determined by the TATA signal at about 25 bp upstream. After the onset of viral DNA replication, production of downstream RNA decreases and a new set of early region RNAs appears, with the 5‘ ends mapping about 20 bp upstream from the TATA box, and they become the predominant species late in infection (39, 108). It is not known whether these RNAs are translated into proteins, or if they play any special role in the biological regulation of viral growth. This finding demonstrates that the early promoter is more complex than previously anticipated, and that it may be composed of two overlapping promoters, one early-early (E-E) operative very early, and the other late-early (L-E) operative later in infection. Attempts have been made recently to dissect the two promoters by mutational analysis (38,90).The elements of the E-E promoter appear to be the TATA box, all six (G + C)-rich motifs, and the 72-bp enhancer, whereas only the last four (G + C)-rich motifs and the 72bp enhancer constitute the elements of the L-E promoter (38, 90). A TATA box-like sequence has not yet been identified for the L-E promoter. In one study, transcription from the L-E promoter remained unaffected in the absence of the 72-bp repeats (39). Synthesis of the E-E and the L-E RNAs can be initiated in the absence of DNA replication (38,90).Mutations in the TATA box that reduce DNA replication and transcription from the E-E promoter dramatically increase transcription at the L-E promoter (38, YO). However, other studies show that DNA replication or amplification is essential to get enhanced transcription from the L-E promoter (39).The role of individual (G + C)-rich motifs in the functioning of both promoters remains to be determined. The physiological function of the L-E RNA and the regulation of the two early promoters are not understood. Since transcription from the upstream site proceeds through ori, it is possible that the upstream promoter plays a role in activating viral DNA replication, either by opening the DNA helix at ori thereby permitting initiation by a DNA primase, or by cleaving the L-E RNA at ori to serve as an RNA primer (39,109,110).In this RNA, there is an open frame upstream from the start site of the T-antigen coding sequences (39).This additional reading frame could code for a protein of 23 amino acids, but it has not yet been identified. This small protein may provide a function needed late in infection. A similar protein (“agnoprotein”), which is encoded within the leader segment of the SV40 16s late RNA, has recently been identified (14).

226

GOKUL C. DAS ET AL.

B. Organization of Late Promoters The SV40 late promoter uses different signals and a different mechanism for its activation. Regulatory elements, which may be involved in modulating the efficiency of late transcription as well as in the temporal expression of the late genes, remain to be defined. SV40 late transcription is initiated at a number of sites over a region of about 220 bp, the major one being at position 325. A second site at 185 is used inefficiently in vivo, but is an important start site in vitro in HeLa cell extracts (30, 41 ). Mutational analysis with a series of deletion mutants indicates the existence of two overlapping promoters that in turn overlap with the early promoter (32,43,45;G. C. Das, unpublished). The promoter for the late RNA initiating at position 185 (LIB5) requires DNA sequences 100 bp upstream from the initiation site and this includes the 21-bp repeats (32, 45). This is also a region of the early promoter where a promoter-specific factor Spl binds (83).For the major late promoter that initiates transcription at 325, two domains that control the efficiency of transcription from this promoter have been identified (41-45). One of these domains is an ll-bp DNA sequence GGTACCTAACC located about 25 bp upstream of the RNA initiation site (41).It acts like a surrogate TATA box and seems to play an equivalent role (41).The “consensus” sequence of the TATA box derived from an analysis of 41 genes is TATATAATA. There is a high conservation of sequence specificity at positions -29, -28, -27, -26, and -24 (where the T at the 5’ end is at -29) (8 8 ).When single base changes are introduced into the surrogate TATA box at -29 and -28 from the 5‘ start site, the level of transcription in vitro is depressed to one half that of the wild type (11 1 ) . Stepwise changes of this sequence toward the same sequence that drives the SV40 early gene result in a stepwise enhancement of transcription ( 111 ). Another critical domain of the L325 promoter operative in vitro is located far upstream inside the (G + C)-rich segment ( 4 2 , 4 3 , 4 5 , 5 7 ) . Three of the six (G + C)-rich motifs are required for full expression of L325 transcription. Additional deletions of the (G + C)-rich motifs sharply reduce the promoter activity. The region containing the six (G + C)-rich mofits constitutes an essential part of both of the E-E, L-E, L325, and LIB5 promoters (38, 39, 42-45, 57; G. C. Das, unpublished). The effect of this region decreases strikingly with increasing distance from the start sites (86, 87). It seems very complex how a single 21-bp region can be shared by the E-E, L-E and late overlapping promoters. To answer this question more thoroughly, mutational analysis of this (G + C)-rich segment is required.

227

sv40 PROMOTERS

Transcription of late genes in uitro requires sequences spanning 250 or more b p from the major initiation site (42, 43, 45, 57), unlike early genes that require only about 100 b p upstream from the initiation site for promoter activity (31-39). It is not understood why the early and late promoters have such different sequence requirements, but these differences may play a role in the temporal regulation of SV40 transcription. Whether the enhancer sequence that plays a crucial role in in uiuo initiation from the early promoter has a less important role for efficient initiation from the late promoter remains to be demonstrated. However, it is reported that the enhancer stimulates transcription from L325 by about 2- to 3-fold (44,57).

II. DNA Binding Property of large T Antigen Of the two early gene products of SV40, the role of “small t” in the lytic cycle is not known. The other, “large T” antigen, plays a vital role in processes such as initiation of viral replication, autoregulation of early transcription, and stimulation of late transcription (15-22). Large-T is a phosphoprotein of molecular weight 96,000 and has an associated ATPase activity (112).But among its biochemical properties, it is the sequence-specific DNA binding activity that has been of intense interest, as this property of T immediately suggests a possible mechanism for its mode of action as a regulatory protein. For binding studies, T has been purified from a variety of sources such as SV40-infected monkey cells (lytic T) (113-11 7), SV40-transformed human cells (SVSO) (118, 119), HeLa cells infected with a hybrid virus (Ad2 D2) of adeno and SV40 (D2) (15,117)or with the hybrid virus Ad-SVR6 (T) (120). Most of the initial binding experiments were carried out with the hybrid D2T protein comprised of approximately 10% of an Ad2 protein at its amino end and 90% of the authentic T antigen at the carhoxy end (121). DNase “footprinting,” filter-binding, and methylation-protection assays demonstrate that D2 protein binds at three closely spaced sites, designated as sites I, 11, and I11 on SV40 DNA over a region of 120-140 bp (15)(Fig. 1).This region spans the origin of DNA replication and a part, if not all, of the early and late promoters of SV40. In a separate study, an additional binding site of this protein has been located between site I and the AUG codon of T (117).Binding studies with authentic T demonstrate the presence of site I and site I1 but have failed to detect site III; this

+

228


is probably due to the limited amount of T used in the in vitro assay

(122). The binding affinity of T for these sites increases in the order site I>>site II>site I11 (15). Although the initial experiments favored a cooperative nature of the interaction (15,117), it has been questioned recently on the basis of studies with mutants in which different binding sites were sequentially deleted (123).Although similar in many respects, the binding of D2 protein is more sensitive to NaCl concentration than that of wild-type T (124).Site I1 binding of authentic T antigen spans the 27-bp palindromic sequences at the origin of viral replication and can be subdivided into two subsites, IIA and IIB, IIA being the early portion of site 11. Affinity of IIA for T is greater than that of IIB, indicating that T binding to a major portion of the replication initiation sequence (Site 11)is the product of at least two interactions (122). Essential nucleotide contacts between the SV40 large T and binding sites I and I1 on the SV40 genome have been inferred from in uitro methylation- and ethylation-interference experiments (123). Each binding site contains two clusters of guanine contacts separated by about 4 turn of the helix, and all four clusters of G contacts are localized within a limited region of the DNA duplex. Three of the G clusters are accessible from one face of the double helix, while the fourth group of contacts, located at the early side of site I, is rotated by approximately 30" relative to the other three. The sequence of these recognition sites corresponds to the consensus family G/TGGGC but not all of these pentanucleotide blocks contain essential contact points (123-125). The molecular details of the binding with site I11 have not yet been well defined, although it is known that this site overlaps with the crucial 21-bp repeat region of the early promoter (15,117). Electron microscopy of the protein-DNA complex suggests that a tetrameric form of large T is bound at each site, although in solution purified D2 protein exists in a variety of aggregates starting from monomer to dodecamer (126). The binding of T does not alter the helical pitch of the DNA (127). It is not known whether the DNA binding property of T is dependent on its various physical states, or whether there is any biochemical or structural difference between old and new T antigen. In only one study have newly synthesized T molecules been shown to have a greater affinity for the SV40 origin region than the older ones, and T molecules phosphorylated to different extents bind DNA with altered affinities (128).

229

SV40 PROMOTERS

111. Regulation of Transcription

A. Regulation of Early Transcription Early genes of SV40 that code for t and T antigens are transcribed before the onset of viral DNA replication. T continues to be synthesized in an exponential fashion up to 96 hours and accumulates in the cell nucleus (129, 130). Temperature-sensitive mutants in this gene (tsA mutants) produce a thermolabile T that fails to function in the initiation of DNA synthesis at a nonpermissive temperature, indicating its positive role in DNA synthesis (129, 130). On the other hand, the rate of synthesis of this protein is more rapid in cells infected with tsA mutants under restrictive conditions than in a corresponding infection with wild-type virus (130).This strongly indicates that T regulates the transcription of early genes, i.e., its own synthesis. Other studies confirming autoregulation by T show that tsA mutants overproduce early RNA even at permissive temperatures and the effect is magnified greatly with a shift to nonpermissive temperatures (19, 131). The loss of autoregulation is at the level of the inhibition of initiation of RNA synthesis. Mutational analysis in vivo indicates that the binding to site I, which has the highest affinity for T, is necessary if not sufficient, for Tmediated autoregulation (132).The role of T in autoregulation is demonstrated directly in vitro by its suppression of early transcription (47-49). Two other observations, one in vitro with a mutant in which binding site I1 is replaced with a 33-bp bacterial DNA fragment, the other in vivo with site I and site I1 deletion mutants, have clearly demonstrated the involvement of these sites in autoregulation (49).It was thought earlier that binding site 111, having the lowest affinity for T, might not be involved in early transcription, since deletion of this site did not affect the normal development of the virus (132).Mutational analysis with base substitution in this site demonstrates that this site also plays a role in the autoregulation, probably by binding T to this region (107). The following lines of evidence indicate that the regulation by T occurs at the level of the initiation of transcription rather than during chain elongation or termination. (1)In vitro transcription from a heterologous promoter with T binding-sites inserted at a downstream location is not blocked by T (31). (2) Binding site I11 overlaps with a crucial part of the early promoter that might serve as the recognition sequence for RNA polymerase I1 (a DNA-directed RNA polymerase,

230

GOKUL C. DAS ET AL.

EC2.7.7.6). Thus, T antigen and RNA polymerase I1 may compete for the same site, resulting in a blocking of transcription initiation (107). ( 3 ) Binding of T may increase the stability of this region, making it unfavorable for the local melting of the double helix before initiation takes place (31).Whatever the exact mechanism, it appears that binding to all three sites is needed for maximum repression of early transcription (49, 107, 132). The mechanism of repression by the A repressor involves a cooperative binding of repressor molecules to sites that compete directly with RNA polymerase for DNA binding (133).However, for inhibition of transcription, occupation of either site is sufficient. Although early studies indicated that binding of T to these sites is cooperative, more recent studies suggest the opposite (15, 117, 123). The site of RNA polymerase binding to the SV40 promoter is not yet mapped, and whether the binding is mediated by some factors remains to be determined. At least one promoter-specific factor, Spl, binds to the (G C)-rich 21-bp repeats which also constitute a part of the T binding site I11 (83).A recent study indicates that the transcription machinery may make a simultaneous contact with the TATA box and the (G + C)-rich region in binding site I11 (107).It is not known whether T competes with S p l or with polymerase I1 for binding sites. There might be another level of regulation of early transcription b y T antigen. Late in infection, new start sites are used for the synthesis of early RNA transcripts (L-E RNA). These are about 40 bp upstream from the E-E RNA transcripts, and some are at the junction of T binding sites I1 and I11 (39,90,108).It was thought earlier that this switchover is mediated by the binding of T at sites I and I1 (90, 108). One recent study indicates that it is not the direct consequence of the presence of T, but that DNA replication is obligatory for this early-tolate switch. Therefore, T is only indirectly involved because of its role in DNA replication (39).It is not known whether the L-E RNA is fully used as template that is translated into protein, or serves some other physiological function, as discussed earlier (Section I,A,4). The relationship between the E-E and L-E promoters and how T antigen regulates the two in the developmental cycle of SV40 are not understood. It is clear that transcription from these two overlapping promoters is coordinated with two stages of virus infection, those events that occur before and those after the onset of viral DNA replication. In addition to its effects on viral functions, T induces a variety of host-cell metabolic functions, including the syntheses of ribosomal RNA, histones, and enzymes involved in DNA metabolism. Perhaps, the high efficiency of E-E promoters is needed not only to supply

+

sv40 PROMOTERS

231

enough T for the initiation of viral D N A replication but also to direct the host cell into a metabolic state that is primed for virus multiplication. The transcriptional gear-shift from the E-E to the L-E promoter may function to adjust the rate of T production to appropriate levels at two different stages of virus growth.

B. Regulation of Late Transcription Increased transcription of the late region begins after the onset of DNA replication and rises very rapidly to make three capsid proteins VP1, VP2, and VP3 (12). The regulation of this promoter is less understood compared to the early promoters. Two different models, not mutually exclusive, of regulation of the late genes are conceivable. One is negative control by a repressor, the other positive control by T. In the former, a protein repressor associated with the viral DNA in the virion may block the initiation or elongation of the late RNA chain until early transcription and DNA replication generate a repressorfree template (134). In the latter, T has either a direct role or an indirect role, because of its role in DNA replication, in activating late transcription (134). Genetic analysis with tsA mutants shows that T is required for late transcription. I n one study, infection with either tsA58 virus or its DNA at the restrictive temperature led to substantial amounts of two early RNAs, but no late RNAs, whereas in cells infected with wildtype virus or its DNA, a significant amount of late mRNA was synthesized even at 7.5 hours after infection. These results indicate that T plays an active role in late transcription (134). One factor in the accumulation of late RNA at late times of infection is due to the amplification of viral genomes, a process in which T is intimately involved. T can also substantially stimulate late promoter activity directly and independently of viral DNA replication (21, 22). Different experimental approaches have led to this conclusion. In one, genome amplification was blocked either by using origin-defective mutants of SV40 or by using inhibitors of DNA synthesis. In the other, normal or altered T, from different cell lines such as COS, C2, C6, and C11, was allowed to exert its regulatory influence on late gene expression. Both kinds of experiments demonstrate the direct role of T in activating late transcription by several fold (21, 22). It is possible that the DNA binding properties of T are involved in late transcription (21,22), but not those required for the initiation of DNA replication. T antigen may stimulate late transcription by positively activating the transcription unit, i.e., by directing the RNA polymerase to a late

232

GOKUL

C. DAS ET AL.

promoter that is unresponsive in its absence. In uiuo competition experiments further suggest the direct interaction of the template DNA with T for the activation of late transcription. Using various competitor fragments from the late region of SV40, an induction of late gene expression can be shown ( J . Brady and G. Khoury, personal communication). This suggests the possibility that cellular transcriptional factors may negatively regulate SV40 late transcription (22). If so, then a positive role for T might be the removal of this repressor by interacting with the template.

IV. Conclusions The two promoters of SV40 are similar to many other eukaryotic promoters, which makes an understanding of their organization and regulation of much importance to eukaryotic molecular biology. One of the early gene products, T antigen, has been purified and it has been possible to study its regulation of transcription in uitro. This DNA-binding protein is thought to be typical of many others present in eukaryotic cells that recognize specific DNA sequences and help to determine which RNAs are to be made in different tissues, or at different stages of development of an organism. We think that a thorough understanding of the regulation of transcription by T antigen will elucidate how a DNA-binding protein acts to regulate gene expression.

REFERENCES 1. J. Tooze, in “DNA tumor Viruses: Molecular Biology of Tumor Viruses (J. Tooze, ed.), Part 2. Cold Spring Harbor Lab., Cold Spring Harbor, NY., 1980. 2. V. B. Reddy, B. Thimmappaya, R. Dhar, K. N. Subramanian, B. S. Zain, J. Pan, K. Ghosh, M. L. Celma and S. M. Weissman, Science 200,494 (1978). 3. W. Fiers, R. Contreras, G. Haegeman, R. Rogiers, A. Van de Voorde, H. Van Heverswyn, J. Van Herreweghe, G. Volckaert, and M. Ysebaert, Noture 273, 113 (1978). 4. J. D. Griffith, Science 204,264 (1979). 5. A. H. Jackson and B. Sugden,J. Virol. 10, 1086 (1972). 6. H. Handa, R. Kaufman, J. L. Manley, M. L. Gefter and P. A. Sharp,JEC 256,478 (1981). 7. M. Revel and Y. Groner, ARE 47, 1079 (1978). 8. Y. Aloni, R. Dhar, 0. Laub, M. Horowitz and G. Khoury, PNAS 74,3686 (1977). 9. Y. Aloni, This Series, 25, 1 (1981). 10. F. Crick, Science 204,264 (1979). 11 . A. R u i n and J. Friedman, This Series, 25,33 (1981). 12. N. H. Acheson, in “DNA Tumor Viruses” (J. Tooze ed.), Part 2, 2nd ed. Cold Spring Harbor Lab., Cold Spring Harbor, NY, 1980. 13. R. Dhar, K. N. Subramanian, J. Pan and S. M. Weissman, €“AS 74, 827 (1977). 14. G. Jay, S. Nomura, C. W. Anderson, and G. Khoury, Nature 291,346 (1981).

Sv40 PROMOTERS

233

15. R. Tjian, Cell 13, 165 (1978). R. Myers and R. Tjian, PNAS 77,6491 (1980).

15a. 16. 17. 18. 19. 20. 21. 22.

D. Shortle and D. Nathans, JMB 131,801 (1979). P. Tegtmeyer, J . Virol. 10,591 (1972). P. Tegtmeyer,]. Virol. 15,613 (1975). G. Khoury and E. May,J. Virol. 23, 167 (1977). S. I. Reed, G. R. Stark and J. C. Alwine, PNAS 73,3083 (1976). J. M. Keller and J. C. Alwine, Cell 36, 381 (1984). J. Brady, J. Bolen, M. Radonovich, N. Salzman and G. Khoury, PNAS 81, 2040 (1984). 23. G. C . Das and S. K. Niyogi, This Series, 25, 187 (1981). 24. P. Lebowitz and S. Weissman, Curr. Toil. Microbiol. Zmmunol. 87,43 (1979). 2.5. A. R. Buchman, L. Burnett and P. Berg, in “DNA Tumor Viruses” (J. Tooze, ed.), Part 2, pp. 799-829. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY, 1980. 26. E. Jakobovitz, S . Bratosin and Y. Aloni, Nuture 285, 263 (1980). 27. K. N. Subramanian and T. Shenk, NARes 5, 3635 (1978). 28. K. N. Subramanian, R. Dhar and S. M. Weissman,JBC 252,355 (1977). 29. V. B. Reddy, P. K. Ghosh, P. Lebowitz, M. Piatak and S. M. Weissman,J. Virol. 30, 279 (1978). 30. P. K. Ghosh, V. B. Reddy, J. Swinscoe, P. Lebowitz and S. M. Weissnian, JMB 126, 813 (1978). 31. R. M. Myers, D. C. Rio, A. K. Robbins and R. Tjian, Cell 25,373 (1981). 32. M. Fromm and P. Berg,J. Mol. Appl. Genet. 1,467 (1982). 33. B. J. Byrne, M. S. Davis, J. Yamaguchi, D. J. Bergsma and K. N. Subramanian, PNAS 80,721 (1983). 34. P. Gruss, R. Dhar and G. Khoury, PNAS 78,943 (1981). 35. C. Benoit and P. Chambon, Nature 290, 304 (1981). 36. D. Mathis and P. Chambon, Nature 290,310 (1981). 37. R. D. Everette, D. Baty and P. Chambon NARes 11,2447 (1983). 38. D. Baty, H. A. Barrera-Saldana, R. D. Everett, M. Vigneron, and P. Chambon NARes 12,915 (1984). 39. A. R. Buchman, M. Fromm and P. Berg, Mol. Cell. Bid. 4, 1900 (1984). 40. R. Contreras, D. Gheysen, J. Knowland, A. Van d e Voorde, and W. Fiers, Nature 300,500 (1982). 41. J. Brady, M. Radonovich, M. Vodkin, V. Natarajan, M. Thoren, G. C. Das, J. Janik and N. P. Salzman, Cell 31, 625 (1982). 42. J. Brady, M. Radonovich, M. Thoren, G. C. Das and N. P. Salzman, Mol. Cell Biol. 4, 133 (1984). 43. U. Hansen and P. Sharp, EMBOJ. 2,2293 (1983). 44. S . W. Hartzell, J. Yamaguchi and K. N. Subramanian, PNAS 81,23 (1983). 45. D. C. Rio and R. Tjian, J . Mol. App. Genet. 2,423 (1984). 46. C. M. Gorman, L. F. Moffat and B. H. Howard, Mol. Cell Biol. 2, 1044 (1982). 47. D. Rio, A. Robbins, R. Myers and R. Tjian, PNAS 77,5706 (1980). 48. U. Hansen, D. G. Tenen, D. M. Livingston and P. Sharp, Cell 27,603 (1981). 49. D. C. Rio and R. Tjian Cell 22, 1227 (1983). 50. P. Moreau, R. Hen, B. Wasylyk, R. Everett, M. P. Gaub and P. Chambon, NARes 9, 6047 (1981). 51. M. Fromm and P. Berg, Mol. Cell Biol. 3,991 (1983). 52. J. d e Villiers, L. Olson, J. Baneji and W. Schaffner, CSHSQB 47,911 (1983).

