Clinical Biochemistry of Domestic Animals, Fifth Edition

Contributors

Delmar R. Finco (441) Department of Physiology and

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

Pharmacology,College of Veterinary Medicine,Uni­ versity of Georgia,Athens,Georgia 30602

Cleta Sue Bailey (785) Department of Surgical and

Radiological Sciences, School of Veterinary Medi­ cine,University of California,Davis,Davis,Califor­ nia 95616

Mats Forsberg (589) Department of Clinical Chemistry,

Duane F. Brobst (353) Department of Veterinary Clini­

Laurel J. Gershwin (139) Department of Pathology,

College of Veterinary Medicine,Swedish University of Agricultural Sciences, Uppsala,Sweden

cal Sciences,College of Veterinary Medicine,Wash­ ington State University,Pullman,Washington 99164

Microbiology, and Immunology, School of Veteri­ nary Medicine, University of California, Davis, Davis, California 95616

Michael L. Bruss (83, 885) Department of Anatomy,

Physiology, and Cell Biology, School of Veterinary Medicine,University of California,Davis,Davis,Cal­ ifornia 95616

Urs Giger (741) Medical Genetics,Veterinary Hospital,

University of Pennsylvania, Philadelphia, Pennsyl­ vania 19104

Charles C. Capen (619) Department of Veterinary Bio­

John W. Harvey (157,885) Department of Physiologi­

sciences, The Ohio State University, Columbus, Ohio 43210

cal Sciences,College of Veterinary Medicine,Univer­ sity of Florida, Gainesville,Florida 32610

George H. Cardinet III (407) Department of Veterinary

Mark Haskins (741) Pathology and Medical Genetics,

Anatomy, Physiology, and Cell Biology, School of Veterinary Medicine,University of California,Davis, Davis,California 95616

School of Veterinary Medicine, University of Penn­ sylvania,Philadelphia,Pennsylvania 19104 Jens G. Hauge (21) Department ofBiochemistry,Physi­

Gary P. Carlson (485) Department of Medicine and

ology,and Nutrition,Norwegian College of Veteri­

Epidemiology, School of Veterinary Medicine, Uni­ versity of California, Davis,Davis, California 95616

nary Medicine,N-0033 Oslo,Norway Walter E. Hoffmann (303) Department of Veterinary

Stan W. Casteel (829) Veterinary Medical Diagnostic

Pathobiology,College of Veterinary Medicine,Uni­

Laboratory,College of Veterinary Medicine,Univer­ sity of Missouri,Columbia,Missouri 65205

versity of Illinois,Urbana,Illinois 61801 William E. Hornbuckle (367) Department of Clinical

W. Jean Dodds (241) Hemopet,Santa Monica,Califor­

Sciences, College of Veterinary Medicine, Cornell

nia 90403

University,Ithaca,New York 14853

Lars-Erik Edqvist (589) Department of Clinical Chem­

Mahendra B. Kabbur (285) Department of Pathology,

istry, National Veterinary Institute, 5-75007 Upp­ sala,Sweden

Microbiology, and Immunology, School of Veteri­ nary Medicine, University of California, Davis, Davis,California 95616

Thomas B. Farver (1) Department of Population Health

and Reproduction, School of Veterinary Medicine, University of California, Davis, Davis, California 95616

J. Jerry Kaneko (45,117,205,571,885) Department of

Pathology, Microbiology, and Immunology, School

xi

xii

Contributors

of Veterinary Medicine, University of California, Davis, Davis, California

John W. Kramer (303) Department of Clinical Medicine

and Surgery, College of Veterinary Medicine, Wash­ ington State University, Pullman, Washington 99164

(845) Ani Lytics, Inc., Gaithersburg, 20877

Walter F. Loeb

Maryland

J. T. Lumeij (857) Division of Avian and Exotic Animal

Medicine, Department of Clinical Sciences of Com­ panion

Animals,

Utrecht

University,

3584 CM

Utrecht, The Netherlands Bruce R. Madewell

Surgical and Radiological Sciences, University of Jan A. Mol

95616

(517, 553) Department of Clinical Sciences

of Companion Animals, Faculty of Veterinary Medi­ cine, Utrecht University, 3584 CM Utrecht, The Neth­ erlands James G. Morris (703) Department of Molecular Biosci­

ences, School of Veterinary Medicine, University of California, Davis, Davis, California

The Ohio State University, Columbus, Ohio Robert B. Rucker

95616

Ad Rijnberk (517,553) Department of Clinical Sciences

43210

(703) Department of Nutrition, Col­

lege of Agriculture and Environmental Sciences, and Department of Biological Chemistry, School of Medi­ cine, University of California, Davis, Davis, Califor­ nia

95616

Joseph E. Smith (223) Department of Diagnostic Medi­

cine/Pathology, Kansas Veterinary Medical Center, Kansas State University, Manhattan, Kansas

(761) Department of Veterinary

California, Davis, Davis, California

(619) Department of Veterinary Pa­

Thomas J. Rosol

thobiology, Department of Veterinary Biosciences,

95616

66506

(327, 367) Department of Clinical Sci­

Bud C. Tennant

ences, College of Veterinary Medicine, Cornell Uni­ versity, Ithaca, New York James R. Turk

14853

(829) Veterinary Medical Diagnostic

Laboratory, College of Veterinary Medicine, Univer­ sity of Missouri, Columbia, Missouri

65205

(785) Consolidated Veterinary Diag­ nostics, Inc., West Sacramento, California 95605

William Vemau

Joseph G. Zinkl

(285) Department of Pathology, Mi­

crobiology, and Immunology, School of Veterinary

of Companion Animals, Utrecht UniverSity, Utrecht,

Medicine, University of California, Davis, Davis,

The Netherlands

California

95616

Preface to the 5th Edition

The first edition of

throughout the text and the appendices. The appendi­ ces in this edition have also been expanded to include new analyte reference ranges for the variety of species with which this volume is concerned. We take this opportunity to acknowledge our grati­ tude to the many contributors who through their con­ scientiousness have eased the incredibly formidable task of assembling a multiauthored volume of this magnitude. All contributors have been extremely en­ thusiastic and cooperative with us as well as with Aca­ demic Press in their efforts to move production of this volume forward in a timely fashion. From the outset, Academic Press, ably represented by editor Chuck Crumly and his staff, has been extremely diligent, pro­ viding constant encouragement, guidance, and assis­ tance for which we are most appreciative. This fifth edition represents a new departure with increased editorship and a host of new and emerging subject areas in the field of clinical biochemistry of domestic animals. It recognizes and welcomes the changing nature of the field as new knowledge is un­ covered and, as a corollary, the need for the wider view that associate editors bring to this effort. All con­ tributors can justly point with pride to this fifth edition. Finally, we express our thanks to our families, who have given us incredible support in this undertaking and have persevered without complaint through our single-minded devotion to this effort.

Clinical Biochemistry of Domestic

Animals

was published in 1963; now, 34 years later, this fifth edition is appearing. We can all appreciate that change is a constant in all human endeavor and

that science in particular is changing at an ever more rapid pace. In recognition of this rapid progress, this fifth edition constitutes yet another major change from the previous edition. In recognition of the breadth, magnitude, and complexity of this change, two associ­ ate editors, Dr. John W. Harvey of the University of Florida at Gainesville and Dr. Michael 1. Bruss of the University of California at Davis, joined a team to guide the preparation of this fifth edition. Many new contributors joined this effort, major revisions of previ­ ous contributions were undertaken, and four new chapters were added: Tumor Markers, Lysosomal Stor­ age Diseases, Oinical Biochemistry in Toxicology, and Avian Oinical Biochemistry. It is also fitting that this volume be dedicated to the memories of Dr. Charles E. Cornelius, co-editor of the first two editions and contributor to all previous edi­ tions, and of Dr. George H. 5tabenfeldt, contributor to the third and fourth editions. Both were great personal friends and outstanding scientists, administrators, and teachers and are sorely missed. The acceptance of the 5ysteme International d'Unites (51 units) continues to be resisted for many reasons, although its use is expanding, albeit slowly, throughout the world. The concept of the SI unit is

J. Jerry Kaneko John W. Harvey Michael L. Bruss

retained in this fifth edition as it was in the fourth with the inclusion of both 51 and conventional units

xiii

C H A P T E R

1 Concepts of Normality in Clinical Biochemistry THOMAS

1. POPULATIONS AND THEIR DISTRIBUTIONS

II. REFERENCE INTERVAL DETERMINATION

AND USE

B.

1

dogs that are free of disease. Whether a given dog belongs to the population of healthy dogs depends on someone's ability to determine if the dog is or is not free of disease. Populations may be finite or infinite in size. A population can be described by quantifiable char­ acteristics frequently called observations or measures. If it were possible to record an observation for all mem­ bers in the population, one most likely would demon­ strate that not all members of the population have the same value for the given observation. This reflects the inherent variability in populations. For a given mea­ sure, the list of possible values that can be assumed with the corresponding frequency with which each value appears in the population relative to the total number of elements in the population is referred to as the distribution of the measure or observation in the population. Distributions can be displayed in tabular or graphical form or summarized in mathematical ex­ pressions. Distributions are classified as discrete distri­ butions or continuous distributions on the basis of val­ ues that the measure can assume. Measures with a continuous distribution can assume essentially an in­ finite number of values over some defined range of values, whereas those with a discrete distribution can assume only a relatively few values within a given range, such as only integer values. Each population distribution can be described by quantities known as parameters. One set of parameters of a population distribution provides information on the center of the distribution or value(s) of the measure that seem to be assumed by a preponderance of the elements in the population. The mean, median, and

2

A. The Gaussian Distribution 2 B. Evaluating Probabilities Using a Gaussian Distribution 2 C. Conventional Method for Determining Reference Intervals 3 D. Methods for Determining Reference Intervals for Analytes Not Having the Gaussian Distribution

5

E. Sensitivity and Specificity of a Decision Based on a Reference Interval 7 F. Predictive Value of a Decision Based on a Reference Interval 8 III. ACCURACY IN ANALYTE MEASUREMENTS

IV.

PRECISION IN ANALYTE MEASUREMENTS

V. INFERENCE FROM SAMPLES A. Simple Random Sampling B. Descriptive Statistics 10 C. Sampling Distributions

10 10

9 9

10

D. Constructing an Interval Estimate of the Population Mean, JL 1 1 E . Comparing the Mean Response of Two Populations 13 F. Comparing the Mean Response of

Three Populations 13 G. Analysis of Variance and Its Uses References

18

14

I. POPULATIONS AND THEIR DISTRIBUTIONS A population is a collection of individuals or items having something in common. For example, one could say that the population of healthy dogs consists of all

CLINICAL BIOCHEMISTRY OF DOMESTIC ANIMALS, FIFTH EDITION

FARVER

1

Copyright © 1997 by Academic Press All rights of reproduction in any fonn reserved.

2

Thomas B. Farver

mode are three members of the class of parameters describing the center of the distribution. Another class of parameters provides information on the spread of the distribution. The spread of the distribution has to do with whether most of the values assumed in the population are close to the center of the distribution or whether a wider range of values is assumed. The

Most analytes cannot take on negative values and so, strictly speaking, cannot have Gaussian distributions. However, the distribution of many analyte values is approximated well by the Gaussian distribution be­ cause virtually all of the values that can be assumed by the analyte are within 4 standard deviations of the

standard deviation, variance, and range are examples of parameters that provide information on the spread of the distribution. The shape of the distribution is very important. Some distributions are symmetric about their center, whereas other distributions are asymmet­ ric, being skewed (having a heavier tail) either to the

bution is Gaussian. Figure 1.2 gives an example. The

right or to the left.

II. REFERENCE INTERVAL DETERMINATION AND USE One task of clinicians is determining whether an animal that enters the clinic has blood and urine ana­ lyte values that are in the normal interval. The conven­ tional method of establishing normalcy for a particular analyte is based on the assumption that the distribution of the analyte in the population of normal animals is the "normal" or Gaussian distribution. To avoid confusion resulting from the use of a single word hav­ ing two different meanings, the "normal" distribution henceforth is referred to as the Gaussian distribution.

A. The Gaussian Distribution Understanding the conventional method for estab­ lishing normalcy requires an understanding of the properties of the Gaussian distribution. Theoretically, a Gaussian distribution is defined by the equation:

y

=

� 1

--

e-(x -/4)2/20'2,

where x is any value that a given measurement can assume, y is the relative frequency of x, IL is the center of the distribution, (T is the standard deviation of the distribution, 1T is the constant 3.1416, and e is the con­

stant 2.7183. Theoretically, x can take on any value from -00 to +00. Figure 1.1 gives an example of a Gaussian distribution and demonstrates that the distribution is symmetric around IL and is bell shaped. Figure 1.1

also shows that 68% of the distribution is accounted for by measurements of x that have a value within 1 standard deviation of the mean, and 95% of the distri­ bution includes those values of x that are within 2 standard deviations of the mean. Nearly all of the dis­ tribution (97.75%) is contained by the bound of 3 stan­ dard deviations of the mean.

mean and, for this range of values, the frequency distri­ figure, adapted from the printout of BMDP2D (detailed data description including frequencies; Engleman, 1992), gives an example of the distribution of glucose values given in Table 1.1 for a sample of 168 dogs from a presumably healthy population. Though not perfectly Gaussian, the distribution is reasonably well approximated by the Gaussian distribution. Support for this claim is that the distribution has the characteris­ tic bell shape and appears to be symmetric about the mean. Also, the mean [estimated to be 96.4 mg/dl (5. 34 mmol/liter)] of this distribution is nearly equal to the median [estimated to be 95.0 mg/dl (5.27 mmol/ liter)], which is characteristic of the Gaussian distribu­ tion. The estimates of the skewness and kurtosis coef­ ficients are close to zero, also characteristic of a Gauss­ ian distribution (Engleman, 1992; Remington and Schork, 1985; Snedecor and Cochran, 1989).

B. Evaluating Probabilities Using a Gaussian Distribution All Gaussian distributions can be standardized to the reference Gaussian distribution, which is called the standard Gaussian distribution. Standardization in general is accomplished by subtracting the center of the distribution from a given element in the distribu­ tion and dividing the result by the standard deviation of the distribution. The distribution of a standardized Gaussian distribution, that is, a Gaussian distribution that has its elements standardized in this form, has its center at zero and has a variance of unity. The elements of the standard Gaussian distribution are traditionally designated by the letter z so that it can be said that z is N(O,l). That all Gaussian distributions can be trans­ formed to the standard Gaussian distribution is conve­ nient in that just a single table is required to summarize the probability structure of the infinite number of Gaussian distributions. Table 1.2 provides an example of such a table and gives the percentiles of the standard Gaussian distribution.

Example 1 Suppose the underlying population of elements is N(4,16) and one element from this popula­ tion is selected. We want to find the probability that the selected element has a value less than 3.0 or greater than 6.1. In solving this problem, the relevant distribu­ tion is specified: x is N(4,16). The probability of observ-

3

Concepts of Normality in Clinical Biochemistry

J-L-2 (T

FIGURE 1.1

The Gaussian distribution.

ing X < 3.0 in the distribution of x is equivalent to the probability of observing z < (3.0 - 4)/4 = -.25 in the standard Gaussian distribution. Going to Table 1.2, Z = .25 is approximately the 60th percentile of the standard Gaussian distribution and by symmetry z = - .25 is approximately the 40th percentile. Thus, the probabil­ ity of observing a z value less than or equal to .25 is approximately 040. The probability of observing x > 6.1 is equivalent to the probability of observing z > (6.1 - 4)/(4) = +.525. Table 1.2 gives the probability of observing a z < .525 as approximately .70, so the proba­ bility of observing a z > .525 approximately equals 1 -

. 70 or .30. The desired probability of observing a sample observation less than 3.0 or greater than 6.1 is the sum of 040 and .30, which is .7 or 7 chances in 10.

C. Conventional Method for Determining Reference Intervals

-

Number of Distinct Values: Sample Size: 168 Estimate Mean Median Mode

96.4 95.0 94.0

The first step in establishing a normal interval by the conventional method involves determining the mean and standard deviation of the distribution of the analyte. This can be accomplished by taking a representative sample (using a sampling design that

H H H HH HHH H 95% Confidence Interval HHH HH Upper Lower HHHHHHH 98.7 94.2 HHHHHHHHHHHH HHHHHHHHHHHHHHH H L----------------------U 5 Each ,-, above =

60

Standard Error (S.E.) 1.12786 1.15470

Maximum 136 Minimum 57 79 Range Variance 213.71 Standard Deviation(s) 14.6

L

U

S

Q 1

MMM

I.......................OEE • N DM

Coefficient -0.09 0.00

•

.

•

•

•

Q 3 •

.

•

S + •

•

•

•

•

•

•

•

EIN

A

N

=

=

40

150

Each 'H' above represents 4 counts

Skewness Kurtosis

M

J-L +2 (T

J-L +1 (T

J-L-1 (T

Each '.' above

•

•

=

•

•

•

.

.

•

•

•

•

M

A X

Coefficient/S.E. -0.49 -0.01 Q1 Q3 S­ S+

87.0 107.0 81.8 111.0

1.50

FIGURE 1.2 Distribution and summary statistics for the sample of canine glucose values (mg/ dl) in Table 1.1. Information adapted from printout of BMDP2D (Engelman, 1992).

4

Thomas B. Farver TABLE 1.1

Glucose (Glu, mgldl) and Alanine Aminotransferase (ALT, Ulliter) for a Sample of 168 Dogs hom the Population of Healthy Dogs'

Cow

Glu

ALT

Cow

Glu

ALT

Cow

Glu

ALT

Cow

Glu

ALT

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

88 104 89 99

60 79 138 58 34 43 47 77 102 34 64 184 82 35 46 29 117 132

43 44 45 46 47 48 49 50 51 52

86 86 115 98 98 99 94 104 107 107 119 114 94 109 110 99 105 102 100 83 83 108 114 105 74 92 97

53 50

85 86 87 88 89

110 78 95 111 116 108 76 111 86 101 106 92 67 75 127 87 136 94 89 72 87 96

54 54 37 25 115 60 36 102 62 43 73 99 50 24 110 65 44 40 18 30 75 66 113

127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168

108 90 100 96 86 100 122 109 77 88 94 92 121 86 84 86 105 91 92 89 123 109 117 115 83 94 92 109 92 93 92 101 113 92 110 116 111 111 70 94 106 102

105 32 25 46 95 99 115 60 67 83 118 44 64 19 68 74 86 47 56 49 78 93 46 31 65 55 52 64 59 49 29 66 53 79 47 46 137 57 49 80

63

97 94 105 86 124 118 112 85 109 96 72 91 94

90 68 84 94 91

90 72 87 94 97 103 70 91 58 89 81 106 94 57 67 93 89 80

112

68 50 95 140 38 146

68 42 43 84 44 84 108 28 75 38

38 26 89 35 69 44 47 41

53

54 55 56 57 58 59 60 61 62

63

64 65 66 67

68 69 70 71

85

80

83 86 110 121 87 88 114 96 107 101

81 82 83 84

1 00 65 95

72 73 74 75 76 77 78 79

90

72 59 80

42 42 116 98 78 56 38 50 47 32 53 97 97 54 36 32 111 63

58 24 96 42 101 46 58 29 115 62 40 78 83 26 19 105 133 56 70

90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126

85

95 96 117 106 113 107 96 94 1 00 127 106 93 99 94 82 130 76 99 81

63

61 62

33

99 97 131 44

68 37 52 113 142 45 80 53 87 36 31

53

128

• These data were provided by Dr. J. J. Kaneko, Department of Pathology, Microbiology and Immunology, School of Veterinary Medicine, University of California, Davis.

has a random component such as simple random sampling) from the population of normal animals and computing the mean and standard deviation of the sample. Once these estimates of JL and (T are obtained, an animal corning into the clinic in the future is classified as being normal for a particular analyte if its value for the analyte is within the bound of some multiple of the standard deviation below the mean and some mul-

tiple of the standard deviation above the mean. The multiple is determined by the degree of certainty that is desired to be placed on the classification scheme. For example, if the multiple chosen is 2, which is the conventional choice, any animal entering the clinic with an analyte value within 2 standard deviations of the mean would be classified as normal, whereas all animals with a value of the analyte outside this bound­ ary would be classified as abnormal. Because 95% of

Concepts of Normality in Clinical Biochemistry TABLE 1.2

ZO.50 ZO.55

=

=

=

ZO.60 ZO.65

ZO.70

%0.75

ZO.8O

=

=

=

=

=

ZO.85

0 0.126 0.253 0.385 0.524 0.674 0.842 1.036

Percentiles of the Standard Gaussian (z) Distributiona,b %0.90 ZO.91 ZO.92 ZO.93 ZO.94 ZO.95 ZO.96 ZO.97

= =

=

=

= =

=

=

1.282 1.341 1.405 1.476 1.555 1.645 1.751 1.881

ZO.9 75

=

=

ZO.98

=

ZO.99 ZO.995 ZO.999 Zo._

%0.99999

=

= =

=

1.960 2.054 2.326 2.576 3.090 3.719 4.265

a

This table is adapted, with kind pennission of the au­ thors and publisher, from Dixon, W. J., and Massey, F. J. (1983). "Introduction to Statistical Analysis," 4th ed., Table A-4, p. 511. McGraw-Hill, New York. b Example: The 75th percentile, or the Z value below which is 75% of the Gaussian distribution, equals 0.674, Z075 0.674. Percentiles smaller than the 50th percentile can be found by noting that the Gaussian distribution is symmetric about zero so that, for example, %0.30 -0.524. =

=

the Gaussian distribution is located within 1.96 or ap­ proximately 2 standard deviations of the mean, with this classification scheme, 2.5% of the normal animals would have a value of the analyte that would be below 2 standard deviations below the mean, and 2.5% of the animals would have an analyte value above 2 standard deviations above the mean. So with this classification scheme, there is a 5% chance that a true normal animal would be classified as being abnormal. Clinicians, by choosing 2 as the multiple, are willing to designate normal animals with extreme values of a particular analyte as being abnormal as the trade-off for not ac­ cepting too many abnormal animals as normals. With this methodology, no consideration is given to the dis­ tribution of abnormal animals because in fact there would be multiple distributions corresponding to the many types of abnormalities. The assumption is that for those cases where an analyte would be useful in identification of abnormal animals, the value of the analyte would be sufficiently above or below the center of the distribution of the analyte for normal animals. The reference interval for glucose based on the distribution from the sample of 168 normal dogs is 96.42857 mg/dl ± (1.96 X 14.61873 mg/dl) or 67.8 mg/dl (3.76 mmol/liter) to 125.1 mg/dl (6.94 mmol/liter). Solberg (1994) gives 1/a as the theoretical minimum sample size for estimation of the 100a and 100(1-a) percentiles. Thus a minimum of 40 animals is required to estimate the 2.5th and 97.5th percentiles but many more than 40 is recommended.

5

D. Methods for Determining Reference Intervals for Analytes Not Having the Gaussian Distribution The conventional procedure for assessing normalcy works quite well provided the distribution of the ana­ lyte is approximately Gaussian. Unfortunately, for many analytes a Gaussian distribution is not a good assumption. For example, Figure 1.3 describes the dis­ tribution of alanine aminotransferase (ALT) values given in Table 1.1 for the same sample of 168 normal dogs. This distribution is visibly asymmetric. The dis­ tribution has a longer tail to the right and is said to be skewed to the right or positively skewed. The skewness value (.91) exceeds the approximate 99th percentile of the distribution for this coefficient for random samples from a population having a Gaussian distribution. That the distribution is not symmetric and hence not Gaus­ sian is also evidenced by the lack of agreement between the mean, median, and mode as shown in Fig. 1.3. Application of the conventional procedure for comput­ ing reference intervals [x ± (1.96 X SD)] reveals a reference interval of 4.4 to 127.7 U/Iiter so that all the low values of the distribution fall above the value, which is 2 standard deviations below the mean of the distribution and more than 2.5% of the high values fall above the value, which is 2 standard deviations above the mean. The following sections gives two approaches that can be followed in such a situation to obtain refer­ ence intervals.

1. Use of Transformations Frequently, some transformation (such as the loga­ rithmic or square root transformation) of the analyte values will make the distribution more Gaussian (Kleinbaumetal., 1988; Neteretal., 1990;Zar, 1984). The boundaries for the reference values are two standard deviations above and below the mean for the distribu­ tion of the transformed analyte values. These bound­ aries then can be expressed in terms of the original analyte values by retransformation. Figure 1.4 de­ scribes the distribution of the ALT analyte values after transformation with natural logarithms. The reference boundaries in logarithmic units are equal to 4.08013 ± (1.96 X 0.47591) or (3.14734, 5.01292), which corre­ spond to (23.3, 150.3 U/liter) in the original units of the analyte.

2. Use of Percentiles The second approach that can be followed in the situation where an assumption of a Gaussian distribu­ tion is not tenable is to choose percentiles as boundaries (Feinstein, 1977; Herrera, 1958; Mainland, 1963; Mas-

6

Thomas B. Farver 86

Number of Distinct Values: Sample Size:

Estimate Mean

Standard Error (S. E • ) 2.42617

66.0 58.5

Median

H H HHH HHHH 95% Confidence Interval HHHHH Lower Upper HHHHHH H H 70.8 61.3 HHHHHHHHHH H H HHHHHHHHHHHHHH L----------------------U

168

3.17543 Not Unique

Mode

Each ' -' above L

Maximum 184 Minimum 18 Range 166 Variance 988.90 Standard Deviation(s) 31.4

U

S I• N

•

•

•

•

•

•

Q 1M •

0

•

•

D E

•

M

E• D I

•

M

E•

•

•

•

•

Q 3

Skewness

0.91

Coefficient/S.E. 4.83

Kurtosis

0.46

1.21

•

Q1 Q3

43.0

M A

S­

34.6

S+

97.5

S •

•

•

+ •

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

•

A N

=

Each '.' above

A

=

10 0 220

Each 'H' above represents 4 counts Coefficient

M

=

=

N

•

X

83.5

3.00

FIGURE 1.3 Distribution and summary statistics for the sample of canine alanine aminotransferase values (U /liter) in Table 1.1. Infonnation adapted from printout of BMDP2D (Engelman, 1992).

animals would be classified as abnormal when having analyte values either below the value of the analyte below which are 2.5% of all normal analyte values or above the value of the analyte below which are 97.5%

sod, 1977; Reed et aI., 1971; Solberg, 1994). For example, if we wanted to misclassify only 5% of normal animals as being abnormal, the 2.5th and 97.5th percentiles could be chosen as the reference boundaries. Thus,

H HHH HHH H HHHHHHHH Estimate 95% Confidence Interval Standard HHHHHHHHH Lower Upper Error (S.E.) HHHHHHHHHHH Mean 4.08013 4.15262 4.00764 0.03672 HHHHHHHHHHHH Median 4.06899 0.05444 H HHHHHHHHHHHHHH Mode Not Unique L--------- -------------U Each ' -' above = 0.15 Maximum 5.21494 Number of Distinct Values: Sample Size:

86

168

Minimum Range

2.89037 2.32457

Variance

0.22649

Standard Deviation(s)

0.47591

L

U

S

Q 1 MM I ............................EE • N M DA

o

=

2.4 5.7

Each 'H' above represents 3 counts Coefficient - 0.08 -0.53

Skewness Kurtosis

M

=

•

•

•

•

•

•

Q

S

3

+

•

•

•

•

•

Q1 •

•

•

•

•

IN

D

A

E

N

Each '.'

Coefficient/S.E. -0.41 -1.39

above

•

•

=

•

•

•

•

M • A

X

Q3 S­ S+

3.76120 4.42483 3.60422 4.55605

0.04

FIGURE 1.4 Distribution and summary statistics for the natural logarithm of the sample of canine ALT values (U /liter) in Table 1.1. Infonnation adapted from printout of BMDP2D (Engelman, 1992).

7

Concepts of Normality in Clinical Biochemistry

of all normal analyte values. This method is attractive because percentiles are reflective of the distribution in­ volved. The 97.5th percentile is estimated as the value of the analyte corresponding to the (n + 1) X 0.975th observation in an ascending array of the analyte values for a sample of n normal animals (Dunn, 1977; Ryan et al., 1985; Snedecor and Cochran, 1989). For the ALT values from the sample of n = 168 animals, (n + 1) X 0.975 = 169 X 0.975 = 164.775. Since there is no 164.775th observation, the 97.5th percentile is found by interpolating between the ALT values correspond­ ing to the 164th and 165th observation in the ascending array commonly referred to as the 164th and 165th order statistics (Ryan et al., 1985; Snedecor and Coch­ ran, 1989). The 164th order statistic is 138 U/liter and the 165th order statistic is 140 U/liter and the interpola­ tion is 138 + .775(140 - 138) = 139.5 U/liter. The 2.5th percentile is estimated similarly as the (n + 1) X 0.025th order statistic, which is the 4.225th order statistic for the sample of ALT values. In this case, the 4th and 5th order statistics are the same, 24 U/liter, which is the estimate of the 2.5th percentile. Note that there is rea­ sonable agreement between this reference interval and that obtained using the logarithmic transformation. This method of using percentiles as reference values can also be used for analytes having a Gaussian distri­ bution. The 2.5th and 97.5th percentiles for the sample of glucose values are 65.4 mg/dl (3.63 mmol/liter) and 126.3 mg/dl (7.01 mmol/liter), respectively. This interval agrees very well with that calculated earlier using the conventional method.

E. Sensitivity and Specificity of a Decision Based on a Reference Interval As alluded to earlier, in addition to the "normal" or healthy population, several diseased populations may be involved, each with its own distribution. Figure 1.5 depicts the distributions of one analyte for a single diseased population and for a normal healthy, nondis­ eased population. Note that there will be some overlap of these distributions. Little overlap may occur when the disease has a major impact on the level of the analyte, whereas extensive overlap could occur if the level of the analyte is unchanged by the disease. Using a reference value based on values of the ana­ lyte for normal animals could lead to two types of mistakes in diagnosis of patients. First, normal patients with values beyond the normal interval would be clas­ sified incorrectly as diseased and would be the false positives. Second, diseased patients having values within the normal interval would be classified as non-

0.03

>­

()

-r----r--,

&0.02 c

'" '" L

w..

.':: 0.0 I '"

'" 0::

0.00

Type D\��:�es

9-95

50

"\ , III

100 150 200 Glucose !mg/dll

M

lIilus Dogs

250

FIGURE 1.5 Overlapping Gaussian distributions of one analyte for a diseased dog population and a healthy, nondiseased dog popu­ lation with upper limit of reference interval, ct, and /3 error rates shown.

diseased, the false negatives. The two kinds of mistakes in classifying patients on the basis of analyte values are called the Type I error (saying a normal anjmal is diseased) and the Type II error (saying an animal is normal when in fact the animal is diseased). The proba­ bilities of making these two errors, the error rates, are a and /3, respectively. The size of these error rates is determined by the reference values, which are the decision points. We call 1 - a the sensitivity of the diagnostic or decision process using reference values (the probability of deciding that a truly normal animal is normal on the basis of the given reference value) and 1 - /3 is the specificity of the decision process (the probability of deciding that a truly diseased animal is diseased). It is possible to change the reference values to reduce the size of a and in so doing increase the sensitivity of the test. However, such an action will also result in an increase in the size of /3 and therefore a reduction in the specificity of the test.

Example 2 Type III diabetic dogs have the chemical form of diabetes mellitus generally regarded as the first level of development of the disease offering the highest likelihood "for successful oral hypoglyce­ mic therapy and/or dietary therapy" (Kaneko, 1977). Thus it would be useful to distinguish Type III dia­ betic dogs from normal dogs. Using the sample mean [155.6 mg/dl (8.63 mmol/liter)] and standard devia­ tion [32.0 mg/dl (1.77 mmol/liter)] of the plasma glucose values given by Kaneko (1977) for five dogs with Type III diabetes mellitus as reasonable estimates of the corresponding parameters for the population of dogs with Type III diabetes mellitus, and assuming that this population distribution is approximately Gaussian, a comparison of this distribution of glucose values can be made with that for the population of

8

Thomas B. Farver

normal dogs described by the approximately Gaussian distribution with parameter estimates given in Fig. 1.2

[/Lx = 96.4 mg/dl (5.35 mmol/liter) and (Ix = 14.6 mg/ dl (0.81 mmol/liter)]. These two distributions are shown in Fig. 1.5; they have reasonably good sepa­ ration with moderate overlap. Based on this informa­ tion, a diagnostic procedure is proposed whereby a dog entering the clinic with a glucose value above 125.1 mg/dl (6.94 mmol/liter), the upper limit of the normal reference interval, will be flagged as possibly having Type III diabetes mellitus, thereby indicating the need for more follow-up. (Note: This is an oversim­ plification of actual practice because a diagnostic deci­ sion of this type would be based on additional informa­ tion, such as the animal's glucose tolerance and insulin response, making the decision rule and subsequent error calculations more complex.) This is an example of a one-sided diagnostic procedure because dogs with a glucose value below the lower limit of the reference interval would not be considered as having Type III diabetes mellitus. Thus the Type I error is concentrated at the upper end of the distribution of glucose values for normal dogs and its magnitude (a) is the area to the right of a glucose value of 125.1 mg/dl in the distribution of glucose values for normal dogs or the

area to the right of the corresponding z value, z = (125.1 - 96.4)/14.6 - 1.96, for the standard Gaussian distribution (see Section II.B); a = 2.5%. The clinician may be interested in determining the sensitivity and the specificity of the diagnostic proce­ dure. The sensitivity is 1 - a = 97.5%. A dog that actually has Type III diabetes mellitus but has a glucose value less than 125.1 mg/ dl would be incorrectly classi­ fied by the proposed diagnostic procedure as being a normal dog. This is a Type II error and the probability of making this type of error is {3, which is the area to the left of a glucose value of 125.1 mg/ dl in the distribution of glucose values for dogs with Type III diabetes mellitus or the area to the left of the corre­ sponding z value, Z = (125.1 - 155.6)/32.0 - -.953, for the standard Gaussian distribution. Using Table A-4 from Dixon and Massey (1983), {3 = 17.62% (or 17.14% by linear interpolation of the information given in Table 1.2), and the specificity of the diagnostic procedure is 1

- {3

= 82.38%.

having a reference value within the normal interval is actually nondiseased or the predictive value of a negative diagnosis, prob(DI-). The predictive value depends on the sensitivity, specificity, and prevalence ( p) of the disease as shown in the following equations: Prob(DI +)

p

X Sensitivity

p

X Sensitivity + (1 -

Prob(DI-)

(1 -

(1 -

p)

p)

p)

X (1 - Specificity)

X Specificity

X Specificity +

p

X (1 - Sensitivity)

Fig. 1.6 demonstrates the extent to which the predictive value of a positive diagnosis changes with the preva­ lence. In general, larger changes are seen in the predic­ tive value of a positive diagnosis for smaller changes in the prevalence for diseases with low prevalence, and smaller changes are seen in the predictive value for larger changes in the prevalence for diseases with high prevalence. In the example of the diagnostic procedure given in the previous section, assuming the prevalence of Type III diabetes mellitus in the dog population was 2%, Prob(DI+)

=

(0.02 X 0.975)/[(0.02 X 0.975) + (0.98 X 0.1762)]

= 0.10 or 10%, and Prob(DI-) = (0.98 X 0.8238)/[(0.98 X 0.8238) + (0.02 X 0.025)] =

0.999 or 99.9%.

To demonstrate how specificity and hence the predictive value of a positive test improve with greater separation of the populations, Kaneko (1977) gives estimates (based on a sample of 11

1.0 0.9 0.8

� .� �

'0 '6 l!! Q.

0.7 0.6 0.5 0.4 0.3 0.2

F. Predictive Value of a Decision Based on a Reference Interval

O��-J__��__L--L����� o 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

0.1

A useful quantity is the probability that a patient having a reference value outside the normal interval actually has the disease. This is known as the predictive value of a positive diagnosis, Prob(DI +). Interest could

of a positive laboratory test having

also be in determining the probability that a patient

ficity.

Prevalence

FIGURE 1.6

Impact of disease prevalence on the predictive value

95% sensitivity and 80% speci­

9

Concepts of Normality in Clinical Biochemistry dogs) of the mean and standard deviation of the plasma glucose values of the population of dogs

assay is run repeatedly on the same sample and the results obtained have little variability, the assay is said

with Type I diabetes mellitus (the juvenile or child­ hood form) as Jl = 415.1 mg/dl (23.02 mmol/liter)

to have high precision. Large variability in the ob­ served results is an indication of low assay precision. Note that precision is defined in reference to what is actually being measured and not to the target value. Clinical analysts have always had a goal of achieving the highest possible level of precision for a particular

of a positive test increases to 78%.

assay within a laboratory. Emphasis is presently placed on meeting an "average laboratory" level of precision

and fr = 114.3 mg/dl (6.34 mmol/liter). If we use these values in the preceding calculations with the diagnostic value remaining at 125.1 mg/ dl, the specificity improves to 99.44% and the predictive value

III. ACCURACY IN ANALYTE MEASUREMENTS Accuracy has to do with the conformity of the actual value being measured to the intended true or target value. An analytical procedure having a high level of accuracy produces measurements that on average are close to the target value. An analytical procedure hav­ ing a low level of accuracy produces measurements that on average are a distance from the target value. Such a procedure in effect measures something other than is intended and is said to be biased. Failure of analytical procedures to produce values that on aver­ age conform to the target values is due to unresolved problems, either known or unknown, in the assay. The degree of accuracy of an analytical procedure has been difficult to quantify due to the fact that the target value is unknown. It is now possible for labora­ tories to compare their assay results with definitive results obtained by the use of isotope dilution-mass spectrometry (Shultz, 1994). Shultz (1994) reports the results of two large surveys of laboratories in the United States (Gilbert, 1978) and Sweden (Bjorkhem et ai., 1981) in which samples from large serum pools were analyzed for frequently tested analytes (calcium, chloride, iron, magnesium, potassium, sodium, choles­ terol, glucose, urea-nitrogen, urate, and creatinine). The laboratory averages were compared with the tar­ get value obtained using definitive methods, and the results of these surveys indicated that, with the excep­ tion of creatinine, all averages expressed as a percent­ age of the target value were within the accuracy goals published by Gilbert (1975). Results from individual laboratories naturally would vary about the average, and many of these laboratories would not have met the accuracy goal.

IV. PRECISION IN ANALYTE MEASUREMENTS Precision has to do with how much variability there is about the actual value being measured when the assay is replicated. If in a given laboratory a particular

(Shultz, 1994). The level of precision is stated quantitatively in terms of the coefficient of variation

(ev).

The

ev

is the

ratio of the standard deviation to the average of a series of replicated assays and its magnitude depends on the concentration of the analyte. Elevitch (1977) and Harris

(1988) provide the guidelines on the desired level of precision in terms of the cv. In the case where the analytical results are intended to assist in the diagnos­ tic process or to assist in monitoring a patient's re­ sponse to treatment, the level of laboratory precision of a given analyte in terms of the cv (eva) need only be

a function of the within day and day-to-day variability or intrasubject variation of healthy subjects. Specifi­ cally,

eVa < -!cVintrasubject.

In the case where analytical test results are to be used to screen a population, the laboratory precision goal in terms of the ev should be a function of the variability in response among healthy subjects or intersubject variation. Specifically,

eVa < -!cVintersubject.

Use of intrasubject variability as a goal for precision has appeal because this source of variability would be considered in decision processes relating to patients. Unfortunately, a given analysis reflects not only this intrasubject variability, but also imprecision in the assay. Shultz (1994) summarizes the results of a large national survey of laboratory precision. With the ex­ ception of high-density lipoprotein and thyroxine (T4), the precision of the assay for the analytes evaluated from the "average" laboratory met or nearly met the precision goals based on the intrasubject variability. This result has to be regarded as encouraging, no doubt reflecting the tremendous emphasis that has been placed on quality control by laboratories as well as the use of automation in analytical work. On the other hand, there were some analytes for which the assay precision for the "average" laboratory was above the precision goal. It also must be remembered that there would be many individual laboratories that would not have assay precision profiles as good as the" average"

10

Thomas B. Farver

laboratory. Assay precision in excess of the precision ?oal ba�ed on physiological variability makes it nearly Imposslb�e to rule out the possibility that very large changes In the level of an analyte are a reflection of assay imprecision.

V. INFERENCE FROM SAMPLES The basis for everything that has been discussed to this point is probability and distributional theory. No other theory is relevant unless one is operating at the level where inference is to be made on the basis of a sample from the underlying population. Most stan­ dard statistical theory assumes that the sample was obtained by simple random sampling.

A. Simple Random Sampling

Simple random sampling (SRS) is a method of sam­ pling whereby, at each step of the sampling process, the elements available for selection have an equally �ik�ly chance of being selected. In most applications, It IS assumed that the elements are selected without replacement, although the elements could be selected with replacement. If the number of elements to be selected is small relative to the number of elements in the population, then it is rather unlikely that an ele­ ment will be selected more than a single time with replacement sampling, so that in such situations sam­ pling with replacement produces essentially the same results as sampling without replacement. It is only when a small finite population is being sampled that differences may be noted between the two methods. Three steps are used to select a sample by SRS with­ out replacement: All elements in the population must first be identified by a unique number from 1 to N, the population size. Then n numbers are selected from a table of random numbers or selected by a random number generator, which gives the numbers 1 to N in random order. Numbers appearing more than once are ignored after their first use. Finally, those elements having numbers corresponding to the n numbers selected constitute the sample. There are other probability-based sampling procedures that should be considered in practice; these methods are found in texts on sampling (Cochran, 1977; Jessen, 1978; Levy and Lemeshow, 1991; Murthy, 1967; Raj, 1968,1972; Scheaf­ fer et ai., 1996).

B. Descriptive Statistics Once the data have been collected, so-called "de­ scriptive statistics" can be computed. As the name sug­ gests, these statistics are useful in describing the under-

lying populations. For example, because complete information for the entire population is not available, it is not possible to know the population mean, J.L = Ix; / N. (Here x; designates the value of the ith element in the population and I indicates summation. Thus the population mean J.L is found by summing the values of all N elements in the population and then dividing the sum by N.) However, a sample mean based on the sample can be computed as x = IXi / n, the sum of the values of all n elements in the sample divided by n. If the sample has been selected in a manner that results in a small bias, x should be a reasonably good estimate of the population mean, J.L, and will be a better estimate as the sample size increases. Other estimates of the measures of central tendency of the population can be obtained from the sample, such as the sample median and the sample mode. Also, sample-based estimates of the measures of dispersion or spread for the population can be obtained. The sample variance is computed as 52 = I(xi - x f/ (n - I), and the sample standard deviation, s, is obtained by taking the square root of S2. The descriptive statistics are called point estimates of the parameters and represent good approximations of the parameters. An alternative to the point estimate is the interoai estimate, which takes into account the underlying probability distribution of the point esti­ mate called the sampling distribution of the statistic.

C. Sampling Distributions

In actual practice, only a single sample is taken from a population and, on the basis of this sample, a single point estimate of the unknown population parameter is computed. If time and resources permitted repeated sampling of the population in the same manner, that is, with the same probability-based sampling design, one point estimate would be obtained for each sample obtained. The estimates would not be the same because the sample would contain different elements of the population. As the number of such repeated sampling operations increases, a more detailed description emerges of the distribution of possible point estimates that could be obtained by sampling the popUlation. This is the sampling distribution of the statistic. Some fundamental facts relating to the sampling distribution of the sample mean follow: (1) The center of the sampling distribution of x is equal to J.L, the center of the underlying distribution of elements in the population. (2) The spread of the sampling distribution of x is smaller than a2, the spread of the underlying distribution of elements in the population. Specifically, the variance of the sampling distribution of x (denoted ai) equals a2/ n, where n is the sample size. So increas­ ing the sample size serves to increase the likelihood

11

Concepts of Normality in Clinical Biochemistry

of obtaining an x close to the center of the distribution because the spread of the sampling distribution is be­ ing reduced. (3) The central limit theorem (Daniel, 1995; Remington and Schork, 1985; Zar, 1996) states that regardless of the underlying distribution of the population of elements from which the sample mean is based, if the sample size is reasonably large (n > 30), the sampling distribution of X is approximated well by the Gaussian distribution. So X drawn from any distribution has a sampling distribution that is approximately N(JL, u2/ n) for n > 30. If the distribution of the underlying population of elements is Gaussian or approximated well by a Gaussian distribution, the sampling distribution of x will be apprOXimated well by the Gaussian distribution regardless of the sample size on which x is based. Probabilities of the sampling distribution of X, N(JL, u2/ n), can be evaluated using the method described in Section II.B. Example 3 Suppose the underlying population of elements is N(4,16) and a sample of size n = 9 is drawn from this population using SRS. We want to find the probability of observing a sample mean less than 3.1 or greater than 6.2. In solving this problem, the relevant sampling distribution is specified: X is N(4, 16/9). Note that the sampling distribution of x is Gaussian because the problem stated that the underlying population was Gaussian. (Otherwise the stated sample size would needed to have been 30 or larger to invoke the central limit theorem.) The probability of observing X < 3.1 in the distribution of X is equivalent to the probability of observing Z < (3.1 4)/ (4/3) = -0.675 in the stan­ dard Gaussian distribution. Going to Table 1.2, Z = 0.675 is approximately the 75th percentile of the stan­ dard Gaussian distribution and by symmetry z = -0.675 is approximately the 25th percentile. Thus, the probability of observing a z value less than or equal to -0.675 is approximately 0.25. The probability of . observing a sample mean greater than 6.2 is equivalent to the probability of observing z > (6.2 4)/(4/3) = + 1.65. Table 1.2 gives the probability of observing a z < 1.65 as approximately 0.95, so the probability of observing a z > 1.65 equals 1 - 0.95 or 0.05. The desired probability of observing a sample mean less than 3.1 or greater than 6.2 is the sum of 0.25 and 0.05, which is 0.3 or 3 chances in 10. -

-

D. Constructing an Interval Estimate of the Population Mean, IL This brief exposure to sampling distributions and their standardized forms provides the framework for generating an interval estimate for JL. Consider the

probability statement Prob(-2 < z < +2) = 0.9544. Because z = (x - JL)/(u/Vn), this probability state­ ment is equivalent to the statement Prob(-2 < (X - JL)/ (u/Vn) < +2) = 0.9544. Some standard algebraic ma­ nipulation of the inequality within the parentheses gives Prob(x + (-2u/Vn) < JL < x + (-2u/Vn» = 0.9544. This is the form of the confidence statement about the unknown parameter JL. With repeated sam­ pling of the underlying population, 95.44% of the inter­ vals constructed by adding and subtracting 2u/Vn to and from the sample mean would be expected to cover the true unknown value of JL. The quantities of x 2u/Vn and x + 2 u/Vn are called the lower and upper confidence limits, respectively, and the interval bounded below by x - 2u/Vn and above by x + 2u/Vn, that is, (X - 2 u/ Vn: x + 2u/Vn), is the 95.44% confidence interval for JL. In practice, only one sample is taken from the population and thus there is a 95.44% chance that the one interval estimate obtained will cover the true value of JL. The value of 95.44% or 0.9544 is called the confidence level. The degree of confidence that is to be had is deter­ mined by the amount of error that is to be tolerated in the estimation procedure. For a 0.9544 level of con­ fidence, the error rate is 1 - 0.9544 = 0.0456. The error rate is designated by alpha, a. The size of a determines the magnitude of the value of z that is multiplied by u/Vn. The convention is to apportion half of a to the lower end and half of a to the upper end of the sam­ pling distribution of x so that the relevant values of z are (1) the z value that has a/2 area to its left, Z,,/2, and (2) the z value that has a/2 to its right or, equivalently (to conform to Table 1.2), the z value that has 1 - (a/ 2) area to its left, ZI-(a/2). Therefore, the most general (Za/2U)/ form of the interval estimate statement is (x Vn: X + (ZI-(a/2)U)/Vn) or x ± (ZI-(a/2W)/Vn because of the symmetry of the Gaussian distribution. -

Example 4 Assuming the distribution in Example 3, construct a 90% confidence interval for JL. A 90% level of confidence implies a tolerated error rate of 10% and the relevant z values are (1) that which has 5% of the distribution of z to its left or ZO.05 -1.645 and (2) that which has 95% of the distribution to its left or ZO.95 = 1.645. The 90% confidence interval for JL is therefore (x ± (1.645 x 4)/3) or (X ± 2.1933) where x is the sample mean obtained by taking a sample of size 9 from the population. The form of the interval estimate given earlier as­ sumes that u, the standard deviation of the underlying population, is also known. Frequently this parameter, like JL, is unknown and must be estimated from the sample drawn from the population using the estimator 2 S2 = �(Xi - X) / (n - 1). For small samples (n < 30), =

12

Thomas B. Farver

the standardized form of X, ex JL)/ (s/ Vn), does not have a standardized Gaussian distribution, N(O,I), but rather has the t distribution corresponding to the effective sample size, n 1, the degrees of freedom. Table 1 .3 gives the percentiles of several t distributions. A given row of Table 1 .3 pertains to the t distribution with the indicated effective sample size or degrees of -

-

TABLE 1.3 df

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 35 40 45 50 60 70 80 90 100 120 140 160 180 200 00

freedom. The entries in the row are percentiles or those points of the given t distribution that have the indicated area to the left. For example, to.95,10 is 1.812, meaning that the 95th percentile of the t distribution with 10 degrees of freedom is 1.812, which is equivalent to saying that 95% of the t distribution with 10 degrees of freedom lies to the left of the t value 1.812. Note

Percentiles of Student's t Distribution''/>

to.ss

to.s

to.75

to.as

to.90

to...

to.m

to.99

to.995

to._

0.158 0.142 0.137 0.134 0.132 0.131 0.130 0.130 0.129 0.129 0.129 0.128 0.128 0.128 0.128 0.128 0.128 0.127 0.127 0.127 0.127 0.127 0.127 0.127 0.127 0.127 0.127 0.127 0.127 0.127 0.127 0.126 0.126 0.126 0.126 0.126 0.126 0.126 0.126 0.126 0.126 0.126 0.126 0.126 0.126

0.510 0.445 0.424 0.414 D.408 0.404 0.402 0.399 0.398 0.397 0.396 0.395 0.394 0.393 0.393 0.392 0.392 0.392 0.391 0.391 0.391 0.390 0.390 0.390 0.390 0.390 0.389 0.389 0.389 0.389 0.388 0.388 0.388 0.388 0.387 0.387 0.387 0.387 0.386 0.386 0.386 0.386 0.386 0.386 0.385

1.000 0.816 0.765 0.741 0.727 0.718 0.711 0.706 0.703 0.'700 0.697 0.695 0.694 0.692 0.691 0.690 0.689 0.688 0.688 0.687 0.686 0.686 0.685 0.685 0.684 0.684 0.684 0.683 0.683 0.683 0.682 0.681 0.680 0.679 0.679 0.678 0.678 0.677 0.667 0.677 0.676 0.676 0.676 0.676 0.674

1.963 1.386

3.078 1.886 1.638 1.533 1.476 1.440 1.415 1.397 1.383 1.372 1.363 1.356 1.350 1.345 1 .341 1.337 1.333 1.330 1.328 1.325 1.323 1.321 1.319 1.318 1.316 1.315 1.314 1.313 1.311 1.310 1.306 1.303 1.301 1.299 1.296 1.294 1.292 1.291 1.290 1.289 1.288 1.287 1.286 1.286 1.282

6.314 2.920 2.353 2.132 2.015 1.943 1.895 1.860 1.833 1.812 1.796 1.782 1.771 1.761 1.753 1.746 1.740 1.734 1.729 1.725 1.721 1.717 1.714 1.711 1.708 1.706 1.703 1.701 1.699 1 .697 1.690 1.684 1.679 1.676 1.671 1.667 1.664 1.662 1.660 1.658 1.656 1.654 1.653 1.653 1.645

12.706 4.303 3.182 2.776 2.571 2.447 2.365 2.306 2.262 2.228 2.201 2.179 2.160 2.145 2.131 2.120 2.110 2.101 2.093 2.086 2.080 2.074 2.069 2.064 2.060 2.056 2.052 2.048 2.045 2.042 2.030 2.021 2.014 2.009 2.000 1.994 1.990 1.987 1.984 1.980 1.977 1.975 1.973 1.972 1.960

31.821 6.965 4.541 3.747 3.365 3.143 2.998 2.896 2.821 2.764 2.718 2.681 2.650 2.624 2.602 2.583 2.567 2.552 2.539 2.528 2.518 2.508 2.500 2.492 2.485 2.479 2.473 2.467 2.462 2.457 2.438 2.423 2.412 2.403 2.390 2.381 2.374 2.368 2.364 2.358 2.353 2.350 2.347 2.345 2.326

63.657 9.925 5.841 4.604 4.032 3.707 3.499 3.355 3.250 3.169 3.106 3.055 3.012 2.971 2.947 2.921 2.898 2.878 2.861 2.845 2.831 2.819 2.807 2.797 2.787 2.779 2.771 2.763 2.756 2.750 2.724 2.704 2.690 2.678 2.660 2.648 2.639 2.632 2.626 2.617 2.611 2.607 2.603 2.601 2.576

636.619 31.599 12.924 8.610 6.869 5.959 5.408 5.041 4.781 4.587 4.437 4.318 4.221 4.140 4.073 4.015 3.965 3.922 3.883 3.850 3.819 3.792 3.768 3.745 3.725 3.707 3.690 3.674 3.659 3.646 3.591 3.551 3.520 3.496 3.460 3.435 3.416 3.402 3.390 3.373 3.361 3.352 3.345 3.340 3.291

1.250

1.190 1.156 1.134 1.119 1.108 1.100 1.093 1.088 1.083 1.079 1.076 1.074 1.071 1.069 1.067 1.066 1.064 1.063 1.061 1.060 1.059 1.058 1.058 1.057 1.056 1.055 1.055 1.052 1.050 1.049 1.047 1.045 1.044 1.043 1.042 1.042 1.041 1.04 0 1.040 1.039 1.039 1.036

• This table is reprinted, with kind pennission of the authors and publisher, from Kleinbaum, D. G., and Kupper, L. L. (1988). "Applied Re/fessing Analysis and Other Multivariable Methods," 2nd ed., Table A-2, p. 647. PWS-Kent, Boston. Example: The 75th percentile of a 1 distribution with 11 degrees of freedom or the 1 value below which is 75% of the 1 distribution with 11 degrees of freedom equals 0.697, 10.75,11 0,697. =

13

Concepts of Normality in Oinical Biochemistry that the 5th percentile of the

t

distribution with

degrees of freedom is - 1.812 because the

t

10

distribu­

tions, like the standard Gaussian distribution, are sym­ metric about zero. Thus, for smaller sample sizes when the value of

u is unknown, the form of a interval for IL is x ± [(t1- a/ 2;n-1S)/Vn].

confidence

As with the mean, interest also centers around esti­

mating the variance,

u2, and standard deviation,

u, of

the population. The estimates for these parameters are

52 and 5,

respectively. One might also be interested in

constructing confidence intervals for these parameters. Because space does not permit further elaboration, the interested reader is referred to several introductory

sponses may not be and most likely are not indepen­ dent so the procedure given above for comparing two groups would not be appropriate. Rather, the differ­ ences in response (pretreatment minus posttreatment response) are formed and the population of interest is the single population of differences, having as one of its parameters the mean difference,

ILd .

The quantity

ILd is estimated by the mean difference for the sample

Ii = Idi 1 n, and an interval estimate (t1-a/ 2;n-15d ) / Vn where 5d is an esti­

is formed by Ii ±

of n differences,

mate of Ud ' If the interval thus constructed covers zero, then it may be that there is no difference between the mean pretreatment and posttreatment values.

statistics books for the relevant formulas and their deri- vations (Daniel, 1995; Dixon and Massey, 1983; Dunn, 1977; Remington and Schork, 1985).

E. Comparing the Mean Response of Two Populations

1. Independent Samples

The presentation thus far has focused on estimation of parameters from a single population. Frequently, interest lies in two populations. For example, in a clini­ cal trial, one group of animals might receive some treatment (t) while a second group of animals serves as a control receiving no treatment (c). Among the several points of interest could be that of estimating the difference in central response for the two popula­

F. Comparing the Mean Response of Three Populations Comparisons of more than two groups is the natural progression from the methodology discussed to this point. Consider the comparison of three independent groups. The approach that immediately comes to mind is that of estimating the three groups' means and stan­ dard deviations and then constructing three sets of confidence intervals (the first group versus the second group, the first group versus the third group, and the second group versus the third group) using the ap­ proach described earlier for two independent groups. However, some modifications need to be made. First, because we are assuming that all three groups have

ILl lLet where the subscripts designate the groups. The point estimate of ILl - ILc is XI - XC' If we assume that the variances of the two populations,

equal variances, pooling of the variances for the three

u2, then the estimate 2 of u is s� = [ (n l 1 1t + ( nc - l )ic] / ( n 1 + nc - 2) (called the pooled variance) and [(XI - xc) - (ILl - ILc)] 1 sp(l 1 nl + 1 1 nc )t- has a t distribution with nl + nc - 2 de­ grees of freedom. A 100 (1 - a)% confidence interval for JLt - ILc is (XI - xc ) ± tl- a/ 2;nt+nc-zsp (1 1 nl + 1 1 nc )t-.

variance is the natural extension of that for two groups,

tions, that is,

that is,

01

and

-

�

are unknown but equal and the

common variance is designated as -

If the interval so constructed covers zero, it may be that

there is no difference between the central responses for the two distributions; otherwise, it could be concluded that the central responses differ Significantly. 2.

Nonindependent Samples The procedure just discussed, in addition to assum­

ing that the variances are homogeneous, also assumes that the two samples are drawn independently. An alternative design for comparing two responses in­ volves using each subject as its own control. For exam­ ple,

a

pretreatment response in an individual might

be compared with a posttreatment response. Clearly

in this design, the pretreatment and posttreatment re-

groups provides a better estimate of the common vari­ ance than does pooling of just the variances for the two groups being compared. The form of the pooled namely: s� = [(ni - l)sy + (n2 - l)s� + (n3 - l)s5] 1 3). When the group simple sizes are (ni + n2 + n3 equal, that is, ni = n 2 = n3t s� = liTl 3. The quantity s� -

is used for all three interval estimates. Second, the error rate of each comparison has to be

adjusted so that the error rate over all three compari­ sons will be a. This is required because theoretically it turns out that the error rate over all three compari­ sons is larger than that for a single comparison. Several approaches are suggested in the literature for circum­ venting this problem. One such approach attributed to Fisher (1966) is called the least significance difference (LSD) procedure (Kleinbaum and Kupper, 1988; Steel and Torrie, 1980). The LSD procedure involves making each single comparison with an error rate of al m where m is the total number of comparisons to be made, which in the present example is 3. This approach gives an error rate over all comparisons of a. In gen­ eral, the form of the interval estimate is (Xi - Xj) ±

[( t1 -a/2m;n1 +n2 +_ . . +nk -k) ( Sp ) (1 1 ni + 1 / nj)t-] where m is the

14

Thomas B. Farver

total number of comparisons to be made and k is the total number of groups. Intervals covering zero would indicate no difference in the central value of the groups being compared.

jected, the process stops or perhaps, at most, the data from all the groups might be pooled together and the parameters of this single population estimated. Tables of the percentiles of the F distributions can be found

G. Analysis of Variance and Its Uses

Schork, 1985), which also provide instruction on how to read the tables. Also, these texts give the generaliza­ tion of the among-group mean square when the sample size is not constant for all groups.

1. Comparing Population Means

The process of deciding whether or not there are differences between groups in the central value of the response being evaluated can be approached using the method of analysis of variance (ANOVA). ANOV A involves decomposing the total variability in a given set of data into parts reflective of the amount of vari­ ability attributable to various sources. One source of variability is that within the group. Because the groups are assumed to have the same spread, this source of variability is estimated as the pooling of the estimated variances for the k groups considered and is equal to s� as defined earlier. The second source of variability results from the variability among groups means. If there is no difference in the groups means, then the k samples can be thought of as being k independent samples from a common population, and the k means, therefore, represent a sample of size k from the sam­ pling distribution of x. Their variance represents an estimate of the variance of the sampling distribution

of X, that is, o-� = I [Xi - I(XJk}F/(k - 1 ), where denotes an estimate. Because o-� = 0-2/ n, where n is the common sample size, inflation of � by n will produce a second estimate of the variance of the underlying population assuming no differences among the group means. These two estimates of u2 should be about A

equal and their ratio close to 1. I f there i s significant separation among some o r all of the group means, the second variance estimate, when computed as described, will be larger than the first estimate, indicating that a component of variance is being estimated beyond that embodied only in an esti­ mate of the within-group variability. This additional variance component being estimated is the variance among group means. ANOVA in this example involves generating the two estimates of u2 (called mean squares) under the hypothesis that all of the group means are equal. The hypothesis is then tested by forming the ratio of the two mean squares (the second mean square divided by the first mean square) called the F statistic. The F statistic has a probability distribution called the

F distribution. If the computed F value is greater than the table distributional value, then the hypothesis is rejected and subsequently confidence intervals are con­ structed as described earlier to determine which group means are different. If the hypothesis cannot be re-

in all introductory texts on statistics (Daniel, 1995; Dixon and Massey, 1983; Dunn, 1977; Remington and

The results of an ANOVA are traditionally summa­ rized in a table called the ANOV A table of which Table

1.4 is such an example. The first column of Table 1.4 shows the sources of variability into which the total variability is decomposed. In the present example, these sources are due to the variability within groups and that which is reflective of variability among group means. Column 4 gives the two independent (under a hypothesis of no difference in response among groups) estimates of u2 or mean squares, mean square among group means (MSA) and mean square within groups (MSw). Columns 3 and 2 give, respectively, the numera­ tor (called the sum of squares) and the denominator (the effective sample size or degrees of freedom) of the corresponding mean square. The sums of squares provide a check on calculations when generating an ANOVA table because the total of the sum of squares attributable to the various sources of variability is equal to the sum of squared deviations of each observation across the k samples from the grand mean of all the observations in the k samples, called the total sum of squares. In other words, the total sum of squares is the numerator for estimating the total variability in the data ignoring group membership. Similarly, the de­ grees of freedom for the sources of variability sum to the effective sample size for estimating the total variability in the data ignoring group membership of the observations [here, kn - 1 = (k - 1) + k(n - 1 )].

2. Identifying Significant Sources of Variability in Experimental Designs a. Experimental Design

1:

Two Factors Crossed

The ANOVA does not represent a superior method to that of the LSD procedure in the context in which it was presented. In fact, if a significant group effect was noted by the ANOVA, the LSD procedure or any other multiple comparison procedure would be used to identify which group differences were contributing to the overall group effect. The reason analysis of vari­ ance has been introduced is that it is a very convenient method for assessing the importance of the various sources of variability encountered in the complex de­ signs of clinical laboratories.

15

Concepts of Normality in Clinical Biochemistry TABLE 1.4

Analysis of Variance Table for Classification of Responses on Basis of One Factor, Equal Responses for Each Class'

Degrees of freedom

Among group means

1

SSA =

- 1)

SSw =

k-

Within group

k (n

Total

kn - 1

Mean square

Sum of squares

Source of variation

(k - I)MSA k (n

SST = SSA

-

+

I)MSw

SSw

k

k

- 2:(xJ k)j2 2:[Xj ;=1 i=1 MSA = l k k 2:sr MSw = � k n

F value

MSA/MSw

a k is the number of classes or groups, n is the number of responses per group, constant over aU groups, S7 is the variability of the responses within the ith group.

As an example of this kind of application of ANOVA, suppose interest centers around an evalua­ tion of the plasma cortisol secretory patterns in dogs with pituitary-dependent hyperadrenocorticism. A part of this type of assessment traditionally involves a determination of the magnitude of intra-animal and interanimal variances. One experiment that would pro­ vide estimates of these variances is as follows: Twenty dogs are randomly selected to participate in the experi­ ment. On three randomly selected days of the study period, plasma cortisol samples are obtained at 11:00 A.M. from all 20 dogs. The assay for plasma cortisol is replicated for each sample. This experimental design involves two factors, "animals" and "days." These fac­ tors are said to be crossed in that each level of factor "animals" (Le., each animal) has been evaluated at each level of factor "days" (Le., at each of the three days on which plasma cortisol samples were drawn). The total variability in this experiment can be decom­ posed by the method of ANOVA into four parts: that which will identify significant among-animals variabil­ ity, significant among-days variability, and significant variability attributable to a lack of parallelism among the animals in their response across the three days samples were drawn (called the interaction between the factors " animals" and " days"), and that which will quantify the residual variability embodying all other sources of variability related to the assay. Table 1.5 lists data that shall be assumed to have been generated from an experiment using the design given in the preceding paragraph. Table 1.6 gives the ANOVA table from these data. In addition to provid­ ing the standard entries of an ANOVA table (sum of squares, degrees of freedom, mean square, and F statis­ tics with corresponding p values), Table 1.6 also gives the expected mean squares, which are the quantities being estimated by the corresponding mean squares in the table. The expected mean squares are functions

of the unknown population parameters. Note that each expected mean square involves one or more of the variance components of the design. The expected mean squares indicate which mean squares are appropriate as the denominator for computing the F statistic to test the hypothesis for a given effect. For example, to test the hypothesis that the variance component giving the magnitude of the variability attributable to the interac­ tion between the factors " animals" and "days" is not Plasma Cortisol Levels (mg/dl) in Dogs with Pituitary-Dependent Hyperadrenocorticism" TABLE 1.5

Dog

Day 1

Day 2

Day 3

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

0.8, 0.8b 2.2, 2.5 2.3, 2.4 3.1, 3.2 4.2, 4.3 3.8, 3.7 2.6, 2.5 2.4, 2.3 2.9, 2.9 2.2, 2.3 3.1, 2.9 5.3, 5.2 3.2, 3.3 4.4, 4.3 4.2, 4.1 2.9, 3.1 3.8, 3.7 5.0, 5.2 3.7, 3.6 2.9, 3.0

1.5, 1.4 4.7, 4.5 2.7, 2.7 2.0, 2.1 3.5, 3.5 4.0, 4.0 3.9, 4.1 4.1, 4.1 3.3, 3.5 2.0, 2.2 3.0, 2.8 4.4, 4.2 2.7, 2.5 4.8, 4.8 4.3, 4.2 3.7, 3.9 3.6, 3.4 5.1, 5.1 1.4, 1.4 2.7, 2.9

4.0, 4.2 3.4, 3.5 2.8, 2.7 2.7, 2.8 2.3, 2.4 2.7, 2.9 2.7, 2.7 2.9, 2.9 2.7, 2.8 1.6, 1.6 2.3, 2.1 3.9, 4.0 2.7, 2.9 3.8, 3.6 3.6, 3.7 4.6, 4.8 3.5, 3.3 4.8, 4.7 2.2, 2.1 3.1, 3.1

a These data were adapted from data provided by Dr. E. C. Feldman, Department of Medicine and Epidemiology, School of Veterinary Medicine, University of California, Davis. b Assay replicated for each sample.

16

Thomas B. Farver TABLE 1.6 Analysis of Variance of Canine Plasma Cortisol Level Data Assuming Experimental Design 1 ("Days" Crossed with "Animals")

Source of variation

df

Sum of squares

Mean square

Among animals Among days Animals X days Error Total

19 2 38 60 119

70.8316 1.1546 42.6787 0.5750 115.2399

3.7280 0.5773 1.1231 0.0096

significant, that is, to test Ho : u"a 0, the test statistic is Fad = MSad / MS., where MS. is the correct denomina­ tor for this test statistic because if the hypothesis is true and uaa 0, the expected mean square for the interaction between " animals" and "days" reduces to =

=

u; so that both MSad and MS. would be estimating u; and it would be expected that the ratio of MSad to MS. (Fad) would not differ significantly from 1. This ratio differing significantly from 1 would be an indication that MSad was estimating something additional to u;, namely, nuaa where n is the number of replications per sample. Likewise, it can be determined by considering the expected mean squares of Table 1.6 that the tests for no significant variability among animals or among days (Ho : u; 0 or Ho : ua 0) are made using MSad =

=

in the denominator of the F statistic. The results of Table 1.6 show that the hypotheses regarding u� and uaa were significant but that regard­ ing ua was not significant. This means that u� and u.a are not equal to zero, indicating significant sources of variability among animals that are attributable to the interaction of animals and days in the plasma cortisol levels recorded in the experiment. The quantity ua, on the other hand, may be equal to zero. Figure 1.7, a graphical presentation of the results obtained from the analysis of variance, gives each ani­ mal's responses (mean of the two determinations) for the three days during which responses were recorded. We can see that for some dogs the response decreased over time, for some the response increased over time, for some the response at day 2 was depressed com­ pared to that at day 1 but then at day 3 was elevated over that observed at day 2, and for others the response at day 2 was elevated compared to that at day 1 but then at day 3 was depressed compared to that observed at day 2. Differing degrees of elevations and depres­ sions were observed with these basic patterns. Figure 1.7 shows that there was a large degree of variability among dogs in the plasma cortisol values on any given day, and that the plasma cortisol profiles across the three days that samples were taken were not parallel across the dogs. This lack of parallelism was identified by the significant interaction in the ANOVA; signifi-

F value

Expected mean square

3.32 0.51 117

� + nndu; + nu.a � + nn,ITa + nIT J � + nuJ �

cance indicated that the observed departures from par­ allelism were larger than would have been expected as being a result of chance. The mean plasma cortisol levels recorded on the three days were 3.25 JLg/dl (89.7 nmol/liter), 3.36 JLg/dl (9.27 nmol/liter), and

3.12 JLg/dl (86.1 nmol/liter) for a range of 0.24 JLg/dl (6.6 nmoilliter). This range is very small relative to the range of mean response of 3.00 JLg/dl (82.8 nmol/liter) observed for the 20 dogs (X max = 4.98 JLg/dl or 137.4 nmollliter and x min = 1.98 JLg/dl or 54.6 nmol/

Thus, although the responses of some dogs in­ creased while those of other dogs decreased from one day to the next day, there was practically no change in the overall variability from day to day. This is reflected in the ANOVA by a nonsignificant variance component attributable to the factor "days." The large variability among animals in mean plasma cortisol is identified in the ANOVA by the rejection of the hypothesis that ui = O.

liter).

b. Estimating Variance Components

Once significance has been established for one or more of the variance components in an experimental deSign, interest focuses on estimating the variance component(s). Estimates are readily obtainable using J.l9/d l 5.0

Gi 4.0 > CD ...J

� 0 "5

3.0

0

2.0

�II)

009 18

009 19

Dog 19

Dog 10

IG

c:: 1.0

Day 1

Day

2

Day

3

FIGURE 1.7 Interaction between "animal" and "day" factors in the canine plasma cortisol level (p,g/ dl) based on data from 20 dogs in Table 1.5.

17

Concepts of Normality in Clinical Biochemistry the appropriate expected mean squares in conjunction with the mean squares obtained from the data. For example, Table 1.6 shows that an estimate of u; is MS. (0.0096). An estimate of u.a can be obtained by noting that MS.d estimates nu.a + u;; solving for uaa followed by substitution of MS. for u; yields (MS.d - MS.)/ n. as the estimate. Based on the data used in the example, iTaa (1.1231 - 0.0096)/2 = 0.5567. Estimates of u� and ua, obtained in a similar manner, are (MS. - MSad ) / ndn (0.4342) and (MSd - MSad) / n.n (-0.0136 or zero), respectively, where na is the number of animals sam­ pled (here 20) and nd is the number of days samples were taken (3). Interval estimates for these variance components can be obtained using these point esti­ mates by methods described elsewhere (Dunn and Oark, 1987; Harter and Lum, 1955; Satterthwaite, ==

1941). c. Estimating the Variance of the Grand Mean Response Another way to visualize the importance of these variance components is to analyze their impact on the estimate of the variance of interest. In some applica­ tions, there would be interest in estimating the grand mean (p,) of the response. In the present example, this would involve estimating the mean plasma cortisol level taking into account any random animal effect, day effect, and the interaction between the animal and day effects. If it is assumed that the main effects, "days" and " animals," are independent of their inter­ action in the response model and other distributional assumptions hold (Dunn and Clark, 1987), then the variance of p." Var(p.,), is given as u; / na + uV nd + uaa / nand + u; / nandn (Dunn and Clark, 1987). In the present example, Var(p.,) is estimated as 0.4342/

20 + 0.0/3 + 0.5567/ 60 + 0.0096/ 120 0.0217 + 0.0000 + 0.0093 + 0.0001 = 0.0311, and by far the ==

greatest contribution to this variance is that due to the variability among animals in their response followed by the contribution of interaction to the response.

d. Estimating the Total Variability of a Single Response In other applications, interest centers about the total variability (utotaD associated with a single response. A single response is a linear combination of the terms in the response model and, using the assumption of the independence of terms in the model, has a variance equal simply to the sum of the variance components. Specifically, utoJ = u� + ua + uaa + u�, which in this example is estimated as 0.4342 + 0.0000 + 0.5567 + 0.0096 = 1 .0005. In this example, the total variability of a single response is determined by two variance components: (1) that associated with the animal vari­ ability and (2) that resident in the interaction between "animals" and "days."

e. Experimental Design 2: One Factor Nested within a Second Factor

A second experimental design that would provide estimates of intra-animal and interanimal variances in the plasma cortisol level is as follows: The 20 dogs are randomly selected over a period of time from the population of dogs identified by a dinic as having pituitary-dependent hyperadrenocorticism. At the time the dog is selected, three days within the next two weeks are randomly selected for plasma cortisol samples to be taken at 1 1:00 A.M. from the dog. Again plasma cortisol assays are replicated for each sample. Because the animals are sampled across time, it is rather unlikely that many and perhaps any of the three days that samples are drawn will be common for the 20 dogs. If the 20 dogs had no sampling days in common, samples would be taken on 60 (20 X 3) different days. Thus, the factor "days" in this design is not crossed with the factor "animals" but is said to be nested within "animals." The total variability in this experiment can be decomposed by the method of ANOVA into three parts, that which will identify significant variability among animals and significant variability among days (nested within animals), and that which quantifies the residual variability. The data in Table 1 .5 were used again assuming that it had been generated from the nested design de­ scribed. Table 1 .7, the ANOVA table, shows that two hypotheses can be tested. The mean square error is the denominator for testing that the variance component for "days nested within animals" (d/a) is equal to zero, that is, Ua/a = O. This hypothesis is rejected based on the data. In this design, Ua/a is the true intradog vari­ ability for this response. The test result that the day effect is significant means that this intradog variability is larger than zero. The second test is that the variance component for "animals" is equal to zero or u; O. The expected mean squares given in Table 1.7 show that the denominator for this test is the mean square for days nested within animals, MSd/a. This test is also significant. The interpretation of a significant u� is the same as that noted for design 1. Once significance has been established, estimates of these variance components can be computed using the method described earlier for design 1 . These estimates 0.0096. The are iT� = 0.4387, iTa I. = 0.5431 and iT� term iT� is the estimate of the interanimal variability, whereas iTa/a estimates the intra-animal variability in plasma cortisol level for the underlying population of dogs. For those interested in estimating the grand mean plasma cortisol level, the importance of each of these components is obtained by observing their impact on the estimate of the variance of the estimate of the grand ==

=

18

Thomas B. Farver TABLE 1.7

Analysis of Variance of Canine Plasma Cortisol Level Data Assuming Experimental Design 2 ("Days" Nested within " Animals")

Source of variation

df

Sum of squares

Mean square

Among animals Days! animals Error Total

19 40 60 119

70.8316 43.8333 0.5750 115.2399

3.7280 1.0958 0.0096

mean, Var([L). The expression for this variance is

Var({L) = a�/ na + ad/�/ nand + a;/ nandn (Little et al., 1991; Neter et ai., 1996), which in the present is esti­ mated as 0.0219 +

0.0091 + 0.0001 = 0.0311 .

Again by

far the greatest contribution to this variance is that due to the variability among animals in their response. The intradog variance component, although slightly larger than the interdog variance component, makes a consid­

erably smaller impact on Var(il). For those interested in estimating the total variability of a single response, �otal = a� + ad /� + a; (Kringle, 1994), which in the present example is estimated as 0.4387 + 0.5431 + 0.0096 = 0.9914. Here the total variability of a single response is divided nearly equally between " animals" and "days nested within animals." Other possible de­ signs could be considered. What has been demon­ strated is that the method of analysis of variance in conjunction with experimental design can be useful in answering a variety of questions. Design 1 allowed an assessment of how parallel the plasma cortisol profiles of the dog population are across time. It also allowed a determination of whether there are major shifts in the daily overall variability in the plasma cortisol level in light of changes in individual animal responses across time. This information is not provided by design rather design 2 provides an estimate of the true intradog variability in the plasma cortisol level. Both

2;

designs provide an estimate of the interdog variability in the response. Design 2 is frequently used to assess sources of vari­ ability in an assay. For example, several laboratories could be involved in doing a particular assay, with several autoanalyzers in each laboratory and multiple technicians running these autoanalyzers. Inference in this context centers around being able to identify if there are significant sources of variation among the laboratories, among auto analyzers within a given labo­ ratory, and among technicians operating a given auto­ analyzer. The goal of analyses of this sort is to identify large sources of variability. Once the larger sources of variability have been identified, changes are made in the system in an effort to reduce the variability associ­ ated with each source. The long-term objective is to have an assay with sources of variability that are as

F value

389 114

Expected mean square

cr. + nndu; O"� + nUd/� u;

+

nu d/;

small as possible. Clinical analysts conventionally di­ vide the square root of the estimates of the variance components (the sample standard deviations) resulting from such assay experiments by the grand mean to obtain coefficients of variation for each source of vari­ ability (Kringle, 1994). These coefficients of variability should be much smaller than those derived as intra­ animal and interanimal variability. Interested readers are strongly encouraged to consult texts written on experimental design and ANOV A (Dunn and Clark, 1987; Neter et ai., 1996).

References Bjorkhem, I., Bergman, A., and Falk, O. (1981). Clin. Chem. 27, 733-735. Cochran, W. C. (1977). "Sampling Techniques," 3rd ed. Wiley, New York. Daniel, W. W. (1995). "Biostatistics: A Foundation for Analysis in the Health Sciences," 6th ed. Wiley, New York. Dixon, W. J., and Massey, F. J. (1983). "Introduction to Statistical Analysis," 4th ed. McGraw-Hill, New York. Dunn O. J. (1977). "Basic Statistics: A Primer for the Biomedical Sciences," 2nd ed. Wiley, New York. Dunn O. J., and Clark, V. A. (1987). "Applied Statistics: Analysis of Variance and Regression," 2nd ed. Wiley, New York. Elevitch, F. R. (1977). Proc. 1976 Conf Analytical Goals in Clinical Chemistry, College of American Pathologists, Skokie, illinois. Engleman, L. (1992). In "BMDP Statistical Software Manual" (W. J. Dixon, ed.), Vol. 1, pp. 141-153. University of California Press, Berkeley. Feinstein, A. R. (1977). "Clinical Biostatistics." C.V. Mosby, St. Louis, Missouri. Fisher, R. A. (1966). "The Design of Experiments," 8th ed. Oliver and Boyd, Edinburgh. Gilbert, R. K. (1975). Am. J. c/in. Pathol. 63, 960-973. Gilbert, R. K. (1978). Am. J. c/in. Pathol. 70, 450-470. Harris, E. K. (1988). Arch. Pathol. Lab. Med. 112, 416-420. Harter, H. L., and Lum, M. D. (1955). "Partially Hierarchical Models in the Analysis of Variance." Wright Air Development Center Technical Report No. 55-33, Wright-Patterson Air Force Base, Ohio. Herrera, L. (1958). J. Lab. c/in. Med. 52, 34-42. Jessen, R. J. (1978). "Statistical Survey Techniques." Wiley, New York. Kaneko, J. J. (1977). J. Small Anim. Pract. 18, 85-94. Kleinbaum, D. G., Kupper, L. L., and Muller, K. E. (1988). "Applied Regression Analysis and Other Multivariable Methods," 2nd ed. PWS-Kent, Boston. ,

,

Concepts of Normality in Clinical Biochemistry Kringle, R. O. (1994). In "Tietz Textbook of Clinical Chemistry" (c. A. Burtis and E. R. Ashwood, eds.), 2nd ed., pp. 384-453. Saunders, Philadelphia, Pennsylvania. Levy, P. S., and Lemeshow, S. (1991). "Sampling of Populations: Methods and Applications." Wiley, New York. Little, R. c., Freund, R. J., and Spector, P. C. (1991). "SAS System for Linear Models," 3rd ed. SAS Institute, Cary, North Carolina. Mainland, D. (1963). "Elementary Medical Statistics," 2nd ed. Saun­ ders, Philadelphia, Pennsylvania. Massod, M. F. (1977). Am. J. Med. Technol. 43, 243-252. Murthy, M. N. (1967). "Sampling Theory and Methods." Statistical Publication Society, Calcutta. Neter, J., Wasserman, W., and Kutner, M. H. (1996). "Applied Linear Statistical Models," 4th ed. Richard D. Irwin, Chicago, illinois. Raj, D. (1968). "Sampling Theory." McGraw-Hill, New York. Raj, D. (1972). "The Design of Sampling Surveys." McGraw-Hill, New York. Reed, A. H., Henry, R. J., and Mason, W. B. (1971). Clin. Chern. 17(4), 275-284.