234

GOKUL C. DAS ET AL.

53. J. Banerji, S. Rusconi and W. Schaffner, Cell 27,299 (1981). 54. R. K. Humphries, T. Ley, P. Turner, A. D. Moulton and A. W. Nienhuis, Cell 30, 173 (1982). 55. R. Treisnian, N. J. Proudfoot, M. Shander and T. Maniatis, Cell 29, 903 (1982). 56. P. Sassone-Corsi, J. P. Dougherty, B. Wasylyk and P. Chamhon, PNAS 81, 308 (1984). 57. M. Vigneron, H. A. Barrera-Saldana, D. Baty, R. E. Everett and P. Chamhon, EMBO 1.3,2373 (1984). 58. M. Tsuda and Y. Suzuki, PNAS 80,7442 (1983). 59. H. Weiher, M. Konig and P. Gruss, Science 219, 626 (1983). 60. J. de Villiers and W. Schaffner, NARes 9, 6251 (1981). 61. J. de Villiers, L. Olson, C. Tyndal and W. Schaffner, NARes 10,7965 (1982). 62. C. Tyndall, G . La Mantia, C. M. Thaker, J. Favoloro and R. Kamen, NARes 9,6231 (1981). 63. S. Kenney, V. Natarajan, D. Strike, G. Khoury and N. P. Salzman, Science 226, 1337 (1984). 64. M. Lusky, L. Berg, H. Weiher and M. Botchan, Mol. Cell B i d . 3, 1108 (1983). 65. P. Hearing and T. Shenk, Cell 33,695 (1983). 66. R. Hen, E. Borelli, P. Sassone-Corsi and P. Chamhon, NARes 11,8747 (1983). 67. N. Rosenthal, M. Kress, P. Gruss and G. Khoury, Science 222,749 (1983). 68. B. Levinson, G. Khoury, G. Van de Woude and P. Gruss, Nature 295,568 (1982). 69. L. A. Laimins, G. Khoury, C. Gorman, B. Howard and P. Gruss, PNAS 79, 6453 (1982). 70. “Enhancers and Eukaryotic Gene Expression” (Y. Gluznian and T. Shenk, eds.), Cold Spring Harbor Lab., Cold Spring Harbor, NY, 1983. 71. S. D. Gillies, S. L. Morrison, V. T. Oi and S. Tonegawa, Cell 33,717 (1983). 72. J. Banerji, L. Olson and W. Schaffner, Cell 33,729 (1983). 73. C. Queen and D. Baltimore, Cell 33, 741 (1983). 74. B. Wasylyk, C. Wasylyk, P. Augereau and P. Chamhon, Cell 32,503 (1983). 7Fi. J. Jongstra, T. Reudelhuher, P. Oudet, C. Benoit, C. B. Chae, J. M. Jeltsch, D. J. Mathis and P. Chamhon, Nature 307, 708 (1984). 76. W. A. Scott and D. J. Wigmore, Cell 15, 1511 (1978). 77. A. J. Varshavsky, 0. H. Sundin and M. J. Bohn, Cell 16,453 (1979). 78. A. Nordheim and A. Rich, Nature 303,674 (1983). 79. M. Katinka, M. Yaniv, M. Vasseur and D. Blangy, Cell 20, 393 (1980). 80. M. Katinka, M. Vasseur and D. Blangy, Nature 290,720 (1981). 81. H. R. Scholer and P. Gruss, Cell 36, 403 (1984). 81u. M. Fromm and P. Berg, 1. M n f . App. Genet. 2, 127 (1983). 82. W. S. Dynan and R. Tjian, Cell 32,669 (1983). 83. W. S. Dynan and R. Tjian, Cell 35, 79 (1983). 84. M. Bensinihon, J. Gaharro-Arpa, R. Ehrlich and C. Reiss, NARes 11,4521 (1983). 85. U. Siebenlist, R . B. Simpson and W. Gilbert, Cell 20,269 (1980). 850. S. Cereghini, P. Herhonnel, J. Jouanneau, S. Saragasti, M. Katinka, B. Bourachot, B. d e Crombrugghe and M. Yaniv. CSHQB 47,935 (1983). 86. H. Mishoe, J. Brady, M. Radonovich and N. P. Salzman, Mol. Cell B i d . 4 , 2911 (1984). 87. N. G. Miyamoto, V. Moncollin, M. Wintzerith, R. Hen, J. M. Egly and P. Chamhon, NARes 12,8779 (1984). 88. R. Breathnach and P. Chambon, ARB 50, 349 (1981).

sv40 PROMOTERS

235

89. P. K. Ghosh, P. Lebowitz, R. J. Frisque and Y. Gluzman, PNAS 78, 100 (1981). 90. B. Wasylyk, C. Wasylyk, H. Matthes, M. Wintzerith and P. Chambon, EMBOJ. 2, 1605 (1983). 91. B. Wasylyk and P. Chambon, NARes 9, 1813 (1981). 92. B. Wasylyk, R. Derbyshire, A. Guy, D. Molko, A. Roget, R. Teoule and P. Chambon, PNAS 77,7024 (1980). 93. M. Concino, R. A. Goldman, M. H. Caruthers and R. Weinmann, JBC 258, 8493, (1983). 94. S. Hirose, K. Takeuchi, H. Hori, T. Hirose, S. Inayama and Y. Suzuki, €“AS 81, 1394 (1984). 95. B. L. Davison, J.-M. Egly, E. R. Mulvihill and P. Chambon, Nature 301, 680 (1983). 96. J. Corden, B. Wasylyk, A. Bu~hwalder. P. Sassone-Corsi, C. Kedinger and P. Chambon, Science 209, 1406 (1980). 97. S. P. Gregory, N. 0. Dillon and P. H. W. Buttenvorth, NARes 10,7581 (1982). 98. S. L. Hu and J. L. Manley, PNAS 78,820 (1981). 99. S. Y. Tsai, M. J. Tsai and B. W. O’Malley, PNAS 78, 879 (1981). 100. G. C. Grosveld, C. K. Shewmaker, P. Jat and R. A. Flavell, Cell 25,215, (1981). 101. R. Grosschedl and M. L. Birnstiel, PNAS 77, 7102 (1980). 102. P. Mellon, V. Parker, Y. Gluzman and T. Maniatis, Cell 27, 279 (1981). 103. S. L. McKnight and R. Kingsbury, Science 217, 316 (1982). 104. S. N. Sinha, R. J. Hellwig, D. P. Allison and S. K. Niyogi NARes 10, 5533 (1982). 105. J. W. Innis and W. Scott, Mol. Cell B i d . 3, 2203 (1983). 106. J. E. Stefan0 and J. D. Gralla, PNAS 79, 1069 (1982). 107. G. C. Das and N. P. Salzman,JMR 182, 229 (1U85). 108. P. K. Ghosh and P. Lebowitz,]. Virol. 40, 224 (1981). 109. T. Itoh and J. Tomizawa, PNAS 77,2450 (1980). 110. H. Masukata and J. Tomizawa, Cell 36,513 (1984). 11 1 . A. Nandi, G. C. Das and N. P. Salznran, Mol. Cell B i d . 5, 591 (1985). 112. R. Tjian and A. Robbins, PNAS 76,610 (1979). 113. J. Reiser, J. Renart, L. V. Crawford and G. R. Stark, J. Virol. 33, 78 (1980). 114. M. Oren, E. Winocour and C. Prives, PNAS 77,220 (1980). 115. R. McKay and D. DiMaio, Nature 289, 810 (1981). 116. P. Tegtmeyer, B. Anderson, S. B. Shaw and V. G. Wilson, Virol. 115, 75 (1981). 117. D. G. Tenen, L. L. Haines and D. M. Livingston, J M B 157, 473 (1982). 118. D. Jessel, T. Landau, J. Hudson, T. Lalor, D. Tenen and D. M. Livingston, Cell 8, 535 (1978). 1 1 8 ~ R. . D. G. McKay,JMB 145, 471 (1981). 119. D. Shalloway, T. Kleinberger and D. M. Livingston, Cell, 20, 411 (1980). 120. R. M. Myers, D. C. Rio, A. K. Robbins and R. Tjian, Cell 25, 373 (1981). 121. J. A. Hassell, E. Lukanidin, G. F e y and J. Sambrook,JMB 120,209 (1978). 122. D. G. Tenen, T. S. Taylor, L. L. Haines, M. K. Bradley, R. G. Martin and D. M. Livingston, JMB 168, 791 (1983). 123. K. A. Jones and R. Tjian, Cell 36, 155 (1984). 124. A. L. DeLucia, B. A. Lewton, R. Tjian and P. Tegtmeyer, J. ViroZ. 46, 143 (1983). 125. P. Tegtmeyer, B. A. Lewton, A. L. DeLucia, V. G. Wilson and K. Ryder,]. Virol. 46, 151 (1983). 126. R. M. Myers, R. C. Williams and R. Tjian,JMB 148, 347 (1981). 127. R. M. Myers, M. Kligman and R . Tjian,JBC 256, 10156 (1981).

236 128. 129. 130. 131. 132. 133. 134.

GOKUL C. DAS ET AL.

M. Oren, E. Winocour and C. Prives, PNAS 77,220 (1980).

J. C. Alwine, S. I. Reed, J. Ferguson and G. R. Stark, Cell 6,529 (1975). P. Tegtmeyer, M. Schwartz, J. K. Collins and K. Rundell,]. Virol. 16, 168 (1975). J. C. Alwine, S. I. Reed and G. R. Stark,]. Virol. 24,22 (1977). D. Maio and D. Nathans, ]ME 156,531 (1982). B. J. Meyer, R. Mauer and M. Ptashne, ] M E 139, 163 (1980). B. A. Parker and G . R. Stark,J. Virol. 31, 360 (1979).

The Role of the Anticodon in Recognition of tRNA by Aminoacyl-tRNA Synthetases

I

LEVL. ISSELEV Institute of Molecular Biology USSR Academy of Sciences Moscow, USSR

I. Concise Background of the Problem. . . . . . . . . . . . . . . . . . . . . . . . . . . . 11. The Role of the Anticodon in Acceptor Function . . . . . . . . . . . . . . . . . A. Prokaryotes ( E . coli). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . B. Eukaryotes (Yeast). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111. General Remarks.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

239 243 243 252 258 263

The problem of interactions between nucleic acids and proteins is illustrated by the mutual recognition of aminoacyl-tRNA synthetases and tRNAs. The first recognition hypothesis dates back to 1964 and postulates that anticodons of tRNAs are involved in this process; this proposal has been corroborated by numerous observations made with various methodological approaches and with tRNAs of different amino-acid specificity. These experimental results are relevant to most tRNAs included in the main structural class with a short varying loop. The recognition of tRNAs with long varying loops remains obscure. The binding of a specific amino acid to a cognate tRNA is catalyzed by its respective aminoacyl-tRNA synthetase (EC 6.1.1).* The enzyme-catalyzed aminoacylation of tRNA is extremely precise owing to the selective activation of amino acids, the selective interaction of tRNA with a cognate synthetgse, and finally, “correction” of the synthesized aminoacyl-tRNA (see reviews (1-3a). The interaction of a tRNA and an aminoacyl-tRNA synthetase is an example of highly specific protein-nucleic acid recognition, which makes it possible to reproduce precisely genetic information in the primary structure of proteins. Actually, the reproduction of a hereditarily predetermined protein structure in cells of one and the same organism as well as in a number of generations depends on three key * The Recommended Name (Enzyme Nomenclature, 1984) is now amino-acidtRNA ligase. Aminoacyl-tRNA synthetase is now another name. (Eds.) 237 Progress in Nucleic Acid Research and Molecular Biology, Vol. 32

Copyright 8 1985 by Academic Press, Inc. All rights of reproduction in any form resewed.

238

LEV L. KISSELEV

processes: (1) DNA-directed mRNA synthesis; (2) codon-anticodon interaction in the ribosome; and (3) enzyme-catalyzed tRNA aminoacylation. These processes are equivalent in terms of the accuracy of protein synthesis although they are realized, in the first and second cases, via homotropic interactions (between nucleic acids) but, in the third case, via heterotropic interactions (between proteins and nucleic acids). In their classical experiments, Chapeville et al. ( 4 )showed that the specificity of translation (reading of mRNA in ribsomes) is determined by the nature of the tRNA rather than by the amino-acid residue attached to it. It has thus become clear that one cannot gain an insight into the mechanism responsible for the high specificity of tRNA aminoacylation without understanding how synthetases and tRNAs interact at the molecular level. Briefly, this problem is referred to as a “recognition” problem. It has been considered in many reviews (1-3 a, 5-9), has presented a challenge to many authors for a number of years, and still is attracting the attention of many laboratories. One may consider three aspects of the recognition problem: (1)the nature of tRNA and synthetase sites in contact with one another; (2) conformational changes induced in the enzyme and its substrate in the course of their interaction; (3) detection of those structural elements by which a particular synthetase discriminates between tRNAs-in other words, chooses its substrate tRNA among 20 or more different types. In the early 1970s, the first and third aspects were not clearly distinguished, while the second did not exist at all since conformational changes of synthetases and tRNAs were definitely established only within recent years (for ref. see 9a-c). Many authors studying tRNA-synthetase pairs of different specificity and from various subjects have arrived at the conclusion that the relationships found for a particular pair cannot be extended.by analogy to another pair. For this reason, we are bound to substantiate our further concepts by referring to enzyme-substrate pairs of different amino-acid specificity. The recognition problem is specific in that the information has been gained by different methods, and only their comparison may provide a certain idea about actual events in the interaction between a synthetase and its tRNA. Finally, prior to presenting the factual material, it should be noted that the recognition problem is not solved yet, although we have gained a better insight into this very complicated area. Much more is known about the primary structure of tRNAs than about aminoacyltRNA synthetases: by the beginning of 1984, over 300 tRNA primary

ROLE OF THE ANTICODON

239

structures had been established ( 7 4 9d) compared to only several synthetase structures. Therefore, we deal below only with the structure of tRNA regions while the protein partner is not discussed. It has been realized for a long time that tRNA structures must include elements by which the molecules differ from one another and that enable an aminoacyl-tRNA synthetase to discriminate between the “cognate” tRNA and a “noncognate” one. Such elements were looked for soon after tRNAs and synthetases were discovered; the search passed through a number of stages and entailed disappointments, errors, hopes and success. Prior to presenting the contemporary state of the problem, it would be relevant to briefly review how these studies progressed from the very beginning.

1. Concise Background of the Problem The first experiments aimed at elucidating the chemical nature of tRNA groups involved in the acceptor function were independently conducted in three laboratories headed by P. Doty, P. Zamecnik, and W. Engelhardt. Their results revealed two important aspects: (1) chemical modification of bases has different consequences for .the acceptor function of a tRNA, depending on its amino-acid specificity (10, 11); (2) nucleotides not involved in intramolecular hydrogen bonding are essential for manifestation of the acceptor function (12, 13).Since the reagents used in these experiments were not selective toward the bases being modified by them, one could not arrive at a conclusion about the nature of the tRNA regions essential for the function. The next step was made a year later (14)when hydroxylamine selectively modifying uracil bases at pH 2 9 was used to modify tRNA (see 15). It was found that tRNA molecules specific for certain amino acids rapidly lost their acceptor activities in the course of modification, whereas tRNAs specific for other amino acids were resistant or inactivated very slowly. In other words, a correlation was established between a modification of certain bases (in this case, uracil) and a change of the acceptor function. For example, yeast tRNASerand tRNAPhe are inactivated by hydroxylamine very slowly whereas tRNALyS,whose anticodon contains 2 or 3 uracils, was rapidly inactivated by “single-hit” kinetics (16). The secondary structure of the tRNA molecules remained intact, indicating that only exposed bases were modified (17). Those tRNAVa’molecules that did not lose their acceptor activity after modification retained their adaptor function as well (16).When this tRNA was modified by O-methylhydroxylamine

240

LEV L. KISSELEV

at pH 5.0 a specific reaction for cytosine bases (see I S ) , some tRNA molecules rapidly lost their acceptor activity, while certain other tRNAs lacking cytosine in their anticodon were inactivated at a much lower rate. The above data, namely (1)the correlation between the susceptibility of the acceptor function to a specific chemical modification and the presence of bases being modified in the anticodon of the tRNAs under study, (2) modification of the exposed bases and the “one-hit” character of the inactivation curve, and (3)the correlation between the preservation of the acceptor and adaptor functions in partly modified molecules, are consistent with the hypothesis that, at least for these tRNAs where the correlation was established experimentally, the anticodon or part of it is required for the acceptor activity. Indeed, it has been concluded that the anticodon is involved in the specific interaction with aminoacyl-tRNA synthetases (16). These findings were extended and supplemented by observing E. coli and liver tRNAs specific for a greater number of amino acids (18, 19). The conclusions made earlier with yeast tRNAs were confirmed and allowed one to state that “the coding triplet (anticodon) plays an important, possibly a key role in the interaction with aminoacyl-tRNA synthetase (heteromolecular recognition between the protein and nucleic acid)” (19). Since all of the tRNAs studied lost their acceptor activity eventually upon modification of their cytosine bases, it has been presumed that the CCA end of the molecule may also play an essential role in the interaction with the enzyme (16-19). Consequently, from the very beginning, the anticodon was considered as a specific (“recognizing”) site rather than solely as the region of contact between tRNA and the enzyme. In parallel with the described work, several laboratories tackled this problem using other tools. The conclusions from these efforts were controversial even within one and the same laboratory. For instance, P. Zamecnik et aZ. first ruled out the possibility that the anticodon might be involved in the acceptor function (20),but later put forward the idea that regions essential for the acceptor and adaptor activities of tRNA could overlap (21).J.-P. Ebel et al. denied the participation of the anticodon in the acceptor function (22),but later reconsidered their opinion (23).J. Carbon showed the same evolution of views, from negation (24) to admission (25). According to other authors (26,27),their findings are not inconsistent with the assumption that the anticodon is involved in the acceptor function of tRNA, although they did not demonstrate such a role. Most of the above


24 1

investigations made use of nonselective chemical modification as a tool for studying the acceptor function of tRNA. Therefore, both the negation and admission of the role played by the anticodon were based on indirect evidence and could not be regarded as unambiguous proof. A new approach was proposed at nearly the same time (19,28); it stemmed from the observation that oligonucleotides from the enzymatic digest of tRNA are competitive inhibitors in aminoacylation reaction (29). It was shown that those polynucleotides and oligonucleotides having the same structure as the anticodon are potent competitive inhibitors of tRNA aminoacylation in the cases studied (19, 28);this is entirely consistent with the idea that the anticodon participates in the acceptor function. As one might expect from these observations, polyuridylate (not polyadenylate) forms a complex with E. coli lysyl-tRNA synthetase that dissociates under the action of tRNALys (30).In this work, the authors did not rule out the possibility that the anticodon of tRNA may be involved in its acceptor function. As was shown by others (31,32),the approach with inhibitors should be used with caution, taking account of possible artifacts; however, these experiments did not refute the conclusion that the anticodon of certain tRNAs might be involved in the acceptor function. In 1965-1967, the primary structures of some yeast tRNAs were deciphered for the first time in the laboratories of R. Holley, H. Zachau and A. Bayev: this created the structural basis for elucidating the role of individual tRNA regions in the interaction with synthetases. In the late 1960s and the early 1970s, new ideas about the location of the tRNA recognition sites appeared, in particular, the acceptor stem (34,35) and the D stem (36, 37) hypotheses. Genetic data about the role of the acceptor stem in the process of recognition were obtained for E. coli tRNATyr.Five suppressor mutants of this tRNA with the substitutions A82 -+ G82, C81 ---* A81, C81 + U81, G2 ---* A2 and G1 + A 1 acquired the ability to accept glutamine (for references, see 37a,b). However, each of those mutant tRNAs also carries alteration in the anticodon. Therefore, from these studies it remains possible that alterations in both anticodon and stem are necessary for the change in acylation specificity. Recently it was proved that a single substitution U) of E. coli tRNALysis sufficient for change in in position 70 (C amino-acid specificity (37c).To my knowledge, this is a unique welldocumented case where a single alteration of the amino-acid acceptor stem causes misacylation of the tRNA, and constitutes a strong argument in favor of participation of this region in enzyme recognition.