19

Remington, R. D. and Schork, M. A. (1985). "Statistics and Applica­ tions to the Biological and Health Sciences," 2nd ed. Prentice Hall, Englewood Cliffs, New Jersey. Ryan, B. F., Joiner, B. L., and Ryan, T. A. (1985). "Minitab Handbook." PWS-Kent, Boston. Sattherthwaite, F. E. (1941). Psychometrika 6, 309-316. Scheaffer, R. L., Mendenhall, W., and Ott, L. (1996). "Elementary Survey Sampling," 5th ed. Duxbury Press, Belmont, California. Shultz, E. K. (1994). In "Tietz Textbook of Clinical Chemistry" (c. A. Burtis and E. R. Ashwood, eds.), 2nd ed., pp. 485-507. Saunders, Philadelphia, Pennsylvania. Snedecor, G. W., and Cochran, W. G. (1989). "Statistical Methods," 8th ed. The Iowa State University Press, Ames. Solberg, H. E. (1994). In "Tietz Textbook of Clinical Chemistry," (c. A. Burtis and E. R. Ashwood, eds.), 2nd ed., pp 454-484. Saunders, Philadelphia, Pennsylvania. Steel, R. G. D., and Torrie, J. H. (1980). "Principles and Procedures of Statistics," 2nd ed. McGraw-Hill, New York. Zar, J. H. (1996). "Biostatistical Analysis," 3rd ed. Prentice Hall, Upper Saddle River, New Jersey.

C H A P T E R

2 DNA Technology in Diagnosis, Breeding, and Therapy JENS G. HAUGE

I. INTRODUCTION

III.

IV.

V.

VI.

VII .

to a large extent been the result of the development of new methods to manipulate, multiply, and study DNA fragments. The same new methods have a number of important applications in human medicine and, gradu­ ally, also in veterinary medicine. For good introduc­

21

II. MOLECULAR GENETICS

21

A. Structure and Replication of Genes 21 B. From Gene to Gene Product 23 GENE MANIPULATION 24 A. Basic Features 24 B. Methods and Tools 24 C. Vectors 27 D. Studies of Gene Structure 27 STUDIES OF METABOLIC AND GENE REGULATION 30 A. Tissue Culture Studies 30 B. Regulation in Transgenic Animals 30 DIAGNOSIS OF GENETIC DISEASE 31 A. Gene Changes in Hereditary Disease 31 B. Detection of Gene Changes with DNA Probes C. DNA Probes and Disease Susceptibility 34 D. Application to Domestic Animals 35 E. Gene Changes in Cancer 38 THERAPY OF GENETIC DISEASES 38 A. Germ Line Gene Therapy 39 B. Somatic Gene Therapy 39 USE OF RECOMBINANT DNA METHODS TO IMPROVE DOMESTIC ANIMALS 39 A. Markers and Maps 39 B. Parentage Identification 40 C. Transgenic Animals 40 References 42

tions, see Old and Primrose (1990) and Watson et al., (1992). An agricultural perspective is given in the book edited by Hansel and Weir (1990). These surveys also discuss the large impact on microbiological diagnosis and vaccine production, as well as hormone produc­ tion, aspects that are not covered here. The development of gene technology rests on the foundations of molecular biology laid during the pre­ ceding 40 years, with the discovery of the double heli­ cal structure of DNA and the breaking of the genetic code as the major milestones. This chapter starts, there­ fore, with a brief review of the molecular biological basis of gene technology.

31

II. MOLECULAR GENETICS A. Structure and Replication of Genes The notion that heredity was linked to matter, to certain cellular structures, appeared for the first time in 1903 when the chromosomal theory of heredity was formulated. Chromosomes were recognized to have properties that could accommodate the hereditary units discovered by Mendel in his experiments. A bridge was built between cytology and genetics. But the nature of the molecules that carried the hereditary information was not thereby identified. Many felt that chromosomal protein was the most likely candidate.

I. INTRODUCTION During the last 20 years an immense expansion has taken place in our understanding of the structure, orga­ nization, and regulation of genes. This expansion has

CUNICAL BIOCHEMISTR Y OF DOMESTIC ANIMALS, FIFTH EDITION

21

Copyright © 1997 by Academic Press All rights of reproduction in any form reserved.

22

Jens G. Hauge

In 1944, however, Avery and coworkers showed con­ vincingly that genes in pneumococcal bacteria are mol­ ecules of deoxyribonucleic acid (DNA). It soon became apparent that pneumococci were no exception in this respect. In animals and plants, as well as in bacteria, hereditary information is carried by DNA. A section of a DNA molecule has the appearance shown in Fig. 2.1a. DNA is built in the form of a chain, with phosphoric acid and the sugar deoxyribose alter­ nating as the links. Phosphoric acid always connects the 5'-OH group in one deoxyribose to the 3'-OH in the next deoxyribose. The heterocyclic bases adenine, guanine, cytosine, or thymine are bound to the l ' car­ bons in deoxyribose. The base-deoxyribose moiety is a nucleoside, and the repeating unit in the chain, the combination base-deoxyribonucleoside-phosphoric acid, a nucleotide. It is impractical to write DNA structures with all the atoms, so simplified writings have come into use. If the bases from the top in Fig. 2.1a are adenine, thy­ mine, and guanine, the most used abbreviation is sim­ ply ATG. The first letter in the name of the base is used to designate the corresponding nucleotide. That such a chainlike DNA molecule could carry hereditary information, genes, was not difficult to con-

a

ceive. Most genes determine the construction of pro­ teins, which are also chainlike molecules, with amino acids making up the links in the chain. One needs only to postulate a one-to-one relationship between an amino acid and a group of DNA bases. The other fundamental property of the hereditary material, that it is accurately duplicated and distributed on both daughter cells when cells divide, was harder to visual­ ize on the basis of the DNA structure, as given in Fig. 2.1a. Watson felt that the solution to the problem perhaps lay hidden in the three-dimensional structure of DNA, and together with Crick he set out to elucidate this structure. The result was the discovery in 1953 of the double-helix structure of DNA. The double helix is built as shown schematically in Fig. 2.1b. A, T, G, and C represent the bases, which point toward each other and are bound to each other through hydrogen bonds. The DNA strands of the double helix have op­ posite directions (a DNA chain has a 5'-end and a 3'­ end, see Fig. 2.1a). The size of the bases and their hydrogen bonding properties make only two pairings possible, namely, A with T and G with C. One tum of the standard double helix has 10 base pairs (bp) and covers a length of 3.4 X 10-6 mm. The diameter of the double helix is 2 X 10-6 mm. It is so slim a structure

b

I

- O - P=O

I o

�I" �

t

H2 s

I Base I

0

H

H

H

3"

o I

� I

- O - P= O o

H 2C H

H

O

H

H

I

o

B ase

H

�

- O - P= O

I o

H 2C H

H I

o

O

H

H

Base

H

FIGURE 2.1 (a) Structure of part of a DNA chain. (b) DNA double helix during replication. Reproduced from Stryer (1981) and Freifelder (1983), with pennission.

�

23

DNA Technology in Diagnosis, Breeding, and Therapy

that if a DNA molecule could reach from the Earth to the moon, it still would weigh only 1 mg ! The model for the DNA molecule at which Watson and Crick (1953) arrived can in a simple way explain how DNA molecules are dupli cated, that is, the process of DNA replication. When the double helix is consid­ ered ( Fig. 2.1b), we can see that the two strands corre­ spond to each other. The base pairing rules determine which base in one strand must be opposite a given base in the other strand, and the strands are said to be complementary. When replication starts, the hydrogen bonds between the two strands are gradually broken (Fig. 2.1b) . New DNA chains are constructed from the four deoxynucleoside triphosphates with the help of the enzyme DNA polymerase, one molecule of pyro­ phosphate being eliminated per nucleotide laid down. Step by step, two new strands, complementary to the original strands, are synthesized . The original strands serve as templates in the process, and the result is two new double helixes that are identical to each other and to the original DNA molecule. The basic mechanism for DNA replication is simple and elegant . The practical execution of this process in the cells, however, involves a large number of proteins, including some additional enzymes. The unwinding of the DNA molecules causes problems, as does the fact that DNA polymerase lays down a new strand from the 5 ' -end toward the 3 '-end only . Furthermore, for this synthesis to start, a primer nucleic acid is needed for the strand to grow. One important auxiliary enzyme is DNA ligase, which links DNA molecules. If DNA is heated to 80-90°C, the molecule will dena­ ture, and the strands will separate. If the temperature is lowered somewhat, the complementary strands will find each other again and reform the double helix. This property is widely exploited in gene technology (Section III.B .1).

B. From Gene to Gene Product When the information stored in DNA is used, organ­ isms utilize another type of nucleic acid, namely, ribo­ nucleic acid, RNA. RNA differs from DNA in that it contains ribose instead of deoxyribose, and uracil instead of thymine. Uracil has the same base pair­ ing properties as thymine . RNA does not normally exist in double-helix form, but RNA may form short stretches of double helix when the RNA strand folds back on itself locally . A DNA strand and a complemen­ tary RNA strand will form a double helix. RNA is made by transcription from one of the DNA strands in a double helix (Fig. 2.2). The DNA strand is used as template in a process somewhat similar to DNA replication. The DNA double helix is opened

�1-'-r-1-l.1-'-.. 1 .L.1- l-1-...l r...l.-I..I.1 .-.J

RNA

[[[[

ANA polymerase

FIGURE 2.2

•

I

RNA synthesis (transcription).

locally, and an RNA strand complementary to the tem­ plate is synthesized with the help of the enzyme RNA polymerase, ribonucleoside triphosphates being used as activated building blocks. Special sequences of bases in front of the genes (promoters) are recognized by the polymerase, causing it to start at the right place and on the right DNA strand. At the other end of the genes, there are termination signals. In all cells there are three main types of RNA : mes­ senger RNA, ribosomal RNA and transfer RNA. Mes­ senger RNA (mRNA) carries the information speci­ fying the amino acid sequences in the proteins. Ribosomal RNA (rRNA) is a group of RNA mole­ cules that forms part of the structure of the protein­ synthesizing machinery of the cells, the ribosomes. Transfer RNA (tRNA) participates in protein synthesis (translation) by carrying amino acids to the ribosomes. The mRNA type is of special importance. For each kind of protein chain the cell makes, there is a corre­ sponding mRNA. The genetic coding system that bio­ chemists discovered between 1960 and 1965 showed that organisms use groups of three nucleotides, triplets, to code for amino acids. It is, therefore, the sequence of triplets in mRNA that directly determines the se­ quence of amino acids in the corresponding polypep­ tide chain. The coding group could not have consisted of only two nucleotides. That would have given only 42 = 16 combinations, and 20 amino acids need coding . But 43 = 64 gives more than enough . Nature has chosen this most economical solution, and done it in a way so that 61 of these 64 triplets code for some amino acid . Almost all the amino acids, therefore, have more than one codon. When ribosomes directed by mRNA link together the amino acids, it is crucial for the result that the ribosomes start at the correct position on the mRNA. Starting one nucleotide too early or late will give a completely different protein. A correct start is secured by having all protein synthesis start with the triplet AUG, which codes for methionine. This means that all proteins initially are synthesized with methionine as their first amino acid. In posttranslational processing, some proteins lose this initial methionine, however . Because methionine also occurs internally in proteins, something more is needed for a correct start. In bacte-

24

Jens G. Hauge

ria, the ribosomes recognize start AUG by the presence of a short stretch of purines a limited distance before the AUG. In animals and plants, the ribosomes choose the AUG that is closest to the 5'-end of the mRNA. The coding triplets UAA, UAG, and UGA do not code for any amino acid, and these codons are used as termination signals. The amino acids do not by themselves have the ability to recognize their codons on mRNA. They must, furthermore, be activated in order to form peptide bonds with each other. Both of these problems are solved by each amino acid becoming bound to a tRNA by means of a specific activating enzyme, utilizing A TP. tRNA has in its structure an anticodon, a set of three nucleotides that forms base pairs with the codon on mRNA for the corresponding amino acid. On the ribosome surface the amino acids leave their tRNA carriers as they are linked together to form polypeptide chains. Briefly then, the information flow is as illus­ trated here for the start of a hypothetical gene: 3' . . TAC AAA AGT GAC TGT CCG . . . 5' DNA . 3' 5' . . . ATG TIT TCA CTG ACA GGC .

5' . .

.

t transcription

AUG

UUU

UCA CUG ACA GGC

Met

Phe

Ser

t translation

Leu

Thr

.

.

.

.

. 3' mRNA

Gly

protein

DNA to be cloned

-

(0

0

!

o!

Opened plasmid

Recombinant DNA

host cells

Into

!

Host chromosome

Growth and selection

a

(0 0 (] (0 0 (] (0 0 (] FIGURE 2.3

Basic steps in cloning of DNA in a plasmid.

amounts that it is possible to study the DNA in the electron microscope, determine its base sequence, in­ ject it in fertilized ova, or use it as a diagnostic probe, to mention some important uses. As is often the case, when a homogeneous DNA preparation is not available as starting material for cloning, but rather a mixture of many types of DNA pieces, special selection techniques are required after­ ward to find the interesting clone or clones. Plasmids are but one type of vector for DNA cloning. Viruses from bacteria or animals are also used quite often.

III. GENE MANIPULATION B. Methods and Tools

A. Basic Features The core of the new gene manipulation techniques is that one can join two pieces of DNA to form a recom­ binant DNA molecule, which then is multiplied, or cloned, in a suitable host organism so that a sufficient amount of the piece of DNA desired is obtained. The one piece of DNA is often used as a plasmid, a ring­ shaped, extrachromosomal DNA molecule that occurs in many bacteria and carries genes, for example, for antibiotic resistance. Figure 2.3 illustrates molecular cloning of a piece of DNA with the help of a plasmid vector. The plasmid vector is opened with the help of a special endonuclease and joined to the foreign DNA with the help of another enzyme. Under proper condi­ tions, some of the bacteria will take in the recombinant DNA molecule, which will be replicated as the bacteria grow and multiply. Most of the bacteria will not have taken in any plasmid. A selection process is therefore necessary, usually based on the plasmid carrying an antibiotic resistance gene. From the bacteria, the plas­ mid and the foreign DNA piece can be isolated in such

1. Nucleic Acid Hybridization

Nucleic acid hybridization is fundamental both in basic and applied gene technology. This is the process whereby sequence-specific DNA : DNA or RNA : DNA duplexes are formed from single-stranded nucleic acids. For most uses, one of the components in the hybridization is radioactively labeled, or labeled by other means, so that the hybrid can be identified. The labeled molecule is referred to as a probe and is used to probe nucleic acid mixtures for the presence of com­ plementary sequences (Fig. 2.4). The factors affecting the stability of the hybrid and the rate of the hybridization reaction are fairly well known (Marmur and Doty, 1961; Nygaard and Hall, 1964; Young and Anderson, 1985). The stability de­ pends on the temperature, the ionic strength of the medium (high ionic strength reduces the repulsion be­ tween the negatively charged phosphate groups and thus stabilizes the duplex), the presence of formamide (which breaks hydrogen bonds), the G + C content of

DNA Technology in Diagnosis, Breeding, and Therapy Sample DNA

1

I I I I II I !! ! ! ! !! II ! !

Denature

- -n�

__ __ Bind

to

support

FIGURE 2.4

the probe (G and C rich probes are bound more tightly because there are three hydrogen bonds linking G to C, compared to two bonds between A and T), and the length of the probe. The temperature at which the duplex is 50% denatured is called its Tm, which can be calculated from the following formula: Tm

=

Positive signal

Labeled probe

ill ill

81SC + 16.6 log M + 0.41(%G + %C) - 500/ n - 0.61 (% formamide).

This formula assumes perfect complementarity be­ tween the hybridizing strands. With mismatches pres­ ent, the Tm is reduced by about 1°C per 1 % mismatch. The factors affecting stability also affect the rate of hybridization. The rate is optimal about 25°C below Tm. For a reaction in solution, the rate is proportional to the concentrations of the reacting species. When the DNA or RNA to be investigated is immobilized on a membrane, which is often the case, the reaction is slower by a factor of 7 to 10, but the general effects of the factors listed earlier are the same. The reaction conditions are manipulated to give hy­ bridizations at a desired level of stringency. High strin­ gency, that is, high temperature and/ or low ionic strength, will allow only the most stable hybrids, those having perfect complementarity, to form. In certain situations, one may want to probe for a family of re­ lated DNA molecules with some sequence differences, and hybridization at a lower stringency will be chosen. The same may be necessary when using a probe from one species to look for a homologous gene in an­ other species. Several methods are in use for labeling a probe with radioactivity (Arrand, 1985). In our experience, the ran­ dom primer method (Feinberg and Vogelstein, 1983) is a convenient way of obtaining high specific activity, better than the so-called nick translation method. The probe DNA is denatured, and both strands are used as template for DNA synthesis, using short random oligonucleotides as primers and one or more of the triphosphates being labeled with 32P. Nonradioactive labeling is also used because it has certain advantages for routine, diagnostic purposes. Attaching biotin to one of the bases (Leary et al., 1983) has received particular attention. The biotinylated

25

r �

Hybridization.

,,�

probe is localized with streptavidin and a biotin­ coupled enzyme for which a substrate is used that gives a colored precipitate. Nonradioactive probes have the drawback of multistep detection procedures and a somewhat lower detection sensitivity than radio­ active probes (Syvanen, 1986).

2. Restriction Endonucleases Restriction endonucleases of type II, usually re­ ferred to as restriction enzymes, are a major tool in gene analysis and gene manipulation. These enzymes bind to particular sequences on DNA and cut both strands within this DNA site. The majority of the enzymes have sites of tetra-, penta-, hexa-, or hepta­ nucleotides which have an axis of rotational symmetry (Fig. 2.5a). Enzymes corresponding to more than 150 different sites have been discovered so far. The names of the enzymes have been constructed from names of the bacteria and the particular substrain from which they have been isolated. The arrows in Fig. 2.5a indicate where the DNA strands are cut by these five enzymes. Treatment of a DNA molecule with a restriction en­ zyme yields a reproducible set of fragments. An en­ zyme with a recognition site of four nucleotide pairs will cut relatively often, statistically once every 44 (i.e., 256) nucleotide pairs, whereas an enzyme with a hexa­ nucleotide recognition site will cut less frequently, once every 46 (i.e., 4096) nucleotide pairs. Figure 2.5b shows the pattern of fragments into which three different restriction enzymes cut the SV40 virus chromosome. After enzyme treatment, the fragments were separated by electrophoresis in an agarose gel and made visible by binding of ethidium bromide, which fluoresces in ultraviolet light. The size of the fragments can be found by running molecular weight markers in the same gel. Short fragments move farther, and long fragments move a shorter distance. The fragments produced by the restriction enzymes will, for most enzymes, have single-stranded, comple­ mentary end sections, because the cuts are not made straight across (Fig. 2.5a), and the fragments are said to have sticky ends. If a plasmid is opened with the same restriction enzyme that has been used to produce

26

Jens G. Hauge a

L

b

5' G G A T C C 3' • 3' C C TA G G 5'

BamH I

S' GAAT T C 3' • 3' C T TAAG 5'

EcoR I

f

L

f

L 5' G G C C 3 ' • 3 ' C C G G 5' f

Hha I

5' C T C G AG 3 ' • 3' G AG C T C 5'

Xho I

f

B

c

Hae I I I

L 5' G C G C 3 ' • 3' C G C G 5' f L

A

(a) Specificities of some restriction enzymes, (b) Gel electro­ phoresis pattern showing SV40 DNA fragments produced with each of three restriction enzymes, Reproduced from Stryer (1981), with permission,

FIGURE 2.5

the DNA fragments to be cloned, this DNA will have sticky ends of a type that allows its insertion in the plasmid. Fig. 2.6a shows what the situation would have been in Fig. 2.3 if the enzyme EcoRI had been used.

3. DNA Ligase

NAD. The ligation reaction is carried out at a low temperature, 4-15°C, in order to increase the stickiness of the sticky ends.

4. Tenninal Deoxynucleotidyltransferase

To fonn a stable, recombinant DNA molecule from two molecules, the strands of these two molecules must be joined. This is effected by the enzyme DNA ligase, which requires for its action a 5'-phosphate and a 3'­ OH group. The enzyme from phage T4 uses ATP as energy donor, whereas the Escherichia coli enzyme uses

� r� r-

GIA-A-T-T-C C-T-T-A-AIG

G-G-G-G-G C-C-C-C-C

Tenninal Deoxynucleotidyltransferase, or briefly, tenninal transferase, is useful when the object is to insert in a plasmid a DNA molecule that has not been produced with the help of a restriction enzyme, a fre­ quently occurring situation. With this enzyme, tails of identical nucleotides can be attached to the 3'-end of DNA strands. The plasmid is opened with a suitable

G IA-A-T-T-C C-T-r-A-AIG

c-c-c-c-c

� j � )

G-G-GoG

FIGURE 2.6 (a) Fitting a piece of DNA into an EcoRI site of a plasmid, (b) Fitting a piece of DNA into a plasmid after tailing the plasmid with G and the insert with C by using terminal transferase.

DNA Technology in Diagnosis, Breeding, and Therapy restriction enzyme and furnished with a tail of about 20 G residues, whereas the DNA to be inserted is fur­ nished with a C tail (Fig. 2.6b).

27

D. Studies of Gene Structure Comprehensive solutions to a wide range of biologi­ cal problems require knowledge of the structure of

5. Adaptors

the gene and regulatory sequences that govern the particular biological phenomenon.

Use of an adaptor DNA containing a restriction site is a common alternative strategy when a DNA mole­ cule is to be inserted in a plasmid opened with the corresponding restriction enzyme. A piece of DNA, 6 to 10 base pairs long and containing the desired restriction site, is synthesized or purchased. The ends of the DNA to be inserted are first made blunt by filling in with DNA polymerase. All 5' -ends are phosphory­ lated with the enzyme, polynucleotide kinase, and the adaptor is attached to both ends of the insert DNA with DNA ligase. Treatment with the restriction en­ zyme then produces the proper sticky ends for inser­ tion in the plasmid. This method illustrates well the role of both chemistry and enzymology in recombinant DNA technology.

C. Vectors Through stepwise modifications of natural

E. coli

plasmids, new plasmids have been constructed that are safe and convenient for different types of gene manipulation. One much used plasmid is pBR322 (Fig. 2.7a). This plasmid carries two antibiotic resistance genes, one for tetracycline and one for ampicillin. Within these two genes, there are several restriction sites for enzymes that have only these sites in the pBR322 molecule. If one uses the BamHI site to insert a foreign DNA, the tetracycline resistance will be elimi­ nated. It is then possible to select for bacteria that have taken up the plasmid by using ampicillin in the medium, and then finding those that have an insert in the BamHI seat by investigating which bacterial colo­ nies have lost tetracycline resistance. This approach is useful since there can be some regeneration of the original plasmid in the ligation step. Lambda virus DNA has also been modified so that it is suitable as a vector. Three of its five EcoR! sites have been removed. The DNA between the two re­ maining sites is not necessary for the infection process and can be exchanged with a foreign piece of DNA of 10-20 kilobase pairs (kbp). The recombinant DNA is packed in vitro using lambda coat and tail proteins. These phage particles enter E. coli bacteria very effi­ ciently and reproduce as in a normal lytic infection. For other bacteria, yeast, and animal and plant cells, other vectors exist. Shuttle vectors, which have speci­ ficity for two different hosts, have also been con­ structed.

1. eDNA Cloning For studies of genes of viruses and bacteria, cloning can start with fragments of their chromosomal DNA. With genes from animals and plants, the situation is different. Here the genome is so large and contains so many noncoding sections that often one must start with mRNA molecules and preform what is called a copy DNA (cDNA) cloning. Total RNA is isolated with chemical methods from a tissue of interest, and mRNA adsorbed on a column of poly(T)-cellulose. mRNA is bound due to the 3' -tail of poly(A), which mRNA possesses in higher organisms. mRNA is eluted and used as the template for RNA-directed DNA polymer­ ase, so-called "reverse transcriptase." This enzyme occurs in retroviruses, where its role during viral infection is to produce double-stranded DNA corres­ ponding to the viral RNA genome. In principle, the same reaction can take place in vitro, and a mixture of cDNA is formed, corresponding to the mixture of mRNA isolated. After attachment of sticky ends, the mixture of cDNA molecules can be inserted in a plasmid. If the bacteria to be transformed have been made competent by a special treatment, 105 to 107 transformed bacteria per microgram of recombinant plasmid can be ex­ pected. The clones from these bacteria make up a cDNA library for that particular tissue. The library will have cDNA clones corresponding to different mRNA in the tissue. Clones for the particular gene of interest can be found by various means. The simplest situation exists if something is known about the amino acid sequence of the protein for which the gene codes. Then, a DNA probe of 17 to 19 bases, corresponding to a suitable hexapeptide, can be syn­ thesized. This probe will be long enough that the search will be specific at high stringency. The probe prepara­ tion will have to be a mixture of DNA molecules, how­ ever, since most amino acids have more than one co­ don. An imprint of the bacterial colonies is made on a nitrocellulose filter, the cells are opened, and their DNA is denatured. If a radioactive probe is used, it will hybridize to a few colonies, those containing cDNA corresponding to the gene of interest. Plasmid DNA is purified, treated with the proper restriction enzyme, and the cDNA isolated from a hybridizing band on an agarose gel after electrophoresis.

28

Jens G. Hauge b

� tl�I:"OP"' ';' '" .0

t d" ;," '"'Ym'

� a

EcoRl

dry paper towels

agarose gel

I

Blot t i ng

"'.lt �'�r���/ �� ? ! n o ,

buffer pBR322

cellulose nit rate f i lter

:Z:::-- .J I

�

gel containing denatu red DNA

filter paper

Hybrid ization with

l2

PcDNA

& Autorad i ography

(a) The cloning vector pBR322 (4362 bp). Amp is the gene for ampicillin resistance, and Tet is the gene for tetracycline resistance. (b) The Southern blot technique. The symbol x indicates the gene or sequence of interest. Reproduced from Emery (1984), with permission.

FIGURE 2.7

was divided into eight pieces,

2. Gene Cloning A collection of clones, which includes all DNA in an organism, is called a

genomic library

and is usually

made in a lambda vector. DNA from the organism is subjected to incomplete digestion with a frequently cutting restriction enzyme, so that fragments of about

larger pieces of noncoding DNA,

exons, separated introns.

by

4. Restriction Maps, Southern Blots, and DNA Sequencing Characterization of gene structure is done in broad

20 kbp are formed. Vector arms are ligated to both sides

outline by restriction mapping and in detail by estab­

of these fragments, and the resulting DNA molecules

lishing the DNA sequence. A restriction map identifies

packed into lambda proteins, as described in Section

the positions of sites for a set of restriction enzymes.

III.e. The lambda clones that result from the infection

The correct order for the fragments produced by a

will, if one has close to a million different clones, repre­ sent all the DNA and genes of a mammal. If one has a cDNA clone for a gene of interest, this cDNA can be radioactively labeled and used as a probe to find the lambda clones that contain the gene or parts of it.

given enzyme is determined in several ways, for exam­ ple, through comparison of DNA fragment lengths cor­ responding to different degrees of partial digestion with the enzyme. The approximate distances between the cut sites are obtained by including molecular weight markers in the electrophoresis. Accurate dis­ tances require DNA sequencing.

3. Splitting of Eukaryotic Genes

If one has a cDNA probe for a gene, it is possible

One of the big surprises in molecular biology re­

to obtain some information on the structure of the

search was a discovery made in 1977 on cloned eukary­

gene without having cloned it by using the technique

otic genes. The chicken gene for ovalbumin was found to be

7700

bp long, while only

2000

bp would be re­

quired to code for the protein. Closer investigation of the cloned DNA with restriction enzymes and DNA sequencing revealed that the coding part of the gene

developed by Southern

(1979)

(Fig.

2.7b).

After diges­

tion with a restriction enzyme and separation by agar­ ose electrophoresis, the DNA fragments are denatured in alkali and transferred to a nitrocellulose or nylon membrane with a flow of buffer. A replica of the gel

DNA Technology in Diagnosis, Breeding, and Therapy

pattern of DNA is thereby produced on the membrane. When the membrane is then treated with the radioac­ tive probe, it will hybridize to those bands that have DNA sequences complementary to the probe and be­ come visible on X-ray film. In this way, information can be obtained about restriction sites in the gene region corresponding to the cDNA. DNA sequencing is carried out by means of a chemi­ cal method developed by Maxam and Gilbert (1977) or by an enzymatic method developed by Sanger et al. (1979). The chemical method starts with a cloned DNA fragment that has a few hundred bases and the 5 '­ end radioactively labeled. From this cloned DNA, a spectrum of radioactive fragments is produced in which all sizes-from the full length down to zero­ are represented. In principle, this procedure is done in four reactions, one reaction each for fragments ending with a G, C, T, or A. The fragments are separated by gel electrophoresis at a voltage of about 3000 V, and an autoradiograph is made from the gel (Fig. 2.8a). The sequence can be read directly from the autoradio-

a G

A

+

G

--

C

+

T

b C

--

G C

A T T C

-

-

3'

--

E

G

T

-

T

-

T

-

-

E

T

-

.

E

E

T

T

-- - C

graph, as indicated in the figure. The enzymatic method gives a similar end result. In this method, DNA polymerase synthesizes DNA in four reactions in the presence of small amounts of 2',3'-dideoxynucleoside triphosphates, which cause chain termination when in­ corporated. DNA sequencing work has yielded a wealth of infor­ mation on gene structure and chromosome organiza­ tion in viruses and bacterial and eukaryotic cells. The following are important examples. Sequencing of the 5375-bp cpX174 virus showed how the DNA could code for proteins of 2000 amino acids in aggregate by having an overlap between some of the genes. The transition between exons and introns was shown to have certain characteristic bases, common to all genes. The pro­ moter region in front of the genes, the site for attach­ ment of RNA polymerase, was also found to have a common base pattern. DNA sequencing has help elucidate the structure and the function of immuno­ globulin genes, histocompatibility genes, and onco­ genes. The sequencing technique has developed quickly. By 1982, the full sequence of lambda DNA with its 48,502 bp was published (Sanger et aI., 1982). Now under way is a gigantic international project that calls for sequencing of the total human genome with its 3 X 109 base pairs by the year 2005. 5. Polymerase Chain Reaction

Probe

G

-

2

E

G

--

1

29

5'

FIGURE 2.8 (a) Autoradiograph of a gel showing labeled frag­ ments obtained in the Maxam and Gilbert DNA sequencing method. The reaction used for adenine affects both adenine and guanine, and the reaction used for cytosine affects both cytosine and thymine. Reproduced from Stryer (1981), with permission. (b) A restriction fragment length polymorphism caused by the presence or absence of a site for restriction enzyme E.

An important use of sequence data is in enzymatic amplification of segments of DNA in the polymerase chain reaction (PCR). PCR amplification involves two oligonucleotide primers that flank the DNA segment to be amplified and repeated cycles of heat denaturation, annealing of the primers to the complementary se­ quences, and extension of the annealed primers with DNA polymerase. The primers hybridize to opposite strands of the target DNA and are oriented so that DNA is synthesized for the region between the prim­ ers. Since the extension products are also complemen­ tary and capable of binding the primers, each cycle doubles the amount of DNA resulting from the previ­ ous cycle. The basic idea of this method was described in 1971 by Kleppe et al. and later developed by Saiki et al. (1985). Its full potential appeared when Saiki et aI. (1988) introduced the thermostable DNA polymerase from Thermus aquaticus (Taq), thus eliminating the need to add fresh polymerase during each cycle. This modi­ fication not only simplified the procedure, making it amenable to automation, but also substantially in­ creased its specificity, sensitivity, yield, and the length of targets that could be amplified. Amplification of 1 0

30

Jens G. Hauge

million times was demonstrated, and segments up to 2000 bp were readily amplified, although longer seg­ ments have reduced yield. The PCR technique has found wide application (Er­ lich, 1989; Erlich et al., 1991b). Isolation of cloned seg­ ments from their vectors is simplified by PCR amplifi­ cation of these, using vector-specific primers that flank the insertion site. RNA and cDNA can be amplified from the products of reverse transcription (Todd et al., 1987). DNA sequencing with the enzymatic method can be done directly on the PCR product without ligation into a sequencing vector. For this purpose, however, it is advantageous to purify the sequenc­ ing template by incorporating biotin in one of the primers. After immobilization of the PCR product on streptavidin-coated magnetic beads, the unbiotinyl­ ated DNA strand is removed by alkali (Hultman et al., 1989). In diagnostic work and genetic mapping, PCR methods are used extensively as will be apparent in later sections. For some applications, it is important to be aware that the Taq enzyme allows incorporation of 1 incorrect nucleotide for about 20,000 nucleotides in­ corporated.

IV. STUDIES OF METABOLIC AND GENE REGULATION Students of metabolic regulation now have new tools at their disposal through the advent of recombi­ nant DNA technology. The impact has been the largest in the area of analysis of gene expression and its regula­ tion by hormones due to the availability of cDNA probes for a broad variety of genes of metabolic inter­ est. These probes make possible quantification of the level of specific messengers. A number of such studies, covering enzymes in glycolysis, gluconeogenesis, lipid metabolism, and amino acid metabolism, are described by Goodridge and Hanson (1986).

A. Tissue Culture Studies Several of the corresponding genes have been iso­ lated, and their promoter-regulator regions are being identified and characterized. One example is the rat gene for cytosolic phosphoenolpyruvate carboxy­ kinase (PEPCK), studied by Hod et al. (1986). The promoter-regulator region had been localized to be within a 621-bp BamH1 / BglII fragment present at the 5' -end of the gene. The hormonal control region was studied by linking the 621-bp fragment to the structural segment of the herpes simplex thymidine kinase (TK) gene in a plasmid vector. This construct was used to

transfect cells of a TK-deficient rat hepatoma cell line. The addition of dibutyryl-cAMP to the cells resulted in a fourfold to sixfold induction of both TK activity and the level of its mRNA. To define the sequences in the PEPCK promoter that are necessary for this cAMP effect, a series of graded deletions was constructed. The transfected chimeric genes retained their respon­ siveness to cAMP as long as they included the sequence from - 61 to - 108. Deletion of this 47-bp fragment completely eliminated cAMP inducibility. Further studies of this and other cAMP-regulated genes have shown that a palindromic DNA sequence, TGACGTCA, termed eRE (cAMP response element) is sufficient for the effect. Other hormones, such as glucocorticoids, androgens, mineralocorticoids, estro­ gen, thyroxin, and retinoic acid, have their specific response segments, where the hormone-receptor com­ plexes bind (Lucas and Granner, 1992). cAMP normally exerts its effect through activation of protein kinase A (PKA), and such is the case here as well. A CRE-binding protein (CREB) was discovered that is phosphorylated by PKA and thereby activated (Yamamoto et al., 1988). PEPCK is actually regulated by several hormones and the dietary state. Further transfection studies have uncovered eight protein-binding domains between -460 and + 73 and identified proteins binding to these response elements (Park et al., 1993). The interaction of these partially tissue-specific transcription factors with the basic transcriptional complex is currently be­ ing studied.

B. Regulation in Transgenic Animals Although much can be learned from studies with cell cultures, including how they react when recombi­ nant DNA is introduced, this approach has its limita­ tions. One fundamental set of regulatory problems is connected to the development of an animal from a fertilized ovum. What sort of mechanisms are responsi­ ble for tissue- and time-specific transcription of genes? Such questions are addressed through the use of trans­ genic organisms, particularly mice (Westphal, 1987; Grosveld and Kollias, 1992). This important technique was introduced by Gordon et al. (1980). The DNA to be investigated has been injected into the male pronu­ cleus of a fertilized ovum, and the ovum implanted. Some of the ova develop into mice with the extra DNA integrated into their chromosomes, usually with many copies arranged head to tail. The first successful experiments dealing with tissue­ specific regulation were reported by Chada et al. (1985). These investigators showed that a human f3-globin gene, in which the front end had been replaced by the

DNA Technology in Diagnosis, Breeding, and Therapy corresponding mouse gene, together with 1200 bp of flanking DNA, was expressed in transgenic mice exclu­ sively in erythroid cells. They also showed that the . hybrid ,a-globin gene was expressed at the proper ti e during development (Magram et al., 19 5). The 5 ­ . flanking region must have been respon�Ible for t s specificity, along with corresponding tissue-speCIfic regulatory proteins. Similar results have been found for a number of other genes and tissues (Palmiter and Brinster, 1986). A particularly striking example is the work on e.last�se I (Swift et al., 1984). Mice were made transgemc WIth the rat elastase I gene and its flanking regions. A marked tissue specificity was observed, rat elastase mRNA being 500,000 times more abundant in the pan­ creas than in the kidneys. Experiments with trimmed 5' -flanking DNA showed that 205 bp were sufficient to uphold the pancreas-specific expression (Ornitz et al., 1985). When this regulatory region was joined to the human growth hormone gene, human growth hor­ mone was found in pancreas and not in other tissues. Furthermore, immunofluorescence analysis demon­ strated that the growth hormone was present in the acinar cells, not in the endocrine or connective tissue cells of the pancreas. The faithful tissue specificity observed with this kind of construct has been exploited in the study of oncogene expression (Section V.E). In one such study, Hanahan (1985) directed the expression of the SV40 oncogene to pancreatic ,a cells using an insulin gene regulator sequence. This technique has also been used, for instance, to further understand the complex and tissue-specific regulation of the PEPCK gene. McGrane et al. (1988) fused the promoter-regulatory sequence of the PEPCK gene to the bovine growth hormone gene as a reporter gene. The transgene was expressed only in liver and kidney and showed dietary and hormonal responsive­ ness. Further, in order to establish their regulatory roles, mutations in specific regulatory domains were produced in transgenic mice (Patel et al., 1994). Another method for producing transgenic animals uses embryonic stem (ES) cells. These ES cells are de­ rived from an early embryo, cultivated, and infected with a retrovirus construct (Robertson et al., 1986) or transfected through the phosphate / DNA precipitation technique (Lovell-Badge et al., 1985). When they are reintroduced into the blastocyst, the ES cells contribute to the production of a chimeric animal. In the next generation, some pure transgenic animals appear. This approach has the advantage that somatic cell genetic techniques can be used to modify and to select cells with a desired potential.

�

�

�

31

v. DIAGNOSIS OF GENETIC DISEASE

A. Gene Changes in Hereditary Disease Mutations could affect the production of a given gene product in many different ways. Studies of hemo­ globin synthesis in humans have found ex �mples of most of these possibilities. Replacement of a smgle base with another base can have the effect that an amino acid is replaced by another amino acid. One hundred and eighty-nine such structural variants are known for the human ,a-globin chain (Little, 1981). Not all of these . variants cause disease, however. The mechamsm of polypeptide chain termination is the source ?f other disturbances. In individuals with hemoglobm Con­ stant Spring, the stop codon TAA for a-globin is re­ placed by CAA. In a type of ,aD-thalassemia (�o produc­ tion of ,a chains), the seventh codon, AAG, 18 replaced by TAG, terminating the chain prematurely. . . Single base mutations may also affect transcn�tion and RNA processing, which is often observed m ,a­ thalassemia (low production of ,a chains). Some muta­ tions in the promoter region reduce transcription markedly. Changes at exon-intron juncti�ns lead to errors in the removal of introns from the pnmary tran­ script, with unusable mRNA as the result. Mutations can also generate new, false exon-intron border se­ quences, again with defective mRNA as the result. Deletions and insertions are other common types of mutations. Deletions or insertions of one, two, or four bases are the cause of several ,ao-thalassemias. Such a mutation destroys the reading frame completely be­ cause the wrong set of triplets is read by the ribosomes. Deletions of n X 3 bases can also occur, and the reading frame is not affected. Mutants with ,a-chain shortening from one to five amino acids are known. Larger dele­ tions also occur; for hemoglobin genes, deletions from 0.6 kbp up to about 20 kbp are known. Studies have uncovered a similar mutational hetero­ geneity for many other genes. In addition, a �ew type of mutation has been discovered for myotomc dystro­ phy (DM) and six other diseases (Miawa, 1994). The DM gene contains the triplet repetition (GCT)n, with n varying somewhat in the population. For affected . individuals n is increased and appears to correlate WIth the severity of the disease.

B. Detection of Gene Changes with DNA Probes Recombinant DNA technology makes possible a di­ rect approach to the study and diagnosis of a genetic disease, instead of being limited to study of its pheno-

32

Jens G. Hauge

typic expression. For animal owners and animal breed­ ers, it is worthwhile to be able to detect carriers of recessive disease genes in order not to propagate the disease gene or to avoid mating the animal to another carrier of the same defect. Furthermore, it is not always the case that the homozygous state for a recessive gene, or the heterozygous state for a dominant gene, is ex­ pressed phenotypically. The disease may be dependant on an environmental factor for its expression or have a late onset in the life of the animal. In this case, DNA analysis could provide an early diagnosis. Application of recombinant DNA methods in diag­ nosis and gene defect studies have made great ad­ vances in the human sector and are well under way in the veterinary sector also. Common to the methods is the need to have cloned or synthetic DNA probes for the disease gene or its neighborhood and / or sequence knowledge that permits the use of PCR methods. 1. Mutations That Create or Destroy a Restriction Site

The ideal situation exists when the mutation either creates a new restriction enzyme site or removes one that was present in the wild type. This is the case for sickle cell anemia in man. In sickle cell anemia, the sixth amino acid in the f3 chain of hemoglobin, gluta­ mate, has been replaced by valine because an A in the corresponding codon has been replaced by T. The restriction enzyme DdeI for normal individuals cuts three places in the f3-globin gene, whenever the se­ quence CTNAG occurs (N can be any base). The muta­ tion has the effect that the second of these sites is destroyed. Sickle cell patients thus will have one longer DNA fragment hybridizing with a f3-globin probe, whereas normal persons will have two shorter frag­ ments, which is revealed by Southern blotting as de­ scribed in Section III.D.4. A heterozygote would be revealed by all three bands being present. Each restriction digest requires 5-10 JLg of patient DNA, which is usually obtained from the buffy coat of peripheral blood samples. The leukocytes present in 10-20 ml of human blood can yield as much as 200 JLg of DNA after treatment with proteinase K and extraction with phenol and chloroform. There are many other examples of diagnosis based on change in a restriction site, for example, in the ras family of oncogenes and for induced H-ras mutations in the rat (Zarbl et ai., 1985). A more convenient version of this test amplifies the region around the mutation with PCR and subjects the PCR product to restriction enzyme treatment. Sufficient DNA is often present to allow direct detection of the restriction fragments after electrophoresis using ethidium bromide and UV light.

2. DeletionlInsertion Mutations A favorable situation exists when deletions or inser­ tions of a few hundred base pairs or more are the cause of the disease. These mutations will directly affect the length of the DNA fragment between two given restric­ tion sites. Deletions occur in human globin genes, the gene causing Duchenne muscular dystrophy, the gene for coagulation factor VIII and others.

3. Single Base Replacement in a Known DNA Sequence Base replacement mutations that do not affect any restriction site may still be diagnosed if one knows the DNA sequence around the mutation site. One can then synthesize a wild-type allele-specific oligonucleotide (ASO) of 18 to 20 bases, label it with radioactivity, and observe its hybridization with DNA fragments on a membrane. Under stringent hybridization conditions, it is possible to detect a mismatch of only one base between the probe and the DNA investigated. Addi­ tionally, it is prudent to make a probe corresponding to the mutant and demonstrate its reduced hybridiza­ tion to normal DNA. This method is used in diagnosis of arantitrypsin deficiency (Kidd et ai., 1984), H-ras mutations in rats (Zarbl et aI., 1985), and several other mutations. The method is more reliable and convenient when carried out on a PCR-amplified segment containing the mutation. The sensitivity thus reached makes possi­ ble the use of nonradioactive, enzyme-labeled ASO probes in dot blot hybridization (Scharf et ai., 199 1 ). The occurrence of many different mutations in the same gene in the population can be handled using a "reverse dot blot" method. Instead of having numer­ ous membranes with the PCR product applied, each to hybridize with a different ASO probe, a membra�e that contains an immobilized array of ASO probes IS hybridized to the PCR product (Erlich et aI., 1991a).

4. Restriction Fragment Length Polymorphisms Linked to the Gene of Interest Earlier, the most widely used and general diagnostic method was that of restriction fragment length poly­ morphism (RFLP) linked to the gene that was investi­ gated. With this technique, one needs no knowledge of the structure of the gene. It is based on the fact that the heterozygosity at the level of DNA is large. If the base sequences of two homologous chromosomes are compared in humans, a difference for each 200 to 400 bp is found (Cooper and Schmidtke, 19�). The majority of these differences are neutral mutations. A few of them reside in the coding triplets, but most of

DNA Technology in Diagnosis, Breeding, and Therapy

them are in introns and in the region between the genes. Some of these mutations will affect known restric­ tion enzymes and be within or close to the gene of interest, so that they can be observed as RFLP with a probe from the same region (Fig. 2.8b). Such an RFLP can be used as a marker for the gene studied. The procedure is essentially the same as in Fig. 2.7h. Pres­ ence and absence of the restriction site will give two different restriction fragment lengths that move differ­ ent distances on the gel.

a. Study of Linked RFLP in Families The use of RFLP for diagnosis has one drawback. An answer cannot be obtained by investigating the DNA of a single individual, as was possible with the methods in Sections V.B.1 through V.B.3. The chromo­ some with the mutant form of the gene will not always have the same restriction site allele, for instance, pres­ ence of the restriction site, that may have been the original constellation. However, only a fraction of chro­ mosomes with this restriction site need have the gene mutation. By recombination, the mutant gene may also have landed on a chromosome with this restriction site absent. If the distance between the restriction site of concern and the gene studied is small, it is probable that the constellation will remain the same within a given family. The use of this method is illustrated in the following example from a family with ,a-thalassemia. Both par­ ents were healthy, but a child was affected. During the following three pregnancies, prenatal diagnosis was performed. Using the restriction enzyme AvaIl, South­ ern blotting, and a labeled ,a-globin probe, the band pattern shown in Fig. 2.9 was observed. Both parents were RFLP + , one chromosome having the restric­ tion site, the other not. The affected child was + + ; therefore, it must thus have been the + chromosome from both parents that carried the disease gene. For the -

33

fetuses, one could conclude that they were noncarrier, affected, and carrier, respectively. The probe cross hy­ bridizes with S-globin sequences common to all.

b. Usefulness of Having More Than One RFLP The example just described was a favorable situa­ tion. It would only have been possible to diagnose 50% of the offspring correctly if the thalassemia gene in one parent had been on an RFLP chromosome. The situation might have been helped if one had another RFLP close by. With two RFLPs, one would have four combinations, so-called "haplotypes." This circum­ stance is equivalent to having four alleles for the marker locus, which gives a favorable degree of hetero­ zygosity. If the distance between the gene and the RFLP site is large, for example, some thousand kilobase pairs, it is important to have a second RFLP situated on the opposite side of the gene. A crossing over during meio­ sis between the first RFLP and the gene would then be detected since a double crossing over is unlikely. Such a flanking RFLP is also useful in work attempting to localize and study the gene if it is not yet charac­ terized.

c. Searching for RFLP The search for RFLP does not have to be a random trial-and-error test of the large battery of commercially available enzymes. Cooper and Schmidtke (1984), Wijs­ man (1984), and Feder et al. (1985) describe rational approaches and give lists of relative efficiency for many enzymes. Enzymes containing the dinucleotide CG in their recognition sequences detect more variation of the base replacement type than other enzymes, proba­ bly because of mutations from methylated CG to TG. Larger probes have been found to be more efficient than small probes. Screening should be limited, as suggested by Skol­ nick and White (1982), to a panel of about 10 unrelated individuals in order not to pick up rare polymor­ phisms. These have low degrees of heterozygosity and therefore are inferior in diagnostic work. When multiple restriction fragments are found, it must be shown that they represent alleles at the same locus. A probe might recognize nonallelic fragments due to sequence homology with a different chromo­ somal region. If the fragments are alleles, they will segregate in a Mendelian fashion.

d. Minisatellites +

FIGURE 2.9 Prenatal diagnosis of ,a-thalassemia. Adapted with permission from Francomano and Kazazian (1986).

Studies of gene structure have uncovered situations in which a single restriction enzyme reveals a multial­ lele polymorphism, a polymorphism that does not af­ fect sites for the enzyme. Between two fixed restriction

34

Jens G. Hauge

sites, one finds a region consisting of tandem repetition of 15- to 100-bp units, and the number of these repeats varies considerably in the population. Such variable tandem repeat (VNTR) regions, also called minisatel­ lites, have been found in humans close to the insulin gene (Bell et al., 1982), a-related globin genes (Proud­ foot et al., 1982), and the c-H-ras oncogene (Capon et al., 1983), but similar VNTR regions appear widely dispersed ( Jeffreys et al., 1985b). The presence near a disease gene of a VNTR region is very useful, because the multiallelic nature of the corresponding RFLP leads to high degrees of heterozygosity. A method for a sys­ tematic search for such RFLP has been reported (Naka­ mura et al., 1987).

e. Microsatellites Weber and May (1989) reported the existence of a large set of highly polymorphic microsatellites, consist­ ing of dinucleotide repeats, that could be typed using PCR. Such microsatellites have become the most im­ portant source of markers for high-resolution genetic maps. A human map was presented in 1992 with 814 microsatellites (Weissenbach et al., 1992). Two years later the map included 2066 (AC). markers, flanked by unique sequences for which PCR primers could be synthesized (Gyapay et al., 1994). The average distance between the markers was 2.9 cM, which means that there are now polymorphic microsatellite markers close enough to most disease genes for diagnostic pur­ poses. These markers are also starting points for work to identify the genes and their mutations. This latter task is being made easier by the parallel development of physical maps of yeast artificial chromosome clones (Cohen et al., 1993) that have been ordered using sequence-tagged sites (Olson et al., 1989).

In other instances, map position has been important in order to verify gene identification, but the availabil­ ity of a plausible candidate protein and its cDNA has made it possible to avoid exhaustive cloning in the specified chromosome region. This positional candi­ date approach will probably become increasingly com­ mon (Ballabio, 1993). In all, about 400 human disease genes can now be diagnosed directly.

6. Genetic Screening Carrier screening programs have been attempted for some human genetic diseases even before DNA tests became available, notably sickle cell anemia, thalassemia, and Tay-Sachs disease. The thalassemia program in Cyprus for testing couples at marriage has been very successful. Pilot programs that test pregnant women for the cystic fibrosis gene and the partner if the test is positive have taken place in Britain and the United States (Williamson, 1993). Although there are legitimate concerns that screening data may be used to coerce families to make reproductive decisions that meet economic or societal objectives rather than per­ sonal wishes, the availability of a screening service is generally welcomed. For human diseases, there is, however, often the problem of genetic heterogeneity. The number of dif­ ferent cystic fibrosis mutations is now greater than 260. Although most of these are rare, 11 mutations making up 91.5% of the mutations in Northwest England (Ferec et al., 1992), a screening program for cystic fibrosis will clearly leave a small residual risk for unexpected development of the disease. For animal genetic dis­ eases, there is less heterogeneity, making screening an effective and widespread tool (Section V.D).

C. DNA Probes and Disease Susceptibility

5. Disease Gene Identification Rapid progress has been made in the identification and characterization of disease genes in humans, a prerequisite for application of the direct methods of diagnosis described in Sections V.B.1 through V.B.3. Many genes have been identified by functional cloning, using information about the protein product (sequence or antibodies). Mapping of the gene then followed cloning. Starting in 1986, another approach, positional cloning, came into use. In this approach, mapping to a precise location on a chromosome is the basis for the cloning effort. Some 20 disease genes have been identified in this way (Collins, 1992). As the Human Genome Project produces high-resolution genetic maps and physical maps based on overlapping clones, positional cloning will be simplified.

For the first time recombinant DNA technology has also made it possible to begin analyzing polygenic diseases, that is, diseases where a combination of unfa­ vorable genes may predispose to the development of disease. 1. HLA Genes

Genes determining susceptibility or resistance to a broad variety of diseases in humans have been local­ ized to the HLA region, the major histocompatibility complex in humans (Vogel and Motulsky, 1986; Kos­ tyu, 1991). Among these diseases are ankylosing spon­ dylitis, celiac disease, dermatitis herpatiformis, and insulin-dependent diabetes mellitus (IDDM). The lat­ ter three are associated with the HLA serologic speci-

DNA Technology in Diagnosis, Breeding, and Therapy

ficity DR3, and IDDM also with DR4. Seventy-five per­ cent of IDDM patients have HLA-DR4, whereas only 32% of controls do. This outcome gives a DR4 individ­ ual a relative IDDM risk of 6 .4, and the risk is increased in DR3,4 heterozygotes (Knip et al., 1986) . The associa­ tion of HLA-DR genes with IDDM susceptibility is related to the fact that IDDM pathogenesis often in­ volves autoimmune phenomena (Nepom and Erlich, 1991) . With recombinant DNA technology, it became pos­ sible to ask whether there might be DNA differences, detectable as restriction fragment length differences, between patients and controls of the same DR3 and/ or DR4 specificity. This was indeed found to be the case by Owerbach et al. (1983) . Using a DQf3 probe, these investigators found a significantly decreased fre­ quency of a BamHl 3.7-kbp fragment among DR4 IDDM patients. These studies have been confirmed and extended by others, including Michelsen and Lernmark (1987) . The full impact of DNA technology came, however, with PCR amplification of cDNA from the first exons of a and {3 chains, the exons containing most of the polymorphisms, followed by sequencing (Todd et al., 1987) . This procedure has identified a large number of a- and {3-chain alleles, denoted by a four-digit code . Typing of DNA sequence polymorph isms at HLA loci is conveniently done by PCR amplification of genomic DNA with appropriate primers, followed by probing with sequence-specific oligonucleotide probes (Erlich et al., 1991a) . Such studies have shown that molecules of DQ( a l *0301,{31 *0302), associated with DR4, and DQ( a l*0501,{31 *0201), associated with DR3 give a rela­ tive IDDM risk of 6 to 18 for Caucasians. In heterozy­ gotes, two further DQ molecules may form by trans­ combination of the subunits (Thorsby and Ronningen, 1993; Nepom and Erlich, 1991). Results of this nature have been reported for malaria, where DRB I *1302DQB l*0501 gives some protection (Hill et al., 1991), and for papillomavirus-associated cervical carcinoma (Apple et al., 1994), where DRB l*IS01-DQB l*0602 gave increased risk (relative risk of 4 .8) .

2. Insulin Gene 5'-VNTR Alleles

As mentioned in Section V.B .4 .d, there is a VNTR associated with the human insulin gene on chromo­ some 11. The variable region falls in three size classes (class I averaging 570 bp, class II averaging 1200 bp, and class III averaging 2200 bp) . A strikingly higher frequency of class I alleles was observed for Caucasian individuals with IDDM (Bell et al., 1984) . By sequencing the insulin gene locus from patients and controls, Lu­ cassen et al. (1993) have identified the region of associa-

35

tion with IDDM to be 4 .1 kbp. Ten polymorphisms, including the VNTR, are in strong linkage disequilib­ rium with each other and show a relative risk for IDDM of 3.8 to 4 .5 . Analysis of other ethnic groups may reveal which exact polymorphism(s) confer susceptibility . It is possible that the VNTR length directly affects tran­ scription. The insulin gene associated polymorphism and the HLA variation are not sufficient to account for the development of IDDM. Extensive mapping of the dis­ ease in the nonobese diabetic mouse, using microsatel­ lite markers, has revealed eight additional loci that make a contribution to this polygenic disease (Ghosh et al., 1993). One of these codes for interleukin-2, which has a role in autoimmunity, and different sequences were found in nonobese diabetic and normal mice.

3. Atherosclerosis The most plausible link between genes and athero­ sclerosis is found in the structure and regulation of genes for lipoproteins, the enzymes that are involved in their metabolism, and lipoprotein receptors. Genes for the apolipoproteins and the low-density lipopro­ tein (LDL) receptor in humans have been cloned and localized on chromosomes. One of every 500 persons is heterozygous for a mutant allele of the LDL receptor, leading to two to three times normal levels of choles­ terol . Hobbs et al. (1992) have unraveled the mecha­ nisms involved and characterized different types of mutations in the gene . Mapping of the defects in a popUlation can start with a determination of mutation­ associated RFLP haplotypes. This information is useful for diagnosis within families (Rodningen et al., 1992) . To establish direct gene tests, DNA from affected indi­ viduals waS screened for mutations (Leren et aI." 1993) . LDL is bound to its receptor via apoBHJO, and hyper­ cholesterolemia may be caused by mutations in the apoBlOO gene as well ( Innerarity et al., 1990) . Mutations are known also in the apoA-I gene and cause low levels of high-density lipoprotein (HDL) accompanied by atherosclerosis. Type III hyperlipoproteinemia patients have a par ticular allele for the apoE gene . With the knowledge of DNA structure, oligonucleotide probes that differentiate between normal and mutant genes can be made, allowing screening of people at risk.

D. Application to Domestic Animals The application of recombinant DNA techniques to diagnosis in domestic animals is now quite wide­ spread. The availability of DNA clones, or knowledge of primer sequences from humans and small experi­ mental animals, has enhanced this development.

36

Jens G. Hauge

1. Diagnosis of Disease Mutations

a. Citrullinemia in Cattle The first disease to benefit from a DNA test was citrullinemia, which is widespread in Australian Friesian cattle. Three conditions were helpful for the establishment of a DNA test for the mutation in 1989: Bovine cDNA libraries were available, sequenced cDNA for the enzyme (argininosuccinate synthetase) for humans and rats was available, and the protein is fairly small, only 412 amino acids. Normal bovine cDNA for the enzyme was isolated with a rat probe and sequenced by Dennis et al. (1989). To identify the mutation, total mRNA was prepared from the liver of an affected animal and cDNA synthe­ sized. The product was amplified with PCR using primers corresponding to sequences immediately be­ fore and after the coding region of the cDNA. These primers had been extended with BamHI linkers, so the amplification product could be ligated into the BamHI site of the DNA sequencing vector. In clones from the mutant, a C to T transition in the first position of codon 86 was found that generates a TGA termination codon leading to a truncated and inactive protein. The change in codon 86 leads to the disappearance of an AvaIl restriction site. This deficiency was utilized by Dennis et al. (1989) to design a simple, PCR-based gene test on genomic DNA. A 194-bp DNA could be amplified from within the exon containing the muta­ tion. A portion of the product was digested with AvaIl, and both digested and undigested DNA were analyzed by electrophoresis. The normal calf has two bands of 72 and 118 bp, the homozygous (affected) calf has one band of 194 bp, and a carrier has all three bands.

b. Malignant Hyperthermia in Pigs Work with malignant hyperthermia (MH) in pigs had started earlier. We isolated clones for the glucose­ phosphate isomerase (GPI ) gene, a closely linked gene (Davies et al., 1987). With this DNA as probe, a multial­ lelic RFLP was detected that was confirmed to be tightly linked to the MH gene (Davies et al., 1988), so that it could serve as a marker. A further search for the MH mutation was based on reports that calcium is more easily released from the sarcoplasmic reticu­ lum of MH pigs. Meissner (1986) had shown that the alkaloid ryanodin interacts directly with the calcium release channel (CRC) of rabbits. Using tritiated rya­ nodin as a label, Lai et al. (1988) purified the channel. This work prepared the way for the cloning of the rabbit CRC cDNA by Takeshima et al. (1989), a 15,lOO-bp-long cDNA. With a small rabbit cDNA clone as a probe, we isolated a porcine cDNA done and showed it hybridized in situ to metaphase chromo-

somes in the same chromosomal band that hybridized with GPI DNA (Harbitz et al., 1990). This result con­ firmed CRC as a good candidate for the product of the MH locus. Sequencing CRC cDNA from a normal and an affected pig, Fujii et al. (1991) identified the muta­ tion, a substitution of T for C at nucleotide 1843, which changes an arginine to cysteine. The mutation was found to be the same in five Canadian pig breeds and in British Landrace (Otsu et al., 1991) and Norwegian Landrace pigs (Harbitz et al., 1992). The mutation destroys a HinP site and creates a HgiAI site. Fujii et al. (1991) amplified a 74-bp piece from the exon containing the mutation and made re­ striction site changes in the product the basis for their diagnosis. A more convenient and reliable method became possible after sequencing of genomic DNA flanking the exon.

c. Leukocyte Adhesion Deficiency in Cattle Leukocyte adhesion deficiency (LAD) in humans is an autosomal recessive disease characterized by greatly reduced expression of the heterodimeric beta2 integrin adhesion glycoproteins on leukocytes. With­ out this protein, neutrophils are unable to enter tissues to destroy invading pathogens. All human cases have been traced to the integrin beta-subunit, CDI8. A granulocytopathy syndrome had been described in Holstein cattle, and Schuster et al. (1992a) investigated whether it was also caused by a CD18 mutation. These investigators isolated bovine CDl8-encoding cDNA by screening a lymphocyte library with a murine CD18 probe and sequenced the clone. Leucocyte RNA was then isolated from a calf with LAD symptoms and from a normal cow (Schuster et al., 1992b). Northern blot analysis revealed a CD18 transcript at normal level and size from the affected calf, ruling out a large dele­ tion or a transcription defect. mRNA was reverse tran­ scribed, and the cDNA was amplified using bovine CDl8-specific primers. To sequence the amplified product, single-stranded DNA was generated with a single primer and sequenced directly. A mutation was detected at nucleotide 383, replacing aspartic acid with glycine (DI28G). The mutation occurs in a row of 26 consecutive amino acids that are identical in normal bovine, human, and murine CDI8. This mutation introduces a HaeIIl site and eliminates a TaqI site. Schuster has utilized this in an assay that PCR amplifies a 58-bp DNA from the mutated exon. Fifteen other LAD calves were found to be homozy­ gous for the D128G mutation. To determine the preva­ lence of this allele, 2025 U.S. Holstein bulls used as semen donors were screened. The carrier frequency was 14.1% before selection against the disease began. In addition, cows (1559 Holsteins) were screened, and

DNA Technology in Diagnosis, Breeding, and Therapy

5.8% were found to be carriers. The LAD incidence in the United States in calves at birth should therefore be about 0.2%. All the LAD calves had pedigrees leading back over five generations to a certain bull, Osborndale Ivanhoe, born in 1952. The carriers with pedigree infor­ mation also descended from this bull or, in a few cases, his immediate relatives. This bull was demonstrated to be a carrier of D128G by testing frozen semen. Recently, an improved detection method has been presented by Batt et al. (1994), using a thermostable ligase in an oligonucleotide ligation assay, combined with PCR (Barany, 1991). Ligation / amplification fails when a single base mismatch is present. Using biotin in one oligonucleotide, for product capture, and a non­ isotopic reporter group in the other, the method can be automated.

d. Hyperkalemic Periodic Paralysis in Quarter Horses Hyperkalemic periodic paralysis (HYPP) is a genetic disease observed among quarter horses. The disease causes attacks of paralysis, inducible by ingestion of potassium. In humans, HYPP is caused by a single base substitution within the skeletal muscle sodium channel gene. Rudolph et al. (1992a) studied a quarter horse pedigree to see if the sodium channel gene was also involved in the horse. Starting with reverse­ transcribed horse mRNA, horse sodium channel cDNA was produced by PCR, using rat primer sequences. This procedure revealed a polymorphism that was used to investigate whether the horse sodium channel gene was genetically linked to HYPP, which was found to be the case. To identify the putative mutation, they used single­ strand conformational polymorphism (SSCP) analysis (Orita et al., 1989) of reverse-transcribed mRNA PCR products from muscle biopsies of normal and affected horses (Rudolph et al., 1992b). This method is capable of detecting existing two-allele polymorphic markers (Neibergs et al., 1993). Rat and human sodium channel sequences were used to design primers. A single­ strand specific for an affected horse was seen in the PCR product from amino acid region 1348-1524. This region was cloned and sequenced. In the affected horse, there was a C to G change, replacing phenylalanine with leucine in transmem­ brane region IV of the channel. All nine normal sodium channel sequences sequenced to date have phenylala­ nine in this position, indicating its importance for activ­ ity. To test whether the leucine substitution cosegre­ gated with the disease, allele-specific hybridization and / or TaqI restriction enzyme analysis of PCR­ amplified genomic DNA sequences was used to study an extended pedigree. All affected horses in the pedi­ gree showed the leucine substitution, whereas all unaf-

37

fected horses had phenylalanine. It was theoretically possible still that this change was in strong linkage disequilibrium with a separate disease-causing muta­ tion. To check this possibility, 176 unaffected horses from a variety of breeds were tested. All were homozy­ gous for the normal allele (Phe / Phe), which strongly argues against linkage disequilibrium with some other mutation. Although this disease gene is dominant, the devel­ oped test is important because affected horses com­ monly do not exhibit signs of disease until they are mature. They may then have episodes of weakness and collapse, posing a hazard to their riders and them­ selves. Current physiological tests do not have the re­ quired sensitivity.

e. Bovine Uridine Monophosphate Synthase Deficiency Deficiency of uridine monophosphate synthase (DUMPS) is an autosomal recessive disorder in Hol­ stein and Red Holstein cattle that results in early em­ bryonic death of homozygous offspring. Heterozy­ gotes are phenotypically normal, but show about one-half the normal enzyme level and elevated levels of orotic acid. Heterozygotes may have a higher genetic merit in milk and protein production. With accurate DNA-based genotyping, it would be possible to pre­ serve the mutation in the population while avoiding matings between carriers. Sch"ber et al. (1992) isolated and sequenced wild­ type bovine UMPS cDNA using a human UMPS­ specific cDNA to screen the library. To identify the mutation responsible for DUMPS, liver RNA from DUMPS heterozygotes was reverse transcribed. Am­ plification of cDNA with sequence-specific primers and subsequent sequencing of the PCR products re­ vealed a C to T transition in codon 405, resulting in a premature stop codon and a truncated protein. The mutation led to the loss of an AvaI site. A genomic DNA-based PCR test was developed that looks for presence of the AvaI site in a 108-nucleotide segment. Complete concurrence between low levels of UMPS and presence of the point mutation was found in a large Holstein pedigree, confirming this mutation as the basic defect in DUMPS cattle.

f Bovine Maple Syrup Urine Disease Maple syrup urine disease (MSUD) results from a deficiency of the branched chain a-ketoacid dehydro­ genase, a mitochondrial multisubunit complex of en­ zymes. The E1 component catalyzes the oxidative de­ carboxylation. MSUD has been identified both in humans and in cattle and is relatively common in the Polled Hereford breed of cattle in Australia.

38

Jens G. Hauge

Hu et al. (1988) isolated and sLquenced cDNA for the bovine Ela subunit, screening an expression library with antiserum to bovine E1. To define the mutation responsible for MSUD, Zhang et al. (1990) produced cDNA from MSUD calf fibroblasts by reverse transcrip­ tion of RNA followed by PCR. These investigators chose primers that amplified the complete coding re­ gion of the Ela subunit. Subcloning of the product and sequencing revealed replacement of a CAG glutamine triplet with TAG, a stop codon, in the partial coding for the leader sequence that directs the protein to mito­ chondria. The mutation was verified by hybridization of amplified cDNA from calves with two radioactive IS-mer allele-specific oligonucleotides. MSUD also occurs in the Polled Shorthorn breed. Amplifying DNA with PCR across the glutamine co­ don and testing for allele-specific oligonucleotide hy­ bridization revealed, however, that the mutation in the Polled Shorthorn must be in a different position from that in the Polled Hereford (Healy and Dennis, 1994).

2. Diagnosis of Disease Susceptibility Association between diseases and alleles of genes in the major histocompatibility complex (MHC) is also known for domestic animals. This is particularly well documented for Marek's disease of chickens, which is caused by a herpes virus (Hanson et al., 1967; Hepkema et al., 1993). Resistance to bovine leukosis, caused by the BLV retrovirus, also has a clear MHC association (Lewin, 1994), and Mejdell et al. (1994) found an influ­ ence of the bovine MHC on resistance to mastitis. A highly significant association was found between the bovine MHC class I antigen BoLA-A8 and chronic pos­ terior spinal paresis, a form of ankylosing spondylitis, in Holstein bulls (Park et al., 1993). The relative risk was 34.6. The special interest in developing RFLP to establish haplotypes for the MHC genes in domestic animals must be viewed in this light. MHC RFLP have been found with human probes in a large number of species (Andersson et al., 1986; Juul-Madsen et al., 1993). More informative studies are now being con­ ducted based on sequencing of the MHC polymorphic exons. Such data were presented for cattle (Andersson et al., 1991), sheep (Fabb et al., 1993), chickens (Moon Sung et al., 1993), horses (Szalai et al., 1993), and pigs (Vage et al., 1994).

E. Gene Changes in Cancer Cancer is, like genetic diseases, a result of changes in the hereditary material, but in the latter case, mainly in somatic cells. A great deal of light has been thrown on the phenomenon of neoplastic growth via the use

of recombinant DNA techniques. The molecular struc­ tures of a long series of viral oncogenes and cellular proto-oncogenes, often coding for proteins participat­ ing in signal transduction from the cell surface to the nucleus, have been elucidated, and the mechanisms for the activation of proto-oncogenes to oncogenes to a large extent clarified (Land et al., 1983; Van de Woude et al., 1984). For the ras family of oncogenes, this activa­ tion takes place by a base replacement, for others, for example, myc, myb, and ets-l, by translocation and / or amplification. Oncogene probes are available for such studies. Translocations, and sometimes mutations, re­ sult in restriction fragment size changes. Amplification leads to stronger bands of hybridizing DNA. Amplifi­ cation, or stimulated transcription, leads to increased hybridization to mRNA. These developments have clinical relevance. DNA methods are becoming useful in tumor diagnosis, and they give new information as to the stage of develop­ ment of the tumor. Oncogenes are highly conserved in evolution, so it is likely that human and murine oncogene probes can be used for investigation of cancer development in domestic animals. Human probes for c-myc, myb, HER-21neu, H-ras, K-ras, and N-ras thus were found to hybridize to DNA fragments from dogs (Hauge et al., 1988). While activated oncogenes act in a dominant man­ ner, mutations in tumor suppressor genes are reces­ sive. Their gene products inhibit cell division, and thus balance the stimulatory properties of the proto­ oncogene products. Loss of function of the suppressor gene on both chromosomes leads to increased cell divi­ sion. Because the first mutation is often transmitted from a parent, whereas the second arises somatically, such tumors are found to be hereditary. Well-studied examples are the genes for retinoblastoma and pS3, as well as several suppressor genes involved in colorectal cancer (Weinberg, 1991). Gene tests for these mutations are of obvious value for early diagnosis and treatment.

VI. THERAPY OF GENETIC DISEASES In the foreseeable future, gene therapy may not be of practical value in regular veterinary medicine be­ cause of the cost involved. But work with gene therapy in animals will nonetheless be important as prepara� tion for such therapy in humans (Anderson, 1984). What needs to be shown in animals is (1) that the gene is integrated in the target cells, (2) that it is expressed on a suitable level, and (3) that the gene does not harm the cells or the animal. The ideal would be to exchange the mutated gene with a wild-type gene. Techniques for accomplishing this exchange by homologous re-

DNA Technology in Diagnosis, Breeding, and Therapy

combination in ES cells have been developed (Smithies

et al., 1985; Thomas and Capecchi, 1986) . Gene therapy can be conceived at different levels: introduction of DNA into the fertilized ovum or early embryo; into cells or tissues that are taken out, modi­ fied, and reimplanted; and into the whole animal with the help of an infective vector.

A. Germ Line Gene Therapy Several successful experiments of this type have been reported. Hammer et al. (1984) describes how injection of the rat growth hormone (GH) gene, linked to the mouse metallothionein promoter, into fertilized ova from dwarf mice with low GH levels yielded mice with normal growth. Le Meur et al. (1985) restored the immune response to a tripeptide in mice from a line with a defective MHC class II E gene, by injection of DNA containing this gene. ,a-Thalassemia has also been successfully corrected by injection of ,a-globin DNA in a murine model of human ,a-thalassemia (Cos­ tantini et al., 1986) .

B. Somatic Gene Therapy Human gene therapy is coming of age (Miller, 1992; Friedman, 1993; Tolstoshev and Anderson, 1993) . Pro­ tocols for treatment of 14 diseases have been approved (Wivel, 1993) based on results obtained with animals . The first disease treated was adenosine deaminase (ADA) deficiency, which causes immunodeficiency . A retrovirus vector was used to introduce the normal cDNA in lymphocytes. Among the diseases recently attacked are cystic fibrosis and hypercholesterolemia . Cystic fibrosis affects about 1 of every 2000 Cauca­ sians. Mouse models for cystic fibrosis have been gen­ erated by homologous recombination in embryonic stem cells, in one case by replacement of part of the cystic fibrosis transmembrane conductance regulator (CFTR) gene with a mutated gene section (Snouwaert et al., 1992), in the other case by insertion of a mutated gene in the host gene (Dorin et al., 1992) . The first type is a null mutation, whereas the latter has a small degree of leakiness. Hyde et al. (1993) showed that CFTR cDNA could be delivered to the lungs of the replace­ ment mutant by direct instillation of a cDNA-liposome cocktail, resulting in correction of the ion conductance defect . Other workers have used adenovirus as a vec­ tor. Unlike retrovirus, adenovirus is capable of infect­ ing terminally differentiated cell types and is naturally drawn to airway epithelial cells. Good expression of human CFTR cDNA, carried on replication-deficient adenovirus, was observed after intratracheal introduc­ tion in cotton rats. Adenovirus is not integrated into

39

host DNA. Studies have therefore been performed on the safety and efficiency of repeated CFTR cD NA trans­ fer in cotton rats and primates (Zabner et al., 1994). Another cDNA that has been carried on adenovirus into rat lung epithelium and expressed is cDNA for aI-antitrypsin (Rosenfeld et al., 1991) . Adenovirus can also be used for muscle diseases. Vincent et al. (1993) have observed long-term correction of dystrophic de­ generation in a mouse model for Duchenne muscular dystrophy after intramuscular injection of adenovirus carrying a human dystrophin minigene. Familial hypercholesterolemia (FH) a ffects individ­ uals heterozygous for LDL receptor mutations (Section V.C.3) and is very severe in those who are homozy­ gous. A strain of rabbits genetically deficient in LDL receptors was used to demonstrate the potential effi­ cacy of ex vivo gene therapy . Part of the liver was resected and perfused with collagenase to free hepato­ cytes. The isolated hepatocytes were transduced with LDL receptor retrovirus and infused into the rabbit liver with good effect (Chowdhury et al., 1991). The experiment was followed up using dogs and baboons. The first results in humans have been reported (Gross­ man et al., 1994) . A 29-year-old woman, homozygous for the disease, underwent the same procedure, and her LDL/ HDL ratio declined from 10 to 13 before treat­ ment to 5 to 8 following gene therapy .

VII. USE OF RECOMBINANT DNA METHODS TO IMPROVE DOMESTIC ANIMALS A. Markers and Maps One important use of DNA methods for the im­ provement of domestic animals has been discussed already : marker-assisted selection against animals that carry a disease gene that has been characterized at the molecular level. On the whole, however, specific genetic disorders are less important than traits affect­ ing growth, reproduction, disease resistance, and the quality of the end product. These traits are usually determined by several genes, by quantitative trait loci (QTL). To enable marker-assisted selection for a QTL, it must be fairly precisely localized on a chromosome . This localization requires genetic and physical maps of high resolution. Much effort has gone into establishing such maps. The work is furthest advanced for pigs. At the First Pig Gene-Mapping Workshop, the number of mapped loci was 170 (Andersson et al., 1993) . Sixty of these loci were anonymous DNA segments, mostly microsatel­ lites, and the rest were polymorphic blood groups and

40

Jens G. Hauge

protein polymorphisms. Using a two-generation refer­ ence population, Rohrer et ai. (1994) established a ge­ netic linkage map with 376 microsatellite and 7 RFLP loci. The average distance between adjacent markers was 5.5 cM. Ellegren et ai. (1994) reported a genetic linkage map for a cross between the European wild boar and a domestic breed (large white) that had a somewhat lower resolution (l1-cM average spacing), but had 60 reference markers for comparative mapping and 47 markers physically assigned by in situ hybrid­ ization. This material has been analyzed with respect to some quantitative traits, and evidence for QTL with large effects on growth, length of the small intestine, and fat deposition was found on chromosome 4 (An­ dersson et ai., 1994). Comparative gene mapping for pig chromosome 4 could indicate candidates for these QTL because clusters of neighboring genes tend to be preserved between species. An example of the comparative approach was the focus on the long arm of human chromosome 19 in the search for the MH locus. It was known that the GPI gene was located here, and that GPI and MH were closely linked in pigs. Another striking example is the observation of linkage for the human secreted phos­ phoprotein 1 (SSPl) locus to the sheep Booroola fecun­ dity (FecB) gene, and the subsequent finding of linkage to human epidermal growth factor and complement I genes, near SSPI in 4q in man. The latter genes define a candidate region on sheep chromosome 6 (Montgom­ ery et ai., 1993). Good progress has also been achieved for genome mapping in cattle. Barendse et al. (1994) mapped 171 loci, with an average distance between markers of 15 cM. Fifty-six loci represent DNA information from other species. An American-Swiss collaboration pro­ duced a bovine map with 313 polymorphic markers, with an average spacing of 8 cM (Bishop et ai., 1994). An extensive comparison between cattle, cats, mice, and humans has been presented by O'Brien et al. (1993). A good foundation now appears to exist for the isola­ tion of bovine QTL, as well as for the study of mamma­ lian genome evolution.

B. Parentage Identification Errors may occur in marking semen portions used in artificial insemination. Piglets may jump from one pen to another and be falsely marked when marking takes place. And it may, at times, be a temptation for a breeder to report a false father or mother. These problems can be addressed using core sequences for minisatellites, which hybridize with the VNTR of many genes, producing a DNA fingerprint ( Jeffreys et ai., 1985a). Because each band represents a minisatellite

locus, the probability that two individuals will have the same pattern is nearly zero except for monozygotic twins. A multilocus microsatellite probe, such as (GTG)s, yields a similar result, with better coverage of the genome (Mersch and Leibenguth, 1994). Alterna­ tively, the PCR results for a set of highly polymorphic microsatellites can be combined (Marklund et ai., 1994).