242

LEV L. KISSELEV

Arguments pertinent to certain other tRNAs are less convincing (see 1 , 7u). Two main arguments are presented in favor of the D stem hypothesis (see 6): the involvement of the D stem in photochemical crosslinking with the synthetase, and the data on heterologous aminoacylation. The first argument refers to contact of the synthetase with tRNA in the D stem region rather than to recognition, which is quite possible. In the second case, the strategy seems to be doubtful since the authors made use of the structural similarity in the D stem region between tRNAs differing in their amino-acid specificity during their aminoacylation by one and the same synthetase to prove that the D stem was involved in the recognition. In my opinion, this is only indicative of a possible contact between this group of tRNAs and the synthetase through the D stem in heterologous interaction, while the question of recognition in a homologous system remains open. The stem hypotheses were further developed in a “binary code,” the main feature of which is the formation of an H-bond between the protein “leg” and any neighboring base except guanine, at which, on the contrary, repulsion occurs ( 3 7 4 . Although some data fit the predictions of this “binary code”, others hardly support it. For example, the serine and leucine isoacceptor tRNA families have no common nucleotides at all in both amino-acid and anticodon stems in each family, although some common bases do exist in the D stems of these tRNAs (37e). On the other hand, the glycine family has no common nucleotides in their D and anticodon stems, whereas the valine family has no common stem nucleotides at all except C31-G39 (37e). It is very tempting to verify the predictions of this model with more than 300 tRNA sequences now available ( 9 d ) .The recognition via two sites, the anticodon and acceptor stem (38),implies that the two recognition sites may coexist; this idea has a general significance and may be fruitful for a number of tRNA-synthetase pairs. It has also been assumed that the fourth (impaired) nucleotide from the 3’ end of tRNA may play the role of a “discriminator” (39). The data now available point against this proposal. For instance, yeast isoacceptor tRNAs specific for Lys and Arg have different bases at the 4th position from the 3‘ end (see 9d). In the glycine isoacceptor family, there is no common base in this position (37e). Moreover, the A73 in t R N A P (E. coZi) can be replaced by U, C or G without any alterations in methionine acceptance with either purified or crude enzyme preparations (39a). No other amino acids esterified this “mutant” tRN A *. The following structural elements can be recognition sites in prin-

243

ROLE O F THE ANTICODON

ciple: the V loop, in particular, in tRNAs belonging to the second structural class and having a long V loop; the anticodon; the D and acceptor stems. These elements are variable enough in sequence and, according to certain observations, may be in contact with the synthetases (see 3, 7a, 8). However, the observations that have been made are too few to consider them here systematically. The T'PC-loop and stem cannot serve as a recognition site owing to the low variability of its structure. Hence, only the anticodon hypothesis has sufficient experimental substantiation and is therefore dealt with below.

II. The Role of the Anticodon in Acceptor Function The role of the anticodon was studied with both eukaryotic (mainly yeast) and prokaryotic (mainly E . coli) tRNAs. Some tRNAs have been the object of intensive and comprehensive studies while others have been examined either casually or not at all. Since the work was conducted for many years and often along parallel lines, the results are not considered in chronological order. The involvement of the anticodon in acceptor function was proved for different tRNAs isolated from various sources. For this reason, the experimental material obtaining for each enzyme-substract pair is presented separately since it has not been proved that it would be correct to extrapolate results from one pair to another.

A. Prokaryotes (E. coh) Glycine tRNAs. Carbon et al. (25, 40-44) described several mutations of isoacceptors of tRNA"'Y and established their primary structures. Certain common structural elements can be detected in comparing the nucleotide sequences of tRNAG'y from prokaryotic and eukaryotic organisms (Fig. 1).One can see that, taking no account of nucleotides in common for all or most the tRNAs, only four sites could be specific for tRNAG*y,viz. the second and third nucleotides of the anticodon, U59, C31*G39, and the acceptor stem region, C20G71G3C70. The middle two sites are found in some other tRNAs and therefore are likely not to be involved in recognition, whereas the anticodon and part of the amino-acid acceptor stem remain good candidates for this role. Indeed, it has been proved that the anticodon participates in the acceptor function. A substitution of U for C involves the third letter of

244

LEV L. KISSELEV

B

A

.. .. .. .. . .. . ......... .. . . ..:. .... .... . .. .. .. 2CG71 3G C70

59

" 0

0 .

0

.

.*-

.

31C

G39

*.

C36 35

* .

.. .. .. .. .. .. ... . :. ... ....... .... : ... . ... .. .. .. A 73

**

*.

.AS4

..

31C

G39

' * ~ C 3 6 35

FIG.1. Generalized structures of tRNAG'y (A) and tRNAVd(B) recognizable by the cognate aminoacyl-tRNA synthetases. Positions are numbered according to the Cold Spring Harbor system. Family-specific homologies are symbolized by the nucleosides in the corresponding positions. Constant, semiconstant and variable nucleosides are shown by dots.

the anticodon in tRNAg'Y and the second letter in tRNAt'Y. In both cases, this transition is accompanied by a posttranscriptional modification of A37, adjacent to the 3'-end of the anticodon. The rate of tRNA aminoacylation falls abruptly in both mutations (for mutant tRNA,C'y, by a factor of lo4 compared with the wild type) owing to a reduced affmity of the enzyme for mutant tRNA. As was noted above, these experiments are consistent with the earlier data on E. co2i tRNAGIY inactivation by hydroxylamine. The correlation between C -+ U conversions of cytosine residues in the anticodon and in the 3'-terminus, on the one hand, and a loss of the acceptor activity, on the other, was studied in detail with tRNA$'Y containing the GCC anticodon (45).The CCA + UUA conversion at the 3' end produced a minor effect on the acceptor activity whereas the transition of C35 and C36 caused an 80%inactivation. The conversion of either C35 or C36 partly inactivated the molecule. These findings fit in well with the earlier observations (46) obtained on a supT tRNAG'y mutant; this mutant retains its acceptor activity in the C35 + U35 conversion, although it is lower than in the wild type, but greater than in the C36 + U36 mutation. Eukaryotic tRNAsCIYwere not studied comprehensively. However, yeast tRNAG1yis rapidly inactivated in acylation by the rat liver enzyme (17)if the exposed cytosines are modified with hydroxylamine.


245

The primary structures of eukaryotic tRNAGIY(see 9d) have only two regions in common containing exposed cytosines: the main part of the anticodon (CC), and the CCA end. Rapid inactivation of these tRNAs after U for C substitution indicates that the anticodon of eukaryotic tRNAscly, just as the anticodon of prokaryotic tRNAsG'Y, may be involved in the interaction with the synthetase. Methionine tRNAs. These tRNAs were studied in detail (47, 48) using a set of chemical reagents, and the results have been summarized (49). The anticodon of these tRNAs is C34-A35-U36 and their acceptor activity is lost when C34 and/or A35 are modified. However, not all the authors agree that the anticodon of E . coli tRNAMetparticipates in the acceptor function. For instance, the rate of aminoacylation did not noticeably decrease in an attempt to inhibit tRNAMet aminoacylation by oligonucleotides which were either an anticodon hairpin or a trinucleotide complementary to the anticodon (50).The negative result of the experiments with competitive inhibition may be attributed to displacement of the oligonucleotides from the complex with tRNA, due either to a much greater binding constant of the enzyme-substrate complex, or to a conformational change of tRNA under the action of the synthetase resulting in a decreased (if any) capability to bind the complementary trinucleotide, or different conformations of the anticodon arm in the isolated state and in the intact tRNA molecule. Synthetic oligodeoxyribonucleotidesof specified composition were used to study their effect on the acceptor activity of E . coli tRNAy". Oligomers complementary to the anticodon loop were found to inhibit the reaction of aminoacylation; the effect of a heptanucleotide was stronger than that of a pentamer, and greater at 0°C than at 10°C (51).The authors concluded that the intact anticodon loop of tRNAY"' is necessary for the interaction with the synthetase. Moreover, they showed that a trinucleotide complementary to the anticodon does not inhibit the reaction, which accounts for the negative result of the preceding work (50). It has recently been shown that T4 RNA ligase can be efficiently used for the synthesis of various polyribonucleotides (51a,b).In particular this approach was successfully applied for the preparation of tRNA molecules substituted in the anticodon bases (51c,d; 52a,b,c). This procedure involves excision of the normal anticodon by limited digestion of intact tRNA with endonuclease, RNA ligase is then used to join the desired oligonucleotide to the 5' half of the tRNA molecule followed by linkage of this modified fragment with the 3' half to regenerate the anticodon loop (Fig. 2). Synthesis of intact tRNAy" ( E . coli) containing the initial antico-

246

LEV L. KISSELEV

P C

3'

scg PG C

I

. . )

n,

C

5

w

6,

7

FIG.2. Preparation of the tRNAPhe(yeast) modified in the anticodon. From Bruce and Uhlenbeck ( S c ) , with permission. Steps involved in the construction of a new anticodon are shown. An internal 32P label can be introduced at position "a" with internally labeled (Ap),C and at position "b" with [y-32P]ATP. don CAU by this procedure yields a product indistinguishable from native tRNAfM"' in its aminoacylation ability with homologous methionyl-tRNA synthetase (52a). However, substitution of any other nucleotide at the wobble position (C34) of the anticodon dramatically impairs the acceptor function. Measurement of methionine acceptance in an excess of pure enzyme showed that the rate of aminoacylation of the AAU, GAU, and UAU anticodon derivatives of tRNA?"' is four to five orders of magnitude slower than that of the tRNA containing C34 (Table I). Furthermore, these derivatives fail to inhibit the aminoacylation of normal tRNApet,indicating that they bind poorly to the enzyme. The authors (52a) conclude that "these results support a model involving direct interaction between methionyl-tRNA synthetase and the C in the wobble position during aminoacylation of tRNAp"'."

247

ROLE O F THE ANTICODON

TABLE I AMINOACYLATION OF tRNA,MetDERIVATIVES CONTAINING BASESUBSTITUTIONS IN THE ANTI COD ON^ Synthesized anticodon sequence 5' -3 3'

Acceptor activity (moles of methionine accepted per mole of tRNA in 30 minutes at 26°C)'

0.4 nM

4 nM

40 nM

1.0 0.32 0.13 0.01 CO.01 6. At acidic pH, hydrolysis of the acetal linkage occurs. The products of modification can be assayed spectrophotometrically after splitting the oligonucleotide moiety, as the dialkylaminobenzaldehyde groups linked with the alkylated residues absorb at 350 nm ( E = 28.8 x 103). 2',3'-0-4-"(2-chloroethyl)-N-methyl]aminobenzylidene derivatives of oligonucleotides (Np),NCHRCl are prepared as a mixture of two isomers (at the C atom of the acetal linkage) present in roughly equimolar amounts. High yields of complementary-addressed modification with these reagents and the absence of detectable self-alkylation suggest that both isomers have favorable conformational states with respect to orientation of the 2-chloroethylamino groups to the bases in the target sequence. Another example of heterobifunctional reagents is 4-[N-(2-chloroethyl)-N-methyl]aminobenzylamine(ClRCHzNH2). The strongly basic amino group of this reagent can be used for coupling to the phosphomonoester groups of oligonucleotides (35,36). This coupling reaction can be performed by activation of the phosphate group under conditions preventing reactions of the 2-chloroethylamino group. I n the methods described below, the phosphate activation is performed either directly in reaction mixtures containing ClRCHzNH2, or the

MODIFICATION OF NUCLEIC ACIDS

297

intermediate oligonucleotide derivative with activated phosphate is isolated and used further in the reaction with the amino component. The efficient method of synthesis of 5’-phosphoroamidates of oligonucleotides, which can be used for preparation of reactive oligonucleotide derivatives, is based on the activation of phosphates of oligonucleotides with 2,Z‘-dipyridyl disulfide and triphenylphosphine. Using this method, a variety of reactive derivatives of oligodeoxyribonucleotides as well as those of oligoribonucleotides were prepared with quantitative yields (37,38).In this procedure, a mixture of amine and oligonucleotide in an organic solvent is treated with an excess of condensing reagent consisting of a mixture of 2,2‘-dipyridyl disulfide and triphenylphosphine at room temperature. The reaction leads to a selective and quantitative amidation of monosubstituted phosphate groups in oligonucleotides. When the reaction is carried out in organic solvents, one must use cetyltrimethylammonium salts of oligonucleotides. It can be performed also in dimethyiformamide containing 10-15% of water; in this case, sodium or triethylamine salts of oligonucleotides can be used. In the latter case, one must use a 10- to 100-fold excess of the condensing reagent. The presence of a strongly basic amine eliminates completely the reaction with internucleotide phosphates. Therefore, the advantage of the described procedure as compared with the traditional use of dicyclohexylcarbodiimide, or mixed anhydrides with diphenylphosphoric and mesitylenezarbonic acids, is that even oligoribonucleotides may be derivatized without splitting and isomerization of‘the 3 -5 internucleotide linkage typical of other procedures (39,40). The bifunctional reagents considered above have two functions of completely different structures. In some cases, compounds with two groups of similar structure differing in their reactivity may be used for the same goal. Thus in 41, the binuclear platinum complexes [H,OPt(dien)] -(CH,),,-[(dien)PtBrI+

were described. The more reactive [ HzOPt(dien)I2+fragment may be used for coupling with an oligonucleotide carrier by substituting guanine for HzO. The platinum oligonucleotide derivative bearing the low-reactive [(dien)PtBr]+ unit is a complementary-addressed reagent, as the substitution of Br- by some nucleic acid base may proceed only inside the complex under conditions of close contact between substituted and attacking ligands. The reactivity of this unit may be additionally regulated by changing the Hal- concentration in the reaction mixture.

298

DMITRI KNORRE AND VALENTIN

V. VLASSOV

The use of the reagents with one potentially reactive group that might be “switched on” after introduction into the addressing oligonucleotide or nucleic acid was proposed a few years ago. NJV‘,N’Tris(2-chloroethyl)-N-4-fo~ylphenyl)trimethylenediamine HOCCGH B N ( C H ~ C H ~ C I ) - ( C H ~ ) ~ N ( C H ~ contains C H & ~ ) ~two highly reactive aliphatic 2-chloroethylamino groups. One of them may alkylate some nucleophilic centers of the oligo- or polynucleotide carrier. Aromatic 2-chloroethylamino groups introduced into the carrier are of low reactivity due to the strong electron acceptor (formyl group) in the pposition of the benzene ring. However, it may easily be activated in the oligonucleotide derivatives by reduction of the HCO group to CHzOH with sodium borohydride (42). [This compound and its fragments are referred to as ClR&RCl and ClR;,dRCI where R h and R h represent the aromatic moiety with either an aldehyde or hydroxymethyl group, and R” the aliphatic moiety of the reagent.] This compound was successfully used to introduce alkylating groups scattered over the guanine residues of polynucleotides in gene-directed mutagenesis (9,31).An attempt is also described to use this compound for alkylation of the 5’ phosphate of oligonucleotides lacking the most reactive guanine residues pd(A-A-T-T-C-C-A-C) and pd(C-T-T-T-CC-A). A CIR:,dR” fragment was introduced at the 5‘ end of oligonucleotides. However only moderate yields (30-40%) of the desired complementary-addressed reagent were obtained (43). Some approaches to attach specific groups to the oligonucleotide carrier convenient for further derivatization with bifunctional reagents were proposed. Among the most promising was the introduction of a strongly basic amino group by attachment of ethylenediamine to a terminal phosphate via intermediate formation of the mixed anhydride of phosphate and mesitylenecarbonic acid should be mentioned. The amino group may be acylated by various carbonic acid derivatives bearing reactive moieties that should be incorporated in the complementary-addressed reagent (44).Enzymatic (45)and chemical (46) approaches to the preparation of oligonucleotides carrying thiophosphate terminal groups have been described. These compounds were successfully converted to reactive derivative by treatment with CIRAxR”CI. Another type of reagent with potentially reactive groups was proposed in (17, 4 7). The heterobifunctional reagent 6-bromo-5,5-dimethoxyhexanoylhydrazide may be introduced at the C-4 position of a cytosine residue. Low reactive CHzBr groups attached to oligo- or polynucleotide carrier may be activated by mild acid treatment con-

299


verting ketal to ketone thus producing rather reactive COCHzBr residues Cyt + NH2NHCO-(CH2)3-C(OCH3)t-CH2Br Cyt-NH-NH-CO-(CHz)3-C(OCH3)2-CH2Br

Cyt-NH-NH-CO-(CH&,-COCH2Br

n+

II. Complementary-Addressed Modification of Model Oligonucleotides and Polynucleotides

A. Specificity of the Complementary-Addressed Modification

The main goal of the complementary-addressed modification is to carry out a reaction within a definite region of the target polynucleotide without touching similar groups in other parts of the same molecule or other species of nucleic acids. At the same time, it is obvious that there is some possibility of direct attack of the same reagent at different points of the polynucleotide without preliminary complex formation. Thus the specificity of the reaction depends primarily on the ratio of the rates of addressed and nonaddressed reaction, Va/Vn. In the most simple case, a biomolecular reaction between reagent and biopolymer, this ratio may be written as

where K = the association constant of biopolymer P with reagent X x = the reagent concentration kl = the rate constant of the first-order reaction inside the con+ plex PX kz = the second order rate constant of the nonaddressed reaction na, 12, = the number of points that may react in the addressed or nonaddressed reaction. This simple relation may not be directly applicable to the single case that has been subjected to rather thorough quantitative investigation, namely, alkylation with 2-chloroeth ylamine derivatives. The nonaddressed reaction in this case proceeds via a rate-limiting conversion of X to the intermediate ethylenimmonium cation X+, which further reacts competitively with definite parts of the nucleic acid and

300

DMITRI KNORRE AND VALENTIN V. VLASSOV

with water and nucleophilic components of the buffer. In 48, it was proposed to characterize the reactivity of a definite jth residue of nucleic acid towards nonaddressed reagent by the value ~j (competition factor)

where ki is the second-order rate constant of reactions of Xf with nucleophiles Nui. The equation for K~ deduced in this paper permits one to express K~ value via initial concentrations of reagent ( x o ) and nucleic acid ( P O ) and the final concentration of modified j th residue zp K. 3

=

z(j)/(xo m

-

#))

. p0

As tcjpo represents the part of the X+ consumed in the reaction with the jth residue, the reaction rate of the nonaddressed modification of this residue may be written as

The expression for the reaction rate of an addressed reaction proceeding via the same intermediate cation according to scheme

P + X G P X (K) PX + PX+ (&I) (rate limiting step) PX++ PZ (modification product) may be written as

.v, = kopoKx/(l + K x ) Therefore

v,/v,

= WK(1

+ kx)

At low x values when VJV, is maximal

VJV,

=

K/K

The K value for the reagents consumed in a parallel reaction in solution may be estimated from the limit value of affinity modification Z , (as in 49):


301

Thus (50)yeast tRNAy' was alkylated with pdC-dG-rA(CHRC1) and it was found that the extent of the modification of 1-35 in the anticodon loop at 20" is 0.48 (SO).In this experiment, po = 11 p M ; xo = 200 FM. Therefore K = 9 X lo3 M-I. In 48, it was found that for nonaddressed alkylation of the same residue with Me-U(CHRC1) K = 21 M-'. Thus VJV, = 9 x 103121 = 430.