C. Transgenic Animals When improving livestock by traditional breeding, one is limited to an exploitation of the alleles that are already present in the animal population. The results 14 years ago of introduction of extra DNA into pronu­ clei of fertilized mouse eggs suggested the possibility of a quicker and larger improvement of production animals. Experiments with injection of growth hor­ mone (GH) in lactating cattle had demonstrated a sig­ nificant boost in milk yields. It was natural to ask whether the same result could be obtained with extra gene copies for the hormone or whether pigs with extra GH genes could reach slaughter weight sooner and with lower feed consumption. The first report on transgenic livestock animals ap­ peared in 1985 (Hammer et ai.). In the following years, these initial experiments were extended. In a review, Pursel and Rexroad (1993) listed 24 reports on trans­ genic pigs, 11 on sheep, 6 on cattle, and 2 on goats. The average frequency of success (transgenic animals / eggs transferred) was 0.8-0.9% for these species, com­ pared to 2-5% routinely obtained with mice. This re­ sult puts a severe constraint on the adoption of this technology. For sheep and cattle there is, however, a promising development in that viable embryos can be obtained by in vitro maturation and fertilization of oocytes removed from slaughterhouse ovaries (Lu et ai., 1987). 1. Improving Productivity Traits

Transfer of genes affecting feed conversion, rate of gain, reduction of fat, and improved quality of meat, milk, and wool would be of great importance for both producer and consumer. These traits are, however, usually determined by several genes, by QTL, of which few are presently known. The mapping efforts dis­ cussed earlier will gradually change this situation. Many of the pig and sheep experiments have in­ volved GH and GH-releasing factor genes from various species. The promoter has often been that of metallo­ thionein from mouse and other species, but other pro­ moters have also been used (mouse albumin, Moloney leukemia virus, rat PEPCK). Pursel et al. (1989) found concentrations of GH in expressing founder pigs to be in the range of

DNA Technology in Diagnosis, Breeding, and Therapy

14-4000 ng/ ml for human GH. The variation could reflect the influence of the chromosomal position for the inserted gene. To obtain a reliable assessment of the effect of the GH transgene on growth, the daily weight gain and feed efficiency were measured for two generations of transgenic and control pigs. The combined transgenic progeny showed 11 % faster weight gain than the controls, and there was a 16-18% increased feed efficiency. Equally dramatic was the effect on subcutaneous fat accretion. Mean back fat thickness was reduced from 21 to 7.5 mm. In a follow­ up study with bovine GH transgenic pigs, Solomon et al. (1994) studied whole-carcass ground tissue lipid composition. At 92 kg, these pigs contained 85% less carcass fat than control pigs, and the reduction in poly­ unsaturated fatty acids was somewhat lower. These favorable traits were, however, offset by con­ siderable deleterious side effects. Many of these could be the results of the lasting, high concentrations of GH. There was a high incidence of joint pathology, gastric ulcers, and infertility. What appears necessary is to achieve a tightly controlled GH gene expression, so that it can be confined to 1-2 months during the period . of rapId growth. It should be possible to devise a pro­ moter construct that is turned on by a substance added in the feed.

2. Disease Resistance One route to specific disease resistance is introduc­ tion of one or more immunoglobulin genes that produce antibodies against a particular pathogen. Model experiments with mice have shown that this could work (Rusconi and Kohler, 1985). WeidIe et al. (1991) produced transgenic pigs that harbored mouse lambda heavy chain and kappa light chain trans­ ge�es from antibodies against the hapten, 4-hydroxy3-rut:0ph�nylacetate. Titers up to 1 mg/ml were obtamed m one foun�er and its progeny. An interesting suggestion for reducing mastitis is to fuse the lysostaphin gene with regulatory elements �om t � betalactoglobulin (BLG) gene before injection m fertihzed bovine eggs. The transgenic cow would then produce lysostaphin in its udder. Lysostaphin hydrolyzes the cell wall of Staphylococcus aureus. The idea is being tested in mice (Clark et al., 1992). �n chickens, Salter and Crittenden (1989) have de­ scnbed a transgenic line carrying a defective ALV ret­ rovirus geno�e that expresses only the envelope pro­ . teIns of the VIruS. This expression leads to resistance since the produced viral protein competes for the virus :ec�pt?r binding sites on the cell surface. This approach IS simIla� to a successful technique for making plants . VIruS reSIstant. The technique has been used to make

�

41

sheep resistant to visna virus, which causes ovine pro­ gressive pneumonia (Pursel and Rexroad, 1993).

3. Milk Modification Various ways of modifying milk protein composi­ tion can be envisaged (Clark et al., 1992). Caseins could be altered in their phosphorylation sites and, thereby, in micelle properties. Reduction in the amount of BLG could reduce allergies and inhibit the synthesis of lac­ tose, a disaccharide not tolerated by 90% of the adult world population. It would also seem useful to try to introduce into cows the gene for human lactoferrin, an antimicrobial agent and iron transporter, in order to make bovine milk better suited for human infants. Ex­ periments have been initiated to achieve this (Krim­ penfort et al., 1991).

4. Phannaceutical Proteins The lack of a-I-antitrypsin (AAT) is one of the most common human genetic disorders. It leads to life­ threatening emphysema, requiring repeated adminis­ trations of intact enzyme (200 g per patient per year). The source has hitherto been human blood plasma with its limitations and dangers. Wright et al. (1991 ) reported that high levels of AAT expression had been obtained in milk of transgenic ewes, in which a sheep BLG regulatory sequence was ligated to a human AAT genomic sequence. One sheep produced AAT at 30 g/ liter, 50% of the milk protein. Earlier investigations using eDNA of AAT or human factor IX resulted in much lower yields. Several lines of sheep have now produced AAT stably during several lactations (Carver et al., 1993). Human protein C (hPC), which has an important r�le in hemostasis, is produced in the milk of transgenic pIgS, the construct here consisting of hPC eDNA in­ serted into the first exon of the mouse whey acidic protein gene (Velander et al., 1992). Milk is not the only possible vehicle for medically important proteins from transgenic animals. Swanson et al. (1992) have succeeded in making pigs transgenic for human globin genes, expressed in pig erythrocytes. Human hemoglobin could be separated from pig he­ moglobin by ion exchange chromatography. Its use would be in crisis treatment. It lacks red cell antigenic components and does not require refrigeration.

5. Use of Embryonic Stem Cells A major limitation in the production of transgenic animals by microinjection of DNA in fertilized ova is that there is no control as to where in the genome the extra DNA finds it place. Sometimes it will disrupt a

42

Jens G. Hauge

gene, introducing defects in development or function of the animal, or it may land in a region where it is poorly expressed. Work with mice has shown that for ES cells in culture, one can achieve a site-specific inte­ gration of the new DNA with the help of homologous recombination. Cells that are shown to have incorpo­ rated the gene properly are then introduced into the blastocyst. Chimeric animals result, but in the next generation, animals with the new gene in their germ cells can be found. For domestic livestock, the ES tech­ nique has met with some problems. However, Notari­ anni et ai. (1990) described the derivation of apparently pluripotent cell lines from porcine and ovine blasto­ cysts. In cattle and sheep (but not in mice), young have been born following transfer of nuclei from cells of the inner cell mass of blastocysts to enucleated oocytes (Marx, 1988). It is possible, therefore, if gene-modified ES cells can be established from these species, that we could transfer nuclei from them and avoid the chi­ meric stage.

References Anderson, W. F. (1984). Science 226, 4OI. Andersson, 1., Boehme, J., Peterson, P. A., and Rask, 1. (1986). Anim. Genet. 17, 295.

Andersson, 1., Sigurdardottir 5., Borsch, c., and Gustafsson, K (1991). Immunogenetics 33, 188. Andersson, 1., Archibald, A 1., Gellin, J., and Schook, 1. B. (1993). Anim. Genet. 24, 205. Andersson, 1., Haley, C. 5., Ellegren, H., Knott, S. A, Johansson, M., Andersson, K, Andersson-Eklund, 1., Edfors-Lilja, I., Fredholm, M., Hansson, I., HAkansson, J., and Lundstmm, K (1994). Science 263, 177I. Apple, R J., Erlich, H. A., Klitz, W., Manos, M. M., Becker, T. M., and Wheeler, C. M. (1994). Nature Genet. 6, 157. Arrand, J. E. (1985). In "Nucleic Acid Hybridization-a Practical Approach" (8. D. Hames and S. J. Higgins, eds.), pp. 17-45. IRL Press, Washington, D.C. Avery, O. T., MacLeod, C. M., and McCarty, M. (1944 ). ]. Exp. Med. 79, 137. Ballabio, A (1993). Nature Genet. 3, 277. Barany, F. (1991). Proc. Nat. Acad. Sci. USA 88, 189. Barendse, W., Armitage, S. M., Kossarek, 1. M., Shalom, A, Kirk­ patrick, B. W., Ryan, A M., Clayton, D., Li, 1., Neibergs, H. 1., Zhang, N., Grosse, W. M., Weiss, J., Creighton, P., McCarthy, F., Ron, M., Teale, A J., Fries, R, McGraw, R A., Moore, 5. 5., Georges, M., Soller, M., Womack, J. E., and Hetzel, D. J. S. (1994). Nature Genet. 6, 227. Batt, C. A, Wagner, P., Wiedmann, M., Luo, J., and Gilbert, R (1994). Anim. Genet. 25, 95. Bell, G. I., Selby, M. J., and Rutter, W. J. (1982). Nature 295, 3I. Bell, G. I., Horita, 5., and Karam, J. H. (1984). Diabetes 33, 176. Bishop, M. D., Kappes, S. M., Keele, J. W., Stone, R T., Sunden, 5. 1. F., Hawkins, G. A., Toldo, s. 5., Fries, R, Grosz, M. D., Yoo, J., and Beattie, C. W. (1994). Genetics 136, 619. Capon, D. J., Chen, E. Y., Levinson, A D., Seeburg, P. H., and Goedde!, D. V. (1983). Nature 302, 33.

Carver, A. 5., Dalrymple, M. A., Wright, G., Cottom, D. 5., Reeves, D. 8., Gibson, Y. H., Keenan J. L., Barrass, J. D., Scott, A R, Colman, A., and Gamer, !. (1993). Bio/Technol. 11, 1263. Chada, K, Magram, J., Raphael, K, Radice, G., Lacy, E., and Costan­ tini, F. (1985). Nature 314, 377. Chowdhury, J. R, Grossman, M., Gupta, S., Chowdhury, N. R., Baker, J. R, and Wilson, J. M. (1991). Science 254, 1802. Clark, A. J., Simons, J. P., and Wilmut, I. (1992). In "Transgenic Animals" (F. Grosveld and G. Kollias, eds.), pp. 249-270. Aca­ demic Press, New York. Cohen, D., Chumakov, 1., and Weissenbach, J. (1993). Nature 366, 698. Collins, F. S. (1992). Nature Genet. 1, 3. Cooper, D. N., and Schmidtke, J. (1984). Hum. Genet. 66, I I . Costantini, F., Chad a, K, and Magram, J . (1986). Science 233, 1192. Davies, W., Harbitz, 1., and Hauge, J. G. (1987). Anim. Genet. 18, 233. Davies, W., Harbitz, 1., Fries, R., Stranzinger G., and Hauge J. G. (1988). Anim. Genet. 19, 203. Dennis, J. A., Healy, P. J., Beaudet, A. L., and O'Brien, W. E. (1989). Proc. Nat. Acad. Sci. USA 86, 7947. Dorin, J. R, Dickinson, P., Alton, E. W. F. W., Smith, S. N., Geddes, D. M., Stevenson, B. J., Kimber, W. L., Fleming, 5., Clarke, A. R, Hooper, M. L., Anderson, L., Beddington, R S. P., and Porteous, D. J. (1992). Nature 359, 21I. Ellegren, H., Chowdhary, B. P., Johansson, M., Marklund, L., Fredholm, M., Gustavsson, I., and Andersson, 1. (1994). Genetics 137, 1089. Emery, A E. H. (1984). "An Introduction to Recombinant DNA." Wiley, Chichester. Erlich, H. A (1989). "PCR Technology. Principles and Applications for DNA Amplification." Stockton Press, New York. Erlich, H., Bugwan, T., Begovich, A B., Scharf, S., Griffith, R, Saiki, R, Higuchi, R, and Walsh, P. S. (1991a). Eur. ]. Immunogen. 18, 33. Erlich, H. A., Gelfand, D., and Sninsky, J. J. (1991b). Science 252, 1643. Fabb, S. A., Maddox, J. F., Gogolin-Ewens, K J., Baker, 1., WU, M-J., and Brandon, M. R (1993). Anim. Genet. 24, 249. Feder, J., Yen, L., Wijsman, E., Wang, L., Wilkins, L., Schroder, J., Spurr, N., Cann, H., Blumenberg, M., and Cavalli-Sforza, L. 1. (1985). Am. ]. Hum. Genet. 37, 635. Feinberg, A P., and Vogelstein, B. A (1983). Anal. Biochem. 132, 6. Ferec, c., Audrezet, M. P., Mercier, B., Guillermit, H., Moullier, P., Quere, 1., and Verlingue, C. (1992). Nature Genet. 1, 188. Francomano, C. A., and Kazazian, H. H. (1986). Ann. Rev. Med. 37, 377. Freifelder, D. (1983). "Molecular Biology." Jones and Bartlett, Boston. Friedman, T. (1993). In "Molecular Genetic Medicine" (T. Friedman, ed.), pp. 1-32. Academic Press, New York. Fujii, J., Otsu, K., Zorzato, F., De Leon, 5., Khanna, V. K, Weiler, J. E., O'Brien, P. J., and MacLennan, D. H. (1991). Science 253, 448. Ghosh, S., Palmer, S. M., Rodrigues, N. R, Cordell, H. J., Hearne, C. M., Cornall, R J., Prins, J.-B., McShane, P., Lathrop, G. M., Peterson, 1. B., Wicker, 1. 5., and Todd, J. A (1993). Nature Genet. 4, 404. Goodridge, A G., and Hanson, R W. (1986). "Metabolic Regulation: Application of Recombinant DNA Techniques." New York Acad. Sci. 478. Gordon, J. W., Scangos, G. A, Plotkin, D. J., Barbos, J. A, and Ruddle, F. H. (1980). Proc. Nat. Acad. Sci. USA 77, 7380. Grossman, M., Raper, S. E., Kozarsky, S. E., Stein, E. A, Engelhardt, J. F., Muller, D., Lupien, P. J., and Wilson, J. M. (1994). Nature Genet. 6, 335. Grosveld, F., and Kollias, G. (1992). "Transgenic Animals." Academic Press, New York.

DNA Technology

in

Diagnosis, Breeding, and Therapy

Gyapay, G., Morisette, J., Vignal, A, Dib, c., Fizames, c., Milasseau, P., Marc, S., Bernardi, G., Lathrop, M., and Weissenbach, J. (1994). Nature Genet. 7, 246. Hammer, R. E., Palmiter, R. D., and Brinster, R 1. (1984). Nature 316, 65.

Hammer, R. E., Pursel, V. G., Rexroad, C. E., Wall, R J., Bolt, D. J., Ebert, K. M., Palmiter, R D., and Brinster, R 1. (1985). Nature 315, 680. Hanahan, D. (1985). Nature 315, 1152. Hansel, W., and Weir, B. J. (1990). In "Genetic Engineering of Ani­ mals: Proceedings for the Second Symposium on Genetic Engi­ neering of Animals" (W. Hansel and B. J. Weir, eds.). J. Reprod.

Fertil. Suppl.

41.

Hanson, M. P., van Zandt, J. N., and Law, G. R (1967). Poult. Sci. 46, 1268. Harbitz, I., Chowdhary, B., Thomsen, P. D., Davies, W., Kaufmann, U., Kran, 5., Gustavsson, I., Christensen, K., and Hauge, J. G. (1990). Genomics 8, 243. Harbitz, 1., Kristensen, T., Bosnes, M., Kran, 5., and Davies, W. (1992). Anim. Genet. 23, 395. Hauge, (1988). Unpublished data. Healy, P. J., and Dennis, J. A. (1994). Anim. Genet. 25, 329. Hepkema. B. G., Blankert. J. J., Albers, G. A. A., Titanus, M. G. J., Egberts, E., van der Zijpp, S. A J., and Hensen, E. J. (1993). Anim. Genet. 24, 283. Hill, A V. 5., Alsopp, C. E. M., Kwiatkowski, D., Anstey, N. M., Twumasi, P., Rowe, P. A, Bennett, S., Brewster, D., McMichael, A J., and Greenwood, B. M. (1991). Nature 352, 595. Hobbs, H. H., Brown, M. S., and Goldstein, J. 1. (1992). Hum. Mut. 1, 445.

Hod, Y., Cook, J. 5., Weldon, S. 1., Short, J. M., Wynshaw-Boris, A, and Hanson, R W. (1986). In "Metabolic Regulation: Application of Recombinant DNA Techniques" (A G. Goodridge and R W. Hanson, eds.), pp. 31-45. New York Acad. Sci. 478. Hu, C-W. c., Lau, K 5., Griffin, T. A., Chuang, J. 1., Fischer, C. W., Cox, R. P., and Chuang, D. T. (1988). J. Bioi. Chem. 263, 9007. Hultman, T., Stahl, S., Homes E., and Uhlen, M. (1989). Nucl. Acids Res. 17, 4937. Hyde, S. c., Gill, D. R, Higgins, C. F., Trezise, A E. 0., MacVinish, 1. J., Cuthbert, A. W., Ratcliff, R, Evans, M. J., and Colledge, W. H. (1993). Nature 362, 250. Innerarity, T. 1., Mahley, R W., Weisgraber, K H., Bersot, T. P., Krauss, R M., Vega, G. I., Grundy, S. M., Friedl, W., Davignon, J., and McCarthy, B. J. (1990). ]. Lip. Res. 8, 1337. Jeffreys, A J., Wilson, V., and Thein, s. 1. (1985a). Nature 314, 67. Jeffreys, A. J., Wilson, V., and Thein, s. 1. (1985b). Nature 316, 76. Juul-Madsen, H. R, Hedemand, J. E., Salomonsen, J., and Simonsen, M. (1993). Anim. Genet. 24, 243. Kidd, V. J., Golbus, M. S., Wallace, R B., Itakura, K, and Woo, S. 1. C. (1984). N. Engl. ]. Med. 310, 639. Kleppe, K, Ohtsuka, E., Kleppe R, Molineux, I., and Khorana, H. G. (1971). f. Mol. BioI. 56, 341. Knip, M., Ilonen, J., Mustonen, A, and Akerblom, H. K (1986). Diabetologia 29, 347. Kostyu, D. D. (1991). Curro Opin. Genet. Devel. 1, 40. Krimpenfort, P. A, Rademakers, A., Eyestone, W., Van de Schans, A, Van den Broek, 5., Koiman, P., Kootwijk, E., Platenburg G., Pieper, F., Strijker, R, and de Boer, H. (1991). Bio/Technol. 9, 844. Lai, F. A, Erickson, H. P., Roussseau, E., Liu, Q.-Y., and Meissner, G. (1988). Nature 331, 315. Land, H., Parada, 1. F., and Weinberg, R A (1983). Science 222, 771. Le Meur, M., Gerlinger, P., Benois, c., and Mathis, D. (1985). Nature 316, 38.

43

Leary, J. J., Brigati, D. J., and Ward, D. C. (1983). Proc. Nat. Acad. Sci. USA 80, 4045. Leren, T. P., Solberg, K, R0dningen, O. K, R0sby, 0., Tonstad, 5., Ose, 1., and Berg, K (1993). Hum. Genet. 92, 6. Lewin, H. A (1994). Anim. Biotech. 5, 183. Little, P. F. R (1981). In "Genetic Engineering 1" (R Williamson, ed.), pp. 61-102. Academic Press, New York. Lovell-Badge, R H., Bygrave, A E., Bradley, A, Robertson, E., Evans, M. J., and Cheah, K S. E. (1985). Cold Spring Harb. Symp. Quant. BioI. 50, 707. Lu, K H., Gordon, I., Gallagher, M., and McGovern, H. (1987). Vet. Rec. 121, 259. Lucas, P. c., and Granner, D. K (1992). Ann. Rev. Biochem. 61, 1131. Lucassen, A M., Julier, c., Beressi, J. P., Beitard, c., Frogue!, P., Lathrop, M., and Bell, J. I. (1993). Nature Genet. 4, 305. McGrane, M. M., deVente, J., Yun, J., Bloom, J., Park, E., Wynshaw­ Boris, A., Wagner, T., Rottman, F. M., and Hanson, R. W. (1988). ]. Bioi. Chem. 263, 11443. Magram, J., Chada, K, and Costantini, F. (1985). Nature 315, 338. Marklund, 5., Ellegren, H., Eriksson, 5., Sandberg, K, and Andersson, 1. (1994). Anim. Genet. 25, 19. Marmur, J., and Doty, P. (1961). J. Mol. BioI. 3, 585. Marx, J. 1. (1988). Science 239, 463. Maxam, A M., and Gilbert, W. (1977). Proc. Nat. Acad. Sci. USA

74, 560.

Meissner, G. (1986). f. Bioi. Chern. 261, 6300. Mejdell, C. M., Lie, 0., Solbu, H., Amet, E. F., and Spooner, R 1. (1994). Anim. Genet. 25, 99. Miawa, S. (1994). Nature Genet. 6, 3. Michelsen, B., and Lemmark, A. (1987). ]. c/in. Invest. 79, 1144. Miller, A D. (1992). Science 357, 455. Montgomery, G. W., Crawford, A M., Penty, J. M., Dodds, K. G., Ede, J., Henry, H. M., Person, C. A, Lord, E. A, Galloway, S. M., Schmack, A E., Sise, J. A., Swarbrick, P. A, Hanraham, V., Buchanan, F. c., and Hill, D. F. (1993). Nature Genet. 4, 410. Moon Sung, A, Nordskog, A W., Lamont, S. J., and Warner, C. M. (1993). Anim. Genet. 24, 227. M0rsch, G., and Leibenguth, F. (1994). Anim. Genet. 25, 25. Nakamura, Y., Leppert, M., O'Connel, P., Wolf, R, Holm, T., Culver, M., Martin, c., Fujimoto, E., Hoff, M., Kumlin, E., and White, R (1987). Science 235, 1616. Neibergs, H. 1., Dietz, A B., and Womack, J. E. (1993). Anim. Genet. 24, 81. Nepom, G. T., and Erlich, H. (1991). Ann. Rev. Immunol. 9, 493. Notarianni, E., Galli, c., Laurie, S., Moor, R M., and Evans, M. J. (1990). Proc. 4th World Congo Genet. Appl. Livest. Prod. 9, 58. Nygaard, A P., and Hall, B. D. (1964). ]. Mol. Bioi. 9, 125. O'Brien, S. J., Womack, J. E., Lyons, 1. A, Moore, K J., Jenkins, N. A, and Copeland, N. G. (1993). Nature Genet. 3, 103. Old, R W., and Primrose, S. B. (1990), "Principles of Gene Manipula­ tion." Blackwell, Boston. Olson, M., Hood, 1., Cantor, c., and Botstein, D. (1989). Science 245, 1434. Orita, M., Suzuki, Y., Sekiya, T., and Hayashi, K (1989). Genomics 5, 874. Ornitz, D. M., Palmiter, R E., Hammer, R. E., Brinster, R 1., Swift, G. E., and MacDonald, R J. (1985). Nature 313, 600. Otsu, K, Khanna, V. K, Archibald, A 1., and MacLennan, D. H. (1991). Genomics 11, 744. Owerbach, D. A, Lemmark, A, Platz, P., Ryder, 1. P., Rask, 1., Peterson, P. A, and Ludvigsson, J. (1983). Nature 303, 815. Palmiter, R D., and Brinster, R 1. (1986). Ann. Rev. Genet. 20, 465. Park, C. A, Hines, H. c., Monke, D. R., and Threlfall, W. T. (1993). Anim. Genet. 24, 53.

44

Jens G. Hauge

Park, E. A, Gurney, A 1., Nizielski, S. E., Hakimi, P., Cao, Z., Moorman, A, Hanson, R W. (1993). ,. Bio/' Chern. 268, 613. Patel, Y. M., Yun, J. S., Liu, J., McGrane, M. M., and Hanson, R W. (1994). ,. BioI. Chern. 269, 5619. Proudfoot, N. J., Gil, A, and Maniatis, T. (1982). Cell 31, 553. Pursel, V. G., and Rexroad, C. E. (1993). ,. Anim. Sci. 71, 10. Pursel, V. G., Pinkert, C. A, Miller, K F., Bolt, D. J., Campbell, R G., Palmiter, R D., Brinster, R 1., and Hammer, R E. (1989). Science 244, 1281. Robertson, E., Bradley, A, Kuehn, M., and Evans, M. (1986). Nature 323, 445. Redningen, O. K, Leren, T. P., Resby, 0., Tonstad, D., Ose, 1., and Berg, K (1992). c/in. Genet. 41, 288. Rohrer, G. A, Alexander, 1. J., Keele, J. W., Smith, T. P., and Beattie, C. W. (1994). Genetics 136, 231. Rosenfeld, M. A, Siegfried, W., Yoshimura, K, Yoneyama, K, Fukay­ ama, M., Stier, 1. E., Paakko, P. K, Gilardi, P., Stratford­ Perricaudet, 1. D., Perricaudet, M., Jallat, S., Pavirani, A, Lelcocq, J. P., and Crystal, R G. (1991). Science 252, 431. Rudolph, J. A, Spier, S. J., Byrns, G., and Hoffman, E. P. (1992a). Anim. Genet. 23, 241. Rudolph, J. A, Spier, S. J., Byrns, G., Rojas, C. V., Bernoco, D., and Hoffman, E. P. (1992b). Nature Genet. 2, 144. Rusconi, S., and Kehler, G. (1985). Nature 314, 330. Saiki, R, Scharf, S., Faloona, F., Mullis, K. B., Horn, G. T., Erlich, H. A., and Amheim, N. (1985). Science 230, 1350. Saiki, R K, Gelfand, D. H., Stoffel, S., Scharf, S. J., Higuchi, R., Horn, G. T., Mullis, K B., and Erlich, H. A (1988). Science 239, 487. Salter, D. W., and Crittenden, 1. B. (1989). Theor. Appl. Genet. 77, 457. Sanger, F., Nicklen, S., and Coulsen, A R (1979). Proc. Nat. Acad. Sci USA 74, 5463. Sanger, F., Coulson, A R, Hong, G. F., Hill, D. F., and Petersen, G. B. (1982). ,. Molec. Bio!. 162, 729. Scharf, S. J., Griffith, R 1., and Erlich, H. A (1991). Hum. Immun. 30, 190. SeMber, S., Simon, D., and Schwenger, B. (1992). Gene 124, 307. Schuster, D. E., Bosworth, B. T., and Kehrli, M. E. (1992a). Gene 114, 267. Schuster, D. E., Kehrli, M. E., Ackerman, M. R, and Gilbert, R O. (1992b). Proc. Nat. Acad. Sci. USA 89, 9225. Skolnick, M. H., and White, R (1982). Cytogenet. Cell Genet. 32, 58. Smithies, 0., Gregg, R c., Boggs, S. S., Koralewski, M. A, and Kucherlapati, R. S. (1985). Nature 317, 230. Snouwaert, J. N., Brigman, K K, Latour, A. M., Malouf, N. N., Boucher, R c., Smithies, 0., and Koller, B. H. (1992). Science 257, 1083. Solomon, M. B., Pursel, V. G., Paroczay, E. W., and Bolt, D. J. (1994). ,. Anim. Sci. 72, 1242. Southern, E. M. (1979). Meth. Enzymol. 68, 152. Stryer, 1. (1981). "Biochemistry." Freeman, San Frandsco. Swanson, M. E., Martin, M. J., O'Donnel, J. K, Hoover, K, Lago, W., Huntress, V., Parsons, C. T., Pinkert, C. A, Pilder, S., and Logan, J. S. (1992). Biotech. 10, 557.

Swift, G. H., Hammer, R. E., MacDonald, R J., and Brinster, R 1. (1984). Cell 38, 639. Syvanen, A C. (1986). Med. BioI. 64, 313. Szalai, G., Bailey, E., Gerber, H., and Lazary, S. (1993). Anim. Genet. 24, 187. Takeshima, H., Nishimura, S., Matsumoto, T., Ishida, H., Kanagawa, K, Minamino, N., Matsua, H., Veda, M., Hanaoka, M., Hirose, T., and Numa, S. (1989). Nature 339, 439. Thomas, K R, and Capecchi, M. R (1986). Nature 324, 34. Thorsby, E., and Renningen, K S. (1993). Diabetologia 36, 371. Todd, J. A, Bell, J. 1., and McDevitt, H. O. (1987). Nature 329, 599. Tolstoshev, P., and Anderson, W. F. (1993). In "Genome Research in Molecular Medicine and Virology" (K W. Adolph, ed.), pp. 35-50. Academic Press, New York. VAge, D. 1., Olsaker, 1., Lingaas, F., and Lie, 0. (1994). Anim. Genet. 25, 73. Van de Woude, G. F., Levine, A J., Topp, W. c., and Watson, J. D. (1984). "Oncogenes and Viral Genes," Cold Spring Harbor Press, New York. Velander, W. H., Johnson, J. 1., Page, R 1., Russel, C. G., Subra­ manian, A., Wilkins, T. D., Gwazdauskas, F. c., Pittius, c., and Drohan, W. N. (1992). Proc. Nat. Acad. Sci. USA 89, 12003. Vincent, N., Ragot, T., Gilgenkrantz, H., Couton, D., Chafey, P., Gregoire, A, Briand, H., Kaplan, J. c., Kahn, A, and Perri-Caudet, M. (1993). Nature Genet. 5, 130. Vogel, F., and Motulsky, A G. (1986). "Human Genetics." Springer, New York. Watson, J. D., and Crick, F. H. C. (1953). Nature 171, 737. Watson, J. D., Gilman, M., Wirkowsky, J., and Zoller, M. (1992). "Recombinant DNA" Freeman, New York. Weber, J. 1., and May, P. E. (1989). Am. ,. Hum. Genet. 44, 388. Weidle, V. H., Lenz, H., and Brem, G. (1991). Gene 98, 185. Weinberg, R A (1991). Science 254, 1138. Weissenbach, J., Gyapay, G., Dib, c., Vignal, A, Morisette, J., Milasseau, P., Vaysseix, G., and Lathrop, M. (1992). Nature 359, 794. Westphal, H. (1987). Bio Essays 6, 73. Wijsman, E. M. (1984). Nucl. Acids Res. 12, 9209 . Williamson, R. (1993). Nature Genet. 3, 195. Wivel, N. A (1993). Trends Biotech. 11, 189. Wright, G., Carver, A, Cottom, D., Reeves, D., Scott, A, Simons, P., Wilmut, 1., Gamer, 1., and Colman, A (1991). Biotech. 9, 830. Yamamoto, K K, Gonzales, G. A, Biggs, W. H., and Montminy, M. R (1988). Nature 334, 494. Young, B. D., and Anderson, M. 1. H. (1985). In "Nucleic Add Hybridization-A Practical Approach" (B. D. Hames and S. J. Higgins, eds.), pp. 47-111. IRL Press, Washington, D.C. Zabner, J., Petersen, D. M., Puga, A P., Graham, S. M., Couture, 1. A, Keyes, 1. D., Lukason, M. J., St. George, J. A, Gregory, R J., Smith, A E., and Welsh, M. J. (1994). Nature Genet. 6, 75. Zarbl, H., Saraswati, S., Arthur, A. V., Dionisio, M. Z., and Barbadd, M. (1985). Nature 315, 382. Zhang, B., Healy, P. J., Crabb, D. W., and Harris, R A (1990). ,. BioI. Chern. 265, 2425.

C H A P T E R

3 Carbohydrate Metabolism and Its Diseases J. JERRY KANEKO

1. INTRODUCTION 45 II. DIGESTION 46 III. ABSORPTION 46 IV. METABOLISM OF ABSORBED CARBOHYDRATE 46 A. General 46 B. Storage as Glycogen 47 C. Glycogen Metabolism 48 D. Catabolism of Glucose 50 V. INTERRELATIONSHIPS OF CARBOHYDRATE, LIPID, AND PROTEIN METABOLISM 56 A. Lipid Metabolism 56 B. The Influence of Glucose Oxidation on Lipid Metabolism 58 VI. INSULIN AND CARBOHYDRATE METABOLISM 58 A. Proinsulin and Insulin 58 B. Insulin Transport 60 C. Glucose Transport 60 D. Insulin Action on Biochemical Systems 60 E. Physiological Effects of Insulin 60 F. Other Pancreatic Islet Hormones 61 VII. BLOOD GLUCOSE AND ITS REGULATION 61 A. General 61 B. Glucose Supply and Removal 62 C. The Role of the Liver 62 D. Glucose Tolerance 63 VIII. METHODOLOGY 63 A. Blood Glucose 63 B. Indirect MOnitoring of Blood Glucose 64 C. Tolerance Tests 65 D. Ketone Bodies 68 IX. DISORDERS OF CARBOHYDRATE METABOLISM 68 A. Diabetes Mellitus 68

CUNICAL BIOCHEMISTRY OF DOMESTIC ANIMALS, FIFTH EDITION

B. Hyperinsulinism 74 C. Hypoglycemia of Baby Pigs 75 D. Glycogen Storage Diseases 75 X. DISORDERS OF RUMINANTS ASSOCIATED WITH HYPOGLYCEMIA 76 A. General 76 B. Carbohydrate Balance 77 C. Biochemical Alterations in Body Fluids 79 D. Ruminant Ketosis 80 References 80

I. INTRODUCTION The biochemical mechanisms by which the chemical energy contained in foodstuffs is made available to the animal are collectively described as metabolism. Thus, the description of the metabolism of a foodstuff encom­ passes the biochemical events that occur from the mo­ ment of ingestion to its final breakdown and excretion. Classically, these biochemical events have been di­ vided into the metabolism of the three major constit­ uents of food: carbohydrates, proteins, and lipids. The metabolism of the lipids and proteins is discussed in other chapters. The major function of ingested carbohydrates is to serve as energy sources and their storage function is relatively minor. Carbohydrates also function as pre­ cursors of essential intermediates for use in synthetic processes. When the metabolic machinery of an animal is disrupted, a disease state prevails, for example, dia­ betes. The literature of the biochemistry of metabolism

45

Copyright © 1997 by Academic Press All rights of reproduction in any fonn reserved.

46

J. Jerry Kaneko

and disease continues to expand as the intricate details of individual and overall reaction mechanisms are con­ tinually refined and elucidated. Additionally, modern molecular approaches have significantly increased our understanding of disease mechanisms and remain fer­ tile fields for investigations into the disease processes. This chapter is presented as a basis for the better under­ standing of the biochemical mechanisms underlying those diseases associated with carbohydrate metabo­ lism. Exhaustive treatment of carbohydrate metabo­ lism in health and disease is beyond the scope of this chapter.

across the intestinal mucosa is thought to be by phos­ phorylation in the mucosal cell. The phosphorylated sugars are then transferred across the mucosal cell, rehydrolyzed, and free glucose appears in the portal circulation for transport to the liver. The possible role of glucose transporters in this event is not yet elucidated but they are very likely to be involved. Glucose transporters are known to be involved in many tissues, including brain, erythro­ cytes, kidney, liver, pancreas, skeletal muscle, heart muscle, and fat cells.

IV. METABOLISM OF ABSORBED CARBOHYDRATE

II. DIGESTION The digestion of carbohydrates in the animal begins with the initial contact of these carbohydrates with the enzymes of salivary juice. Starch of plant foods and glycogen of meat are split into their constituent mono­ saccharides by the action of amylase and maltase. This activity ceases as the food matter passes into the stom­ ach, where the enzymatic action is destroyed by the hydrochlOric acid. Within the stomach, acid hydrolysis may occur, but the stomach empties too rapidly for complete hydrolysis to take place. Thus, only a small portion of the ingested carbohydrate is hydrolyzed prior to passage into the small intestine. In the small intestine, digestion of carbohydrate takes place quickly by the carbohydrate splitting enzymes contained in the copious quantities of pancreatic juice and in the succus entericus. Starch and glycogen are hydrolyzed to glu­ cose by amylase and maltase; lactose to glucose and galactose by lactase; and sucrose to glucose and fruc­ tose by sucrase (sucrose-a-glucosidase). The monosac­ charide products of enzymatic hydrolysis of carbo­ hydrates, glucose, fructose, and galactose, are the principal forms in which absorption occurs in the mono­ gastric animal.

III. ABSORPTION The monosaccharides are almost completely ab­ sorbed through the mucosa of the small intestine and appear in the portal circulation as the free sugars. Ab­ sorption occurs by two methods: (1) facilitated diffu­ sion and (2) sodium-dependent active transport. Glu­ cose and galactose are absorbed rapidly and by both methods. Fructose is absorbed at about half the rate of glucose with a portion being converted to glucose in the process. Other monosaccharides, for example, mannose, are absorbed slowly at a rate consistent with a diffusion process. The active absorption of glucose

A. General Liver cells are readily permeable to the absorbed glucose. This process is facilitated by glucose trans­ porter (GLUT) proteins within the plasma membrane, in particular, GLUT-2 is the transporter in the liver cell plasma membrane (Thorens et al., 1988). Within the liver, there are several pathways by which the immedi­ ate fate of the absorbed hexose is determined. Glucose, fructose, and galactose first enter the general metabolic scheme through a series of complex reactions to form glucose phosphates (Fig. 3.1). The enzyme, galactose­ I-P uridyl transferase, which catalyzes the reaction galactose-l-P + UDP-glucose � UDP-galactose + glucose-l-P, is blocked or deficient in congenital galactosemia of humans. The glucose phosphates are then converted

0 OOP-G

Gal-1-P

UTP

��---.

l

UDP-Ga\ ----- UDP-Glucose

Glucose-1-P

I Glucose � Gluco Fructose

+ATP

L

l ---s-p

n

Fructose-S-P

I '6� �"G +ATP

Mannose-S-P

Fructose-1-S-P

Fructose-1-P

H

A-3-P

Glyceraldehyde

FIGURE 3.1 Pathways for hexose metabolism. ATP, adenosine triphosphate; UTP, uridine triphosphate; UDP-G, uridine diphos­ phate glucose; DHAP, dihydroxy acetone phosphate; GA-3-P, glycer­ aldehyde-3-phosphate.

47

Carbohydrate Metabolism and Its Diseases TABLE 3.1

Liver Glycogen Content of Animals

Species

Glycogen in liver (%)

Reference

Dog Sheep Cow (lactating) Cow (nonlactating) Baby pig Baby pig (newborn)

6.1 3.8 1.0 3.0 5.2 14.8

Lusk (1928) Roderick et al. (1933) Kronfeld et al. (1960) Kronfeld et al. (1960) Morrill (1952) Swiatek et al. (1968)

to and stored as glycogen, catabolized to CO2 and water or, as free glucose, returned to the general circulation. Essentially, intermediate carbohydrate metabolism of animals evolves about the metabolism of glucose, and the liver is the organ of prime importance.

cosyl units. The external chains beyond the last 1-6 link are longer and contain between 7 to 10 glucose units. The molecular weight of glycogen may be as high as 4 X 106 Mr and contain about 20,000 glucosyl units. In Table 3.2, the amount of carbohydrate available to meet the theoretical requirements of a hypothetical dog is shown. The amount present is sufficient for about half a day. It is apparent that the needs of the body which must be continually met are satisfied by alternate means and not solely dependent on continu­ ous ingestion of carbohydrates. During and after feed­ ing (postprandial), absorbed hexoses are converted to glucose by the liver and enter the general circulation. Excesses are stored as glycogen or as fat. During the fasting or postabsorptive state, glucose is supplied by the conversion of protein (gluconeogenesis) and by the breakdown of glycogen (glycogenolysis). The contin­ ued rapid synthesis and breakdown of glycogen that

B. Storage as Glycogen Glycogen is the chief storage form of carbohydrate in animals and is analogous to the storage of starch by plants. It is found primarily in liver and in muscle, where it occurs at about 3-6% and about 0.5%, respec­ tively (Table 3.1). Glycogen is comprised solely of a­ D-glucose units linked together through carbon atoms 1 and 4 or 1 and 6. Straight chains of glucose units are formed by the 1-4 links and these are cross-linked by the 1-6 links. The result is a complex ramification of chains of glucosyl units with branch points at the site of the 1-6 links (Fig. 3.2). The internal chains of the glycogen molecule have an average length of four glu-

TABLE 3.2

Carbohydrate Content of a Dog"

Muscle glycogen (0.5%) Liver glycogen (6%) Carbohydrate in fluids 5.5 rnrnol/liter (100 mg/ dl)

Caloric value (45.2 X 4 kcaI/ g) 181 KcaTCaloric requirement (70 kgf = 70 X 5.6) 392 kcal / day 181 X 24 hours 11 hours 392 =

25.0 g 18.0 g

� 45.2 g

=

=

a Body weight, 10 kg; liver weight, 300 g; muscle weight, 5 kg; volume of blood and extracellular fluid, 2.2 liters.

FIGURE 3.2 Glycogen structure. Note that hydrolysis of a 1-6 link by the de­ brancher enzyme yields a mole of free glucose.

48

J. Jerry Kaneko

is, turnover, is well illustrated by the biological half­ time of glycogen, which is about a day.

C. Glycogen Metabolism

1. Glycogenesis

The initial reaction required for the entrance of glu­ cose into the series of metabolic reactions that culmi­ �ates in the synthesis of glycogen is the phosphoryla­ tion of glucose at the C-6 position. Glucose is phosphorylated with adenosine triphosphate (ATP) in liver by an irreversible enzymatic reaction catalyzed by a specific glucokinase (GK): glucose + ATP

GIC

�

(Hl .....I Cl

50

Cl

0

0"-

0

2

TIME (hours)

3

0

FIGURE 3.15 Insulin tolerance in the dog. Curves falling in the shaded areas are described as noted.

3. Glucagon Stimulation Test

4. Epinephrine Tolerance Test Epinephrine also has a postinjection hyperglycemic effect via hepatic glycogenolysis. The blood glucose level rises to a maximum of 50% above the fasting level in 40-60 minutes and returns to the original level in 1 .5-2 hours. The test is performed by obtaining a fast­ ing blood sample (0 time), injecting 1 ml of 1 : 1000

68

J. Jerry Kaneko

epinephrine-HCL (in the dog) intramuscularly and ob­ taining blood samples every 30 minutes for 3 hours. The characteristic increase in blood glucose is used as an index of the availability of liver glycogen for the production of blood glucose. On the basis of a lowered response to epinephrine, liver glycogen can indirectly be shown to be depleted in bovine ketosis. This can be confirmed directly by measurement of glycogen in biopsy samples. A lowered glycemic response is also a characteristic response of the glycogen storage diseases where glycogenolysis is inhibited by enzyme defi­ ciencies.

5. Leucine-Induced Hypoglycemia The oral administration of L-leucine induces a marked and persistent hypoglycemia in hyperinsulin­ ism due to pancreatic islet cell tumors. The hypoglyce­ mia is associated with a rise in plasma insulin due to increased release of insulin by the tumorous islet cells. The test is performed by the oral administration of 150 mg L-leucine / kg body weight as an aqueous suspen­ sion to the fasting dog. A fasting blood glucose sample is taken before administration (0 time) and every 30 minutes for 6 hours. A hypoglycemic effect is seen quickly at 0.5-1 hour and may persist for as long as 6 hours in hyperinsulinism. The normal dog exhibits no hypoglycemic effect.

6. Tolbutamide Test The intravenous administration of tolbutamide, an oral hypoglycemic agent, induces the release of insulin from the pancreas and is used as a test of the availabil­ ity of insulin from the pancreas. The blood glucose curve during the test parallels the insulin tolerance test. This test has not been used in animals.

D. Ketone Bodies The methodology and role of ketone bodies in the carbohydrate economy of animals in health and disease are discussed in the chapter on lipid metabolism. The major ketone bodies are acetone, AcAc, and 3-0H­ butyrate (3-0H-B). The 3-0H-B is the precursor of ace­ tone and AcAc so that the measurement of any or all in body fluids is a standard method to evaluate ketosis and ketoacidosis. Additionally, 3-0H-B constitutes one-half or more of the total ketone bodies. Recently,

J.OH·BD

3-0H-butyrate + NAD+ � acetoacetate + NADH + H+ NADH + NBT (ox)

�

NAD+ + NBT (red).

A rapid, reliable method would have decided advan­ tages in the management of diabetic ketoacidosis, bo­ vine ketosis, and ovine pregnancy toxemia.

IX. DISORDERS OF CARBOHYDRATE METABOLISM Although alterations in blood glucose levels occur in a wide variety of disease states, they are of particular importance in the endocrine disorders. Normal blood glucose levels are the result of a finely balanced system of hormonal interaction affecting the mechanisms of supply and removal from the circulation. When a hor­ monal imbalance occurs, a new equilibrium is estab­ lished. Whether this equilibrium is clinically evident as a persistent hypoglycemia or hyperglycemia de­ pends on the total interaction of the hormonal influ­ ences on carbohydrate metabolism. Further discus­ sions concerning the disorders of the pituitary, adrenals, and the thyroids are given in their respective chapters. The following sections discuss the conditions in which the principal manifestations are closely re­ lated to derangements in carbohydrate metabolisms.

A. Diabetes Mellitus Although diabetes mellitus has been reported in virtually all laboratory animals (gerbils, guinea pigs, hamsters, mice, rats, nonhuman primates) and in horses, cattle, sheep, and pigs, it is most frequently found in dogs and cats. Estimates of the incidence of diabetes range as high as 1 : 66 (1 .52%) for dogs and 1 : 800 for cats. Diabetes mellitus in animals has been frequently reviewed (Cotton et al., 1971; Foster, 1975; Ling et al., 1977; Engerman and Kramer, 1982; Kaneko and Howard, 1989).

1. Natural History of Diabetes The disease in dogs occurs most frequently in the mature or older female, often in association with estrus, and in all breeds. In contrast, male cats appear to be more commonly affected than females. In the dog, dia­

a point-of-care enzymatic and colorimetric method has been developed for assay of plasma 3-0H-B (GDS Di­ agnostics, Elkhart, IN). The method is based on the enzyme 3-0H-B dehydrogenase (3-OH-BD) and nitro­

betes is frequently associated with obesity and it is

blue tetrazolium (NBT):

is significantly impaired, suggesting that obesity also

now known that obesity is the single most important contributing factor to the development of diabetes (Mattheeuws et al., 1984). In the obese cat, the GTT

69

Carbohydrate Metabolism and Its Diseases

absence or relative (Type II) due to insulin antibodies, receptor defects, or deficiencies.

predisposes cats to diabetes (Nelson et al., 1990). The obese cat also has a GST response like that of the Type II diabetic (Kirk et al., 1993). Little is known of the genetic aspects of diabetes in animals as compared to humans in which the hereditary predisposition is well known. Insulin genes are identified, cloned, and hu­ man insulin produced by biotechnology that is now widely used. Diabetes has, however, been reported in the offspring of diabetic dogs (Gershwin, 1975; Kaneko et al., 1977). Kramer (1977, 1981) and Kramer et al. (1980) reported their observations on hereditary diabetes in a family of keeshonds and Williams et al., (1981) re­ ported hereditary diabetes in golden retrievers. On the basis of serum insulin (I) response patterns during the IVGTT, diabetes mellitus of dogs can be divided into at least three types, Types I, II, and III (Table 3.8; Kaneko et al., 1977, 1979). Type I dogs are characterized by no or very low initial [ (1 0) level and no [ response to the glucose load similar to juvenile diabetes (Type I) in children. Type II is characterized by a normal or high [ 0 and, again, no increment of [ response to the glucose, which are features of the matu­ rity onset form (Type II) of diabetes in humans. Type III is characterized by a normal [ 0, a normal or delayed I response to the glucose, and a delayed return of I to normal at 60 minutes as seen in chemical diabetes (Type III). Types II and III were each later further subdivided into the obese and nonobese types (Mat­ theeuws et al., 1984). All diabetics are glucose intolerant so their separa­ tion into the various types depends on their insulin response patterns in the IVGTT. The importance of defining their type is the likelihood that Types II and III obese diabetics are the most likely subjects for suc­ cessful dietary therapy and/ or oral hypoglycemic ther­ apy. Nelson et al. (1993) have successfully treated some diabetic cats with oral sulfonylureas. It is also likely that in the natural history of diabetes, Type III (chemical) diabetes precedes the development of Types I and II depending on the nature of the insulin deficiency, whether absolute (Type I) due to islet cell

TABLE 3.8

2. The Etiology of Diabetes The most frequent contributory factors to the onset of diabetes are pancreatitis, obesity, infection, stress, and estrum. The possibility of a viral etiology has also been reviewed (Steinke and Taylor, 1974), and Yoon et al. (1979) isolated a virus from a patient who died and the virus produced diabetes in mice. Currently, autoimmunity is considered to be the fundamental cause of Type I diabetes, possibly as an aftermath to the viral infection. Autoimmunity is evi­ denced by the lymphocytic infiltration associated with immune processes, and lymphocytic infiltration is found in diabetes of cattle and humans. It is also ob­ served in cattle or rabbits immunized with bovine insu­ lin. Eisenbarth (1986) reviewed the evidence that Type I diabetes in the biobreeding (BB) rat and the nonobese diabetic (NOD) mouse and Type I diabetes of humans is an autoimmune disease. Taniyama et al. (1995) found lymphocytic infiltration in all pancreata of four diabetic cattle and their studies were suggestive of Type I diabe­ tes as an autoimmune disease. In the cat, several studies indicate that there is a strong correlation between pancreatic insular amyloid­ osis and diabetes although the amyloid does not ap­ pear to be the primary cause (O'Brien et al., 1985). The high estimates of the incidence of diabetes is an indication of its importance as a clinical disease entity. Furthermore, the similarities of the clinical pic­ ture of diabetes with other wasting diseases character­ ized by polyuria and polydypsia attest to the impor­ tance of laboratory examinations for the early and accurate diagnosis of diabetes. In no other disease is an understanding of the metabolic alternations so im­ portant in diagnosis and proper treatment. The fundamental defect in diabetes mellitus is an absolute (due to absence of pancreatic (3 cells) or rela­ tive (insulin resistance due to insulin antibodies, recep­ tor defect, or deficiency) lack of insulin resulting in an

Diagnostic Criteria for Types of Diabetes Mellitus in Dogs IV Gn

Diabetes type I

II

ill

Normal

Insulin'

Fasting glucose Go (mgldl)

Tt (m)

k (%/min)

> 200 > 200 100-200 70-110

>70 >70 >45 15-45

20 JLU I liter) with a hypoglycemia of 110 mg / dl) and, unless the pig is fed, its blood glucose drops rapidly to hypoglycemic levels within 24-36 hours. The liver glycogen, which is high (14.8%) at birth, is almost totally absent at death. In contrast, newborn lambs, calves, and foals are able to resist starvation hypoglycemia for more than a week. If the baby pig suckles, its ability to withstand starva­ tion progressively increases from the day of birth. A lO-day-old baby pig can be starved up to 3 weeks before symptoms of hypoglycemia occur. Gluconeogenic mechanisms are undeveloped in the newborn pig, which indicates that the gluconeogenic enzymes of the baby pig are inadequate at birth. This also indicates that these enzymes need to be induced by feeding so they can reach their maximal activities within 1 or 2 weeks after birth. The precise hepatic gluconeogenic enzymes and their induceability by feeding have not yet been identified. The association of baby pig hypoglycemia with com­ plete or partial starvation is shown by the findings that their stomachs are empty at necropsy and the syndrome itself is indistinguishable from experimental starvation of the newborn baby pig. Starvation of the newborn pig under natural conditions can occur due to factors relating to the sow (agalactia, metritis, etc.) or to the health of the baby pig (anemia, infections, etc.), either case resulting in inadequate food intake. The requirement for feeding to induce the hepatic glu­ coneogenic mechanisms in the newborn baby pig ex­ plains its inability to withstand starvation in contrast to the newborn lamb, calf, or foal, which is born with fully functioning hepatic gluconeogenesis.

D. Glycogen Storage Diseases The glycogen storage diseases are characterized by the pathological accumulation of glycogen in tissues. Based on their patterns of glycogen accumulation, their clinical pathological findings, their enzymes of glyco­ gen metabolism, and the structural analyses of their glycogen, the GSDs in humans have been classified into eight types (Howell and Williams, 1983). All have an autosomal recessive mode of inheritance except for GSD VIII, which is sex linked. Their glycogen struc­ tures are normal except for Types III and IV.

76

J. Jerry Kaneko

Type I or classical von Gierke's disease is character­ ized by increased liver glycogen leading to a marked hepatomegaly. There is a marked hypoglycemia and the blood glucose response to epinephrine or glucagon is minimal or absent. The liver glycogen structure is normal. The defect in this disease is a deficiency of the enzyme G-6-Pase. Type II or Pompe's disease is a generalized glycogenosis with lysosomal accumula­ tion of glycogen and early death. The defect in this disease is a deficiency of acid-a-glucosidase (AAGase). In Type III or Cori's disease, the debrancher enzyme is deficient, which leads to the accumulation of glyco­ gen of abnormal structure. The branches are abnor­ mally short and there are an increased number of branch points; it is a limit dextrin and the disease is sometimes called a limit dextrinosis. There is a variable hypoglycemia, little or no response to epinephrine or glucagon, hepatomegaly, cardiomegaly, and early death. In Type IV or Andersen's disease the brancher enzyme is deficient, which leads to a glycogen with abnormally long branches and few branch points. It is clinically similar to Type III. In Type V or McArdle's disease muscle phosphorylase (MPase) is deficient, whereas in Type VI, it is liver phosphorylase (LPase), which is deficient. Type VII or Tarui's disease is charac­ terized by a deficiency of muscle phosphofructokinase (PFK) with accumulation of glycogen in muscle, and Type VIII is deficient in leukocyte or hepatic phos­ phorylase b kinase (PBK). This disease is uniquely sex linked. Of these eight types in humans, only Types l, II, III, and VIII are found in animals. Other forms of glycogen storage in animals are described as GSD-like based on their pathological patterns of glycogen accumulation. GSD in animals has been reviewed by Walvoort (1983). There is an inherited PFK deficiency in the springer spaniel dog but unlike human Type VII GSD, there is no muscle pathology or glycogen accumulation in muscle. The deficiency in the dog is expressed as a hemolytic anemia caused by a deficiency of the PFK isoenzyme in the erythrocytes and is rightly considered to be an inherited erthrocyte enzyme deficiency rather than a GSD (Giger et al., 1985). Mammalian PFK is present in tissues as tetramers composed of combina­ tions of three different subunits: PFK-M (muscle), PFK­ L (liver), and PFK-P (platelets). Human and dog muscle and liver have homogenous tetrameric PFK-� and PFK-L4 respectively. Human erythrocyte PFK is a ' mixed tetramer, PFK-Lz l PFK-M2, whereas the dog erythrocyte PFK is a mixed tetramer, PFK-Mz l PFK-P2 (Vora et al., 1985). In PFK-M subunit deficiency in the dog erythrocyte, PFK-L replaces PFK-M; PFK-Lz l PFK­ P2• In the human erythrocyte, PFK-P replaces PFK-M; PFK-L2 / PFK-P2• Although the substituted PFK in the

erythrocyte is the same in dog or human, the deficiency in the human is expressed as a GSD, whereas in the dog it is expressed as an exertional hemolytic anemia. The anemia occurs after heavy exertional respiratory stress such as is experienced in vigorous hunting or exercise. Hyperventilation induces a respiratory alka­ losis, which in tum increases the fragility of the eryth­ rocyte and the hemolytic anemia occurs (Giger et al., 1985). A radiation-induced Type I GSD occurs as an auto­ somal recessive condition in the C3H mouse and is characterized by hypoglycemia, early death, and a de­ ficiency of liver G-6-Pase (Gluecksohn-Welch, 1979). Type n GSD has been described in Brahman cattle (O'Sullivan et al., 1981 ), the Lapland dog (Walvoort et al., 1982), and in the Japanese quail (Murakami et al., 1980). In the Brahman cattle, Type II is characterized by early death, generalized glycogen deposition, and a marked decrease in AAGase activity. It is inherited as an autosomal recessive. In the Lapland dog, there is also early death, generalized glycogen deposition, hepatomegaly, and cardiomegaly. There is also a marked decrease in heart and liver AAGase. The Japa­ nese quail with Type II is also characterized by early death, glycogen depOSition in the heart, liver, and mus­ cles, and decreased AAGase. Type III occurs in the German shepherd dog and is characterized by early death, little or no response to epinephrine or glucagon, hepatomegaly, and cardio­ megaly with glycogen accumulation. The glycogen has a limit dextrin structure and there is a very low debrancher enzyme activity in liver and muscle (Svenkerud and Hauge, 1978; Ceh et al., 1976). Type VIII is seen in the rat and the mouse. In the rat, the disease is inherited as an autosomal recessive, appears healthy but is hypoglycemic, has hepatomeg­ aly due to glycogen accumulation in the liver, and also has a very low liver phosphorylase kinase activity (Clark et al., 1980). The affected mouse is apparently healthy but has increased glycogen accumulation in the muscle with a very low muscle PBK. The inheritance is sex linked (Gross, 1975).

x. DISORDERS OF RUMINANTS ASSOCIATED WITH HYPOGLYCEMIA

A. General The principal disorders of domestic ruminants in which hypoglycemia is a salient feature are bovine ketosis and ovine pregnancy toxemia. Pregnancy toxe­ mia characteristically is a widespread disease of high mortality occurring in the pregnant ewe just prior to

77

Carbohydrate Metabolism and Its Diseases

one-third to one-half of the daily glucose turnover of 100 g is utilized by the fetus. A good approach to assess the glucose requirements of an animal is to measure its turnover rate or the rate at which glucose enters or leaves the circulation. This is best measured by the use of isotopically labeled glucose and has been used in lactating cows. It has been estimated to be 1440 g/ day (60 g / hour) in cows and about 144 g/ day (6 g / hour) in normal pregnant ewes just prior to term.

term, the time when carbohydrate demands are high­ est, especially in those ewes carrying more than one fetus. Bovine ketosis, on the other hand, occurs in the high-producing dairy cow characteristically during the early stages of lactation, when milk production is gen­ erally the highest. Abnormally high levels of the ketone bodies, acetone, AcAc, 3-0H-B, and isopropanol ap­ pear in blood, urine, and the milk. These alterations are accompanied by the clinical signs of ketosis: loss of appetite, weight loss, decrease in milk production, and nervous disturbances. The energy metabolism of the ru�inant is focused on the utilization of the volatile fatty acids produced by rumen fermentation rather than on carbohydrates as in the nonruminant. The carbohydrate economy of the ruminant is significantly different from that of the nonruminant and an appreciation of these differences is important to the understanding of these metabolic disorders of the ruminant.

2. Glucose Sources The large amounts of indigestible carbohydrates in­ gested by ruminants are fermented to volatile fatty acids by the rumen microflora. Little, if any, of the digestible carbohydrates (starch, glucose) in the diet escapes this fermentation so that glucose absorption from the digestive tract accounts for virtually none of the daily glucose requirement of ruminants. However, if any glucose escapes rumen fermentation, for exam­ ple, in gastrointestinal disease, it is readily absorbed. An indirect source of blood glucose is ruminal lactic acid. Lactic acid is a product of many fermentation reactions and ruminal lactate can be absorbed. The blood lactate can be a source of blood glucose via the lactic acid cycle (Fig. 3.4). However, the principal source of blood lactate is the breakdown of muscle glycogen. Therefore, some of the ruminant's glucose requirement may be met by lactate but this is minimal because excess lactic acid in the rumen is toxic.

B. Carbohydrate Balance 1. Glucose Requirements The heavy demands for glucose in early lactation and in late pregnancy are well known. In the late 1950s, Kleiber calculated that about 60% of the lactating cows' daily glucose requirement is for the production of milk. The balance sheet (Table 3.10) indicates a total daily glucose requirement of 1 140 g of which 700 g appear in the milk. For sheep in late pregnancy, about TABLE 3.10

Carbohydrate Balance Sheet"

A. Cow's daily glucose flux 1. In 12.5 kg milk: 610 g lactose 462 g milk fat with 58 g glycerol Carbohydrate carbon in milk/ day 2. Daily glucose catabolism: Cow produced daily 3288 liters CO2 1762 g C Transfer quotient plasma gluocse --+ CO2 is 0.1 Thus glucose to CO2/ day = 1 + 2 daily flux of glucose

Carbohydrate carbon 257 g C/ day 23 g C/ day 280 g C/ day

=

=

72 X 456 1 80

=

1140 g glucose/ day

B. Cows glucose sources Cow secreted daily in urine 34 g N, indicating catabolism of 213 g protein less c in urea = Maximum available for glucose synthesis from protein = Glucose flow in milk and respiration Thus glucose flow from nonprotein sources = =

�20 X 360

a

176 g C/ day 456 g C/ day

=

100 g C/ day 14 g C/ day

96 g C/ day 456 g C / day 360 g C/ day

= 900 g glucose daily must have been supplied from a non-protein source

From Kleiber (1959).

78

J. Jerry Kaneko

The carbohydrate balance she-.:t (Table 3.10) pro­ vides the contribution of protein as a source of carbohy­ drate for the lactating cow. Because glucose absorption in the ruminant is minimal, the balance sheet also illus­ trates the importance of an alternate nonprotein source of carbohydrate carbon. These sources are the ruminal volatile fatty acids. The principal products of rumen fermentation are the volatile fatty acids, acetate, propi­ onate, and butyrate. These acids are absorbed across the rumen wall and are the major source of nutriment for the ruminant. Various authors have used a variety of techniques to estimate the amounts of production and absorption of these acids. These fatty acids are found in blood in proportions of about acetate, 65; propionate, 20; and butyrate, 10. Further details of fatty acid production and absorption by the ruminant may be found in the chapter on lipid metabolism. In general, carbon atoms of acetate, although they appear in carbo­ hydrate (blood glucose, milk lactose) through the mechanism of rearrangement in the TCA cycle (Fig. 3.9), cannot theoretically contribute to the net synthesis of carbohydrate. Thus, acetate is not a glucogenic com­ pound. The large amounts of acetate provided by ru­ men fermentation are utilized for energy purposes and for the synthesis of fat. A possible mechanism for the direct incorporation of acetate into a glucose precursor is the so-called "glyoxylate pathway," which occurs in plants but not in animals. Propionate, on the other hand, is a well-known pre­ cursor of carbohydrate. The pathway leading to a net synthesis of glucose from propionate is available via the reaction Propionate + CO2

�

succinate,

as shown in Fig. 3.9. According to the scheme, 2 moles of propionate are required for the synthesis of 1 mole of glucose so 1 g of propionate theoretically can pro­ vide 1.23 g of glucose. The amounts of propionate available from rumen fermentation can theoretically supply the glucose requirements not accounted for by protein sources. Butyrate, the third major fatty acid of rumen fermen­ tation, influences glucogenesis but does not contribute carbon directly to glucose. Butyrate stimulates glucose production by liver by increasing phosphorylases and gluconeogenesis. The AcCoA derived from {3oxidation of butyrate also activates pyruvate carboxyl­ ase, a key gluconeogenic enzyme that further promotes gluconeogenesis.

3. Utilization of Glucose The overall utilization of glucose by the ruminant has significant differences from that of other animals. Acetate oxidation rather than glucose plays the impor-

tant role in energy metabolism of the ruminant. Only about 10% of the respiratory CO2 arises from glucose oxidation, which is considerably less than the 25-60% for the rat, dog, and human. The glucose tolerance of the cow, however, is the same as in other animals. The plasma clearance TI/2 of 33 minutes in the cow is similar to that of dogs (Kaneko et aL, 1977) and humans. About 60% of the glucose oxidized in the mammary gland of the lactating cow occurs via the HMP pathway (Fig. 3.6), the same as in the rat mammary gland. HMP pathway activity in the ruminant mammary gland is also evidenced by the high activities of the HMP en­ zymes, G-6-PD and 6-P-GD in sheep and cows' mam­ mary glands. Thus, even though overall glucose utili­ zation is lower in ruminants, their pathways of glucose catabolism are the same as in other animals. As in other animals, the HMP pathway is the major provider of the reductive atmosphere for the synthetic processes of the mammary gland. Through the TCA cycle pathway, carbons from ace­ tate, from whatever source, appear in milk products (Fig. 3.17). Glucose carbon atoms may be given off as CO2, appear in the amino acids of milk protein via transamination of oxaloacetate and a-ketoglutarate, or appear in milk fat. The short-chain fatty acids of butter­ fat are synthesized from acetate in the mammary gland, whereas the long-chain acids of butterfat are derived from blood lipids. The synthetic pathway for fatty acids in the gland is the same as that in other animal tissues (see Section IX). The major portion of glucose uptake by the mam­ mary gland provides for the biosynthesis of milk. The glucose and galactose moieties of lactose are derived from blood glucose. The rate of lactose synthesis is

j � t

c i r c ulation

glucose

®

@

>

r

pyru

� � @§J'> � C02 cet

te

oxaloacetate

�

8

ketone bodies fat t y acid

fC02

citrate

/

a-ketoglutarate ,

lactose

,

� jI protein

fat

M ILK

FIGURE 3.17 Summary of some metabolic pathways in the mammary gland.

Carbohydrate Metabolism and Its Diseases also constant over a wide range of blood glucose con­ centrations, 1.1-4.4 mmol / liter (20-80 mg / dl), which indicates that lactose synthesis is maximal even under hypoglycemic conditions. The mammary gland, there­ fore, is a glucose utilizing tissue, principally for biosyn­ thesis and considerably less is oxidized. The principal metabolic pathways involved are summarized in Fig. 3.17. Ruminant nervous tissue, that is, brain, is also simi­ lar to that of other animals in being an obligatory glucose-utilizing tissue. The HK activity of sheep brain, however, is significantly lower than that of rat brain. This means that even though there is the same obliga­ tory glucose requirement between the ruminant and nonruminant, glucose utilization by ruminant nervous tissue is lower than in the nonruminant. Similarly, ru­ minant intestine and muscle use less glucose than non­ ruminants. With regard to organ distribution of gluconeogenic enzymes, the highest G-6-Pase activities are found in ruminant livers as compared to other organs of rumi­ nants and are generally equal to or slightly lower than the activities found in nonruminant livers. During early lactation, the period when a cow's glucose re­ quirement is highest, hepatic G-6-Pase does not in­ crease. Similarly, cow liver PEPCK, a key gluconeo­ genic enzyme, is already very high in comparison to that of rat liver. All of this is in keeping with the concept that liver is primarily a glucose producing tissue. This also means that the high-producing dairy cow that has been genetically selected for these qualities is already synthesizing glucose maximally under normal condi­ tions. It follows that any additional demands for glu­ cose from physical stress, disease, etc., are unlikely to be met by increased glucose production. This glucose shortage leads to ketosis, the primary form from excess milk production or secondary form from the stress of a disease. To summarize, the ruminant appears to be an ani­ mal well adapted to a carbohydrate economy based on the endogenous synthesis of glucose from noncar­ bohydrate sources (gluconeogenesis). The enzymatic mechanisms for gluconeogenesis are already operating at near maximal levels in the high-producing dairy cow. Glucose oxidation by individual tissues as well as by the intact animal is lower in ruminants than in nonruminants. Although overall partitioning of glu­ cose oxidation may be different in ruminants, the path­ ways by which this oxidation is accomplished are simi­ lar to those of other animals (Fig. 3.17). The endocrine relationships of ruminants are also qualitatively similar to those of nonruminants so that the normally low blood glucose concentrations of ruminants are a reflec-

79

tion of their degree of influence or balance rather than their type of action.

C. Biochemical Alterations in Body Fluids 1. Hypoglycemia Hypoglycemia is such a consistent finding in bovine ketosis and in ovine pregnancy toxemia that hypoglyce­ mia has been suggested as another name for bovine ketosis. . This hypoglycemia has played an important role in ketosis, as a rationale for therapy and as a basis for the concept of ketosis and pregnancy toxemia as manifestations of a carbohydrate deficiency that occurs under conditions of excessive and insurmountable de­ mands.

2. Ketonemia Ketonemia is a consistent feature of bovine ketosis and ovine pregnancy toxemia from which the name ketosis is derived. The ketone bodies are the same as those previously mentioned (Section V.3), AcAc, 3-0H­ B, and acetone. A fourth compound, isopropanol, is included for the ruminant and interconversions can occur between these ketone bodies. The fundamental mechanism of ketosis is covered in a separate chapter and a brief outline is presented here.

a. Site of Ketone Body Production Increased ketogenesis occurs under conditions that favor the accumulation of acetate. In the nonruminant, the liver is the sole source of ketone bodies and the ketone bodies appear in the body fluids when produc­ tion exceeds the capacity for utilization. They are low renal threshold substances as well as being highly vola­ tile so they readily appear in urine, milk, and in the breath. In the ruminant, the rumen epithelium and mammary gland are also sources of ketone bodies. The extent of their contribution to the ketone body pool, however, is uncertain but it could be considerable in the ketotic animal.

b. Hyperketonemia Hyperketonemia is influenced by a number of con­ ditions, all of which relate to the carbohydrate econ­ omy of the ruminant. Starvation is the most well known cause of ketosis. Mild to moderate ketonemia is often seen without detrimental effects during early lactation, late pregnancy, underfeeding, and with high fat diets. In all instances, the constant demands of the body for glucose are not adequately met. Hyperketonemia is a consistent finding in bovine ketosis and pregnancy toxemia, though the degree of ketonemia does not necessarily parallel the severity of

80

J. Jerry Kaneko

the clinical signs. Normally, total blood ketones in sheep or cows are less than 0.10 mol/ liter (0.6 mg / dl).

D. Ruminant Ketosis The finely balanced carbohydrate economy of the ruminant plays a central role in the development of ketosis in cows and sheep. In the cow, large amounts of glucose must be produced by the liver to meet the eavy demands for lactose, particularly in early lacta­ tion, when the demand is highest. Similarly, in the pregnant ewe, especially when carrying twins, there is a progressive and large obligatory demand for hexoses (fructose), which is maximal near term. The precise mechanisms whereby internal metabolic imbalances occur in trying to meet these demands and manifest as ketosis, however, are uncertain. The fundamental imbalance is an inability of hepatic gluconeogenesis to respond to meet the demands of lactation or pregnancy and a hypoglycemia occurs, followed by hyperketo­ nemia, and clinical ketosis develops.

�

References Arai, T., Washizu, T., Hamada, S., Sako, T., Takagi, S., Yashiki, K., and Motoyoshi, S. (1994). Vet. Res. Comm. 18, 417. Armbruster, D.A. (1987). CIin. Chem. 33, 2153. Ashmore, J., and Weber, G. (1968). In "Carbohydrate Metabolism and Its Disorders" (F. Dickens, P. J. Randle, and W. J. Whelan, eds.), p. 336. Academic Press, New York. Atkins, C. E., and Chin, H. P. (1983). Am. J. Vet. Res. 44, 596. Atkins, C. E., Hill, J. R, and Johnson, R K (1979). J. Am. Vet. Med. Assoc. 175, 362. Balasse, E. 0., and Havel, R. J. (1971). J. CIin. Invest. 50, SOL Banauch, D., Brummer, W., Ebeling, W., Metz, H., Rinfrey, H., Ley­ bold, K, and Rick, W. (1975). Z. Klin. Chem. 13, 101. Bell, R R, and Jones, E. R (1945). J. Compo Pathol. Therap. 55, 117. Bolli, G. B., Gottesman, I. 5., Campbell, P. J., Haymond, M. W., Cryer, P. E., and Gerich, J. E. (1984). N. Engl. J. Med. 311, 1214. Buchanan, K. D. (1975). "Diabetes: Its Physiological and Biochemical Bases," p. 63. MTP Press, Lancaster, UK. Bunn, H. F., Gabbay, K. H., and Gallop, P. M. (1978). Science 200, 21. Ceh, L., Hauge, J. G., Svenkerud, R, and Strande, A. (1976). Acta. Vet. Scand. 17, 210. Cherrington, A. D., and Exton, J. H. (1976). Metabolism 25, 1351. Clark, D. G., Topping, D. L., TIlman, R J., Trimble, R P., and Malthus, R S. (19SO). Metabolism 29, 415. Cornelius, L. M., and Mahaffey, E. A. (1981). J. Am. Vet. Med. Assoc. 180, 635. Cotton, R B., Cornelius, L. M., and Theran, P. (1971). ]. Am. Vet. Med. Assoc. 159, 863. Cutler, J. T. (1934). ]. Bioi. Chem. 106, 653. Dobbs, R, Sakurai, H., Sakai, H., Faloona, G., Valverde, 1., Baetens, D., Orci, L., and Unger, R (1975). Science 187, 544 . Duckworth, W. c., and Kitabchi, A. E. (1981). Endocrin. Rev. 2, 21 0. Eisenbarth, G. S. (1986). N. Engl. J. Med. 314, 1360. Engerman, R L., and Kramer, J. W. (1982).]. Am. Diabetes Assoc. 31,26. Felig, P., Wahren, J., Sherwin, R, and Hendler, R (1976). Diabetes 25, 1091. Foster, S. J. (1975). J. Sm. Anim. Pract. 16, 295.

Gabbay, K. H., Hasty, K., Breslow, J. L., Ellison, R c., Bunn, H. F., and Gallop, P. M. (1977). J. CIin. Endocrinol. Metab. 44, 859. Garber, A. J., Menzel, P. H., Boden, G., and Owen, O. E. (1974). J. CIin. Invest. 54, 981. Gershwin, L. J. (1975). J. Am. Vet. Med. Assoc. 167, 479. Giger, U., Harvey, J. W., Yamaguchi, R A., McNulty, P. K, Chiapella, A., and Beutler, E. (1985). Blood 65, 345. Gluecksohn-Waelsch, S. (1979). Cell 16, 225. Gross, S. R (1975). West. J. Med. 123, 194. Hill, F. W. G., Pearson, H., Kelly, D. F., and Weaver, B. M. Q. (1974). J. Small Anim. Pract. 15, 119. Hooghuis, H., Rodriguez, M., and Castano, M. (1995). Vet. CIin. Pathol. 23, 1 10. Howell, R R, and Williams, J. C. (1983). In "The Metabolic Basis of Inherited Disease" 0. B. Stanbury, J. B. Wymgaarden, D. S. Frederickson, Goldstein, J. L., and Brown, M. S., eds.) 5th ed., p. 141. McGraw-Hill, New York. Jensen, A.L., and Aaes, H. (1992). Vet. Res. Commun. 16, 317. Johnson, R K, and Atkins, C. E. (1977). In "Current Veterinary Therapy" (R W. Kirk, ed.), Vol. 6, p. 1010. Saunders, Philadel­ phia, Pennsylvania. Kaneko, J. J., and Howard, C. F., Jr. (1989). In "Clinical Chemistry of Laboratory Animals" (W. F. Loeb and F. W. Quinby, eds.), p. 73. Pergamon Press, New York. Kaneko, J. J., Mattheeuws, D., Rottiers, R P., and Vermeulen, A. (1977). J. Small Anim. Pract. 18, 85. Kaneko, J. J., Mattheeuws, D., Rottiers, R P., Van Der Stock, J., and Vermeulen, A. (1978a). Acta Endocrinol. 87, 1l3. Kaneko, J. J., Mattheeuws, D., Rottiers, R P., and Vermeulen, A. (1978b). Am. J. Vet. Res. 39, S07. Kaneko, J. J., Mattheeuws, D., Rottiers, R P., and Vermeulen, A. (1979). Cornell Vet. 69, 375. Kaneko, J. J., Kawamoto, M., Heusner, A. A., Feldman, E. c., and Koizumi, 1. (1992). Am. J. Vet. Res. 53, 1797. Kawamoto, M., Kaneko, J. J., Heusner, A. A., Feldman, E. c., and Koizurni, 1. (1992). Am. J. Vet. Res. 53, 851. Khoo, J. c., Steinberg, D., Thompson, D., and Mayer, S. E. (1973). J. Bioi. Chem. 248, 3823. Kirk, C. c., Feldman, E. c., and Nelson, R W. (1993). Am. J. Vet.

Res. 54, 463.

Kitabchi, A. E. (1977). Metabolism 26, 547. Kleiber, M. (1959). Proc. Interam. Symp. Peaceful Appl. Nucl. Energy, 2nd, Buenos Aires, p. 161. Kramer, J. W. (1977). Fed. Am. Soc. Exp. Bioi. 36, 279. Kramer, J. W. (1981). Am. J. Pathol. lOS, 194. Kramer, J. W., Nottingham, 5., Robinette, J., Leaz, G., Sylvester, S., and Dessouky, M. (19SO). Diabetes 29, 558. Kreisberg, R. A. (1978). Ann. Intern. Med. 88, 681. Kronfeld, D. 5., Simesen, M. G., and Dungworth, D. L. (1960). Res.

Kruth, S. A., Feldman, E. c., and Kennedy, P. C. (1982). J. Am. Vet. Med. Assoc. 181, 54. Ling, G. c., Lowenstine, L. J., Pulley, L. T., and Kaneko, J. J. (1977). J. Am. Vet. Med. Assoc. 170, 521. Lusk, G. (1928). "The Elements of the Science of Nutrition," 4th ed., p. 321. Saunders, Philadelphia, Pennsylvania. McCandless, E. L., Woodward, B. A., and Dye, J. A. (1948). Am. ]. Physiol. 154, 94. McGarry, J. D., and Foster, D. W. (1976). Am. J. Med. 61, 9. McGarry, J. D., Wright, P. H., and Foster, D. W. (1975). J. CIin. Invest. 55, 1202. McMillan, F. D., and Feldman, E. C. (1986). ]. Am. Vet. Med. Assoc.

Vet. Sci. 1, 242.

188, 1426. Markussen, J. (1971). Int. J. Protein Res. 3, 149.

Carbohydrate Metabolism and Its Diseases Mattheeuws, D., Rottiers, R., Deijcke, J., DeRick, A, and DeSchepper, ]. (1976). ,. Small Anim. Pract. 7, 313. Mattheeuws, D., Rottiers, R., Kaneko, J. J., and Vermeulen, A (1984). Am. J. Vet. Res. 45, 98. Morrill, C. C. (1952). Am. ,. Vet. Res. 13, 164. Murakami, H, Takagi, A, Nonaka, S., Ishiura, S., Sugita, H., and Mizutani, M. (1980). Exp. Anim. 29, 475. Naithani, G., Steffans, G., Tager, H S., Buse, G., Rubenstein, A F., and Steiner, D. F. (1984). Hoppe-Seyler's Z. Physiol. Chern. 365, 57l. Nathan, D. M., Singer, D. E., Hurxthal, K, and Goodson, J. D. (1984). N. Engl. J. Med. 310, 34l. Nelson, R. W., Himsel, C. A, and Feldman, E. C. (1990). Am J. Vet Res. 51, 1357. Nelson, R. W., Feldman, E. c., and Ford, S. L. (1993). J. Am. Vet. Med. Assoc. 203, 82I. O'Brien, T. D., Hayden, D. W., Johnson, K W., and Stevens, J. B. (1985). Vet. Pathol. 22, 250. O'Sullivan, B. M., Healy, P. J., Fraser, 1. R., Nieper, R. E., Whittle, R. J., and Sewell, C. A (1981). Aust. Vet. J. 57, 227. Pfeifer, M. A, Halter, J. B., and Porte, D. (1981). Am. J. Med. 70, 579. Phillips, R. W., Knox, K L., Pierson, R. E., and Tasker, J. B. (1971). Cornell Vet. 61, 114. Priester, W. A (1974). ,. Nat. Cancer Inst. 53, 227. Rafaguzzaman, M., Svenkerud, R., Strande, A, and Hauge, J. G. (1976). Acta Vet. Scand. 17, 196. Raptis, S., and Dimitriadis, G. (1985). Clin. Physiol. Biochem. 3, 29. Renold, A E., and Cahill, G. F., Jr. (1966). In "The Metabolic Basis of Inherited Disease" 0. B. Stanbury, J. B. Wyngaarden, and D. S. Fredrickson, eds.), p. 69. McGraw-Hill, New York. Roderick, L. M., Harshfield, G. S., and Merchant, W. R. (1933). Cornell Vet. 23, 348. Rogers, W. A, Donovan, E. F., and Kociba, G. J. (1975). J. Am. Vet. Med. Assoc. 166, 1092. Ross, S. A., Brown, J. c., and Dupre, J. (1974). Diabetologia 10, 384. Schade, D. S., and Eaton, R. P. (1985). N. Engl. J. Med. 312, 1120.