B. Complementary-Addressed Modification with Oligonucleotide Derivatives

T h e first investigations of complementary-addressed modification were carried out in the early 1970s when nucleic acids of known sequences as well as heterogeneous oligonucleotides were hardly available. Therefore, the first demonstrations of the possibility of effective intracomplex reactions between complementary-addressed reagents and target polynucleotides and elucidation of some essential features of these reactions were performed using reactive derivatives of oligoadenylates as complementary-addressed reagents and ribosomal RNA as a target. When rRNA was treated with 2',3'-0-4-"-(2chloroethy1)-N-methyl]aminobenzylidene derivatives of oligoadenylates, it was found that efficient alkylation occurs showing all the most typical features of the intracomplex reaction. The extent of modification of polynucleotides reaches the plateau level with increase of the ratio of the reagent and rRNA concentrations, and decreases with increase in temperature. The maximal extent of modification approaches 90% of the limit value of noncovalent complementary binding of the oligonucleotide reagent to rRNA. The maximal extent of alkylation increases with the length of the carrier (51). The kinetics of the reaction (fj2)is characterized by reaction rate constant ko of the same order of magnitude as the conversion of model 2-chloroethylamino derivatives U(CHRC1) and Me-pU(CHRC1) to ethyleneimmonium cation (53).Estimation of V J V , performed as described in the previous section for alklation of rRNA with (Ap)sA(CHRCl)at 5" using the data presented in (51 ) ( p o = 0.01 p M , xo = 1.42 pM, Z J p , = 0.25, K = 10 M-') leads to a value of 28 x lo3. Similar results were obtained for the alkylation of denatured E. c d i DNA with CI(RCHz)NH(pA)7at 20" (54).Starting from experimental values po = 27.6 x M, x0 = 0.74 X lop6M, Z,lpO = 0.1, K = 10 M-l, the VJV, value was found to be 16 x 103.

302


The preparation and the use of [14C]CH3 labeled methylating addressed reagents (p-CH30SO2C~H&H2CH2)(pdA)sand (p-CH3 O S O ~ C ~ ~ H & H ~ C H for Z ) alkylation ( ~ ~ T ) ~ of E. coli rRNA has been described (55). For 10 days, 70% of the former and 50% of the latter reagent (in a 30-fold molar excess over rRNA) was incorporated into RNA. In the nonaddressed version when p-CH30SOz-CfiH4-CH20H was used as the alkylating reagent under the same conditions, less than 1%of the reagent was consumed. Very rapid alkylation with (Np),-N(CHRC1) of several 23-S rRNA sites with a rate four orders of magnitude greater than that of the ethyleneimmonium cation formation has been described (56). The composition of alkylated bases strongly differs for addressed and nonaddressed modification. With Me-pU(CHRCI), guanines are predominantly alkylated in tRNA (57,58). With U(CHRCI), besides 7alkylguanine, 15-35% of alkylcytosine was found in the alkylation products of tRNA depending on the pH of the reaction medium (59). Similar results were obtained in the study of alkylation of beef spleen DNA (60). When the analysis was performed for rRNA modified with (Ap)&(CHRCI), alkylated guanine, adenine, and cytosine represented 32, 25, and 21% of the alkylation products (for comparison, rRNA contains 32% of Gua, 25% of Ade, and 21% of Cyt). Some indications of the alkylation of phosphodiester groups were obtained (61). Alkylation of beef spleen DNA with the same reagent results in modification of guanines and adenines (133-alkyl-Ade78 7-alkyl-Gua, and 4 7-alkyl-Ade per 1000 bases at 20") (62). Alkylation with (ClRCH2) NH(pA)G results in formation of 3-alkyl-Cyt, 7-alkyl-Ade, 3-alkyl-Ade7 and 7-alkyl-Ade in the ratio 39 : 19 : 16 : 7 (54).This change in specificity means that the reactive group of the bound complementary-addressed reagent cannot move freely along the target sequence but rather is limited in mobility and thus is forced to react with the nearest nucleophile irrespective of its chemical structure. The first study on localization of the points of alkylation in polynucleotide modified with complementary addressed reagent was done with yeast tRNAy' which was modified with the reagent pdC-dGrA(CHRC1). Analysis of the sites of alkylation in the tRNAy' structure (Fig. 1) suggests that the reaction occurs according to a complementary-addressed mechanism and that it is the third nucleotide from the 5' terminus of the target sequence that is alkylated (50).Thorough investigation of complementary-addressed modification with alkylating derivates of oligonucleotides were performed later using synthetic heterogeneous oligonucleotides as the targets. In these experiments,

303


0 0

0 0 0 0 0 0 0 0

0 0 . 0

oN::"H;H*c,

FIG.1. Sites of modification in the yeast tRNAp' (shown with arrows) by the alkylating oligonucleotide derivative pdC-dG-rA(CHRC1) (SO).(R = -C6H4N(CH~)CH&H2-).

the alkylating derivative of the synthetic oligonucleotide T-G-C-C-AA-A-C-rA(CHRC1)was used for modification of a series of complementary oligonucleotide targets possessing various nucleotides in the region to which the reactive groups were delivered (63, 64): 5' dp(X-X-C-C-T-G-T-T-T-G-G-C) 3'

3' CHRCI-rA-d(C-A-A-A-C-C-G-T) 5' I: X = G , 11: X = C, 111: X = A, IV: X = T

All four oligonucleotide models are modified efficiently with the formation of alkylated guanosine (at the N7 atom), cytidine (at the N3 atom), adenosine (at N1, N3, and N7 atoms), and some unidentified thymidylate derivatives, respectively. In target I, the position of the alkylation was determined as guanosine position 2. In all four cases, 85-95% of the complementary-addressed reagent bound to the target oligonucleotide was consumed in the alkylation reaction. This means that the alkylating group of the reagent is limited in mobility and the only nucleotide of the target available for modification by the group is the third nucleotide from the 5' terminus of the target sequence. It should be mentioned that reaction between the complementaryaddressed reagent and the target in this system at 0" proceeds with the rate of four orders of magnitude greater than that of the formation of

304

DMITRI K N O W AND VALENTIN V. VLASSOV

ethyleneimmonium cation at the same temperature (64).The reason of this acceleration remains unclear. An alkylating complementary-addressed reagent, the 3’-thiophosphate derivative of the oligonucleotide, was used for modification of synthetic 5‘-32P-labeled oligonucleotide (46). 5 pd(T-G-A-G-T-C-G-T-A-T-T-A) 3’ 3’ ClR~Jl”-Spd(A-G-C-A-T-A-A-T) 5

In this case efficient reaction occurs also. Up to 95% of the oligonucleotide reagent present in the complementary complex binds covalently to the target. The only site of alkylation was at G-4 of the target. Thus one may conclude that in this case the alkylating group is also restricted in mobility. Alkylation of DNA at definite points may be used for subsequent specific fragmentation. The possibility of obtaining a restricted number of fragments by this approach was demonstrated for E. coli DNA, the DNAs of T2, T4, T7, and A phages, and of Dictiostelium discoideum (7,65).However, this approach was not used for the isolation of homogeneous fragments and sequencing of the split region. When a complementary-addressed reagent is used for the modification of polynucleotides, some effects that cannot be observed in experiments on modification of short oligonucleotides should be taken into account. One of these effects is related to the specificity of modification. Complementary-addressed reagents can modify nucleic acids not only within the target sequences but also within sequences which are partially complementary to the carrier oligonucleotide. The relative extent of modification of these sequences is obviously controlled by the corresponding association constants. Another effect is related to the tertiary structure of polynucleotides; some remote sequences of the polynucleotide may be juxtaposed to the target sequence. The reactive group of the bound reagent can interact with nucleotides in these sequences resulting in the modification of some residues of the polymer that have nothing to do with the target sequence. The high accuracy of complementary-addressed modification of certain sequences in polynucleotides was demonstrated in experiments on modification of a 365-nucleotides-long DNA fragment (66). The sequence of the fragment was identical with the sequence in the 3‘-terminal region of encephalomyocarditis virus RNA in positions from 6686 to 7051 (67). As a complementary-addressed reagent the oligonucleotide derivative ClRCHZNHpd(C-C-C-T-C-T-T-T-C-T-T)


305

was used, This reagent is complementary to the sequence 6892-6902 of cDNA. In order to identify positions of alkylation in the DNA fragments, the modified fragments were split at the positions of alkylated purine nucleotides and the fragments produced were separated by gel-electrophoresis. The only nucleotides modified in the DNA fragment were G-6900, G-6901, and G-6902 residues located in the close vicinity of the 5' terminus of the oligonucleotide reagent. The total yield of modification at these residues was -70%. In the control experiments with alkylation of the complementary DNA fragment, no substantial modification was observed. The effect of interference of tertiary nucleic acid structure with the complementaiy-addressed modification was observed in similar experiments on modification of a 32P-labeled 203-nucleotide-long DNA fragment isolated from Hue111 digest of single-stranded DNA of bacteriophage M 13 (68).The alkylating derivative used as reagent was prepared from 5'-thiophosphorylated oligonucleotide

Modification occured at definite positions of the fragment; however, these positions are quite far from the sequence complementary to the carrier. The olignoucleotide carrier was complementary to the sequence 39-47 of the fragment while alkylation proceeded in the region 80-90 and around the nucleotide residue in position 150. In experiments with oligonucleotide derivatives, some properties of platinating groups, photoactivatable groups and DNA-cleaving groups were investigated. Oligonucleotides C a derivatized by cis-aquahydroxodiamminoplatinum were used for complementary-addressed modification of poly(1). This platinum compound is a heterobifunctional reagent possessing two positions that can be substituted, the aqua group and the hydroxy group. The treatment of oligo(C) with the platinum complex results in the substitution of the N-3 of cytosine residues for the aqua ligand. Derivatives with a random positioning of reactive groups retain their capacity to form complementary complexes with poly(I) when the extent of their modification is up to 18%. In the complementary complexes of the derivatives and poly( I), crosslinking occurs. The optimal extent of base derivatization in oligocytidylates was 14%. In this case, the yield of crosslinking was 40% (18). Complementaryaddressed reagents with platinum reactive groups were also quite efficient in modifying synthetic oligonucleotides. The derivative of

306


V. VLASSOV

pd(T-C-C-G-C-C-T-T-T) carrying a [dienPtBr] + group at a single guanine residue was used for modification of 32P-labeledcomplementary pd(G-G-C-G-G-A) 5' pd(T-C-C-G*-C-C-T-T-T) 3' 3' d(A-G-G-C -G-G)p 5'

where G* is derivatized guanosine containing [GuaPt(dien)12+-(CH&[( dien)PtBr]+

Under conditions providing oligonucleotide complex formation, crosslinking of the oligonucleotide reagent to the target oligonucleotide occurs with -85% yield. The reaction was complete within a few minutes. The product of reaction was subjected to partial purinespecific splitting by formic acid and diphenylamine treatment. As the platination stabilizes glycosyl bonds in modified nucleosides, there was no splitting at these positions. Therefore gel-electrophoretic analysis of the oligonucleotide partially split by depurination reveals positions of the platinated residues. The platination proceeded mainly at 5' terminal guanosine residue of the target (19). Poly(A) was split in the presence of nonathymidylate carrying a dansylaminoethyl residue attached to 5'-phosphate under laser irradiation (337 nm) via a double quantum absorption mechanism. Noncomplementary poly(C) and poly(U) remained unchanged under the same conditions (69). In the mixtures of target oligonucleotides pd(AG-C-T-T-C-T-C-G-A-G-G-A-G) and pd(T-G-A-G-T-C-G-T-A-T-T-A) with dansylaminopropyl(R) derivatives of the respective complementary carriers Rpd(G-A-G-A-A-G-C-T) and pd(T-A-A-T-A-C-G-A)pR, crosslinking of the reagent to the target proceeded in parallel with double quantum splitting under laser irradiation. Another DNA-splitting group, EDTA-Fe(II), was investigated in experiments where octathymidylates carrying these groups at their 5'termini were used for modification of poly(A). The oligothymidylate derivatives degrade poly(A) in the presence of oxygen and reducing agents, and the process is specific with respect to the polynucleotides. The efficiency of reaction of the EDTA-Fe(I1) group was rather low (-10%) under the conditions used (71).In the experiment described above, it was demonstrated that various reactive groups when attached to oligonucleotides acquire the property to modify efficiently those nucleic acids that possess complementary targets for the oligonucleotide carriers. Modification within these target-sequences proceeds with high yields. The effects observed are specific ones.

MODIFICATION O F NUCLEIC ACIDS

307

Modification of the non-complementary sequences in control experiments usually does not occur to a measurable extent. In all cases where the point of modification was localized, the reagents investigated modified nucleic acids with high accuracy with respect to the position of the nucleotides attacked by the reactive groups.

C. Complementary-Addressed Modification with Polynucleotide Derivatives The binding of polynucleotides to nucleic acids is usually achieved by various hybridization procedures. In the course of these procedures, reactions within the reactive polynucleotide derivatives can proceed as well as reactions with solvent constituents. Therefore, in the case of polynucleotide derivatives, one must use the lowreactive groups or groups with latent activity that need special activation. Alkylating derivatives of early phage DNA transcripts were used for complementary-addressed modification of the phage DNA (9). Transcripts possessing 4-5% nucleotides modified with ClR~,R”Cl can still interact with the complementary H-strand of the phage DNA. Binding of the derivatives to the L-strand of the phage DNA did not occur, suggesting that this is a specific process. Activation of the ClRA,R”- groups by the sodium borohydride treatment resulted in covalent crosslinking of polynucleotides. For modification of complementary sequences in intact T7 DNA, the derivatized transcripts were hybridized to T7 DNA using the technique of R-loop formation. Polyalkylating transcripts hybridize to the corresponding regions in the DNA and the sodium borohydride activation results in covalent binding of the transcripts to the DNA with yields close to quantitative. Single-stranded DNA fragments can also serve as carriers for the C1R‘R”-groups(72). These groups coupled to DNA survive the hybridization procedure at moderate temperatures. Therefore, this derivatization can be used for preparation of complementary-addressed reagents of DNA restriction fragments, which are readily available. The derivatives of T7 DNA restriction fragments hybridize with T7 DNA and modify it when activated. They do not interact with noncomplementary DNA from other sources, demonstrating that the process of the modification is specific in this case. Similar experiments have been performed with supercoiled plasmid pBR322 (73). Alkylating derivatives of T7 RNA were prepared also using the bifunctional reagent 6-bromo-5’,5-dimethoxyhexanoyl-

308

DMITRl KNORRE AND VALENTIN V. VLASSOV

hydrazide with a latent alkylating group. In these experiments, two

RNA derivative preparations were used, with 10 and 25%of cytidine residues modified. The reactive groups of the derivatives were activated before the annealing or after the complex formation. In both cases intensive crosslinking occurred when the RNA derivatives are annealed with T7 DNA. These derivatives did not modify foreign DNA in control experiments, thus showing the specificity of modification. When activation is performed before the annealing the crosslinking efficiences are 20 and 7%, respectively, for RNA preparations with 25 and 10%of cytidines derivatized. When alkylating groups are activated after the hybrid complex formation, the efficiences of crosslinking are higher: 25 and 26%,respectively, for RNA derivatives with 25 and 10%of the cytidines modified (47).These differences in the level of crosslinking suggest the existence of an intramolecular crosslinking reaction, a process that interferes with the complementary-addressed reaction. Another explanation may be the poor ability of the heavily substituted RNA to take part in the complementary interaction. These effects should be taken into account when the carriers are subjected to intensive derivatization. Ribosomal 1 6 4 RNA carrying aminomethyltrimethylpsoralen adducts has been used for complementary-addressed modification of pCS3 plasmids containing the DNA sequences complementary to 16S RNA (74). The RNA derivative carrying four psoralen adducts per RNA molecule on the average was hybridized to the supercoiled plasmid and irradiated with UV light (A = 360 nm). Under irradiation, the crosslinking between the RNA derivative and the rDNA proceeds with rather high yield: about 50%of the psoralen adducts delivered by the RNA derivative to the complementary DNA sequence bind covalently to the DNA. The experiments described above demonstrate that the polynucleotide derivatives carrying multiple activatable reactive groups can be prepared from DNA and RNA fragments of various origins, and can be used for specific modification of complementary sequences in nucleic acids. As the complementary sequences in this case are long, the yields of crosslinking with them are high despite the use of inefficient groups and despite the intramolecular reactions that will obviously occur in such derivatives. On the other hand, the crosslinking occurs within the extended nucleic acid sequences complementary to the carriers, and one does not know whether the reaction occurs randomly along these sequences or if the crosslinks are concentrated within some small area of these sequences.


309

111. Biochemical Applications of Complementary-Addressed Modification

A. Site-Directed Mutagenesis The first experiments on gene-directed mutagenesis in viruses and plasmids were performed using polyalkylating derivatives of RNA and DNA fragments. Bacteriophage T7 early transcripts carrying ClR;,dR’- reactive groups (polyalkylating transcripts) were used for modification of the corresponding genes in the T7 DNA (9). Transcripts of early gene 1.3 and mixture of transcripts of early genes 0.3 and 1.1were used as carriers. The function of gene 1.3 is to code T7 DNA ligase. One of the functions of gene 0.3 is to overcome host restriction, and the function of gene 1.1 is unknown. T7 DNA modified with the polyalkylating transcript of gene 1.3 was packaged i n vitro into T7 phage particles and the particles were used for infection of E . coZi B. The plaques obtained after the infectioning were tested for ability to grow on the ligase-deficient E . coli BL2 strain. Four of 140 plaques contained T7 mutants defective in the gene 1.3. As the mutations affected only one chain of the phage DNA, each of these plaques had to contain progenies of the intact and modified T7 DNA strands, and the last one produced phage mutants. The amounts of mutant phages in each of the four plaques varied from 0.7 to 1.5%. Control experiments were performed identically, with the exception that derivatization of RNA transcripts was omitted. In these experiments, no mutants were found. Similar experiments were performed with T7 DNA modified at genes 0.3 and 1.1. The modified DNA was used for transfection of suppressor E . coZi strain with subsequent selection of the mutants on the E . coli strain that had a normal restriction system. Both genes 0.3 and 1.1 were affected by the modification. Three of the 24 plaques produced by transfection contained mutants, and the amounts of mutant phages varied from 2 to 10%in different plaques. No mutations in other genes were observed. Some suggestions were made about the nature of mutagenic action of the polyalkylating RNA derivatives. RNA covalently bound to one of the DNA strains would deprive it of its replicative capacity and there would be no mutations produced. However, the transcript crosslinked to T7 DNA may be hydrolyzed enzymatically in vivo, and in the course of the subsequent replication, mutations may occur due to some scattered residual damages of the DNA chain.

310


Another set of experiments were performed with the plasmid pBR322 (73). Gene Tc’ of the plasmid was chosen as the target. The carriers were prepared of the 377 base-pair DNA fragment related to the Tc‘ gene which was isolated from the EcoRl and BamHI digest of the plasmid. Digestion of this fragment with exonuclease I11 was used to prepare the shorter single-stranded DNA fragments. These fragments were derivatized with ClR~,R”Cl.The polyalkylating fragments were bound to the supercoiled plasmid using the technique of Dloops formation. Up to 20% of plasmids formed D-loops. After crosslinking the polyalkylating fragments to the plasmid DNA, the latter was used for transfection of E. coli C 600 (rec A+) and HB 101 (rec A-). In both cases, mutations in the Tc‘ gene appeared with yield 0.4%. Sequencing of the isolated mutant plasmids revealed an interesting mechanism of mutagenesis functioning when a complementary polynucleotide fragment binds to DNA. In the mutant plasmids analyzed, the main type of mutation was the appearance of a 7- or 8-basepair tandem direct repeat in a fixed site of the Tc‘ gene downstream of a palindrome. It was proposed that the palindrome is responsible for the repeat appearance. According to the proposed scheme, the polyalkylating DNA fragment bound to the plasmid has a hairpin structure due to the presence of the palindrome, which is stabilized by crosslinking with two aliphatic 2-chloroethylamino groups of the reagent. Removal of the modified group from the DNA and the bound fragment, and substitution of the bound fragment for the displaced DNA strand, resulted in the repeat formation (Fig. 2). A 164 ribosomal RNA derivative carrying photoactivatable aminomethyltrimethylpsoralen monoadducts was used for modification of the complementary DNA sequence in a supercoiled plasmid pCS3 (74). The plasmid is a derivative of pBR313 containing an insert of part of the E. cob cistron including the 3‘-60% of the 16-S rDNA. Superhelical plasmid DNA took up the RNA derivative to form Rloops in about 30%yield of the R-looped molecules. RNA in the loops was further crosslinked to the DNA by irradiation and the covalent hybrids were isolated. Electron microscope investigation of the linearized plasmid molecules showed that the RNA is bound to a unique site on the DNA. The covalently modified R-looped plasmids were used to examine the effect of several types of damage on the ability of the plasmid to transform E. coli and survive. R-looped DNA was treated with RNase or S1 nuclease or alkali. The latter treatment was done in order to split DNA at the psoralen-modified residues. Each DNA sample was used to transform E. coli C-600 cells and the ampicillin-resistant colonies formed by transformed cells were studied.


311

FIG.2. Mechanism of tandem repeat formation in the Tcrgene of plasmid pBR322 modified with polyalkylating DNA fragment (73). (A) Plasmid DNA and the complementary polyalkylating DNA fragment. (B) Formation of the D-loop and crosslinking. (C) Degradation of the displaced DNA strand and removal of alkylated residues by the reparation systems. (D) Heteroduplex formed as the final result of the repair process. (E) DNA molecules produced by replication of the heteroduplex.

Nine percent of the colonies from the transformation done with Rlooped plasmid treated with S1 nuclease showed plasmid DNA with large deletions covering rDNA sequence. As checked by control experiments, these deletions were not the result of S1 nuclease treatment alone. Colonies screened from transformants with (undamaged) psoralen-containing R-looped plasmids not treated with alkali or nucleases all showed normal plasmid molecular weights as did control samples having no R-loops. The experiments described demonstrate the high efficiency of complementary-addressed modification as an approach to gene-directed mutagenesis in uitro. These first experiments show also that mutations including long DNA stretches, rather than scattered point mutations, can be predominant in the case of modification with polynucleotide derivatives with multiple reactive groups.

B. Complementary-Addressed Modification of Cellular Nucleic Acids To effect complementary-addressed modification of cellular nucleic acids, reactive oligonucleotide derivatives must find their way to targets, surviving the attack of cellular enzymes on their oligonu-

312


cleotide carriers and the attacks of various nucleophiles at their reactive groups. Delivery of complementary-addressed reagents into a cell is also a problem, for cellular membranes are not easily permeable to charged substances such as oligonucleotides. The question of permeability of cellular membranes to polynucleotides was investigated thoroughly and it was found that animal cells have a natural mechanism for taking up nucleic acids. Evidence has been presented demonstrating the uptake of substantial quantities of DNA, RNA and synthetic polynucleotides by cells (75-77). The permeability of cells can be enhanced by detergent treatment (78, 79). Another way to enhance the uptake of exogenous nucleic acids is to treat cells with nucleic acids together with polycationic substances such as DEAE-dextran, spermine, polylysine, polyarginine, and polyornithine (80-83). Precipitation with calcium phosphate has been used to enhance the uptake of DNA into cells (84)and this method was used for the delivery of oligoadenylates into cells (85, 86). Various short oligonucleotides can be precipitated by calcium phosphate (87). The use of the liposome technique (88,89)for the delivery of complementary-addressed reagents into cells can be another efficient approach to achieve this goal. A straightforward approach to design complementary-addressed reagents capable of entering living cells is the use of nonionic oligonucleotide analogs. Oligonucleotide ethylphosphotriesters and oligonucleoside methylphosphonates penetrate cellular membrane easily and are capable of complementary interactions (90-98). A variety of oligonucleotide phosphotriesters with different alkyl substituents can be prepared according to a general method, namely transesterification of oligonucleotide chlorophenyl phosphotriesters (93). The use of nonionic oligonucleotide carriers solves also the problem of stability of complementary-addressed reagents within cells, for blocked phosphates are not attacked by nucleases. Several attempts to use synthetic oligonucleotides for affecting the intracellular processes have been reported. 2'-5'-Oligoadenylates have been introduced into animal cells using the calcium phosphate precipitation technique in investigations of the mechanism of interferon action (85, 86). The observed biological effects indicated that the oligonucleotides entered the cells. It has been reported that a synthetic oligodeoxyribonucleotide complementary to both the 3' and 5' termini of Rous sarcoma virus 35-S RNA blocks translation of the viral RNA in uitro. When this oligomer was added to chick embryo fibroblast cells infected with RSV, production of virus by the cells was inhibited and there was decreased cell transformation (94,95).It was


313

inferred that the oligonucleotide entered the cells and annealed with viral sequences therein, interfering in some manner with viral production and cell transformation. It is interesting that the oligonucleotide with hydrophobic protecting groups at terminal hydroxyl groups demonstrated more pronounced biological effects than the unblocked oligonucleotide. This fact may be ascribed to the enhanced uptake of the blocked oligonucleotide or to its higher efficiency in interfering with the RNA functions. Nonionic derivatives of oligonucleotides complementary to the AC-C-A terminal sequence of tRNAs and to the anticodon loop of tFiNALys inhibit growth of mammalian cells in culture (90, 92). Treatment of the permeable mutant E. coli M L 308-225 with oligodeoxyribonucleoside methylphosphonates complementary to the ShineDalgarno sequence in 16-S ribosomal RNA resulted in inhibition of both protein biosynthesis and colony formation (91). This effect can obviously be attributed to the blocking of the ribosomal mRNA binding site. The experimental data show that nonionic oligonucleotide derivatives and even unmodified oligonucleotides can enter animal cells and affect to some extent biochemical processes therein, most probably by annealing with complementary single-stranded sequences of nucleic acids and interfering with their functions. Oligonucleotides carrying acridine dye groups have recently been synthesized that are capable of very tight binding to the complementary sequences (96). The excess binding energy is provided in this case by the intercalating dye groups, which stabilize the complementary complexes. Such oligonucleotide derivatives that do not damage polynucleotides can be the efficient specific inhibitors of nucleic acid function. Modification of cellular nucleic acids with complementary-addressed reagents was studied by use of alkylating derivatives of oligothymidylates, the expected targets for which are the poly(A) sequences of cellular messenger RNAs (97-101). Krebs ascites tumor cells were treated with (Tp)gU(CHRCl)at 20". Up to 4% of the reagent introduced into the reaction mixture bound to the cells, and 40% of the bound reagent formed covalent bonds with the cellular biopolymers. The alkylating oligonucleotide ethylphosphotriester derivatives bound to cells more efficiently. Up to 20% of the derivatives introduced into the reaction mixture bound to cells and up to 60% of this amount reacted with cellular biopolymers. Analyses of modified biopolymers from cells treated with alkylating oligothymidylate derivatives have revealed that the derivatives alkylate both cellular proteins and nucleic acids.

314


Derivatives of oligothymidylates with unblocked phosphates were relatively more efficient than oligothymidylate ethylphosphotriesters in reaction with proteins. This fact can obviously be ascribed to the electrostatic effects, which play an important role in nucleic acidprotein interactions. Nucleic acids were alkylated both in nuclei and cytoplasm. Half of the modified sequences in RNA were represented by poly(A) sequences which comprise about 1% of the total cellular RNA content. This fact of preferential modification of poly(A) sequences demonstrates that complementary-addressed modification of cellular RNA took place. The effect of higher temperature on the selectivity of alkylation is in accordance with this conclusion. At 20" the efficiency of alkylation of poly(A) sequences is 110 times higher than that of modification of the rest of RNA. At 37", where the complementary complexes are not so stable, this difference is 70-fold. Stability of the carrier oligonucleotides in cells was checked in parallel experiments with oligothymidylate derivatives labeled either at the 5' terminus with 32P or in the alkylating group carrying a 14Clabel (97). The extent of modification of cellular nucleic acids determined from 32Pand I4C incorporations are similar. Therefore the carrier oligonucleotide is not degraded to any appreciable extent in the cell during the 4-hour incubation at 37". Interaction of reactive polynucleotide derivatives with cells has been investigated in experiments with an alkylating derivative of transfer RNA, tRNA(CHRC1) (102).L-1210 cells took up 0.6% of the tRNA(CHRC1)added to the medium, and 55%of the bound derivative concentrated in the cell nuclei. Degradation of the tRNA(CHRC1) in cells was rather slow. In 3 hours of incubation at 37", half of the derivative molecules were still intact. The possibility of damage of preselected mRNAs by means of complementary-addressed reagents was investigated using modification of mRNA coding for the K light chain of immunoglobulin G in mouse MOPC-21 myeloma cells (103).The cells were treated with the oligonucleotide reagent (ClRCH2NH)pT-G-C-T-C-T-G-G-T-T-T, complementary to the fragment of mRNA coding for the variable sequence of the K chain. The treatment substantially suppresses synthesis of IgG in the cells while synthesis of other cellular proteins is only slightly affected. In control experiments where cells were treated with the same oligonucleotide derivative cariying a deliberately inactivated (hydrolyzed) 2-chloroethyl group or with the alkylating derivative of noncomplementary oligonucleotide PA-C-A-C-A-C the effect on the IgG synthesis was negligible. An attempt has been made to suppress viral multiplication with


315

complementary-addressed reagents specific to the viral nucleic acids. Chicken fibroblasts infected with fowl plague virus were treated with alkylating oligonucleotide derivatives. (CIRCH,NH)d(pC-C-T-T-G-T-T-T-C-T) (A) d(pA-C-C-A-A-A-A-G-C)rA(CHRCI) (B) (CIRCH,NH)d(pT-T-T-T-C-C-C-T-T-T-T) (C) and d(pC-C-C-A-A-A-C)rA(CHRCl) (D)

Reagent (A) was complementary to the 5'-universal terminal region of the viral RNAs in positions 4-13. Reagent (B) was complementary to the universal 3' terminal sequence in the viral RNAs. Reagent (C) was complementary to a fragment in the central part of the mRNA coding for the hemagglutinin, and was identical in sequence with the fragment 1035-1046 of the corresponding viral RNA. Reagent (D) had no particular complementary sequence in viral nucleic acids and was used as a control. Treatment with reagent (A) results in decrease of the virus production in the system by factor of 15 x lo3. Reagent (C) is 1% as efficient, and the other two reagents showed no effect. The results demonstrate the specific effect of the oligonucleotide derivatives on virus multiplication, which is determined by the oligonucleotide parts of the reagents.

IV. Concluding Remarks The results presented clearly demonstrate that a variety of chemical reactions may be carried out within oligonucleotide duplexes. The reaction of oligonucleotide derivatives with complementary oligonucleotides and nucleic acids was the topic of our review. However, the possibilities of condensation of two oligonucleotides or analogs put in close contact by binding to a common complementary strand is intensively being investigated as an approach to chemical ligation (105). One may expect that two complementary oligonucleotides each bearing some reactive groups, one at the 3' and the other at the 5' end, will be widely used in the near future to study reactions proceeding in close contact, thus modelling essential features of enzymatic reactions and affinity labeling of proteins. Therefore, one may say that a new field of bioorganic chemistry, namely organic chemistry of nucleic acid duplexes is appearing. (106). Complementary-addressed modification with derivatives cleaving nucleic acids may provide a useful addition to the use of restriction endonucleases. The approach also provides the possibility to introduce some labels (fluorescent, paramagnetic) into definite points of

316


V. VLASSOV

nucleic acids for functional investigations. Specific scission of DNA in polytene chromosomes in situ by alkylating oligoadenylate derivatives has been demonstrated, and it was concluded that the approach may be useful for the investigation of chromosome structure and functions. The elaborated version of gene-directed mutagenesis with reactive polynucleotide derivatives is not as precise as the most popular method based on the priming of DNA replication with synthetic oligonucleotides containing desired point mutations. However, the version described in this review although applied up-to-now only to in uitro experiments may easily be transformed to in uiuo mutagensis. Of course, the most promising applications of complementary-addressed modification are related to modification of definite nucleotide sequences in uiuo to suppress viral multiplication, as has already been done with influenza virus. The selective arrest of the synthesis of MOPC-21 immunoglobulins demonstrates the possibility to suppress the expression of immunoglobulin genes in definite clones, thus providing the possibility to substitute selective immunodepression for total unspecific immunodepression in the cases of some antoimmune diseases, and in transplantation to overcome the histocompatibility barrier. To suppress the multiplication of definite (e.g., malignant) cells, one must find an approach to carry out complementary addressed modification of DNA in the cell. This has not been done so far. However, the composition of alkylated bases in the small fraction of DNA modified with [ d T ~ ( E t ) ] n u ( c H R Cin l ) ~Krebs I1 carcinoma cells is in favor of the addressed character of this alkylation. Maybe, this modification of DNA with a complementary-addressed reagent might be done in the replication fork or in special states of chromosomes. It was recently found that DNA of the condensed metaphase chromosomes behaves as denatured towards alkylation with dimethylsulfate (108). Concluding, we may state that complementary-addressed modification of single-stranded nucleic acids is rather advanced in its chemical part and now comes into the period of intensive search for biological applications.

REFERENCES 1. J. Watson and F. Crick, Nature 171,737 (1953). 2. A. M.Belikova, V. F. Zarytova and N. I. Grineva, Tetrahedron Lett. No. 37,3557

(1967). 3

Oligothymidylate esterified with ethyl residues at internucleotide phosphates.


317

3. N. I. Grineva, V. F. Zarytova and D. G. Knorre, Zzu. Sib. Otd. Akad. Nauk SSSR, Ser. Khim.Nauk 5, 118 (1968). 4. N. I. Grineva, Biokhimiya 42, 370 (1977). 5. B. R. Baker, “Design of Active-Site Directed Irreversible Enzyme Inhibitors.” John Wiley, New York, 1967. 6. “Methods in Enzymology” (W. B. Jakoby and M. Wilchek, eds.), Vol. 47, Academic Press, New York, 1977. 7. “Affinity Modification of Biopolymers” (D. G. Knorre, ed.). Nauka, Novosibirsk, 1983. 8. J. Summerton, J . Theor. B i d . 78, 77 (1979). 9. R. I. Salganik, G. L. Dianov, L. P. Ovchinnikova, E. N. Voronina, E. B. Kokoza, and A. V. Mazin, PNAS 77,2796 (1980). 10. C. C. Price, G. M. Gaucher, P. Koneru, R. Shibakawa, I. R. Sowa and M. Yamaguchi, BBA 166,327 (1968). 1 1 . A. M. Belikova, N. I. Grineva and G. G. Karpova, lzu. Sib. Otd. Akad. Nauk, Ser. Khim. Nauk (1972) iss. 4, 101. 12. N. I. Grineva, D. G. Knorre and V. A. Kurbatov, Mot. Biol. (USSR) 4, 814 (1970). 13. “Organic Chemistry of Nucleic Acids” (N. K. Kochetkov and E. I. Budovsky, eds.), p. 575. “Khymiya,” Moscow, 1970. 14. A. Holy and K. H. Scheit, BBA 138,230 (1967). 15. B. Singer, This series 15,219 (1975). 16. W. C. J. Ross, “Biological Alkylating Agents.” Butterworth, London and Washington, 1962. 17. J. Summerton,J. Theor. Biol. 78,61 (1979). 18. V. V. Vlassov and S. A. Kazakov, Bioorg. Khym. (USSR) 8,499 (1982). 19. V. V. Vlassov, V. V. Corn, E. M. Ivanova, S. A. Kazakov and S. V. Mamayev, FEBS Lett. 162, 286 (1983). 20. J. J. Roberts and A. J. Thomson, This series 22, 71 (1979). 21. N. P. Johnson, J. P. Macquet, J. L. Wiebers and B. Monsarrat, NARes 10, 5255 (1982). 22. W. Bauer, S. L. Gonias, S. K. Kam, K. S. Wu and S. J. Lippard, Bchem 17, 1060 ( 1978). 23. J. Filipski, K. W. Kohn, R. Prather and W. M. Bonner, Science 204, 181 (1979). 24. “Bleomycin: Chemical, Biochemical and Biological Aspects” (S. M. Hecht, ed.), Springer-Verlag, New York, 1979. (See also Miller and Zahn in Vol. 20 of this series.) 25. R. P. Hertzberg and P. B. Dervan, JACS 104,313 (1982). 26. P. G. Schultz, J. S. Taylor and P. B. Dervan,JACS 104,6861 (1982). 27. P. G. Schultz and P. B. Dervan, JACS 105,7748 (1983). 28. M. I. Stockman, Phys. Lett. 76A, 191 (1980). 29. L. Z. Benimetskaya, A. L. Kozionov, S. Yu. Novozhilov and M. I. Stockman, Dokl. Akad. Nauk S S S R 272,217 (1983). 30. N. V. Bulychev, A. V. Lebedev, L. Z. Benemetskaya, A. L. Kozinov, Yu. E. Nesterikhin, S. Yu. Novozhilov, S. G. Rautian and M. I. Stockman, Bioorg. Khym. (USSR) 10,520 (1984). 31. R. I. Salganik, G. L. Dianov, V. A. Kurbatov, G. V. Shishkin and A. G. Gall, Dokl. Akad. Nauk SSSR 239,217 (1978). 32. V. V. Vlassov, 0. I. Lavrik, S. V. Mamayev, S. N. Khodyreva, V. E. Chizhikov and A. F. Shvalye, Mol. Biol. (USSR) 14, 531 (1980).

3 18


33. V.V. Vlassov, D. M. Graifer, G. G. Karpova and V. E. Chizhikov, Bioorg. Khym. (USSR) 7,787 (1981). 34. V. K. Ryte, G. G. Karpova and N. I. Grineva, Bioorg. Khym. (USSR) 3, 31 (1977). 35. N. I. Grineva and T. S. Lomakina, Zh. Obshch. Khym. (USSR) 42,1630 (1972). 36. 0. I. Gimautdinova, N. I. Grineva, G. G. Karpova, T. S. Lomakina and E. L. Shelpakova, Bioorg. Khym. (USSR) 4, 917, (1978). 37. G. F. Mishenina, V. V. Samukov and T. N. Shubina, Bioorg. Khym. (USSR) 5,886 (1979). 38. 0.I. Gimautdinova, G. G. Karpova and N. A. Kozyreva, Mol. B i d . (USSR) 16,752 (1982). 39. V. F. Zaxytova, V. K. Ryte and T. S. Chernikova, Bioorg. Khym. (USSR) 3, 1626 (1977). 40. V. L. Drutsa, V. F. Zrytova, D. G. Knorre, A. V. Lebedev, N. I. Sokolova and Z. A. Shabarova, NARes 5, 185 (1978). 41. V. V. Vlassov and S. A. Kazakov, Bioorg. Khym. (USSR) 9,530 (1983). 42. A. A. Gall, V. A. Kurbatov, A. A. Mustayev and G . V. Shishkin, Izv. Sib. Otd. Akad. Nauk S S S R , Ser. Khim. Nauk 2,99 (1979). 43. S. I. Oshevski, M. A. Grachev and A. A. Mustayev, Bioorg. Khym. (USSR) 9,958 (1983). 44. M. B. Gottikh, M. G. Ivanovskaya, V. P. Veiko and 2.A. Shabarova, Bioorg. Khym. (USSR) 7, 1310 (1981). 45. S. I. Oshevski, FEBS Lett. 143, 119 (1982). 46. V. V. Vlassov, A. A. Gall, A. A. Godovikov, V. F. Zarytova, I. V. Kutyavin, I. P. Motoviliva and G. V. Shishkin, Dokl. Akad. Nauk S S S R 274, 1244 (1984). 47. J. Summerton and P. A. Bartlett,/MB 122, 145 (1978). 48. V. V. Vlassov and D. G. Knorre, Mol. Biol. (USSR) 8,234 (1974). 49. D. G. Knorre and T. A. Chimitova, FEBS Lett. 131,249 (1981). 50. N. I. Grineva, G. G. Karpova, L. M. Kuznetsova, T. V. Venkstern and A. A. Bayev, NARes 4,1609 (1977). 51. N. I. Grineva and G. G. Karpova, Mol. Biol. (USSR) 8,832 (1974) 52. L. Z. Benimetskaya, N. I. Grineva, G. G. Karpova, N. P. Pichko and T. A. Chimitova, Bioorg. Khim. (USSR) 3, 903 (1977). 53. V. V. Vlassov, N. I. Grineva and D. G. Knorre, Izv. Sib. Otd. Akad. Nauk SSSR, Ser. Khim. Nauk 1 , 104 (1969). 54. 0.I. Gimautdinova, G. G. Karpova, T. S. Lomakina, E. L. Shelpakova, A. I. Chemassova and N. I. Grineva, Bioorg. Khym. (USSR) 6 , 7 0 (1980). 55. N. I. Grineva and E. G. Saikovich, Bioorg. Khym. (USSR) 5,563 (1979). 56. G. G. Karpova, N. P. Pichko, T. A. Chimitova and N. I. Grineva. Mol. Biol. (USSR) 13, 1012 (1979). 57. V. V. Vlassov, N. I. Grineva and D. G. Knorre, FEBS Lett. 20,66 (1972). 58. V. V. Vlassov and L. M. Skobeltsina, Bioorg. Khym. (USSR) 4,550 (1978). 59. A. M. Belikova and N. I. Grineva, Izv. Sib. Otd. Akad. Nauk SSSR, Ser. Khim. Nauk 5, 119 (1971). 60. A. M. Belikova, N. I. Grineva, D. G. Knorre and S. D. Mysina, Dokl. Akad. Nauk S S S R 212 876 (1973). 61. N. I. Grineva and G. G. Karpova, Bioorg. Khym. (USSR) 1,588 (1975). 62. L. Z. Benemetskaya, G. G. Karpova and N. I. Grineva, Bioorg. Khym. (USSR) 4, 1372 (1978). 63. V. V. Gorn, G. G. Karpova, D. G. Knorre, I. V. Kutyavin and N. P. Pichko, Bioorg. Khym. (USSR) 8,1225 (1982).