81

Shannon, J. A., Farber, S., and Troast, L. (1941). Am. J. Physiol. 133, 752. Sherwin, R. S., Fisher, M., Hendler, R., and Felig, P. (1976). N. Engl. J. Med. 294, 455. Spieth, K (1973). Prakt. Thierzt. 54, 292 . Steiner, D. E. (1977). Diabetes 26, 332. Steinke, J., and Taylor, K W. (1974). Diabetes, 23, 631. Stewart, J., and Holman, H. H. (1940). Vet. Rec. 52, 157. Svenkerud, R., and Hauge, J. G. (1978). Camp. Pathol. Bull. 10, 2. Swiatek, K R., Kipnis, D. M., Mason, G., Chao, K, and Comblath, M. (1968). Am. J. Physiol. 214, 400. Taniyama, H., Ushiki, T., Tajima, M., Kurosawa, T., Kitamura, N., Takahashi, K, Matsukawa, K, and Itokura, C. (1995). Vet. Pathol. 32, 221. Taylor, S. 1. (1995). In "The Metabolic and Molecular Bases of Inher­ ited Disease," (c. R. Scriver, A L. Beaudette, W. S. Sly, and D. Valle, eds.), 7th ed., Vol. 2, p. 843. McGraw-Hill, New York. Teunissen, G. H B., Rijnberk, A., Schotman, W., and Hackeng, W. H L. (1970). Kleint. Prax. 15, 29. Thorens, B., Sakhar, H. K, Kaback, H R., and Lodish, H F. (1988). Cell 55, 28I. Unger, R. H. (1981). Diabetologia 20, 1 . Unger, R . H., and Orci, L . (1975). Lancet 1, 14. Unger, R. H, and Orci, L. (1976). Physiol. Rev. 56, 778. Vora, S., Giger, U., Turchen, S., and Harvey, J. W. (1985). Proc. Nat. Acad. Sci. USA. 82, 8109. Walvoort, H. C. (1983). J. Inher. Metab. Dis. 6, 3. Walvoort, H. c., Slee, R. G., and Koster, J. F. (1982). Biochim. Biophys. Acta 715, 63. Wilder, R. M., Allan, F. N., Power, M. H., and Robertson, H. E. (1927). J. Am. Med. Assoc. 89, 348. Williams, M., Gregory, R., Schall, W., Gossain, V., Bull, R., and Padgett, G. (1981). Fed. Proc. 40, 740. Wood, P. A, and Smith, J. E. (1980). ,. Am. Vet. Med. Assoc. 176, 1267. Yoon, J.-W., Marshall, A, Onodera, T., and Notkins, A L. (1979). N. Engl. J. Med. 300, 1173.

C H A P T E R

4 Lipids and Ketones MICHAEL L. BRUSS

1. INTRODUCTION 83 II. LONG-CHAIN FATTY ACIDS 84 A. Structure, Properties, and Assay of Long-Chain Fatty Acids 84 B. Synthesis of Long-Chain Fatty Acids 84 C. Catabolism of Long-Chain Fatty Acids 86 III. TRIACYLGLYCEROL 87 A. Structure, Properties, and Assay of Triacylglycerol 87 B. Synthesis of Triacylglycerol 88 C. Catabolism of Triacylglycerol 90 IV. PHOSPHOLIPIDS 90 A. Structure and Properties of Phospholipids 90 B. Synthesis of Phospholipids 90 C. Catabolism of Phospholipids 90 V. CHOLESTEROL 91 A. Structure, Properties, and Assay of Cholesterol 91 B. Metabolism of Cholesterol 92 VI. LIPOPROTEINS 93 A. Structure, Properties, and Assay of Lipoproteins 93 B. Apolipoproteins 94 C. Digestion of Fat and Formation of Chylomicrons 95 D. Very-Low-Density Lipoproteins: Synthesis, Export, and Metabolism 96 E. Metabolism of High-Density Lipoproteins 97 VII. HYPERLIPIDEMIA 97 A. Introduction 97 B. Canine Fasting Hyperlipidemias 98 C. Feline Fasting HyperIipideroias 99 D. Equine Fasting Hyperlipidemia 100 VIII. KETOGENESIS AND KETOSIS 100 A. Introduction 100 B. Chemistry of Ketones 100

CUNICAL BIOCHEMISTRY OF DOMESTIC ANIMALS. FIFTH EDITION

C. D. E. F. G. H.

SyntheSiS of Ketones 102 Catabolism of Ketones 104 Pathophysiology of Ketonemia 106 Fasting Ketosis 106 Diabetic Ketosis 107 Ketosis Associated with Pregnancy and Lactation 108 1. Postexercise Ketosis 111 References 111

I. INTRODUCTION This chapter covers the biochemistry and clinical chemistry of long-chain fatty acids, triacylglycerols, phospholipids, cholesterol, and ketones, a list that in­ cludes the majority of lipids found in vertebrates. The only remaining major classes are sphingolipids and waxes, which are not discussed. Although lipids have many functions, two of the most important are energy storage and membrane structure. Triacylglycerols are by far the most important lipid with regard to energy storage, and phospholipids and cholesterol are the most important lipoid membrane constituents. Lipids serve other functions, including being precursors for steroids and bile acids (cholesterol), thermal insulation (triacylglycerols), and electrical insulation (various lip­ ids). Virtually all lipids are insoluble in water, which greatly complicates their handling in the body. Because of their insolubility, lipids must rely on proteins for transport for any significant distance in the body and various proteins have evolved to provide this function. The insolubility of lipids is an asset as well as a liability.

83

Copyright : chai� . Most fetal Hb is replaced by adult Hb types In rumI­ nants during the first month(s) after birth (Aufderheide et al., 1980; Blunt, 1972; Huisman et al., 1969; Kitchen and Brett, 1974; Lee et al., 1971). This switch from pro­ duction of fetal Hb to the production of adult Hb ap­ pears to result from an inherent programming of hema­ topoietic stem cells (Wood et al., 1985). In cats, dogs, horses, and pigs, embryonal Hbs are replaced by adult Hb types during the fetal period. Hb types present in fetuses are identical to those found in adults (Bunn and Kitchen, 1973; Kitchen and Brett, 1974). Considerable heterogeneity of Hb types occurs in adult animals. With the possible exception of pigs, two or more types are reported to occur in domestic animal species (Braend, 1988; Kitchen, 1969). Most polymor­ phism of animal Hbs is determined genetically and usually caused by multiple amino acid interchanges (Kitchen, 1974). Nongenetic alterations in Hb structure can also contribute to apparent Hb heterogeneity. Ex­ amples include the N-acetylation of {3 chains in cat HbB (Taketa et al., 1972) and glycosylation of Hb, a function of intracellular glucose concentration and RBC lifespan (Higgins et al., 1982; Rendell et al., 1985). Increased glycosylation of Hb has been reported in diabetic dogs (Mahaffey and Cornelius, 1982; Wood and Smith, 1980). A unique occurrence in sheep and goats is the syn­ thesis of HbC in response to anemia (Huisman and Kitchen, 1968). This Hb switching from synthesis of HbA in sheep and HbA and HbB in goats to HbC is mediated by EPO (Barker et al., 1980). Carbon dioxide decreases oxygen affinity for HbC more than it does for normal adult Hbs (Huisman and Kitchen, 1968; Winslow et al., 1989).

D. Reticulocytes 1. Fonnation Although reticulocytes may be formed by denuclea­ tion as metarubricytes pass through endothelial cells into vascular sinuses of bone marrow, most are formed within the extravascular space of the bone marrow by a process of nuclear extrusion that requires functional microtubules (Chasis et al., 1989) and is likened to mito­ sis (Bessis, 1973; Simpson and Kling, 1967). Extruded nuclei are bound and phagocytosed by a novel receptor

164

John W. Harvey

on the surface of bone marrow macrophages (Qui et al., 1995). Early reticulocytes have polylobulated surfaces. Their cytoplasm contains ribosomes, polyribosomes, and mitochondria necessary for the completion of Hb synthesis. Reticulocytes derive their name from a net­ work or reticulum that appears when stained with basic dyes such as methylene blue and brilliant cresyl green. That network is not preexisting but is an artifact formed by the precipitation of ribosomal ribonucleic acids and proteins. As reticulocytes mature, the amount of ribosomal material decreases until only a few basophilic specks can be visualized with reticulo­ cyte staining procedures. These mature reticulocytes have been referred to as type IV (Houwen, 1992) or punctate reticulocytes (Alsaker et al., 1977; Perkins and Grindem, 1995). To reduce the chance that a staining artifact would result in misclassifying a mature RBC as a punctate reticulocyte using a reticulocyte stain, the cell in question should have two or more discreet blue granules that are visible without requiring fine focus adjustment of the cell being evaluated to be clas­ sified as a punctate reticulocyte.

2. Metabolism Reticulocyte metabolism and maturation have been reviewed by Rapoport (1986) and are summ arized

here. Immature reticulocytes continue to synthesize protein (primarily globin chains) with residual mRNA, tRNA, and rRNA formed prior to denucleation. Syn­ thesis of fatty acids is minimal, but phospholipids are synthesized from preformed fatty acids. Substrates for protein and lipid synthesis and for energy metabolism are provided from endogenous sources (breakdown of mitochondria and ribosomes) as well as from plasma. The reticulocyte can synthesize adenine and guanine nucleotides de novo. Most ATP is generated in reticulocytes by oxidative phosphorylation in mitochondria. Glucose is the major substrate, but amino acids and fatty acids can also be

cles (exosomes) extracellularly (Johnstone, 1992). This is a highly selective process where some proteins (e.g., transferrin receptor and fibronectin receptor) are lost and cytoskeletal proteins (e.g., spectrin) and firmly bound transmembrane proteins (e.g., the anion trans­ porter and glycophorin A) are retained and concen­ trated (Johnstone, 1992; Rapoport, 1986). Some membrane proteins such as the nucleoside transporter, glucose transporter, Na,K-ATPase, insulin receptor and adrenergic receptors decrease to variable degrees depending on the species involved (Chasis et al., 1989; Johnstone, 1992). Examples where reticulo­ cytes from a species exhibit a complete or nearly com­ plete loss of a protein, which is retained in mature RBCs from other species, include the adenosine trans­ porter in sheep (Jarvis and Young, 1982), the glucose transporter in pigs (Zeidler and Kim, 1982), and Na,K­ ATPase in dogs (Maede and Inaba, 1985). Although loss of membrane components accounts for much of the change in membrane protein composi­ tion during reticulocyte maturation, certain proteins such as protein 4.1 and glycophorin C increase because they are still being synthesized in reticulocytes (Chasis

et al., 1989). Mitochondria undergo degenerative changes owing to lipoxygenase attack and subsequent ATP-dependent proteolysis (Rapoport, 1986). Degenerating mitochon­ dria are either digested or extruded following entrap­ ment in structures resembling autophagic vacuoles (Simpson and Kling, 1968). The polysomes separate into monosomes and decrease in number and disap­ pear as reticulocytes mature into RBCs. The degrada­ tion of ribosomes appears to be energy dependent; it presumably involves proteases and RNAases (Rapo­ port, 1986).

4. Species Differences in Marrow Release

utilized for energy (Rapoport, 1986).

Reticulocyte maturation begins in the bone marrow and is completed in the peripheral blood and spleen in dogs, cats, and pigs. As reticulocytes mature, they

3. Maturation into Erythrocytes

lose the surface receptors needed to adhere to fibro­ nectin and thrombospondin components of the extra­

Reticulocyte maturation into mature RBCs is a grad­ ual process that requires a variable number of days

cellular matrix, thereby facilitating their release from the bone marrow (Long and Dixit, 1990; Patel et al., 1985; Vuillet-Gaugler et al., 1990). Residual adhesion

depending on the species involved. Consequently the morphologic and physiologic properties of reticulo­ cytes vary with the stage of maturation. The cell surface undergoes extensive remodeling with loss of mem­ brane material and ultimately the formation of the bi­ concave shape of mature RBCs (Bessis, 1973). The loss of membrane protein and lipid components appears to require ATP (Weigensberg and Blostein, 1983) and involve formation of intracellular multivesicular bod­ ies that fuse with the plasma membrane releasing vesi-

molecule receptors on newly released reticulocytes may explain their tendency to concentrate in the reticu­ lar meshwork of the spleen (Patel

et al., 1985).

Reticulocytes become progressively more deform­ able as they mature, a characteristic that also facilitates their release from the marrow (Waugh, 1991). To exit the extravascular space of the marrow, reticulocytes press against the abluminal surfaces of endothelial cells that make up the sinus wall. Cytoplasm thins and small

The Erythrocyte pores (0.5-2 ILm) develop in endothelial cells that allow reticulocytes to be pushed through by a small pressure gradient across the sinus wall (Lichtman and Santillo, 1986; Waugh, 1991). Pores apparently close after cell passage. Relatively immature aggregate-type reticulocytes are released from canine bone marrow; consequently, most of these cells appear polychromatophilic when viewed following routine blood film staining proce­ dures (Laber et ai., 1974). Absolute reticulocyte counts oscillate with a periodicity of approximately 14 days in some dogs, suggesting that canine erythropoiesis may have a homeostatically controlled physiologic rhythm (Morley and Stohlman, 1969). Reticulocytes are normally not released from feline bone marrow until they mature to punctate-type reticulocytes; conse­ quently, few or no aggregate reticulocytes « 0.4%), but up to 10% punctate reticulocytes, are found in blood from normal adult cats (Cramer and Lewis, 1972). The high percentage of punctate reticulocytes results from a long maturation time (Fan et ai., 1978) with delayed degradation of organelles. Reticulocytes are generally absent in peripheral blood of healthy adult cattle and goats, but a small number of punctate types (0.5%) may occur in adult sheep Gain, 1986). Equine reticulo­ cytes are absent from blood normally and are not re­ leased in response to anemia. 5. "Stress" Reticulocytes

Except for horses, increased numbers of reticulo­ cytes are released in response to anemia, with better responses to hemolytic anemias than to hemorrhage. When the degree of anemia is severe, basophilic macro­ reticulocytes or so-called stress reticulocytes may be released into blood. It is proposed that a generation in the maturation sequence is skipped, and immature reticulocytes that are twice the normal size are released (Rapoport, 1986). Increased EPO results in a diminu­ tion in the adventitial cell and endothelial cell barrier separating marrow hematopoietic cells from the sinus, thereby potentiating the premature release of stress reticulocytes from the marrow (Chamberlain et al., 1975). Although a portion of these macroreticulocytes apparently is rapidly removed from the circulation (Noble et ai., 1990), it is clear from studies in cats that some can mature into macrocytic RBCs with relatively normal lifespans (Weiser and Kociba, 1982).

E. Abnormalities in Erythroid Development

1. Ineffective Erythropoiesis Ineffective erythropoiesis is used to describe the de­ struction of developing erythroid cells in marrow. Nor­ mally, minimal cells die within the marrow (Odart-

165

chenko et ai., 1971), but ineffective erythropoiesis is prominent in disorders of nucleic acid, heme, or globin synthesis. Examples include folate deficiency, iron deficiency, vitamin B6 deficiency, lead poisoning, and thalassemia in humans (Jandl, 1987). Ineffective eryth­ ropoiesis also occurs in association with myeloprolifer­ ative and myelodysplastic disorders (Durando et ai., 1994; Jain, 1986) and congenital dyserythropoiesis (Holland et ai., 1991; Steffen et ai., 1992).

2. Vitamin and Mineral Deficiencies Folate is required for normal DNA synthesis. Folate deficiency impairs the activity of the folate-requiring enzyme thymidylate synthase (Jandl, 1987). Not only is deoxythymidylate triphosphate (dTTP) synthesis de­ creased, but deoxyuridylate triphosphate (dUTP) accu­ mulates secondarily in the cell such that some becomes incorporated into DNA in place of dTTP. Cycles of excision and attempts to repair these copy errors, with limited thymidine available, result in chromosomal breaks and malformations and slOwing of the S phase in the cell cycle. Consequently, erythroid precursors are often large with deranged appearing nuclear chro­ matin; such cells are classified as megaloblastic cells. Folate deficiency in people causes macrocytic anemia because fewer divisions occur as a result of retarded nucleic acid syntheSiS in the presence of normal protein synthesis (Jandl, 1987). Possible causes of folate defi­ ciency include dietary deficiency, impaired absorption, and drugs that interfere with folate metabolism. Macrocytic anemias resulting from folate deficiency are rarely reported in animals. A possible case was reported in a dog on anticonvulsant therapy (Lewis and Rebar; 1979), but serum folate was not measured. Megaloblastic precursors are present in bone marrow of cats with experimental dietary folate deficiency, but packed cell volumes (PCVs) and mean cell volumes (MCVs) remained normal (Thenen and Rasmussen, 1978). Macrocytic anemia occurs in folate-deficient pigs (Bush et ai., 1956) but not lambs (Stokstad, 1968). Vitamin B12 (cobalamin) deficiency in people causes hematologic abnormalities similar to folate deficiency because vitamin B12 is necessary for normal folate me­ tabolism in humans (Chanarin et ai., 1985). In contrast, vitamin B12 deficiency does not cause macrocytic ane­ mia in any animal species (Chanarin et ai., 1985). Ane­ mia has been reported in some experimental animal studies, but RBCs were of normal size (Stokstad, 1968; Underwood, 1977), although slight increases in MCV have been reported in B12-deficient goats fed diets defi­ cient in cobalt (Mgongo et ai., 1981 ). Cobalamin defi­ ciency has been reported secondary to an inherited malabsorption of cobalamin in giant schnauzer dogs (Fyfe et ai., 1989, 1991). Affected animals have normo-

166

John W. Harvey

cytic, nonregenerative anemia with increased aniso­ cytosis and poikilocytosis, neutropenia with hyperseg­ mented neutrophils and giant platelets. Megaloblastic changes in the bone marrow were particularly evident in the myeloid cell line. The malabsorption of cobal­ amin in these dogs apparently results from the absence of an intrinsic factor-cobalamin receptor in the apical brush border of the ileum (Fyfe et al., 1991). No blood or bone marrow abnormalities were recognized in kittens fed a B12-deficient diet for several months (Morris, 1977), but a normocytic nonregenerative anemia was present in a cobalamin-deficient cat that probably had an inherited disorder of cobalamin absorption (Vaden et al., 1992). A number of disorders exhibit macrocytic anemias with megaloblastic abnormalities in the marrow that mimic findings in human folate or B12 deficiency, but have had normal serum levels of these vitamins when measured. Examples include cats infected with the fe­ line leukemia virus (Dunn et al., 1984; Hirsch and Dunn, 1983; Weiser and Kociba, 1983a), cattle with congenital dyserythropoiesis (Steffen et al., 1992), and myelodysplastic syndrome in a horse (Durando et al., 1994). In addition, some miniature and toy poodles exhibit macrocytosis without anemia and variable megaloblastic abnormalities in the bone marrow with normal serum folate and B12 values (Canfield and Wat­ son, 1989; Schalm, 1976). Abnormalities in heme or globin synthesis can result in the formation of microcytic hypochromic RBCs. Cel­ lular division is normal, but Hb synthesis is delayed; consequently, one or more extra divisions occur in RBC development, resulting in smaller cells than normal. Pyridoxine, vitamin B6t is required for the first step in heme synthesis. Although natural cases of pyridox­ ine deficiency have not been documented in domestic animals, microcytic anemias with high serum iron values have been produced experimentally in dogs (McKibbin et al., 1942), cats (Bai et al., 1989; Carvalho da Silva et al., 1959), and pigs (Deiss et al., 1966) with dietary pyridoxine deficiency. With the exception of young growing animals, iron deficiency in domestic animals usually results from blood loss. Milk contains little iron; consequently, nurs­ ing animals can easily deplete body iron store as they grow (Furugouri, 1972; Harvey et al., 1987; Holter et al., 1991; Siimes et al., 1980). Microcytic RBCs are pro­ duced in response to iron deficiency (Holman and Drew, 1964, 1966; Holter et al., 1991; Reece et al., 1984) but a low MCV may not develop postnatally in species where the MCV is above adult values at birth (Weiser and Kociba, 1983b). The potential for development of severe iron deficiency in young animals appears to be less in species that begin to eat food at an early age.

Chronic iron deficiency anemia with microcytic RBCs is common in adult dogs in areas where hook­ worm and flea infestations are severe (Harvey et al., 1982; Weiser and O'Grady, 1983). Severe iron defi­ ciency appears to be rare in adult cats (French et al., 1987; Fulton et al., 1988) and horses (Smith et al., 1986), but it occurs frequently in ruminants that are heavily parasitized with blood-sucking parasites such as

Haemonchus contort us. Prolonged copper deficiency generally results in anemia in mammals (Brewer, 1987; Lahey et al., 1952), although it was not a feature of experimental copper deficiency in the cat (Doong et al., 1983). The anemia is generally microcytic hypochrOmiC; however, normo­ cytic anemia has been reported in experimental studies in dogs, and normocytic or macrocytic anemias have been reported in cattle and adult sheep (Brewer, 1987). Copper deficiency results in impaired iron metabolism in at least two ways. In experimental studies in pigs, serum iron concentration is low in early copper defi­ ciency when iron stores are normal (Lahey et al., 1952). Functional iron deficiency results from inadequate iron absorption and mobilization of iron stores caused by a decreased concentration of circulating ceruloplasmin (Lee et al., 1968). Ceruloplasmin is the major copper­ containing protein in plasma, and its ferroxidase activ­ ity appears to be important in the oxidation of ferrous iron (released from ferritin stores) to the ferric state for binding to transferrin (Frieden, 1983). If experimental copper deficiency is prolonged, hy­ perferremia occurs and bone marrow sideroblasts in­ crease (Lee et al., 1968). Reticulocyte mitochondria from copper-deficient pigs are unable to synthesize heme at the normal rate using ferric iron (Williams et al., 1976). A deficiency in copper-containing cytochrome oxidase within mitochondria may slow the reduction of ferric to ferrous iron within mitochondria. That would, in tum, limit heme synthesis, which requires iron in the ferrous state (Porra and Jones, 1963).

3. Deficiencies in Globin Synthesis Hereditary deficiencies in synthesis of the globin a chain (a-thalassemia) and {3 chain ({3-thalassemia) cause microcytic hypochromic anemias in humans with variable degrees of poikilocytosis Gandl, 1987). Both a- and {3-thalassemia occur in mice, but hereditary hemoglobinopathies have not been reported in domes· tic animals (Kaneko, 1987).

4. Aplastic Anemia Aplastic anemia is generally used to describe ane­ mias where granulocytic, megakaryocytic, and eryth­ rocytic cell lines are markedly reduced in the bone

The Erythrocyte

marrow. When only the erythroid cell line is reduced or absent, terms such as pure red cell aplasia or selective erythroid aplasia or hypoplasia are used. These anemias can result from insufficient numbers of stem cells, ab­ normalities in the hematopoietic microenvironment, or abnormal humoral or cellular control of hematopoiesis (Appelbaum and Fefer, 1981; Juneja and Gardner, 1985). The factors mentioned are interrelated and the exact defect in a given disorder is usually unknown. Drug-induced causes of aplastic anemia or general­ ized marrow hypoplasia in animals include estrogen toxicity in dogs (Crafts, 1948; Gaunt and Pierce, 1986; Sherding et ai., 1981) and ferrets (Bernard et ai., 1983; Kociba and Caputo, 1981), phenylbutazone toxicity in dogs (Schalm, 1979; Watson et ai., 1980; Weiss and Klausner, 1990), trimethoprim-sulfadiazine adminis­ tration in dogs (Fox et al., 1993; Weiss and Klausner, 1990), bracken fern poisoning in cattle and sheep (Par­ ker and McCrea, 1965; Sippel, 1952; Tustin et ai., 1968), trichloroethylene-extracted soybean meal in cattle (Strafuss and Sautter, 1967), and various cancer chemo­ therapeutic agents. Thiacetarsamide, meclofenamic acid, and quinidine (Watson, 1979; Weiss and Klausner, 1990) have also been incriminated as poten­ tial causes of aplastic anemia in dogs, as has griseoful­ vin in a cat (Rottman et al., 1991). Human parvovirus, hepatitis viruses, Epstein-Barr virus, dengue, cytomegalovirus, and human immuno­ deficiency virus 1 have, in low percentages of infec­ tions, resulted in aplastic anemia in people (Rosenfeld and Young, 1991). Parvovirus infections can cause se­ vere erythroid hypoplasia, as well as myeloid hypopla­ sia in canine pups (Robinson et ai., 1980), but animals usually do not become anemic because of the long lifespans of RBCs. Either affected pups die acutely or the bone marrow returns rapidly to normal before ane­ mia can develop. In contrast to its effects in pups, parvovirus is reported to have a minimal effect on erythroid progenitors in adult dogs (Brock et ai., 1989). Parvovirus inhibits colony formation of both myeloid and erythroid progenitor cells in feline bone marrow cultures (Kurtzman et ai., 1989), but only myeloid hypoplasia was reported during histologic examina­ tion of bone marrow from viremic cats (Larsen et ai.,

1976). Although some degree of marrow hypoplasia and/ or dysplasia often occurs in cats with feline leukemia virus (PeL V) infections, true aplastic anemia is not a well-documented sequela (Rojko and Olsen, 1984). Anemia is a common finding in ill cats infected with feline immunodeficiency virus (FN). It does not ap­ pear to result from a direct action(s) of the virus on

167

erythroid progenitor cells, but generally occurs second­ ary to inflammation, dysplasia, or neoplasia (linen­ berger et ai., 1991; Shelton et ai., 1990). Dogs that enter the chronic stage of ehrlichiosis generally have some degree of marrow hypoplasia and in severe cases can develop aplastic anemias (Buhles, et ai., 1975). Idiopathic aplastic anemias have also been reported in dogs (Eldor et ai., 1978; Weiss and Christopher, 1985) and horses (Berggren, 1981; Lavoie et ai., 1987). CFU­ Es were not detected in bone marrOW culture from a dog with an idiopathic aplastic anemia (Weiss and Christopher, 1985). One case of erythroid and myeloid aplasia, with normal megakaryocyte numbers, has been reported in a horse, the etiology of which was unknown (Ward et ai., 1980).

5. Selective Erythroid Aplasia Selective erythroid aplasia (pure red cell aplasia) occurs as either a congenital or acquired disorder in people (Dessypris, 1991). Acquired erythroid aplasia is often associated with abnormalities of the immune system. ErythrOid aplasia may also occur secondary to disorders including infections, malignancy, and drug or chemical toxicities. Acquired erythroid hypoplasia or aplasia occurs in dogs (Weiss, 1986; Weiss et ai., 1982). Some cases have immune-mediated etiologies based on positive re­ sponses to immunosuppressive therapy and the pres­ ence of antibodies that inhibit CFU-E development in marrow cultures (Weiss, 1986). Erythroid hypoplasia or dysplasia is reported to be a rare sequela to vaccina­ tion against parvovirus in dogs (Dodds, 1983). High doses of chloramphenicol cause reversible erythroid hypoplasia in some dogs (Watson, 1977) and erythroid aplasia in cats (Watson and Middleton, 1978). A dog has been reported to have congenital erythroid aplasia based on histopathologic examination of bone marrow at necropsy, but the M:E ratio was normal when aspi­ rate smears were examined several days prior to eutha­ nasia (Hotston Moore et ai., 1993). Selective erythroid aplasia occurs in cats infected with subgroup C FeLV, but not in cats infected only with subgroups A or B (Onions et ai., 1982; Riedel et ai., 1986). Although both myeloid and erythroid pro­ genitor cells are infected with virus, only erythroid progenitor cell numbers are decreased in cat marrow. Recent studies suggest that CFU-E numbers are mark­ edly decreased because the envelope glycoprotein gp70 of subgroup C FeLV is uniquely cytopathic for BFU-E or impairs differentiation of BFU-E into CFU­ E (Abkowitz, 1991; Dean et ai., 1992; Dornsife et ai.,

1989).

John W. Harvey

168 IV. THE MATURE ERYTHROCYTE

Values for RBC glucose utilization, ion concentra­ tions, and survival times in normal animals are given in Table 7.1 . Enzyme activities are given in Tables 7.2 and 7.3, and chemical constituents in RBCs are given in Tables 7.4 and 7.5. These are not, however, compre­ hensive lists. Other values are provided by Friedemann and Rapoport (1974) and in various chapters of a refer­ ence book edited by Agar and Board (1983a). Anemia induced by phlebotomy or by hemolytic drugs pro­ duces changes in many of the given values owing to the influx of young RBCs into the circulation in response to the anemia (Agar and Board, 1983a). Methods for enzyme assays vary considerably; consequently, each laboratory will need to establish its own normal ranges if enzyme studies are to be done.

A. Membrane Structure The RBC membrane is composed of a hydrophobic lipid bilayer with a protein skeletal meshwork attached to its inner surface by binding to integral (transmem­ brane) proteins (Fig. 7.2). Membrane proteins from RBCs have been numbered by their migration loca­ tion (Smith, 1987) on sodium dodecyl sulfate­ polyacrylamide gel electrophoresis (SOS-PAGE); some have also been given one or more names. Electro­ phoretic patterns of membrane proteins on SOS-PAGE are species variable (Gillis and Anastassiadis, 1985;

TABLE 7.1

Species

Smith et ai., 1983a; Kobylka et ai., 1972; Whitfield et ai., 1983). 1. Lipids

The lipid bilayer and associated transmembrane proteins chemically isolate and regulate the cell interior. The bilayer consists of phospholipids arranged with hydrophobic hydrocarbon chains of fatty acids to the center of the bilayer and the polar ends of the molecules in contact with both intracellular and extracellular aqueous environments. Molecules of unesterified cho­ lesterol are intercalated between fatty acid chains in molar concentrations approximately equal to the sum of the molar concentrations of phospholipids. Phos­ pholipids are asymmetrically arranged, with anionic amino-containing phospholipids (phosphatidylserine and most of the phosphatidylethanolamine) located in the inner layer of the bilaminar membrane. These phospholipids are shuttled across the membrane by an ATP-dependent aminophospholipid-specific trans­ porter (Zwaal et ai., 1993). Most of the cationic choline­ containing phospholipids, phosphatidylcholine (leci­ thin) and sphingomyelin, are located in the outer layer (Kuypers et al., 1993). These choline-containing phos­ pholipids are readily exchangeable with plasma phos­ pholipids, whereas the aminophospholipids are not (Reed, 1968). Species vary in RBC membrane phospho­ lipid compositions (Engen and Clark, 1990; Garnier et al., 1984; Nelson, 1967; Wessels and Veerkamp, 1973).

Erythrocyte Glucose Utilization, Ion Concentrations, and Survival Times of Various Mammals'

RBC glucose utilization (I'mol/hr/ml)

Human Dog Cat Horse Cattle Sheep

1.48 1.33 0.94 0.64 0.56 0.69

Goat Pig Rabbit Guinea pig Mouse Hamster Rat

1 .94 (4) 0.09 (5) 2.26 :t 0.30 (2) 1 .44 (6) 2.85 :t 0.20 (2) 2.38

:t :t :t :t

:t

:t

:t

0.11 0.12 0.09 0.10 0.05 0.19

(1) (1) (1) (1) (2) (3)

0.20 (2)

RBC Na+ (mmo1/1iter)

RBC K+ (mmo1/1iter)

6.2 :t 0.8 (7)

102 :t 3.9 (7) 5.7 :t 1.0 (8) 5.9 :t 1.9 (8) 120 :t 1 1 .1 (9) 22.0 :t 4.5 (8) HK+, 98.7 (10) LK+, 39.4 (10) 76.1 (4) 105.9 :t 12.7 (8) 110.1 :t 6.0 (8) 107.2 :t 10.1 (8)

92.8 :t 11.1 (8) 105.8 :t 14.4 (8) 10.4 :t 1.8 (9) 79.1 :t 14.6 (8) HK+, 17.1 (10) LK+, 73.7 (10) 13.4 (4) 15.6 :t 1.8 (8) 16.8 :t 6.3 (8) 24.4 :t 5.4 (8) 17.2 (12) 33.5 :t 3.5 (8)

.. 92.0 (12) 104.7 :t 15.4 (8)

RBC survival (days) 120 100 72 143 130 135

(11) (11) (11) (11) (11) (11)

115 67 57 80 43 50 56

(11) (11) (11) (11) (11) (11) (11)

• Mean values have been recalculated at times to pennit direct comparisons between species. Standard deviation values are given where indicated. Figures in parentheses are the references cited below. Abbreviations used are given in the footnote at the beginning of the chapter. References: (1) Harvey and Kaneko (1976b), (2) Magnani et a/. (1980), (3) Leng and Annison (1962), (4) Harkness et al. (1970), (5) Kim and McManus (1971 ), (6) Laris (1958), (7) Beutler (1995), (8) Coldman and Good (1 967), (9) Contreras et a/. (1986), (10) Tucker and Ellory (1971), (1 1) Vacha (1983), (12) Miseta et al. (1983).

169

The Erythrocyte TABLE 7.2 Enzyme HK

Human

Dog

Cat

Horse

1.78 :t 0.38 (1)

2.05 :t 0.86 (2)

5.04 :t 0.84 (2)

3.75 :t 0.42 (2)

2.2 :t 0.7 (2)

8.4 :t 1.8 (2)

GPI

24.1

PFK

11.0

Aldol

Erythrocyte Enzymes of Various Animal Species'

:t

1.0 (3)

:t

2.3 (1)

GAPD MPGM

54.4 :t 2.8 (4)

59.0 (6)

37.7 :t 5.6 (1)

4.04 :t 1.19 (4)

6.35 (6)

15.0

:t

2.0 (1)

3.02 :t 0.67 (1)

AST

436

:t

0.32 (4) 70 (40)

89.6 :t 9.4 (4)

200 :t 26 (1)

LDH

:t

1.3 (2)

226 :t 42 (1)

5.39 :t 0.83 (1)

Enol PK

:t

2.36

320 :t 36 (1)

PGK

9.7

49.0 :t 8.3 (3)

3.19 :t 0.86 (1)

2111 :t 397 (1)

TPI

16.3 :t 1.8 (3)

0.84 :t 0.16 (4) 8.4 :t 2.0 (2)

52.2 :t 5.0 (4)

3.14 :t 1.10 (2)

27.3

:t

2.71 :t 0.98 (9) 57.2 :t 14.2 (9)

69.1 :t 19.8 (9)

29.2 (6)

5.7 (6)

22.9 :t 5.5 (2)

15.1 :t 2.2 (8)

13.6 :t 3.9 (9) 1.3 :t 0.7 (2)

32.3 :t 3.6 (9) 1.53

:t

DPGM

2.00 (6)

1.02 (6)

0.08 (6)

0.6 (6)

DPGP

0.021 (6)

0.010 (6)

0.005 (6)

0.006 (6)

G6PD

8.34 :t 1.59 (1)

1 1 .33 :t 1.95 (2)

8.78 :t 0.78 (1)

6PGD

3.37

GR

:t

0.69 (3)

30.8 :t 4.6 (1)

GPx SOD Catalase (X 10')

153

:t

0.51 :t 0.12 (3)

Na+, K+-ATPase

8

:t

0.15 (3)

82.3 :t 24.0 (2) 9 :t 4 (2)

24 (1)

19.2 :t 3.8 (1)

NADPH-Diaph

:t

1.30 (2)

2118 (5)

2352 (5)

NADH-MR

1 .38

:t

3.2 :t 0.9 (2)

6.7 :t 1.8 (1)

GST

6.73

2 (10)

22.2 :t 2.8 (2)

0.33 :t 0.06 (3)

Nil (10)

5.9 (3)

15.16 :t 3.15 (2) 7.00 :t 1.19 (2)

3.69

:t

1.02 (3)

35.6

:t

8.5 (2)

168.7 :t 24.0 (2) 2885 (5)

161 :t 43 (2)

20.0 :t 4.6 (2)

0.39 :t 0.06 (3)

Nil (10)

0.27

(7)

19.19 :t 3.00 (2) 2.69

:t

0.89

:t

6.4

:t

0.66 (2) 0.06 (3)

42.6 :t 17.9 (2) 1.7 (2)

105 :t 21 (2)

25.9 :t 5.5 (2)

0.70 :t 0.06 (3) 5 :t 2 (10)

• All enzyme units are IU/g Hb, except SOD given in units/ g Hb and ATPase given in �moles phosphorus liberated / g Hb/hr. Mean values have been recalculated at times to permit direct comparisons between species. Standard deviation values are given where indicated. Figures in parentheses are the references cited below. Abbreviations used are given in the footnote at the beginning of the chapter. Temperatures above 25°C are included in reference citations. References: (1) Beutler (1984) (at 37"C), (2) Harvey (1994) (at 37"C), (3) Harvey and Kaneko (1975a), (4) Maede and Inaba (1987) (at 37"C), (5) Kurata et al. (1993), (6) Harkness et al. (1969), (7) Franken and Schotman (1977), (8) Schechter

et al.

(1973), (9) Smith

et al.

(1972), (10) Gupta

et al.

(1974) (at 44°C).

Early studies indicated that ruminant RBCs lack phos­ phatidylcholine, a finding at odds with a more recent study (Engen and Clark, 1990). Although RBC choles­ terol! phospholipid ratios remain relatively constant, differences occur in membrane phospholipid composi­ tion between neonate and adult RBCs (Marin et al., 1990). A small amount of glycolipid is also located in the outer layer. Species differ in the dominant glycolipids of RBCs (Eberlein and Gercken, 1971; Yamakawa, 1983). Based on studies of human blood group anti­ gens, it is assumed that many animal blood group antigens are also glycolipids and that specificity resides in the carbohydrate moieties.

2. Integral Membrane Proteins Integral membrane proteins penetrate the lipid bi­ layer. These glycoproteins express carbohydrate resi­ dues on the outside surface of the cell. They contribute negative charge to the cell surface, function as receptors or transport proteins and carry RBC antigens (Chasis and Mohandas, 1992; Mohandas and Chasis, 1993;

Schrier, 1985). Band 3 (protein 3) is the major integral protein accounting for approximately one-fourth of the total membrane protein (Schrier, 1985). It is important as an anion transporter and provides a site for binding of the cytoskeleton internally. Additional transmem­ brane glycoproteins called glycophorins also help an­ chor and stabilize the cytoskeleton (Chasis and Mohan­ das, 1992).

3. Membrane Skeletal Proteins The membrane skeleton appears as a dense sweater­ like meshwork bound to the inner surface of the lipid bilayer. It is a major determining factor of membrane shape, deformability, and durability (Mohandas and Chasis, 1993). Major skeletal proteins include spectrin, actin, protein 4.1, ankyrin, and adducin (Fig. 7.2). Spec­ trin is a heterodimer of long, flexible a and f3 chains twisted around one another. Heterodimers are bound together by self-association at their head ends to form tetramers. Multiple spectrin tail ends (average 6) are joined by binding to common short filaments of actin to form an anastomosed meshwork of polygons (Fig.

170

John W. Harvey Erythrocyte Enzymes of Various Animal Species'

TABLE 7.3 Enzyme

Cattle

HK GPI PFK Aldol TPI GAPD PGK MPGM Enol PK LDH DPGM DPGP G6PD 6PGD GR GPx GST SOD Catalase (X 1()3) NADH-MR Na+, K+-ATPase

0.36 ::!: 0.14 (1) 17.0 ::!: 1.0 (1 ) 2.43 ::!: 1.26 (1) 1.46 (2)

0.58 19.8 1.53 1.28

43.3 (2) 16.2 ::!: 7.3 (1) 10.0 ::!: 4.5 (1) 3.15 ::!: 0.98 (1) 5.72 ::!: 1.13 (1) 23 (2) 0.42 (6) 0.03 (6) 5.40 ::!: 0.49 (1) 0.84 ::!: 0.19 (1) 0.69 ::!: 0.47 (7) 165 (8) 4.7 ::!: 0.5 (9) 2060 ::!: 75 (10) 81 (8) 1.83 ::!: 0.24 (1) Nil (11)

57.2 49.6 18.7 8.15 2.82 33.6

Sheep

300

::!: 0.07 (3) ::!: 2.6 (3)

::!: 0.14 (3) ::!: 0.13 (3) ::!: 17 (3) ::!: 5.2 (3) ::!: 2.6 (3) ::!: 1.5 (3) ::!: 0.51 (3) ::!: 0.22 (3) ::!: 3.5 (3)

0.76 ::!: 0.13 (3) 2.60 ::!: 164 ::!: 7.8 ::!: 1910 ::!: 16.6 ::!: 2.00 ::!: HK+, 3 ::!: LK+ Nil

0.22 (3) 21 (3) 1.0 (9) 100 (10) 1.5 (12) 0.12 (3) 1 (11)

Pig

Goat 0.52 ::!: 0.15 (4) 79.1 ::!: 8.5 (4) 2.33 ::!: 0.46 (4) 1.44 ::!: 0.71 (4) 589 ::!: 122 (4) 73.7 ::!: 10.2 (4) 87.8 ::!: 18.3 (4) 19.3 ::!: 6.7 (4) 13.9 ::!: 2.0 (4) 5.00 ::!: 1.23 (4) 17.5 ::!: 3.05 (4) 0.04 (6) 0.01 (6) 2.06 ::!: 0.46 (4) 0.58 ::!: 0.24 (4) 3.87 ::!: 1.57 (4) 179 ::!: 55 (4) 8.0 (13)

0.18 ::!: 0.12 (5) 90.9 ::!: 16.3 (5) 1.23 ::!: 0.85 (5) 0.78 ::!: 0.67 (5) 719 ::!: 211 (5) 40.6 (5) 94.2 ::!: 63.7 (7) 27.3 ::!: 12.6 (5) 8.58 ::!: 6.45 (5) 12.5 ::!: 4.8 (5) 28.7 ::!: 5.9 (5) 0.58 (6) 0.01 (6) 17.3 ::!: 1.2 (5) 3.3 ::!: 1.4 (5) 2.6 ::!: 0.8 (7) 1.4 ::!: 0.2 (9) 1240 ::!: 100 (10) 10 ::!: 2 (11)

a All enzyme units are IU / g Hb, except SOD given in units/ g HB and ATPase given in ILmoles phosphorus liberated/ g Hb/hr. Mean values have been recalculated at times to permit direct comparisons between species. Standard deviation values are given where indicated. Figures in parentheses are the references cited below. Abbreviations used are given in the footnote at the beginning of the chapter. Temperatures above 25°C are included in reference citations. References: (1) Zinkl and Kaneko (1973a), (2) Smith et al. (1972), (3) Agar and Smith (1973, 1974), (5) McManus (1967), (6) Harkness et al. (1969), (7) Agar et al. (1974), (8) Kurata et al. (1993), (9) Del Boccio et al. (1986), (10) Maral et al. (1977), (11) Gupta et al. (1974) (at 44°C), (12) Suzuki and Agar (1983), (13) Board and Agar (1983).

TABLE 7.4 Analyte

Human

G6P F6P FOP DHAP 3PG 2PG PEP Pyruvate Lactate AMP ADP ATP 2,3DPG Pi GSH GSSG

27.8 ::!: 7.5 (1) 9.3 ::!: 2.0 (1) 1.9 ::!: 0.6 (1) 9.4 ::!: 2.8 (1) 44.9 ::!: 5.1 (1) 7.3 ::!: 2.5 (1) 12.2 ::!: 2.2 (1) 53 ::!: 33 (1) 932 ::!: 211 (1) 21.2 ::!: 3.4 (1) 216 ::!: 36 (1) 1438 ::!: 99 (1 ) . 4171 ::!: 636 (1) 480 (5) 2234 ::!: 354 (1) 4.2 ::!: 1.5 (1)

Erythrocyte Chemical Constituents of Various Species' Dog 17.1 5.4 1.4 6.7 48.8 17.6 20.7 24 940

::!: 1.8 (3) 0.5 (3) ::!: 0.2 (3) ::!: 1.0 (3) ::!: 4.2 (3) ::!: 4.6 (3) ::!: 4.1 (3) ::!: 9 (6) ::!: 517 (6) 35 ::!: 6 (3) 211 ::!: 50 (3) 639 ::!: 140 (2) 5989 ::!: 632 (2) 350 (5) 2171 ::!: 344 (2) 6.9 ::!: 1.7 (7)

Horse

Cat 9.1 3.5 3.9 7.8 22.4 24.4 5.3

::!:

82 (5) 529 ::!: 176 (2) 874 ::!: 317 (2) 260 (5) 2106 ::!: 264 (2)

::!:

::!:

::!: ::!: ::!: ::!: ::!:

600 ::!: 2.3 ::!: 16 ::!: 370 ::!: 6220 ::!: 210 (5) 2552 ::!:

2.6 1.1 1.8 4.5 6.6 4.5 1.3

(4) (4) (4) (4) (4) (4) (4)

100 (8) 1.3 (4) 3 (4) 74 (2) 1071 (2) 382 (2)

a Given in nmole/rnl RBC, except lactate and pyruvate which are given in nmole/ rnl whole blood. Mean values have been recalculated at times to permit direct comparisons between species. Standard deviation values are given where indicated. Figures in parentheses are the references cited below. Abbreviations used are given in the footnote at the beginning of the chapter. References: (1) Beutler (1984), (2) Harvey (1994), (3) Harvey et al. (1992b), (4) Smith and Agar (1976), (5) Harkness et al. (1969), (6) Maede and Inaba (1987), (7) Maede et al. (1982), (8) Snow and Martin (1990).

171

The Erythrocyte TABLE 7.5

Erythrocyte Chemical Constituents of Various Species'

G6P F6P FDP DHAP 3PG 2PG PEP Pyruvate Lactate AMP ADP ATP 2,3DPG Pi GSH GSSG

52 28 16 35 32 9 19 54 1989

:!: 15 (1) :!: 11 (1) :!: 12 (1) :!: 11 (1) :!: 12 (1) :!: 4 (1) :!: 7 (2) :!: 24 (1) :!: 758 (1)

73 :!: 22 (1) 633 :!: 115 (1) 289 (5) 400 (6) 2490 :!: 350 (7)

Pigs

Goats

Sheep

Cattle

Analyte

3.2 :!: 0.8 (3) 1.2 :!: 0.7 (3)

28 :!: 10 (2) 11 :!: 2 (2) 25 :!: 10 (2) 10 :!: 3 (2) 11 :!: 21 (2) 14 :!: 10 (2) 19 :!: 14 (2) 87 :!: 21 (2) 1623 :!: 1203 (2) 40 :!: 23 (2) 138 :!: 31 (2) 532 :!: 126 (8) 21 :!: 16 (8) 666 :!: 206 (9) 2257 :!: 130 (10) 40 seconds). Patients with defective intrinsic clotting, thrombocytopenia, or thrombopathia cannot consume prothrombin properly and there is an excess of prothrombin in their serum. The PCT is therefore short in severe defects such as hemophilia, where it may be less than 10 seconds. Thrombocytopenia or thrombopathia can be ruled out by allowing the whole blood for this test to clot in the presence of a source of PF-3 activity such as inosithin.

2. Specific Assays a. Functional Activity Assays Specific quantitative assays of intrinsic clotting ac­ tivity are relatively difficult to perform, time consum­ ing, and costly. These assays measure factors VIII:C, IX, XI, and XII and are based on either manual or automated one-stage (PTT) or two-stage (thromboplas­ tin generation) techniques. The clotting factor in ques­ tion is measured in dilutions of test plasma mixed with standard aliquots of substrate plasma specifically deficient in this factor. The clotting activities of the diluted test plasma are compared by double logarith­ mic plots (calculated manually or by automated instru­ mentation) to those obtained for a standard pool of normal plasma from the same species and are reported as a percentage of the normal plasma or reported as units per milliliter. (One unit is defined as the clotting activity present in 1 ml of normal plasma.) As long as the control and test plasmas are homologous, the deficient substrate plasma can be heterologous (animal or human source). Specific quantitative assays of extrinsic clotting ac­ tivity are based on the PT and / or RVVT tests. Measure­ ments of factors II, V, VII, and X are usually based on these techniques. Animal or human plasma known to be deficient in one of these factors serves as the sub­ strate and is mixed with dilutions of test plasma (Pog­ gio et al., 1991). Equivalent dilutions of a normal,

263

Hemostasis pooled homologous plasma serve as the control. The results are compared by double logarithmic plot and the activity of the patient's sample is reported as the percentage of normal or as units per milliliter. Factor X assays usually use either brain tissue extract or Stypven for activation but may use both of these activators. Factor V and VII assays use the PT technique although factor V can also be measured by RVVT and

by a more complex, two-stage technique. Quantitative assays of prothrombin (factor II) are

levels (Murtaugh and Dodds, 1988; Dodds, 1991b, 1992; Joseph et al., 1996). Immunological assays have also been developed for the antigens related to most of the other coagulation proteins (Dodds, 1989a), and include a new ELISA method for thrombin-ATIll complexes

(Topper et al., 1996). These assays include neutralizing,

precipitating, and radioimmunoassays and have been used to examine whether the plasma of patients with specific functional deficiencies of clotting factors con­ tains normal, reduced, or undetectable amounts of the related antigens (Howard

et al.,

1994). ,

either one-stage methods, based on specific activation of factor II by snake venoms (Smith and Brinkhous, 1991), or two-stage clottability assays. Although the one-stage assay is more easily performed, availability

c.

of the Taipan or tiger snake venom used in the reaction is limited by its expense.

measurements of ristocetin- and venom-induced plate­ let agglutination and macroscopic platelet aggregation

Other Assays Quantitative assays of vWF activity include various

Fibrinogen assays are discussed in Section III.C.I.d. The clot solubility time depends on the concentration of factor XIII and is thus a measure of its activity. Recalcified PRP or PPP clots are put into a solution of

tests (Ermens et al., 1995; Konkle, 1995). There are many demonstrable species differences in the interaction of

5 M urea or 1 % monochloroacetic acid and held at room temperature for 48 hours. Normal clots do not dissolve, whereas factor XIII-deficient clots usually dis­ solve within 24 hours.

modification of existing assays designed for measuring ristocetin cofactor in human plasma to permit its quan­ titation in animal plasmas. One such modification for canines involves diluting normal and patient plasmas

b. Immunological Assays The most widely used immunolOgical assay for di­ agnosis of bleeding disorders measures vWF : Ag [for­ merly called factor VIII-related antigen (FVIIIR : Ag)]. The functions of this component of the factor VIII com­ plex are discussed in Section II.C.2.g. vWF : Ag is usu­ ally quantitated by the very reliable and sensitive ELISA technique, including the commercially available patented test developed by this author and colleagues (Zymtec vWF, Iatric Corp., Tempe, Arizona) (Benson et al., 1991, 1992), and other ELISA-based methods (Meyers et al., 1990a, 1990b; Johnstone and Crane, 1991;

mammalian plasmas with ristocetin (Feldman, 1988; Dodds, 1989a). These differences have necessitated

to a final concentration of not more than 2.5% in the reaction mixture (50 JLI diluted test plasma; 350 JLI washed, formalin-fixed human platelets at 200,000 / JLI; and 100 JLI of ristocetin at 10 mg/ ml). The optical den­ sity settings of a platelet aggregometer are set at zero baseline for each plasma dilution using a blank tube containing 350 JLI Tris buffered saline, 50 JLl diluted test plasma and 100 JLI ristocetin at 10 mg/ ml. It is important to recognize that specific assays for the platelet-related property of human vWF may not be applicable to other species. The snake venom (botro­ cetin) cofactor assay appears to work well with many species and thus avoids the problems encountered

Hoogstraten-Miller et al., 1995; Meinkoth and Meyers, 1995). Alternative methods are the quantitative radio­ immunoassay of Kraus et al., 1989 (described in Feld­ man, 1988), and the original Laurell electroimmunoas­ say used until the ELISA technology was perfected (Dodds, 1989a). Humans and imimals affected with the common forms of vWD (types 1 and 3) will typically have low levels of plasma vWF : Ag (Brooks et al., 1991b, 1992, 1996b; Dodds et al., 1993; Johnstone et al., 1993; Ruggeri and Ware, 1993; Ermens et al., 1995; Konkle, 1995), and patients with severe vWD may also have low FVIII : C (Stokol et al., 1995). When used in

with ristocetin (Smith and Brinkhous, 1991). Chromogenic substrates are also used to assay sev­ eral coagulation and kinin components. For example, chromogenic assays are available for prothrombin, thrombin, ATIll, factors VII and X, plasminogen, kinin­ ogen, and prekallikrein (Tollefsen, 1990; Welles et al.,

conjunction with the FVIII : C assay, the ratio of FVIII : C to vWF : Ag is then calculated to confirm a suspected diagnosis of hemophilia A and can also be used to assist in identifying hemophilic carrier females because of their lowered FVIII.: C but normal vWF : Ag

Tests of the fibrinolytic system are used with the appropriate coagulation and platelet function assays to distinguish between primary fibrinolysis, a rare en­ tity, and the fibrinolysis commonly seen secondary to DIC and consumption coagulopathy (Feldman, 1988;

1990a; Andreasen and Lovelady, 1993).

D. Evaluation of Fibrinolysis 1. General Screening Tests

264

w. Jean Dodds

Comp, 1990a; McKeever et al., 1990; Hargis and Feld­ man, 1991; Welch et al., 1992; Joist et al., 1994; Fareed et al., 1995; Lijnen and Collen, 1995; Verstraete, 1995).

a. Clot Lysis Time The time required for whole blood clots to lyse at 37°C is proportional to the plasmin activity of the blood. It is also dependent on the degree of clot retrac­ tion and fibrinogen content of the sample. Poor clot retraction retards clot lysis whereas hypofibrinogen­ emic clots are friable. A simple, sensitive method for measuring clot retraction and lysis in dilute whole blood is described in Section ill.B.l.b. Normal human blood clots take about 1 6-36 hours to lyse. Most ani­ mals have more active fibrinolytic mechanisms and their blood is usually lysed by 8-20 hours. Clot lysis tends to be more active in females than males and to increase with age and stress.

b. Euglobulin Lysis Time The euglobulins of plasma, such as fibrinogen, plas­ minogen, plasmin, and plasminogen activator, precipi­ tate on dilution with water. Fibrinolytic inhibitors, such as antiplasmin and antiactivator, do not. The plasma euglobulin lysis test measures the time for re­ dissolved, thrombin-clotted euglobulin precipitates to be lysed by endogenous plasmin. The assay should be performed on fresh PPP prepared at 4°C since platelets contain fibrinolytic inhibitor activity and chilling re­ tards the inactivation of plasminogen activator. The use of a tourniquet or other trauma at the site of veni­ puncture should be minimal because it enhances fibri­ nolysis.

c. Fibrin Plate Test The ability of plasma to lyse standardized fibrin plates is a measure of its lytic capacity and depends on the plasminogen and plasminogen activator concen­ tration. A standard fibrinogen solution (250 mg / ml) is clotted uniformly with bovine thrombin (50 NIH units /ml) in a petri dish. The fresh test plasma or its euglobulin precipitate and a standard urokinase solution are applied to duplicate fibrin plates and incu­ bated for 18 hours at 3�C. The same precautions apply to plasma collection and preparation as described for the euglobulin lysis test. The area of lysis on each fibrin plate is expressed in square millimeters as the product of its two perpendicular diameters.

2. Specific Assays a. Functional Activity Assays An accurate functional measure of plasma plasmin­ ogen content is obtained with a caseinolytic assay based on the ability of plasmin to digest a standard

solution of a-casein. The plasma is first acidified to decrease its antiplasmin content and then converted by streptokinase (for human plasma) or urokinase (for animal plasmas) to plasmin. The caseinolytic activity of the resultant plasmin is quantified and compared to that of a known plasmin standard.

b. Other Assays A variety of other assays are used to measure plas­ minogen levels by quantifying the plasmin generated. These include use of chromogenic substrates, rate assays of plasmin activity using basic amino acid ester substrates, specific active-site titration, fluorometric immunoassays, radioimmunoassays, protein inhibitor, and affinity chromatographic methods (Fareed et al., 1 995; Verstraete, 1995). Assays for TPA are also avail­ able (Gerding et al., 1 994; Fareed et al., 1995; Lijnen and Collen, 1 995).

E. Interpretation of Practical Screening Tests Table 10.5 outlines the diagnosis of bleeding disor­ ders based on results of screening tests. These can be expanded on the basis of results for the APTT, PT, TT, and platelet count. Similar approaches are used in humans to interpret the significance of hemostatic tests (Hathaway and Corrigan, 1991; Blomback et al., 1992; Schmidt et al., 1 993; Andrew, 1995; Exner, 1 995).

1. Elevated IT, APIT, and PT, Variable Platelet Count These cases usually have a variable reduction in platelet count and may have abnormally shaped red blood cells on the blood smear. The prolonged TT indi­ cates a reduced amount of functional fibrinogen most likely caused by the presence of in vivo coagulation with production of PDP. The AP1T and PT are pro­ longed whenever fibrinogen-thrombin clottability is impaired. The probable diagnosis is DIC with its requi­ site underlying cause.

2. Elevated APTf (Slight1 Nonnal PT and IT, Decreased Platelet Count Thrombocytopenia results in less available PF-3 to promote intrinsic coagulation and the AP1T is thus slightly prolonged. vWD may present this picture espe­ cially if there is concomitant hypothyroidism. A vWD test is needed to confirm the diagnosis.

3. Elevated APIT (Moderate), Nonnal PT and IT, Nonnal or Increased Platelet Count A moderate intrinsic system clotting defect is pres­ ent. If the patient is a young male with an active bleed­ ing history, hemophilia is a primary consideration.

265

Hemostasis TABLE 10.5

Diagnosis of Bleeding Disorders Test Results

PT

APIT

Bleeding time

IT

Platelet count

vWF :tDecreased

Probable diagnoses Disseminated intravascular coagulation (DIC), acute

Elevated

Elevated

Elevated

Elevated

:tDecreased'

:tElevated

:tElevated

:tElevated

:tElevated

:tDecreased

:tDecreased

DIe, chronic Thrombocytopenia

Elevated

Slightly elevated

Normal

Normal

Decreased

Normal

Elevated

Slightly elevated

Normal

Normal

:tDecreased

Decreased

vWD (especially with concomitant hypothyroidism)

:tElevatedb

Moderately elevated

Normal

Normal

:tElevated

Normal

Hemophilias A or B; Factors X or XI deficiencies

:tElevatedb

Markedly elevated

Normal

Normal

:tElevated

Normal

Hemophilias A or B; Factors X or XI deficiencies

Normal

Markedly elevated

Normal

Normal

Normal

Normal

Factor XII deficiency (Hageman trait), especially in cats; prekallikrein deficiency (Fletcher trait)

:tElevated

Markedly elevated

Normal

Normal

Decreased

Normal

Factor XII deficiency with concomitant thrombocytopenia (cats)

:tElevated

Slightly to moderately elevated

Moderately to markedly elevated

Normal

:tDecreased

Normal

Rodenticide toxicosis, severe liver disease

Normal

Slightly to moderately elevated

Slightly to moderately elevated

Normal

Normal

Normal

Factors II or X deficiencies

Normal

Normal

Moderately to markedly elevated

Normal

Normal

Normal

Factor VII deficiency or early liver disease

• :t,

variably. Primary (platelet-dependent) bleeding time is usually normal but secondary (fibrin-dependent) bleeding time is prolonged; rebleeding is also commonly seen. b

During the stress caused by bleeding, fibrinogen levels will increase rapidly with resultant shortening of the APTf despite the presence of underlying hemophilia. ate defect of intrinsic coagulation (e.g., factor X or XI

XI deficiency). In the absence of a bleeding history,

trinsic clotting defect may be present (e.g., factor X or especially in cats with very long (>60 seconds) APTfs,

Alternatively, this profile may indicate another moder­

congenital factor XII deficiency (Hageman trait) should

defiCiency), but these are relatively rare. Hemophilia

be the first choice. This defect by itself is not of clinical consequence. In animals with a mild to moderate

should be ruled out first. Specific assays of factors VIII : C (hemophilia A) and

IX

(hemophilia B )are

needed for confirmation. The vWF levels should be normal unless both hemophilia and vWD are present

bleeding history and another disease process (e.g., kid­ ney or liver disease), a very long APTT may reflect Fletcher trait (prekallikrein deficiency). A prolonged PT with normal APTT and TT is most

alent).

likely caused by factor VII deficiency or early liver disease. If the defect is vitamin K responsive, liver disease is the likely cause.

4. Elevated APIT (Marked), Normal PT and IT, Normal or Increased Platelet Count

5. Elevated APIT (Marked� Normal PT and IT, Decreased Platelet Count

(e.g., in Doberman pinschers where vWD is so prev­

In a young male with an active bleeding history, hemo­ A severe, intrinsic system clotting defect is present.

philia is most likely. Both hemophilia and vWD could

be present in an affected Doberman pinscher. In other cases of a significant bleeding history, some other in-

When a mild bleeding diathesis is present, especially in cats, the very long APTT is usually caused by an incidental familial factor XII deficiency (Hageman trait), and the bleeding results from concomitant thrombocytopenia.

266

w. Jean Dodds

6. Elevated APIT and PT (Moderate to Marked1 Normal IT and Platelet Count

A significantly prolonged PT with a concomitant prolonged APTf and normal IT is most likely caused by rodenticide toxicosis or severe liver disease. The platelet count may also be reduced. The source of the rodenticide should be investigated and treatment with vitamin Kl instituted for 1-6 weeks depending on the type of rodenticide involved (Mount and Kass, 1989). 7. Elevated APIT, PT (Slight to Moderate1 Normal

IT and Platelet Count

Mild to moderate prothrombin or factor X deficien­ cies can prolong both the APTf and PT somewhat but the IT is normal. 8. Normal APTf, IT, and Platelet Count, Elevated PT

A prolonged PT with normal APIT and IT is most likely caused by factor VII deficiency or early liver disease. If the defect is vitamin K responsive, liver disease is the likely cause. 9. Diagnosis of von Wille brand's Disease

Table 10.6 gives a breakdown of the various catego­ ries for diagnosing vWD based on the bleeding time and vWF antigen level.

10. Distinguishing Between Rodenticide Toxicosis and Thrombosis Animals admitted on an emergency basis for acute nontraumatic bleeding pose a difficult diagnostic chal­ lenge. While the most likely diagnosis is rodenticide toxicosis, the clinician can be misled if the client insists that there has been no opportunity for exposure. With the newer toxicants available today, a nontarget com­ panion animal can become poisoned secondarily, with­ out the owner's knowledge upon ingesting a poisoned rodent that enters the premises. The presumptive diag­ nosis of rodenticide poisoning must be distinguished from bleeding associated with Ole because the clinical signs and presentation can be similar. In the case of thrombosis, an underlying cause must be present but may not be immediately obvious. A complete coagulation profile can readily distin­ guish between rodenticide poisoning and DIe if it de­ tects the presence of FOP either by direct measurement or with the IT. This test is dependent not only on the quantity of fibrinogen present, but also on its ability to be coagulated by thrombin. In the presence of sig­ nificant amounts of FOP, the IT is elevated because

the fibrinogen molecule has been cleaved by the in vivo enzymatic action of thrombin and thereby has been rendered less coagulable. The typical results of coagulation screening tests for either rodenticide toxicosis or Ole are mild to moder­ ate thrombocytopenia; variable prolongation of the PT and APIT-often exceeding 60 seconds each; and quantitative fibrinogen levels varying from low to ele­ vated. At this point, the critical assay that distinguishes between these two potentially life-threatening prob­ lems is the IT. If the IT is normal, the diagnosis is rodenticide toxicosis; if it is elevated, Ole is present. In mild cases of rodenticide exposure or thrombosis, prolongation of the PT and APTf may be subtle. For example, patients undergoing the stress of a bleeding episode may have elevated fibrinogen and platelet numbers as an acute phase response so that the PT and APIT are shortened below the normal ranges. This also occurs in the early hypercoagulative phase of a thrombotic event. In the case of rodenticide poisoning or acute, generalized liver disease, the PT may appear relatively prolonged in comparison to the APTf, IT, and fibrinogen levels, which are all shortened from increased activity. Thus, the PT may be at the upper limit of normal range or slightly prolonged when the APTf and IT are at the lower range or below the normal range. This discordance (elevated PT with shortened APIT and IT) is sufficient to diagnose impairment of the vitamin K-dependent prothrombin complex by either rodenticide exposure, liver disease, or even vitamin K­ deficiency in neonates or from sterilization of the bowel by prolonged antibiotic usage. This pattern of coagula­ tion test results is explained by two opposing actions: the increased generation of thromboplastic activity fol­ lowing a stress event, which shortens the intrinsic coag­ ulation pathway and elevates fibrinogen, as measured by the APTf and IT, respectively, versus the inhibition of prothrombin complex / vitamin K metabolism by ro­ denticide, which blocks shortening of the PT in parallel. The difference between the PT and APIT results can be as little as a 1 -2 seconds lengthening of the PT and a 1-2 seconds shortening of the APTf, with a resulting gap of 4-5 seconds between the end points of both tests.

IV. BLEEDING DISORDERS A. General During the past 40 years, a variety of hemorrhagic diseases have been recognized and studied in animals and humans. Historically, these disorders in animals

Hemostasis TABLE 10.6

Bleeding time"

267

Diagnosis of von Willebrand's Disease

von Willebrand factor antigen (%)b

Interpretation

Normal

� 70

Normal range is 70-180%.

Normal'

70-79

Lower end of normal range; caution advised for breeding stock. Mates should have higher

Normal

50-69

Borderline normal (equivocal result) or heterozygous carrier of vWD gene. Recommend

Normal or elevated'


+A

M . ADP . P + A *

('4(.. �

� �'io'ii. o

o�«: 12 months Male Female

Canineb

>12 months

0-4 months

Caprine' Equine'"

Both sexes Female

3-12 years

Std. Mean dev.

Range

4.7 3.0 2.4 1.8 1.8 0.8 1 .2 0.8

:to.3 :to.2 :to.2 ±0.2 :to.2 :to.2 :to.3 :to.1

22.7

:t7.3

9.9-28.8

0.9

:to.9

0.0-2.5

1 .2-8.2 1 .2-5.5 0.1-6.6 0.4-6.2 0.4-5.6 0.0-7.5 0.2-2.6 0.0-1.8

1.3

:to.9

0.0-3.6

Feline'

Male Female

1.7 1.9

:t 1 .1 :t 1.3

0.4-3.4 0.0-4.5

Ovine'

Male Female

0.8 0.5

:to.6 :to.3

0.0-2.9 0.0-0.9

Gallus 27-134 days domes ticus d 1 year

38.1-59.1 Male Female

109 50

' Enzyme activities determined without reducing agents. Data from Cardinet (1969). b Enzyme activities determined in the presence of glutathione. Data from Heffron et al. (1976). Unexercised; not in training. Enzyme activities determined with­ out reducing agents. Data from Cardinet (1969). d Enzyme activities determined in the presence of cysteine. Data from Holliday et al. (1965). C

425

liberated following intramuscular injection of a drug depends on the properties of the injected solution and on such muscle factors as species differences in muscle CK activity, local blood flow, susceptibility of the mus­ cles and local muscle binding of the drug (Steiness et ai., 1978). Therefore, an accurate history is important in evaluating CK activities. There are three principal isoenzymic forms of CK. Creatine kinase has a dimeric structure consisting of M (muscle) subunits and B (brain) subunits, which combine to form the three heterogeneous MM (or CK3), MB (or CK2), and BB (or CKl) isoenzymes (Dawson et al., 1965, 1967). A fourth variant form, CK-Mt, is found in mitochondrial membranes and may account for up to 15% of the total cardiac CK activity. The isoenzymes can be separated by three different meth­ ods: (1) electrophoresis, (2) immunological techniques, and (3) ion-exchange chromatography (Fiolet et al., 1977). The pattern of isoenzyme distribution varies among the organs of different species. Thus, identification of the isoenzymes present can be used to help determine the tissue source of elevations in CK (Eppenberger et ai., 1964a; Dawson et al., 1965; Dawson and Fine, 1967; Van der Veen and Willebrands, 1966; Sherwin et ai., 1969; Klein et ai., 1973). Ontogenic studies in the rat revealed that all organs investigated contained only BB-CK in early stages of fetal development. In skeletal muscle, BB-CK forms slowly disappear and are initially replaced by MB-CK forms followed by CK-MM forms. Mixtures of isoenzymes occur during the transition. In the adult pattern, MB-CK forms have been variously reported to be present or absent. The inconsistency in noting the presence of MB-CK forms in skeletal muscle of animals may be due to the source of skeletal muscle sampled since all skeletal muscles may not contain the MB-CK isoenzyme (Sherwin et al., 1967; Thorstensson et ai., 1976). However, other studies indicate that, al­ though there is more MM-CK in white muscle of the rat than in red, there is no MB-CK fraction in either (Dawson and Fine, 1967). Reasons for the discrepancies in the detection of MB-CK forms in mammalian skele­ tal muscles are not evident. The adult isoenzyme pat­ tern in muscles of the rat appears at 90 days after birth, whereas in cardiac muscle, the shift occurs earlier and the adult pattern has both MB-CK and MM-CK forms. In the brain, BB-CK is the major isoenzyme throughout life (Eppenberger et ai., 1964a). An unusual sex linkage of muscle CK isoforms has been reported in Harris' hawks (Morizot et ai., 1987). The total CK activity of breast muscle was similar in males and females; how­ ever, the females had three isoenzymic forms, but males had only one, which was judged to be BB-CK.

426

George H. Cardinet

Determinations of serum isoenzyme patterns have found clinical application in human medicine. The de­ termination of MB forms is generally advocated as the best biochemical diagnostic tool for acute myocardial infarction (Fiolet et a/., 1977). However, studies of the isoenzyme composition in cardiac muscle of the horse reveals that less than 1 .5 to 3.9% of the total CK activity is attributable to the MB-CK form. Hence, its determi­ nation cannot be used for detection of myocardial dis­ orders in the horse (Fujii et al., 1980; Argiroudis et a/., 1982). Changes in serum isoenzyme patterns have been observed in various neuromuscular disorders and Du­ chenne muscular dystrophy (Goto et a/., 1969; Somer et a/., 1976). In the horse, One study found that horses with a previous history of rhabdomyolysis had higher serum MM-CK and BB-CK activities, but lower MB­ CK activities than control horses ( Johnson and Perce, 1981). With modest exercise the affected horse had a significant decrease in serum MM-CK and BB-CK, and increase in MB-CK. The significance of those findings has not been elucidated. Elevations in total CK activities have been reported in the hereditary muscular dystrophies of chickens (Holliday et aI., 1965; Wagner et aI., 1971) and Syrian hamsters (Eppenberger et aI., 1964b); in selenium­ vitamin E deficiencies of cattle (Allen et aI., 1975; Van Vleet et aI., 1977), sheep (Whanger et aI., 1970), and swine (Ruth and Van Vleet, 1974); in myodegeneration due to ingestion of toxic plants in cattle (Henson et aI., 1965); in the arthrogryposis-hydranencephaly syn­ drome in calves (Hamada, 1974); in capture myopa­ thies involving a moose (Haigh et aI., 1977) and prong­ horns (Chalmers and Barrett, 1977); and in paralytic myoglobinuria involving horses (Gerber, 1964; Cardi­ net et aI., 1967; Lindholm et a/., 1974b) and a double­ muscled heifer (Holmes et aI., 1972). Elevations in CK activities have also been reported in polioencephalo­ malacia and focal symmetric encephalomalacia of sheep and it has been suggested that CK determina­ tions may be of value in diagnosing diseases of the central nervous system (Smith and Healy, 1968). How­ ever, in diseases of the central nervous system that involve motor function, elevations of CK may be from skeletal muscle rather than from the central nervous system. This appears to be the case in central nervous system disorders in man (Cao et aI., 1969). Diseases of muscle are classified, whenever possible, as to the origin or site of the primary lesion. Myopa­ thies are those diseases in which the primary defect or disease process is considered to be limited to the myofibers, and neuropathies are those diseases of mus­ cle which are secondary changes due to defects or diseases of the neuron (e.g., denervation). Although

III

CK determinations may be specific for diseases of mus­ cle, they do not provide information relative to the origin of the disease process. However, elevations of CK are generally higher in myopathies than neuropa­ thies because myonecrosis is much more prevalent in myopathies than neuropathies. More precise informa­ tion regarding the origin of muscle diseases can be obtained by the use of histological and histochemical examination of muscle biopsies. In a comparative study, the behavior of CK was found to be distinctly different from that of serum aspartate aminotransferase (AST or GOT) during the course of paralytic myoglobinuria in horses (Fig. 16.8). The latter enzyme is discussed in the next section. Ele­ vations in AST activities were present for weeks after the onset of clinical disease, whereas CK activities re­ mained elevated for only a few days. The course of elevations of these enzymes in this disease can be di­ rectly attributed to different disappearance rates of their activity in the plasma (Fig. 16.9). Although CK is more specific for myonecrosis than AST, the simultaneous determinations of AST and CK in the horse are potentially valuable diagnostic and prognostic aids owing to the different disappearance rates of their serum or plasma activities. (1) Elevated CK activities indicate that myonecrosis is active or has recently occurred; (2) persistent elevations of CK indi­ cate that myonecrosis continues to be active; and (3) elevated AST due to myonecrosis accompanied by de­ creasing or normal CK activities indicates that myo­ necrosis is nO longer active. It has not been established whether there are similar differences in the disappear­ ance rates of AST and CK activities in the plasma of other animal species. Therefore, it is not known whether the same assessment of myonecrosis through � § ...

.� :�

2800 �-------' O CK

2400

•

2 000

AST

c:; as

�

"C C as � (.)

o �������� o

6

12

...- Hours

18

24 2 4 6

-t

8 10 12 14 16 18 20 22 24 26 28 Days

I

Time after onset of clinical signs of paralytic myoglobinuria

FIGURE 16.8 The difference in the time course of elevations in AST and CK activities due to muscle necrosis (equine paralytic myoglobinuria). The AST activity remained elevated for much longer periods than CK activity. Adapted from Cardinet et al. (1967).

Skeletal Muscle Function

� §

.� :; � '"

Ien c(

200

i- ..--e-e - - e. - - . - ... - - - ..- - - - ..- - - -e- - - - -.- - - - -.-

1 00 50

"0 c::

'"

::.:: ()

e

1 0 L-__�__-L__-L__� o 50 1 00 1 50 200

e

e �� L-__�__ -L

__ __

Minutes postinjection 250

300

350

FIGURE 16.9 Disappearance of AST (dotted-dashed line) and CK (dotted-solid line) activities in the serum of a horse following the intravenous injection of AST and CK. The differences in the course of serum elevations of these two enzymes due to necrosis in the horse are the result of differences in their disappearance rates in the serum. Adapted from Cardinet e/ Ill. (1967).

the simultaneous determination of AST and CK activi­ ties can be applied to species other than the horse. The normal CK values presented in Table 16.2 were determined by the method of Tanzer and Gilvarg (1959) . Storage of serum at 0 to -5°C results in a loss of activity; however, the addition of reducing agents such as cysteine or glutathione to the incubating me­ dium increases the serum activities, and the losses due to storage are minimized (Okinaka et aI., 1964; Weis­ mann et aI., 1966; Hess et aI., 1968). The cysteine or glutathione reactivates the CK activity lost during stor­ age, therefore, the addition of those reducing agents to the incubating medium is the method of choice and necessitates the establishment of normal "activated" values.

2. Aspartate Aminotransferase Another enzyme that has been used as a diagnostic aid in neuromuscular disorders of domestic animals is serum aspartate aminotransferase (AST) formerly called glutamic oxaloacetic transaminase (GOT). Some normal values of AST activity reported for domestic animals are summarized in Table 16.3. Normal values do not appear to differ greatly between sexes, although reported values for cows (Cornelius et aI., 1959a) are somewhat higher than values for bulls (Roussel and Stallcup, 1966). Differences associated with age have been reported in sheep (Lagace et aI., 1961) and there are seasonal differences in bulls (Roussel and Stallcup, 1966). Also, physical activity is associated with higher values in horses (Cornelius and Burnham, 1963; Cardi­ net et aI., 1963, 1967; Blackmore and Elton, 1975). Elevations of AST activities have been reported in white muscle disease of lambs, cows, and swine (Blin-

427

coe and Dye, 1958; Kuttler and Marble, 1958; Swingle et aI., 1959; Blincoe and Marble, 1960; Whanger et aI., 1970; Ruth and Van Vleet, 1974), tying up and paralytic myoglobinuria in horses (Cornelius and Burnham, 1963; Cardinet et aI., 1963, 1967; Lindholm et aI., 1974b), azoturia in a double-muscled heifer (Holmes et aI., 1972), hereditary muscular dystrophy in chickens (Cor­ nelius et aI., 1959b), myodegeneration due to ingestion of toxic plants in cattle (Henson et al., 1965), and cap­ ture myopathy in a moose (Haigh et aI., 1977) and in pronghorns (Chalmers and Barrett, 1977). Although the use of AST determinations has proven valuable as a diagnostic aid, the enzyme lacks organ specificity since in addition to high concentrations in skeletal and cardiac muscle, AST activities are also high in the liver as well as other organs and tissues, including the red blood cells (RBCs) (Cornelius et aI., 1959a; Nagode et aI., 1966; Cardinet et aI., 1967).

3. Lactate Dehydrogenase Lactate dehydrogenase (LDH) activities are high in various tissues of the body. Therefore, measurements of LDH are not organ specific. Molecules of LDH are tetrameric, made up of four subunits of the two parent molecules, M (muscle) and H (heart). Various combina­ tions of those subunits result in five isoenzymes of LDH, which can be separated by electrophoresis. The M monomer is found in purest form in skeletal muscle as the isoenzyme M4 (or LDH5), whereas the H mono­ mer is found predominantly in the heart muscle as the isoenzyme H4 (or LDH1). The other three forms are molecular hybrids forming the isoenzymes M3H (or LDH4), M2H2 (or LDH3), and MH3 (or LDH2), and they are found in various amounts in different organs. The H4 isoenzyme is maximally active at low concen­ trations of pyruvate and strongly inhibited by excess pyruvate (Dawson et aI., 1964), which favors the oxida­ tion of lactate (Dawson et aI., 1964; Sund, 1969). The M form, on the other hand, maintains activity at relatively high pyruvate concentrations (Dawson et al., 1964), which favors anaerobic reduction of pyruvate (Dawson et aI., 1964; Sund, 1969). Thus, tissues with essentially aerobic metabolism, such as heart muscle, contain mostly heart-specific isoenzymes, whereas tissues with more viable or flexible metabolic properties, such as skeletal muscle, contain predominantly the muscle­ specific isoenzyme (Dawson et aI., 1964). The LDH isoenzyme pattern within skeletal muscle seems to be under genetic control (Dawson et aI., 1964), but is influenced by environmental factors (Sjodin, 1976). Total LDH activity and muscle-specific LDH activity are higher in fast-twitch myofibers than in slow-twitch myofibers. Positive correlations have been

428

George H. Cardinet III TABLE 16.3

Normal Values of Serum Aspartate Aminotransaminase Activity in Some Animals

AST activity (UlIiter) Species Bovine

Canine

Equine

Comment Bull, 1-97 weeks Cows, 2-10 years Calves, 7-27 days

11.4 21.0 1 1 .3

±8.3 ±2.7 ±1.8

2 2

>9 months 5 years

10.9 12.8 10.6 12.4 lOA

±2.6 ±0.8 ±2.6 ±3.7 ±2.5

6-23 6-25 7-14

2 3 4 4 4

165 79.2 89.3 72.5 166 120

±33.8 ± 16.2 ±8.6 ±25.0 ± 75A ±14.9

58-161 53-91 48-170 48-456 88-154

2 5 6 7 5 6

9.1

±2.3

6-13

8

56.5

± 14.9 21-25 10 weeks At capture, day 1 After handling (day 3) After captivity (day 15) Grand mean

65.8 244 65.3 125

Adult

107

Domestic sheep Caprine

Reference'

Std. dev.