3 19

64. V. V. Corn, G. G. Karpova, D. G. Knorre, I. V. Kutyavin and N. P. Pichko, Dokl. Akad. Nauk SSSR 270,613 (1983). 65. N. I. Grineva, Vestnik Akad. Med. Nauk SSSR No 2, 83 (1981). 66. V. V. Vlassov, S. A. Gaidamakov, V. V. Corn and S. A. Grachev, F E B S Lett., submitted. 67. A. C. Palnenberg, E. M. Kirby, M. R. Janda, N. L. Drake, G. M. Duke, K. F. Potratz and M. S. Collett, NARes 12, 2969 (1984). 68. M. A. Grachev and S. I. Oshevski, Dokl. Akad. Nauk SSSR 272,1259 (1983). 69. L. Z. Benemetskaya, N. V. Bulychev, A. L. Kozionov, A. V. Lebedev, Yu. E. Nesterikhin, S. Yu. Novozhilov, S . G. Rautian and M. 1. Stockman, FEBS Lett. 163, 144 (1983). 70. L. Z. Benemetskaya, N. V. Bulychev, A, L. Kozionov, A. V. Lebedev, S. Yn. Novozhilov and M. 1. Stockman, NARes Symp. Ser. 14,323 (1984). 71. A. S. Boutorin, V. V. Vlassov, S. A. Kazakov, 1. V. Kutyavin and M. A. Podyminogin, FEBS Lett. 173,43 (1984). 72. A. V. Mazin, G. L. Dianov and R. I. Salganik, Mol. Biol. (USSR) 15,252 (1981). 73. R. I. Salganik, A. V. Mazin, G. L. Dianov and L. P. Ovchinnikova, Genetika (USSR)20,1244 (1984). 74. P. K. Chatterjee and C. R. Cantor, JBC 257,9173, (1982). 75. P. L. Schell, BBA 262,467 (1972). 76. H. Woodland and S. E. Ayers, Biochem. J . 144, 11 (1974). 77. F. Herrera, R. H. Adamson and R. C. Gallo, PNAS 64, 1543 (1970). 78. D. Billen and A. C. Olson,J. Cell. Biol. 69, 732 (1976). 79. M. R. Miller, J. J. Castellot and A. B. Pardee, Bchem 17, 1073 (1978). 80. F. E. Farber, J. L. Melnick and J. S. Butel, BBA 390, 298 (1975). 81. B. V. Howard, M. K. Estes and J. S. Pagano, BBA 228, 105 (1971). 82. P. L. Schell BBA 340,323 (1974). 83. M. Ehrlich, L. P. Sarafyan and D. J. Myers, BBA 454,397 (1976). 84. F. L. Garaham and A. J. Eb, Virology 52,456 (1973). 85. A. G. Hovanessian, J. N. Wood, E. Mews and L. Montagnier, PNAS 76, 3261 (1979). 86. A. G. Hovanessian and J. N. Wood, Virology 101,81 (1980). 87. K. Stenberg, B. Oberg and J. B. Chattopadhyaya, BBA 697, 170 (1982). 88. G. Gregoriadis and A. Allison “Liposomes in Biological Systems.” John Wiley, New York, 1980. 89. L. E. Stephanovich, S. A. Linde, G. G. Karpova, N. I. Grineva and S. A. Magarill, Biokhimiya 44, 1289 (1979). 90. P. S. Miller, L. T. Braitennan and P.O.P. Ts’o, Bchem 16, 1988 (1977). 91. K. Jayaraman, K. McParland, P. Miller and P.O.P. Ts’o, PNAS 78, 1537 (1981). 92. P. S. Miller, K. B. McParland, K. Jayaraman and P.O.P. Ts’o. Bchem 20, 1874 (1981). 93. N. K. Daniluk, V. A. Petrenko, P. I. Pozdniakov, G. F. Sivolobova and T. N. Shubina, Bioorg. Khim. 7 , 703 (1981). 94. P. C. Zamecnik and M. L. Stephenson, PNAS 75,280 (1978). 95. M. L. Stephenson and P. C. Zamecnik, PNAS 75,285 (1978). 96. U. Asseline, M. Delarne, G . Lancelot, F. Toulme, N. T. Thuong, T. MontenayGarestier and C. Helene, PNAS 81,3297 (1984). 97. E. M. Ivanova, G. G. Karpova, D. G. Knorre, V. S. Popova, A. S. Ryte and L. E. Stephanovich, Mol. Biol. (USSR) 18, (1984). 98. D. G. Knorre, V. F. Zarytova, G. G. Karpova and L. E. Stephanovich, NARes Symp. Ser. 8, 195 (1981).

320

DMITRI K N O W AND VALENTIN V. VLASSOV

99. G. G. Karpova, D. G. Knorre, A. S. Ryte and L. E. Stephanovich, FEES Lett. 122, 21 (1980). 100. V. F. Zarytova, E. M. Ivanova, G. G. Karpova, D. G. Knorre, N. P. Pichko, A. S. Ryte and L.E. Stephanovich, Bioorg. Khym. 7, 1512 (1981). 101, V. V. Vlassov, G. G. Karpova, D. G. Knorre, N. P. Pichko and A. S. Ryte, Dokl. Akad. Nauk SSSR 274,965 (1984). 102. G. Y. Soloviev, N. I. Drize, V. L. Surin, E. Y. Krinetski and N. I. Grineva, M o l . Genet. Virusol. (USSR) No. 12,34(1983). 103. V. V. Vlassov, A. A. Godovikov, V. F. Zarytova, E. M. Ivanova, D. G. Knorre and I. V. Kutiavin, Dokl. Akad. Nauk SSSR 276, 1263 (1984). 104. V. V. Vlassov, V. F. Zarytova, I. V. Kutiavin, L. V. Yurchenko, A. G. Bukrinskaja and N.K. Sharova, Voprosy Vimsol., submitted. 105. Z. A. Shabarova, Soviet Scientific Reviews section D. Phys. B i d . Reu. 5, l(1984). 106. D. G. Knorre and Z. A. Shabarova, Bioorg. Khym. (USSR) 10, 129 (1984). 107. G. M. Dymchits, G. V. Rumyantseva, L. A. Frumgarts, A. D. Grusdev, G. A. Zainiev, I. E. Shilova, G. G. Karpova, V. I. Yamkovoy and N. I. Grineva, Mol. Biol. (USSR) 15,86 (1981). 108. N. D. Belyaev, V. G. Budker, V. A. Dubrovskaya and N. M. Matveeva, FEES Lett. 154,285 (1983).

Addendum: Hypermodified Nucleosides of tRNA A synthesis of optically pure (S)-wybutine (LXII) has been reported recently (1). In this procedure, 1-benzylwyeine (LXVIII) is first transformed into its 7-formyl derivative (Vilsmeier-Haack reac(R)-{2-carboxy-2-[(methoxytion) and then coupled with carbonyl)amino]ethyl}triphenylphosphonium chloride (Wittig reaction) obtained from L-serine. The reaction product after methylation of the carboxyl group and the removal of the benzyl protection group by hydrogenolysis gives (S)-wybutine. It has been suggested that reduction in the steric repulsion between 4-methyl and S-p-~-ribofuranosylgroups is a major driving force of acid-catalysed hydrolysis of wyosine (XX)(2). In another report it was shown that the acid hydrolysis of wyosine (XX) and its 5’phosphate proceeds b y an A-1 mechanism and, contrary to the common nucleosides and their phosphates, with comparable rates (3). For the first time, the structure of the degradation product of the tricyclic core of wyosine (XX) observed under alkaline conditions was established. At a pH above 9, apart from the competitive glycosylic bond cleavage, the inner ring ofwyosine opens up due to hydroxyl ion attack at the carbonyl site (C-9) (2).

REFERENCES 1. T. Itaya and A. Mizutani, NARes Synip. Ser. 15, 13 (1984). 2. T. Itaya and T. Harada,/CS Cheni. Cornmuti. 858 (1984). 3. R. Golankiewicz, E. Zielonacka-Lis, and W. Folkman, NARes 13,2443 (1985).

32 1


Index

A Adducts, in DNA in cells, radiationinduced, 147-148 Adenosines, hypermodified chemistry, 46-49 codon-anticodon interactions and, 57-

B Bases damage to DNA in cells, irradiation and, 143-146 products formed in oxygenated solutions by ionizing radiations, 123-

59

129

synthesis, 42-46 Alkali-labile bonds, in DNA irradiated in cells, 140-143 Aminoacyl-transfer ribonucleic acid synthetases, role of anticodon in recognition of concise backround of problem, 239-

243 general remarks, 258-262 Anoxic solutions, products formed from DNA by ionizing radiations, 130-

131 Anticodon role in acceptor function in eukaryotes, 252-258 in prokaryotes, 243-252 role in recognition of tRNA by aminoacyl-tRNA synthetases concise background of problem,

239-243 general remarks, 258-262 Anticodon loop fragments, studies on,

59-61 Anticodon loop structure simulation studies, 63-64 wyosine in tRNAPhein, 61-63 Antigen gene expression, in trypanosomes, DNA rearrangement and, 3-4 Antigenic variation in trypanosomes differential gene expression and,

2-3 gene conversion as mechanism for,

unaltered, release from D N A by ionizing radiations, 123

C Cell(s), irradiation of DNA in 138-140 base damage, 143-146 clusters of damage sites, 146-147 radiation-induced adducts, 147-148 single-strand breaks and alkali-labile bonds, 140-143 Codon-anticodon interactions, of hypermodified nucleosides of tRNA, 53-

54,64-67 within anticodon loop architecture,

59-64 unique features imposed by, 54-59 Complementary-addressed modifications biochemical applications cellular nucleic acids, 311-315 site-directed mutagenesis, 309-31 1 of model compounds oligonucleotides, 301-307 polynucleotides, 307-308 specificity of, 299-301 Complementary-addressed reagents, synthesis of oligonucleotide derivatives, 294-299 reactive groups for, 292-294 Computer, search for coordinated base changes in 16-S-like rRNA and,

4-7

199-214 323

324

INDEX

D

Damage sites, from ionizing radiation of DNA, clusters of, 146-147 Deoxyribonucleic acid effects of ionizing radiations in solid state direct action, 132 products, 134-138 yields of chemical changes, 132-134 effects of reactive species formed from water by ionizing radiation, 117118 base products in oxygenated solutions, 123-129 deoxyribose products in oxygenated solutions, 118-122 hydroxyl radical attack on double helical DNA, 129-130 products in anoxic solutions, 130-131 release of unaltered bases, 123 single strand breaks, 122-123 irradiated, quantitative measurements that should be made on, 148-149 irradiation in cells, 138-140 base damage, 143-146 clusters of damage sites, 146-147 radiation-induced DNA adducts, 147-148 single-strand breaks and alkali-labile bonds, 140-143 mechanisms of action of ionizing radiations on, 116-1 17 rearrangement, antigenic gene expression in trypanosomes and, 3-4 of SV40 large T antigen, binding properties, 227-228 Deoxyribose, in oxygenated solution, products of ionizing radiations, 118122

E Early promoter, of SV40, organization of, 220-225

EF-G models implying energy contribution of, to ribosomal translocation, 9091

sequence of events promoted by, 8486 eIF-2, ternary initiation complex and, 274-276 eIF-3, small ribosomal subunit and, 277-279 Electron microscopy, immune, arrangement of proteins in small ribosomal subunit, 271-274 Energetics, of ribosomal translocation catalysis of, 88-89 general considerations, 86-88 models implying all displacements to be thermodynamically spontaneOUS,91-93 models implying energy contributions of EF-G with GTP, 90-91 Eukaryotes, role of anticodon in receptor function of, 252-258

G Gene conversion, in trypanosomes antigenic repertoire evolution and, 17-20 endpoints, 14 extent, 7-14 frequency, 15 as mechanism for antigenic variation, 4-7 transcription and, 15-17 H

Hydroxyl radicals, attack on double helical DNA, 129-130 I

Ionizing radiations effects of reactive species formed from water on DNA in dilute solution, 117-118 base products in oxygenated solutions, 123-129 deoxyribose products in oxygenated solutions, 118-122 hydroxyl radical attack on double helical DNA, 129-130

325

INDEX

products in anoxic solutions, 130131 release of unaltered bases, 123 single strand breaks, 122-123 effects on DNA in solid state, 131-132 direct action, 132 products, 134-138 yields of chemical changes, 132-134 mechanisms of action on DNA, 116117

with oligonucleotide derivatives, 301-307 with polynucleotides, 307-308 specificity of, 299-301 Oxygenated solutions, effects of ionizing radiations on DNA in base products, 123-129 deoxyribose products, 118-122

K

Peptidyl-transfer ribonucleic acid, helical displacement of, 97-101 Polynucleotides, complementary-addressed modifications of, 307-308 Preinitiation complexes, function and arrangement of components in eIF-2 and ternary initiation complex, 274-276 interactions between eIF-3 and small ribosomal subunit, 277-279 interactions between eIF-3 mRNA and small ribosomal subunit, 279-281 interactions between ternary initiation complex and small ribosomal subunit, 276-277 Prokaryotes, role of anticodon in acceptor function of, 243-252 Protein(s), arrangement in small ribosomal subunit, 268-269 cross-linking experiments, 269-27 1 immune electron microscopy, 271-274

Kinematics, of ribosomal translocation helical displacement of peptidyl-tRNA, 97-101 movements of tRNA on ribosome, 101-105 mutual orientation of two tRNAs, 9397 L

Late promoters, of SV40, organization of, 226-227 M

Messenger ribonucleic acid, interaction between eIF-3 and small ribosomal subunit and, 279-281 Mutagenesis, site-directed, complementary-addressed derivatives and, 309311 N

Nucleic acids, cellular, complementaryaddressed modification of, 311-315 Nucleotides, phylogenetically conserved, distribution in 16-S-like rRNA, 191197

0 Oligonucleotides complementary-addressed derivatives, synthesis of, 294-299 complementary-addressed, modifications of

P

R

Reactive groups, for complementaryaddressed reagents, 292-294 Regulatory region, of SV40, early promoter, 220-225 late promoters, 226-227 Ribonucleic acid, see Messenger RNA; Ribosomal RNA; Transfer RNA Ribosomal ribonucleic acid conservation, functional sites and, 197-199 16-S-like, comparative anatomy of comparative sequence analysis and secondary structure, 156-158 comparison of secondary structure, 158-182

INDEX

distribution of phylogenetically conserved nucleotides in, 191197 functional sites and rRNA conservation, 197-199 phylogenetically variable secondary structure elements, 182-191 le-S-like, computer-assisted search for coordinated base changes in, 199200 searching for, 201-214 tertiary interactions and constraints on primary tRNA structure, 200-201 Ribosomal translocation conformational movements of ribosome and, 105-108 definition, 75-77 energetics of catalysis of, 88-89 general considerations, 86-88 models implying all displacements to be thermodynamically spontaneous, 91-93 models implying energy contributions of EF-G with GTP, 90-91 experimental tests, 77-78 kinematics of helical displacement of peptidyltRNA, 97-101 movements of tRNA on ribosome, 101-105 mutual orientation of two tRNAs, 93-97 main facts concerning, 81-84 sequence of events promoted by EF-G, 84-86 two-tRNA-site model for ribosomal elongation cycle, 79-81 Ribosome arrangement of proteins in small subunit, 268-269 cross-linking experiments, 269-271 immune electron microscopy, 271274 small subunit interactions eIF-3 and, 277-279 eIF-3 and mRNA, 279-281 ternary initiation complex and, 276277

5

Secondary structures of 16-S-like rRNAs comparison of, 158-182 phylogenetically variable elements of, 182-191 Sequence analysis, and RNA structure, of 16-S-like rRNA, 156-158 Simian virus 40 DNA binding properties of large T antigen, 227-228 regulation of transcription by early, 229-231 late, 231-232 regulatory region of, 218-220 organization of early promoters, 220-225 organization of late promoters, 226227 Single-strand breaks in DNA, ionizing radiations and, 122123 irradiation of DNA in cells and, 140143 Specificity, of complementary-addressed modification of oligonucleotides, 299-301 T

T antigen, of SV40, DNA binding property of, 227-228 Ternary initiation complex eIF-2 and, 274-276 small ribosomal subunit and, 276-277 Transcription, regulation in SV40 early, 229-231 late, 231-232 Transfer ribonucleic acid constraints on primary structure, tertiary reactions with 16-S-like rRNA, 200-201 hypermodified nucleosides of, bioorganic chemist’s point of view, 27-35 structural features and codon-anticodon interactions, 53-67 synthesis and chemistry adenosines, 42-49

327

INDEX

uridines, 35-42 wyosines, 4 9 5 3 movements on ribosome, 101-105 mutual orientation of two on ribosome,

93-97 two-site model for ribosomal elongation, 79-81 Transfer ribonucleic acid aminoacyl synthetases, role of anticodon in recognition of concise backround of problem, 239-

frequency of gene conversion in, 15 gene conversion as mechanism for antigenic variation in, 4-7 gene conversion endpoints in, 14 orientation of gene conversion mechanism in relationship with transcription, 15-17 sexual conjugation in, evolution of antigen gene repertoire and, 21

U

243 general remarks, 258-262 Translocation, see Ribosomal translocation Trypanosomes antigen evolution in, gene conversion and, 17-20 antigen gene expression in, DNA rearrangements and, 3-4 antigenic variation due to differential gene expression, 2-3 extent of gene conversion in, degree of homology _.between recombinant sequences and, 7-14

Uridines hypermodified chemistry, 38-42 synthesis, 35-38 wobble, codon-anticodon interactions and, 54-57

W Wyosines, hypermodified in anticodon loop structure, 61-63 chemistry, 51-53 synthesis, 49-51


Contents of Previous Volumes Volume 1 “Primer” i n DNA Polymerase Reactions-F. J . B o l h n The Biosynthesis of Ribonucleic Acid in Animal Systems-R. M . S. S m e l l i e The Role of DNA in RNA Synthesis-Jerard H u r w i t z and J . T. A u g u s t Polynucleatide Phosphorylase-M. Grunberg-Manago Messenger Ribonucleic Acid-Fritz Lipmann The Recent Excitement in the Coding Problem-F. H. C . Crick Some Thoughts on the Double-Stranded Model of Deoxyribonucleic Acid-Aaron B e n d i c h and H e r b e r t S. Rosenkranz Denoturation and Renaturation of Deoxyribonucleic Acid-J. Murmur, R . Rownd, and C . L. Schildkraut Some Problems Concerning the Macromolecular Structure of Ribonucleic Acids-A. S. Spirin The Structure of DNA as Determined by X-Ray Scattering Techniques-Vittoria Luzzati Molecular Mechanisms of Radiation Effects-A. Wacker Volume 2 Nucleic Acids and Information Transfer-Liebe F. C a u u l i e r i and Barbara H. Rosenberg Nuclear Ribonucleic Acid-Henry Harris Plant Virus Nucleic Acids-Roy Markhum The Nucleases of Escherichio coli--I. R . L e h r n u n Specificity of Chemical Mutagenasis-Dauid R . K r i e g Column Chromatography of Oligonucleotides and Palynucleotides-Matthys Stoehelin Mechanism of Action and Application of Azopyrimidines-J. Skoda The Function of the Pyrimidine Base in the Ribonuclease Reaction-Herbert U’itzel Preparation, Fractionation, and Properties of sRNA-G. L. Brown Volume 3 Isolation and Fractianotion of Nucleic Acids-K.

S . Kirby

Cellular Sites of RNA Synthesis-David M . Prescott Ribanucleases in Taka-Diastase: Properties, Chemical Noture, and Applications-Fujio Egami, K e n j i Takahashi, and Tsuneko U c h i d u Chemical Effects of Ionizing Radiations on Nucleic Acids and Related Compounds-Joseph J. Weiss The Regulation of RNA Synthesis in Bacteria-Frederick C . Neidhardt Actinomycin and Nucleic Acid Function-E. Reich und 1. H. G o l d b e r g De Novo Protein in Synthesis in Vitro-B. N i s m a n and]. P d r n o n t Free Nucleotides in Animal Tissues-P. Mundel Volume 4 Fluorinated Pyrimidines-Charles

Heidelberger

Genetic Recombination in Bacteriophage-E. Volkin DNA Polymerores from Mammalian Cells-H. M. K e i r The Evolution of Bare Sequences in Polynucleotides-B. J . McCarthy Biosynthesis of Ribosomes in Bacterial Cells-Syozo 0sarc;a 5-Hydroxymethylpyrimidines and Their Derivotives-T. L. V. Ulbright

329

330

CONTENTS OF PREVIOUS VOLUMES

Amino Acid €sten of RNA. Nucleotides, and Related Compounds-H.

C. Zachau a n d H .

Feldmann Uptake of DNA by living Cells-L.

Ledoux

Volume 5 Introduction to the Biochemistry of 4-Arobinosyl Nucleosides-Seymour S. Cohen Effects of Some Chemical Mutagens and Carcinogens an Nucleic Acids-P. D.Lawley Nucleic Acids in Chloroplasts and Metabolic DNA-Tatsuichi Iwamura Enzymatic Alteration of Macromolecular Structure-P. R . Srinioasan and Ernest Borek Hormones and the Synthesis and Utilization of Ribonucleic Acids-J. R . Tutu Nucleoside Antibiotics-Jack J . Fox, Kyoichi A . Watanabe, a n d Alexander Bloch Recombination of DNA Molecules-Charles A. Thomas, J r . Appendix I. Recombination of a Pool of DNA Fragments with Complementary Single-Chain Ends-C. S. Watson, W. K . Smith, a n d Charles A . Thomas, Jr. Appendix II. Proof that Sequences of A, C, G, and T Can Be Assembled to Produce Chains of Ultimate length, Avoiding Repetitions Everywhere-A. S. Fraenkel a n d J . Gillis The Chemistry of Pseudouridine-Robert Warner Chambers The Biochemistry of Pseudouridine-Eugene Goldwasser and Robert L. Heinrikson Volume 6 Nucleic Acids and Mutability-Stephen Zarnenhof Specificity in the Structure of Transfer RNA-Kin-ichiro

Miuru

Synthetic Polynucleotides-A. M. Michelson, J . Massou1ie;und W. Guschbauer The DNA of Chloroplasts, Mitochondria, and Centrioles-S. Granick and Aharon Gibor Behavior, Neural Function, and RNA-H. Hyden The Nucleolus and the Synthesis of Ribosomes-Robert P. Perry The Nature and Biosynthesis of Nuclear Ribonucleic Acids-C. P. Ceorgieo Replication of Phage RNA-Charles Weissmunn a n d Seoero Ochoa Volume 7 Autoradiographic Studies on DNA Replication in Normal and leukemic Human ChromosomesFelice Guoosto Proteins of the Cell Nucleus-Lubornir S. Hnilicu The Present Status of the Genetic Code-Curl R . W’oese The Search for the Messenger RNA of Hemoglobin-H. Chantrenne, A. Burny, a n d C. baix Ribonucleic Acids and Information Transfer i n Animol Cells-A. A. Hadjiolol: Transfer of Genetic Information during Embryogenesis-Martin Nemer Enzymatic Reduction of Ribonucleotides-Agne Lursson u n d Peter Reichurd H . Phillips und D . M. Brown The Mutagenic Action of Hydroxylomine-J. Mammalian Nucleolytic Enzymes and Their Sierakowska Volume 8 Nucleic Acids-The

First Hundred Years-].

N.

Localization-Duoid

Shugar und

Mar-

Hafina

Daoidson

Nucleic k i d s and Protomine in Salmon Testes-Gordon H. Dixon and Michael Smith Experimental Approaches to the Determination of the Nucleotide Sequences of large Oligonucleotides and Small Nucleic Acids-Robert W.Holleg Alterations of DNA BOK Composition in Bacteria-G. F. Cause Chemistry of Guanine and Its Biologically Significant Derivatives-Robert Shupiro


331

Bacteriophage 4x1 74 and Related Viruses-Robert L. Sinsheimer W. Rushizky a n d The Preparation and Characterization of Large Oligonucleotides-George H e r b e r t A. Sober Purine N-Oxides and Cancer-George Bosworth Brown The Photochemistry, Photobialogy, and Repair of Polynucleotides-R. B . Setlow What Really I s DNA? Remarks on the Changing Aspects of a Scientific Concept-Erwin Churgufl Recent Nucleic Acid Research in China-Tien-Hsi Cheng a n d Roy H . D o i

Volume 9 The Role of Conformation in Chemicol Mutagenasis-B. Singer a n d H . Fraenkel-Conrat Polaragraphic Techniques in Nucleic Acid Research4. Puletek RNA Polymerase and the Control of RNA Synthesis-John Y.Richardson Radiation-Induced Alterations in the Structure af Deoxyribonucleic Acid and Their Biological Consequences-D. T.Kunazir Optical Rotatory Dispersion and Circular Dichroism of Nucleic Acids-]en Tsi Yung a n d Tatsuya Samejima The Specificity of Molecular Hybridization in Relation to Studies on Higher Organisms-P. M . B . Wulker Quantum-Mechanical Investigations of the Electronic Structure of Nucleic Acids and Their Constituents-Bernurd Pullman a n d Alberte Pullman The Chemical Modification of Nucleic Acids-N. K . Kochetkoo a n d E . 1. Btidowsky Volume 10 Induced Activation of Amino Acid Activating Enzymes by Amino Acids and tRNA-Ahn H. M e h l e r Transfer RNA and Cell Differentiation-NohorU Sueoka a n d Tamiko Kano-Sueoku N6-(A*-lsapentenyl)adenosine: Chemical Reactions, Biosynthesis, Metabolism, and Significance to the Structure and Function of tRNA-Ross H. Hull Nucleotide Biosynthesis from Preformed Purines in Mammalian Cells: Regulatory Mechanisms and Biological Significance-A. \V. Murruy, D a p h n e C . E l l i o t t , and M . R . Atkinson Ribosome Specificity of Protein Synthesis in Vitro-Orio Ciferri und Bruno Parisi Synthetic Nucleotide-peptides-zoe? A. Shobarooa The Crystal Structures of Purines, Pyrimidines and Their Intermolecular Complexes-Donuld Voet and Alexander R i c h Volume 1 1 The Induction of Interferon by Natural and Synthetic Polynucleotides-C/aretlce C d b y , Ir. Ribonucleic Acid Maturation in Animal Cells-R. H. B u r d o n Liporibonucleopratein as an Integral Part of Animal Cell Membranes-V. S. Shupot und S. Ya. Davidooa Uptake of Nonviral Nucleic Acids by Mammalian Cells-Pushpa M . Bhargaoa a n d G . Shun-

mugam The Relaxed Control Phenomenon-Ann M . Ryun a n d Ernest Borek Molecular Aspects of Genetic Recombination-Cedric I . Dauern E. Kennel/ Principles and Practices of Nucleic Acid Hybridization-Duoid Recent Studies Concerning the Coding Mechanism-Thomas H . lukes a n d Lih Gatfin The Ribosomal RNA Cistrons-M. L. Birnstiel. M . Chipchase, a n d J . Speirs Three-Dimensional Structure of tRNA-Friedrich Cramer Becker a n d l e r a r d H u r w i t z Curront Thoughts an the Replication of DNA-Andrew Reaction of Aminoacyl-tRNA Synthetases with Heterologous tRNA's-K. Bruce Jacobson On the Recognition of tRNA by Its AminoacylrtRNA Ligase-Robert W. Chambers

332


Volume 12 Ultroviolet Photochemistry as a Probe of Polyribonucleotide Conformation-A. J . Lomant and Jacques R . Fresco Same Recent Developments in DNA Enzymology-Mehran Coulian Minor Components in Transfer RNA: Their Choroctwirotion, Location, and Function-susumu Nishimrrra The Machoniun of Aminoacylotion of Transfer RNA-Robert E . Loftfield Regulotion of RNA Synthesis-Ekkehard K . F. Eautz The Poly(dA-dT) of Crob-M. Laskowski, Sr. The Chemicol Synthesis and the Biochemicol Properties of Peptidyl-tRNA-Yehuda Nathan

Lopidot and

de Croot

Volume 13 Reactions of Nucleic Acids and Nucleoprohins with Formaldehyde-M. Yo. Feldman Synthesis and Functions of the -C-C-A Terminus of Transfer RNA-hfut~ay P. Deutscher Mommalion RNA Polymeroses-Samson T. Jacob Polybdonorim diphosphote ribox)-Takashi Sugimura T k Stereochemistry of Actinomycin Binding to DNA and Its Implications in Moleculor BiologyHenry M. Sobell Resistance Factors ond Their Ecologicol Importance to Bocterio and to h n - M . H. Richmond Lysogenic Induction-Ernest Eorek and Anrr Ryan Recognition in NucleK Acids ond the Anticodon Families-Jacques Ninio Tronslation and Tronrcription of the Tryptophan Operon-Fumio Imarnoto lymphoid Cell RNA's and Immunity-A. Arthur Cottlieb Volume 14 DNA Modification ond Restriction-Werner Arber Mechanism of Bacteriol Transformation and Tronsfection-Nihd K . Notuni a n d June K . Setlow L. cefter DNA Polymeroxs II ond 111 of Eschnichia coli-hfa/colrn The Primary Structure of DNA-Kenneth M u r r u y und Robert \V. Old RNA-Directed DNA Polymerase-Properties and Functions in Oncogenic RNA Viruses and CellsMaurice Green a n d Gray F. Cerurd Volume 15 informotion Transfer in Cells Infected by RNA Tumor Viruses ond Extension to Humon NeoplosioD. Cillespie, W. C . Saxinger, und R . C. Gullo Mommolion DNA Polymeroses-F. J. E o l l u m Eukaryotic RNA Polymerases and the Factors That Control Them-E. B . Eiswas, A. Gunguly, a n d

D. Dus Structural and Energetic Consequences of Noncomplementory B o x Oppositions in Nucleic Acid Helices-A. J . Lomant und Jacques R . Fresco The Chemicol Effects of Nucleic Acid Alkylation and Their Relation to Mutagonesis ond Carcinogenesis-E. Singer €ffects of the Antibiotics Netropsin ond Distamycin A on the Structure and Function of Nucleic Acids-Christoph Zimmer Volume 16 Initiation of Enzymic Synthesis of Deoxyribonucleic Acid by Ribonucleic Acid Primers-Erwin Chargaff Transcription ond Processing of Transfer RNA Precursors-John D.Smith

333

CONTENTS OF PREVIOUS VOLUhIES Bisulfite Modification of Nucleic Acids and Their Constituents-Hikoya Hayatsu I . Budowsky The Mechanism of the Mutagenic Action of Hydroxylamines-E. Diethyl Pyrocarbonate in Nucleic Acid Research-L. Ehrenberg, 1. Fedorcsbk, and

F . Sol-

YmosY Volume 17 The Enzymic Mechanism of Guanosine 5'. 3'-Polyphosphate Synthesis-Fritz

Lipmann a n d Jose

SY Effects of Polyamines on the Structure and Reactivity of tRNA-Ted T. Sakai and Seymour S. Cohen Information Transfer and Sperm Uptake by Mammalian Somatic Cells-Aaron Bendich, Ellen Borenfreund, Steven S. Witkins, Delia Beju, u n d Paul J. Higgins Studies on the Ribosome and Its Components-Pnina Spitnik-Efson a n d D a u i d Elson Classical and Postclassical Modes of Regulation of the Synthesis of Degradotive Bacterial Enzymes- Boris Magasanik Characteristics and Significance of the Polyadenylate Sequence in Mammalian Messenger RNAGeorge Brawerman Polyadenylate Polymerases-Mury Edmonds a n d Munj A n n \$'inters Three-Dimensional Structure of Transfer RNA-Sung-Hou K i m Insights into Protein Biosynthesis and Ribosome Function through Inhibitors-Sidney Pestku Interaction with Nucleic Acids of Carcinogenic and Mutagenic N-Nitroso Compounds-W. Lijinsky Biochemistty and Physiology of Bocterial Ribonuclease-Alok

K . D a t t u und Sufi1 K . N i y o g i

Volume 18 The Ribosome of Escherichio coli-R. Brintucomhe, K . H. Nierhuus, R . A. Garrett u n d H . G. Wittmann Structure and Function of 5 S and 5.8 S RNA-Vofker A. Erdmunn High-Resolution Nuclear Magnetic Resonance Investigations of the Structure of tRNA in SolutionD a o i d R . Kearns Premelting Changes in DNA Conformation-E. PuleEek Quantum-Mechanical Studies on the Conformation of Nucleic Acids and Their ConstituentsBernard Pullman und Anil Sorun Volume 19: Symposium on mRNA: The Relation of Structure a n d Function 1. The 5'-Terminal Sequence (''Cap'') of mRNAs Caps in Eukaryotic mRNAs: Mechanism of Formation of Reovirus mRNA 5'-Tenninal m'GpppGmC-Y. Furuichi, S. Muthukrishnun. J. Tomust a n d A. Shutkin Nucleotide Methylation Patterns i n Eukaryotic mRNA-Fritz M. Rottman, Ronald C . D e srosiers a n d Karen Friderici Structural and Functional Studies on the "5'-Cap": A Survey Method of mRNA-Harris Busch, Friedrich Hirsch, Kaushul Kumur C u p t a , Munchunuhulli Roo, Willium Spohn und Benjamin C. Wu Modification of the 5'-Terminals of mRNAs by Viral and Cellulor Enzymes-Bernard Moss, Scott A. Martin, Marcio J . Ensinger, Robert F. Boone und Cha-Mer Wei

I.

Blocked and Unblocked 5' Termini in Vesicular Stomatitis Virus Product RNA in Vitro: Their Possible Role in mRNA Biosynthesis-Richard J . Colonno, Cordon Abraham and Amiya K . Banerjee The Genome of Poliovirur Is an Exceptional Eukaryotic mRNA--f'uon Fon Lee, Akio Nomoto and Eckard Wimmer

334


II. Sequences ond Conformations of mRNAs Transcribed Oligonucleotide Sequences in Hela Cell hnRNA and mRNA-Mary Edmonds, Hiroshi Nakazato, E . L. Korwek a n d S. Venkatesan Harris a n d Leon Dure Polyadenylylation of Stored mRNA in Cowon Seed Germination-Barry

111 mRNAs Containing and lacking PolyjA) Function as Separate and Distinct Classes during Embryonic Development-Martin Nemer u n d Saul Surrey Sequence Analyris of Eukaryotic mRNA-N. J. Proudfoot, C . C . Cheng a n d C. G. Brownlee The Structure and Function of Protamine mRNA from Developing Trout Testis-P. L. Davies, C . H. Dixon, L. N. Femer, L. Gedamu a n d K . l a t r o u The Primary Structure of Regions of SV40 DNA Encoding the Ends of mRNA-Kiranur N. Subramanian, Prabhat K . Ghoshi, Ravi Dhar, Bayar Thimmuppaya, Sayeeda €3. Zain, Julian Pan a n d Sherman M . Weissman Nucleotide Sequence Analysis of Coding and Noncading Regions of Human P-Globin mRNACharles A. Marotta, Bernard G . Forget, Michuel CohenlSolal a n d Sherman M . Weissman Determination of Globin mRNA Sequences and Their Insertion into Bacterial Plasmidr-Winston Sulser, Jeff Browne, Pat Ciarke, H o w a r d Heindell, Russell Higuchi, Gary Paddock, John Roberts, Gary Studnicku and Paul Zakar The Chromosomal Arrangement of Coding Sequences i n a Family of Repeated Genes-G. M . Rubin, D.J. Finnegan a n d D. S. Hogness Mutation Roter in Globin Genes: The Genetic load and Haldane’s Dilemma-Winston Salser a n d Judith Strommer lsiiacson Heterogeneity of the 3’ Portion of Sequences Related to Immunoglobulin K-Chain mRNA-Ursuh Storb Structural Studies on Intact and Deadenylylated Robbit Globin mRNA-John N. Vournakis, Marcia S. Flashner, MuryAnn Katopes, G a r y A. Kitos, Nikos C . Vamvukopoulos, Matthew S. Sell u n d Regina M . Wurst Molecular Weight Distribution of RNA Fractionated on Aqueous and 70% Formamide Sucrose Gradients-Helga Boedtker u n d Hans Lehruch 111. Processing of mRNAs Bacteriophages 17 and 13 as Model Systems for RNA Synthesis and Processing-J. J. Dunn, C. U‘. Anderson, J. F. Atkins, D. C . Burtelt a n d K! C. Cruckett The Relationship between hnRNA and mRNA-Robert P. Perry, Enzo Bard, B. D a v i d Homes, D a w n E . Kelley u n d U e l i Schihler A Comparison of Nuclear and Cytoplasmic Viral RNAs Synthesized Early in Productive Infection with Adenovirus 2-Heschel J. Raskas and Elizabeth A. Craig Biogenesis of Silk Fibrain mRNA: An Example of Very Rapid Processing?-Pad M. L i z a r d i Visualization of the Silk Fibroin Transcription Unit and Nascent Silk Fibroin Molecules on Polyribosomer of Bombyx mori-Steven L. .%Knight, Nelda L. Sullivan a n d Oscar L. Miller, Jr. Production and Fate of Balbiani Ring Products-B. Daneholt, S. T. Case, J. Hyde, L. Nelson a n d L. Wieslander Distribution of hnRNA and mRNA Sequences in Nuclear Ribonucleoprotein Complexes-Alan J. Kinniburgh, Peter B. Billings, Thomas J. Quinlan a n d Terence E . M a r t i n IV. Chromatin Structure and Template Activity The Structure of Specific Genes in Chromatin-Richard Axe/ The Structure of DNA in Native Chromatin as Determined by Ethidium Bromide Binding-J. Paoletti, B. B . Magee a n d P. T. Magee Cellular Skeletons and RNA Messages-Ronald Herman, Gary Zieve, Jeffrey Williams, Robert Lenk a n d Sheldon Penman

335


The Mechanism of Steroid-Hormone Regulation of Transcription of Specik Eukoryotic Genes-Bert W. O’Malley and Anthony R . Means Nonhistone Chromosomal Proteins and Histone Gene Transcription-Gary Stein, Janet Stein, L e w i s Kleinsmith, W i l l i a m Park, Robert Jansing and Judith Thomson Selective Transcription of DNA Medioted by Nonhistone Proteins-Tung Y. Wang, Ninu C. Kostraba and Ruth S. N e w m a n V. Control of Translation Kaesberg Structure and Function of the RNAs of Brome Mosaic Virus-Paul Effect of 5’-Terminal Structures on the Binding of Ribopolymers to Eukoryotic Ribosomes-S. M u t h u k r i s h n a n , Y. F u r u i c h i , G. W. B o t h and A. J. Shatkin Translational Control in Embryonic Muscle-Stuart M . H e y w o o d and D o r i s S. Kennedy Protein and mRNA Synthesis in Cultured Muscle Cells-R. G. Whalen, M . E . B u c k i n g h a m a n d F. G r o s VI. Summary mRNA Structure and Function-James E. D a r n e l l

Volume 20 Correlation of Biological Activities with Structural Features of Transfer RNA-B. F. C . C l a r k Bleomycin, an Antibiotic That Removes Thymine from Double-Stranded DNA--\.Vemer E . G. MU[ler a n d Rudolf K . Z a h n Mammalian Nucleolytic Enzymes-Halinu Sierukotcska and D a v i d Shugar Transfer RNA in RNA Tumor Viruses-LarrrJ C. Waters a n d B e t h C. Mullin Integration versus Degradation of Exogenous DNA in Plants: An Open Question4‘aul F. Lurquin lnitiotion Mechonisms of Protein Synthesis-Murianne Grunberg-Munago a n d Francois G r o s

Volume 21 Informoromas and Their Protein Components: The Present Stote of Knowledge-A. razhensky und A . S. S p i r i n Energetics of the Ribosome-A. S . S p i r i n Mechonisms in Polypeptide Chain Elongation on Ribosomes-Engin Bermek

A. Preob-

Synthetic Oligodeoxynucleotides for Analysis of DNA Structure and Function-Ray WU, Chander P. Bahl, a n d Saran A. Narang The Transfer RNAs of Eukaryotic Organelles-W’. E d g a r B u m e t t , S . D.Schtcurtzbach, a n d L.

1. Hecker Regulation of the Biosynthesis of Aminoacid:tRNA Ligoses and of tRNA-suson D i e t e r Sol1

D . Morgan and

Volume 22 The -C-C-A End of tRNA and Its Role in Protein Biosynthesis-Mathias S p r i n d a n d F r i e d r i c h Cramer The Mechanism of Action of Antitumor Platinum Compounds-J. J. Roberts a n d A. J. Thomson DNA Glycosyloses, Endonucleases for Apurinic/Apyrimidinic Sites, and Base Excision-RepoirThomas L i n d a h l Naturally Occurring Nucleosids ond Nucleotide Antibiotics--Robert J. Suhadolnik Genetically Controlled Voriation in the Shapes of Enzymes-George Johnson Transcription Units for mRNA Production in Eukaryotic Cells and Their DNA Viruses-James E. Darnell, Jr.