Exercised, in training > 1 year

Wild bighorn sheep

Range

Mean

± 7.2 ±48.5 ±52.8 ±94.1

11

61.4

Porcine

1-3 years

Gal/lis domestiClls

6 months

1.5 178

11 11 11 11

±23.l

12

± 6.8

2

±89.3

13

, 1, Roussel and Stallcup (1966); 2, Cornelius et aI. (1959a); 3, Hibbs and Coles (1965); 4, Crawley and Swenson (1965); 5, Cornelius and Burnham (1963); 6, Cardinet et al. (1963); 7, Cardinet et al. (1963); 8, Cornelius and Kaneko (1960); 9; Blincoe and Marble (1960); 10, Lagace et al. (1961); 11, Franzmann and Thome (1970); 12, Lewis (1976); 13, Cornelius (1960).

found between both the individual percentage of fast­ twitch myofibers and muscle lactate concentration, and between lactate concentration and total LDH activity and muscle-specific LDH activity, respectively (Tesch et ai., 1978). Because the fast-twitch myofibers derive more of their energy for contraction via anaerobic me­ tabolism than do slow-twitch myofibers, it seems rea­ sonable that fast-twitch myofibers should contain more of the muscle-specific LDH isoenzymes. Skeletal mus­ cle LDH activity was found to increase with training in men (Suominen and Heikkinen, 1975; Sjodin et al., 1976) and swine ( Jorgensen and Hyldgaard-Jensen, 1975), whereas it has been reported to decrease with training in horses (Guy and Snow, 1977). The reasons for this difference are not obvious. Elevated LDH activities have been reported in selenium-vitamin E deficiency of cattle (Allen et ai., 1975), sheep (Whanger et ai., 1970), and swine (Ruth

and Van Vleet, 1974) and in myoglobinuria in a double­ muscled heifer (Holmes et ai., 1972) and in horses (Rose et ai., 1977). Elevations in LDH activity have also been reported in a variety of hepatic disorders. Therefore, unless isoenzyme analysis is utilized, the measure­ ments of LDH elevations are not organ specific.

4. Serum Pyruvate Kinase Measurements of serum pyruvate kinase (PK) activ­ ities have been found to be more discriminatory than creatine kinase in distinguishing between stress­ susceptible pigs (Duthie and Arthur, 1987; Duthie et al., 1988). Employing PK measurements, homozygous, halothane-reacting (malignant hyperthermic) pigs could be discriminated from homozygous nonreacting pigs. Isoenzymes of PK exist and their distinction may be important in the application of this enzymic assay.

Skeletal Muscle Function

B. Muscle-Specific Serum Proteins and Antibodies With the advent of immunochemical procedures such as enzyme-linked immunosorbent assay (ELISA), radial immunodiffusion assays, radioimmunoassays, and immunocytochemical assay, sensitive tests for the detection of tissue-specific proteins and antibodies in serum are being applied to the diagnosis of neuromus­ cular disorders. The potential sensitivity and specific­ ity of these procedures will no doubt in the future replace a large number of the clinical enzymology assays currently employed.

1. Myoglobin Myoglobin is a 17,500-Da heme protein that stores and transports oxygen in myofibers. Elevated levels of myoglobin have been found in myopathies of humans (Penn, 1986). The specificity of myoglobin for skeletal and cardiac muscle and its plasma clearance make myoglobin determinations a potentially effective method for monitoring myonecrosis (Holmgren and Valberg, 1992). Plasma myoglobin concentrations were compared to aspartate aminotransferase (AST) and cre­ atine kinase (CK) activities following exercise in horses susceptible to rhabdomyolysis (Valberg et al., 1993). The authors report that the detection of myoglobin was the most sensitive indicator of myonecrosis.

2. Myosin Light Chains and Troponin Application of myosin light-chain assays are evolv­ ing for use in the diagnosis of inflammatory muscle disease (Mader et al., 1994) and myocardial infarctions (Nicol et al., 1993), whereas troponin I assays are being promoted for use in the diagnosis of acute myocardial infarctions in humans (Edleman, 1995). These assays offer the potential for improved specificity and sensi­ tivity and are likely to find applications in veterinary diagnostic clinical chemistry.

3. Acetylcholine Receptor Antibodies Detection of circulating autoantibodies to acetylcho­ line receptors (AChR) by an immunopreciptation assay (Lindstrom et al., 1981) is a valuable adjunct to the diagnosis of immune-mediated myasthenia gravis (MG) in humans (Engel, 1994c) and dogs (Lennon et al., 1978; Shelton et al., 1988). It has been estimated that approximately 80% of human patients with MG have detectable AChR antibodies (Engel, 1994c). A valuable immunocytochemical screening test for circulating AChR antibodies in canine MG employs staphylococcal protein A conjugated to horseradish peroxidase (SPA-HRP), a reagent that localizes IgG.

429

Control sections of muscle containing neuromuscular junctions when incubated with canine MG patient sera and subsequently with SPA-HRP localizes IgG at neu­ romuscular junctions (Cardinet and Bandman, 1985). This immunoreagent has also served to detect antinu­ clear antibodies, antistrial antibodies, and sarcolemmal associated antibodies in immune-mediated myasthe­ nia gravis and inflammatory muscle disorders in the dog (Shelton and Cardinet, 1987).

VII. MUSCLE BIOPSY AND HISTOCHEMISTRY IN THE DIAGNOSIS OF NEUROMUSCULAR DISORDERS The use of muscle biopsies in the evaluation of neu­ romuscular disorders (motor unit diseases) allows ex­ amination of myofibers, neuromuscular junctions, in­ tramuscular nerve branches, connective tissues, and blood vessels. As discussed previously, histochemical examination of skeletal muscle provides information relative to the morphological, biochemical, and pre­ sumed physiological properties of myofibers. There­ fore, the application of histochemical techniques in conjunction with routine light and electron micro­ scopic examination of muscle biopsies offers the po­ tential to evaluate and integrate the pathoanatomical, biochemical, and physiological manifestations of neu­ romuscular disorders. Further, the advent of immuno­ cytochemistry has extended our ability to recognize immunopathologic mechanisms and disorders as well. The application of histochemical techniques has be­ come an essential diagnostic procedure for the evalua­ tion of neuromuscular disorders in humans (Engel, 1962, 1965, 1967, 1970; Dubowitz, 1968; Dubowitz et al., 1973). Their application has been most helpful in determining which portion of the motor unit (neuron, myofiber, or both) is involved in the disease process. Their use has been particularly successful in distin­ guishing between neuropathies and myopathies and / or providing profiles specific for selected neuromuscu­ lar disorders (Engel, 1970; Dubowitz et al., 1973). Rou­ tine histochemical procedures have been less informa­ tive in the evaluation of junctionopathies (disorders of neuromuscular transmission). However, immunocyto­ chemical methods add to the evaluation of immune­ mediated diseases of the neuromuscular junction in which deposits of immune complexes are localized along the end plate in myasthenia gravis (Engel et al., 1977; Pflugfelder et al., 1981). Histochemical techniques have been employed in the evaluation of neuromuscular disorders in animals of comparative biomedical research interest such as in the hereditary muscular dystrophies of animals. How-

430

George H. Cardinet

ever, only in recent years have such techniques begun to be routinely applied to the evaluation of muscle biopsies in clinical cases of neuromuscular diseases in animals. A detailed consideration of muscle biopsy techniques in the evaluation of neuromuscular diseases is beyond the scope of this chapter. For details, the reader is referred to Dubowitz et al., (1973), Engel (1994b), and Banker and Engel (1994). During the past two decades, there has been an evolving recognition of the broad spectrum of neuro­ muscular disorders that occur in animals. Neuromus­ cular disorders in animals are associated with sponta­ neous and inherited endocrine, immune-mediated, infectious, metabolic, and neoplastic diseases. With continued application of those procedures and newer technologies, undoubtedly many heretofore unrecog­ nized neuromuscular disorders in animals will be rec­ ognized (Cardinet and Holliday, 1979). To date, disor­ ders of carbohydrate, lipid, electron transport and electrolyte metabolism have been recognized in skele­ tal muscles of domestic animals common to clinical veterinary medicine.

VIII. SELECTED NEUROMUSCULAR DISORDERS OF DOMESTIC ANIMALS A. Ion Channelopathies

1. Acetylcholine Receptor Ion Channels and Myasthenia Gravis Myasthenia gravis is a disorder of neuromuscular transmission in which there is a reduction in the num­ ber of ligand-gated AChR, ion channels on the postsyn­ aptic sarcolemmal membrane (PSM) . This condition results in weakness due to the reduced sensitivity of the PSM to the transmitter, ACh. Two basic forms of MG exist: (1) acquired autoimmune MG and (2) con­ genital MG. Different mechanisms are responsible for the reduction of AChRs in these two disorders.

a. Acquired Autoimmune Myasthenia Gravis Acquired MG is an immune-mediated disorder of humans (Engel, 1994c), dogs (Lennon et al., 1978, 1981 ), and cats (Indrieri et al., 1983) in which autoantibodies are produced against AChRs. The density of AChRs is reduced by the complement-mediated destruction, accelerated internalization, and degradation of AChRs by cross-linking of the receptors by antibody. In hu­ mans (Tzartos ct al., 1982) and dogs (Shelton et al., 1988), the autoantibody response is heterogeneous. Most antibodies are IgG and directed against the main immunogenic region (MIR), a specific external portion of the a-subunit that is distinct from the ACh-binding

III

site; however, autoantibodies are also produced against all of the other subunits. Only a small percent­ age of antibodies is directed against the ACh-binding sites on the a-subunits. Dogs with MG usually exhibit some form of muscu­ lar weakness; however, this can be quite variable and may include the following presentations: acute quadri­ plegia, degrees of exercise-related weakness, gait ab­ normalities, or no apparent limb muscle weakness with dysphasia or regurgitation associated with a mega­ esophagus. There appears to be a bimodal distribution in the onset of this disease in dogs (early and late onset). Dogs are rarely affected before one year of age and peak frequencies were found at 3 and 10 years of age, respectively; the prevalence did not appear to be gender related (Shelton et al., 1988). Frequently, overt signs may be limited to esophageal dysfunction. In a study of 152 dogs afflicted with idiopathic megaeso­ phagus, 40-57 dogs (26% to 38%) had MG (Shelton et al., 1990). A definitive diagnosis of MG is provided by detec­ tion of circulating antibodies to the AChR. Additional diagnostic tests that provide for a presumptive diagno­ sis include clinical, pharmacologic, electrodiagnostic, and immunocytochemical methods of evaluation. When clinical signs permit the objective assessment of strength, pharmacologic testing can be employed through the intravenous administration of 1-10 mg oj edrophonium chloride, an ultra-short-acting anticho· linesterase agent. Improved strength with edropho· nium provides a presumptive diagnosis of MG. A pre· sumptive diagnosis of MG is also suggested when thE application of low-frequency (2-10 Hz), repetitiVE nerve stimulation results in the reduced amplitude oj the first few evoked compound MAPs (decrement­ ing response). Two immunocytochemical procedures provide sen· sitive presumptive tests for MG. In muscle biopsies 01 human and canine MG patients that contain neuromus· cular junctions, it is possible to localize the IgG bounc to the PSM using the immunoreagent SPA-HRP (Enge et al., 1977; Pflugfelder et al., 1981). Similarly, patien sera applied to control sections and subsequentl) treated with SPA-HRP will detect circulating antibod· ies that bind to the PSM of neuromuscular junction� (Cardinet and Bandman, 1985; Cain et al., 1 986; Sheltor and Cardinet, 1987).

b. Congenital Myasthenia Gravis Congenital MG is a developmental disorder of hu mans (Engel, 1994d) and dogs (Lennon et al., 1981 Miller et al., 1983; Oda et al., 1984a). In the dog, th{ syntheSiS of AChRs appears to be normal and degrada tion does not appear to be accelerated. The reducec

Skeletal Muscle Function

AChR density is believed due to a low insertion rate of AChRs into the PSM (Oda et al., 1984b). In humans, several congenital myasthenic syndromes are recog­ nized and include putative failure of packaging or re­ syntheSiS of ACh, the absence of AChE from the basal lamina, and a deficiency of AChRs and abnormal chan­ nel function. Because congenital MG is not immune mediated, the immunodiagnostic tests used in ac­ quired MG are of no value in establishing the diagnosis of congenital MG.

2. Sodium Ion Channels and Periodic Paralysis Equine hyperkalemic periodic paralysis (HYPP) is a dominantly inherited disorder of muscle that causes episodes of stiffness (myotonia), weakness, tremors, and / or paralysis in association with elevated serum potassium (Spier et al., 1990, 1993). Weakness or paraly­ sis can be induced by the ingestion of potassium. Linkage studies have revealed that HYPP cosegre­ gates with the equine adult skeletal muscle sodium channel a-subunit gene (Rudolph et al., 1992a) involv­ ing a Phe to Leu mutation in the transmembrane domain IVS3 (Rudolph et al., 1992b). Patients are het­ erozygous and express both normal and mutant a­ subunits. The primary physiological defect in mutant sodium channels is impaired inactivation (Cannon et al., 1995). With incomplete inactivation, sodium ion conductance into myofibers is sustained, which in turn sustains depolarization of the sarcolemma. It appears likely that the clinical variability and severity of signs is associated with the ratio of mutant to normal sodium ion channels expressed in the skeletal muscles of af­ fected horses (Zhou et al., 1994). This disorder in horses is similar to primary HYPP in humans in which there is also a mutation involving the gene encoding the sodium ion channel a-subunit. A number of mutations involving the gene encoding the a-subunit of the skeletal muscle sodium ion chan­ nel have been identified in humans that are responsible for the HYPP disorder and paramyotonia congenita, an exercise and cold-induced stiffness and weakness (Lehmann-Horn et al., 1994). Diagnosis of equine HYPP is aided by knowledge of the breed and pedigree, clinical signs, and the assess­ ment of serum potassium ion concentrations during an episode of weakness or stiffness. Oral administration of potassium chloride can be used as a test to induce weakness in susceptible horses. A definitive diagnosis is possible by base-pair analysis of the DNA sequence responsible for encoding of the a-subunit (Rudolph et al., 1992b).

431

3. Chloride Ion Channels and Myotonia Myotonia is a clinical sign in which the uncontrolled, prolonged, and painless contraction of skeletal muscles occurs. The condition is due to hyperexcitability of the sarcolemma and the abnormal production of repetitive depolarizations of the sarcolemma followed by de­ layed repolarization and relaxation. Affected patients exhibit varying degrees of muscle stiffness with the onset of exercise. The stiffness will often subside with continued exercise or repeated movements and is not aggravated by cold. Muscles may be grossly hypertro­ phied with well-defined muscle groups. Percussion of muscles results in local contractions that create dim­ pling of the surface overlying the contracting muscles. Myotonia congenita (MC) occurs in goats in which there is impaired conductance of chloride ions across the sarcolemma of T-tubules (Bryant and Morales­ Aguilera, 1971; Adrian and Bryant, 1974). Unlike, HYPP, MC is a nonprogressive disorder that improves with exercise ("warm-up") and is not induced by cold. Histopathologic changes in skeletal muscle are usually minimal and nonspecific. Two principal myotonic disorders occur in humans: (1) myotonia congenita (Rudel et al., 1994) and (2) myo­ tonic dystrophy (MD; Harper and Rudel, 1994). Myoto­ nia congenita is a nonprogressive childhood disorder in which there is a diminished chloride conductance across the sarcolemma caused by mutations of the skel­ etal muscle chloride ion channel. Two forms of MC are recognized in which the mutations of the chloride ion channel involve autosomal dominant (Thomsen' s disease) and recessive (Becker's myotonia) modes of inheritance (Koch et al., 1992). Myotonic dystrophy is an autosomal dominant myotonic disorder in humans that differs from MC in that it is progressive and variably involves a variety of other systems (e.g., smooth muscle of hollow organs, heart, brain and peripheral nerves, endocrine glands, eyes, skeletal system, and integument). The underlying cause of this disorder has not been established even though the gene has been localized. Onset of signs may be recognized in childhood or later in adults, hence MD can also have a congenital presentation. The systemic features are most helpful in differentiating between MC and MD in addition to histopathologic features, which include increased central nuclei, ringed fibers, sarcoplasmic masses, and type 1 fiber atrophy. In addition to goats, myotonic disorders have been described in horses (Steinberg and Botelho, 1962; Jami­ son et al., 1987; Reed et al., 1988) and dogs (Kortz, 1989). In attempting to draw comparisons between these dis­ orders and the human disorders, some confusion exists

432

George H. Cardinet III

since most reports describe a congenital onset compati­ ble with MC but physical and histopathologic features that are similar to MD. In the horse, two reports pro­ vide clinical and histopathologic features in foals that appear comparable to MD ( Jamison et a/., 1987; Reed et a/., 1988). In dogs, congenital presentations are de­ scribed in chow chows (Griffiths and Duncan, 1973; Wentink et a/., 1974; Amann et a/., 1985) among a num­ ber of other breeds. The histopathologic features ob­ served in these dogs have been variable. More detailed studies of membrane function and molecular genetics will be required to elucidate fully the pathogenesis and comparative features of these disorders.

4. Voltage-Sensitive Calcium Ion Channels of the Sarcoplasmic Reticulum and Malignant Hyperthermia Malignant hyperthermia (MH) is an inherited, au­ tosomal dominant, pharmacogenetic disorder of hu­ mans and swine, and possibly dogs and horses, in which there is a rapid increase in body temperature and muscle rigidity associated with the uncontrolled contraction and metabolism of muscles leading to rhabdomyolysis, and death (Gronert, 1994). The condi­ tion is triggered by the administration of volatile anes­ thetic agents and exacerbated by succinylcholine. In swine, stresses such as fighting, transport, and exercise also trigger its onset. In swine, MH is due to an alteration in the calcium ion release channel (ryanodine receptor) of the SR due to a mutation in the gene for encoding the ryanodine receptor (Fujii et a/., 1991). The alteration increases the sensitivity of the calcium release channels and results in their prolonged open state and release of calcium into the sarcoplasm. In humans, the same mutation has been observed in some families; however, other mutations are likely and some cases do not appear to be linked to the ryanodine receptor gene (MacLennan and Phillips, 1992). Diagnostic testing for susceptibility to MH involves a standardized in vitro contracture test employing mus­ cle biopsy samples in which contractures induced by caffeine or halothane are measured (Larach, 1989). During an attack, serum enzyme activities are mark­ edly elevated due to extensive myonecrosis. No spe­ cific histopathologic features are present in susceptible individuals. With the identification of the mutation in the gene for encoding the ryanodine receptor, molecu­ lar genetic testing will provide a definitive test in swine (Houde et a/., 1993; O'Brien et a/., 1993). There are reports of MH-like presentations in horses associated with halothane anesthesia (Hildebrand and

Howitt, 1983; Manley, et al., 1983) and in dogs associ­ ated with exercise (Rand and O'Brien, 1987). Contrac­ ture testing (Nelson, 1991) and calcium homeostasis defects (O'Brien et a/., 1990) suggest the stress / exercise­ induced syndrome in dogs may be MH.

B. Cytoskeletal Dystrophin Deficiency and Muscular Dystrophy Duchenne muscular dystrophy is an X-linked reces­ sive disorder of skeletal muscle in humans (Engel et al., 1994), dogs (Kornegay et al., 1988; Cooper et a/., 1988), cats (Carpenter et al., 1989; Gaschen et al., 1992) and mice (Bulfield et a/., 1984). The disorder is due to a deficiency of dystrophin (Hoffman et a/., 1987), a sub sarcolemmal cytoskeletal protein that participates in the attachment of myofibrils to the sarcolemma. Dystrophin in concert with a trans­ membrane protein complex (dystrophin-associated protein) is believed to provide stability to the sarco­ lemma. Presumably, the deficiency of dystrophin cre­ ates instability of the sarcolemma; however, the precise mechanisms for initiating myofiber necrosis are not known. A number of mutations of the dystrophin gene have been identified. In the dog, dystrophin deficiency is due to a splice-site mutation in which there is an RNA processing error (Sharp et al., 1992). The mutation in the cat has not been identified. The expression of the disease varies among species but may be characterized as a progressive degenerative disorder of muscle in which there is a marked increase in serum enzyme activities, gross hypertrophy of some muscle groups, and a stiff, "bunny hopping" gait in dogs and cats. Affected individuals may also have a cardiomyopathy since the dystrophin deficiency also involves cardiac myofibers (Moise et a/., 1991). In dogs, the onset of clinical signs is usually evident by 2-4 months of age and somewhat later in cats. In dogs, this disorder was first observed in golden retrievers and has subsequently been identified in other breeds, including Irish terriers, rottweilers, and samoyeds. Histologic sections reveal focal lesions consisting of myofiber clusters undergoing the spectrum of change from myonecrosis through macrophage infiltration and phagocytosis to regeneration. Individual fibers may be atrophic or hypertrophic, calcified, and hyper­ contracted and possess central nuclei. The focal cluster­ ing of necrotic fibers may be due to the local effects of mast cell accumulations and degranulation (Gorospe et al., 1994). Beyond clinical signs and biopsy features, immu­ nob lotting and immunocytochemical staining for dys­ trophin are valuable diagnostic methods for detecting

Skeletal Muscle Function

deficiencies of dystrophin within muscle biopsies (Cooper et al., 1990; Gaschen et al., 1992).

C. Immune-Mediated Canine Masticatory Muscle Myositis The muscles of mastication are selectively affected in two inflammatory muscle disorders in dogs known as eosinophilic and atrophic myositis. Limb muscles are essentially spared. These disorders may be distinct and separate entities or merely variations of the same disor­ der that distinguish between acute and chronic on­ sets, respectively. The muscles of mastication in the dog are principally composed of type 2M myofibers, fast-twitch fibers that possess a unique myosin isoform, heavy chain, and light chains (Shelton et al. 1985a). Recent evidence sug­ gests this disorder is an autoimmune disease (Shelton et al., 1985b, 1987). Dogs afflicted with this disorder produce autoantibodies directed against type 2M fibers and there is little cross-reaction of the antibodies against type 2A fibers of limb muscles. The specificity of these antibodies for 2M fibers provides presumptive evidence for its immune pathogenesis and the limited distribution of lesions to the muscles of mastication. Diagnosis of the disorder includes muscle biopsy demonstration of an inflammatory disorder, localiza­ tion of immunoglobulins fixed to type 2M fibers within �he biop sy, and demonstration of circulating antibod­ Ies agamst type 2M fibers employing SPA-HRP.

D. Disorders of Glyco(geno)lysis Affecting Skeletal Muscle Disorders of glyco(geno )lysis affecting skeletal mus­ cl�s �ariably involve some excess storage of glycogen wlthm affected myofibers, resulting in the presence of glycog:n-containing vacuoles. In humans, glycogen storage dIseases (GSD) affecting muscle include defi­ ciencies of a-1,4-glucosidase (GSD-II), debranching en­ zyme (GSD-III), branching enzyme (GSD-IV), myo­ phosphorylase (GSD-V), phosphofructokinase (GSD­ VII), phosphoglycerate kinase (GSD-IX), phosphoglyc­ erate mutase (GSD-X), and lactate dehydrogenase (GSD-XI; Engel and Hirschhorn, 1994; DiMauro and Tsujino, 1994). Though variably documented, some of these disorders also occur in domestic animals (Wal­ voort, 1983).

433

childhood (infantile and juvenile) and adult forms, which are variations of the same disorder based on the age of onset and tissue and organ involvement. This disorder is inherited as an autosomal recessive trait (Engel and Hirschhorn, 1994). This disorder has also been reported in shorthorn and Brahman cattle ( Jolly et al., 1977; Richards et al., 1977; O'Sullivan et al., 1981). Both infantile and late onset equivalent variations have been described (How­ ell et al., 1981). Clinical signs in Brahman calves become evident at 2-3 months of age with a loss of condition, poor growth, and lethargy followed by incoordination and muscle tremors with death by 9 months of age. Although the onset of clinical signs may also occur within the first 2-3 months of age, some affected short­ horn calves appear clinically normal until 5-9 months of age when weight gains are not maintained and pro­ gressive muscular weakness develops with death by 12-16 months of age. Excessive accumulation of glyco­ gen occurs in skeletal and cardiac muscle, brain, and spinal cord. The disorder in cattle is also inherited as an autosomal recessive trait (Howell et al., 1981). In other species, single cases of generalized glyco­ genoses in which a-glucosidase deficiency were dem­ onstrated in the Lapland dog (Walvoort et al., 1982) and Japanese quail (Murakami et al., 1980). In addition, generalized glycogenoses have also been described in sheep (Manktelow and Hartley, 1975), cats (Sandstrom et al., 1969) and dogs (Mastofa, 1970; Walvoort et al., 1981); however; those conditions were not character­ ized biochemically. 2.

Debranching Enzyme Deficiency (GSD III)

Debranching enzyme possesses two activities, a-1,4glucan transferase and a-1,6-glucosidase. In the hydro­ lysis of glycogen, myophosphorylase acts on the a-1,4 linkages of the terminal glucose residues up to the last four glucose residues preceding the a-1,4 linkages. The a-1,4-glucan transferase transfers the last three resi­ dues to another branch and the a-1,6-glucosidase hy­ drolyses the a-1,6 branch point. Several presumed cases of debranching enzyme de­ ficiency have been described in the dog (Rafiquzzaman et al., 1976; Ceh et al., 1976; Otani and Mochizuki, 1977). Biochemical demonstration of debranching enzyme deficiency is limited to a single case (Ceh et al., 1976). There is diffuse organ involvement with glycogen stor­ age. Onset of signs appear at 2 months of age and the disorder is progressive with death by 10-15 months.

1. a-1,4-Glucosidase Deficiency (GSD II)

3. Branching Enzyme Deficiency (GSD IV)

Lysosomal a-1,4-glucosidase deficiency, also known as Po�pe' s disease, generalized type II glycogenosis, and aCid maltase deficiency, occurs in humans with

Glycogen storage associated with branching en­ zyme deficiency has been reported in a family of Nor­ wegian forest cats (Fyfe et al., 1992). Clinical signs and

434

George H. Cardinet III

involved organs include skeletal muscle, heart, and the central nervous system. Abnormal glycogen was evident in tissues at birth. Hepatic involvement and function was reportedly normal.

4. Myophosphorylase Deficiency (GSD V) Myophosphorylase deficiency (McArdle' s disease) is an inherited, autosomal recessive, glycogenosis in humans (DiMauro and Tsujino, 1994) and Charolais cattle (Angelos et aI., 1995). In cattle, clinical signs include recumbency and fa­ tigue with forced exercise and elevated serum en­ zymes. Muscle glycogen concentrations were elevated 1.6 times higher than controls. Histopathologic changes are modest with some vacuolated myofibers; however, the vacuoles did not contain glycogen. A rapid diagnosis is possible by employing the histo­ chemical staining method for demonstrating myo­ phosphorylase activity in frozen sections of biopsy specimens, which may be confirmed by quantitative analyses. 5. Phosphofructokinase Deficiency (GSD VII)

Phosphofructokinase (PFK) is a key enzyme of the Embden-Meyerhof pathway in all tissues and inher­ ited deficiencies of this enzyme in humans are ex­ pressed primarily as a myopathy in which it is desig­ nated as type VII glycogen storage disease (GSD-VII). There is a partial expression of hemolysis and the defi­ ciency is otherwise quite heterogeneous (Rowland et al., 1986; Kaneko, 1987). Deficiency of PFK was reported in springer spaniel dogs that presented a clinical picture remarkably simi­ lar to that seen in humans. The dogs presented with a history of intermittent severe hemolytic episodes, and total erythrocyte PFK activity was 10% of normal controls (Giger et aI., 1985). The disorder �s inherited as an autosomal recessive trait (Giger et ai., 1986). Mammalian PFK is present in different tissues as tetramers of three subunits, PFK-M (muscle), PFK-L (liver), and PFK-P (platelets). Human muscle and liver contain homogenous tetrameric PFK-M4 and PFK-L4, respectively. The erythrocytes contain an admixture of PFK-M and PFK-L tetramers. In normal dogs, a similar isoenzymic distribution pattern exists except in the erythrocytes where the PFK tetramers consist of an admixture of PFK-M and PFK-P subunits (Vora et aI., 1985). As in humans, PFK deficiency was found to be a deficiency of the PFK-M subunits and the erythrocytic hybrids consisted of the PFK-L and PFK-P subunits. Giger et al. (1985) speculate that exertional stresses inducing hyperventilation and respiratory alkalosis in

turn directly or indirectly enhance the alkaline fragility of their erythrocytes and subsequent hemolysis. Early reports suggested that affected dogs did not manifest severe muscle-related signs, which were pos­ sibly masked by signs referable to hemolysiS. How­ ever, biochemical studies have revealed reduced gly­ colysis in muscle (Giger et aI., 1988a, 1988b), and pathologic studies have established the presence of a myopathy that included the presence of PAS positively stained polysaccharide storage vacuoles in up to 10% of the myofibers (Harvey et aI., 1990).

6. Polysaccharide Storage Myopathy A polysaccharide storage disorder has been de­ scribed in quarterhorses, quarterhorse crossbreds, american paints, and appaloosa horses (Valberg et aI., 1992). The horses had recurrent episodes of exertional rhabdomyolysis and myoglobinuria. Up to 5% of the type 2 muscle fibers had vacuoles that contained acid mucopolysaccharide inclusions that were brilliantly stained with the periodic-acid Schiff ' s (PAS) stain. These inclusions consisted of ,a-glycogen particles and fibrillar material. When treated with amylase, the PAS staining intensity was reduced only slightly. Glycogen concentrations were approximately 45% greater in the muscles of horses with these polysaccharide inclu­ sions. An in vitro biochemical method was employed fOI screening deficiencies of enzymes involved in anaero­ bic glyco(geno )lysis. Affected muscles were capablE of utilizing glycogen, glucose-I-phosphate, glucose6-phosphate, fructose-6-phosphate, and fructose-I,6phosphate as substrates to produce lactate; however, the amount of lactate produced was significantly lowel than the controls. This study strongly points to an underlying metabolic abnormality that remains to be precisely defined.

E. Mitochondrial Myopathies A large number of complex mitochondrial myopa­ thies have been described in humans (Morgan-Hughes, 1994) and, recently, there have been reports of mito· chondrial muscle disorders in animals. When the utili· zation of oxygen is reduced, lactate formation iE favored and results in elevated blood lactate even al low-intensity levels of exercise. Disorders of the respi­ ratory chain often result in increased numbers of mi· tochondria and aggregates of mitochondria undel the sarcolemma. In sections, these mitochondrial ago gregates impart a "ragged-red" appearance to the pe­ riphery of myofibers when stained with the modified trichrome stain. The presence of ragged red myofiben

435

Skeletal Muscle Function

and elevated lactate concentrations with exercise war­ rant investigation of metabolic abnormalities involving mitochondria.

1. Respiratory Chain Complex I (NADH Ubiquinone Oxioreductase) Deficiency A deficiency of Complex I was observed in an Ara­ bian mare with severe limitations to even mild exercise in which there was an elevated venous p02l 10w oxygen consumption, and lactic acidosis (Valberg et al., 1994). Complex I of the respiratory chain is One of four com­ plexes involved in oxidative phosphorylation and transfers electrons from NADH to CoQ in the Conver­ sion of oxygen to water.

2. Cytochrome c Oxidase Deficiency An episodic weakness reported in Old English sheepdog littermates is accompanied by elevated serum enzymes, lactic acidosis, and increased p02 (Breitschwerdt et ai., 1992). This disorder may involve a reduction in cytochrome c oxidase (Vijayasarathy et aI., 1994).

F. Endocrine Myopathies Signs referable to muscle weakness are frequently observed as part of the clinical presentations of endo­ crine disorders.

1. Corticosteroid Myopathy Hyperadrencorticism causes muscle wasting (atro­ phy) and weakness in dogs with Cushing's disease and following corticosteroid administration (Braund et al., 1980a, 1980b; Duncan and Griffiths, 1977). The muscle wasting is due to a rather selective anguloid­ to-angular atrophy of type 2 myofibers; however, quantitative studies reveal atrophy of type 1 fibers as well. Myotonia is a variable accompanying sign of this disorder (Duncan and Griffiths, 1977). Muscle wasting appears to be due to catabolism with decreased protein synthesis and increased protein degradation mediated by altered transcription in protein metabolism (Kamin­ ski and Ruff, 1994).

2. Hypothyroid Myopathy Thyroid status has a profound affect on skeletal muscle and hypothyroid states are often accompanied by manifestations of neuromuscular disease. However, descriptions of muscle disorders in clinical veterinary medicine are limited (Braund et al., 1981). Selective type 2 myofiber atrophy and type 1 myofiber predomi­ nance (or type 2 myofiber paucity) are common find-

ings in canine hypothyroidism. Experimental studies reveal that the proportion of myofiber types is influ­ enced by thyroid status in which thyroidectomy results in type 1 myofiber predominance, and thyroid excess results in type 2 myofiber predominance (Ianuzzo et aI., 1980). These changes may be mediated via neural influences since denervation abolishes these effects of thyrOidectomy ( Johnson et al., 1980).

References Adrian, R. H., and Bryant, S. H. (1974). J. Physiol. 240, 505. Allen, W. M., Bradley, R., Berrett, S., Parr, W. H., Swannack, K, Barton, C. R. Q., and Macphee, A. (1975). Br. Vet. J. 131, 292. Amann, J. F., Tomilson, J., and Hakinson, J. K (1985). JAVMA 187, 415. Anderson, M. G. (1975). Equine Vet J. 7, 160. Angelos, S., Valberg, S. J., Smith, B. P., McQuarrie, P. 5., Shanske, 5., Tsujino, 5., DiMauro, S., and Cardinet III, G. H. (1995). Muscle Nerve 18, 736. Argiroudis, S. A., Kent, J. E., and Blackmore, D. J. (1982). Equine Vet. J. 14, 317. Ashmore, C. R., and Doerr, L. (1971). Exp. Neurol. 31, 408. Asmundson, V. 5., and Julian, L. M. (1956). J. Hered. 47, 248. Baldwin, K. M., Reitman, J. 5., Terjung, R. L., Winder, W. W., and Holloszy, J. O. (1973a). Am. J. Physiol. 225, 1045. Baldwin, K. M., Winder, W. W., Terjung, R. L., and Hol!oszy, J. O. (1973b). Am. J. Physiol. 225, 962. Banker, B. Q., and Engel, A. G. (1994). In "Myology" (A. G. Engel and C. Franzini-Armstrong, eds.), Vol. 1, pp. 832-888. McGraw­ Hill, New York. Barany, M. (1967). J. Cen. Physiol. 50, 197. Barany, M., and Close, R. 1. (1971). J. Physiol. 213, 455. Barany, M., Barany, K, Reckard, T., and Volpe, A. (1965). Arch. Biochem. Biophys. 109, 185. Barden, J. A., and Mason, P. (1978). Science 199, 1212. Barnard, R. J., Edgerton, V. R., and Peter, J. B. (1970). J. Appl. Physiol. 28, 762. Bass, A., Brdiczka, D., Eyer, P., Hofer, S., and Pette, D. (1969). Eur. J. Biochem. 10, 198. Beatty, C. H., and Bocek, R. M. (1970). In "Physiology and Biochemis­ try of Muscle as a Food" (E. J. Briskey, R. G. Cassens, and B. B. Marsh, eds.), Vo!' 2, p ISS. Univ. of Wisconsin Press, Madison, Wisconsin. Beatty, C. H., Peterson, R. D., and Bocek, R. M. (1963). Am. J. Physiol. 204, 939. Beatty, C. H., Basinger, G. M., and Bocek, R. M. (1966a). Arch. Biochem. Biophys. 117, 275. Beatty, C. H., Basinger, G. M., Dully, C. c., and Bocek, R. M. (1966b). J. Histochem. Cytochem. 14, 590. Beatty, C. H., Peterson, R. D., Basinger, G. M., and Bocek, R. M. (1966c). Am. J. Physiol. 210, 404. Beatty, C. H., Basinger, G. M., and Bocek, R. M. (1967). J. Histochelil. Cytochem. 15, 93. Beecher, G. R., Kastenschmidt, L. L., Cassens, R. G., Hoekstra, W. G., and Briskey, E. J. (1968). J. Food Sci. 33, 84. Billeter, R., Weber, H., Lutz, H., Howard, H., Eppenberger, H. M., and Jenny, E. (1980). Histochem. 65, 249. Blackmore, D. J., and Elton, D. (1975). Equine Vet. J. 7, 34. Blanchaer, M. C. (1964). Am. J. Physiol. 206, 1015. Blincoe, c., and Dye, W. B. (1958). J. Anim. Sci. 17, 224. Blincoe, c., and Marble, D. W. (1960). Am. J. Vet. Res. 21, 866.

436

George H. Cardinet III

Bocek, R. M., and Beatty, C. H. (1966). ]. Histochelll. Cytocilelll. 14, 549. Bocek, R. M., Basinger, G. M., and Beatty, C. H. (1966a). Am. ]. Physiol. 210, 1108. Bocek, R. M., Peterson, R. D., and Beatty, C. H. (1966b). Am. J. Physiol. 210, 1 1Ol. Braund, K. G., and Lincoln, C. E. (1981). Am. J. Vet. Res. 42, 407. Braund, K. G., Edwin, J. H., and Richardson, K. E. Y. (1978). Am. J. Vet. Res. 39, 56l. Braund, K. G., Dillon, A. R., and Mikeal, R. 1. (1980a). Exp. Neural. 68, 50. Braund, K. G., Dillon, A. R., Mikeal, R. 1., and August, J. R (1980b). Vet. Pathol. 17, 134. Braund, K. G., Dillon, A. R, August, J. R, and Ganjam, V. K. (1981). Vet. Pat/wi. 18, 589. Breitschwerdt, E. B., Kornegay, J. N., Wheeler, s. J., Stevens, J. B., and Baty, C. J. (1992). ]AVMA 201, 73l. Brockman, K. G., and McKenzie, K. K. (1983). Am. J. Physiol. 244, H552. Brooke, M. H., and Kaiser, K. K. (1970). Arch. Neural. 23, 369. Brooke, M. H., Williamson, E., and Kaiser, K. K. (1971). Arch. Neural. 25, 360. Bryant, S. H., and Morales-Aguilera, A. (1971). J. Physiol. 219, 367. Bulfield, G., Siller, W. G., and Wright, P. A. 1. (1984) PNAS 81, 1 189. Bullard, H. H. (1912). All!. J. Anat. 14, l. Buller, A. J., and Lewis, D. M. (1965). J. Physiol. 178, 343. Buller, A. J., Eccles, J. c., and Eccles, R. M. (1960). J. Physiol. 150, 417. Burke, R. E. (1994). In "Myology" (A. G. Engel and C. FranziniArmstrong, eds.), Vol. 1, pp. 464-483. McGraw-Hill, New York. Burke, R. E., and Tsairis, P. (1973). ]. Physiol. 234, 749. Burke, R. E., Levine, D. N., Tsairis, P., and Zajac, F. E. (1973). J. Physiol. 234, 723. Cain, G. R, Cardinet, III, G. H., Cuddon, P.A., Gale, R P., and Champlin, R. (1986). Tmnsplmltation 41, 2l. Cannon, S. c., Hayward, 1. J., Beech, J., and Brown, R. H. (1995). J. NeufOphys. 73, 1892. Cao, A., De Virgilis, 5., Lippi, c., and Trabalza, N. (1969). Clin. Chim. Acta. 23, 475. Cardinet, G. H., III (1969). Unpublished observations. Cardinet, G. H., III, and Bandman, E. (1985). J. Vet. Med, Ser. C, Anat. Histol. Embrylogia. 14(1), 8l. Cardinet, G. H., III, and Holliday, T. A. (1979). Am!. N . Y. Acad. Sci. 317, 290. Cardinet, G. H., III, Fowler, M. E., and Tyler, W. S. (1963). AIIl . J. Vet. Res. 24, 980. Cardinet, G. H., III, Litterell, J. F., and Freedland, R. A. (1967). Res. Vet. Sci. 8, 219. Cardinet, G. H., III, Freedland, R A., Tyler, W. 5., and Julian, 1. M. (1972). Am. J. Vet. Res. 33, 1671. Cardinet, G. H., III, Leong, C. 1., and Means, P. S. (1982). Muscle Nerve 5, 665. Cardinet, G. H., III, Shelton, G. D., and Willis, S. E. (1986). J. Vet. Med, Ser. C, Allot. Histol. Embryologio. 15, 168. Carpenter, J. 1., Hoffman, E. P., Romanul, F. c., Kunkel, 1. M., Rosales, R K., Ma, N. 5., Dasbach, J. J., Rae, J. F., Moore, F. M., and McAfee, M. B. (1989). Am. J. Patlwl. 135, 909. Ceh, 1., Hauge, J. G., Svenkerud, R, and Strande, A. (1976). Acta Vet. Scnnd. 17, 210. Chalmers, G. A., and Barrett, M. W. (1977). J. Am. Vet. Med. Assoc. 171, 918. Citoler, P., Benitez, 1., and Maurer, W. (1966). Exp. Cell. Res. 45, 195. Close, R. (1965). Nature (London) 206, 83l. Close, R. 1. (1972). PlllJsiol. Rev. 52, 129. Colombo, J. P., Richterich, R., and Rossi, E. (1962). Klill. Woe/leuse/lr. 40, 37.

Cooper, B. J., Winand, N. J., Stedman, H., Valentine, B. A., Hoffman, E. P., Kunkel, 1. M., Scott, M. 0., Fischbeck, K. H., Kornegay, J. N., and Avery, R. J. (1988). Na ture 334, 154. Cooper, B. J., Gallagher, E. A., Smith, C. A., Valentine, B. A., and Winand, N. }. (1990). Lab. Invest. 62, 17l. Cornelius, C. E. (1960). Calif. Vet. 13, 22. Cornelius, C. E. (1967). Proc. Am. Anim. Hosp. Assoc. 34tlz Ann. Meet­ illg, p. 82. Cornelius, C. E., and Burnham, 1. G. (1963). ]. Am. Vet. Med. Assoc. 142, 639. Cornelius, C. E., and Kaneko, J. J. (1960). J. Am. Vet. Med. Assoc. 137, 62. Cornelius, C. E., Bishop, J., Switzer, J., and Rhode, E. A. (1959a). Cornell Vet. 49, 1 16. Cornelius, C. E., Law, G. R. J., Julian, 1. M., and Asmundson, V. S. (1959b). Prac. Soc. Exp. Bioi. Med. 101, 4l. Cosmos, E. (1966). Oev. Bioi. 13, 163. Cosmos, E., and Butier, J. (1967). III "Exploratory Concepts in Muscu· lar Dystrophy and Related Disorders" (A. T. Milhorat, ed.), pp. 197-204. Excerpto Med. Found., Amsterdam. Crabtree, B., and Newsholme, E. A. (1972). Biochem. J. 126, 49. Crawley, G. J., and Swenson, M. J. (1965). All!. J. Vet. Res. 26, 1468. Cutmore, C. M. M., Snow, D. H., and Newsholme, E. A. (1985). Equine Vet. J. 17, 354. Cutmore, C. M. M., Snow, D. H., and Newsholme, E. A. (1986) Equine Vet. J. 18, 72. Dawson, D. M., and Fine, J. H. (1967). Arch. Neural. 16, 175. Dawson, D. M., and Kaplan, N. O. (1965). J. Bioi. Chell!. 240, 3215. Dawson, D. M., and Romanul, F. C. A. (1964). Arch. Neural. 11, 369 Dawson, D. M., Goodfriend, T. 1., and Kaplan, N. O. (1964). Sciencl 143, 929. Dawson, D. M., Eppenberger, H. M., and Kaplan, N. O. (1965) Biochem. Biophys. Res. Commllll. 21, 346. Dawson, D. M., Eppenberger, H. M., and Kaplan, N. O. (1967) J. BioI. Chem. 242, 210. Denny-Brown, D. E. (1929). Prac. R. Soc. LOlldon, Ser. B 104, 371. DiMauro, S., and Tsujino, S. (1994). III "Myology" (A. G. Engel ane C. Franzini-Armstrong, eds.), Vol. 2, p. 1554-1576. McGraw-Hill New York. Domonkos, J. (1961). Arch. Biochelll. Biophys. 95, 138. Domonkos, J., and Latzkovits, 1. (1961a). Arch. Biochem. Biophys 95, 144. Domonkos, J., and Latzkovits, 1. (1961b). Arch. Biocllem. Biophys 95, 147. Dubowitz, V. (1967). J. Physiol. 193, 48l. Dubowitz, V. (1968). "Developing and Diseased Muscle." Spastici Int. Med. Publ., London. Dubowitz, V., and Pearse, A. G. E. (1960). Histochemic 2, 105. Dubowitz, V., Brooke, M. H., and Neville, H. E. (1973). "Musch Biopsy: A Modern Approach." Saunders, Philadelphia, Pennsyl vania. Duncan, 1. D., and Griffiths, 1. R. (1977). Vet. Rec. 100, 30. Duthie, G. G., and Arthur, J. R (1987). Am. J. Vet. Res. 48, 309. Duthie, G. G., Arthur, J. R, Simpson, S. P., and Nicol, F. (1988). Am J. Vet. Res. 49, 508. Ebashi, 5., Endo, M., and Ohtsuki, 1. (1969). Quart. Rev. Biophys. 2, 351 Eccles, J. c., Eccles, R. M., and Lundberg, A. (1958). ]. Physiol. 142, 275 Edleman, R (1995). Clin. Lab. News 21, l. Edstrom, 1., and Kugelberg, E. (1968). J. Neurol. Neurosurg. Psych iat. 31, 424. Edwards, R. H. T., Melcher, A., Hesser, C. M., and Wigertz, O. (1971) Adv. Exp. Med. BioI. 11, 425. Eisenberg, E., and Kielley, W. W. (1974). J. Bioi. Chem. 249, 4742.

Skeletal Muscle Function Engel, A. G. (1994a). In "Myology" (A. G. Engel and C. Franzini­ Armstrong, eds.), Vol. 1, p. 261. McGraw-Hill, New York. Engel, A. G. (1994b). In "Myology" (A. G. Engel and C. Franzini­ Armstrong, eds.), Vol. 1, p. 822. McGraw-Hill, New York. Engel, A. G. (1994c). In "Myology" (A. G. Engel and C. Franzini­ Armstrong, eds.), Vol. 2, p. 1769. McGraw-Hill, New York. Engel, A. G. (1994d). In "Myology" (A. G. Engel and C. Franzini­ Armstrong, eds.), Vol. 2, p. 1798. McGraw-Hill, New York. Engel, A. G., and Hirschhorn, R (1994). In "Myology" (A. G. Engel and C. Franzini-Armstrong, eds.), Vol. 2, p. 1533. McGraw-Hill, New York. Engel, A. G., Lambert, E. H., and Howard, F. M. (1977). Mayo Clin. Froc. 52, 267. Engel, A. G., Yamamoto, M., and Fischbeck, K H (1994). In "Myol­ ogy" (A. G. Engel, and C. Franzini-Armstrong, eds.), Vol. 2, p. 1 133. McGraw-Hill, New York. Engel, W. K (1962). Neltrology 12, 778. Engel, W. K (1965). III "Neurohistochemistry" (c. W. M. Adams, ed.), p. 622. Elsevier, Amsterdam. Engel, W. K (1967). Pedial. Clin. North Am. 14, 963 . Engel, W. K (1970). Arch. Neurol. 22, 97. Eppenberger, H. M., von Fellenberg, R, Richterich, R, and Aebi, H. (1962). Ellzymol. BioI. Clin. 2, 139. Eppenberger, H. M., Eppenberger, M., Richterich, R, and Aebi, H. (1964a). Oev. Bioi. 10, 1 . Eppenberger, M., Nixon, C . W., Baker, J . R , and Homburger, F. (1964b). Proc. Soc. Exp. BioI. Med. 117, 465. Essen-Gustavsson, B., and Lindholm, A. (1985). Equine Vet. f. 17, 434. Essen-Gustavsson, B., Karlstrom, K, and Lindholm, A. (1984). Equine Vet. f. 16, 197. Felig, P. (1977). Ann. N. Y. Acad. Sci. 301, 57. Felig, P., and Wahren, J. (1971). Adv. Exp. Med. BioI. 11, 205. Fiolet, J. W. T., Willebrands, A. F., Lie, K 1., and Ter Welle, H. F. (1977). Clin. Chim. Acta 80, 23 . Franzmann, A. W., and Thorne, E. T. (1970). ]. Am. Vet. Med. Assoc. 157, 647. Froberg, S. 0., Carlson, L. A., and Ekelund, L. G. (1971). Adv. Exp. Med. Bioi. 11, 307 . Fujii, J., Otsu, K, Zorzato, F., de Leon, S., Khanna, V. K, Weiler, J. E., O'Brien, P. J., and MacLennan, D. H. (1991). Science 253, 448. Fujii, Y., Ikeda, S., and Watanabe, H. (1980). Bull. Equine Res. Insl. 17, 21. Fyfe, J. c., Giger, U., Van Winkle, T. J" Haskins, M. E., Steinberg, S. A., Wang, P., and Patterson, D. F. (1992). Ped. Res. 32, 719. Gaschen, F. P., Hoffman, E. P., Gorospe, J. R, Uhl, E. W., Senior, D. F., Cardinet, G. H., and Pearce, L. K (1992). f. Neural. Sci. 110, 149.

Gauthier, G. F. (1969). Z. Zelljorsch. Mikrosk. Anat. 95, 462. Gauthier, G. F., and Padykula, H. A. (1966). f. Cell. BioI. 28, 333. Gazith, J., Himmelfarb, S., and Harrington, W. F. (1970). f. BioI. Chem. 245, 15. George, J. c., and Bokdawala, F. D. (1964). f. Anim. Morphol. Physiol.

George, J. c., and Iype, P. T. (1964). Ind. ]. Omitlwl. 2, 84. George, J. c., and Talesara, C. L. (1961). f. Cell. Camp. Physiol. 58, 253. Gerber, H. (1964). Zentralbl. Veterillaermed. 11, 135 . Gershman, L. c., Stracher, A., and Dreizen, P. (1969). f. BioI. Chelll. 244, 2726. Giger, U., Harvey, J. W., Yamaguchi, R A., MeNutty, P. K, Chiapella, A., and Buetler, E. (1985). Blood 65, 345. Giger, U., Reilly, M. P., Asakura, T., Baldwin, C. J., and Harvey, J. W. (1986). Anim. Gelletics 17, 15. Giger, U., Argov, Z., Schnall, M., Bank, W. J., and Chance, B. (1988a). Muscle Nerve 11, 1260. 11, 124.

437

Giger, U. Kelly, A. M., and Teno, P. S. (1988b). Enzy111e 40, 25. Glock, G. E., and McLean, P. (1954). Bioclzem. ]. 56, 171. Goldberg, A. L. (1967). Nailire (London) 21 6, 1219. Gollnick, P. D., and King, D. W. (1969). Am. f. Physiol. 21 6, 1502. Gollnick, P. D., Armstrong, R B., Saubert, C. W., Piehl, K, and Saltin, B. (1972). f. Appl. Physiol. 33, 312. Gorospe, J. R, Tharp, M. D., Hinckley, J., Kornegay, J. N., and Hoffman, E. P. (1994). f. Neural. Sci. 122, 44. Goto, 1., Nagamine, M., and Katsuki, S. (1969). Arch. Neural. 20, 422. Granit, R, Phillips, C. G., Skoglund, S., and Steg, G. (1957). f. NellrD­ physiol. 20, 470. Green, D. E. (1951). In "Enzymes and enzyme systems" ( J. T. Edsall, ed.), p. 15. Harvard Univ. Press, Cambridge, Massachusetts. Green, H. J., Klug, G. A., Reichmann, H, Seedorf, U., Wiehrer, W., and Pette, D. (1984). Pflugers Arch. 400, 432. Green, M. R, and Landau, B. R (1965). Arch. Bioclzem. Biophys. 111, 569.

Griffiths, 1. R, and Duncan, 1. D. (1973). Vel. Rec. 93 , 184. Gronert, G. A. (1994). In "Myology" (A. G. Engel and C. FranziniArmstrong, eds.), Vol. 2, p. 1661. McGraw-Hill, New York. Gunn, H. M. (1978). f. Anal. 127, 615. Guth, L. (1968). Patlwl. Rev. 48, 645. Gutmann, E. (1962). "The Denervated Muscle." Publ. House Czech. Acad. Sci., Prague. Guy, P. S., and Snow, D. H (1977). Bioclzem. Biophys. Res. COlllmZl11. 75, 863.

Guy, P. S., and Snow, D. H (1981). Res . Vet. Sci. 31, 244. Hagenfeldt, L., and Wahren, J. (1971). Adv. Exp. Med. BioI. 11, 153. Haigh, J. c., Stewart, R R, Wobeser, G., and MacWilliams, P. S. (1977). ]. Am. Vet. Med. Assoc. 171, 924. Hall-Craggs, E. C. B. (1968). f. Anat. 102, 241. Hamada, T. (1974). Vet. Record 95, 441. Hamilton, M. J., Hegreberg, G. A., and Gorham, J. R (1974). Am. f. Vet. Res. 35, 1321. Harper, J. A., and Parker, J. E. (1964). Poult. Sci. 43, 1326. Harper, J. A., and Parker, J. E. (1967). f. Hered. 58, 189. Harper, P. S., and Rudel, R (1994). In "Myology" (A. G. Engel and C. Franzini--Armstrong, eds.), Vol. 2, p. 1192. McGraw-Hill, New York. Harris, J. B. (1979). Ann. N.Y. Acad. Sci. 317, 1. Harvey, J. W., Calderwood Mays, M. B., Gropp, K E., and Denaro, F. J. (1990). Vet. Patlwl. 27, 1. Havel, R J. (1971). Adv. Exp. Med. Bioi. 11, 315. Heffron, J. J. A., Bomzon, L., and Pattinson, R A. (1976). Vet. Record 98, 338.

Hegreberg, G. A., Camacho, Z., and Gorham, J. R (1974). Arch. Pat/wi. 97, 225. Hennig, R, and Lomo, T. (1987). Acta Physiol. Scand. 130, 1 15. Henson, J. B., Dollahite, J. W., Bridges, C. H., and Rao, R. R (1965). f. Am. Vet. Med. Assoc. 147, 142. Hermansen, L. (1971). Adv. Exp. Med. BioI. 11, 401. Hess, J. W., Murdock, K J., and Natho, G. J. W. (1968). Am. f. Clin. Pathol. 50, 89. Hess, R., and Pearse, A. G. E. (1961). Proc. Soc. Exp. BioI. Med. 107, 569. Hibbs, C. M., and Coles, E. H. (1965). Proc. Soc. Exp. Bioi. Med. 118, 1059. Hildebrand, S. V., and Howitt, G. A. (1983). Am. f. Vet. Res. 44, 2280. Hodgson, D. R., Rose, R J., Allen, J. R, and DiMauro, J. (1984a). Comell Vet. 75, 366.

Hodgson, D. R, Rose, R J., Allen, J. R, and DiMauro, J. (1984b). Res. Vel. Sci. 3 6, 169. Hodgson, D. R, Rose, R J., DiMauro, J., and Allen, J. R (1985). Equine Vet. f. 17, 300. Hoffman, E. P., Brown, R H, and Kunkel, L. M. (1987). Cell 5 1, 919.

438

George H. Cardinet III

Holliday, T. A, Asmundson, V. S., and Julian, 1. M. (1965). Enzymol. Bioi. Clin. 5, 209. Holloszy, J. O. (1967). J. Bioi. Chell!. 242, 2278. Holloszy, J. O. (1982). Arch. Phys. Med. Rehabil. 63, 231 . Holloszy, J. 0., and Coyle, E. F. (1984). J. Appl. Physiol: Respirat. Environ. Exer. Physiol. 56, 831. Holloszy, J. 0., Oscai, 1. B., Mole, P. A., and Don, 1. J. (1971). Adv. Exp. Med. 11, 51. Holloszy, J. 0., Rennie, M. J., Hickson, R c., Conlee, R K, and Hagberg, J.M. (1977). Ann. N.Y. Acad. Sci. 301, 440. Holloszy, J. 0., Dalsky, G. P., Nemeth, P. M., Hurley, B. F., Martin, W. H. III, and Hagberg, J. M. (1986). III "International Series on Sport Sciences" (B. Saltin, ed.), Vol. 16, p. 183. Human Kinetics, Champaign, Illinois. Holmes, H. G., Ashmore, C. R, Robinson, D. W., Finn, J. P., and O'Dell, J. (19 72 ) . Vet. Record 90, 625 . Holmgren, N., and Valberg, S. (1992). Am. J. Vet. Res. 53, 957. Homburger, F., Baker, J. R, Nixon, C. W., and Wilgram, G. (1962). Arch. Intern. Med. 110, 660. Horowitz, P. (1994). Excitation-contraction coupling in skeletal mus­ cle. In "Myology" (A G. Engel and C. Franzini-Armstrong, eds.), Vol. 1, p. 423. McGraw-Hill, New York. Houde, A, Pommier, S. A., and Roy, R (1993). J. Anim. Sci. 71, 1414. Howell, J. M. c., Dorling, P. R, Cook, R D., Robinson, W. F., Bradley, S., and Gawthorne, J. M. (1981). J. Pathol. 134, 266. Hultman, E., and Nilsson, 1. H. (1971). Adv. Exp. Med. Bioi. 11, 143. Huxley, H. E. (1983). In "Muscle and Nonmuscle Motility" (A. Stracher, ed.), Vol. 1, p. 1. Academic Press, New York. Ianuzzo, C. D., Patel, P., Chen, V., and O'Brien, P. (1980). In "Plastic­ ity of Muscle" (D. Pette, ed.), p. 593. Walter de Gruyter, Berlin. Indrieri, R J., Creighton, S. R, Lambert, E. H., and Lennon, V. A (1983). JAVMA 182, 57. Innes, J. R. M. (1951). Br. Vet. J. 107, 131. Issekutz, 8., Jr., Miller, H. I., Paul, P., and Rodahl, K (1965). J. Appl.

Physiol. 20, 293.

Issekutz, 8., Jr., Miller, H. I., and Rodahl, K (1966). Fed. Proc., Fed. Am. Soc. Exp. BioI. 25, 1415. Issekutz, 8., Jr., Shaw, W. A S., and Issekutz, T. B. (1975). J. Appl. Physiol. 39, 349. Jamison, J. M., Baird, J. D., Smith-Maxie, 1. 1., and Hulland, T. J. (1987). Equine Vet. J. 19, 353. Johnson, B. D., and Perce, R B. (1981). Equine Pract. 3, 4. Johnson, M. A., Mastaglia, F. 1., Montgomery, A, Pope, B., and Weeds, A. G. (1980). In "Plasticity of Muscle" (D. Pette, ed.), p. 607. Walter de Gruyter, Berlin. Jolly, R D., Van-De-Water, N. S., Richards, R B., and Dorling, P. R (1977). Aust. J. Exp. Bioi. Med. Sci. 55, 141. Jorfeldt, 1. (1971). Adv. Exp. Med. BioI. 11, 407. Jorgensen, P. F., and Hyldgaard-Jensen, J. F. (1975). Acta Vet. Scand. 16, 368. Kaneko, J. J. (1987). Clill. Chim. Acta 165, 1. Kaminski, H. J., and Ruff, R 1. (1994). In "Myology" (A. G. Engel and C. Franzini-Armstrong, eds.), Vol. 2, p. 1726. McGraw-Hill, New York. Karlsson, J. (1971). Adv. Exp. Med. BioI. 11, 383. Klein, M. S., Shell, W. E., and Sobel, 8. E. (1973). Cardiovasc. Res. 7, 412. Koch, M. c., Steinmeyer, K, Lorenz, c., Ricker, K, Wolf, F., Otto, M., Zoll, 8., Lehmann-Horn, F., Grzeschik, K H., and Jentsch, T. J. (1992). Sciellce 257, 797. Kornegay, J. N., Tuler, S. M., Miller, D. M., and Levesque, D. C. (1988). Muscle Nerve 11, 1056. Kortz, G. (1989). Semin. Vet. Med. and SlIrg. (Small Animal ) 4, 141. Kowalski, K, Gordon, E. E., Marinez, A, and Adamek, J. (1969). J. Histochem. Cytochem. 17, 601.

Kubista, V., Kubistova, J., and Pette, D. (1971). Eur. J. Biochem. 18, 553. Kuttler, K 1., and Marble, D. W. (1958). Am. J. Vet. Res. 19, 632. Lagace, A, Bell, D. S., Moxon, A. 1., and Pounden, W. D. (1961). Am. J. Vet. Res. 22, 686. Lamb, D. R., Peter, J. B., Jeffress, R. N., and Wallace, H. A. (1969). Am. J. Physiol. 217, 1628. Larach, M. G. (1989). Anesth. Analg. 69, 511. Lawrie, R A. (1952). Nature (London) 170, 122. Lawrie, R. A (1953). Biochem. J. 55, 298. Lee, F. 5., Guenther, A E., and Meleney, H. E. (1916). Am. J. Physiol.

40, 446.

Lehmann-Hom, F., Engel, A. G., Ricker, K, and Rudel, R (1994). In "Myology" (A G. Engel and C. Franzini-Armstrong, eds.), Vol. 2, p. 1303. McGraw-Hill, New York. Lehmann-Horn, F., Sipos, I. Jurkat-Rott, K., Heine, R., Brinkmeier, H., Fontaine, B., Kovacs, 1., and Melzer, W. (1995). Soc. Gell. Physiol. Ser. 50, 101. Lennon, V. A., Palmer, A. c., Pfiugfelder, c., and Indrieri, R J. (1978). III "Genetic Control of Autoimmune Disease" (N. R Rose, P. E. Bigazzi, and N. 1. Warner, eds.), Vol. 1. p. 295. Elsevier North Holland, New York. Lennon, V. A., Lambert, E. H., Palmer, A. c., Cunningham, J. G., and Christie, T. R (1981). III "Myasthenia Gravis: Pathogenesis and Treatment" (E. Satoyoshi, ed.). p. 41. Univ. of Tokyo Press, Tokyo. Lewis, J. H. (1976). Am. J. Vet. Res. 37, 601. Lindholm, A, and Saltin, B. (1974). Acta Vet. Scalld. 15, 310. Lindholm, A., Bjerneld, H., and Sa; in, n l ; //4a). Acta Physiol. Scand.

90, 475.

Lindholm, A., Johansson, H. E., and Kjaersgaard, P. (1974b). Acta

Vet. Scand. 15, 325. Lindstrom, J. (1994). In "Myology" (A. G. Engel and C. Franzini­ Armstrong, eds.), Vol. 1, p. 585. McGraw-Hill, New York. Lindstrom, J., Einarson, 8. and Tzartos, S. (1981). Methods Ellzymol.

74, 432.

Lomo, T. (1986). In "International Series on Sport Sciences: Biochem­ istry of Exercise VI" (B. Saltin, ed.), Vol. 16, p. 27. Human Kinetics, Champaign, Illinois. Lomo, T., Westgaard, R. H., and Dahl, H. A. (1974). Proc. R. Soc. London Ser. B 187, 99. Lowenstein, J. M. (1972). Physiol. Rev. 52, 382. Lowey, S. (1994). In "Myology" (A. G. Engel and C. Franzini-Annstrong,

eds.), Vol. 1, p. 485. McGraw-Hill, New York. Lowey, S., and Risby, D. (1971). Nature (LoI1don) 234, 81.

Lowey, S., Benfield, P. A., Silberstein, 1., . � ang, 1. M. (1979). Nature (London) 282, 522. Luginbuhl, A. J., Dudley, G. A, and Stan;, · ' 3. (1984). Histochemistry 81, 55. Lymn, R W., and Taylor, E. W. (1971). Bio( hemistry 10, 4617. McGavin, M. D., and Baynes, 1. D. (1969). Path. Vet. 6, 513. MacLennan, D. H., and Phillips, M. S. (1992). Science 256, 789. Mader, R., Nicol, P. D., Turley, J. J., Billbao, J., and Keystone, E. C. (1994). J. Med. Sci. 30, 902. Magleby, K 1. (1994). III "Myology" (A G. Engel and C. FranziniArmstrong, eds.), Vol. 1, p.442. McGraw-Hill, New York. Manktelow, B. W., and Hartley, W. J. (1975). J. Compo Pathol. 85, 139. Manley, S. V., Kelly, A. B., and Hodgson, D. (1983). JAVMA 183, 85 Maruyama, K (1986). Int. Rev. Cytol. 104, 81. Mastofa, I. E. (1970). Acta Vet. Scand. 11, 197. Maxwell, 1. c., Carlson, D. 5., and Brangwin, C. E. (1980). Histochem.

J. 12, 209.

Michelson, A M., Russell, E. S., and Harman, P. J. (1955). Proc. Nat/ Acad. Sci. USA 41, 1079.

Skeletal Muscle Function Miller, L. M., Lennon, V. A., Lambert, E. H., Reed, S. M., Hegreberg, G. A., Miller, J. B., and Ott, R 1. (1983). JAVMA 182, 694. Moise, N. S., Valentine, B. A., Brown, C. A., Erb, H. N., Beck, K. A., Cooper, B. J., and Gilmour, R F. (1991). J. Am. College Cm·diol. 17, 812. Mole, P., Os cal, 1. B., and Holloszy, J. O. (1971).J. Clin. Invest. 50, 2323. Mommaerts, W. F. H. M. (1969). Physiol. Rev. 49, 427. Morgan, T. E., Cobb, 1. A., Short, F. A., Ross, R, and Gunn, D. R (1972). Adv. Exp. Med. Bioi. 11, 87. Morgan-Hughes, J. A. (1994). In "Myology" (A. G. Engel and C. Franzini-Armstrong, eds.), Vol. 2, p. 1610. McGraw-Hill, New York. Morizot, D. c., Bednarz, J. c., and Ferrell, R E. (1987). Cytogenet. Cell Genet. 44, 89. Murakami, H., Takagi, A., Nonaka, S., Ishiura, S. Sugita, H., and Mizutani, M. (1980). Exp. Allim. (Tokyo) 29, 475. Nachmias, V. T., and Padykula, H. A. (1958). J. Biophysic. Biochem. Cytol. 4, 47. Nagode, L. A., Frajola, W. J., and Loeb, W. F. (1966). Am. J. Vet. Res. 27, 1385. Needham, D. M. (1926). Physiol. Rev. 6, 1. Nelson, T. E. (1991). JA VMA 198, 989. Nevins, M. A., Saran, M., Bright, M., and Lyon, 1. J. (1973). J. Am. Vet. Med. Assoc. 224, 1382. Nicol, P. D., Matsueda, G. R, Haber, E., and Khaw, B. A. (1993). J. Nucl. Med. 34, 2144. O'Brien, P. J., Pook, H. A., Klip, A., Britt, B. A., Kalow, 8. I., McLaughlin, R N., Scott, E., and Elliott, M. E. (1990). Res. Vet. Sci. 48, 124. O'Brien, P. J., Shen, H., Cory, C. R, and Zhang, X. (1993). JAVMA 203, 842. Oda, K., Lambert, E. H., Lennon, V. A., and Palmer, A. C. (1984a). Mliscle Nerve 7, 705. Oda, K., Lennon, V. A., Lambert, E. H., and Palmer, A. C. (1984b). Muscle Nerve 7, 717. Ogata, T. (1960). J. Biochem. 47, 726. Okinaka, S., Sugita, H., Momoi, H., Toyokura, Y., Watanabe, T., Ebashi, F., and Ebashi, S. (1964). J. Lab. Clin. Med. 64, 299. Oliver, I. T. (1955). Bioelzem. J. 61, 1 16. Orvis, J. A., and Cardinet, G. H., III (1981). Mliscle Nerve 4, 354. Oscai, L. B., and Holloszy, J. O. (1971). J. BioI. Chem. 246, 6968. O'Sullivan, B. M., Healy, P. J., Fraser, I. R, Nieper, R E., Whittle, R. J., and Sewell, C. A. (1981). Aust. Vet. J. 57, 227. Otani, T., and Mochizuki, H. (1977). Exp. Anim. (Tokyo) 26, 172. Padykula, H. A. (1952). Am. J. AHllt. 91, 1 07. Padykula, H. A., and Gauthier, G. F. (1963). J. Cell. Bioi. 18, 87. Padykula, H. A., and Gauthier, G. F. (1967). In "Exploratory Concepts in Muscular Dystrophy and Related Disorders" (A. T. Milhorat, ed.), p. 1 1 7. Excerpta Med. Found., Amsterdam. Pande, S. V., and Blanchaer, M. C. (1971). Am. J. Physiol. 220, 549. Pardo, J. V., Siciliano, J. D. F., and Craig, S. W. (1983). PNAS 80, 1 008 . Paul, P. (1971). Adv. Exp. Med. Bioi. 11, 225. Paul, P., and Issekutz, B. (1967). J. Appl. Physiol. 22, 615. Penn, A. S. (1986). In "Myology" (A. G. Engel and B. Q. Banker, eds.), Vol. 2, p. 1 785. McGraw-Hill, New York. Peter, J. B., Jeffress, R N., and Lamb, D. R (1968). Science 160, 200. Peter, J. B., Barnard, R J., Edgerton, V. R, Gillespie, C. A., and Stempel, K. E. (1972). Biochemistry 11, 2627. Pette, D. (1975). Acta Histoellem. 14, 47. Pette, D., and Vrbova, G. (1985). Muscle Nerve 8, 676. Pette, D., Smith, M. E., Staudte, H. W., and Vrbova, G. (1973). Pflugers Arch. 338, 257. PPflugfelder, C. M., Cardinet, G. H., III, Lutz, H., Holliday, T. A., and Hansen, R. J. (1981). Muscle Nerve 4, 289.

439

Pierbon-Bormoli, S., Betto, R, and Salviati, G. (1989). J. Muscle Res. Cell Moti/. 10, 446. Rafiquzzaman, M., Svenkerud, R, Strande, A., and Hauge, J. G. (1976). Acta Vet. Scand. 17, 196. Rand, J. S., and O'Brien, P. J. (1987). JAVMA 190, 1013. Reed, S. M., Hegreberg, G. A., Bayly, W. M., Brown, C. M., Paradis, M. R, and Clemmons, R M. (1988). Muscle Nerve 11, 291. Richards, R 8., Edwards, J. R, Cook, R D., and White, R R (1977). Neuropathol. App/. Neurabiol. 3, 45. Robbins, N., Karpati, G, and Engel, W. K. (1969). Arch. Neural. 20, 318. Romanul, F. C. A. (1964). Arch. Neural. 11, 355. Romanul, F. C. A., and Van Der Meulen, J. P. (1966). Nature (London) 212, 1369. Romanul, F. c., and Van Der Meulen, J. P. (1967). Arch. Neurol. 17, 387. Rose, R J., Purdue, R. A., and Hensley, W. (1977). Equine Vet. J. 9, 122. Roussel, J. D., and Stallcup, o. T. (1966). Am. J. Vet. Res. 27, 1527. Rowell, 1. B. (1971). Adv. Exp. Med. Bioi. 11, 127. Rowland, L. P. , DiMauro, S., and Layzer, R B. (1986). In "Myology" (A. G. Engel and B. Q. Banker, eds.), Vol. 2, p. 1603. McGraw­ Hill, New York. Rowlerson, A., Pope, B., Murray, J., Whalen, R B., and Weeds, A. G. (1981). J. Muscle Res. Cell Moti/. 2, 415. Rowlerson, A., Mascarello, F., and Veggetti, A. (1983). J. Muscle Res. Cell Moti/. 4, 443. Rudel, R, Lehmann-Horn, F., and Ricker, K. (1994). In "Myology" (A. G. Engel and C. Franzini-Armstrong, eds.), Vol. 2, p. 1291. McGraw-Hill, New York. Rudolph, J. A., Spier, S. J., Byrns, and Hoffman, E. P. (1992a). Anim. Genet. 23, 241 . Rudolph, J. A., Spier, S. J., Byrns, G., Rojas, C. V., Bernoco, D., and Hoffman, E. P. (1992b). Nature Genet. 2, 144. Ruth, G. R, and Van Vleet, J. F. (1974). Am. J. Vet. Res. 35, 237. Sahlin, K. (1986). In "International Series on Sports Sciences: Bio­ chemistry of Exercise VI" (B. Saltin, ed.), Vol. 16, p. 323 . Human Killetics Publishers, Champaign, Illinois. Salmons, S., and Vrbova, G. (1969). J. Physio/' 201, 535. Saltin, B., and Essen, B. (1971). Adv. Exp. Med. Bio/. 11, 419. Saltin, B., and Karlsson, J. (1971). Adv. Exp. Med. BioI. 11, 289. Saltin, B., Henriksson, J., Nygaard, E., Andersen, P., and Jansson, E. (1977). Ann. N.Y. Acad. Sci. 301, 3. Sandstrom, B. Westman, J., and Ockerman, P. A. (1969). Acta. Neuro­ pat/wi. 14, 194. Sarkar, S., Sreter, F. A., and Gergely, J. (1971). Proc. Natl. Acad. Sci. USA 68, 946. Seidel, J. C. (1967). J. Bioi. Chell!. 242, 5623. Sharp, N. J., Kornegay, J. N., Van Camp, S. D., Herbstreith, M. H., Secore, S. 1., Kettle, S., Hung, W. Y., Constantinou, C. D., Dykstra, M. J., and Roses, A. D. (1992). Genomics 13, 115. Shelton, G D., and Cardinet, G. H., III (1987). J. Vet. Intel'll. Med. 1, 36. Shelton, G D., Bandman, E., and Cardinet, G. H., III (1985a). Am. J. Vet. Res. 46, 493. Shelton, G. D., Cardinet, G. H., III, and Bandman, E. (1985b). Muscle Nerve 8, 783. Shelton, G. D., Cardinet, G. H., III, and Bandman, E. (1987). Mllscle Nerve 10, 753. Shelton, G D., Cardinet, G. H., III, and Lindstrom, J. M. (1988). NeurologJJ 38, 1417. Shelton, G. D., Willard, M. D., Cardinet, III, G. H., and Lindstrom, J. (1990). J. Vet. Intel'll. Med. 4, 281. Sherwin, A. 1., Siber, G. R, and Elhilali, M. M. (1967). Clin. Chim. Acta. 17, 245. Sherwin, A. 1., Norris, J. W., and Bulcke, J. A. (1969). NeuralogJJ 19, 993.

440

George H. Cardinet III

Simoneau, J. A., Lortie, G., Boulay, M. R, Thibault, M. c., and Bouchard, C. (1985). Eur. J. App/. Physiol. 54, 250. Sjodin, B. (1976). Acta Physiol. Scand. 436, 1 . Sjodin, B., Thorstensson, A., Frith, K , and Karlsson, J . (1976) . Acta Physiol. Scalld. 97, 150. Smith, J. B., and Healy, P. L. (1968) . Clill. C/zim. Acta. 21, 295. Snow, D. H., and Guy, P. S. (1980). Res. Vet. Sci. 28, 137. Snow, D. H., and Mackenzie, G. (1977). Equille Vet. J. 9, 134. Snow, D. H., Billeter, R, and Jenny, E. (1981). Res. Vet. Sci. 30, 381. Snow, D. H., Billeter, R, Mascarello, F., Carpene, E., Rowlerson, A., and Jenny, E. (1982a). Histoelzemistry. 75, 53. Snow, D. H., Kerr, M. G., Nimmo, M. A., and Abbott, E. M. (1982b). Vet. Record. 110, 377. Somer, H., Dubowitz, V., and Donner, M. (1976). J. Neural. Sci. 29, 129. Spier, S. J., Carlson, G. P., Holliday, T. A., Cardinet, III, G. H., and Pickar, J. G. (1990). JAVMA 197, 1009. Spier, S. J., Carlson, G. P., Harrold, D., Bowling, A., Byrns, G., and Bernoco, D. (1993). JAVMA 202, 933. Sreter, F. A., Gergely, J., Salmons, S., and Romanul, F. (1973). Nature (LOlldoll), New BioI. 241, 17. Sreter, F. A., Luff, A. R, and Gergely, J. (1975). J. Gell. Physiol. 66, 81 1 . Staron, R S., and Pette, D. (1986) . Histochemistry 86, 1 9. Staudte, H. W., and Pette, D. (1972). Camp. Bioelzem. Physiol. 31, 533. Stein, J. M., and Padykula, H. A. (1962). Am. J. Allat. 110, 103. Steinberg, S., and Botelho, S. (1 962) . Sciellce 137, 979. Steiness, E., Rasmussen, F., Svendsen, 0., and Nielsen, P. (1978). Acta Pharmacal. Toxicol. 42, 357. Stubbs, S. S., and Blanchaer, M. C. (1965). Call. J. Bioelzem. 43, 463. Sund, H. (1969). III "Biological Oxidations" (T. P. Singer, ed.), p. 603. Wiley (Interscience), New York. Suominen, H., and Hiekkinen, E. (1975). Eur. J. Appl. Plzysiol. 34, 249. Swingle, K F., Young, S., and Dang, H. C. (1959). Am. J. Vet. Res. 20, 75. Tanzer, M. L., and Gilvarg, C. (1959). J. BioI. Cizem. 234, 3201. Tarrant, P. V., and McVeigh, J. M. (1979). Res. Vet. Sci. 27, 325. Terjung, R L., Baldwin, K M., Mole, P. A., Klinkerfuss, G. H., and Holloszy, J. O. (1 972). Am. J. Physiol. 223, 549. Terjung, R L., Dudley, G. A., Meyer, R A., Hood, D. A., and Gorski, J. (1986). III "International Series on Sport Sciences: Biochemistry of Exercise VI" (B. Saltin, ed.), Vol. 16, p. 131. Human Kinetics, Champaign, Illinois. Tesch, P., Sjodin, B., and Karlsson, J. (1978). Acta Physiol. Scalld. 103, 40. Therriault, D. G., Beller, G. A., Smoake, J. A., and Hartley, L. H. (1973). J. Lipid Res. 14, 54. Thorstensson, A., Elwin, K, Sjodin, B., and Karlsson, J. (1976) . Scalld. J. Clin. Lab. 36, 821.

Tokuyasu, K T., Dutton, A. H., and Singer, S. J. (1983). J. Mol. BioI. 96, 1727. Tzartos, 5., Seybold, M., and Lindstrom, J. M. (1982). Pmc. Natl. Acad. Sci. USA 79, 188. Valberg, S. (1986). Equille Vet. J. 18, 479 . Valberg, S. J., Cardinet, III, G. H., Carlson, G. P., and DiMauro, S. (1992). NeufOmllsc. Disord. 2, 351. Valberg, S., Jonsson, L., Lindholm, A., and Holmgren, N. (1993). Eqllille Vet. ,. 25, 1 1 . Valberg, S. J., Carlson, G. P., Cardinet, III, G. H., Birks, E. K , Jones, J. H., Chomyn, A., and DiMauro, S. (1994). Muscle Nerve 17, 305. Van Der Veen, K J., and Willebrands, A. F. (1966). Clill. C/lim. Acta. 13, 312. Van Vleet, J. F., Crawley, R. R., and Amstutz, H. E. (1977). J. Am. Vet. Med. Assoc. 171, 443. Vijayasarathy, c., Giger, U., Procuiuk, U., Patterson, D. F., Breitschw­ erdt, E. B., and Avadhani, N. G. (1994). Camp. Bioc!zem. Physiol. A Physiol 109, 887. Vora, S., Giger, U., Turchen, 5., and Harvey, J. W. (1 985). Proc. Nat/. Acad. Sci. USA 82, 8109. Vrbova, G. (1963). J. Plzysiol. 169, 513. Wachstein, M., and Meisel, E. (1955). J. Biophys. Bioellem. Cytol. 1, 483. Wagner, P. D., and Weeds, A. G. (1977). J. Mol. BioI. 109, 455. Wagner, W. D., Peterson, R A., and Anido, V. (1971). Am. J. Vet. Res. 32, 2091. Wahren, J. (1 977). Ann. N.Y. Acad. Sci. 301, 45. Wahren, J., Ahlborg, G., Felig, P., and Jorfeldt, L. (1971). Adv. Exp. Med. BioI. 11, 189. Walvoort, H. C. (1983). J. Illlzer. Metab. Dis. 6, 3. Walvoort, H. c., Van der Ingh, T. S. G. M., and Van Nes, J. J. (1981). Ber/. MUllch. Tiemrztl. Wochwscllr. 94, 39. Walvoort, H. c., Slee, R. G., and Koster, J. F. (1982). Bioellilll. Biophys. Acta 715, 63. Wang, K, and Ramirez-Mitchell, R. (1983). J. Cell Bioi. 96, 562. Weeds, A. G. (1969). Nature (LOlldo1l) 223, 1362. Weeds, A. G., and Lowey, S. (1971). J. Mol. BioI. 61, 701. Weeds, A. G., Trentham, D. R, Kean, C. J. c., and Buller, A. J. (1974). Natllre (Londoll) 247, 135 . Wentink, G. H., Hartman, W., and Koeman, J. P. (1974). Tijdsclzr. Diergelleesk. 99, 729. Whanger, P. D., Weswig, P. H., Muth, O. H., and Oldfield, J. E. (1970). Am. J. Vet. Res. 31, 965. Wiesmann, U., Colombo, J. P., Adam, A., and Richterich, R (1966). E1lzym. BioI. Clill. 7, 266. William, P. E., and Goldspink, G. (1971). J. Cell Sci. 9, 75I. Zhou, L Spier, S. J., Beech, J., and Hoffman, E. P. (1994). Hllman Mol. Genet. 3, 1599.

C H A P T E R

17 Kidney Function DELMAR R. FINCO

C. D. E. F, G,

Clearance Methods 476 Quantitative Renal Scintigraphy 477 Single-Injection Techniques 478 Urinary Enzymes 479 Single-Sample ("Spot") Fractional Excretion Determinations 479 H. A Perspective of Renal Function Tests and Their Use 479 References 480

I. INTRODUCTION 441 II. NORMAL RENAL FUNCTIONS 442 A. The Nephron as a Structural and Functional Unit 442 B. Nephron Heterogeneity 443 C. Renal Perfusion 443 D. Glomerular Filtration 444 E. Tubular Modification of Filtrate 445 F. Conservation of Nutrients 446 G. Electrolyte Homeostasis 446 H. Acid-Base Balance 448 I. Water Homeostasis 449 J. Renal Maturation and Senescence 451 K. Renal Metabolism 452 III. ALTERATIONS IN RENAL FUNCTION DUE TO EXTRARENAL FACTORS 453 A. Diet and Protein Catabolism 453 B. Renal Perfusion 453 C. Renal Outflow Impairment 454 IV. PRIMARY RENAL DYSFUNCTION 455 A. Acute Renal Failure 455 B. Chronic Renal Failure 455 C. Consequences of Renal Failure 457 V. URINALYSIS 460 A. Urine Color, Odor, and Turbidity 461 B. Specific Gravity 461 C. Urine Protein Determination 462 D. Occult Blood Reactions 464 E. Glucosuria 464 F. Urine pH 465 G. Acetonuria 465 H. Bile Pigments in Urine 466 I. Urine Sediment 466 VI. TESTS OF RENAL FUNCTION 468 A. Tests for Azotemia 468 B. Urine Concentration Tests 473 CLINICAL BIOCHEMISTRY OF DOMESTIC ANIMALS, FIFTH EDITION

I. INTRODUCTION In mammals, the kidneys function as a major excre­ tory organ for elimination of metabolic wastes from the body. In most species, death occurs within a week after total cessation of renal function. Partial loss of renal function results in variable deviations from nor­ mal, depending on the quantity of functional tissue remaining. The term azotemia refers to accumulation of nitrogenous wastes in the blood. Blood concentra­ tions of creatinine and urea are measured as indices of azotemia, although neither imparts significant toxicity because of its accumulation. Animals with moderate to severe azotemia may have a constellation of clinical signs, including lethargy, anorexia, mucosal ulcers, vomiting, diarrhea, weight loss, anemia, and altered urine output. These signs are referred to as renal failure, uremia, or uremic syndrome, and reflect the develop­ ment of abnormalities in a multitude of tissues second­ ary to subnormal renal functions. The role of the kidneys in maintaining life is a com­ posite of several functions. Water and many electro-

441

Copyright © 1997 by Academic Press

All rights of reproduction in any form reserved.

442

Delmar R. Finco

lytes are conserved by the kidneys in times of negative body balance and are excreted in times of positive balance. During the processing of body fluids, the kid­ neys avidly conserve nutrients such as glucose and amino acids; consequently, urine is almost devoid of such materials. Hydrogen ions are excreted or con­ served so that blood pH is kept within narrow limits. End products of nitrogen metabolism, such as urea, creatinine, and allantoin, are removed from the body via the urine so that blood levels remain low and rela­ tively constant. The kidneys produce several hor­ mones, including renin, erythropoietin, and prosta­ glandins, and perform a vital hydroxylation of vitamin D that is required for its activation. The kidneys also respond to several hormones, including antidiuretic hormone (ADH), parathyroid hormone (PTH), aldoste­ rone, thyroid hormone, and others. The story of the evolution of the kidneys to fullfill the changing needs of the organism has been told elo­ quently by Homer Smith (1959), a pioneer in the study of the kidney. Single-celled and primitive organisms had little need for excretory specialization because sim­ ple diffusion of materials to and from their iso-osmotic aqueous environment was adequate both for intake and for excretory processes. As the complexity of or­ ganisms grew, an extracellular space developed that mimicked the osmolality and electrolyte makeup of the original aqueous environment. Simple excretory organs sufficed because extracellular water and wastes could be extruded, and needs could be replaced easily by ingestion. However, the threat of osmotic imbalance occurred as the seas increased in salinity. Some animals such as the shark solved this problem by retaining urea as a compound to balance the osmolality of the seas. Others depended on renal excretion of excess salt via renal tubular systems. The animals that eventually moved to freshwater from the ocean faced a different problem. The osmolal­ ity of their environment being nil, there was an influx of water from the environment into their tissues. A high glomerular filtration rate evolved to facilitate the removal of this water. But then there was a movement of some animals from freshwater to land and a need to conserve water rather than to excrete it. The high­ filtration kidney was then modified so that most of the filtered water was reabsorbed while solute wastes were still excreted. An understanding of the evolutionary processes previously outlined provides insight into an apparent paradox involving the mammalian kidney. In mam­ mals, there is a continuous potential loss of sodium­ rich fluid equal to 4-6% of cardiac output via glomeru­ lar filtration. Energy is then expended by the tubules to reabsorb more than 99% of the filtered sodium, to-

gether with accompanying water. Certainly an engi­ neer asked to design an excretory organ would be unlikely to choose this devious approach to the end result. The evolutionary development of the kidney has clinical relevance. In renal failure, one must deal not only with problems of inadequate excretion because of decreased glomerular filtration, but also with prob­ lems of excess losses arising from alterations in func­ tion of the tubular system reclaiming filtrate. Polyuria of renal failure is an example of such a problem.

II. NORMAL RENAL FUNCTIONS A. The Nephron as a Structural and Functional Unit Most renal functions are sums of the activities of thousands of nephrons. Each nephron consists of sev­ eral anatomical divisions (Fig. 17.1). Glomeruli are composed of a tuft of capillaries supplied by an afferent arteriole and drained by an efferent arteriole. Bow­ man's capsule surrounds the glomerular capillaries

FIGURE 17.1 Nephrons as functional units in the kidney. Outer cortical and juxtamedullary nephrons ha ve some differences in struc­ ture and in function. Subdivisions of nephrons include the glomeru­ lus (G), proximal tubule (PT), loop of Henle (LH), distal tubule (OT), and the collecting duct (CD).

443

Kidney Function

and channels filtrate into the proximal tubule. The con­ voluted portion of the proximal tubule assumes a me­ andering path within the cortex before the straight portion (pars recta) descends toward the medulla. A marked thinning of the straight portion occurs abruptly as it descends; this constitutes the beginning of the loop of Henle. This structure courses into the medulla, makes a hairpin turn, and ascends into the cortex. At some point in its ascent, the epithelium be­ comes thicker, which persists to the point of apposition of the loop of Henle with the glomerulus from which the tubule originates. The afferent and efferent arteri­ oles of the glomerulus are adjacent to the ascending tubule. These structures, at the site of their confluence, are referred to as the juxtaglomerular apparatus. Modi­ fied tubular cells at this site are called the macula densa. The distal tubule is considered to be that segment of the nephron between the macula densa and the first union of two distal tubules into a collecting duct. Col­ lecting ducts empty into the renal pelvis. The number of nephrons varies among species, with larger species having more of them. For example, the cat has about 190,000, the dog about 430,000, and the cow about 4 million per kidney (Smith, 1951). Within the canine species, where there are large body size differences between breeds, nephron size rather than nephron number varies with body size (Finco and Duncan, 1972; Zhu et al., 1992). A simple view of nephron function can be summa­ rized as follows. Perfusion of glomerular capillaries with blood under pressure results in the passage of material (filtrate) through the capillary walls into Bowman's space. Filtrate then passes through the tubu­ lar system of each nephron and is modified in composi­ tion during its passage. Modification occurs primarily by reabsorption of lumen contents by tubular cells for return to the body. In addition, important secretory functions are performed by tubule cells, whereby mate­ rials delivered to the renal interstitium by the blood are transported from the interstitium to the tubule lu­ men for excretion. The end result of these tubular activ­ ities is a marked reduction in the volume of filtrate and a modification of its character.

B. Nephron Heterogeneity Common clinical methods for measuring renal func­ tion determine the sum of operation of all functional nephrons. However, both anatomical and physiologi­ cal data indicate that all nephrons are not the same. Internephron heterogeneity has been described in a variety of species (Lameire et al., 1977; Valtin, 1977; Sands et al., 1992). Anatomic differences between outer cortical (OC) and juxtamedullary OM) nephrons are

most apparent. Physiologic differences appear to exist also (Walker and Valtin, 1982). Shunting of blood from the OC nephrons to the JM nephrons could alter renal function by decreasing function of the OC nephrons and increasing function of the JM nephrons. However, studies that have investigated the role of blood shunt­ ing in pathologic processes have not led to unanimity of opinion on its relevance (Lameire et al., 1977; Brenner et al., 1986; Steinhausen et al., 1990).

C. Renal Perfusion

1. Renal Vasculature Although some variations occur, vascular patterns of renal perfusion are similar in the mammalian species that have been studied. The renal artery divides into several interlobar arteries that are of about equal size. Further vascular subdivisions occur (arcuate, interlob­ ular arteries) until afferent arterioles supply the glo­ meruli. Efferent arterioles drain the glomeruli and di­ vide into another capillary network. The site perfused by this network is dependent on the anatomical loca­ tion of the glomeruli from which the efferent arteriole originates (Beeuwkes, 1971). In the superficial cortex, efferent arterioles supply only the tubule of the parent glomerulus. In the mid cortex, efferent arterioles supply either superficial or deep tubules of different glomeru­ lar origin. In the juxtamedullary area, efferent arteri­ oles provide capillary networks for the juxtamedullary zone and straight vessels (vasa recta) that course into the medulla. Venous drainage of renal blood is similar in most mammals, but the cat has a system of subcapsular veins that drains the outer cortex exclusively (Nissen and Galskov, 1972). Owing to lack of collateral circulation, abrupt blockage of renal arterial vasculature results in infarc­ tion of the renal tissue normally supplied (Siegel and Levinsky, 1977). However, slow occlusion of the renal artery divisions may result in survival of some renal parenchyma because of communications between ex­ trarenal arteries and interlobar or arcuate arteries. Such communications apparently do not carry significant quantities of blood under usual conditions, but do di­ late when normal renal vasculature is slowly compro­ mised (Christie, 1980). 2.

Renal Blood Flow

The kidneys receive roughly 15-25% of cardiac out­ put (Kaikara et al., 1969). This rate of perfusion is very high compared to that of most other tissues. Clearance procedures (discussed in Section VI) and electromag­ netic flowmeters provide reliable methods for measur-

444

Delmar R. Finco

ing total renal blood flow (RBF) but give no informa­ tion on intrarenal distribution of blood. Other methods (inert gas washout, microsphere injection) have been used for intrarenal blood flow measurements but all have been criticized (Dworkin and Brenner, 1991). However, renal medullary blood flow rate is known to be comparable to that of other tissues of the body and much lower than the high cortical flow rate.

capillary lumen (Pitts, 1974). Another difference is that glomerular filtrate is not totally devoid of protein. Fil­ trate harvested from proximal tubules of dogs by mi­ cropuncture techniques had albumin concentrations of 20 to 30 mg/ liter (2.0 to 3.0 mg / dl), and contained lower molecular weight proteins as well (Dirks et al., 1965; Maack et al., 1992). The relevance of this protein is discussed in Section V.c.

3. Autoregulation

2.

Both RBF and glomerular filtration rate (GFR) re­ main fairly constant during variations in systemic arte­ rial pressure from 10.7 to 24 kPa (80 to 180 mm Hg) (Pitts, 1974). This phenomenon is known as autoregula­ tion, and it occurs secondary to modulation of afferent arteriolar resistance. It occurs in the isolated perfused kidney and thus is not dependent on renal innervation or on extrarenal humoral agents. Two major theories are provided to explain autoregulation (Dworkin and Brenner, 1991). The myogenic theory postulates a di­ rect vascular response to changes in systemic arterial pressure. The tubuloglomerular feedback theory pos­ tulates that some intratubular signal is transmitted via the macula densa to the adjacent arterioles, causing their response to changing arterial pressure. Autoregulation of RBF and GFR operates under physiologic conditions, but it may be abolished with disease, which has significance in clinical medicine.

In order for material to pass from the glomerular capillary lumen into Bowman' s space, it must traverse three anatomic barriers: the capillary endothelium, the basement membrane, and the capillary epithelium. Glomerular capillaries are different from other capillar­ ies of the body in several ways. The endothelial fenes­ trae are larger and account for proportionately more surface area, the basement membranes are thicker, ad­ ventitial connective tissue is absent, and the epithelial cells have an elaborate arrangement of interdigitating foot processes (Farquhar, 1975). An intercapillary cell, the mesangial cell, acts as a scavenger cell for particles lodged in the capillary wall. The mesangial cell also has contractile properties and may influence the filtra­ tion area of the glomerulus (Raij and Keane, 1985). It is generally acknowledged that the endothelium can function only to restrain cellular elements since the fenestrae are 50 to 100 nm in diameter. Athough the gaps between the epithelial foot processes are very wide (25 nm ), slit membranes which cover the gaps could act as filters. The uninterrupted nature of the basement membrane suggests that it is likely to be the major filtration barrier (Farquhar, 1975).

4. Role of Neural Stimuli in Renal Perfusion The kidney receives sympathetic fibers, with evidence for direct innervation of renal tubular cells (Barajas and Mueller, 1973), which may play some role in renal function (Gottschalk et al., 1985). Current thought suggests that the kidneys are under minimal basal neural tone so that denervation produces mini­ mal changes in renal function. However, neural stimu­ lation beyond the basal tone may alter renal function, particularly by vasoconstriction and decreased renal blood flow. An example is given in Section III.B.2.

D. Glomerular Filtration

1. Nature of Filtrate Material that passes from the capillary lumen into Bowman's space has been collected by micropuncture techniques and compared to the composition of blood. The composition of glomerular filtrate is nearly identi­ cal to that of blood from which the cells and protein have been removed. A small difference in ion concen­ tration exists as a consequence of Gibbs-Donnan fac­ tors generated secondary to protein restriction to the

Anatomic Considerations for Filtration

3. Glomerular Filtration Dynamics Forces relevant to glomerular filtration are shown in Fig. 1 7.2. Hydraulic (hydrostatic) pressure of cardiac origin provides the energy for glomerular filtration. The major force opposing filtration is the oncotic pres­ sure of plasma proteins. An additional force opposing filtration is hydrostatic force in Bowman's space. One potential force, oncotic pressure in Bowman's space, is usually ignored because protein concentration in filtrate is so low compared to that of plasma. The permeability and the surface area of the capil­ lary filtering area are other factors that influence the rate of filtration (Maddox and Brenner, 1991). As fluid traverses the glomerular capillary bed, plasma proteins are concentrated because filtrate leaves the capillary. This concentration results in a progressive rise in on­ cotic pressure as the efferent arteriole is approached. In rats, oncotic pressure in the capillary matches net effective hydraulic pressure before the end of the capil-

445

Kidney Function 70 m m

AA

�tjJ� �t-=28 m m

V

For quantitative comparisons, the passage of macro­ molecules through the glomerular barrier can be com­ pared to water movement through the barrier (Table 17.1). EA

36 mm

FIGURE 17.2 Schematic representation of forces significant in glomerular filtration. Hydraulic (blood) pressure is the major factor forcing fluid from the capillary lumen into Bowman's space. This pressure is opposed by plasma oncotic pressure and hydraulic pres­ sure in Bowman's space.

lary, causing filtration to cease prematurely. This is called filtration pressure equilibrium. With these condi­ tions, an increase in plasma flow rate increases GFR as follows. An increase in flow rate would decrease the proportion of plasma that became filtrate. Intraluminal oncotic pressure would not rise as rapidly as it would at a slower flow rate, so a positive filtration pressure would exist over a greater span of the capillary loop. Thus, with filtration pressure equilibrium, plasma flow rate markedly influences GFR (Tucker and Blantz, 1977). However, filtration pressure equilibrium does not exist in dogs and probably does not exist in most other species (Fried and Stein, 1983). 4.

Filtration of Macromolecules

Common methods of detecting proteinuria that are used as a part of urinalysis are negative in normal animals because the glomerular filter does not allow passage of large quantities of protein into filtrate. The idea that pores in the filtration barrier prevent mac­ romolecules and cells from entering filtrate led to the concept that the size and shape of molecules influ­ enced their filtration (Pitts, 1974). In addition, filtra­ tion is influenced by electrical charge (Rennke and Venkatachalam, 1977; Brenner et al., 1978). Negatively charged sialoproteins are components of all three lay­ ers of the capillary wall. These negative charges facili­ tate passage of cationic macromolecules, and impede passage of anionic macromolecules through the filter. Albumin has a size and shape that marginally prevents its passage through the glomerular capillary wall. However, its negative charge at blood pH further im­ pedes its loss into filtrate. If the negatively charged sialoproteins lining the filtration barrier are removed experimentally or with disease, then albuminuria oc­ curs (Brenner et al., 1978).

E. Tubular Modification of Filtrate Death from hypovolemia and electrolyte depletion would occur in less than 1 hour if all the glomerular filtrate formed were lost from the body. The renal tu­ bules modify glomerular filtrate to achieve body ho­ meostasis for water and most electrolytes. Large quan­ tities of filtrate (65-80%) are reabsorbed in the proximal tubules. As the reduced volume of filtrate traverses the more distal portions of the nephron, further reab­ sorption takes place. Secretion of certain endogenous materials (i.e., protons, potassium, organic anions) as well as exogenous materials (certain drugs, toxins) from tubular cells into the tubular lumen also is impor­ tant in renal excretory functions. The concept of tubular maxima (Tm ) for reabsorption and secretion applies to many materials handled by the tubules. This concept originated before cell recep­ tors and transport kinetics were studied in other sys­ tems. Whole kidney Tm values, however, represent the composite of all segments of the nephron and the action of both the luminal and the basolateral cell membranes. The Tm value for maximal tubular reabsorption or se­ cretion by the kidney is expressed as milligrams trans­ ported per 100 ml of filtrate, or milligrams transported per unit time. Another measurement that evaluates the role of the tubules in body homeostasis is the determination of fractional excretion (FE). The FE is defined as that frac­ tion of filtered substance (S ) that is excreted in the urine (i.e., not reabsorbed by tubules). The FE of a substance S can be calculated from values for its filtered load and its urinary excretion. For an 5 that passes freely through the glomerular barrier, the filtered load is calculated as the product of GFR X plasma concenTABLE 17.1

Permeability of the Glomerular Capillary Barrier

Substance

MW

Permeability coefficient"

Urea Creatinine Glucose Inulin Myoglobin Hemoglobin Albumin

60 113 180 5,500 17,000 65,000 69,000

1 1 1 0.98 0.75 0.03 (J Q.

14 PCV

PROTEIN 12

50 10 45

8

40 35

0

10

20

30

PLASMA VOLUME DEFICIT

�

�

(%)

40

50

S Z jjj l0 a:: Q. < :E en < ..J Q.

6

FIGURE lS.7 Effects of decreasing plasma volume on pev and plasma protein concentration as calculated using Eqs. (18.13) and (18.14).

507

Fluid, Electrolyte, and Acid-Base Balance ,00

II PCV o PROTEIN

80 60 40

INCREASE

('Yo)

20

0

o

5

10

PLASMA VOLUME DEFICIT (%) 15

20

25

30

35

40

45

50

FIGURE 18.8 Calculated effects of plasma volume depletion. Percentage increase in PCV and plasma protein concentration with progressive plasma volume deficits.

should be correlated with clinical evidence such as return of normal pulse pressure, capillary refill time, and jugular distensibility.

B. Serum Sodium Serum sodium concentration varies within rela­ tively narrow limits in the normal individual but there is substantial interspecies variation in the normal range of sodium, chloride, and osmolality as indicated in Table 18.4. A serum sodium concentration of 134 mEq/ liter (134 mmol/ liter), while quite normal for a horse or cow, represents a significant hyponatremia in a dog or cat. Before proceeding with a discussion of the signifi­ cance of alterations of sodium concentration, some comment on the methods used for electrolyte determi­ nation is appropriate. Flame photometry for many years had been the standard method for the determina­ tion of both sodium and potassium. Recently, methods employing ion-specific electrodes have achieved wider use. Although both procedures yield accurate and re­ producible results, there may be consistent differences between the two procedures. Flame photometry yields results in mEq or mmol/ liter of plasma or serum, whereas the ion-specific electrodes are generally con­ sidered to measure the electrolyte concentration only in the aqueous phase of the sample. The water content of serum or plasma samples is normally between 93 and 94%, with most of the remaining volume occupied by protein. For this reason, electrolyte determinations performed by flame photometry tend to be 6-7% lower than those obtained using ion-specific electrodes. Inter­ estingly, ion-specific electrode instruments that dilute

samples tend to yield values which are very similar to values reported from the flame photometer. This may be a function of the mathematical algorithm employed in these devices to convert changes in electric potential to electrolyte concentration. There may be interfering substances in the urine of some animal species which render urine potassium determinations inaccurate when assessed by ion-specific potentiometry (Brooks et al., 1988) The relationship between protein concen­ tration and serum or plasma water content has been determined for man and the horse (Eiseman et al., 1936; Carlson and Harrold, 1977; Harris et al., 1987). This relationship can be used to correct serum or plasma electrolyte concentration determined by flame pho­ tometry for the effects of variation in protein concentra­ tion and thus water content. Electrolyte concentration then can be expressed per liter of serum or plasma water. Sodium mEq/ liter H20 Measured sodium (mEq/ liter) = X 100 % Water

(18.15)

1. Hyponatremia The common causes of hyponatremia are listed in Table 18.5. A falsely low sodium concentration may be noted when there is marked hyperlipemia or hyper­ proteinemia. Large quantities of lipid or protein oc­ cupy a significant volume in a serum or plasma sample and because electrolytes are dissolved only in the aque­ ous phase the measured concentrations will be falsely low. The presence of obvious lipemia or markedly ele­ vated serum protein concentration should alert the cli­ nician to the probable cause of an accompanying hypo-

508

Gary P. Carlson TABLE 18.5

Causes of Hyponatremia

False hyponatremia Hyperlipidemia Hyperproteinemia Hyperglycemia Hyponatremia (relative water excess) Decreased effective circulating volume Vomiting Diarrhea Excessive sweating Cutaneous loss, burns Blood loss Repeated pleural drainage Pleuritis, chylothorax Adrenal insufficiency "Third space problems" Sequestration of fluid Peritonitis Ascites Ruptured bladder Excess circulating volume Congestive heart failure Chronic liver failure Nephrotic syndrome Normal effective circulating volume Physchogenic water drinking Renal disease Inappropriate ADH secretion

natremia. Should this information not be available, the presence of a false hyponatremia can be determined by comparison of the measured serum osmolality and the calculated osmolality based on sodium, glucose, and urea concentration as explained in Section VIlLE. This potential cause for confusion in interpretation of hyponatremia can be avoided if ion-specific electrodes are used for electrolyte determination. Marked hyperglycemia associated with diabetes mellitus or the administration of glucose at an exces­ sive rate generally produces a hyponatremia. As glu­ cose concentration increases in the ECF, osmotic forces are generated that result in the movement of cellular water into the ECF, diluting serum sodium concentra­ tion. The actual mechanics of this process are compli­ cated by the fact that cells and tissues are variably permeable to glucose and, thus, the glucose space clearly exceeds the ECF volume. For practical pur­ poses, we can anticipate that serum sodium concentra­ tion will decline 1.6 mEq/ liter (1.6 mmol / liter) for each 100 mg / dl (5.55 mmol/ liter) increase in glucose con­ centration (Saxton and Seldin, 1986). Serum osmolality may be increased by hyperglycemia but this should not cause a large disparity between the measured and calculated serum osmolality. Changes in water balance are principally responsi­ ble for changes in serum sodium concentration (Leaf,

1962). Hyponatremia should be considered an indica­ tion of a relative water excess (Scribner, 1969). Hypona­ tremia is often but not invariably associated with con­ ditions which cause sodium depletion and resultant decreases in effective circulating volume. These condi­ tions include vomiting, diarrhea, excessive sweat losses, and adrenal insufficiency. Dehydration and vol­ ume depletion induce neurohormonal responses that result in increased water consumption via increased thirst and enhanced renal conservation of water as well as sodium (Rose, 1984). Fluid losses in these forms of dehydration are most often hypotonic or isotonic and initial fluid and electrolyte deficits do not result in hyponatremia until water intake and / or renal water retention disturb the balance between the remaining exchangeable cations and the total body water. Thus, while substantial sodium and potassium deficits are associated with these conditions, plasma sodium con­ centration does not always reflect these deficits and diagnosis of sodium depletion should be based on other grounds (Scribner, 1969). The accumulation of sodium-containing fluid within body cavities as a result of ascites, peritonitis, or a ruptured bladder is referred to as a third space problem (Rose, 1984). The fluid that accumulates in the "third space" has a composition similar to the ECF. When this accumulation of fluid occurs rapidly, plasma volume is reduced and serum sodium concen­ tration then may decrease as the compensating re­ sponses result in water retention. A classic example of this situation is the marked hyponatremia associated with ruptured bladder in neonatal foals. As the dilute urine of the neonate accumulates in the abdomen, os­ motic equilibrium is established first with the ECF and then with all of the body fluid compartments. Sodium and chloride as well as other ions are drawn from the rest of the ECF into this progressively expanding fluid compartment and hyponatremia develops despite the fact that there has been no appreciable loss of sodium from the body. The severe neurologic signs associated with ruptured bladder in foals are related in large part to the effects of the sudden and marked hypotoniC hyponatremia on the central nervous system. Hypona­ tremia associated with excessive retention of water also can occur without the development of significant so­ dium depletion or decreases in effective circulating fluid volume. Hyponatremia may be observed with psychogenic polydipsia if the rate of water consump­ tion exceeds renal capacity for free water clearance due to intrinsic renal disease or renal medullary washout (Tyler et al., 1987). This also may occur in patients with impaired free water clearance due to renal disease or when under the influence of inappropriate release of ADH (McKeown, 1984). Hyponatremia and associated

509

Fluid, Electrolyte, and Acid-Base Balance

produce an ADH-like material, with a variety of brain diseases which apparently stimulate synthesis and re­ lease of ADH, or in association with certain pulmonary diseases which may result in abnormal neural inputs from the lung that trigger the inappropriate ADH re­ lease from the pituitary. The diagnosis of SIADH in the absence of readily available procedures for ADH determination is depen­ dent on ruling out a number of other potential causes for persistent hyponatremia. The diagnostic criterion for SIADH for human subjects include the following (McKeown, 1984):

neurologic disturbances can develop with naturally occurring disease (Lakritz et al., 1992) or iatrogenically if excessive amounts of free water are administered to patients with altered renal function (Arieff, 1986; Sterns et al., 1986). Urine sodium concentration can be useful in the differentiation of the causes of hyponatremia as indi­ cated in Fig. 18.9. Renal adaptive responses normally result in sodium retention and production of urine with very low sodium concentration in patients with sodium depletion resulting from vomiting, diarrhea, excessive sweat loss, or third space problems (Rose, 1984). Hyponatremia and hypokalemia of adrenal in­ sufficiency are generally associated with a relatively high urine sodium concentration (Rose, 1984). With the syndrome of inappropriate secretion of antidiuretic hormone (SIADH), urine sodium tends to be high in the presence of hyponatremia, whereas urine sodium concentration tends to be low in animals with psycho­ genic polydipsia. Urine sodium concentration in ani­ mals with renal failure can be quite variable. A form of moderate to pronounced hyponatremia associated with elevated levels of ADH which devel­ ops in the absence of an appropriate osmotic or volume stimuli has been well recognized in human subjects. This syndrome has been associated with several sys­ temic disorders including malignant neoplasms which

1. Demonstration of a hypotonic hyponatremia without hypovolemia or edema 2. Normal renal, adrenal, and thyroid function 3. Inappropriately elevated urine osmolality relative to plasma osmolality 4. Relatively high urine sodium concentration 5. Correction of the hyponatremia by strict fluid restriction SIADH has been reported in the dog (Breitschwerdt and Root, 1979; Giger and Gorman, 1984; Houston et al., 1989; Crow and Stockham, 1985), but the incidence and importance of this problem in other animal species is uncertain. A variant of SIADH has been described in certain chronically ill and malnourished human sub-

HYPONATREMIA (RENAL RESPONSES) •

DECREASED BLOOD VOLUME

NORMAL BLOOD VOLUME �

ADRENAL INSUFFICIENCY (Urine Na High)

I

SODIUM DEPLETION (Urine Na Low)

PSYCHOGENIC DRINKER (Urine Na Low)

BLOOD LOSS (Urine Na Low)

INAPPROPRIATE ADH (Urine Na High)

H EART or LIVER FAILURE (Urine Na Low)

RENAL FAILURE (Urine Na Variable)

I

1 I

I

I

SEQUESTERED FLUID (Urine Na Low)

FIGURE 18.9 The urinary sodium concentration as an aid in the differentiation of possible causes of hyponatremia. Urinary electrolyte concentration is also influenced by dietary intake.

510

Gary P. Carlson

jects that maintain a mild but persistent hyponatremia. In these individuals ADH secretion remains under os­ motic control but it would appear that the osmorecep­ tor threshold functions at a lower value than normal (McKeown, 1984). Thus, it is the resetting of the osmo­ stat which results in persistent chronic hyponatremia. Resolution of the underlying medical problem usually results in correction of this variant form of SIADH. 2. Hypernatremia

Hypernatremia almost always is associated with ele­ vation of serum osmolality. Hypernatremia occurs in dehydrated subjects when water losses exceed losses of sodium and potassium and should be considered as an indication of a reiative water deficit (Scribner, 1969). This can occur in the initial stages of diarrhea, vomit­ ing, or renal disease if losses of water exceed the elec­ trolytes lost (Saxton and Seldin, 1986). Common causes of hypernatremia are listed in Table 1 8.6. As water losses are replaced by increased water consumption and / or enhanced renal water retention, serum sodium concentration tends to decline into or below the normal range. Hypernatremia also develops as the result of an essentially pure water loss, such as the evaporative respiratory water loss in panting animals (Tasker, 1980). Hypernatremia is associated with excessive re­ nal free water loss with either central or nephrogenic diabetes insipidus if water intake is restricted (BreuTABLE 18.6

Causes of Hypematremia

Water losses in excess of electrolyte loss Digestive Vomiting Diarrhea Cutaneous Bums Renal Diuretics Diabetes mellitus Manitol Intrinsic renal disease Pure water losses Insensible Panting Diabetes insipidus Central Nephrogenic Inadequate water intake Water deprivation Abnormal thirst mechanism Sodium excess (water restriction) Hypertonic saline or sodium bicarbonate Salt poisoning Mineralocorticoid excess

kink et ai., 1983). Food and water deprivation in normal individuals is associated with substantial reduction of renal and fecal output (Tasker, 1 967b; Carlson et ai., 1979b; Brobst and Bayly, 1982; Genetzky et ai., 1987). However, continued cutaneous and respiratory insen­ sible water loss may result in hypernatremia (Elkinton and Taffel, 1942; Rumsey and Bond, 1976; Carlson et ai., 1979b; Genetzky et ai., 1987). Abnormal thirst mechanisms with resultant hyper­ natremia have been reported in young dogs (Crawford et al., 1984; Hoskins and Rothschmitt, 1984). Hyperna­ tremia may occur transiently following administration of hypertOnic saline or sodium bicarbonate if water intake is restricted or impaired. The hypernatremia observed with salt poisoning in cattle and swine is, in fact, triggered by water restriction in animals with a high salt intake (Padovan, 1980; Pearson and Kallfelz, 1 982). As long as adequate water is available, salt poi­ soning does not occur. In human subjects, hyperna­ tremia is reported with mineralocorticoid excess (McKeown, 1984).

C. Serum Potassium Serum potassium concentration does not always re­ flect potassium balance but is influenced by factors which alter internal balance, that is, the distribution of potassium across the cell membrane between the ECF and ICF as well as factors which change external bal­ ance, that is, potassium intake and output (Patrick, 1977; Rose, 1984; Brobst, 1986). The effective adjust­ ment of external and internal balance in normal indi­ viduals in response to either large potassium loads or excessive potassium losses usually maintains serum potassium concentration within normal limits. How­ ever, changes in potassium concentration occur in a wide variety of clinical circumstances and have pro­ found neuromuscular effects largely due to changes in cell membrane potential (Brobst, 1986). The compen­ sating responses to change in circulating fluid volume distribution and acid-base balance can result in con­ fusing and contradictory findings. As an example, calves with acute diarrhea may develop significant depletion of body potassium stores due to excessive losses and inadequate intake (Phillips and Knox, 1969). However, the serum potassium concentration in these calves may actually be normal to increased as the result of renal shutdown and the metabolic acidosis induced by dehydration and sodium depletion with resultant decreases in the effective circulating fluid volume. Electrolyte replacement fluids given to these calves should include potassium (Fettman et al., 1986). Correct interpretation of serum potassium concentration thus requires a knowledge of probable intake and sources

511

Fluid, Electrolyte, and Acid-Base Balance

of excessive loss as well as the status of renal function and acid-base balance. Measurement of erythrocyte potassium concentration has been suggested as an aid in the assessment of the potassium status of horses with exercise-related myopathy (Muylle et al., 1984a, 1984b; Bain and Merritt, 1990). However, experimental studies have failed to demonstrate a close correlation between erythrocyte potassium concentration and po­ tassium depletion.

1. Hypokalemia Hypokalemia occurs relatively frequently in domes­ tic animals and may result from depletion of the body' s potassium stores or from a redistribution of potassium from the ECF into the ICF space (Brobst, 1986), as indicated in Table 18.7. Hypokalemia most often is associated with excessive potassium losses from the gastrointestinal tract as the result of vomiting or diar­ rhea (Tasker, 1967c). Excessive renal loss of potassium results from the action of mineralocorticoid excess, the action of certain diuretics, and as the result of altered renal tubular function in animals with renal tubular acidosis or postobstructive states. Chronic dietary pot­ assium deficiency eventually can lead to modest hypo­ kalemia even in normal individuals (Aitken, 1976; Dow et al., 1987b). A rapidly developing and profound hypo­ kalemia can develop in animals with reduced dietary intake due to anorexia when coupled with other causes of excessive potassium loss (Tasker, 1980). Hypokalemia may develop without potassium depletion as the result of intracellular movement of potassium from the ECF space. This occurs with an acute alkalosis (Burnell et al., 1966) and in patients treated with insulin and glucose infusions (Tannen, 1984). In fact, medical management of severe lifeTABLE 18.7

Causes of Hypokalemia

Hypokalemia due to altered external balance Gastrointestinal losses Vomiting Diarrhea Renal losses Mineralocorticoid excess Diuretics Renal tubular acidosis Postobstruction diuresis Hypokalemic nephropathy in cats Dietary deficiency Hypokalemia due to altered internal balance Excessively rapid bicarbonate administration Insulin with glucose administration Catecholamine release Hypokalemic periodic paralysis

threatening hyperkalemia often involves the adminis­ tration of glucose, insulin, and in some circumstances bicarbonate in an effort to shift potassium intracellu­ larly. Catecholamines exert a biphasic effect on potass­ ium concentration (Tannen, 1984). The initial response to catecholamine administration is a modest transient increase in potassium as the result of a-adrenergic stimulation followed by hypokalemia as the result of ,B-adrenergic receptor responses (Clausen et a!., 1980; Bendheim et al., 1985; Williams et al., 1985). Beta­ adrenergic agents have been used for the treatment of hyperkalemic periodic paralysis (Bendheim et al., 1985). The relationship between catecholamine release, receptor stimulation, and potassium balance may play a significant role in the development of exertional fa­ tigue in human and equine athletes (Sjogaard et al., 1985; Carlson, 1987).

2. Hyperkalemia Hyperkalemia can be produced iatrogenically by the excessive administration of potassium salts but does not usually result from high dietary intake in individuals with normal renal function (Aitken, 1976). The three major causes of hyperkalemia are indicated in Table 18.8. Hyperkalemia may develop in vitro in blood sample containers due to hemolysis or pro­ longed storage of blood samples prior to the separation of serum or plasma from erythrocytes in the sample. This leakage of erythrocyte potassium can result in significant errors in those species with a high potas­ sium content in their erythrocytes; the horse, pig, and most cattle (Tasker, 1980). The erythrocytes of cats and most dogs have a high sodium content and relatively low potassium. Slight hemolysis will have little effect on serum or plasma potassium concentration in these TABLE 18.8

Causes of Hyperkalemia

False hyperkalemia Hemolysis Markedly elevated leukocyte or platelet count Hyperkalemia due to altered external balance Addison's disease Renal disease Ruptured bladder Urethral obstruction Hypovolemia with renal shutdown Hyperkalemia due to altered internal balance Metabolic acidosis Diabetes Tissue necrosis Hyperkalemic periodic paralysis Vigorous exercise

512

Gary P. Carlson

species. Polymorphism however, occurs in intracellu­ lar cation content of certain breeds of sheep, cattle, and dogs which relate to sodium-potassium ATPase activity of mature erythrocytes. Release of potassium from leukocytes or platelets into the serum following clot formation in the sample collection vial is a poten­ tial cause of hyperkalemia in subjects with a marked leukocytosis or thrombocytosis (Mandell et. al., 1988; Degan, 1986). Hyperkalemia may be associated with excessive re­ nal potassium retention in conditions such as Addi­ son's disease, acute renal failure, and renal shutdown secondary to inadequate effective circulatory fluid vol­ ume. Hyperkalemia results, in a number of circum­ stances, from the movement of intracellular potassium into the ECF without change in external potassium balance. Hyperkalemia develops in association with a metabolic acidosis, particularly when the acidosis results from volume depletion complicated by renal shutdown. Hyperkalemia may be noted in patients with diabetes or transiently in animals with massive muscle necrosis. Interestingly, most horses with exer­ tional rhabdomyolysis do not develop hyperkalemia or a metabolic acidosis (Koterba and Carlson, 1982). Vigorous short-term exercise at high intensity results in a profound hyperkalemia in horses. Serum potas­ sium as high as 9-10 mEq / liter (9-10 mmol / liter) has been observed transiently in horses exercising at maxi­ mal intensity on a high-performance treadmill (Harris and Snow, 1986). Muscular exhaustion in these horses appeared to be related to the hyperkalemia and the profound lactic acidosis seen in the immediate postex­ ercise state (Kryzwanek, 1974; Kryzwanek et al., 1976; Milne et al., 1976; Carlson, 1987). An interesting and unusual condition of episodic hyperkalemia and mus­ cular weakness has been recognized in the dog ( Jezyk, 1982) and certain quarterhorses (Cox, 1985; Steiss and Naylor, 1986; Spier et al., 1990, 1993; Pickar et al., 1991). In the horse the condition closely resembles the herita­ ble disease hyperkalemic periodic paralysis, which has been reported in human subjects, and is due to an alteration in the voltage regulated sodium channel (Ru­ dolph et al., 1992a, 1992b). Sudden marked increases in serum potassium concentration result from the transcellular movement of potassium. Serum po­ tassium concentration can reach 8-9 mEq / liter (89 mmol/ liter) and is associated with profound electro­ cardiographic abnormalities and fluid shifts which re­ sult in marked increases in PCV and protein concentra­ tion. The disease is inherited as an autosomal dominant and all affected horses can be traced back to a common ancestor. A DNA test which can detect the single base pair substitution responsible for this disease has been developed (Rudolph et al., 1992a). Using this procedure

it is possible to identify individuals which are hetero­ zygous or homozygous for this trait.

D. Serum Chloride It long has been assumed that the anion chloride which combines with sodium to form common salt simply follows sodium in the physiologic processes which regulate body fluid and electrolyte balance. It is becoming increasingly apparent that this may not always be true and that some of the problems ascribed to sodium retention do not occur unless chloride is present in excess as well (Kurtz et al., 1987). Causes of alterations in chloride concentration are given in Table 18.9. The hyperchloremia and hypochloremia which are normally seen in association with roughly propor­ tional changes in sodium concentration are due to changes in body water balance. Changes in chloride concentration which are not associated with a similar change in sodium concentra­ tion are usually associated with acid-base imbalances (Divers et al., 1986). Chloride concentration tends to vary inversely with bicarbonate concentration. A dis­ proportionate increase in chloride most commonly is associated with a normal to low anion gap hyperchlor­ emic metabolic acidosis, and may be seen as a compen­ sating response for a primary respiratory alkalosis (Saxton and Seldin, 1986). Disproportionate decreases in chloride characteristically are seen in a metabolic alkalosis but also may be seen as part of the compensat­ ing response for a chronic primary respiratory acidosis (Saxton and Seldin, 1986). E. Osmolality

It has been demonstrated that the concentration of sodium in serum water is closely correlated with the serum osmolality over an extremely wide range TABLE 18.9 Causes of Alterations in Chloride Concentration Hyperchloremia With proportional increase in sodium Dehydration (relative water deficit) Without proportional increase in sodium Hyperchloremic metabolic acidosis Compensation for respiratory alkalosis Hypochloremia With proportional decrease in sodium Overhydration (relative water excess) Without proportional decrease in sodium Hypochloremic metabolic alkalosis Compensation for respiratory acidosis

513

Fluid, Electrolyte, and Acid-Base Balance

of physiologic and pathologic states provided ap­ propriate corrections are made for the contributions made by variations in glucose and urea concentrations (Edelman et al., 1958). The measurement of serum os­ molality has two specific and very useful purposes (Gennari, 1984). First, to determine whether serum wa­ ter content deviates widely from normal, and second to screen for the presence of foreign low-molecular­ weight substances in the blood. Interpretation of serum osmolality for these purposes requires simultaneous comparison of the measured osmolality and the calcu­ lated osmolality as determined from the measured con­ centrations of the major solutes in serum (sodium, glu­ cose, and urea). The difference between the measured and the calculated osmolality is sometimes referred to as the osmolal gap (Feldman and Rosenberg, 1981; Shull, 1978, 1981). Sodium is the principal cation in serum and sodium is balanced by a number of different anions (chloride, bicarbonate, protein, sulphate, and phosphate). So­ dium concentration thus provides a reasonable esti­ mate of the total electrolyte concentration (anions and cations) in the sample, and in this calculation is usually multiplied by a factor of 2. As has been mentioned earlier the water content of serum samples is approxi­ mately 94%. Correction for the water content of serum is not necessary since, fortuitously, it is counterbal­ anced by the fact that sodium chloride does not dissoci­ ate completely and has an osmotic coefficient of 0.93 in serum (Wolf, 1966; Dahms et al., 1968; Rose, 1984). The osmolality can thus be calculated using the mea­ sured sodium concentration and the concentration of the two nonelectrolyte components of serum which are normally present in amounts sufficient to influ­ ence osmolality: mOsm / kg H20

=

2

x

Sodium + Glucose + Urea.

(18.16)

This calculation is valid if concentrations of sodium, glucose, and urea are expressed in mmol/ liter. Conver­ sion of glucose concentration from mg / dl to mmol/ liter requires division by 18, whereas urea concentra­ tion in mg / dl can be converted to mmol/ liter by divid­ ing by 2.8. Decreases in osmolar gap indicate laboratory error. Increases in osmolar gap (>10 mOsm/ kg) could also represent laboratory error but generally result from one of two circumstances: either a decrease in serum water content or the addition of low-molecular-weight substances in serum. Decreases in serum water content occur with marked hyperlipidemia or hyperproteine­ mia and the calculated osmolality will exceed the mea­ sured osmolality. This is the cause of the false hypona­ tremia discussed in Section VLB. The presence of a

similar disparity between the measured and calculated osmolality in the absence of hyperlipidemia or hyper­ proteinemia should prompt an alert for the presence of exogenous substances in abnormally high concentra­ tions in the serum. These substances could include a variety of exogenous and potentially toxic compounds such as mannitol, ethanol, methanol, ethylene glycol, isopropanol, ethyl ether, acetone, trichlorethane, and paraldehyde (Saxton and Seldin, 1986).

References

Adrogue, H. L Brensilver, J., and Madias, N. E. (1978). Am. J. Physiol. 235, F291-F297. Aitken, F.e. (1976). In "Sodium and Potassium in Nutrition of Mam­ mals," Technical Communication No. 26, Commonwealth Bureau of Nutrition, Commonwealth Agricultural Bureaux, Farnham Royal, England. Arieff, A. 1. (1986). N. Engl. J. Med. 314, 1529-1535. Bain, F. T., and Merritt, A. M. (1990) J. Am. Vet. Med. Assoc. 196, 12591261. Bendheim, P. E., Reale, E. 0., and Berg, B. O. (1985). Neurology 35, 746-749. Bennett, E. J. (1975). Anesthesiology 43, 210-224. Bianca, W., Findlay, J. D., and McLean, J. A. (1965). Res. Vet. Sci. 6, 38-55. Billig, D. M., and Jordan, P. H. (1969). Surg. Gynecol. Obstet. 129, 1274 1282. Billig, E. (1984). J. Dairy Sci. 67, 2939-2948. Block, E. (1984). J. Dairy Sci. 67, 2939-2948. Brooks, e. 1., Garry, F., and Swartout, M. S. (1988). Am. J. Vet. Res. 49, 710-714. Boyd, J. W. (1981). Br. Vet. J. 137, 166-172. Breitschwerdt, E. B., and Root, e. R. (1979). J. Am. Vet. Med. Assoc. 175, 181-186. Breukink, H. J., Van Wegen, P., and Schotman, J. H. (1983). Equine Vet. J. IS, 284-287. Bristol, D. G. (1982). J. Am. Vet. Med. Assoc. 181, 63-65. Brobst, D. (1984). Calif. Vet. 38(9), 20-23, 48. Brobst, D. (1986). J. Am. Vet. Med. Assoc. 188, 1019-1025. Brobst, D. F., and Bayly, W. M. (1982). J. Equine Vet. Sci. 2, 51-56. Buck, W. B., Osweiler, G. D., and Van Gelder, G. A. (1976). In "Clinical and Diagnostic Veterinary Toxicology" (G. A. Van Gelder, ed.), 2nd ed., pp. 83-86. Kendall/ Hunt Publishing Co., Dubuque, Iowa. Burnell, J. M., Villamil, M. F., Uyeno, B. T., and Scribner, B. H. (1966). J. Clin. Invest. 35, 935-939. Carlson, G. P. (1975). Proc. Int. Symp. Equine Hematol., 1st, East Lansing, Michigan, pp. 473-480. Am. Assoc. Equine Pract., Golden, Colorado. Carlson, G. P. (1979a). Proc. Annu. Meet. Am. Assoc. Equine Pract., 25th, Miami, Florida pp. 459-468. Carlson, G.P. (1979b). Vet. Clin. N. Am. (Lg. Anim. Pract.) I, 313-329. Carlson, G. P. (1983a). In "Current Therapy in Equine Medicine" (N. E. Robinson, ed.), pp. 311-318. Saunders, Philadelphia, Penn­ sylvania. Carlson, G. P. (1983b). In "Equine Exercise Physiology" (D. H. Snow, S. G. B. Persson, and R. J. Rose, eds.), pp. 291-309. Granta Edi­ tions, Cambridge, England. Carlson, G .P. (1983c). III "Current Therapy in Equine Medicine" (N. E. Robinson, ed.), pp. 318-321. Saunders Co., Philadelphia, Pennsylvania. -

514

Gary P. Carlson

Carlson, G. P. (1987). III "Equine Exercise Physiology 2" 0. R Gillespie and N. E. Robinson, eds.), pp. 393-426. ICEEP Publica­ tions, Davis, California. Carlson, G. P. (1995). Equille Vet J. Suppl. 18, 261-265. Carlson, G. P., and Harrold, D. R (1977). Vet. Clill. Patlzal. 6, 18-20. Carlson, G. P., and Rumbaugh, G. E. (1983). Am. J. Vet. Res. 44, 964-968. Carlson, G. P., Harrold, D., and Rumbaugh, G. E. (1979a). Am. J. Vet. Res. 40, 587-589. Carlson, G. P., Rumbaugh, G. E., and Harrold, D. R (1979b). Am. J. Vet. Res. 40, 982-985. Chinard, F. P. (1951). III "Methods in Medical Research" (M. B. Vessehers, ed.), Vol. 4, pp. 38-48. Year Book Medical Publish­ ers, Chicago. Christensen, G. (1993). Scmld. J. Lab. Invest. 53, 203-209. Clark, A M. (1980). Vet. Rec. 106, 146-149. Clausen, T., Wang, P., Orskov, H., and Kristensen, O. (1980). Scalld. J. Clill. Lab. Illvest. 40, 211-220. Cornelius, L. M., Finco, D. R, and Culver, D. H. (1978). Am. J. Vet. Res. 39, 1185-1190. Cox, J. H. (1985). Proc. Amlll. Meet. Am. Assoc. Equine Pract., 31st, Toronto, pp. 383-391. Crawford, M. A, Kittleson, M. D. and Fink, G. D. (1984). J. Am. Vet. Med. Assoc. 184, 818-821. Crow, S. E. and Stockham, S. L. (1985). J. Am. Anim. Hasp. Assoc, 21,393-400. Dahms, H. Rock, R., and Seligson, D. (1968). Ciill. Chem. 14, 859-870. Dalton, R G. (1964). Br. Vet. J. 120, 378-384. Degan, M. A. (1986). Vet. Ciill. Path. 15, 20-22. De Morais, H. S. A. (1992a). III "Fluid Therapy in Small Animal Practice" (S. P. DiBartola ed), pp. 276-296. Saunders, Philadel­ phia, Pennsylvania. De Morais, H. S. A. (1992b). III "Fluid Therapy in Small Animal Practice" (S. P. DiBartola ed), pp 297-320. Saunders, Philadel­ phia, Pennsylvania. Dhein, C R and Wardrop, K. J. (1995). J. Am. Vet. Med. Assoc. 206, 1565-1566. DiBartola, S. P. (1992a). III "Fluid Therapy in Small Animal Practice" (S. P. DiBartola ed), pp. 193-215. Saunders, Philadelphia, Penn­ sylvania. DiBartola, S. P. (1992b). III "Fluid Therapy in Small Animal Practice" (S. P. DiBartola ed), pp. 216-243. Saunders, Philadelphia, Penn­ sylvania. DiBartola, S. P. (1992c). In "Fluid Therapy in Small Animal Practice" (S. P. DiBartola ed), pp. 244-257. Saunders, Philadelphia, Penn­ sylvania. DiBartola, S. P., and De Morais, H. S. A. (1992). III "Fluid Therapy in Small Animal Practice" (S. P. DiBaratola, ed.), pp. 258-275. Saunders, Philadelphia, Pennsylvania. Dieterich, R A, and Holleman, D. F. (1973). Call J. Zool. 51, 867-873. Divers, T. J., Freeman, D. E., Ziemer, E. L., and Becht, J. L. (1986). Proc. Amlu. Meet. Am. Assoc. Equille Pract., 321ld, Nashville, Ten­ nessee, pp. 69-80. Dow, S. W., LeCouteur, R A, Fettman, M. J., and Spurgeon, T. L. (1987a). J. Am. Vet. Med. Assoc. 191, 1563-1568. Dow, S. W., Fettman, M. J., LeCouteur, R A, and Hamar, D. W. J. (1987b). J. Am. Vet. Med. Assoc. 191, 1569-1575. Earley, L. E., and Daugharty, T. M. (1969). N. Ellgl. J. Med. 281, 72-86. Edelman, I. S., and Leibman, J. (1959). Am. J. Med. 27, 256-277. Edelman, I. S., Leibman, J., O'Meara, P., and Birkenfeld, L. W. (1958). J. Clill. Illvest. 37, 1236-1256. Eiseman, A J., MacKenzie, L. B., and Peters, J. P. (1936). J. BioI. Chelll. 116, 35-45. Elkinton, J. R, and Danowski, T. S. (1955). "The Body Fluids. Basic

Physiology and Practical Therapeutics." Williams and Wilkins, Baltimore. pp. 68-111. Elkinton, J. R, and Taffel, M. (1942). J. Clin. Invest. 21, 787-794. Elkinton, J. R, and Winkler, A W. (1944). J. Clill. Illvest. 23, 93-101. Emmett, M. and Nairns, R G. (1977). Medicine 56, 38-54. English, P. B. (1966a). Res. Vet. Sci. 7, 233-257. English, P. B. (1966b). Res. Vet. Sci. 7, 258-275. Epstein, V. (1984). Equille Vet. J. 16, 453-456. Evans, J. W. (1972). J. Allim. Sci. 33, 1001-1004. Feldman, B. F., and Rosenberg, D. P. (1981). J. Am. Vet. Med. Assoc. 178, 396-398. Fencil, V. and Leith D. H. (1993) Resp. Physiol. 91, 1-16. Fencil, V., and Rossing, T. H. (1989). Alln. Rev. Med. 40, 17-29. Fettman, M. J., Brooks, P. A, Burrows, K. P., and Phillips, R W. (1986). J. Am. Vet. Med. Assoc. 188, 397-401. Fisher, E. W., and Martinez, A. A (1976). Res. Vet. Sci. 20, 302-305. Fleming, S. A, Hunt, E. L., Brownie, C, Rakes, A, and McDaniel, B. (1992). Am. J. Vet. Res. 53, 222-224. Fonnesbeck, P. V. (1968). J. Allim. Sci. 27, 1350-1356. Fredeen, A H., DePeters, E. J., and Baldwin, R L. (1988). J. Allim. Sci. 66, 174-84. Frischrneyer, K. J., and Moon, P.F. (1994). Am. J. Vet. Res. 55, 11531157. Gamble, J. L., and Robertson, J. S. (1959). Am. J. Physiol. 179, 659-667. Garella, S., Chang, B. S., and Kahn, S. I. (1975). Kidlley lilt. 8, 279-283. Garry, F., and Rings, D. M. (1987). Pmc. Amlll. Vet. Med. FaYUm, 5th, Washington, D.C, pp. 823-826. Genetzky, R M., Loparco, F. V., and Ledet, A E. (1987). Am. J. Vet. Res. 48, 1007-1011. Gennari, F. J. (1984). N. Ellgl. J. Med. 310, 102-105. George, J. W. (1994). III "Veterinary Laboratory Medicine Clinical Pathology" 0. R Duncan, K. W. Prasse, and E. A. Mahaffey, eds.), 3rd ed., pp. 94-111. Iowa State University Press, Ames. Giger, U. and Gorman, N.T. (1984). Compelld. COllt. Ed. Pract. Vet. 6, 805-812. Gilfix B. M., Bique, M., and Magder, S. (1993), J. Crit. Care 8, 187-197. Gingerich, D. A, and Murdick, P. W. (1975a). Am. J. Vet. Res. 36, 663-668. Gingerich, D. A, and Murdick, P. W. (1975b). J. Am. Vet. Med. Assoc. 166, 227-230. Glazier, D. B., Littledike, E. T., and Evans, R D. (1979). J. Equille Med. Sztrg. 3, 489-494. Glazier, D. B., Littledike, E. T., and Evans, R. D. (1982). Am. J. Vet. Res. 43, 1934-1937. Gossett, K. A, and French, D. D. (1983). Am. J. Vet. Res. 44, 1744-1745. Hankes, G. H., Nelson, A. W., and Swan, H. (1973). Am. J. Vet. Res. 34, 221-229. Hannon J. P., Bossone, C A, and Wade, C E. (1990). Lab. Allim. Sci. 40, 293-299. Hansard, S. L. (1964). Am. J. Physiol. 206, 1369-1372. Harris, P., and Snow, D. H. (1986). J. Physiol. 376, 46P. Harris, R C, Harman, J. C, Marlin, D. J. and Snow, D. H. (1987). In "Equine Exercise Physiology 2" ( J.R Gillespie and N.R Robinson, eds.),pp. 464-475. ICEEP Publications, Davis, Cali­ fornia. Hayter, C J., Clapham, W. F., Mills, I. H., and Wardener, H. E. (1962). Clill. Sci. 23, 229-236. Hedley-Whyte, J., and Laver, M. B. (1964). J. Appl. Physiol. 19, 901. Hinton, M. (1978). Equine Vet. J. 10, 27-31. Hintz, H. F., and Schryver, H. F. (1976). J. Allim. Sci. 3, 637-643. Hix, E. L., Evans, L. E., and Underbjerg, G. K. L. (1953). J. Allim. Sci. 12, 459-473. Hix, E. L., Underbjerg, G. K. L., and Hughes, J. S. (1959). Am. J. Vet. Res. 20, 184-191.

Fluid, Electrolyte, and Acid-Base Balance

Hopper, L Elkinton, J. R., and Winkler, A W. (1959). I. c/in. Invest. 23, 111-117. Hoskins, J. D., and Rothschmitt, J. (1984). Vet. Med. 79, 489-491Houston, D. M., Allen, D. G., and Kruth, S. A(1989). Can Vet. J. 30, 423-425. Humes, H. D. (1984). In "Fluids and Electrolytes" 0. P. Kokko and R. 1. Tannen, eds.), pp. 119-149. Saunders, Philadelphia, Pennsyl­ vania. Ilkiw J. E., Rose, R. L and Martin L C. A (1991). J. Vet. Int. Med. 5, 294-298. Inagami, T. (1994). J c/ill. Pharmacol. 34, 424-426. Irvine, R. O. H., and Dow, J. W. (1968). Aust. Ann. Med. 17, 206-213. Jezyk, P. F. (1982). I. Am. Anim. Hosp. Assoc. 18, 977-980. Jones, N. L. (1990). c/in. Biochem. 23, 189-195. Judson, G. J., and Mooney, G. J. (1983). In "Equine Exercise Physiol­ ogy" (D. H. Snow, S. G. B. Persson, and R. J. Rose, eds.), Granta Editions, Cambridge, England. Kamal, T. H., Shehata, 0., and Elbanna, I. M. (1972). In "Isotope Studies on the Physiology of Domestic Animals." Proc. Symp. Inl. Atomic EnergJ) Agency, Vienna, pp. 95-102. Kami, G., Merritt, A M., and Duelly, P. (1984). Equine Vet. I. 16, 356-358. Kasari, T. R., and Naylor, J. M. (1984). Can. Vet. I. 25, 394-399. Kasari, T.R., and Naylor, J. M. (1986). Can. J. Vet. Res. 50, 502-508. Kohn, C. W. (1979). Vet. c/in. N. Am. (Large Anim. Pract.) 1, 289-311Kohn, C. W. and DiBartola, S.P. (1992). In "Fluid Therapy in Small Animal Practice" (S. P. DiBaratola, ed.), pp. 1-34. Saunders, Phila­ delphia, Pennsylvania. Koterba, A., and Carlson, G. P. (1982). I. Am. Vet. Med. Assoc. 180, 303-306. Kowalchuk, J. M., and Scheuermann, B. W. (1993). Can. I. Physiol. Pharmacol. 72, 818-826. Kryzwanek, H. (1974). I.S. Afr. Vet. Assoc. 45, 355-360. Kryzwanek, H., Milne, D. W., Gabel, A. A, and Smith, 1. G. (1976). Am. J. Vet. Res. 37, 291-294. Kurtz, T. W., AI-Bander, H. A, and Morris, R. C. (1987). New Engl. I. Med. 317, 1043-1048. Lakritz, L Madigan, L and Carlson, G. P. (1992). I. Am. Vet. Med. Assoc. 200, 1114-1116. Landis, E. M., and Pappenheimer, J. P. (1963). In "Handbook of Physiology" (W. F. Hamilton, ed.), Sect. 2, Vol. II, pp. 961-1034, Am. Physiol. Soc. Williams and Wilkins, Baltimore. Leaf, A. (1962). N. Eng!. I. Med. 267, 77-83. Leitch, L, and Thomson, M. A (1944). Nutr. Abstr. Rev. 14, 197-223. Littlejohn, A, and Mitchell, B. (1969). Res. Vet. Sci. 10, 260-262. McAuliffe, J. J., Lind, L. J., Leith, D. E., and Fencl, V. (1986). Am. I. Med. 81, 86-90. McCance, R. A. (1937). BiocJzem. J. 31, 1278-1284. McCance, R. A. (1938). I. Physiol. 92, 208-218. McCoy, C. P., and Edwards, W. C. (1979). Oklahoma Vet. 31, 12-14. McDougall, J. G., Coghlan, J. P., Scoggins, B. A., and Wright, R. D. (1974). Am. J. Vet. Res. 35, 923-929. McGuirk, S. M., and Butler, D. G. (1980). J. Am. Vet. Med. Assoc. 177, 551-554. McKeever, K. H., Hinchcliff, K. W., Schmall, 1. M., Lamb, D. R. and Muir, W. W. (1991). In "Equine Exercise Physiology 3" (S. G. B. Persson, A. Lindholm, and 1. B. Jeffcott, eds.), pp. 368-373. ICEEP Publications, Davis, California. McKeown, J. W. (1984). In "Fluids and Electrolytes" 0. P. Kokko and R .1. Tannen, eds.), pp. 63-117. Saunders, Philadelphia, Penn­ sylvania. Madias, N. E., Ayus, J. c., and Adrogue, H.J. (1979). N. Eng/. J. Med. 300, 1421-1423.

515

Madigan, J. E., Thomas, W.P., Backus, K.Q., and Powell, W.E. (1992).

Equine Vet. I. 24, 399-401Man, S. 0., and Carroll, H. J. (1977). N. Engl. I. Med. 297, 814-817. Mandell, C. P., Degen, M. A, Hopper, P. E., and Zinkl, J. G. (1988). Vet. c/ill. Path. 17, 32-33. Mason, D. E., Muir, W. W., and Wade, A (1987). J. Am. Vet. Med.

Assoc. 190, 989-994. Mattheeuws, D. R. G., Kaneko, J. J., Loy, R. G., Cornelius, C. E., and Wheat, J. D. (1966). Am. I. Vet. Res. 27, 699-705. Merritt, A. M., Kohn, C. W., Ramberg, C. F., Cimprich, R. E., Reid, C. F., and Bolton, J. R. (1977). Am. J. Vet. Res. 38, 1769-1774. Michell, A R. (1974). B1". Vet. J. 130, vi-viii. Michell, A R. (1985). Vet. Rec. 116, 653-657. Milne, D. W., Skarda, R. T., Gabel, A. A., Smith, L. G., and Ault, B. S. (1976). Am. I. Vet. Res. 37, 285-290. Morris, D. D., Divers, T. J., and Whitlock, R. H. (1984). Am. I. Vet. Res. 45, 2431-435. Muylle, E., Van den Hende, c., Nuytten, J., Deprez, P., Vlaminck, K., and Oyaert, W. (1984a). Equine Vet. J. 16, 447-449. Muylle, E., Nuytten, J., Van den Hende, c., Vlaminck, K., and Oyaert, W. (1984b). Equine Vet. J. 16, 450-451Nairns, R. G., and Emmett, M. (1980). Medicine 59, 161-187. Padovan, D. (1980). Cornell Vet. 70, 153-159. Patrick, J. (1977). Kidney Int. 11, 476-490. Pearson, E. G., and Kallfelz, F. A (1982). Cornell Vet. 72, 142-149. Persson, S. (1967). Acta Vet. Scmzd. Supp/. 19, 1-189. Persson, S. G. B., Ekman, 1., Lydin, G., and Tufvesson, G. (1973). Zentmlbl. Veterinaermed. Reihe A. 20, 441-455. Phillips, R. W., and Knox, K. 1. (1969). I. Compo Lab. Med. 3, 1-3. Phillips, R. W., Lewis, L. D., and Knox, K. L. (1971). Ann. N. Y. Acad. Sci. 176, 231-243. Phillipson, A. T. (1977). In "Duke's Physiology of Domestic Animals" (M. J. Swenson, ed.), 9th ed., pp. 250-286. Comstock Publishing Co., Ithaca, New York. Pickar, J. G., Spier, S. L Snyder, J. R., and Carlsen, R. C. (1991). Am. I. Physiol. 260, C926-C933. Pierce, A. W. (1957). Vet. Rev. Annot. 3, 37-43. Polzin, D. J., Stevens, J. B., and Osborne, C. A (1982). Compo Cont. Ed. Pract. Vet. 4, 1021-1036. Pownall, R., and Dalton, R. G. (1973). BI". Vet. J. 129, 583-588. Reeves et al., (1982). J. App/. Plzysiol. 53, 87-95. Robb, J., Harper, H. F., Hintz, H. F., Reid, J. T., Lowe, J. E., Schryver, H. F., and Rhee, M. S. S. (1972). Anim. Prod. 14, 25-34. Rose, B. D. (1984). In "Clinical Physiology of Acid-Base and Electro­ lyte Disorders," 2nd ed. McGraw-Hill, New York. Rose, R. J., and Carter, J. (1979). J. Vet. Pharmacol. Ther. 2, 215-221. Rose, R. L Carter, R. J., and Ilkiw, J. E. (1980). J. Vet. Pharmacol. Ther. 3, 111-119. Rose, R. L Gibson, K. T., and Suann, C. J. (1986). Vet. Rec. 19, 522-525. Rossing, R. G., and Cain, S. M. (1966). J. Appl. Physiol. 21, 195-201Rudolph, J. A, Spier, S. J., Byrns, G., Rojas, C. V., Bernoco, D. and Hoffman, E. P. (1992a) Nat. Genet. 2, 144-147. Rudolph, J. A, Spier, S. J., Byrns, G. and Hoffman, E. P. (1992b) Anim. Genet. 23, 241-250. Rumbaugh, G. E., Carlson, G. P., and Harrold, D. R. (1981). I. Am. Vet. Med. Assoc. 178, 267-271Rumbaugh, G. E., Carlson, G. P., and Harrold, D. (1982). Am. J. Vet. Res. 43, 735-737. Rumsey, T. S., and Bond, J. (1976). J. Amin. Sci. 42, 1227-1238. Sandals, W. C. D. (1978). Call. Vet. J. 19, 136-137. Saxton, D. R., and Seldin, D. W. (1986). In "Fluids and Electrolytes" 0. P. Kokko and R. 1. Tannen, eds.), pp. 3-62. Saunders, Philadel­ phia, Pennsylvania.

516

Gary P. Carlson

Schultze, G., Kirsch, K, and Rocker, 1. (1972). Pfliigers Arch. 337, 351-366. Scribner, B. H. (1969). In "Teaching Syllabus for the Course on Fluid and Electrolyte Balance." School of Medicine, University of Washington, Seattle. Senay, 1. C. (1970). J. Physiol. (London) 210, 617-635. Shull, R (1978). Vet. c/in. Pathol. 7, 7-14. Shull, R (1981). J. Am. Vet. Med. Assoc. 178, 1119-1123. Singer, R B., and Hastings, A. B. (1948). Medicine 27, 223-243. Sjogaard, G., Adams, R P., and Saltin, B. (1985). Alii. J. Physiol. 248, R190-R196. Speirs, V. C. (1980). Alii. J. Vet. Res. 41, 199-203. Spensley, M. S., Carlson, G. P., and Harrold, D. (1987). Am. J. Vet. Res. 48, 1703-1707. Spier, S. J., Carlson, G. P., Holliday, T. A., Cardinet, G. H. D., and Pickar, J. G. (1990). J. Alii. Vet. Med. Assoc. 197, 1009-1017. Spier, S. J., Carlson, G. P., Harrold, D., Bowling, A., Byrns, G., and Bernoco, D. (1993). J. Alii. Vet. Med. Assoc. 202, 933-937. Spurlock, G. H., Landry, S. 1., Sams, R, McGuirk, S., and Muir, W. W. (1985). Am. J. Vet. Res. 46, 1117-1120. Steffey, E. P., Wheat, J. D., Meagher, D. M., Norrie, R D., Mckee, J. Brown, M., and Arnold, J. (1977). Alii. J. Vet. Res. 38, 379-382. Steiss, J. E., and Naylor, J. M. (1986). Can. Vet. J. 27, 332-335. Stems, R H., Riggs, J. E., and Schochet, S. S. (1986). N. Engl. J. Med. 314, 1535-1542. Stewart, P. A. (1981). In "How to Understand Acid-Base: A Quantita­ tive Acid-Base Primer for Biology and Medicine." Elsevier North Holland, New York. Stewart, P. A. (1983). Can. J. Physiol. Pharmacal. 61, 1444-146I. Stick, J. A., Robinson, N. E., and Krehbiel, J. D. (1981). Alii. J. Vet. Res. 42, 733-737. Strombeck, D. R (1979). In "Small Animal Gastroenterology." Stonegate Publishing, Davis, California. Sufit, E., Houpt, K A., and Sweeting, M. (1985). Equine Vet. J. 17, 12-16. Tannen, R 1. (1984). In "Fluids and Electrolytes." ( J. P. Kokko and R 1. Tannen, eds.), pp. 150-228. Saunders, Philadelphia, Pennsyl­ vania. Tasker, J. B. (1967a). Cornell Vet. 57, 649-657. Tasker, J. B. (1967b). Cornell Vet. 57, 658-667.

Tasker, J. B. (1967c). Cornell Vet. 57, 668-677. Tasker, J. B. (1980). In "Clinical Biochemistry of Domestic Animals" ( J. J. Kaneko, ed.), 2nd ed., pp. 402-447. Academic Press, New York. Thornton, J. R, and English, P. B. (1977). Res. Vet. Sci. 22, 298-302. Tollertz, G. (1964). Acta. Vet. Scand. 5, 24-34. Tremblay, R R M., Butler, D. G., Allen, J. W., and Hoffman, A. M. (1991). Can. Vet. J. 32, 308-310. Tyler, R D., Qualls, c. W., Heald, R D., Cowell, R 1., and Clinkenbeard, K D. (1987). /. Alii. Vet. Med. Assoc. 191, 1095-1097. van Beaumont, W., Strand, J. c., Petrofsky, J. 5., Hipskind, S. G., and Greenleaf, J. E. (1973). J. Appl. Physiol. 34, 102-106. van Sluijs, F. J., de Vries, H. W., De Bruijne, J. J., and van den Brom, W. E. (1983). Am. J. Vet. Res. 44, 459-462. Wade, 1., and Sasser, 1. B. (1970). Am. J. Vet. Res. 31, 1375-1378. Weldon, A. D., Moise, N. S. and Rebhun, W. C. (1992) J. Vet. Int. Med. 6, 294-297. Whitehair, K J., Haskins, S. c., Whitehair, J. G., and Pascoe, P. J. (1995) J. Vet. Int. Med. 9, I-II. Whitlock, R H., Kessler, M. J., and Tasker, J. B. (1975a). Cornell Vet. 65, 512-526. Whitlock, R H., Tasker, J. B., and Tennant, B. (1975b). Dig. Dis. Sci. 20, 595-596. Williams, M. E., Gervino, E. V., Rosa, R M., Landsberg, 1., Young, J. B., Silva, P., and Epstein, F. H. (1985). N. Engl. J. Med. 312, 823-827. Wilson, E. A., and Green, R A. (1985). Camp. Cant. Ed. Pract. Vet. 7, S364-S37I. Winkler, A. W., Elkington, J. R, Hopper, J., and Hoff, H. (1944). J. c/in. Invest. 23, 103-109. Wolf, A. V. (1966). In "Aqueous Solutions and Body Fluids: Their Concentrative Properties and Conversion Tables." Harper & Row, New York. Yoshida, S., Sudo, H., Noro, K, and Nokaya, C. (1967). Exp. Rep. Equine Hlth. Lab. 4, 30-36. Ziemer, E. 1., Parker, H. R, Carlson, G. P., Smith, B. P., and Ishizaki, G. (1987a). J. Am. Vet. Med. Assoc. 190, 289-293. Ziemer, E. 1., Parker, H. R, Carlson, G. P., and Smith, B. P. (1987b). J. Alii. Vet. Med. Assoc. 190, 294-296. Zweens, J., Frankena, H., Rispens, P., and Zijlstra, W. G. (1975). Pflugers Arch. 357, 275-290.

C H A P T E R

19 Pituitary Function JAN A . MOL AND AD RIJNBERK

1.

The neurosecretory pathway runs from the anterior hypothalamus, traverses the floor of the ventral hypo­ thalamus, and terminates in the neural lobe (NL) of the neurohypophysis on fenestrated blood vessels (Page, 1986). The system is involved in osmoregulation through the production and release of vasopressin, and in parturition and nursing through the secretion of oxytocin. In the intermediate lobe (lL) the secretory activity is regulated via direct neuronal inhibitory and stimula­ tory influences (Tilders et al., 1 985). In amphibians (Verburg-van Kemenade et al., 1986) and reptiles (Dores et aI., 1987) the IL plays an important role in adaptation to background color. The function of IL cells in mammals has not been fully established, but they may play a role in opioid-regulated functions.

HYPOTHALAMUS-PITUITARY SYSTEM 517 A. Anatomical Considerations 517 B. Regulation of Pituitary Functions 521 II. ANTERIOR LOBE AND INTERMEDIATE LOBE 523 A. Proopiomelanocortin-Derived Peptides 523 B. Glycoprotein Hormones 529 C. Somatomammotropic Hormones 533 III. NEUROHYPOPHYSIS 541 A. Vasopressin 541 B. Oxytocin 543 IV. ASSESSMENT OF PITUITARY FUNCTION 545 A. Adenohypophysis 545 B. Neurohypophysis 546 References 546

I. HYPOTHALAMUS-PITUITARY SYSTEM

A. Anatomical Considerations

The hypothalamus-pituitary system is a preemi­ nent example of integration of neural and endocrine control. It consists of three major systems: (1 ) a neuro­ endocrine system connected to an endocrine system by a portal circulation, (2) a neurosecretory pathway, and (3) a direct neural regulation of endocrine secretion (Fig. 19.1). The neuroendocrine system involves clusters of peptide- and monoamine-secreting cells in the anterior and midportion of the ventral hypothalamus. Their products reach the median eminence by axonal trans­ port. From there they are released into the capillary vessels of the hypothalamus-pituitary portal system and transported to the pituitary to regulate the secre­ tion of hormones from the anterior lobe (AL) of the adenohypophysis.

CLINICAL BIOCHEMISTRY OF DOMESTIC ANIMALS, FIFTH EDITION

The hypothalamus and pituitary control vital func­ tions such as growth, reproduction, lactation, basal metabolism, stress response, parameters of immune function, and the state of hydration. Understanding of the complicated functional relationship of the hypo­ thalamus to the pituitary requires an appreciation of the anatomical relationships.

1. Hypothalamus Hypophysiotropic neurohormones are produced in several areas of the hypothalamus. For example, in an immunofluorescence study of hypothalami of dogs, the majority of cell bodies with immunoreactivity to corticotropin-releasing hormone (CRH) were found in

517

Copyright © 1997 by Academic Press All rights of reproduction in any form reserved.

518

Jan A. Mol and Ad Rijnberk

h) poph)�IOUOPI hormonc�

{

GnRH

CRII

OthC'IS

p'IU".') """"",,,, '

ACTII Gil PRL

II LlI I- H

oMSH

\,p

or

FIGURE 19.1 Schematic representation of the relationship be­ tween the hypothalamus and the pituitary in a generalized mammal. The hypothalamus exerts control over the anterior lobe (AL) by releasing and inhibiting factors. These hormones reach the AL cells via capillaries of the pituitary portal system. The neural lobe (NL) of the pituitary is a downward projection of the hypothalamus. The intermediate lobe (IL) is under direct neurotransmitter control.

the region of the periventricular and para ventricular nuclei, but they were also found in the supraoptic and suprachiasmatic area as well as craniodorsal to the mamillary bodies (Stolp et ai., 1987). The cell bodies of the neurohormone-producing neurons that project to the median eminence are in part intermingled with cell bodies that also synthesize these neurohormones but project to other brain areas. The majority of the neurons that project to the median eminence are found in the preoptic and suprachias­ matic region of the hypothalamus. Axons containing the same neurohormone may have synaptic contacts that enable regulation of cellular function between these neurons. The neurohormone-producing cells receive a com­ plex neural input from a variety of chemical messen­ gers, such as neurotransmitters and other neurohor­ mones. Not only the neurohormones CRH and arginine vasopressin (AVP), but in general the combi­ nation of a neurohormone and another chemical mes­ senger, may colocalize within a single neuron. The neurons of the neurohypophyseal system repre­ sent a more anatomically distinct entity, with cell bod­ ies located in the paraventricular and supraoptic hypo­ thalamic nuclei (Sawchenko and Swanson, 1983). However, within these areas there are also neurons producing a variety of other neuropeptides. 2. Neurotransmitter Systems

The major neurotransmitter systems for intercellular communication within the central nervous system con­ sist of monoamines and peptides. These chemical mes-

sengers regulate the biosynthesis and release of the hypophysiotropic neurohormones. Through a network of axodendritic and axoaxonic contacts, these neurons are connected to the neurohormone-producing cells. In addition, many monoamines and peptides are found within the hypophysiotropic hormone-producing cells, where they are released together with the neuro­ hormones into the portal system and modify the effect of hypothalamic hormones on the pituitary. The biogenic amine neurotransmitters, known to play a regulating/ modulating role in the hypothala­ mus-pituitary system, include catecholamines (dopa­ mine, noradrenalin, adrenalin), indolamines (seroto­ nin, melatonin), acetylcholine, y-aminobutyric acid (GABA), and histamine. Neuropharmacologic agents can be used to alter neurotransmitter effects and, as a consequence, hypothalamic and pituitary hormone re­ lease. Many of the peptides with potential effects on hypo­ physiotropic hormone release are widely distributed in hypothalamic and extrahypothalamic areas of the brain. They include, among many others, peptides common to the gastrointestinal tract, such as gastrin, cholecystokinin, and pancreatic polypeptide, as well as bombesin, angiotensin II, galanin, substance P, neu­ rotensin, enkephalins, neuropeptide Y, natriuretic pep­ tide, and, from a common precursor, the vasoactive intestinal peptide (VIP) and the peptide histidine iso­ leucine (PHI). The last three peptides may, through vasoconstriction and vasodilatation activities, play an important role in the control of the portal blood flow.