336


Volume 23 Tho Poptidyhmnrfomw Contor of Riboromosdlexander A. Krayeusky and M a r l t ~K. Kukkanooa Patkrnr of Nucloic Acid Synthrb in Physmnn polpphah-GeofTrey Tuff~ock Biahomical Efhch of fho Modification of Nuclek k i d s by Cortoin Polmclic Aromatic carcinqmnr4ezider Gmnberger and 1. Bernard Weinstein Participationof Modihd Nucl.oridor in Translation and 1rancription-B. Stngw and M. Kruger Tho Accuracy of Tranrlation-hfichae~ Yarns Structuro. Function, and Evolution of Tmnrfor RNAr (with Appondix Giving C0mpl.t. Squoncor of 178 tRNAr)-Ram P. Singhal and Pamela A. M . Fallis Volume 24 Structuro of Transcribing Chromatin-Diane Mathis, Pierre Oudet, and Pierre Chambon ligand-lnducod Confamotional Changor in Ribonucloic Acids-Hans Cilntw Gassen Roplicatiw DNA Polymonxr and Mochanhmr at a Roplication Fork-Robert K . Fujimura and Shishir K . Das Antibodios Spocific for Modifiod Nuclooridor: An Immunochomical Approach for tho Idotion and Charachrirationof Nucloic kids-Theodore W. Munns and M . K a t h q n Liszewskt DNA Structuro and Gono Roplication-R. D. Wells, T. C.Goodman, W. Hillen, C.T. Horn. R. D. Klein, J. E . Larson, U. R. Milller, S. K . Neuendorf, N. Panoyotatos, and S. M . Stirdioant Volume 25 Splicing of Viral mRNAs-Yosef Aloni DNA kthylation and I t s Porriblo Biolqical Rolor-Aharm Razin and Joseph Friedman Mochonirmrof DNA Roplicationand Mutogonosir in Uhraviolot-lrradiatod Bachria and Mammalian Colls-Jennifer D. H a l l and Daofd W. Mount Tho Rogulation of Initiation of Mammalian Protoin Synth.rir-RosemaryJagus, W. French Anderson, and Brian Safer Structuro, Roplication, and Transcription of tho SV40 Gonomo-Cokul C. DUS and Salil K . Niyogi Volume 26: Symposium on DNA: Multiprotein Interactions Introduction: DNA- Mukiprotoin Intoractions i n Transcription, Roplication, and Ropair-R. K. Fujimura Roplicatiw DNA Polymorau and Its Complox: Summary-Daoid K w ~ Enrymo Studios of 4x174 DNA Roplication-Ken-ichi Arai, Naoko AraiJoseph ShlomaiJoan Kobori, Laurien Polder, Robert Low. Ulrich Hilbscher. L.eRoy Bertsch, and Arthur Kornberg of Erch.richia d: Structure and Function of tho onXontaining Tho DNA Roplication Origin (M’) DNA Fragmont-Yukinori Hirota, Masao Yamadu, Akiko Hishimura. Atsuhiro Oka, Kazunori Sugimoto, Kiyozo Asada, and Mitsuru Takanami Roplication of h o a r Duplox DNA in Viho with Bachriophogo 15 DNA Polymoraso-R. K. Fujimura, S. K . Das. D. P. Allison, and B. C. Roop Muhanirnrr of Catalysis of Human DNA Polynwrasosaand B-Dautd K m , Paul A. Fisher, and Teresa S.-F. Wang Structural and Functional Promior of Calf Thymus DNA Polymorau 6-Marietta Y. w. Tsang Lee, Cheng-Keat Tan. Kathleen M. Downey, and Antero G . So kchanhmr of Transcription: Summary-R. K. Fujimura


337

Regulotory Circuits of Bacteriophoge Lambda-s. L. Adhya, S. Garges, and D. F. Ward Chromatin Tronwription and Replication: Summory-Ronald L. Seale Site of Histone Assembly-Ronald L. Seale Chromatin Replication in Tetrohymeno pyrifomis-A. T.Annunziato and C. L. F. Woodcock Role of Chromatin Structure, Histone Acetylotion, and the Primary Sequence of DNA in the Expression of SV40 ond Polyomo in Normal or Terotocorcinomo C e l l s 4 . Moyne, M . Katinka, Saragosti, A. Chestier, and M. Yaniv Control of Transcription in Eukoryotes: Summary-William I. Rutter Repair Replication Schemes in Bacteria and Humon CellS-Philip c. Hanawalt, Priscilla K . Cooper, and Charles Allen Smith Recent Developments in the Enzymology of Excision Repair of DNA-EWOZ c. Friedberg, Currie T. M. Anderson, Thomas Bonura, Richard Cone, Eric H. Radany, and Richard 1. Reynolds Multiprotein lnteroction in Strand Cleovoge of DNA Damaged by UV and Chemicals-Erling Seeberg In VitFo Packaging of Damaged Bacteriophage 77 DNA-Warren E. Masker, Nancy B. Kuemmerle, and Lori A. Dodson The Inducible Repair of Alkyloted DNA-]ohn Cairns, Peter Robins, Barbara Sedgwick, and Phillipa Talmud Functions Induced by Damaged DNA: Summary-Edyn M. Witkin Inducible Error-Prone Repair and Induction of Prophoge Lombdo in Enherichio di-Raymond Deuoret DNA and Nucleoside Triphosphate Binding Properties of recA Protein from Escherichio coli-K. McEntee, G . M. Weinstock, and 1. R . Lehman Molecular Mechonism for the induction of "SOS" Functions-Michi0 Oishi, Robert M. Zrbe, and Lee M. E. Morin Induction and Enhanced Reoctivotion of Mommalion Viruses by Light-kfTy E . Bockstahler Comparative Induction Studies-Ernest C. Pollard, D. 1.Fluke, and Deno Kazanis Concluding Remarks-Ernest C . Pollard

s.

Volume 27 Poly(odenosine diphosphote ribose)-Pad Mandel, Hideo Okazaki, and Cloude Niederga ng The Regulatory Function of Poly(A) and Adjacent 3' Sequences in Transloted RNA-uriel Littauer and Hermona Soreq tRNA-like Structures in the Genomes of RNA Viruses-Anne-Lise Haenni, Sadhna ]oshi, and Francois Chapeoille Mechanism of lntarferon Action: Progress toword Its UnderstandingAanes c. Sen Thomas and Wlodzimierz Szer RNA-Helix-Oestobilizing Proteins--John 0. Nucleotide CycIoses-Luurence S. Bradham and Wai Yiu Cheung Cyclic Nucleotide Control of Protein Kinoses-R. K . Sharma

z.

Volume 28 The Structure of Ribosomal RNA ond Its Orgoniration Relative to Ribosomal Protein-Richard Brimacombe, Peter Maly, and Christian Zwieb Structure, Biosynthesis, and Function of Queuosine in Transfer RNA-sUSumU Nishimuru Queuine: An Addendum-Ram P. Singhul The Fidelity of Tronslotion-Abraham K . Abraham Structure and Functions of Ribosomal Protein S1 -Alap-Raman Subramanian

338


The Yeost Cell Cycle: Coordination of Growth ond Division Rates-SteUen S. McLuughlin

G. E l l i o t t a n d Caluin

Prokoryotic and Eukaryotic 5 S RNAs: Primary Sequences and Proposed Secondary StructuresRam P. Singhal andJoni K . Shaw Structure of Tronrfer RNAs: Listing of 150 Addition01 Sequences-Ram Roberts, a n d Vikram N. Vakharia

P.

Singhal,

Volume 29: Genetic Mechanisms in Carcinogenesis Introduction-W. K . Yang 1. Genetic Factors in Cancer Evolution of RNA Tumor Viruses Anolagy for Nonviral Carcinogenesis-Howard

M.

Ed& F.

Temin

G. Knudson, Jr. Model Hereditary Cancers of Man-Alfred Bacterial "Inserted Sequence" Elements and Their Influence on Genetic Stability and EvolutionWerner Arber Significance of Specific Chromosomol Tranrlocationr and Trisomies for the Genesis of Murine and Humon Tumors of the Lymphocyte- Plasmocyte Lineage-George

Klein (Summary prepared

b y W. W. Au) Short Communications Quantitation of One Aspect of Koryotype Instability Associated with Neoplastic Transformation in Chinese Hamster Cells-L. S. Cram, M. F. Bartholdi, F. A. Ray, G. L. Travis, H.Jett, a n d P. M . Kraemer Mechanism of Mutation at the Adenine PhosphoribosyltronsferaseLocus i n CHO cells-A. Simon

I.

and M . W. Taylor Development of a Tronsplantable Mouse Myeloid Leukemio Model System: A Preliminary Report-

].

W. W. Au, H.E. Luippold, a n d A. Otten II. Genetic Elements in Radiation and Chemical Carcinogenesis Molecular Studies of the Radiation Leukemia Virus (RadLV) and Related Retroviruses of C57BUKa Mice-R. A. Grymes, M. L. Scott, J. P. Kim, K . E. Fry, a n d Henry S. K a p h Endogenous Retrovirus and Radiation-Induced leukemia in the RFM Mouse-Raymond w. Tennant, L. R. Boone, P. A. Lalley, a n d W. K . Yang Genetic ond Probability Aspects of Cell Transformation by Chemical Carcinogens-Charles Heidelberger, 1. R. Landolph, R. E. K . Fournier, A. Fernandez, a n d A. R . Peterson Short Communications T. Snow, R. S. Foote, a n d Replication and Demethylation of O'-Methylguanine in DNA-E. S. Mitra Quantitative Assay of Low Level of Benzo[o]pyrenediol Epoxide Bound to DNA by Acid-Induced Liberation of Tetraols followed by Chromatography ond Fluorometric Detection-R. 0. Rahn,

J. M . Holland, a n d L. R. Shugart Tronsfer of Phorbol Ester Promotability by Transfection of DNA from Promotable into Nonpromotoble Cells-N. H. Colburn, C. B. Talmadge, and T.D.Gindhart Specificity of Interaction between Carcinogenic Polynuclear Aromatic Hydrocorbons and Nuclear Proteins: Widespread Occurrence of a Restricted Pattern of Histone Binding in Intact Cells-

].

M. C. MacLeod, C. Pelling, T. ]. Slaga, P. A. Noghrei-Nikbakht, B. K . Mansfield, a n d J . K . Selkirk 111. Mechanirm of Viral Cardnogenesis Role of Polyoma T Antigens in Malignant Cell Transformation-Walter Eckhart Avian Leukosis Viruses and Concer: Genetics of Insertional Mutogenesis-Harriet L. Robinson Short Communications U s e of a Viral Probe to Study Recombinational Exchonges in Mammalian Cells-G. Duigou a n d S. G. Zimmer

I.


339

Comporotive Tryptic Peptide Anolysis of P85 gag-mos of Mo-MuSV ti-110 and the P38-P23 mosReloted Products of Wild-Type Virus-E. C. M u r p h y , Jr. and R. B. ArZinghaus Genomic Complexity and Molecular Cloning of a Proviral DNA Specific for a Feral Rat Endogenous C-Type Virus, Originated from o 3-Methylcholanthrene-Induced Fibrosarcoma-S. S. Yang a n d R. M o d u l i An SV40 Mammalian lnductest for Putative Carcinogens-S. P. Moore a n d T. P. C o o h i l l Characterization of a Cellular Protein That Promotes SV40 Infection in Human Cells-V. F. Righth a n d and C. Bagshaw IV. Regulatory Functions and Genetic Control Concer as o Problem in lntercellulor Communication: Regulation by Growth-Inhibiting Factors (Chalones)-Van R . Potter Restriction of Murine Leukemio Viruses by Fv-1: A Model for Studying Host Genetic Control of Retroviral Gene Movement and Leukemogenesis-Wen K . Yang, L. R . Boone, R . W. Tennant,

].

a n d A. B r o w n Expression of a Viral Oncogene under Control of the Mouse Mammary Tumor Virus Promoter: A New System for the Study of Glucocorticoid Regulation-Gordon L. H a g e r Short Communications Variation of Long-Terminal-Repeat Size in Molecular Clones of the BALB/c Endogenous Ecotropic R . Boone, F. E. Myer, D.M . Yang, /. 0.Kiggans, C. Koh, Murine Leukemia Virus-,!,. R . W. Tennant, and W. K . Yang Role of Nu-Methylodenotine in Expression of Rous Sarcoma Virus RNA: Anolyses Utilizing Immunoglobulin Specific for N6-Methylodenosine-R. J . Resnick, D. Noreen, T. W. Munns, and M. L. Perdue V. Growth and Differentiation in Neoplastic Transformation Role of Tyrosine Phasphorylation in Malignant Tronsformation by Viruses and in Cellular Growth Control-Tony H u n t e r a n d J. A. Cooper Molecular Interaction of the sm Gene Product with Cellular Adhesion Plaques-Larry R. Rohrschneider, M. J. Rosok, and L. E. G e n t r y The Receptor for Epidermal Growth Factor Functions as a Tyrosyl-Specific Kinass-Stanley Cohen Short Communications An in Vitro Model of Epithelial Cell Neoplastic Progression: Growth Properties and Polypeptide Composition of Cell Lines-K. D.Somers, M. M. Murphey, a n d D.G. Stark Chromosomal Protein Antigens Formed in Experimental Hepatocarcinogenesis by Aro Dyes-W. N. Schmidt, B. J. Gronert, R . C. Briggs, D.L. Page, a n d L. S. H n i l i c a Cross-linking of Nuclear Antigens to DNA in HeLa Cells-Z. M. Banjar, R . C. Briggs, L. S. HniZica, J. Stein, a n d G. Stein Expression of Mutated Actin Gene Associated with Malignant Transformation-H. Hamada, J. Leauitt, a n d T. Kakunaga Phenotypic Changer in Epithelial Cell Populotion Undergoing Neoplastic Transformation in Vitro-

G. R. Braslawsky, S. J. Kennel, and P. Nettesheim VI. The Seorch for Human Transforming Genes

F. Parada, C. Shih, M. Murray, and The Oncogene of a Human Bladder Carcinoma-L. R o b e r t A. Weinberg Tronsforming Genes of Neoplasms-Geoffrey M. Cooper Short Communications A Sequence Homologous to Rous Sarcoma Virus v-SK Is on Human Chromosome 20-A. Y. Sakaguchi, S. L. Naylor, a n d T. B. Shows Variable Differentiative Response of 6-Thioguonine-Resistant HMO Sublines: Possible Relationship E . GaZZagher, A. C. Ferrari, A. W. Zulich, a n d to Double-Minute Chromosomes-R. J. R . Testa

340


Volume 30 RNA Procesring in o Unicellular Microorgonism: Implicationsfor Eukaryotic Cells-David Apirion Neorest-Neighbor Effects in the Structure ond Function of Nucleic Acids-E. Bubienko, P. C m z , J . F . Thomason, and P. N. Borer The Elongotion Foctor EF-TUand L Two EncodingGenes-L. Bosch, B . Kraal, P. H. Van der Meide, F . J. Duisterwinkel, and J . M . Van Noort Smoll Nuclear RNAs and RNA Procersing-Ram Reddy and Harris Busch Ribosome Evolution: The Struchtml Bases of Protein Synthesis in Archoebocterio, Eubacteria, and Eukoryotes-James A. Lake Analysis of the Expression of Genes Encoding Animal mRNA by in Kftu Techniques-James L. Manley Synthesis, Processing, and Gene Structure of Vosopressin and Oxytocin-Dietmar Richter

Volume 31 Irnmunoozra)r of Corcinogen-Modified DNA-Paul

T . Strickland and John M . Boyle

On the Biologicol Significonce of Modified Nucleosides in tRNA-Helga

Kersten The Orgonization ond Tmnscription of Eukoryotic Ribosomal RNA Genes-Radha K. Man&[ Strudure, Function, and Evolution of 5-5 Riboromol RNh-Nicholas Delihas,Janet Andersen, and Rani P. Singhal Optimization of Tmnslotionol Accumcy-C. C . Kurlnnd and Mdns Ehrenberg Molecular Aspects of Development in the Brine Shrimp Artemia-Albert J. Wahha and Charles L. Woodley Translotional Control Involving o Novel Cyioplasmic RNA and Ribonudeoprotein-Satyapriyo Sarkar The Hypoxonthine PhosphoribosyltronsfemseGene: A Model for the Study of Mutation in Mommolian Cells-A. Craig Chinault and C . Thomas Caskey The Molecular Genetics of Humon Hemoglobin-Francis S . Collins and Shernian M. Weissman

Progress in Nucleic Acid Research and Molecular Biology, Volume 32

Progress in Nucleic Acid Research and Molecular Biology, Volume 50

Progress in Nucleic Acid Research and Molecular Biology, Volume 6

Progress in Nucleic Acid Research and Molecular Biology, Volume 82

Progress in Nucleic Acid Research and Molecular Biology, Volume 65

Progress in Nucleic Acid Research and Molecular Biology, Volume 31

Progress in Nucleic Acid Research and Molecular Biology, Volume 57

Progress in Nucleic Acid Research and Molecular Biology, Volume 58

Progress in Nucleic Acid Research and Molecular Biology, Volume 40

Progress in Nucleic Acid Research and Molecular Biology, Volume 56

Progress in Nucleic Acid Research and Molecular Biology, Volume 77

Progress in Nucleic Acid Research and Molecular Biology, Volume 51

Progress in Nucleic Acid Research and Molecular Biology, Volume 61

Progress in Nucleic Acid Research and Molecular Biology, Volume 55

Progress in Nucleic Acid Research and Molecular Biology, Volume 70

Progress in Nucleic Acid Research and Molecular Biology, Volume 44

Progress in Nucleic Acid Research and Molecular Biology, Volume 47

Progress in Nucleic Acid Research and Molecular Biology, Volume 67

Progress in Nucleic Acid Research and Molecular Biology, Volume 17

Progress in Nucleic Acid Research and Molecular Biology, Volume 41

Progress in Nucleic Acid Research and Molecular Biology, Volume 19

Progress in Nucleic Acid Research and Molecular Biology, Volume 69

Progress in Nucleic Acid Research and Molecular Biology, Volume 45

Progress in Nucleic Acid Research and Molecular Biology, Volume 72

Progress in Nucleic Acid Research and Molecular Biology, Volume 52

Progress in Nucleic Acid Research and Molecular Biology, Volume 51

Progress in Nucleic Acid Research and Molecular Biology, Volume 28

Progress in Nucleic Acid Research and Molecular Biology, Volume 54

Progress in Nucleic Acid Research and Molecular Biology, Volume 49

Progress in Nucleic Acid Research and Molecular Biology, Volume 38

Progress in Nucleic Acid Research and Molecular Biology, Volume 46

Progress in Nucleic Acid Research and Molecular Biology, Volume 32

Progress in Nucleic Acid Research and Molecular Biology, Volume 50

Progress in Nucleic Acid Research and Molecular Biology, Volume 6

Progress in Nucleic Acid Research and Molecular Biology, Volume 82

Progress in Nucleic Acid Research and Molecular Biology, Volume 65

Progress in Nucleic Acid Research and Molecular Biology, Volume 31

Progress in Nucleic Acid Research and Molecular Biology, Volume 57

Progress in Nucleic Acid Research and Molecular Biology, Volume 58

Progress in Nucleic Acid Research and Molecular Biology, Volume 40

Progress in Nucleic Acid Research and Molecular Biology, Volume 56

Progress in Nucleic Acid Research and Molecular Biology, Volume 77

Progress in Nucleic Acid Research and Molecular Biology, Volume 51

Progress in Nucleic Acid Research and Molecular Biology, Volume 61

Progress in Nucleic Acid Research and Molecular Biology, Volume 55

Progress in Nucleic Acid Research and Molecular Biology, Volume 70

Progress in Nucleic Acid Research and Molecular Biology, Volume 44

Progress in Nucleic Acid Research and Molecular Biology, Volume 47

Progress in Nucleic Acid Research and Molecular Biology, Volume 67

Progress in Nucleic Acid Research and Molecular Biology, Volume 17

Progress in Nucleic Acid Research and Molecular Biology, Volume 41

Progress in Nucleic Acid Research and Molecular Biology, Volume 19

Progress in Nucleic Acid Research and Molecular Biology, Volume 69

Progress in Nucleic Acid Research and Molecular Biology, Volume 45

Progress in Nucleic Acid Research and Molecular Biology, Volume 72

Progress in Nucleic Acid Research and Molecular Biology, Volume 52

Progress in Nucleic Acid Research and Molecular Biology, Volume 51

Progress in Nucleic Acid Research and Molecular Biology, Volume 28

Progress in Nucleic Acid Research and Molecular Biology, Volume 54

Progress in Nucleic Acid Research and Molecular Biology, Volume 49

Progress in Nucleic Acid Research and Molecular Biology, Volume 38

Progress in Nucleic Acid Research and Molecular Biology, Volume 46

Recommend Documents