3. Vascular System The releasing and inhibiting hormones are stored in nerve terminals in the median eminence, where their concentrations are 10 to 100 times as great as elsewhere in the hypothalamus. The uniquely organiz�d capi�la�y plexus (see Page, 1986) of the median emmence Is m . close proximity to nerve terminals of the hypophyslO­ tropic neurons. In contrast to other brain regions, the blood-brain barrier in the area of the median eminence is incomplete, permitting protein and peptide hor­ mones as well as other charged particles to move to the intercapillary spaces and the nerve terminals contained therein. These terminals respond to humoral and neu­ ronal stimuli by secreting releasing and inhibiting fac­ tors into the portal system. The portal capillaries coalesce into a series of vessels that descend through the pituitary stalk and form a second capillary plexus that surrounds the AL cells (Fig. 19.1). Inferior hypophyseal arteries supply the neurohy­ pophysiS. From the primary plexus of the neural lobe

Pituitary Function

(NL), blood flows not only to the systemic circulation but also to the AL and the hypothalamus. There is evidence for some degree of circulatory flow within the pituitary, that is, from the AL to the NL, from there to the infundibulum, and then back to the AL. The primary capillary plexus of the NL appears to be well positioned in the mini-circulatory system, controlling all of the afferent vascular events and many of the efferent vascular events in the pituitary (Page, 1986). The vascularization of the intermediate lobe is closely linked to that of the neurohypophysis. How­ ever, in spite of the rich blood supply of the NL, the IL is a poorly vascularized structure.

4. Pituitary During embryogenesis the adenohypophysis devel­ ops from Rathke's pouch, which arises from the primi­ tive roof of the mouth in contact with the base of the brain. Rathke's pouch subsequently separates by con­ striction from the oral cavity. The anterior wall thickens and forms the anterior lobe of the adenohypophysis. This largest portion of the adenohypophysis remains separated from the intermediate lobe by the hypophy­ seal cleft, which is the residual lumen of Rathke' s pouch. In several species (Fig. 19.2) the adenohypophy­ sis also extends into a pars tuberalis that forms a cuff or collar around the proximal neurohypophysis and may even envelop part of the median eminence (Hul­ linger, 1979; Batten and Ingleton, 1987). The posterior wall of Rathke' s pouch is closely op­ posed to the neural tissue of the NL, thereby forming

Man

Cow

Goat

Dog

519

the intermediate lobe (IL), which is well developed in most mammals, but not in man and birds (Fig. 19.2). In man only during fetal life is a distinct IL found (Visser and Swaab, 1979), whereas there is some debate about the view that in adults "invading" cells of the posterior lobe are homologous with the IL cells of lower vertebrates (Coates et al., 1986; McNicol, 1986; Lamberts, 1987). The pituitary stalk (infundibulum) and the neurohy­ pophysis (posterior lobe, neural lobe) develop from the basal outgrowth of the diencephalon in connection with the development of Rathke' s pouch. The cells of the diencephalic outgrowth later develop into glial cells (pituicytes), whereas nerve fibers from the supra­ optic and paraventricular nuclei grow into the NL.

5. Cells of the Anterior Lobe The peptide hormones secreted by the AL can be divided into three categories: (1) the somatomammo­ tropic hormones growth hormone (GH) and prolactin (PRL), (2) the glycoprotein hormones thyrotropin (TSH), follicle-stimulating hormone (FSH), and lutein­ izing hormone (LH), and (3) the corticomelanotropins, which include a-melanotropin (MSH), adrenocorti­ cotropin (ACTH), /3-endorphin (/3-END), and /3lipotropin (/3-LPH). Over the years identification of the adenohypophy­ seal cells has been developed by histochemical staining techniques, immunocytological methods, and ultra­ structural studies. Staining and immune reactions de-

Horse

Cat

FIGURE 19.2 Schematic illustration of median sections through mammalian and avian pituitaries. Key D , anterior lobe; ., intermediate lobe; EJ neural lobe. Redrawn from Seiferle (1975) and Batten and Ingleton (1987).

Pig

Bird

520

Jan A. Mol and Ad Rijnberk

pend on the chemical nature of the hormones that are stored in granules within the cytoplasm. Special stains allow histochemical identification of particular cell types. A general identification of adenohypophyseal cells, as derived from Baker (1974), Bentley (1976), and Batten and Ingleton (1987), is given in Table 19.1, al­ though it must be realized that the various cell types do not always give exactly the same staining reaction in different animal species. As far as ultrastructural features are concerned, the adenohypohyseal cells have the typical cytoplasmic organelles for peptide synthesis and release. Character­ istics such as the form and location of the various organelles and the size and shape of the cells allow identification of adenohypophyseal cells at the ultra­ structural level (Mikami, 1986; Batten and Ingleton, 1987). In the embryOnic development of the cells of the AL, two main lineages arise from a progenitor cell, the acidophilic (A) cells that result in the mammasomato­ tropic cells and the basophilic (B) cells that differentiate into gonadotrope, thyrotrope, and corticotrope. Dur­ ing the fetal development of the mouse AL the expres­ sion of the a-subunit of the glycoproteins is among the first that becomes apparent. Then hormone biosynthe­ sis is found in the corticotropes and then in the gonado­ tropes. The differentiation of the thyrotropes and so­ matomammotropes depends on the expression of a specific transcription factor called Pit-l or GHF-l. Mu­ tations in the Pit-l gene may result in the complete absence of TSH-, GH-, and PRL-producing cells in the pituitary. The concept of one type of pituitary cell producing one hormone now appears to be an oversimplification. Not only are the majority of gonadotropes and somato­ mammotropes multihormonal cells that contain LH TABLE 19.1

Cell type

Hormone

and FSH or GH and PRL, respectively, but other combi­ nations such as gonadotrope hormones and ACTH or TSH and ACTH have been described. As already has been discussed for the hypothalamus, the hormone­ producing AL cells, or the so-called "folliculostellate" cells, also may syntheSize a variety of chemical messen­ gers, intrapituitary growth factors, and cytokines that exert local paracrine effects on cell function and prolif­ eration (Denef, 1994).

6. Cells of the Intennediate Lobe The predominant IL cell in mammals is the melano­ trope, a cell with immunoreactivity for a-MSH, which is very sparse in the AL (Halmi and Krieger, 1983). In some species, including the dog (Halmi et al., 1981) and the horse (Amann et al., 1987), the IL is cytologically heterogeneous. ACTH-containing cells (B cells) have been found to be dispersed among the predominant melanotropes (A cells). In the dog the immunoreactive ACTH content of the IL even exceeds that in the AL (Halmi et al., 1981). In agreement with the previously mentioned direct neural regulation of IL secretory ac­ tivity, the presence of neural elements was demon­ strated in the bovine IL (Boyd, 1987). Evidence is accu­ mulating that the cells of the intermediate lobe are also involved in the biosynthesis and release of a yet­ unknown prolactin-releasing factor. The blood supply of the IL is very poor, and therefore the peptides re­ leased from the IL are thought to act mainly by diffu­ sion in a paracrine manner. 7. Cells of the Neural Lobe

The NL contains axonal nerve fibers, often swollen by being packed with neurosecretory granules. These nerve fibers and the glial cells (pituicytes) have synap-

Molecular Weights of Adenohypophyseal Hormones and Staining Properties of Adenohypophyseal Cells

MW'

Subunits

Staining

Somatotrope

GH

22,000

Acidophil

Lactotrope

PRL

22,500

Acidophil

Gonadotrope

LH FSH

30,000 32,000

a / (3 a / (3

Thyrotrope

TSH

28,000

a / (3

Corticotrope

ACTH (3-END

4,500 3,500

Basophil Basophil

Melanotrope

a-MSH

1,700

Basophil

Basophil Basophil Basophil

Molecular weight. Periodic acid Schiff reaction. C Color obtained after co-staining with PAS-Orange G and performic acid-Alcian blue.

a

b

Orange G

+ +

PASb

Color' Yellow

+ +

+ + + +

Yellow Blue Blue Blue Red Red Red

Pituitary Function

toid contacts. The pituicytes appear to play an interme� diary role in the regulation of the release of vasopressin and oxytocin (Van Leeuwen and De Vries, 1983). B. Regulation of Pituitary Functions

For the regulation of each of the five major adenohy� pophyseal hormone systems (ACTH, LH and FSH, T5H, GH, and PRL) there is a feedback (closed�loop) system. AL hormone and hypophysiotropic hormone secretion are suppressed by the products of target en� docrine glands such as the thyroid, adrenal, and gonad. Apart from this long�loop feedback, some hormones (e.g., PRL) regulate their own secretion directly by act� ing on the hypothalamus (short�loop feedback). On this powerful feedback control with primarily blood� borne signals, other signals are superimposed. These may originate within the central nervous system (open loop) and can be mediated through neurotransmitters and hypophysiotropic hormones. Thus, influences are exerted that represent the environment (temperature, light-dark), stress (pain, fear), and intrinsic rhyth� micity. These regulatory factors influence peptide synthesis and / or release in adenohypophyseal cells, where each of the steps in hormone synthesis and ultimate secre� tion represents a potential control point in the regula­ tion of circulating hormone levels.

1. Hypophysiotropic Hormones The main hypothalamic neurohormones may stimu­ late or inhibit the release of a single hormone, or affect several hormone-producing cells (Fig. 19.3). The pre­ dominant influence of the hypothalamic hormones on the pituitary is stimulatory. Interference with the integ­ rity of the hypothalamus-pituitary connections results in decreased secretion of pituitary hormones. The ex­ ception is PRL, the secretion of which is increased when hypothalamic influence is removed. Hypothalamus

Adenohy pophysis FIGURE 19.3

Hypophysiotropic regulation of the secretion of pituitary hormones. Solid lines denote hormones whose structures have been determined. Dashed lines indicate a factor whose identity is still unknown.

521

As the complexity of the peptide structure of the hypophysiotropic hormones increases, species varia­ tion in sequence may occur. Whereas the structures of TRH, GnRH, and 5S (3, 10, and 14 amino acids, respectively) are identical in all mammals, those of GHRH and CRH (44 and 41 amino acids) exhibit spe­ cies specificity. The one nonpeptide hypophysiotropic hormone is dopamine. In addition to its major role as a neurotransmitter, dopamine is the most important inhibitor of PRL.

2. Regulation of Gene Expression The developments in recombinant DNA technology have provided better understanding of the genes en­ coding pituitary hormones and the regulatory ele­ ments involved in gene transcription. The main ele­ ments regulating the gene transcription will be mentioned briefly (Fig. 19.4). Eukaryotic genes encod­ ing peptide hormones consist of a promoter (regula­ tory) unit and a transcription unit encoding the pri­ mary transcript that after appropriate processing will form the messenger RNA (mRNA). The promoter unit is the upstream (5') part of the gene. The promoter has specific DNA sequences (re­ sponse elements) permitting the binding of transcrip­ tion factors that enhance or inhibit gene expression by changes in the stability of the RNA polymerase-TATA box factors complex at the constitutive promoter. At a greater distance from the gene, specific enhancer elements may also regulate gene transcription. The highly specialized hormone production is also regu­ lated by tissue-specific silencers and enhancers that are mandatory for gene expression. Specific binding of hypothalamic neurohormones to membrane receptors of individual AL cells will result in changes in second messenger concentrations of, for instance, cyclic AMP or inositol triphosphate within the cell, a process called signal transduction. These changes will result in differences in phosphorylation of transcription factors that bind to a cAMP response element (CRE) or tetradecanoylphorbol acetate re­ sponse element (TRE) in the promotor unit. Complexes of steroid hormones, thyroid hormone, or the retinoids with their respective receptors may also bind to specific areas of the promoter. For example, a glucocorticoid response element (GRE) in the promoter of the gene encoding pro-opiomelanocortin binds the glucocorti­ coid-receptor complex, resulting in inhibition of gene transcription (negative GRE). Transcription is thus reg­ ulated by responses to extracellular signals, but tissue­ specific intracellular transcription factors also play a crucial role in determining gene expression.

522

Jan A. Mol and Ad Rijnberk

Exon 2

Exon I lntron

3'

---1-1+ 1--Transcription

hrurnA C:==-===C====

---- J+-_+t_ Post-transcriptional modification

mRNA c:=c:::==

®

2 (1

ij -------i-t-

..

Translation

preprohormonc prohormone

Second mcssengers(s) (cAMP. cOMPo RAS.

-----1+ Post-translational modification

Calcium. Diacylglycerol)

®

®

Secretion hormone

FIGURE 19.4 Schematic representation of potential control points in pituitary hormone synthesis and secretion.

After initiation of transcription a primary transcript is made, which contains an RNA copy of the entire transcription unit, the heteronuclear RNA (hnRNA). After excision of the intron areas, a process called splic­ ing, the formation of a 7-methylguanine cap at the 5' end, and the addition of a poly(A) tail at the 3' end, a mature mRNA is formed. Through alternative splicing reactions, length variants of the mRNA may ultimately result in variation of the coding sequence (see section on growth hormone) or changes in mRNA stability as found for the insulin-like growth factors (IGFs). Through differences in exon coupling, even completely different peptides can be obtained from a single gene, as in the case of the gene encoding calcitonin.

3. Prohormone Biosynthesis, Processing, and Release The process of peptide synthesis occurs principally on the rough endoplasmatic reticulum (rER). mRNA, encoded by nuclear DNA, passes to cytosolic ribo-

somes, and by sequential processing of transfer RNAs (tRNA) with their attached amino acids, the translation process of mRNA to a peptide starts. The beginning of the growing peptide forms a specific signal peptide that facilitates the attachment of the translation com­ plex to the rER and enables the passage into the lumen of the rER. The signal sequence of the pre-prohormone is cleaved and the remaining prohormone undergoes several modifications such as disulfide formation and glycosylation. The peptide passes along the rER lumen into the Golgi complex, where peptides are packaged and released into the cytoplasm as membrane-bound granules. During storage of these granules the prohor­ mone is further processed by specific proteolytic cleav­ age, C-terminal amidation, or N-terminal carboxyla­ tion. Characteristic proteolytic cleavage sites are pairs of the basic amino acids arginine and lysine. The gran­ ules are stored until the hormone is released by exo­ cytosis. This process involves fusion of the granule membrane with the cell membrane.

523

Pituitary Function

resulting in increased cAMP production within the cell ( Jacobson and Drouin, 1994). The sequence of CRH appears to be very homologous among species. The peptide sequence CRH in the dog is identical to that in the human, rat, and horse (Mol et a/., 1994), and there are only small differences in CRH in cattle and sheep. Glucocorticoids inhibit POMC gene expression via type II glucocorticoid receptors in the AL cortico­ trope. POMC expression in the IL melanotrope is also stim­ ulated by cAMP. In vitro experiments in various species point to a possibility that CRH activates this signal transduction system. However, chronic infusion of CRH stimulates the POMC mRNA concentration in the AL, but inhibits mRNA concentration in the IL (Hollt and Haarmann, 1984). The POMC gene expres­ sion in the IL is inhibited by dopaminergic and GABAergic compounds (Chronwall et al., 1987; Loef­ fler et al., 1986), but not by glucocorticoids.

II. ANTERIOR LOBE AND INTERMEDIATE LOBE A. Proopiomelanocortin-Derived Peptides The corticotropic cells of the AL and the melano­ tropic cells of the IL are both able to synthesize pro­ opiomelanocortin (POMC), the common prohormone for ACTH, a-MSH, ,B-LPH, and a family of ,B-END­ related peptides (Fig. 19.5).

1. ACTH/a-MSH a. Gene Expression

The gene encoding POMC contains three exon areas. Exon 1 encodes a 5' untranslated region (5'-UTR) and exon 2 encodes the signal peptide and the N-terminal part of the POMC prohormone. The majority of the prohormone, including y-MSH, is encoded by exon 3. After the splicing process a mRNA of approximately 1 100 nucleotides is formed in the pituitary of most species, including the dog (Mol et al., 1991). By alterna­ tive splicing of the intron A sequence a longer tran­ script is found in the hypothalamus. Ectopic human ACTH-producing tumors contain shorter POMC mRNA sequences ( Jacobson and Drouin, 1994). The expression of the POMC gene in corticotropes and melanotropes is regulated by the same promoter elements, as shown by experiments with transgenic mice. Synergistic stimulation of gene expression re­ quires binding of yet partially characterized transcrip­ tion factors in the distal and central part of the POMC promoter. Binding of corticotropin-releasing hormone mediates the activation of the corticotrope in the AL,

'Yl.1MSH 'Y,MSH

JP

b. (Pro)hormone POMC prohormone sequences are known for a vari­ ety of species and share a common structure (Fig. 19.5). Among species a high sequence homology is found between the N-terminal site (N-POC) of the prohor­ mone (including y-MSH), the ACTH region, and the l3-endorphin region. In contrast, the regions between y-MSH and ACTH and the first part of l3-lipotropin are rather heterogeneous (Numa and Imura, 1985; Mol et al., 1991). Species differences in the sequence of ACTH occur only in the N-terminal part (Fig. 19.6), whereas the ACTH(1-24) sequence, necessary for full biological activity, is identical among mammals.

MSH

CLIP

fi-MSH

ACTII fi-END

N-POC

fi-LPH

FIGURE 19.5 Schematic representation of preproopiomelanocortin (horizontal bar), with four main do­ mains: the signal peptide, which is cleaved after entrance to the lumen of the RER; the N-terminal peptide (N-POC), containing the (pro- )yMSH sequences and the joining peptide ( JP); the ACTH domain, from which MSH and corticotropin-like intermediate lobe peptide (CLIP) can be generated; and the .B-lipotropin (.B­ LPH) domain, including the endorphin (END) family of pep tides and a metenkephalin sequence (Enk). Pairs of basic amino acid residues are indicated with vertical lines, representing potential sites of proteolytic cleavage. In the anterior lobe, major cleavage products are N-POC, ACTH, and .B-LPH. In the intermediate lobe, the major products are N-POC, y-MSH, a-MSH, and .B-endorphin.

524

Jan A. Mol and Ad Rijnberk MSH

Dog Mink Pig Cattle Elephant Rabbi t Mouse Rat Human

--- ---------- -- --------- -- -- - - -- ---

-

--

--- -------

CLIP ---

--

---

- -

-- -----

- - -- -L L - ---Q - L -

VKVYPNGAEDESAEAFPVEF --L

- --

-----------

-

----

-

-

------

--

-v

------ - ------- ---------- G------- L------------------------------ - ------V--N-------L--------------- -------V --N -------L--------------- ----------- -------L-* * * * * * * * * * * * * * - * * * * * * * * * ** * * * * * *

39 39 39 39 39

39 39 39 39

FI�URE 1�.6 Comparison of the ACTH(1-39) sequence from vanous species using the one-letter code for amino acids. Identical �min? acids are given by a single line (-). The asterisk ( ) means an Identical sequence for all species. The shaded box indicates the two pairs of basic amino acids prone to proteolytic cleavage. *

The proteolytic enzymes PC1 / PC3 and PC2 are in­ volved in the processing of prohormones in neuroen­ docrine cells. They cleave the precursor at pairs of basic am�no acids, resulting in the formation of biologically actIve hormones. The PC1 / PC3 enzyme present in the AL cleaves POMC to ACTH and f3-LPH. PC2 generates the ACTH(l-13) fragment in both AL and IL. The com­ bination of both enzymes is present in the IL. C-Terminal amidation and N-terminal acetylation of ACT� (l-13) results in a-MSH. Various degrees of acetylatIon of MSH result in the storage of desacetyl-, monoa�etyl-, and diacetyl-a-MSH in secretory gran­ ules (Elpper and Mains, 1980; Krieger, 1980). The desa­ cetyl form is predominant in the AL. In the rat the IL contains diacetyl-a-MSH, whereas the canine IL con­ tains predominantly the monoacetyl-a-MSH (Young et al., 1992).

c. Secretion by the AL ACTH is released in frequent pulses, as demon­ strated in the pituitary venous effluent of the horse (Redekopp et ai., 1986a,b). By measurements in periph­ eral blood, Kemppainen and Sartin (1984) documented episodic secretion of ACTH in dogs, with an average of nine peaks per 24-hour period. The release of ACTH by the AL is influenced by many factors (Table 19.2). A number of these factors modulate the release of CRH and AVP, which are con­ sidered to be the predominant stimulating neurohor­ mones in vivo (Antoni, 1986; Keller-Wood and Dall­ man, 1984). The relative contribution of CRH and AVP to ACTH release varies among species and circum­ stances. In dogs (Van Wijk et al., 1994) and pigs (Minton and Parsons, 1993) both exogenously administered CRH and LVP induce comparably high plasma ACTH concentrations. In sheep, AVP is the predominant ACTH-releasing factor. In the rat the stimulation ' of ACTH release by CRH and AVP is synergistic, mean­ ing that the response to CRH + AVP is greater than

the sum of the reactions to CRH and AVP separately (Buckingham, 1987). The basal release of ACTH is regulated by the occu­ pancy of type I or mineralocorticoid receptors (MR) in the hippocampus ( Jacobson and Sapolsky, 1991). Decreases in brain MR binding capacity of the dog during aging results in enh anced basal activity of the hypothalamus-pituitary-adrenal axis (Rothuizen et ai., 1993). In sheep, basal ACTH release is inh ibited by active immunization against CRH (Guillaume et ai., 1992a) but not by immunization against AVP (Guil­ laume et ai., 1992b). Exercise and insulin-induced hypoglycemia stimu­ late ACTH release. In the horse, exercise-induced ACTH release is accompanied by increased AVP con­ centration in pituitary venous blood without changes in CRH concentrations (Alexander et al., 1991). In sheep, mild hypoglycemia is accompanied by increases in AVP and CRH concentrations in portal blood, but in severe hypoglycemia the AVP secretion is relatively much higher than the CRH release (Caraty et al., 1990). AVP also regulates the ACTH response on hypoglyce­ mia in the neonatal rat (Muret et ai., 1992). Hypotension induced by nitroprusside (Kemp­ painen and Sartin, 1987) or by hemorrhage (Lilly et al., 1983) causes ACTH release in the dog, together with large increases in plasma AVP derived from the NL (Raff et ai., 1988). Selective neurohypophysectomy re­ sults in greatly attenuated ACTH and AVP responses to hypotension and angiotensin II, whereas the ACTH response to CRH injection remains unchanged. By sub­ stitution with adequate AVP infusion, the ACTH re­ sponse to hypotension can be restored (Raff et al., 1992). In sheep, chronic absence of ovarian hormones after ovariectomy reduces the ACTH response to hypoten­ sion also, but not to CRH, AVP, or hypoglycemia (Pecins-Thompson and Keller-Wood, 1994). TABLE 19.2

Factors Modulati ng ACTH Release from the Adenohypophysis Stimulating

Inhibiting

Hormones

CRH Vasopressin Angiotensin II Cholecystokinin-8 TRH, GnRH (in the dog)

Glucocorticoids Enkephalin

Biogenic amines

(Nor)adrenalin Serotonin GABA

Dopamine

Others

Cytokines Hypoglycemia Hypoxia Stress (physical, emotional)

525

Pituitary Function TABLE 19.3

Species (n)

Hormone

Basal Concentrations of POMC-Derived Pep tides in Plasma of Healthy Animals"

Time of sampling (hours)

Range (pmollliter) (mean 4.4-22

Dog (31)

ACTH

Dog (15)

ACTH

08.00-14.00

Dog (160)

ACTH

08.00-10.00

2.2-19.8

Dog (19)

ACTH

08.00

(17.6 ± 2.2)

(7.0 ± 0.7)

(7.2 ± 0.6)

±

SEM)

Reference Feldman et al. (1977) Kemppainen et al. (1986) Peterson et al. (1986) Rijnberk et al. (1987) Kemppainen and Sartin (1986)

Dog (11)

a-MSH

08.00-14.00

Dog (160)

a-MSH

08.00-10.00

Dog (14)

a-MSH

08.00

(12.6 ± 1.9)

Rijnberk et al. (1987)

Dog (160)

,I3-END

08.00-10.00

1.5-17.4

Peterson et al. (1986)

1.5-15

(16 ± 3)

Peterson et al. (1986)

Willemse et al. (1993)

Cat (6)

ACTH

08.00-13.00

Cat (100)

ACTH

09.00

Cat (100)

a-MSH a-MSH

08.00-13.00 09.00

3.6-200

Peterson et al. (1994)

Cat (100)

,I3-END/ LPH

09.00

3.8-130

Peterson et al. (1994)

Horse

ACTH

Horse

a-MSH

Cat (6)

1.1-22

(77 ± 7)

(7 ± 1)

(14.4 ± 1.2)

Horse

,I3-END

(18.3 ± 4.4)

Horse

ACTH

3.5-15.0

Sheep

ACTH

Sheep

a-MSH

10.00 10.00

(116 ± 19) 0-4.2

Peterson et al. (1994) Willemse et al. (1993)

Orth et al. (1982) Orth et al. (1982) Orth et al. (1982) Hodson et al. (1986) Clarke et al. (1986) Clarke et al. (1986)

Most of the values have been converted to SI units. 1 ng ACTH /liter = 0.22 pmol/ liter; 1 ng a-MSH/ liter = 0.60 pmol/ liter; 1 ng ,13END/liter = 0.29 pmol/ liter. fl

In the dog the release of ACTH is stimulated by f3adrenergic agonists (isoproteronol), dopaminergic an­ tagonists (haloperidol), and serotoninergic agonists (quipazine maleate). TRH and GnRH stimulate the re­ lease of cortisol, probably by stimulating ACTH release (Stolp et al., 1982). Although direct stimulatory effects of catecholamines on the in vitro release of ACTH from ovine pituitary cells has been found, central stimula­ tion of predominantly noradrenergic, but also adrener­ gic, pathways evokes the highest ACTH response (Liu et al., 1991). Endogenous opiates (metenkephalin, dynorphin, and /3-endorphin) inhibit the release of ACTH in man (Besser et al., 1987). Conflicting results have been re­ ported on the effect of metenkephalin in the rat, but f3-endorphin and dynorphin may exert tonic inhibition of CRH release (Plotsky, 1986). The metenkephalin ag­ onist DAMME stimulates the release of ACTH in the dog (Meij et al., 1990). Activation of the immune system by infections results in the enh anced production of the cytokine interleukin-1 (IL-1/3), which has the ability to stimulate CRH secretion from the hypothalamus and thus acti­ vates the hypothalamus-pituitary-adrenal axis. The

biosynthesis and release of IL-6 has been found in the folliculostellate cells of the AL. IL-6 also stimulates the HPA-axis, at both the hypothalamic and the pituitary level (Sweep et al., 1991). Endogenous corticosteroids inhibit ACTH release predominantly at hypothalamic sites. Synthetic ste­ roids such as dexamethasone may act primarily at the pituitary level (De Kloet et .al., 1974). In a review on corticosteroid-mediated feedback, Keller-Wood and Dallman (1984) suggested three different time sched­ ules: a fast feedback that acts on the corticotropic cell but may not be related to nuclear receptor binding, an intermediate feedback that probably acts by inhibition of CRH release, and a slow feedback that acts by a decrease in mRNA encoding POMC in the pituitary gland. The delay in inhibiting ACTH production may be caused by the high stability of the mRNA encoding POMC and not by the absence of direct inhibition of the transcription. Dexamethasone inhibits gene tran­ scription in vivo within 30 minutes in the rat (Fremeau et al., 1986). In the dog the intermediate-delayed feed­ back is determined by the mean change in corticoste­ roid concentration over time (Keller-Wood, 1989). Acute lowering of plasma cortisol in the horse by inhi-

526

Jan A. Mol and Ad Rijnberk

bition of the synthesis in the adrenal gland resulted first in an increased ratio of ACTH : CRH in pituitary venous blood before CRH concentrations started to rise (Alexander et al., 1993), indicating that the first effect is the opposite of the fast feedback and is medi­ ated by increased sensitivity of the corticotrope.

d. Secretion by the IL

The release of POMC-derived peptides by the IL is under direct neural control. The rat and the mouse have been the mammals in which most studies on the regulation and processing of POMC in the IL have been carried out thus far. In the rat the release of POMC-derived pep tides is regulated predominantly via tonic dopaminergic inhibition and �-adrenergic stimulation (Berkenbosch et a/., 1981; Tilders et al., 1985), although GABAergic innervation of the IL has also been demonstrated (Oertel et a/., 1982). In addition, Proulx-Ferland et a/. (1982) demonstrated that CRH is a potent stimulator of a-MSH secretion by the 11. In line with the absence of a glucocorticoid receptor in the IL (Antakly et al., 1985), the a-MSH response to CRH could not be suppressed by dexamethasone ad­ ministration. The expression of the glucocorticoid re­ ceptor in the IL is suppressed by dopamine (Antakly et al., 1987), whereas the CRH receptor content of the rat IL is stimulated by dopamine (Shiver et al., 1992). In the dog, in vitro studies (Mol et a/., 1987) and in vivo and immunocytochemical observations (Middle­ ton et al., 1987a) have revealed the IL to be resistant to glucocorticoid suppression. There is also evidence from in vivo studies that dopaminergic pathways play a regulatory role in canine IL function (Kemppainen and Sartin, 1986). However, in other respects the situa­ tion in the dog is different from that in the rat with regard not only to the heterogeneous cytology (see Section LA.5), but also to some of the regulation charac­ teristics. Despite the fact that CRH-immunoreactive fibers have been identified in the canine neurointerme­ diate lobe (Stolp et al., 1987) and although in vitro CRH stimulates ACTH release from the neurointermediate lobe (Mol et al., 1987), there is no convincing evidence that CRH can stimulate release of ACTH from the IL in vivo (Kemppainen and Sartin, 1986, 1987; Middleton et a/., 1987b), whereas no (Kemppainen and Sartin, 1986) or a very small (Rijnberk et al., 1987) a-MSH response to CRH stimulation has been observed. In contrast with a significant increase of plasma MSH concentrations after administration of the dopamine antagonist haloperidol, even after administration of dexamethasone to inhibit the contribution of the AL, no release of IL ACTH is observed after haloperidol. In the cat as well, no stimulation of MSH occurs after CRH administration (Willemse and Mol, 1994).

However, cats undergoing handling and skin testing without anesthesia show significant increases in plasma MSH concentrations (Willemse et a/., 1993). In vitro experiments revealed the sensitivity of feline IL MSH release to dopaminergic inhibition (Willemse and Mol, 1994). In fetal and newborn lambs and in adult sheep (Newman et al., 1987) the administration of a dopamine-receptor antagonist results in a-MSH re­ lease. Elimination of the inhibitory hypothalamic con­ trol in sheep by hypothalamus-pituitary disconnection results in increased a-MSH release (Clarke et al., 1986). The dopamine inhibition of ACTH secretion in the hyperadrenocorticoid horse (Wilson et a/., 1982) sug­ gests that in the normal horse the IL is under dopamin­ ergic control. In the rabbit the dopaminergic control of the IL is absent (Schimchowitsch et al., 1986).

e. Action The predominant action of ACTH is stimulation of steroidogenesis and corticosteroid release from the ad­ renaIs (see the chapter on adrenal function). ACTH also exerts a growth-stimulating effect on the adrenal cortex. Moreover, non-ACTH portions of POMe, that is, N-terminal POMC peptides, are involved in adreno­ cortical growth (Lowry et a/., 1987). In pharmacological dosages ACTH may promote lipolysis in fat cells and amino acid uptake in muscle. The role of ACTH pro­ duced in hypothalamic neurons projecting to higher brain centers remains to be elucidated. Distribution patterns have been reported for the cat (Rao et a/., 1986). The biological effect of a-MSH in mammals remains uncertain, although the presence of specific a-MSH receptors in many tissues suggests that this peptide may affect the function of several organs (Tatro and Reichlin, 1987). Stimulation of nerve regeneration by a-MSH has been reported (Verhaagen et al., 1987; De Koning and Gispen, 1987). The C-terminal fragment of ACTH(18-39), called corticotropin-like intermediate lobe peptide (CLIP), has been found to be a potent stimulator of in vitro adrenal DNA synthesis (Wulffraat et al., 1987). a-MSH is present in the fetal pituitary of all species studied, and there is evidence that it may play a role in the regulation of intrauterine growth (Mulchahey et al., 1987).

f Disease Lesions at the hypothalamic and pituitary level may result in altered syntheSiS and release of POMC­ derived peptides. There have been no reports of the occurrence of isolated ACTH deficiency in domestic animals. There are a few reports of dogs with tumorous (supra)hypophyseal lesions, with indirect evidence for multiple adenohypophyseal and neurohypophyseal

Pituitary Function

deficits (Rijnberk, 1971; Eigenmann et al., 1983a), and one dog reported to have secondary hypoadrenocorti­ cism without information about other pituitary func­ tions (Peterson et al., 1992). In contrast to the few descriptions of ACTH defi­ ciency, pituitary-dependent hyperadrenocorticism is a common disorder in the dog (Peterson, 1987a), in which the adenomas producing the ACTH excess may originate in the AL or the IL (Capen et al., 1967; Peterson et al., 1986). From the significantly lower CRH concen­ trations in cerebrospinal fluid of dogs with pituitary­ dependent hyperadrenocorticism as compared to con­ trol dogs, it is concluded that the excessive ACTH secretion is not caused by chronic hyperstimulation with CRH (Van Wijk et aI., 1992). The ACTH secretion appeared also to be less sensitive to stimulation with CRH than with LVP (Van Wijk et al., 1994). Evidence for a genetic involvement in tumorigenesis was found in a family of Dandie Dinmont terriers (Scholten-Sloof et al., 1992). In the horse the disease originates primarily in the IL (Wilson et al., 1982; Orth et aI., 1982). In agree­ ment with the characteristics described earlier for the secretion of POMC-derived peptides by the IL, ACTH release by tumors of IL origin in both the dog and the horse tends to be strongly resistant to suppression by dexamethasone (Orth et al., 1982; Peterson et al., 1986). In dogs the highest plasma a-MSH concentrations are found in individuals with dexamethasone-resistant ACTH secretion. This suggests an IL origin of the dis­ ease, although there is evidence that the pituitary le­ sions do not always maintain the characteristics of the lobe of origin (Rijnberk et al., 1988b). A dog with diabe­ tes insipidus has been described, in which the pituitary tumor released primarily biologically inactive POMC­ derived peptides (Goossens et al., 1995). In contrast with the equine IL tumors, these tumors in dogs re­ spond poorly to administration of dopamine agonists in terms of diminished ACTH secretion (Rijnberk et al., 1988b). For details on clinical manifestations, laboratory findings, diagnostics, and treatment of these diseases, including iatrogenic hypoadrenocorticism due to corti­ costeroid therapy, the reader is referred to textbooks (Feldman and Nelson, 1996; Rijnberk, 1996) and the chapter on adrenocortical function in this volume.

g. Tests

Basal levels of circulating POMC-derived peptides are measured for diagnostic purposes in situations of suspected hypo- as well as hypersecretion. ACTH val­ ues below or just within the reference range (Table 19.3) may be found in cases of hypothalamus-pituitary disease, as well as in situations in which endogenous or exogenous glucocorticoid excess suppresses hormone

527

synthesis in the corticotropic cells. This makes the mea­ surement of basal ACTH levels a useful tool in the differentiation between pituitary-dependent hyper­ adrenocorticism and hyperadrenocorticism due to adrenocortical tumor (Feldman, 1983; Peterson, 1986). In pituitary-dependent hyperadrenocorticism, ACTH values exceeding the reference range may be found, but there is considerable overlap. Basal MSH concen­ trations have been found to be elevated in horses (Orth et al., 1982) and dogs (Peterson et al., 1986; Rijnberk et al., 1987) with pituitary-dependent hyperadrenocorti­ cism of IL origin. Of the dynamic tests, the dexamethasone suppres­ sion tests (see chapter on adrenocortical function) are still the best for the diagnosis and differential diagnosis of excessive ACTH and glucocorticoid secretion. In animals suspected of having pituitary-adrenocortical insufficiency, the secretory capacity for POMC-derived peptides can be measured by provocative testing. CRH and AVP are also used in the differentiation between pituitary-dependent hyperadrenocorticism and hyperadrenocorticism due to adrenocortical tu­ mor. The chronically suppressed corticotropic cells are presumed to be unresponsive to these stimuli. How­ ever, Meijer et al. (1978) found considerable overlap in plasma cortisol values in the AVP test, whereas much less is claimed to occur in the CRH test (Peterson, 1986). It has been demonstrated that LVP stimulates cortisol release by adrenal tumors in a direct way (Van Wijk et al., 1994). It has been shown that dogs with pituitary­ dependent hyperadrenocorticsm usually do not re­ spond supranormally to CRH administration (Rijnberk et al., 1987). The cells of these pituitary lesions were found to be less responsive to CRH in vitro than were normal corticotropes (Mol et al., 1987). As far as the possibility for manipulation of canine IL function is concerned, of the substances tested only the dopamine agonist haloperidol (Kemppainen and Sartin, 1986; Rijnberk et al., 1987) caused significant increases in circulating a-MSH. More detailed informa­ tion pertaining to some of these tests is presented in Section IV of this chapter and in Table 19.4.

2. fJ-EndorphinlfJ-Lipotropin

a. Gene Expression and Biosynthesis /3-Lipotropin (/3-LPH) consists of the 91 C-terminal amino acids of the POMC precursor. It is synthesized in the corticotropic cells of both the AL and IL. The C-terminal sequence (36-91) is remarkably similar in human, porcine, and ovine pituitaries, whereas the N­ terminal sequence (1-36) is rather heterogeneous. /3-

528

Jan A. Mol and Ad Rijnberk TABLE 19.4

Maximal Plasma Concentrations of POMC-Derived Peptides Following Intravenous Administration of a Variety of Stimulants

Species (n)

Substance

Dose

ACTH Dog (16) Dog (6) Dog (16) Dog (19) Dog (6) Cat (6) Cat (4) Sheep (5) Sheep (5) Pig (4) Pig (3)

CRH AVP LVP Haloperidol Insulin CRH Haloperidol CRH AVP CRH LVP

1 /Lg/kg 0.6 /Lg/kg 0.2 U / kg 0.06 mg/kg 0.5 U / kg l /Lg/kg 2 mg/kg l /Lg / kg l /Lg / kg 0.5 /Lg/kg l /Lg / kg

(48.6 ::!:: (50.4 ::!:: (48.6 ::!:: (61.6 ::!:: (84.9 ::!:: (29.0 ::!:: (81 ::!:: (125 ::!:: (202 ::!:: 109 ::!:: 452 ::!::

a-MSH Dog (6) Dog (19) Cat (6) Cat (4)

Haloperidol Haloperidol CRH Haloperidol

0.2 mg /kg 0.06 mg/kg l /Lg / kg 2 mg/ kg

94.2' (39.6 ::!:: 3.6) no effect (125 ::!:: 36)

Kemppainen and Sartin (1987) Rijnberk et al. (1987) Willemse and Mol (1994) Peterson et al. (1994)

f:l-END Dog (6)

Haloperidol

0.2 mg/kg

179'

Kemppainen and Sartin (1987)

Range (pmol/liter) (mean ± SEM) 11.7)' 15.0) 4.8) 6.6) 19.6) 3.2) 54) 52) 77) 2%" 5%b

Reference van Wijk et al. (1994) Kemppainen and Sartin (1987) van Wijk et al. (1994) Rijnberk et al. (1987) Kemppainen and Sartin (1987) Willemse and Mol (1994) Peterson et al. (1994) Pradier et al. (1986) Pradier et al. (1986) Minton and Parsons (1993) Minton and Parsons (1993)

a

Increment. Value relative to basal concentration. c Mean. b

MSH (sequence 37-58) appears to be an extraction artifact and plays no physiological role. By proteolytic cleavage at amino acid 61, y-LPH (f3-LPH(1-61)) and a family of endorphins is formed. Through specific deletion of C-terminal amino acids and N-terminal acetylations, various f3-END-related peptides are formed. The three main compounds are 13END(61-91), y-END(61-77), and a-END(61-76). The corticotropic cells of the AL synthesize approximately equal or higher concentrations of f3-LPH than of 13END(1-31) in most species. The equine AL, however, contains primarily f3-END(1-31) and N-acetyl-f3END(1-27) (Millington et al., 1992). The corticotropic / melanotropic cells of the IL produce relatively more f3-END and related peptides. In the IL of the normal dog the ratio f3-LPH /f3-END is less than 0.1 (Krieger, 1983). f3-END(1-27) is the most abundant form in the canine IL, followed in decreasing order of concentra­ tion by the nonacetylated f3-END(1-31) and (1-26) forms, and the N-acetyl (1-27), (1-26), and (1-31) forms (Young and Kemppainen, 1994).

b. Secretion The stimuli that induce ACTH or a-MSH release from either AL or IL cells also cause a release of the f3-LPH-related peptides from the same cell. The cellular subset of specific proteolytic, amidating, and acetylat­ ing enzymes present in the secretory vesicles define the ultimate composition of POMC-derived peptides that are secreted into the blood.

c. Action The main action of f3-LPH is to mobilize fat from adipose tissues, as demonstrated in the rabbit. Its bio­ logical function in man and other species has not been fully elucidated. Brain tissue can break down f3-LPH to form f3-END-related peptides. However, it is ques­ tionable whether these pituitary pep tides are involved in brain function; in conscious sheep, hemorrhagic stress elevates f3-END concentrations in plasma but not in cerebrospinal fluid (Smith et ai., 1986). f3-END is an endogenous opiate with a potent mor­ phinomimetic action. It is also produced in the hypo­ thalamus where the entire POMC precursor is present. Brain and pituitary endorphin are probably part of separate systems. The role of f3-END and other opioid peptides in the secretion of pituitary hormones is has been studied with long-acting analogues, opiates, and opioid receptor antagonists (Besser, 1987; Van Wimersma Greidanus, 1987). It appears that there are opioid mechanisms in the hypothalamus and that stim­ ulation, for example, causes secretion of PRL (Gross­ man et al., 1981; Delitala et al., 1982) and inhibition oj the release of oxytocin, whereas the release of vasopres· sin is differentially affected (van Wimersma Greida· nus, 1987; Hellebrekers et al., 1987). Cross-species stud· ies have revealed marked differences in the amount oj hypothalamic opiate receptors of the various subtype� of opioid peptides.

529

Pituitary Function

b. (Pro)hormone Translation of the mRNA results in the formation of a precursor peptide with a molecular weight of ap­ proximately 13,500. The a-subunit has a high degree of homology among species (Fig. 19.7). The human sequence contains 92 amino acids, rather than 96 as found in all other species studied thus far. The shorter human sequence is due to a deletion of 12 nucleotides at the beginning of exon 3. After cleavage of the signal peptide, five disulfide bridges are formed. Two aspara­ gine residues are prone to N-glycosylation, but there is also a putative O-glycosylation site (Fig. 19.7). The a-subunit is produced in excess of the ,8-subunit that determines the hormone specificity. The ,8-subunit for­ mation is rate limiting in the formation of the hetero a,8 dimer.

d. Disease/Tests Hypersecretion of ,8-END is associated with the ACTH hypersecretion of IL tumors in both equine and canine pituitary-dependent hyperadrenocorticism (Orth et aI., 1982; Peterson et aI., 1986; Rijnberk et aI., 1987). As the regulation of the secretion of ,8-END is similar to that of other POMC-derived peptides from the AL and IL, testing procedures can be applied as mentioned in the section on ACTH and MSH. Elevation of plasma ,8-END has been documented in horses fol­ lowing running and shipping (Li and Chen, 1987) and in sheep during electroimmobilization and shearing procedures ( Jephcott et aI., 1987). Treadmill exercise of Thoroughbred horses induces a variable response in plasma ,8-END concentration (Art et al., 1994). Elevated plasma ,8-END concentrations are found in dogs with congestive heart failure (Himura et aI., 1994) and are further elevated after naloxone treatment.

1. TSH The thyrotropic cells of the anterior pituitary pro­ duce thyroid-stimulating hormone (TSH), which stim­ ulates both the synthesis and secretion of thyroid hormone.

B. Glycoprotein Hormones The gonadotropes and thyrotropes synthesize the hormones LH, FSH, and TSH, each of which consists of two different peptide chains, termed the a- and ,8subunits. The amino acid sequence of the a-subunit is identical for LH, FSH, and TSH within a species, as it is for placental chorionic gonadotropin. The f3-subunits have unique structures and determine the hormone specificity. Carbohydrate substituents account for 1020% of the molecular weights of these hormones.

a. Gene Expression The rat and human TSH gene consists of three exons, whereas the mouse gene has five exons. The most prominent stimulator of gene transcription is the hypo­ thalamic thyrotropin-releasing hormone (TRH). TRH activates the thyrotrope via a signal transduction pro­ cess that is mediated by activation of the phospholipase C pathway. At least two ultimately activated transcrip­ tion factors are known. The first is Pit-1 / GHF-1, which is essential for cells producing TSH, GH, and PRL. Secondly, heterodimers of c-fos/c-jun forming an AP-1 complex activate the TSH-,8 gene. Both a-subunit and TSH-,8-subunit expression are strongly inhibited by a negative thyroid hormone response element (nTRE) in the pituitary thyrotrope. The a and ,8 types of the thyroid hormone receptor may form homodimers, and heterodimers, or even form heterodimers with the reti­ noic acid receptors RAR and RXR. Which forms are

1. a-Subunit a. Gene Expression The gene encoding the a-subunit varies between 8 and 16.5 kb among the species and contains four exons. The a-subunit comes to expression in a variety of cells such as thyrotropes, gonadotropes, and syncytiotro­ phoblasts. Its promotor sequence contains a cAMP re­ sponse element (CRE), a negative thyroid hormone response element (nTRE), a negative glucocorticoid re­ sponse element (nGRE), and thyrotrope-specific and GnRH / TRH response sites.

!��:�/cattle ���������� � . � � �� ���� �� � Rat � � � � Turkey Horse Human

�

�

I i

-- ----------------- --- - ---L--- -- --G--RF--- P------ T-------T-D - - --R----- F---V- - -- KV -[ A-*. ** * * ** * ** ** *

------

------------ ------M---------- - ------s--- - - v---s*** * * * * * * * * ** * * * * * *

------

•

--0

--- S ------------

1 -LKO-VKI -----IRV-----IKL --- S yNRV---- GFK-

-------

***

*

· ��•· .�m

; I

� --�Q � �. ---------R . - jI'-Q- -PF--Q P----L- -- - ----- -L--------Q � � �. ����- ������ � - ���: . ������� �- �� . FIGURE 19.7 L---OLII--

--0:.

*

--ll **

-

·.

-- - -

-- - H --I - -** * * *

Sequence comparison of the a-subunit for glycoprotein hormones (LH, FSH, and TSH). Lines between shaded boxes represent intra chain disulfide bridges. The open boxes indicate potential glycosyl­ ation sites. See legend to Fig . 19.6.

�

96 96 96 96 96 96 92

530

Jan A. Mol and Ad Rijnberk

important for negative feedback is the subject of cur­ rent research.

e. Disease/Tests Sensitive TSH assays became available in 1995 (Klin­ gler et al., 1995; Nachreiner et al., 1995). At the time this chapter was written, the first clinical evaluations were encouraging for the usefulness of one of these assays for the diagnosis of primary hypothyroidism (Williams et al., 1996). Long-term glucocorticoid excess reduces T4 levels in dogs (Belshaw, 1983; Peterson et al., 1984). As in man, this may be the result of diminished pituitary sensitivity to TRH (Visser and Lamberts, 1981) or to inhibited TRH release. However, in dogs there is some indirect evidence for intact TSH release, whereas it has been postulated that glucocorticoids interfere with basal T4 secretion by inhibiting lysosomal hydrolysis of colloid in the thyroid follicular cell (Kemppainen et al., 1983). Low plasma thyroxine values may also be due to decreased plasma TBG concentrations. Also, the use of certain drugs, such as antiepileptics, re­ sults in low plasma total thyroxine concentrations in the dog. In the dog hypothyroidism is most often due to a primary defect in the thyroid gland. Secondary hypo­ thyroidism may be caused by a pituitary tumor (Rijn­ berk, 1971; Chastain et al., 1979; Dunbar and Ward, 1982) or panhypopituitarism caused by a suprasellar tumor (Eigenmann et al., 1983a). Low plasma thyroxine values in combination with low TSH values may thus be caused by either secondary hypothyroidism or hy­ peradrenocorticism and certain drugs. In the first cir­ cumstance a true hypothyroidism exists, whereas in the latter cases the dogs should be essentially euthy­ roid. Unfortunately, free thyroxine measurements in the dog have not been found to discriminate between these circumstances, and thus true secondary hypothy­ roidism due to impaired TSH secretion should be docu-

b. (Pro)hormone The TSH-.8 chain consists of 118 amino acids and forms six intrachain disulfide bonds (Fig. 19.8). The TSH-13 of the species in this figure share 74% identical amino acids. There are no free cysteine residues, consis­ tent with the fact that subunits form heterodimers non­ covalently. The .8-chain of TSH has one putative N­ glycosylation site.

c. Secretion The release of TSH is stimulated by the hypothala­ mic TRH. Somatostatin (SS) has an inhibitory effect on TSH release. Thyroid hormone exerts a long-loop negative feedback. Thus, secretion of TSH is regulated by the interplay of TRH, SS, and thyroid hormone feedback (Reichlin, 1982). Although this feedback de­ pends on 3,3',5-triiodothyronine (T3) binding to the nuclear receptor of the thyrotrope, plasma TSH corre­ lates better with the thyroxine (T4) plasma concentra­ tion (Larsen, 1982). A specific 5'-deiodinating enzyme is responsible for T3 generation in the thyrotropic cell.

d. Action TSH stimulates both synthesis and secretion of thy­ roid hormones. After receptor binding, TSH stimulates the production of cAMP, which acts as a second mes­ senger. In addition, intracellular Ca2+ may modulate the biological effect of TSH via the phosphoinositol pathway. As a result of receptor activation, T4, and to a much lesser extent T3, are secreted into the blood. Prolonged stimulation of the thyroid with TSH results not only in hypersecretion of thyroid hormone, but also in enlargement of the thyroid gland.

P ig

Cattle Rat M ouse Human

Pi g

Cattle Rat Mouse Human

F �:��� e;.--� . -----y-o··

..

-- T

* *'i***** '"

TVEIPG -A----

*

1

R----T-Y -O-R- !----R- 1--

****

... ...

�:i:' -

jIr ; * * -

-

-

-

-

*

�G �TRDFNGKLFLPKYALSQDV --

xvx-------------

---1----------------

**i � * *

L..[>HHVTPYFSY PVAI �l&;l�T OYS[ -x--�

R--------------A-------- L - - - - -- - - F---V

L--A--------L-!+4+·+f----- · **

****

***

•

-

** *

1

--- 1 --------------

--N*

*

---------------

**************

�YRDFMYK ---- - --- - -T-R - - -- 1 -R -- - - 1 -R

**** *

HEAI KTN� irKPQK SYVLEFSI ------M ---TF- LGG-- G ---VR-- -- SF-LGG- -V -------- - ---'- --LVG - -V

T---V---

***

*

**

�

*

*

*

*

60 60 60 60 60

118 1 13 118 118 118

FIGURE 19.8 Sequence comparison of the /3-subunit of thyroid-stimulating hormone (TSH). Lines between shaded boxes represent intrachain disulfide bridges. The open box indicates a potential glycosylation site. See legend to Fig. 19.6.

531

Pituitary Function

mented by performing a TRH test and measurement of TSH secretion. For further details the reader is re­ ferred to the chapter on thyroid function.

2. LH and FSH Luteinizing hormone (LH), which is identical to the interstitial-cell stimulating hormone, and follicle­ stimulating hormone (FSH) are produced by the go­ nadotropic cells of the AL.

a. Gene Expression In primates multiple copies of LHj3-related genes have been identified, the chorionic gonadotropins (CG), which are expressed in the placenta. The horse is the only known nonprimate in which a CGj3 protein is formed in the placenta for the maintenance of early pregnancy. In the horse this CGj3 in the placenta is derived from the same gene that encodes pituitary LHj3. Pulsatile GnRH and castration both stimulate the LHj3 mRNA levels in the pituitary. Pulsatile treatment with GnRH also increases the length of the polyadeny­ lation of the LHj3 mRNA. The consequences for mRNA stability or translation efficiency are unknown. In the horse three transcription start sites have been found (Sherman et ai., 1992). The gene encoding FSHj3 contains three exons and two introns. Exon 1 encodes only untranslated mRNA sequences, exon 2 encodes the signal peptide and the first 18 amino acids of the prohormone, and exon 3 encodes the remaining amino acids of the prohormone and a 3' -untranslated region (3'UTR) that is long rela­ tive to other j3-subunits. The bovine gene is transcribed into a single mRNA, whereas in humans there are several transcripts of different lengths as the result

Pig

SRGPLRP

Dog Rat Cattle Sheep Human Chicken

Pig Dog Rat Cattle Sheep Human

Chicken

--E ----

LG-GG-**

:j:��

V -

-

------

of alternative splicing in exon 1 or variations in the polyadenylation length. Little is known about the response elements in the FSHj3-promoter. However, it has been demonstrated that GnRH, activin, inh ibin, and gonadal steroids regu­ late the gene expression.

b. (Pro)hormone After cleavage of the signal peptide, LHj3 proteins are formed with a variation in amino acid content among the species of 115 to 121 amino acids (Fig. 19.9). The mature FSHj3 chain contains 1 1 1 amino acids in most species, except in the horse, in which it con­ tains 118 amino acids (Fig. 19.10). Both subunits contain six intrachain disulfide bridges. LHj3 has a single N-glycosylation site, whereas FSHj3 has two N-glycosylation sites. The carbohydrate chain formed by the glycosylation of LHj3 is sulfated, in contrast to FSHj3, in which a sialic acid residue is added to the carbohydrate chain. This difference may play a role in the hormone sorting within the gonadotrope and certainly results in prolonged plasma half-life of FSH. c. Secretion and Action As with other pituitary hormones, the gonadotro­ pins are released in a pulsatile fashion, stimulated by the hypothalamic gonadotropin-releasing hormone GnRH. The pulsatile secretion of GnRH has been dem­ onstrated in sheep (Clarke and Cummins, 1982) and the horse (Irvine and Alexander, 1986). Continuous infusion of GnRH in rhesus monkeys switched off go­ nadotropin secretion (Belchetz et ai., 1978) because of down regulation of GnRH receptors (Naor et al., 1984). In the female, FSH stimulates follicle development and

F

- -----K-

-H -H

- -----K- I --VQK- -V-V -KD * * *

ELSFASIRLP --H--------- R---V----

--R---V----- R---V ---DVR-E ------

A-RYERWA-W-

PGVDPTVSFPVAL -----M- - --------- I ------- ----M- - ---------M------R----V------I-S--R-LL ---* ** * ****

-G *

-

LSSS --N---T -- -T R -T -

MAT-

-

�

-

*'

GPRAQPL ----- S ---T--M ---T -- --- T-----KDH-VQGLG-A

RPLLPGLLFL - - -- - - - --L- H-----LF H- P- - DI --H- P -- DI ---QHS G- FGGE A*

60 60 59 60 60 59 60

121 121 120 121 121 115 119

FIGURE 19.9 Sequence comparison of the ,B-subunit for luteinizing hormones (LH). Lines between shaded boxes represent intrachain disulfide bridges. The open box indicates a potential glycosylation site. See legend to Fig. 19.6.

532

Jan A. Mol and Ad Rijnberk

YTRDLVYKDPARPN IQK

Pig Sheep Cattle Horse H uman Rat

Pig Sheep Cattle Horse Human Rat

-

-

*

- - -----R------------- - - --- - - - -K- ---- ----------K----------------- T-******* ***** **

H HADS LYTYPVAT - - --- - - -- - -- - - -- ---- - - -- -

- - ----- -- - --- - - - - ----- - - R- S ---------* * * * * * * * * * * ..

60

--------

60 60

- - ---------- - --

--------

********

FSEMKE -- DIR-R-I--GD- -XYPVAL SY -G---*

***

-G----

60

FKELVYET --------

60 60

111 111 111 118 111 111

FIGURE 19.10 Sequence comparison of the f3-subunit for follicle-stimulating hormone (FSH). Lines between shaded boxes represent intrachain disulfide bridges. The open boxes indicate potential glycosylation sites. See legend to Fig. 19.6.

ripening in the ovaries. Within the theca interna, LH stimulates the production of androstenedione that is converted, by FSH stimulation, into estradiol-17,B. Es­ tradiol stimulates the pulse frequency of the GnRH release. Above a threshold level of estradiol, a massive LH surge is stimulated by a hypothalamic GnRH surge, resulting in initiation of ovulation. Administration of PMSG in cows results in increased follicular growth. The increased number of preovulatory follicles se­ verely suppresses endogenous FSH secretion (Bevers et al., 1989). After ovulation, LH stimulates the formation of a corpus luteum in the ovary, and this secretes high concentrations of progesterone into the plasma. This progesterone inhibits the release of gonadotropins by decreasing the pulse frequency of GnRH release, or by decreasing the sensitivity of the gonadotrope to stimulation with GnRH. During the course of early to late anestrus, the sensitivity of pituitary LH release of the female dog to stimulation with GnRH increases (Van Haaften et al., 1994). The ovaries also synthesize a nonsteroidal factor that attenuates or inhibits the midcycle gonadotropin surge. After partial purification by which all detectable inhibin and follistatin activity (see later discussion) was removed, the so-called gonadotropin surge inhibiting factor" (GnSIF) appeared to inhibit the GnRH­ stimulated LH secretion in vitro without affecting the basal FSH release (Danforth and Cheng, 1993). In the male animal, LH stimulates the synthesis and release of testosterone in the Leydig cells of the testis, which may in tum exert a negative feedback on LH secretion. This feedback may depend upon aromatiza­ tion of testosterone to estradiol in the brain. In the male dog both testosterone and estradiol are the major inhibitors of LH and FSH release (Winter et al., 1982). The opioid agonist bremazocine exerts a stimulatory II

effect on the LH peak level, but not on the peak fre­ quency during the follicular phase of the estrous cycle of heifers. Both peak height and frequency are de­ creased after administration of an opioid antagonist. In pigs and sheep there are also reports on the role of opioid control in various reproductive states (Short et al., 1987). FSH secretion is also stimulated by GnRH. The LH surge is accompanied by an FSH surge. The release of FSH can be selectively inhibited in vitro by inhibin, a glycoprotein hormone produced in the Sertoli cells of the testis and the granulosa cells of the ovary (De Tong, 1979). A simple bioassay has been developed for inhibin-like activity using ovariectomized ewes. Injec­ tion of follicular fluid, which is rich in inhibin activity, resulted in a selective lowering of the plasma FSH (Miller and Bolt, 1987). FSH in tum has been found to increase inhibin secretion by cultured Sertoli cells and granulosa cells (Ultee-Van Gessel and De Tong, 1987). The amino acid sequences of bovine and porcine in­ hibin have been elucidated (Forage et al., 1986). In dogs with Sertoli-cell tumors, increased inhibin bioactivity and mRNA for the a- and ,BB-subunits is found in tumor tissue. The increased inhibin concentrations in plasma are likely to be the cause of the suppressed FSH plasma concentrations, although plasma LH and testosterone concentrations were also suppressed (Grootenhuis et al., 1990). The activins are formed in the pituitary gonado­ trope. Originally these proteins, consisting of homo­ and heterodimers of activin-A and activin-B, were pu­ rified from ovarian follicular fluid. Within the pituitary the activin-B homodimer stimulates FSH synthesis without affecting LH synthesis. The action of the activ­ ins is antagonized by the glycoprotein follistatin, which is also produced both in folliculostellate cells and in gonadotropes of the pituitary.

Pituitary Function

d. Tests Plasma LH can be detected by heterologous radio­ immunoassays. Using a f3-unit-specific monoclonal an­ tibody against equine LH, parallel displacement curves have been reported for LH preparations from the horse, sheep, pig, cow, dog, cat, rabbit, rat, and kangaroo (Matteri et al., 1987). Cross reactivity ranged from 45 to 265% for LH of these species, but there was only 15% cross reactivity for human LH. For details readers are referred to the chapter on reproduction.

533

does not result in enhanced plasma GH concentrations in vivo, because glucocorticoids also stimulate the re­ lease of somatostatin from the hypoth alamus and thus inhibit GH release from the pituitary. The gene encoding GH-N contains five exon areas. By alternative splicing, the 5' part of exon 3 is missing in a minority of pituitary transcripts, and this results in the formation of a GH protein missing 15 amino acids in the N-terminal part of the molecule (Bau­ mann, 1991).

b. (Pro)hormone

C. Somatomammotropic Hormones Growth hormone (GH), prolactin (PRL), and placen­ tal lactogen show homology in amino acid composition and some biological activities, and thus may be grouped together as a family of somatolactotropic hor­ mones. There is increasing evidence that these hor­ mones evolved from a single ancestral gene (Wallis, 1984; Seo, 1985).

1. GH The somatotropic cells, producing GH, are the most abundant cells of the anterior lobe.

a. Gene Expression In primates multiplication of the GH gene has re­ sulted in a cluster of five GH-related genes. The GH­ N gene is expressed in the pituitary and the other four, consisting of the chorionic somatomammotropins (so­ called "placental lactogens") and a variant gene (GH­ V), are expressed in the placenta during pregnancy. In nonprimates a single gene encodes GH, but a family of PRL-like genes (also including placental lactogens) are present. The Pit-l (also called GHF-l) transcription factor is essential for the development of pituitary cells produc­ ing GH, PRL, and TSH. Within the somatotrope the Pit1 gene expression is stimulated by the hypothalamic growth-hormone-releasing hormone (GHRH) and in­ hibited by somatostatin (55). Both neurohormones modulate GH gene transcription by stimulation or in­ hibition, respectively, of the adenyl ate cyclase activity. The activins as formed by the gonatropes may also inhibit the expression of the GH gene. This effect is additive to the inhibition by 55. A synergistic stimulation of GH gene transcription is mediated by retinoic acid and by both thyroid hormone and glucocorticoids. The retinoic acid receptor (RAR) and the thyroid hormone receptor (TRa or TRf3) may form heterodimers with the 9-cis retinoic acid receptor (RXR) and thus strongly promote GH gene transcrip­ tion. The glucocorticoid-mediated gene transcription

GH is a single-chain polypeptide. It contains two intrachain disulfide bridges and has a molecular weight of approximately 22,000. The amino acid se­ quence of GH belongs to the best-known sequences of pituitary hormones among species (Fig. 19.11). Using DNA technology the cDNAs for human and bovine GH have been cloned in bacteria and brought to ex­ pression, leading to the production of recombinant GH (Olson et al., 1981), which is used to stimulate human growth and to enhance bovine milk production (Bau­ man et al., 1985). In cattle a polymorphism in the GH gene exists resulting in four GH variants that can arise from two possible N-terminal amino acids (phenylalanine or ala­ nine) and two amino acids in position 127 (leucine or valine). The transmission of the trait of high milk production was greater for homozygous leucine-127 in Holstein cows and valine-127 in Jersey cows (Lucy et al., 1993).

c. Secretion A variety of factors modulate the release of GH (Table 19.5). The integration of all of these factors re­ sults in pulsatile release of GH with pulse intervals of 4-6 hours, as shown in the dog (Watson et al., 1987; French et al., 1987). The hypothalamic neurohormones GHRH and 5S play a central role in the pulsatile GH release, where bursts of GHRH result in the episodic release of GH and the intervening troughs in GH secre­ tion are maintained by 55 (Fig. 19.12). GHRH was first isolated from human pancreatic islet cell tumors (Rivier et al., 1982; Guillernin et al., 1982). Using this knowledge, the amino acid sequence of hypothalamic GHRH of various species was eluci­ dated. A GHRH(I-44)NHz and a GHRH(I-40)OH form is found in the hypothalamus; both have GH­ releasing capacity because the biologic activity resides in the 1-31 sequence. Outside the central nervous sys­ tem, GHRH mRNA or immunoreactivity is found in leukocytes, testis, ovary, and placenta. Continuous in­ fusion of a high dose of GHRH results in elevated GH levels, followed by a decline to low values due to

534 Dog/ P ig Red F ox

Am. Mink H orse

Jan A. Mol and Ad Rijnberk SET I PAPTGKDEAQQRS DVELLRFSLLL IQSWLGPVQFLSRVFTNS --------------------------V--------L ---------------------------M--------------------------------------------M------------------ L--------

FPAMPL SSL FANAVLRAQH LHQLAADTYKEFERAYI PEGQRY S I QNAQ ----------------------------- D-------------

Elephant Rat Mouse

----------E -----T-M----------------------I-------------E-----T-M---------------------- I ------------- S-----------------------------------------N----K--L ----I-----------L------------------------------------------TV F----- T S G Cattle � ----------N----K--L----I----------- L --------------------------------------T--------S S / F T V G Goat heep ----------- D---K--M-------V------T---Y--K----N -------N-------------L---E------- S ---- D--HT NK-S-Duck ----------- D---K--M-- P----V------T---Y--K----N n ----------- D---K--M-------V------T-M- Y--K----N -- S --T - SNRE- T--K-NL----I--------- -----RS -- --- T I---R-- D--M---HR----- F---Q---E----KE-K-- FL -- P -TS Human E A -- S --T - SNRVKT--K-NL----I----T----E---L -RS--A-Human (var ) -- T I---R-- D--M---RR- Y--- Y--- Q---E---LKE -K-- FL -- P- TS * * ** * ** * * * * * * * * ** ** ** ** * * * • ** * ** ** * * * ** * * **

��!��;

==;====:==;==========t===�=======i====g===i :=�=== '

Dog/Pig Red F ox

LVFGTS DR VYEKLKDLEEGIQALMRELEDGS PRAGQ ILKQTYDKFDTNLRS DDALLKNYGLL ------------- ------------------------ ---------------------

-

-

-

-------- --------------------------- P------------------ -------­ -------- -----R-----------------------------------------------­ -------- ------------------------- P--V-----------M------------El ephant -M------ ---------------- Q-------- I------------A-M------------Rat -M------ ----------------Q--------v------------A-M------------Mouse ________ ------------ L---------T-----------------M------------Cattle Goat / Sheep ________ ------------L--------VT-----------------M-------------------- -F-------------------R---GP - L-- P------I H --N E---------Duck -------- - F------------------- R--- GP -L -RP------I H --NE---------Chicken -------- - F-------------------R--- GP -L -RP--- R--I H ---E---------Turkey -- Y-A--SN-- DL---------T-- GR-------T---F ---- S ----- SHN-- ________ _ Human Human (var ) -- Y-A--SN--RH--------- T--WR-------T---FN-S- S----KSHN----------* * ****** ** ** **********

Am . Mink Horse

".

-

:... I

I

DLHKAETYLR ---------------------- ----

.

FVE S ---------

F

:::::::::::::: :;. ::::: -------------- �", --A-R----- T------­ R-----T ------­ ------V----K-------V----K-------V----K-R--MD-V-- F--I­ R--MD-V-- F--IV ** * ** *

I I I

99 99 99 99 99 99 99 99 99 98 99 99 100 100

190 1 90 190 190 190 190 190 1 90 190 189 190 190 1 91 191

FIGURE 19.11 Sequence comparison of growth honnone (GH). Lines between shaded boxes represent intrachain disulfide bridges. See legend to Fig. 19.6.

desensitization of the somatotropic cell as a result of down regulation of GHRH binding sites. This phenom­ enon is a time- and concentration-dependent process (Bilezikjian et ai., 1986). The GHRH-induced GH re­ lease is potentiated by protein kinase C activation, indi­ cating that the phosphoinositol system also plays a role in the GH release. A candidate for this activation is the yet unknown endogenous counterpart of the synthetic GH-releasing peptide (GHRP-6) that stimu­ lates GH release in vivo in a variety of species, including the dog (Muruais et ai., 1993). TheTRH effect on GH release is more complex. In healthy mammals, includ­ ing the dog (Rutteman et ai., 1987a), TRH does not TABLE 19.5

stimulate GH release (5canes et ai., 1986). In cattle, TRH induces some GH release, but when used together with GHRH these two releasing hormones act in strong synergy (Lapierre et ai., 1987) (see also Table 19.7). 50matostatin (55) denotes a family of peptides in­ cluding the 14-amino-acid-containing 5-14 and the N­ terminal-extended 5-28 (Reichlin, 1982). Passive immu­ nization of chickens with 55 antiserum increases the basal plasma GH concentrations and augments TRH­ induced GH secretion (Harvey et ai., 1986). In compari-

Putative Factors Modulating GH Release Stimulating

Inhibiting

Hormones

GHRH Glucagon Pentagastrin Enkephalin

Somatostatin (55) IGF-l, IGF-II Corticosteroid excess Hypothyroidism

Biogenic amines

a-Adrenergic agonists ,B-Adrenergic antagonists Dopamine

a-Adrenergic antagonists ,B-Adrenergic agonists Serotonin antagonists (cyproheptadine)

Others

Hypoglycemia Fall in free fatty acids Amino acids (arginine) Sleep Stress (emotional) Exercise

Hyperglycemia Rise in free fatty acids

I Growth

r growth-"romoting actions Indirect

Skeletal growth

Honnone

I

Tissue growth Cellular differentiation

'I

l -.--1--. rn:

Direct Antilin

Lipolysis

Increased blood sugar

FIGURE 19.12 Hypothalamic regulation of pituitary GH release, peripheral actions of GH, and the effects of IGFs (feedback), glucocor­ ticoids, and sex steroids on the hypothalamic-pituitary system.

Pituitary Function

SOn with that of 5-14, the suppressive effect of 5-28 on GH release is much longer. 55 appears mainly to inhibit GH release and to have little effect on the synthesis of GH. The apparent ineffectiveness of 55 in some studies may be due to its extremely short half-life (Harvey and Scanes, 1987), which was found to be 1.82 minutes in the dog (5chusdziarra et aI., 1979). However, in dogs, GH pulsatile secretion could also not be inhibited by the administration of the long-acting SS analogue SMS 201-995 in a dose of 1 /Lg / kg (Watson et aI., 1987). Continuous infusion of SS in a dose of 0.15 /Lg / kg / min abolished the secretory bursts, but nO effect was seen on basal secretion rates (Cowan et al., 1984). In the chicken, bolus injections of SS also failed to inhibit basal GH release (Harvey et aI., 1986). The pulsatile GH secretion is age and gender related. Although there are conflicting reports on the effect of aging on basal GH release, it is generally assumed that the pulse height of GH decreases with age in man. This age-related decline has been confirmed in domes­ tic fowl (Harvey et al., 1987) and cattle (Verde and Trenkle, 1987). In the rat there is a gender-related pat­ tern of GH secretion. The male rat has high pulses and a very low trough, whereas the female rat has lower bursts but higher basal levels (see also Section 11.C.1.e.). There are also indirect neuronal influences that modulate GH secretion (see also Table 19.5) (Hall et ai., 1986). The main physiological stimuli for GH secre­ tion are sleep, physical exercise, stress, fasting, cate­ cholamines, hypoglycemia, and certain amino acids, but not all apply to all species. For example, GH secre­ tion in dogs is not related to sleep or day-night cycles, although GH peaks may occur after forced wakeful­ ness at the onset of sleep (Takahashi et al., 1981). Insulin-induced hypoglycemia and arginine adminis­ tration do not consistently result in GH release in the dog (Eigenmann and Eigenmann, 1981 a). Of the neuro­ transmitter systems involved, adrenergic systems seem to play a major role. 0!2-Adrenergic agonists promote GH secretion, whereas l3-adrenergic agonists are inhib­ itory. Thus, clonidine, a central 0!2-adrenergic agonist, is also an effective stimulator of GH secretion in the dog (Eigenmann and Eigenmann, 1981a). There is increasing evidence that the GH-stimulated release of insulin-like growth factor I (IGF-I) exerts a negative long-loop feedback on pituitary GH release. IGF-I stimulates the release of somatostatin by the hy­ pothalamus and also inhibits pituitary GH release di­ rectly (Ceda et ai., 1987). In the domestic fowl, injection of IGF-I results in decreased GH responses to stimula­ tion with TRH and GHRH (Buonomo et ai., 1987). The elevated GH levels in fasting dogs have also been ex­ plained as the result of impaired feedback inhibition due to low IGF-I levels in plasma (Eigenmann et al., 1985). A low level of feeding also results in reduced

535

plasma IGF-I concentrations and significantly in­ creased plasma GH concentrations in steers (Breier et aI., 1986). GH responsiveness of the liver appears to change under different nutritional conditions through changes in receptor number, thereby having a domi­ nant influence on the regulation of GH release. After secretion, GH may circulate in the blood par­ tially bound to specific GH-binding proteins. One of these binding proteins appears to be identical to the amino acid sequence of the extracellular domain of the GH receptor, whereas the other seems to be unrelated to the GH receptor (Bingham et aI., 1994). There was 2 no difference in the binding of 1 51-labeled hGH to se­ rum proteins of poodles differing considerably in body size, but binding to serum of dwarf pigs was reduced in comparison to binding to serum of nor­ mal pigs (Lauteriq et aI., 1988).

d. Action The effects of growth hormone can be divided into two main categories: rapid or metabolic actions and slow or hypertrophic actions. The (acute) metabolic re­ sponses are due to direct interaction of growth hormone with the target cell, whereas the slow hypertrophic ef­ fects, or those on cartilage, bone, and other tissues, are indirect. The direct effects of GH result in enhanced li­ polysis and restricted glucose transport due to insulin resistance (Eigenmann, 1984). Administration of recom­ binant-DNA-derived human GH induces glucose intol­ erance in the dog (Shaar et al., 1986). Treatment of pigs for 7 days with purified pituitary GH or recombinant­ DNA GH causes insulin resistance together with de­ creases in insulin- and IGF-I-stimulated lipogenesis of adipose tissue (Walton et aI., 1987). The GH responsive­ ness of adipose tissue is less dependent On nutrition than that of the liver, which may explain the reduced adiposity in the absence of growth enhancement in GH­ treated ruminants (Breier et ai., 1986). In contrast to these direct catabolic effects, the indi­ rect actions are anabolic and appear to be mediated by the insulin-like growth factors, IGF-I and IGF-II (Daughaday et aI., 1987). In their chemical structure the IGFs have approximately 50% homology with insulin/ proinsulin, suggesting they have evolved from a com­ mon ancestral molecule. The IGFs are bound to a family of six different IGF binding proteins (IGFBP). The main IGFBP in plasma is IGFBP-3, which is formed and released by the liver upon stimulation with GH. The IGF-IGFBP complex binds in plasma to an acid-labile subunit, resulting in a 150-kDa complex. As a conse­ quence of this binding the IGFs have a very long plasma half-life, which is consistent with their long­ term growth-promoting action. Insulin and IGF seem to complement each other, insulin being the acute and IGF the long-term regulator of anabolic processes (for

536

Jan A. Mol and Ad Rijnberk

a review, see Eigenmann, 1987). The other IGFBPs play a more local role in the modulation of the IGF effects in specific tissues. These effects may be inhibitory as well as stimulatory. Competitive binding experiments have revealed that IGF receptors are distinct from insulin receptors and that there are two subtypes of IGF receptors (type I and type II), which differ in their affinity for IGF-I and IGF-IL These two subtypes of IGF receptors have been characterized in terms of structure, phosphoryla­ tion, and regulation (Rechler and Nissley, 1985). There is increasing evidence that the action of GH cannot be categorized by two opposing biological ac­ tivities in such absolute terms as described earlier. GH exerts its growth-promoting effect not only via IGF-I produced in the liver, but also by a direct effect on cells in the growth plate. It has been proposed that GH stimulates cell differentiation directly and clonal expansion indirectly through the local production of insulin-like growth factors (Nilsson et al., 1986). Circulating IGF-I concentrations appear to be associ­ ated with body size. In a study in different breeds of dogs the IGF levels were found to correlate with body size (Eigenmann et al., 1984a). By studying genetic sub­ groups within one breed, i.e., standard, miniature, and toy poodles, Eigenmann et al. (1984b) found IGF-I lev­ els (and not GH) to parallel body size. Dwarf chickens also have relatively low circulating levels of this growth factor (Huybrechts et al., 1987).

e. Extrapituitary GH Synthesis In dogs, endogenous progesterone and exogenous progestins may induce considerable rises in plasma growth hormone levels, resulting in acromegalic changes and insulin resistance with the possibility of development of frank diabetes mellitus (Eigenmann and Rijnberk, 1981; Eigenmann et ai., 1983b; Selman et al., 1994b). GH excess has only been found to oc­ cur in intact female dogs during diestrus (proges­ terone phase) or in dogs treated with progestins (Con­ cannon et al., 1980; Rijnberk et al., 1980; Eigenmann and Rijnberk, 1981). Estradiol priming may enhance the effect of exogenous progestins in inducing GH overproduction (Eigenmann and Eigenmann, 1981b). Progestin withdrawal and / or ovariohysterectomy re­ sults in a reduction of the elevated plasma GH and IGF-I concentrations (Eigenmann et aI., 1984a) and re­ versal of the soft-tissue changes and the insulin resis­ tance (Eigenmann et al., 1983b). A decrease in the plasma GH levels has also been achieved by adminis­ tration of a synthetic antiprogestin (Watson et al., 1987). Selman et al. (1991) made progress in the character­ ization of this phenomenon. Progestin-induced GH ex­ cess was insensitive to stimulation with GHRH or clon-

idine and also insensitive to inhibition with the SS analogue SMS 201-995, whereas the dosage SMS used inhibited the GHRH- and clonidine-stimulated GH re­ lease in healthy male dogs (Selman et al., 1991). Taken together these observations strongly supported the au­ tonomous character of the progestin-induced GH se­ cretion. Hypophysectomy of progestin-treated ovario­ hysterectomized female dogs did not result in decreases in plasma GH concentrations (Selman et al., 1994a), supporting an autonomous, nonpituitary pro­ duction. Analyses of various tissues revealed the high­ est immunoreactive GH concentrations in the mam­ mary gland. The mammary origin of elevated plasma GH was then confirmed by an arteriovenous gradient across the mammary gland, a rapid decrease and nor­ malization of plasma GH concentrations after complete mammectomy (Selman et al., 1994a), and the presence of GH mRNA in the canine mammary gland by reverse transcriptase PCR (Mol et al., 1995a). From the sequence identity between the pituitary and the mammary PCR product it was concluded that a single gene encodes pituitary and mammary GH in the dog. Until recently, the progestin-induced GH synthesis was only found in the dog. In cats Peterson (1987b) found no significant rise in plasma GH concentrations during 12 months of treatment with megestrol acetate. However, it was demonstrated recently that in cats with progestin-induced fibroadenomatous changes of the mammary gland the GH mRNA is also present in these tissues (Mol et al., 1995a). Moreover, GH mRNA has now been demonstrated in the normal and neoplas­ tic human mammary gland as well (Mol et al., 1995b, 1996). In normal cyclic female dogs, significantly higher basal plasma concentration of GH and IGF-I are found during the progesterone-dominated metestrus than during anestrus. This increased basal concentration is accompanied by a decreased responsiveness to stimu­ lation with GHRH or clonidine (Rutteman et aI., 1987b; Selman et al., 1991). In the light of the progesterone­ induced synthesis in mammary tissue, it is possible that these changes during metestrus are caused by a contribution of the mammary gland to the circulating plasma GH concentrations. Whether extrapituitary GH synthesis is also the cause of gender-related differences in GH secretion in other species remains to be eluci­ dated.

f Disease Inadequate GH secretion at an early age causes re­ tardation of growth. Apart from occasional reports on dwarfism in dogs and cats, GH-deficiency dwarfism seems to occur primarily as a genetically transmitted condition in the German shepherd (Andresen et aI.,

537

Pituitary Function

1974) and in the Carelian bear dog (Andresen and Willeberg, 1976). The abnormality has been docu­ mented as a primary GH deficiency (Scott et ai., 1978; Eigenmann, 1981) with low levels of circulating IGF-I (Eigenmann et al., 1984c) and seems to he due to pres­ sure atrophy of the adenohypophysis by cysts of Rathke's pouch (Miiller-Peddinghaus et al., 1980). In adults, GH deficiency causes much less impres­ sive changes. The clinical features remain confined to the skin and coat. Affected dogs present with alopecia and hyperpigmentation of the skin (Parker and Scott, 1980; Eigenmann and Patterson, 1984). The skin prob­ lems may improve by treatment with heterologous GH, although the administration of hGH to the dog is not effective and gives rise to GH antibody formation (Van Herpen et al., 1993). The condition is not well defined and may be an isolated GH deficiency. It has been suggested that in some cases of alopecia the low basal GH levels and the absence of response to stimula­ tion by clonidine or GHRH were due to enhanced somatostatin release as a result of mild and fluctuating hyperadrenocorticism (Rijnberk et al., 1993). In cats, as in man (Melmed et al., 1983), excessive amounts of growth hormone may be secreted by a pituitary tumor. Since the publication of Gembardt and Loppnow (1976) on adenohypophyseal adenomas in TABLE

19.6

Basal GH and

diabetic cats, there has been growing awareness that GH hypersecretion may be involved in the pathogene­ sis of diabetes mellitus (Peterson et al., 1990). Hypothyroidism leads to a markedly reduced growth-hormone secretion, with lack of response to provocative stimuli (Chernausek et al., 1983). Apart from low serum GH levels, hypothyroidism is associ­ ated with low GHBP, IGF-I, and IGFBP-3 concentra­ tions (Rodriguez-Arnao et al., 1993). In hyperadrenocorticism, GH secretion is decreased and the response to provocative stimuli is impaired. In hyperadrenocorticoid dogs this proved to be reversible when normocorticism was achieved, suggesting that the impaired GH secretion is the result of hypercorti­ cism per se (Peterson and Altszuler, 1981) and medi­ ated by glucocorticoid-induced increase of the hypo­ thalamic 55 tone (Takahashi et al., 1992). g. Tests

As GH secretion is pulsatile, single values are of no great diagnostic value. In healthy individuals, resting plasma GH concentrations (Table 19.6) may be very low, even below the detection limit of the assay. Hence when GH deficiency is suspected, a stimulation test is needed. In the dog, stimulation with clonidine, xylaz­ ine, or GHRH has proven to be reliable for this purpose,

IGF-I Concentrations in Plasma of Healthy Animals"

Species (n)

Age, breed (condition)

Range (ltg/liter) (mean ± SEM)

Reference

GH Dog (63) Dog (5) Dog (5) Dog (6) Dog (6) Cat (25) Mare (12) Calf (6) Steer (6) Pig (4) Pig (4) Sheep (6)

Adult 7 weeks, Great Dane 27 weeks, Great Dane Beagle (anestrous) Beagle (metestrous) Adult 7-11 year, Selle Franc;ais 3 days 267-303 days 5 days 170 days

(1.9 :+: 0.1) (13.7 :+: 2.2) (0.7 :+: 0.4) (0.5 :+: 0.1) (2.2 :+: 0.4) (3.2 :+: 0.7) (2.8 :+: 0.7) (11.0 :+: 1.8) 15.3-32.6 (9.0 :+: 1.4) (1.5 :+: 0.1) (3.0 :+: 0.2)

Eigenmann and Eigenmann (1981a) Nap et al. (1993) Nap et al. (1993) Selman et al. (1991) Selman et al. (1991) Eigenmann et al. (1984d) Davicco et al. (1994) Coxam et al. (1987) Wheaton et al. (1986) Lee et al. (1993a) Lee et al. (1993a) Krysl et al. (1987)

IGF-I Dog (8) Dog (10) Dog (10) Dog (13) Mare (12) Pig (4)

Cocker spaniel Beagle Keeshond German shepherd 7-11 year, Selle Fran�ais 170 days

36 :+: 27 87 :+: 33 117 :+: 34 280 :+: 23 154-318 (182 :+: 30)

Eigenmann et al. (1984a) Eigenmann et al. (1984a) Eigenmann et al. (1984a) Eigenmann et al. (1984a) Davicco et a/. (1994) Lee et a/. (1993a)

IGF-II Dog (8) Pig (4)

7 :+: 0.5 year, beagle 170 days

(390 :+: 55) (326 :+: 19)

Mol et a/. (1997) Lee et al. (1993a)

" The values can be converted to SI units: 1 /Lg GH/liter = 0.0455 nmol /liter; 1 /Lg rGF/ liter of determinations in samples collected at 15-min intervals for 24 hours.

b Means

=

0.133 nmol /liter.

538

Jan A. Mol and Ad Rijnberk TABLE

Species (n)

19.7

Substance

Maximal Plasma Concentrations of GH Following Intravenous Administration of a Variety of Stimulants·

Dose

Range (pglliter) (mean

�

SEM)

Reference

Dog (6)

Cloriidine

10 jLg/ kg

(19.2 ,± 6.2)

Selman et Ill. (1991)

Dog (6)

GHRH

1 jLg/kg

(10.0 ± 5.1 )

Selman et Ill. (1991)

Cow (4)

GHRH

0.4 jLg/kg

16.5

Enright ct Ill. (1987)

Calf (4)

GHRH

0.4 jLg/kg

(107 ± 55)

Enright et Ill. (1987)

Cow (6)

GHRH

3.3 jLg/kg

20.4

Lapierre et Ill. (1987)

Cow (6)

TRH

TRH + GHRH

1.1 jLg/kg

4.0

Lapierre et al. (1987)

1.1 jLg / kg each

71.4

Lapierre et Ill. (1987)

GHRH

0.5 jLg/kg

(50.1 ± 19.1)

Barenton e t Ill. (1987)

Cow (6) Lamb •

See also footnote to Table 19.6.

whereas in calves and lambs GHRH stimulation has been used (Table 19.7). Elevated GH values are not definitive proof of acro­ megaly, not only because of the pulsatile nature of the GH secretion but also because environmental factors may cause sharp increases. Although hypersecretion should be tested by an inhibition test, in the case of progestin-induced GH secretion a blunted stimulation of high basal GH concentrations may support the diag­ nosis, together with elevated plasma IGF-I concentra­ tions (Selman et al., 1991). It should be kept in mind that the reference values for the IGF-I concentrations in the dog are breed ( body size) dependent. =

2. Prolactin

a. Gene Expression Prolactin (PRL) is encoded by a single gene in the human genome. In nonprimates such as cattle, how­ ever, a family of PRL-related genes is found, and only one gene encodes GH. The PRL-related genes of the rat can be divided into the lactogens (placental lactogen I, II, I-variant, and the decidual luteotropin) and the nonlactogens (PRL-like protein (PLP) A, PLP-B, PLP­ C, and the proliferin-1 and -2 proteins). The appearance of specific PRL-producing lacto­ tropes is one of the latest events in pituitary develop­ ment. It has been proposed that most, if not all, lacto­ tropes are derived from presomatotropes under the influence of the lactotrope-specific transcription factor (LSF-1). In humans and rats the majority of PRL­ producing cells are also capable of synthesizing GH and are called mammosomatotropes. Even in adult­ hood the balance between somatotropes and lacto­ tropes may change in the direction of PRL-producing cells under the influence of chronic stimulation with GnRH (Cooke, 1995). It is not clear whether this is due to conversion of somatotropes to lactotropes or

whether new lactotropes are generated from pituitary stem cells. The Pit-1 transcription factor is a prerequisite for the expression of PRL, as it is for expression of GH and TSH. The Pit-1 factor appears to interact with the estrogen response element (ERE) located on a more distal enhancer site, resulting in augmented transcrip­ tion. More specific for the PRL promoter is the binding of the LSF-1 transcription factor. Glucocorticoids may inhibit PRL gene expression by a negative GRE. The PRL gene, like the GH gene, has five exon areas. In the extrapituitary expression of PRL in decidua and lymphocytes, a different promoter and exon 1 are used, located upstream to the pituitary promotor.

b. (Pro)hormone Prolactin (PRL) is synthesized as a single polypep­ tide that, after cleavage of the signal peptide, has a molecular weight of approximately 22,500 and three intrachain disulfide bridges. There is a good deal of variability in the sequences of PRLs of different species (Fig. 19.13). A minority of prolactin molecules may be present in the pituitary in a glycosylated form. This form has a decreased potency in receptor binding and biological assays. Expression of the PRL gene within the hypothala­ mus-neurohypophyseal system results in full-length PRL mRNA. After translation the hormone is pro­ cessed differently from the AL by specific proteolytic enzymes that generate a 16- or 14-kDa N-terminal frag­ ment. This fragment is able to inhibit angiogenesis by a specific receptor (Clapp et al., 1994).

c. Secretion For two reasons PRL has a unique place among the AL hormones. First, it is the only hormone that is under tonic inhibition by the hypothalamus. After transplan-

539

Pituitary Function Pig Am . Mink Horse Human Cattle Sheep

Chicken

Turkey E lephant Rat Mouse

TSS LS TPEDKEQAQQI HHEVLLNLILRVLRSWNDPLYHLVT MS LR DLFDRAV IL SHYIHNLSS EMFNE F DKRYAQGRGFI TKAINS ----- S ------------- ------------------- - ------------ -------D-- -------------

100 76

��F1-: ����������������l����j����������������:; ��������������������������j��� I I R- S -

G-

-

�

---I---- --D----T-----MD---GL--------D--AS V-- P-------M------ S --- D -- H --N-Q-- L----- PR------A---------KVPP --------S LVH ------ FQ- I T P- PE ----V -M------T-YTD--I----Q-V- D-E --A----IT ---E----V--------T-YTD--I----Q-V-D- E-MV-V-- - ----A-------- LKVPP--------SLVQ-SS-- F Q * ** * * ***** * *** * * * ** ***** *

D K EVRGMQEAPDA I L S RAI EIEEQNKRLLEGMEKIVGQVHPG l KENEVYSVWSGLPS LQMADEDT RLFAFYNLL C RRDSHKI NYL LL ----------S-------- ----R----------------VR- - ----------- -------S---- ----- � - - - - - - - - - - - - - - ----------------------E ---- K-------- R-------------Q-R--------------- -------S-------Horse ----------------- -----E---- K-V-----T-------- L--S --- -ET----I-P--------- --- -ES--S-Y- --Human -----KG--------------E ---------M-F--- I--A--T- P-P---------TK-- -A- YS ------ � ----S---T----­ Cattle -----KGV-------------E ---------M-F--- I --A--T- P- P---------TK---A- H S ------ � ----S---T----­ S heep -- QRIK----T--WK-V------------------R --S -DAG-- I --H - D------ L----S------ --- � ------------V-Chicken --QR IK----T--WK-V------------------ RI- S-DAG--- F-Q- D ------ L---- S --------- • ------------V-Turkey . E lephant --H SL PK-- S - L-TK-T-VK- E -Q----- I----D-----A-- - KA-----------TT---A- ------- F � --------S-----------V----- FGLG- I H ------I---K------------ I---IS-AY-EA-G--I- L ---Q-----GV--ESKDL----N I Rat Mouse G- G-I----EY- --- -K------- Q---- V--- I S -AY-EA-G- GI - F---Q----- GV--ESKI LS LR-T I � ------V-- F --V** * **** * * * ** * ***** * * * * **** * *

Pig Am . Mi n k

-V----- HNN---NN---NN-

L-H-- ­ L -H-N - ­ -V-NN-VHKN -AHQN-"" *

1 00 98 98

199 175 199 199 199 199 199 199 199 197 1 97

FIGURE 19.13 Sequence comparison of prolactin (PRL). Lines between shaded boxes represent intrachain disulfide bridges. See legend to Fig. 19.6.

tation of the pituitary under the kidney capsule, PRL synthesis is remarkably enhanced. Secondly, PRL lacks a specific target organ that produces factors exerting negative feedback. The main hypothalamic prolactin-inJtibiting factor (PIF) is dopamine. After the release in the median eminence, it may reach the lactotrope via the portal vasculature or by diffusion from the synaptic contact of dopaminergic terminals that end On the melano­ tropes of the 11. The relative contribution of these sys­ tems is not clear. Several isoforms of the endothelins, known for their vasoconstrictor properties, are po­ tent inhibitors of PRL release in vitro, but nO effect has been demonstrated in vivo. Both basal and TRH­ induced PRL secretion is also inhibited by calcitonin

in vitro. The suckling-, stress-, or estrogen-induced PRL surge cannot be explained by changes in dopaminergic inhibition alone. Therefore, the presence of a prolactin­ releasing factor (PRF) has been suggested. Several can­ didates for PRF have been proposed. One of the com­ ponents of hypothalamic extracts known to have a stimulatory effect On PRL release is TRH. This was demonstrated in several mammalian species, including the pig (Bevers and Willemse, 1982). However, TRH is probably neither the sole nor the major physiological PRF. Another strong candidate for PRF is vasoactive intestinal peptide (VIP) (Shimatsu et aI., 1985). The rat posterior pituitary was found to contain a potent PRF (Hyde et al., 1987). Further analysis demon­ strated that two compounds, oxytocin from the NL and an unidentified peptide from the IL, could function

as PRF. The low potency of oxytocin made it a less likely candidate for the physiological regulation of PRL release. The PRF from the IL appeared to be a small peptide that is present in the posterior pituitary of many species. Its chemical nature remains to be deter­ mined. Like other pituitary hormones, PRL is released in a pulsatile manner, with fluctuations during different stages of the reproductive cycle. Apart from an increase around the time of ovulation (McNeilly et ai., 1982), PRL concentrations in plasma tend to increase in the luteal phase of the sexual cycle in dogs (DeCoster et aI., 1983; Okkens et aI., 1985) and cows (Dieleman et al., 1986), but not in cats (Banks et aI., 1983). During lactation very high PRL concentrations have been found in the sow by Bevers et aI. (1978) and in the dog by Concannon et aI. (1978). In addition, distinct increases in PRL concentration have been found in relation to parturition (Taverne et aI., 1978 / 1979; Con­ cannon et aI., 1978). The effect of the lighting regime On PRL secretion in rams is thought to be a direct effect of melatonin' On the pituitary gland (Lincoln and Clarke, 1994). Important modulating factors in the control of PRL secretion are estrogens, especially 17-,B-estradiol. Es­ trogens modulate the TRH receptor levels in the pitu­ itary (DeLean et aI., 1977) and cause a biphasic increase in transcription of the prolactin gene (Gorski et aI., 1985). From experiments in rats, Deis and Alonso (1985) concluded that progesterone may also stimulate the release of PR1. Rutteman et al. (1987a) found that estradiol rapidly induces an enhanced PRL response

540

Jan A. Mol and Ad Rijnberk

to TRH in dogs, without changing basal PRL levels. Subsequent administration of medroxyprogesterone acetate did not further affect these findings. Neurogenic factors also influence PRL secretion. Milking and suckling are almost immediately followed by PRL release. Removal of litters from their dams, for example, piglets from sows (Bevers et al., 1978), results in a rapid decline in PRL levels in plasma. Following return of the litters, PRL concentrations rise again.

d. Action The most familiar role of PRL in mammals is stimu­ lation of mammary gland growth and lactation. PRL increases mitosis of mammary gland epithelial cells not only during development, but also during preg­ nancy and lactation. It has been relatively difficult to demonstrate in vitro effects of PRL on cell proliferation in mammary glands (Friesen et al., 1985). This can be explained by the increasing evidence that the growth­ promoting effect of PRL has much in common with that of GH, i.e., intermediate factors comparable to IGF are required for effects on cell proliferation (Nicoll et al., 1985). PRL has a wide variety of other physiological ac­ tions among vertebrates. It may affect water and elec­ trolyte balance, metabolism, gonadal function, and be­ havior. Of these, the effect on the ovary has received much attention. PRL has a luteotrophic effect in some animals, such as rodents, sheep, and ferrets (see the review by McNeilly et al., 1982), but not in the cow (Bevers and Dieleman, 1987). Bovine follicles do not bind (ovine) PRL (Bevers et al., 1987). The reciprocal relationship between PRL and LH has been well demonstrated in several species, includ­ ing the sow (Bevers et al., 1983) and the cow (Dieleman et ai., 1986). This can explain the reduced fertility dur­ ing lactation that is known in many species. It has been suggested that PRL may also interfere directly at the ovarian level (McNeilly, 1982). These effects of PRL also appear to play a role in the maintenance of the long interestrous interval in the bitch. Treatment of bitches with the dopamine agonist bromocriptine re­ sults in considerable shortening of the interestrous in­ terval (Okkens et al., 1 985). Mammalian maternal behavior in several species consists of nest building and caring for offspring. Prolactin-induced maternal behavior has been demon­ strated in some animals. However, the primary role of PRL in inducing mammalian maternal behavior has been questioned (Scapagnini et ai., 1985). There appears to be diversity in the hormonal basis of maternal behav­ ior, and in some species estrogens and progesterone play a crucial role (Rosenblatt, 1984).

e. Disease The prolactinoma is the most frequently diagnosed functioning pituitary tumor in man. Mixed GH-PRL adenomas are known to occur in a substantial number of patients with acromegaly, and the concomitant production and secretion of PRL by corticotrope ade­ nomas is not unusual (Ishibashi and Yamaji, 1985). There have been no well-documented reports on the occurrence of prolactinomas in animals, although in dogs with pituitary-dependent hyperadrenocorticism (PDH) Stolp et al. (1986) found PRL concentrations in plasma to be higher than those in healthy control dogs and in dogs with hyperadrenocorticism due to adreno­ cortical tumor. Moreover, the PRL concentrations in dogs with PDH responded supranormally to stimula­ tion with TRH. This suggests that the pituitary lesions of dogs with PDH contain prolactin cells. Thus, there do not seem to be pathological hyper­ prolactinemic states in domestic animals that require treatment with dopaminergic drugs. These drugs are nevertheless often used, the main application being in a phYSiological condition in the bitch called pseudo­ pregnancy (Arbeiter and Windig, 1977; Janssens, 1981; Mialot et al., 1981 ). Administration of the dopamine agonist bromocriptine results in rapid disappearance of the signs suggesting impending parturition. As treatment with bromocriptine lowers PRL concentra­ tions in plasma (Okkens et al., 1985; Stolp et ai., 1986), it is likely that the lactation and maternal behavior of the diestrual bitch are PRL dependent.

f Tests Despite the variations in amino acid sequences of PRLs from different species, plasma concentrations can be measured in heterologous assays, for example, em­ ploying labeled porcine PRL with an antibody against ovine PRL in an assay for canine PRL (Stolp et ai., 1986). PRL levels in the blood may vary during the reproductive cycle, but the basal levels in both male and anestrous female dogs are within similar narrow limits (Stolp et al., 1986). Reference values are presented in Table 19.8. When there is a suspicion of prolactin deficiency, the secretory capacity can be tested with TRH. In healthy dogs intravenous administration of 10 p,g TRH per kg body weight resulted in PRL increments oj 13. 1-42.5 p,g/ liter (Rutteman et al., 1987a). In cats, 75 p,g TRH caused PRL elevation to seven times the basal levels (Banks et ai., 1983). Prolactin concentrations car be suppressed with the dopamine agonist bromocrip· tine. Both 20 p,g / kg intravenously (Stolp et al., 1986: and 10 p,g / kg orally (Rijnberk et ai., 1987) resultec in protracted decreases lasting at least 8 hours aftel administration.

541

Pituitary Function TABLE

19.8

Concentrations of PRL in Plasma of Healthy Animals·

Condition

Species (n)

Anestrous female

Dog

Reference

Range (uglliter) (mean ± SEM) (9.1 :': 1.2)

Knight et al. (1977)

Dog (6)

1.5 weeks of lactation

(86 :': 19)

Concannon et al. (1978)

Dog (6)

8-32 hours prepartum

(117 :': 24)

Concannon et al. (1978)

(0.9 :': 10.5jb

Stolp et al. (1986)

Dog (23)

Males + anestrous females

Dog (5)

Ovariectomized females

Cat (8)

Early gestation

Cat (8)

7.9-11.5

Rutteman et al. (1987a)

(7.0 :': 0.3)

Banks et al. (1983)

End of gestation

(43.5 :': 4.5)

Banks et al. (1983)

Cat (8)

4 weeks post partum

(40.6 :': 7.2)

Banks et al. (1983)

Cow (5)

Luteal phase

(23.3 :': 4.8)

Dieleman et al. (1986)

Follicular phase

(15.8 :': 2.7)

Dieleman et al. (1986)

Cow (5) Sow (3)

2nd week of lactation

(9.1 :': 1.4-26.1 :': 5.0)'

Bevers et al. (1978)

Sow (3)

Piglets removed

(1.4 :': 0.3-1.9 :': 0.1)'

Bevers et al. (1978)

Sheep (8)

Males

68.6 :': 15.3

Matthews and Parrott (1992)

a When

necessary PRL concentrations are converted to SI units: 1 f.Lg PRL/ liter = 0.0444 nrnol/ liter. of 6 measurements throughout the day. e Means of 8 measurements at IS-min intervals.

b Means

III. NEUROHYPOPHYSIS A. Vasopressin

1. Gene Expression Although there is considerable homology between the genes for vasopressin (VP) and oxytocin (OT), each of the relevant hypothalamic neurons produces only one hormone, either VP or �T. The VP gene consists of three exons. Exon 1 encodes the signal peptide, the nonapeptide VP, and the N-terminal part of neurophy­ sin II (NP-II ). Exon 2 encodes the central part of NP­ II, and exon 3 encodes the C-terminal part of NP-II and a glycoprotein (GP). The major factor that regulates VP synthesis is osmotic stimulation. Rats given hypertonic saline chronically have increased VP and OT mRNA concentrations in the magnocellular neurons in the hy­ pothalamus. Similar increases in VP mRNA are ob­ served in hypovolemic conditions. Acute increases in plasma osmolality or hypovolemia result in a rapid lengthening of the poly(A) tail of the mRNA, followed by a slow increase in mRNA abundance.

2. (Pro )hormone After cleavage of the signal peptide, the GP moiety of the VP prohormone is glycosylated, disulfide bonds are generated, and then the prohormone is packaged into a secretory vesicle. The correct sorting of the pro­ hormone into the secretory pathway requires the for­ mation of aggregates. Specific association of VP and NP-II results in dimeric and tetrameric units, which is

essential for sorting (De Bree and Burbach, 1994). The secretory vesicles are axonally transported to the axon endings in the posterior pituitary. Studies in the dog suggest that approximately 1.5 hours are required from the time of synthesis to possible release of the nonapep­ tide (Richter and Ivell, 1985). During this transport the neurophysin II, VP, and glyprotein are liberated by proteolytic cleavage. The C-terminal glycine residue of the VP molecule is finally amidated. The mature VP nonapeptide has a disulfide bridge between cysteine residues at positions 1 and 6 (Fig. 19.14). In most mammals VP contains an arginine residue at position 8 (arginine vasopressin, AVP). In the mem­ bers of the pig family, VP contains a lysine residue at position 8. This lysine vasopressin (LVP) is the only form present in the domestic pig. Heterozygotic pecca­ ries, warthogs, and hippopotami may possess both AVP and LVP. Substitution of the phenylalanine resi­ due at position 3 by isoleucine leads to arginine vaso­ tocin (AVT). This peptide has been identified as a uniquely nonmammalian substance and is probably the most primitive and certainly the most commonly occurring neurohypophyseal peptide. Neurohypophy­ sis of some strains of chickens (but not ducks or tur­ keys) contain not only AVT but also AVP (Choy and Watkins, 1986). In the rat and the calf the cDNAs encoding the neu­ rophysins of AVP and oxytocin have a high degree of homology. Amino acid analysis of the AVP- and oxytocin-related neurophysins also shows a strong ho­ mology for bovine, porcine, and equine neurophysin.

542

Jan A. Mol and Ad Rijnberk

Arginine vasopre s s in (AV P ) vasopres s i n (LVP ) Lysine (AVT ) Arginine vasotocin Conopressin 5 ( s na i l venom) Mesotocin ( MT ) (OT) Oxytocin

Cy . T yr . Phe . Gln . Asn .

"-

lIe Phe . Ile . Arg Ile I le -

y

. Pro . Arg . Gly-NHz Lys -

*

•

-

Ile Leu

*

FIGURE 19.14 Sequence comparison of vasopressin (VP) and oxytocin (OT). Lines between shaded boxes represent intra chain disulfide bridges. See legend to Fig. 19.6.

The glycoprotein sequence appears to be highly con­ served among various species, but its physiological role remains to be clarified (Majzoub, 1985). 3. Secretion The major determinant in the release of VP is plasma osmolality. Below a certain plasma osmolality thresh­ old, which may vary considerably between individu­ als, plasma VP concentration is suppressed to levels that allow maximal free water clearance (Wade et al., 1982). Once plasma osmolality rises to the threshold, the pituitary secretes VP. Although the concentration of plasma VP causing maximal antidiuresis is about 5-10 pmol / liter, the release of VP is related to plasma osmolality over a much broader range (Robertson, 1983). Biewenga et al. (1987) developed a nomogram for this relationship in the dog, which allows analysis of the osmoregulation of VP secretion in terms of sensi­ tivity and threshold. Apart from the influence of osmolality, a significant decrease in circulating blood volume may also cause enhanced VP secretion. The decrease in left atrial pres­ sure triggers receptors. Denervation of these receptors prevents the hypovolemia-mediated VP release. The functional properties of this regulatory system differ from those of the osmoregulatory system. Although the relationship between plasma osmolality and VP appears to be linear, the relationship between blood volume and plasma VP is best described by an expo­ nential function (Vokes and Robertson, 1985a). Under ordinary circumstances, changes in VP levels are not observed until blood volume decreases by 510%. With further decreases, plasma VP rises exponen­ tially. The relative insensitivity of VP to modest changes in blood volume means that this regulatory mechanism plays little or no role in the physiological control of water balance. Under ordinary circum­ stances, total body water rarely changes by more than 1-2%, an amount far too small to affect VP secretion through hemodynamic influences (Wang and Goetz, 1985). In dogs, 24 hours of fluid deprivation increases plasma VP because of changes in both extracellular volume and tonicity, but the increase in tonicity plays

a greater role than the reduction in volume (Wade et

ai., 1983). It has been proposed that hemodynamic influences modulate VP secretion by raising or lowering the os­ mostat (Robertson, 1978). According to this concept, a decrease in plasma volume and / or pressure will per­ mit normal osmoregulation of VP, but the threshold for release will be lowered by an amount proportional to the degree of hypovolemia or hypotension, associ­ ated with increased sensitivity of the VP response to rising plasma osmolality. Conversely, hypervolemia or hypertension increases the threshold and decreases the sensitivity of release (Quillen and Cowley, 1983). There are numerous other factors that may influence VP secretion. Of these, nausea / vomiting (Raschler et al., 1986) and insulin-induced hypoglycemia (Vokes and Robertson, 1985b) are the most potent. The effects of opiates and opioid pep tides on VP secretion have been studied for many years and have been the subject of some controversy. Both stimulatory and inhibitory effects have been reported. Apart from the substances, doses, and routes of administration used, the animal species is an important factor (Van Wimersma Greida­ nus, 1987). Intracerebroventricular administration of .a-END in conscious rats decreases basal and stimu­ lated VP release (ten Haaf et ai., 1986). An opioid inhibi­ tion of dehydration-induced VP release has been re­ ported in dogs (Wade, 1985). However, Hellebrekers et al. (1988) could not confirm this endogenous opioid modulation of osmolality-regulated VP release in con­ scious dogs subjected to hypertonic saline infusion. In contrast, sharp increases in plasma VP have been found in conscious dogs after intravenous administration of the ,u-type opiate receptor agonist methadone (Helle­ brekers et al., 1987).

4. Action The major role of VP is to regulate body fluid ho­ meostasis by affecting water resorption. An increase in plasma VP results in increased water retention, which maintains plasma osmolality between narrow limits. The antidiuretic effect is achieved by promoting the reabsorption of solute-free water in the distal and col-

543

Pituitary Function

lecting tubules of the kidney. In the absence of VP, this portion of the nephron is not permeable to water and the hypotonic filtrate of the ascending limb of Henle's loop passes unmodified through the distal tubule and collecting duct. In this condition urine osmolality de­ creases to around 80 mOsm / kg. The action of VP on the tubular cells is mediated by V2 receptors that, after VP binding, activate a signal transduction system lead­ ing to an intracellular increase in cAMP ( Jard, 1983). Besides this well-known role in the regulation of fluid homeostasis, VP exerts a large variety of effects. Among these is the vasopressor effect, which is medi­ ated by VI receptors (Liard, 1986). In addition, VP increases glucogenolysis by liver cells, increases ACTH release by adenohypophysis (Lowry et al., 1987), and has several effects on animal behavior (De Wied and Versteeg, 1979). The extrarenal effects on hepatocytes and vascular smooth muscle are exerted through cAMP-independent, calcium-dependent VI receptors ( Jard, 1983). There is some evidence that in birds AVT, apart from its established antidiuretic action, also has oxytocic properties and participates in normal oviposition (Shi­ mada et al., 1986, 1987). MT does not seem to function in oviposition, and its release was found to be negatively correlated with plasma osmolality (Koike et al., 1986). 5. Disease

Disorders of the hypothalamic-neurohypophyseal system resulting in deficiency or excess of VP are known to occur in domestic animals. Deficient VP re­ lease causes the syndrome of diabetes insipidus, which is characterized by persistent inappropriately dilute urine in the presence of strong osmotic stimuli for VP release. The disease is not uncommon in the dog and cat (see Feldman and Nelson, 1996; Rijnberk, 1996) and rare in other species. VP excess is known as the syndrome of inappropri­ ate ADH (SIADH) secretion or the Schwartz-Bartter syndrome. The VP release is inappropriately high in relation to the plasma osmolality. The resulting defect in water excretion causes hyponatremia, which is the hallmark of SIADH. This condition has been docu­ mented as a disease of the dog, and at the same time studies of the threshold and sensitivity of VP secretion revealed that it occurred in two forms (Rijnberk et al., 1988; Houston et al., 1989). The glucocorticoid excess in Cushing' s syndrome is accompanied by (mild) polyuria in the dog. A marked impairment of the osmolality-regulated AVP release at the pituitary level may, in concert with a partial AVP resistance at the kidney level, be the cause of this corticosteroid-induced polyuria (Biewenga et al., 1991 ).

Chronic liver disease is accompanied by both enhanced activity of the pituitary-adrenocortical axis and distur­ bances in sodium and water homeostasis. Also in these cases a profoundly impaired osmoregulation of AVP release was found (Rothuizen et al., 1995). In dogs with spontaneous pericardial effusion, moderately elevated plasma AVP levels declined rapidly after pericardio­ centesis (Stokhof et al., 1994).

6. Tests The diagnosis of VP deficiency requires that it be differentiated from other causes of water diuresis. Fol­ lowing exclusion of conditions such as hyperadreno­ corticism, hyperthyroidism, and hypercalcemia, both nephrogenic diabetes insipidus and primary polydip­ sia remain as the main differential diagnoses. Slightly elevated plasma osmolality may suggest neurogenic or nephrogenic diabetes insipidus, whereas low plasma osmolality may be observed in primary polydipsia. However, in many cases this parameter is not con­ clusive. The procedure that is widely used to differentiate these disorders is the modified water deprivation test, as introduced for the dog by Mulnix et al. (1976). In this test maximal urine concentration is induced by several hours of dehydration. Once a plateau in urine osmolality is reached, the effect of an injection with VP is investigated. A further increase in urine osmolal­ ity by 50% or more is regarded as diagnostic for VP de­ ficiency. Although these indirect criteria for VP secretion can usually differentiate between complete neurogenic and complete nephrogenic diabetes insipidus, they cannot differentiate among partial neurogenic, partial nephro­ genic, and dipsogenic polyurias. In these situations, direct measurements of plasma VP during hypertonic saline infusion are required (Biewenga et al., 1989). Hypersecretion of VP can only be diagnosed by measurements of the circulating concentrations of the hormone (Table 19.9). The basic criterion is the pres­ ence of "inappropriately" high VP concentrations in relation to the hypoosmolality of the extracellular fluid. The type of osmoregulatory defect can best be judged by repeatedly measuring plasma osmolality and VP during administration of hypertonic saline (Rijnberk et al., 1988). Details of this test are given in Section IV of this chapter.

B. Oxytocin 1. Gene Expression The oxytocin (OT) gene expression is predominantly found in the hypothalamic supraoptic nucleus and paraventricular nucleus (Ivell, 1986). The neurons proj-

544

Jan A. Mol and Ad Rijnberk TABLE

Species (n)

19.9

Concentrations of Neurohypophyseal Hormones in Plasma of Healthy Animals'

Peptide

Condition

Range (pmollliter) (mean

±

SEM)

Reference

Dog (5)

AVP

Basal

0.5-2.9

Dogterom et al. (1978)

Dog (12)

AVP

Basal

(3.3 ::': 0.4)

Simon-Oppermann et al. (1983)

Dog (9)

AVP

24 hours dehydration

(7.7 ::': 1.5)

Szczepanska-Sadowska et al. (1983)

Dog (25)

AVP

Basal

0.9-6.0

Biewenga et al. (1987)

Dog (10)

aT

First week of lactation

15-66

Eriksson et al. (1987)

Cow (334)

aT

Before milking

(1.6 ::': 0.6)

Schams (1983)

Sow (9)

aT

6-12 days postpartum

500 mg/ dl (> 12.9 mmol / liter), and diabetes mellitus is elimi­ nated, serum cholesterol's diagnostic accuracy is greatly increased. Therefore, increased cholesterol is a signal for further investigation of thyroid disease. Recently, cholesterol used in conjunction with the fT4 was found to have the highest predictive value among a battery of thyroid function tests (Larsson, 1988). Similarly, hypocholesterolemia has little value as an index of hyperthyroidism. On the other hand, choles­ terol decreases consistently in response to thyroxine replacement therapy, so it has value as a guide to

30-70 45 ::t 18

Free (mg/dl)

Ester (mg/dl)

Reference

30 ::t 10

63 ::t 23

Kritchevsky (1958) Boyd (1942) Morris and Courtice (1955)

37 ::t 15

73 ::t 15

51 ::t 20

59 ::t 19

34 ::t 9

66 ::t 19

Kaneko (1980) Kritchevsky (1958) Boyd (1944) Morris and Courtice (1955) Boyd and Oliver (1958) Boyd and Oliver (1958) Kritchevsky (1958) Boyd (1942) Leveille et al. (1957) Leveille et al. (1957) Kritchevsky (1958)

15.7

81.1

22 ::t 13

23 ::t 12

Kritchevsky (1958) Norcia and Furman (1959) Morris and Courtice (1955) Kritchevsky (1958) Boyd (1942) Boyd and Oliver (1958) Morris and Courtice (1955)

70 64 ::t 12 35.6 ::t 53.7

Kritchevsky (1958) Boyd (1942) Lennon and Mixner (1957) Lennon and Mixner (1957) Lennon and Mixner (1957) Lennon and Mixner (1957)

5.7-10.9

28-48

Reiser et al. (1959)

Thyroid Function

therapeutic response. In thyroidectomized horses with clinical evidence of hypothyroidism, Lowe et al. (1974) demonstrated a 50% decrease in serum cholesterol shortly after feeding iodinated casein.

B. Direct Tests of Thyroid Function A direct approach to thyroid evaluation is to mea­ sure the amount of the hormones in the blood. Because there are thyroid inhibitory effects among a wide vari­ ety of iodine-containing compounds, it is critical that any form of iodine-containing medication, including thyroid hormones, be uncovered in the history of the patient. As a general rule, any iodine-containing medi­ cation or thyroid hormones being given to the patient should be withdrawn for at least 2 weeks before any thyroid function tests are undertaken.

1. Serum Protein-Bound Iodine The classic protein-bound iodine test measured all the precipitable protein-bound hormonal forms (T4t T3, rT3) plus all coprecipitable iodine. Clearly, this method overestimated the amounts of hormone, but it was a useful index of thyroidal activity because the PBI is proportional to the amount of thyroid hormones. Be­ cause most of the iodine bound to plasma protein is hormonal iodine, the PBI mainly represents hormone concentration. The partitioning between hormonal and nonhormonal forms is quite constant, so that the PBI is directly proportional to the amount of thyroid hormone. The PBI method is fraught with logistical and techni­ cal difficulties and, though useful in its time, is no longer used. The sample and the chemical procedure are readily contaminated by any source of iodine: anti­ septics, internal or external iodine-containing medica­ ments, iodinated radiographic contrast media, and thy­ roid medications (Table 21.4). Reliability is improved by prior extraction of the thyrOid hormones from the serum with butanol (butanol-extractable iodine, BEl). A further refinement was to separate the hormones by column chromatography (Pilleggi et al., 1961; Fisher et al., 1965). Automation of the iodine assay increased its reliability, but immunoassays for thyroid hormones are now the tests of choice. Reference ranges for ani­ mals are given in Table 21.5; note that these ranges are considerably lower than in humans.

2. Competitive Protein-Binding Assays The elucidation of the protein-binding equilibrium and the competition between labeled and nonlabeled hormones for binding sites on the proteins provided

579

the basis for the determination of T4t and indirectly, the TBG and fT4:

(21.1 ) Since the TBG-T4 is the total T4t one can substitute and reformulate Eq. (21. 1 ):

(21.2) It can be seen that the amount of fT4 is directly proportional to the amount of T4 and inversely propor­ tional to the amount of TBG, and that knowing these two factors permits the calculation of fT4. The calcu­ lated fT4 is expressed as the free thyroxine index (FTI), but it is not valid for animals. T4 is now measured directly by competitive protein binding (CPB) assay, by radioimmunoassay (RIA), or by enzyme-labeled im­ munoassay (EIA). The TBG is measured indirectly by competitive binding of labeled T3 to a secondary bind­ ing agent such as red cells, resins, charcoal, or hemoglo­ bin. TBG can also be assayed directly by immunoassay.

a. Thyroxine by Competitive Protein Binding The competitive protein binding (T4-CPB) method was the forerunner of all the competitive protein bind­ ing methods for T4 and does not depend on the colori­ metric determination of iodine (Murphy and Pattee, 1964). The method is based on the competitive binding of TBG for patient T4 and l3l1-labeled T4. The labeled T4 and patient T4 bind to the TBG in proportion to their concentrations, so that labeled T4 binding is inversely proportional to patient T4 concentration. The use of TBG has been replaced by immunoglobulins as the binding proteins, so CPB is now seldom used. Refer­ ence values in animals are given in Table 21.6. There is minimal interference from diphenylhydan­ toin and salicylates when given in large amounts (Spar­ agana et al., 1969), so T4-CPB is a reliable method (Lucis et al., 1969). Comparisons with T4-RIA indicated no significant difference (Kaneko et al., 1975). Also, in horses, T4-CPB decreased dramatically to almost zero at 2 days after thyroidectomy (Lowe et al., 1974). Feed­ ing of iodinated casein to these horses resulted in a rapid rise of T4 to above normal. Iodinated casein contains about 2.5 f.Lg (3.85 nmol) T4 and 1 .25 f.Lg (1.61 nmol) T3 per gram (Kaneko, 1979).

b. Thyroxine by Radioimmunoassay Thyroxine by radioimmunoassay (T4-RIA) is a typi­ cal immunoassay method in which an antibody is used as the binding protein. Owing to their high specificities, antibodies are useful for the assay of proteins, polypep­ tides, and haptens, including most of the hormones.

580

J. Jerry Kaneko TABLE 21.4

Some Drugs Affecting Thyroid Function Tests

Duration of effect (average period)

Compound

Effect on 1311 uptake

Effect on PBI

Iodides Lugol's, cough syrup, vitamin preparations Iodine antiseptics Iodine-containing drugs

Decrease Decrease Decrease

Increase Increase Increase

None None None

10-30 days 10-30 days 10-30 days

X-Ray contrast media Iodoalphionic acid (pheniodal) Iodopyracet (Diodrast) Iodized oil (oleum iodatum) Most iv contrast media Most gallbladder media

Decrease Decrease Decrease Decrease Decrease

Increase Increase Increase Increase Increase

None None None None None

3-12 months 2-7 days 0.5-3 years 2-6 weeks 3 weeks-3 mo

Hormones Thyroid extract Triiodothyronine ACTH Estrogens Oral contraceptives Androgens

Decrease Decrease Decrease None None None

Increase Decrease None Increase Increase None

Increase Decrease Increase None None Increase

4-6 weeks 2-4 weeks 2 weeks

Thiocarbamide compounds Thiouracil Propylthiouracil Thiocyanate Pheny lbu tazone Bromides Antihistamines Diphenylhydantoin (dilantin)

Decrease Decrease Decrease Decrease Decrease Decrease None

Decrease Decrease Decrease None None None Decrease

None None None Increase

5-7 days 5-7 days 14-21 days 14 days 10-30 days 7 days 7-10 days

None

None

Increase'

Salicylates a

Effect on T3 uptake

Increase

Effect depends on the method.

The principles and details of radioimmunoassay are described in the chapter on clinical enzymology. Poly­ clonal antibodies give accurate results and are usually used for these hormone assays. Reference values in animals are given in Table 21.6. Belshaw and Rijnberk (1979) report that other than in hypothyroidism of dogs, only diphenylhydantoin, corticosteroids, or Cushing's disease result in very low concentrations of either T4-RIA or T3-RIA. This further emphasizes the need for a careful history of the patient prior to assay for thyroid hormones. The mean normal T4-RIA in dogs is 2.3 � 0.8 JLg/ dl (29.6 � 10.3 nmol/ liter) with an observed range of 0.6-3.6 JLg/ dl (7.7-46.3 nmol / liter) (Kaneko et al., 1978), which is comparable to the 1 .53.6 JLg/ dl (19.3-46.3 nmol/ liter) of Sims et al. (1977) and Belshaw and Rijnberk (1979). T4-RIA is now widely used in cats because of the high prevalence of hyper­ thyroidism in this species. The reference range for cats is 0.1-2.5 JLg/ dl (1 .3-32.3 nmol/ liter). In the horse, T4RIA is also quite low , 0.9-2.8 JLg/ dl (11.6-36.0 nmol/ liter). Messer et al. (1995) report a similar finding, a mean T4-RIA of 21.42 � 3.46 nmol / liter in 12 adult horses.

c.

Thyroxine by Enzyme-Labeled Immunoassay

An important advance in hormone assay is the de­ velopment of an enzyme-labeled immunoassay (EIA) test comparable in every way to radioimmunoassay except that the labeling is by an enzyme rather than radioiodine. This method has some obvious advan­ tages in that there is no need to use radioactivity and the enzyme can be assayed in any laboratory. One system labels T4 with malate dehydrogenase (MD), which, in competition with unlabeled T41 binds to anti­ body. This method has the acronym EMIT, for enzyme multiplied immunosorbent test.! This method has an­ other advantage in that the labeled and unlabeled frac­ tions need not be separated. The MD is inactive when bound to T� when it binds to the immunoglobulin, it is activated. The activity of malate dehydrogenase is assayed by standard enzyme methodology, and the T4 is read from a standard curve as in a radioimmuno­ assay. Another system uses two separate recombinant fragments of the enzyme f3-galactosidase (Horn et al., I BioRad Laboratories, Richmond, CA 94804; ICL Scientific, Foun­ tain Valley, CA 92708.

581

Thyroid Function TABLE 21.5

Concentration of Serum Protein-Bound Iodine in Domestic Animals

Species

PHI (pg/dl serum)"

Reference

Dog

2.3 (1.5-3.5) 3.4 ± 1.0 (1.5-5.1) 2.3 ± 0.8 (1.1-4.3) 1.99 ± 0.24 2.6 ± 0.18 2.7 (1.8-4.5)

Siegel and Belshaw (1968) Mallo and Harris (1967) Quinlan and Michaelson (1967) Theran and Thornton (1966) O'Neal and Heinbecker (1953) Kaneko (1980)

Cat

1.86 ± 0.29 (1.2-2.5) 2.2 ± 0.6 (1.5-3.5) (1.67-2.7)

Kaneko (1980) Irvine (1967b) Kaneko (1964) Trum and Wasserman (1956)

Pig Landrace Large white

2.7 ± 0.1 4.4 ± 0.2

Sorenson (1962) Sorenson (1962)

Dairy Cattle Lactating Pregnant heifers Nonpregnant heifers

(2.73-4.11) 3.7 ± 0.3 5.0 ± 0.7 3.3 ± 0.1

Long et al. (1951) Sorenson (1962) Sorenson (1962) Sorenson (1962)

Beef cattle

2.19

Long et al. (1951)

Horse, Thoroughbred

Steer

3.5 (2.5-6.0)

2.5 ± 0.3

Sorenson (1962)

Sheep Lamb

3.8 ± 1.0 (3.6-4.0) (4-13)

Hackett et al. (1957) Falconer (1987)

Goat, miniature

4 (2-5)

Ragan et al. (1966)

a Values are means with their standard

deviations, if available. Ranges are given in parentheses.

1991). The individual fragments, enzyme donor (ED) and enzyme acceptor (EA), are inactive, but when they recombine, they form the active enzyme. Thyroxine is bound to ED. In the presence of thyroxine antibody, the thyroxine-ED-antibody complex inhibits recombi­ nation with EA and no active f3-galactosidase is formed. When sample thyroxine and a standard amount of thyroxine-ED are mixed, they compete for a standard amount of antibodies. Thus, unbound thyroxine-ED accumulates in direct proportion to the amount of sample thyroxine. The unbound thyrox­ ine-ED is free to bind with EA to form the active f3galactosidase. In this case, enzyme activity is directly proportional to the amount of thyroxine. This method had a very high correlation coefficient compared to that of an RIA (r = 0.969), an EIA (r = 0.966), and a fluorescence polarization immunoassay (FPIA, r =

0.939). Another variation of the nonisotopic immunoassay is the fluorescence immunoassay, in which a fluoro­ chrome is tagged to the T4• The T4 concentration is inversely proportional to the fluorescence, as in the RIA. This is also a very sensitive test, but it requires a sensitive spectrofluorometer for the assay. Enzyme immunoassays have not replaced radioim­ munoassays for hormones, but the principle is now widely used for antigen or antibody assays using horseradish peroxidase (HRP) as the enzyme label. The

procedure is popularly known as the enzyme-labeled immunoadsorbent assay, or by its acronym, ELISA. This procedure has been adapted for T4 by labeling T4 with HRP, then coupling the HRP to a dye that indi­ cates enzyme activity and hence T4 concentration.

d. Triiodothyronine by Radioimmunoassay Triiodothyronine is assayed by radioimmunoassay (TrRIA), and reference values in animals are given in Table 21.6. In the dog, the mean normal TrRIA is 107 ± 18 ng/ dl (1.6 ± 0.3 nmol/ liter) with an ob­ served range of 82-138 ng/ dl (1.26-2.12 nmol / liter) (Kaneko, 1980). It closely parallels T4-RIA in the dog, so that the simultaneous determination of T4-RIA and TrRIA will increase the diagnostic accuracy of either one alone. In cats, TrRIA is less widely used than T4-RIA. The reference range for cats is 15-50 ng/ dl (0.23-0.77 nmol! liter). In the horse, Messer et al. (1995) report a mean TrRIA of 0.85 ± 0.52 nmol/ liter.

3. "Free" Thyroxine and "Free" Triiodothyronine "Free" thyroXine (fT4) is the unbound fraction of the total circulating T4t and its concentration is con­ trolled by the equilibrium between TBG and TBG-T4 (Section III). Equilibrium dialysis or ultrafiltration are the best methods for determining the free hormones, but they are too labor intensive for use in most clinical

582

J. Jerry Kaneko TABLE 21.6 Species

Dog

Serum Thyroxine and Triiodothyronine Concentrations in Animals"

T, CPB (lLgldl)

T, RIA (lLgldl)

T, RIA (ngldl)

0.3-2.3 (1.3 ::':: 0.5)

Kaneko (1980)

0.6-3.6 (2.3 ::':: 0.8) 0.74-4.1 (2.48 ::':: 0.52) 1.13-4.71 (3.1 ::':: 1.1)

Male Female

45-175 (94 ::':: 24) 82-138 (107 ::':: 18) 45.4-1 17.5

0.1-2.5 (0.1 ::':: 0.5) 1.0-3.6 (1.7 ::':: 0.5) 0.8-3.7 (2.1 ::':: 0.6) (0.95 ::':: 0.5)

Larsson (1988)

Bigler (1976)

15-104 (52.3 ::':: 1.8)

Ling et al. (1974) Peterson et al. (1983) Kaneko (1980)

(1.55 ::':: 0.27) 2.0-3.3 (2.67 ::':: 0.50) 2.1-2.8

Weanling

Kaneko (1980)

Bigler (1976)

0.9-2.8 (1.9) 1.46-3.38 (2.57 ::':: 0.71) (2.70)

Foal

Belshaw and Rijnberk (1979)

Kaneko (1980)

0.8-3.8 (2.1 ::':: 0.1) Horse

Kaneko et al. (1978)

Kallfelz (1973)

1.4-3.6 Cat

Reference

(55.34 ::':: 33.85) 110-130 (117 ::':: 12) 48-62

Kallfelz and Lowe (1970) Thomas and Adams (1978) Irvine and Evans (1975) Messer et al. (1995) Osame and Ichijo (1994) Glade and Luba (1987)

McCrady et al. (1973)

Cow

4.2-8.6 (6.3 ::':: 1.0) (5.10 ::':: 1.30)

Pig

(2.10 ::':: 0.42)

Kallfelz and Erali (1973)

Goat

(5.25 ::':: 2.08)

Kallfelz and Erali (1973)

Sheep

(6.05 ::':: 1.64)

Kallfelz and Erali (1973)

Monkey (Macaca mlliata)

(4.1 ::':: 0.6)

Baboon

(160 ::':: 34)

(9.9 ::':: 2.2)

(121 ::':: 18)

a

Kallfelz and Erali (1973)

Be1chetz e/ al. (1978) Maul et al. (1982)

Observed ranges; means ::':: SO in parentheses.

laboratories. An equilibrium dialysis method for free thyroxine has recently become commercially available. Other methods for free thyroid hormone assay have been reviewed by Wilke (1986). Labeled thyroxine­ analog methods of determining free T4 are inaccurate and should not be used (Alexander, 1986). Free T4 as determined by RIA (fT4-RIA), however, is accurate and ought to be more widely used in conjunction with the T4-RIA and TrRIA. The fT4 concentration for dogs is 0.52-2.7 ng/ dl (6.7-34.7 pmol / liter) (Larsson, 1988) and for the beagle dog is 3.53 ± 0.34 ng / dl (45.4 ± 4.4 pmol/ liter) (Mi­ chaelson, 1969). Eckersall and Williams (1983) found the total T4-RIA by a commercial kit for humans to be

inaccurate, but the fT4-RIA was highly accurate and readily discriminated hypothyroid dogs. Larsson (1988) found that the fT4 and the serum cholesterol were the best combination of initial screening tests to detect hypothyroid dogs. On the other hand, in the hyperthyroid cat, Hays et al. (1988) found no differ­ ences in dialyzable T4 from the normal. They infer that the total T4 is sufficient for diagnosis and that the free hormone is not needed in the cat. In the horse, Messer et al. (1995) report a mean fT4-RIA of 14.0 ± 1 .16 pmol/ liter. The fTrRIA parallels fT4-RIA in its binding charac­ teristics. Since fT3 is the physiologically active form of the hormone, it is potentially the single most reliable

583

Thyroid Function

test of thyroid function. In the horse, Messer et al. (1995) report a mean fT3-RIA of 0.89 ± 0.53 pmol / liter.

tivity of ,B-galactosidase is inversely proportional to the amount of TBG in the serum.

4. Triiodothyronine Uptake

6. Free Thyroxine Index

The uptake of radiolabeled triiodothyronine by red cells (T3 uptake) in vitro as a test of thyroid function was reported by Hamolsky et ai. (1959). This test mea­ sured the partitioning of radioiodine-labeled T3 be­ tween TBG and a secondary binding agent, in this case, RBC. Serum TBG is the primary binding agent, and it binds T4 more firmly than it does T3 (Section VI). 3 Therefore when l 1I-T3 is added to the serum, it will bind first to any unbound receptor sites on the TBG. When RBC or another secondary binding agent such as resin or charcoal is added, excess unbound labeled T3 will bind to the secondary agent. The radioactivity of the RBC (or resin) will thus be inversely proportional to the amount of unbound TBG. The T3-uptake test is therefore an indirect measure of the amount of un­ bound TBG. The uptake is low in hypothyroidism where the low T4 leaves excess unbound binding sites on the TBG and high in hyperthyroidism where there are few binding sites on the TBG. The T3 uptake should not be confused with the TrRIA. There are many modifications to the test, basically in the choice of secondary binding agents. Pain and Oldfield (1969) surveyed six T3-uptake methods in­ cluding the red-cell uptake, resin uptake, resin-sponge uptake, Sephadex, charcoal-hemoglobin, and the thy­ roid binding index. They concluded that the original RBC T3 uptake gave overall satisfactory results, but only two of the newer methods, Sephadex and char­ coal, gave satisfactory results for both hyper- and hy­ pothyroidism. On the basis of ease of performance, the T3 uptake by charcoal-hemoglobin (Irvine and Stande­ ven, 1968) is the test of choice. The Truptake test is widely used in humans for the indirect assay of unbound TBG, from which the free thyroxine index (FTI) is calculated. In animals, how­ ever, the results are so poorly correlated with thyroid activity as to be virtually useless. Wide variations in the hormone binding proteins and their degree of binding account for these inaccuracies (Table 21.1).

The total T4 represents the thyroxine bound to TBG (TcTBG) and is readily measured as the T4-RlA. The T3 uptake or the T4 uptake as described earlier repre­ sents the unbound TBG and, as indicated previously, is also easily measured. Referring to the equilibrium equation (21.2), it can be readily seen that

5. Thyroxine Uptake

This assay is a modification of the ,B-galactosidase EIA described in Section VIII.B. (Horn et ai., 1991). In this case, no antibodies are added to the system, and the thyroxine enzyme donor fragment is free to bind with TBG in the serum. The remaining thyroxine en­ zyme donor fragment binds to the thyroxine acceptor fragment to generate the active enzyme. Thus, the ac-

fT4 = k

TcTBG = T4 TBG

X

T3 uptake (or T4 uptake) = FTI.

The product, T4 X T3 uptake, is the free thyroxine index (FTI) and is the calculated index of the amount of fTT4. The FTI is still widely used in humans, where it correlates well with thyroid disease (Wilke and East­ ment, 1986), but is invalid for use in animals. The free hormones should be determined in animals using di­ rect fT3 and fT4 immunoassays. 7. Thyroid-Binding Globulin, Thyroglobulin, and Thyroid Autoantibodies

Thyroid-binding globulin (TBG) and thyroglobulin (Tg) or colloid are measured by RIA. The standard technique for thyroglobulin antibodies (TgAb) is hem­ agglutination, but it is being supplanted by the ELISA (Voller et al., 1980) as a superior method (Roman et al., 1984 ) . Haines et al. (1984), using ELISA, detected TgAb in a high percentage of dogs with hypothyroidism, dogs with other endocrine diseases, and dogs that were closely related to the TgAb-positive dogs, but found a low percentage in healthy unrelated dogs. They con­ clude that thyroid autoimmunity is strongly geneti­ cally influenced in the dog. 8. Thyroid-Stimulating Hormone

Serum TSH is measured by radioimmunoassay (TSH-RIA) or by the two-site immunoradiometric assay (IRMA) in humans. The cross reactivity of the anti-human-TSH antibody for dog TSH is insufficient for it to be used as an accurate diagnostic test in the dog. Using the human TSH-RIA assay method for the dog, the normal values were 5.9 ± 4. 1 /LU / ml (Kaneko et ai., 1975), and 3.50 ::t:: 1 .67 /LU / ml (Larsson, 1981). The human assay method used in monkeys (M. mu­ iatta ) gave 0.2-2.6 /LU/ ml ( 1 .53 /LU/ ml) (Belchetz et ai., 1978 ) . A TSH-RIA using an anti-canine TSH has become commercially available specifically for use in dogs.

584

9. Thyrotrophin-Releasing Factor

J. Jerry Kaneko

As with TSH, no reliable thyrotrophin-releasing fac­ tor (TRF) assay has been developed for use in domestic animals. Purified TRH, however, is available and is used in the TRF response test.

C. Radionuclide Uptake Tests of Thyroid Function 1. Radioiodine Uptake Radioiodine is taken up by the thyroid gland in exactly the same manner as the nonradioactive isotope, and its uptake remains as one of the most definitive tests for thyroid function. With the development of direct hormone assays, radioiodine uptake tests are now largely used in nuclear medicine for imaging and localizing of active thyroid nodules and "hot spots" and for estimating therapeutic doses of radioiodine. Radioactive 99mTc-pertechnetate is also taken up and is now widely used for imaging. The normal uptake at 72 hours is 10-40% in the dog (Kaneko et al., 1959), compared to a range of 7-37% reported by Lombardi et al. (1962). The normal uptake at 24 hours is 7-35% and at 48 hours is 8-38%. The correlation of the 72-hour uptake with thyroid function has made it a reliable diagnostic test of thyroid disease in dogs. The normal uptake in cats at 24 hours is 9.2% (Peterson et al., 1983) whereas the maximal uptake is about 33% at 3-5 days postinjection (Broome et al., 1988). Iodine-containing compounds, including exoge­ nous thyroid hormone, also interfere with thyroid up­ take (Table 21.4), so the uptake test should be deferred for at least 2 weeks if iodine compounds have been administered.

2. The Conversion Ratio The conversion ratio (CR) gives the fraction of the injected radioiodine that has been converted to thyrox­ ine during the radioiodine uptake test. The serum pro­ teins with their bound radioactive hormones are sepa­ rated from non-protein-bound serum iodides using a resin and then counted for radioactivity (Scott and Reilly, 1954; Zieve et al., 1956). The result, when ex­ pressed as the ratio of protein-bound radioactivity to total plasma radioactivity, is the conversion ratio. Therefore, it reflects the amount of labeled thyrOid hormone that was formed by the gland from the in­ jected radioiodine. A CR of 2-6% has been reported (Lombardi et al., 1962) for normal dogs. This test is useful in the diagnosis of hyperthyroidism in humans, but its value in animals is unknown.

3. Thyroxine Secretion Rate The output of thyroxine by the thyroid gland is also a direct indicator of thyroid function and can be determined by several methods. Most data on thyrox­ ine secretion rate (TSR) in domestic animals have been obtained by a technique based on the amount of exoge­ nous L-thyroxine necessary to inhibit the release of radioactivity by the thyroid, that is, by thyroxine sup­ pression. Another method has been to calculate the TSR from the fractional turnover rate of injected radio­ thyroxine and the T4 or PBI. The TSR has been deter­ mined in many domestic animals, including the cow (Pipes et al., 1963), sheep (Henneman et al., 1955), goat (Flamboe and Reineke, 1959), pig (Sorenson, 1962), horse (Irvine, 1967), and dog (Kallfelz, 1973). A wide variation in the reported values is probably the result of differences in technique and conditions of study. The TSR is also affected by age, lactation, diet, season, and training in the case of horses. In animals, the TSR appears to vary from a low of about 0.108 mg (0.139 /Lmol) 1 100 kg 1 day in the horse to a high of about 0.46 mg (0.59 /Lmol ) / kg l day in the cow or 0.49 mg (0.63 /Lmol)/ kgl day for the dog.

D. Trophic Hormone Response Tests

1. Thyroid-Stimulating Honnone Response The response of the thyroid to TSH injection is a means of evaluating thyroid activity, as well as of dif­ ferentiating a primary hypothyroidism due to a thy­ roid lesion from a hypothyroidism secondary to a pitu­ itary lesion. The responsiveness of the thyroid to the TSH injection is evaluated by an increase (or failure to increase) as evidence of thyrOid activity (or lack of activity). A variety of dosages of TSH have been used, as well as a variety of procedures, to detect this increase in activity, among them the radioiodine uptake test (Kaneko et al., 1959), the radioiodine uptake curve (Siegel and Belshaw, 1968), the PBI (Siegel and Bel­ shaw, 1968), the TrCPB (Hoge et al., 1974), the T4-RlA (Kaneko et al., 1978), and, more recently, imaging. In the general procedure, thyroidal activity (or lack of activity) is first established by the measurement of se­ rum hormones, and this is followed by the TSH re­ sponse test. In a primary hypothyroidism where the lesion is localized in the thyroid, there is no response to the exogenous TSH. If the hypothyroidism is due to a pituitary hypofunction with a deficiency of TSH, or a hypothalamic lesion with a lack of TRH, the thy­ roid will respond to the exogenous TSH, as shown by a significant increase in serum hormone concentrations. Glucocorticoids and phenylbutazone are also well known to depress thyroid activity, so the TSH response

585

Thyroid Function

test is useful in detecting low hormone concentrations due to drugs or in Cushing's disease. A useful procedure for the TSH response test is first to obtain a serum sample for baseline T4 or T3, then to inject 10 IU of bovine TSH intravenously. Oliver and Waldrop (1983) and Held and Oliver (1984) recom­ mend a minimum of 5 IU for the dog and horse, respec­ tively. After 4 hours, a second serum sample is taken and hormone again measured. The normal response in dogs is a doubling or more of the hormone above baseline level. In the primary hypothyroid individual, there a virtual absence of a response. In the drug­ induced or Cushing's patient with low hormone there will be a response to well within the normal hormone concentrations. The secondary (TSH) or tertiary (TRH) hypothyroid patient will have a response similar to the drug-induced or Cushing's patient. In cats, Peterson et ai. (1983) found T4 to increase almost threefold above the baseline at 4 hours post-TSH. In the baboon, the TSH response test had peaks of more than double the baseline at 8 and 12 hours for T3 and T4I respectively (Maul et ai., 1982). In horses, Messer et ai. (1995), at 6 hours after TSH administration, found significant increases in T4 , T3, and IT4I but not significantly in IT3. They suggest that the TSH response test may not be as valuable for thyroid disease diagnostics in the horse as it is in the dog and cat.

2. Thyrotrophin-Releasing Factor Response The response to thyrotrophin-releasing factor (TRF) as developed for humans has been used in dogs and cats. In the dog, there was no T4-RIA, TTRIA, or T3 uptake response to exogenous human TRF (Kraft and Gerbig, 1977). TSH could not be measured, but presum­ ably the human TRF did not stimulate TSH release by the pituitary. Larsson (1981) measured TSH after giving TRH, and, although a significant difference was found, the overlap was too great to be of diagnostic value. Lothrop et al. (1984), however, found a doubling of T4-RIA at 6 hours postinjection in dogs and cats. Therefore, the TRF response test must be further evalu­ ated in animals. Some ( Jones, 1993) recommend the TRH response test in cats in preference to the TSH test, particularly in hyperthyroid cats. At 4 hours after the IV administration of TRH, normal cats have a greater than 50% increase in their serum T41 whereas hyperthy­ roid cats have a minimal or no increase.

3. Triiodothyronine Suppression Test In cats, T3 has an inhibitory effect on the secretion of TSH and injection of T3 is followed by a fall in serum T4 levels. In hyperthyroid cats, because of their autonomous production of T4I administered T3 will

have no suppressing effect on their high levels of serum T4. For the test, after a pretest serum sample is taken, 25 J.Lg T3 is given t.i.d. for 2 days. Four hours after the last dose, a serum sample for T4 is taken. In hyperthy­ roid cats, there is little or no fall in serum T4I in contrast to a greater than 50% suppression in normal cats ( Jones, 1993).

XI. DISEASES OF THE THYROID Thyroid disease has been most extensively studied and reviewed in the dog (Belshaw, 1983). In the cat, the incidence of hyperthyroidism has increased greatly in the last decade and is the most frequently encoun­ tered endocrinopathy in this species (Peterson et al., 1994). In the horse, Lowe et al. (1974) described the clinical effects of experimental thyroidectomy in mares and stallions, and hypothyroidism remains as a fre­ quent consideration in breeding problems. In rumi­ nants, a congenital goiter has been described in Merino sheep and in Afrikander cattle. Local enlargements or nodules are also seen in all animals; these are usually benign tumors.

A. Goiter Goiter may be defined as an enlargement of the thyroid gland that is not due to inflammation or malig­ nancy. There are two general types of goiters: (1) non­ toxic goiters, which produce either normal amounts of hormone (simple goiter) or below-normal amounts of hormone (hypothyroid), and (2) toxic goiters, which produce excess amounts of hormone (hypertrophy). Furthermore, a defect or deficiency at any trophic step can also result in thyroid disease. Iodine deficiency (endemic goiter) is well known in iodine-deficient ar­ eas of the world. Goitrogenic materials, either natural substances or drugs, induce goiters by their blocking effects on steps in the hormonogenic pathways. There are also rare types of familial goiters associated with defects in hormone synthesis (dyshormonogenesis) in humans (Stanbury and Dumont, 1983) and that find their counterparts in Merino sheep (Rac et ai., 1968) and in Afrikander cattle (Van Zyl et al., 1965). Falconer (1987) reviewed the congenital goiter in the Merino sheep in which the fundamental defect is a failure of thyroglobulin synthesis. The goiter is inherited as an autosomal recessive and is frequently seen in Australia. The similar congenital goiter in the Afrikander cattle (Ricketts et al., 1985) is also inherited as an autosomal recessive. These cattle have a thyroglobulin synthesis defect involving defective gene splicing of the thyro­ globulin gene transcript. In Bongo antelopes, goiter

586

J. Jerry Kaneko

seen in a group of adults was associated with synthesis of an abnormal 19S thyroglobulin (Schiller et ai., 1995). The goitrous antelopes were hormonally euthyroid, but had other manifestations of hypothyroidism such as reproductive difficulties. Iodine-deficiency goiter observed in seven thoroughbred foals in the northern island of Hokkaido, Japan (Osame and Ichijo, 1994), was attributed to iodine deficiency in the soil of the region. All foals had readily detectable thyroid enlarge­ ments, and four of the seven had clinical signs of thy­ roid deficiency as well. The goiters receded after iodine supplementation of the feed. Simple goiter is a compensatory increase in thyroid glandular mass (hyperplasia and hypertrophy) so that the gland maximizes iodine uptake and is able to syn­ thesize and release a normal amount of T4. At this time, the patient is physiologically normal, but the gland can become quite large. Ultimately, in iodine defi­ ciency, the goitrous gland fails to synthesize sufficient T4 and hypothyroidism occurs.

B. Hypothyroidism Hypothyroidism may be the result of a variety of causative factors. Thyroiditis, with similarities to Hashimoto's thyroiditis in humans, has been reported in about 12% of beagle dogs (Beierwaltes and Nishi­ yama, 1968). Antithyroglobulin antibodies were found in these dogs. In the adult dog, follicular atrophy is probably the most common cause of hypothyroidism (Clark and Meier, 1958). Finally, hypothyroidism may be secondary to a pituitary insufficiency. The hypothyroid dog is typically obese, lethargic, has myxedema, a dry skin, and a sparse hair coat. Hypothyroidism is therefore an important differential in the diagnoses of dermatoses. The requirement of T4 for normal reproduction, growth, and development is well known, so hypothyroidism is an important differ­ ential in reproductive failures. Experimentally thyroid­ ectomized mares and stallions failed to grow, were lethargic, had coarse, dull hair coats and increased serum cholesterol (Lowe et al., 1974). Hypothyroid horses tend toward obesity and crestiness. In the initial screen, an increased cholesterol is often the first clue to hypothyroidism. Larsson (1988) con­ cluded that IT4 and cholesterol are the best indicators of canine hypothyroidism. Definitive laboratory find­ ings in hypothyroidism of animals are a low T4 and / or T3 with little or no response to the TSH response test. Therefore, the recommended algorithm is to first obtain the total T4 and T3 (and the IT4 and IT3 if avail­ able). If the results are equivocal, this is followed by the TSH response test. Other definitive studies such 3 as the 1 11 uptake and the 99mTc-pertechnetate scans can

be used in specialized hospital settings for the identifi­ cation of isolated thyroid nodules. In human thyroid disease diagnostics, TSH is now considered to be the single best test of thyrOid status (Beckett, 1994) and to be more cost-effective than T4. TSH can now be readily assayed with a functional sensitivity of 0.01-0.02 mU / liter. Because of this sensi­ tivity, human thyroidologists now recommend that TSH be used as the initial test and that T4 or preferably IT4 be used only on a selected basis. In the event that similar degrees of accuracy and sensitivity evolve for dog TSH, this may well become the definitive test for hypothyroidism in the dog.

C. Hyperthyroidism Hyperthyroidism or toxic goiter is characterized by weight loss, hyperactivity, a voracious appetite, and increased thyroid hormones. Hyperthyroidism is rarely seen in dogs (Meier and Clark, 1958), but in the cat, in the past decade, the high incidence of hyperthy­ roidism has been recognized as a common endocrinop­ athy (Holzworth et ai., 1980; Peterson et ai., 1983). The most common form of hyperthyroidism in the cat is a functional thyroid adenoma. Increases in T4 and / or T3 are virtually pathogno­ monic signs of hyperthyroidism in the cat. In 131 cases, T4 was increased to between 4.0-54.1 J,Lg / dl (51.5696.3 nmol/ liter) in all cats, and T3 was between 541000 ng / dl (0.83-15.36 nmol/ liter) in 97% of the cats (Peterson et ai., 1983). Hays et ai. (1988) suggest that T4 and T3 are sufficient for diagnosis and that the free hormones are not needed for the diagnosis. In the 131 cases, the mean 24-hour 1311 uptake was 39.1%, com­ pared to 9.2% in normal cats. There was no increment of response above the baseline value to the TSH re­ sponse test in these cats, as would be expected for a tumor. 99mTc-pertechnetate scans demonstrated in­ creased uptake and size in one or both lobes of the thyroid. The TRH response test is equivocal for use in the hyperthyroid cat, but Jones (1993) found little or no increase in serum T4 in these cases. The T3-suppression test similarly had no decrease in serum T4 in hyperthy­ roid cats and is currently considered to be the most useful of the function tests in hyperthyroid cats.

D. Tumors of the Thyroid Gland Except for the dog and cat, tumors of the thyroid are rare in animals (Lucke, 1964). In the dog, Brodey and Kelly (1968) found no evidence of clinical thyroid disease associated with thyroid tumors. Loar (1986), however, found that about 20% of thyroid tumors were

Thyroid Function

functional. Belshaw (1983), using scintigraphic im­ aging in dogs, identified functional thyroid tumors and their metastases. Interestingly, a majority of thyroid carcinomas in dogs were associated with high serum thyroglobulin (Tg), but there was no direct correlation with T4 (Verschueren and Gosling, 1992). In feline hyperthyroidism, a functional thyroid ade­ noma is the most common finding.

References Adams, D. D. (1958). J. c/in. Endocrinol. Metab. 18, 699. Akasha, M., and Anderson, R. R. (1984). Proc. Soc. Exp. Bioi. Med. 177, 360. Alexander, N. M. (1986). c/in. Chem. 32, 417. Barua, J., Cragle, R. G., and Miller, J. K. (1964). f. Dairy Sci. 47, 539. Beckett, G. J. (1994). JIFCC 6, 368. Beierwaltes, W. H., and Nishiyama, R. H. (1968). Endocrinology 83, 501. Belchetz, P. E., Gredley, G., and Himsworth, R. (1978). J. Endocrinol. 76, 427. Belshaw, B. E. (1983). In "Textbook of Veterinary Internal Medicine," Vo!' II (S. J. Ettinger, ed.), 2nd ed., p. 1592. Saunders, Philadel­ phia, Pennsylvania. Belshaw, B. E., and Rijnberk, A. (1979). J. Am. Allim. Hosp. Assoc. 15, 17. Belshaw, B. E., Cooper, T. B., and Beeker, D. V. (1975). Endocrinologtj

96, 1280. Bigler, B. (1976). Schweiz. Arch. Tierheilk. 118, 417. Boyd, E. M. (1942). J. BioI. Chem. 143, 131. Boyd, E. M. (1944). Ct11lt1d. f. Res. 22, 39. Boyd, G. S., and Oliver, M. F. (1958). III "Cholesterol" (R. P. Cook, ed.), p. 187. Academic Press, New York. Brodey, R. S., and Kelly, D. (1968). Callcer 22, 406. Broome, M. R., Turre\, J. M., and Hays, M. T. (1988). Am. f. Vet. Res. 49, 193. Bustad, 1. K., George, 1. A, Jr ., Marks, S., Warner, D. E., Barnes, C. M., Herde, K. E., and Kornberg, H. A (1957). Radiat. Res. 6, 380. Clark, S. T., and Meier, H. (1958). Zentmlbl. Veterillaermed. 5, 17. Cline, M. J., and Berlin, N. 1. (1963). Am. J. Physiol. 204, 415. Eckersall, P. D., and Williams, M. E. (1983). f. Small Allim. Pmct. 24, 525. Edelman, I. S. (1974). N. Engl. f. Med. 290, 1303. Falconer, I. R. (1987). Compo Path. Bull. 19(3), 2. Fisher, D. A., addie, T. H., and Epperson, J. (1965). J. c/ill. Endocrinol. Metab. 25, 1580. Flamboe, E. E., and Reineke, E. P. (1959). f. Animal Sci. 18, 1135. Freinke\, N., and Ingbar, S. H. (1955). f. Clill. Endocrillol. 15, 598. Glade, M. J., and Luba, N. K. (1987). Am. J. Vet. Res. 48, 578. Hackett, P. 1., Gaylor, D. W., and Bustad, 1. K. (1957). Am. J. Vet. Res. 18, 338. Haines, D. M., Lording, P. M., and Penhale, W. J. (1984). Am. J. Vet. Res. 45, 1493. Hamolsky, M. W., Golodetz, A., and Freedberg, A S. (1959). J. Clill. Endocrillol. Metab. 19, 103. Hays, M. T., Turrel, J. M. and Broome, M. R. (1988). f. Am. Vet. Med. Assoc. 192, 1. Henneman, H. A., Reineke, E. P., and Griffin, S. A. (1955). J. Allim. Sci. 14, 419. Hoge, W. R., Lund, J. E., and Blakemore, J. C. (1974). J. Am. Anim. Hosp. Assoc. 10, 167.

587

Holzworth, J., Theran, P., Carpenter, J. 1., Harpster, N. K. and Todoroff, R. J. (1980). J. Am. Vet. Med. Assoc. 176, 345. Horn, K., Castineiras, M. J., Ortola, J., Kock, R., Perriard, F. c., Bittner, S., Pairet, J. V., Ers, P., Boulanger, J., Zeidner, S., Maier, R., Boege, F., Dubois, H., McGovern, M., and Opitz, 1. (1991). Eur. J. Clill. Chem. Clin. Biochem. 29, 697. Irvine, C. H. G. (1967). f. Endocrinol. 39, 313. Irvine, C. H. G., and Evans, M. J. (1975). J. Reprod. Fertil. Suppl. 23, 709. Irvine, W. J., and Standeven, R. M. (1968). f. Endocrillol. 41, 31. Jones, B.R. (1993). Vet. Focus 3, 3. Kallfelz, F. A (1973). Am. f. Vet. Res. 34, 535. Kallfelz, F. A., and Erali, R. P. (1973). Am. J. Vet. Res. 34, 1449. Kallfelz, F. A, and Lowe, J. E. (1970). J. Am. Vet. Med. Assoc. 156, 1888. Kaneko, J. J. (1964). Proc. 10th Anllu. Conv. Am. Assoc. Equine Practitioners, De1lver, p. 125. Kaneko, J. J. (1979). Unpublished. Kaneko, J. J. (1980). In "Clinical Biochemistry of Domestic Animals" 0. J. Kaneko, ed.), 3rd ed., p. 491. Academic Press, New York. Kaneko, J. J., Tyler, W. S., Wind, A P., and Cornelius, C. E. (1959). f. Am. Vet. Med. Assoc. 135, 10. Kaneko, J. J., Baker, B., and Mills, R. (1975). Bull. Am. Soc. Vet. Clill. Path. 4, 30. Kaneko, J. J., Comer, K. M., and Ling, G. V. (1978). Calif Vet. 32, 9. Koppers, 1. E., and Palumbo, P. J. (1972). Med. Clill. North Am. 56, 1013. Kraft, W., and Gerbig, T. (1977). Dtsch. Tierarztl. Woche1lschr. 84, 185. Kritchevsky, D. (1958). "Cholesterol," p. 279. John Wiley & Son, New York. Larsson, M. (1981). Acta Vet. SCt111d. 22, 589. Larsson, M. (1988). f. Am. Anim. Hosp. Assoc. 24, 209. Larsson, M., Pettersson, T. and Carlstrom, A. (1985). Gen. Compo Endocrinol. 58, 360. Lennon, H. D., and Mixner, J. P. (1957). f. Dairy Sci. 40, 1424. Leveille, G., Fisher, H., and Weiss, H. S. (1957). Proc. Soc. Exp. BioI. Med. 94, 383. Ling, G. V., Lowenstine, 1. W., and Kaneko, J. J. (1974). Am. f. Vet. Res. 35, 1247. Lipman, 1. M. (1967). Am. J. Med. 43, 486. Loar, A S. (1986). In "Kirk's Current Veterinary Therapy" (R. W. Kirk, ed.), 9th ed., p. 1033. W. B. Saunders, Philadelphia. Lombardi, M. H., Comar, C. L., and Kirk, R. W. (1962). Am. f. Vet. Res. 23, 412. Long, J. F., Gilmore, 1. 0., Curtis, G. M., and Rife, D. C. (1951). J. Anim. Sci. 10, 1027. Lothrop, C. D., Jr., Tamas, P. M., and Fadok, V. A (1984). Am. J. Vet. Res. 45, 2310. Lowe, J. E., Baldwin, B. H., Foote, R. H., Hillman, R. B., and Kallfelz, F. A (1974). Cornell Vet. 64, 276. Lucis, O. M., Cummings, G. T., Matthews, S., and Burry, C. (1969). f. Nucl. Med. 10, 160. Lucke, V. M. (1964). J. Small Anim. Pmct. 5, 351. McCrady, J. D., Hightower, D., Williams, G. D., Kyzar, J. R., and Bridges, C. H. (1973). Cardiovasc. Res. Centro Bull. 12, 21. McKenzie, J. M. (1958). Endocrinology 63, 372. Mahley, R. W., and Weisgraber, K. H. (1974). Circ. Res. 35, 713. Mallo, G. 1., and Harris, A 1. (1967). Vet. Med. Small Anim. Clin. 62, 533. Maul, D. H., Rosenberg, D. P., Henrickson, R. V. and Kaneko, J. J. (1982). Lab. Anim. Sci. 32, 267. Messer, M. T., Ganjam, V. K., Nachreiner, R. F., and Krause, G. F. (1995). J. Am. Vet. Med. Assoc. 206, 63. Michaelson, S. M. (1969). Mod. Vet. Practice 50, 43. Morris, B., and Courtice, F. C. (1955). Q. J. Exp. Physiol. Cog. Med. Sci. 40, 127.

588

J. Jerry Kaneko

Murphy, B. E. P., and Pattee, C. J. (1964). J. Clin. Endocrinol. Metab. 24, 187. Norcia, L. M., and Furman, R H. (1959). Proc. Soc. Exp. Bioi. Med. 100, 759. Oliver, J. W. and Waldrop, V. (1983). J. Am. Vet. Med. Assoc. 182, 486. O'Neal, L. W., and Heinbecker, P. (1953). Endocrinology 53, 60. Osame, S., and Ichijo, S. (1994). J. Vet. Med. Sci. 56, 771. Pain, R W., and Oldfield, R K (1969). Tech. Bull. Regist. Med. Teclmol. 39, 139. Peterson, M. E., Kintzer, P. P., Cavanagh, P. G., Fox, P. R, Ferguson, D. c., Johnson, G. F., and Becker, D. V. (1983). J. Am. Vet. Med. Assoc. 183, 103. Peterson, M. E., Randolph, J. F., and Mooney, C. T. (1994). In "The Cat: Diseases and Clinical Management" (R G. Sherding, ed.), p. 1403. Churchill Livingstone, New York. Pilleggi, V. J., Lee, N. D., Golub, O. J., and Henry, R J. (1961). J. Clin. Endrocinol. Metab. 21, 1272Pipes, G. W., Bauman, T. R, Brooks, J. R, Comfort, J. E., and Turner, C. W. (1963). J. Anim. Sci. 22, 476. Pliam, N. B., and Goldfine, 1. D. (1977). Biochem. Bioplzys. Res. Commllll. 79, 166. Quinlan, W., and Michaelson, S. M. (1967). Am. J. Vet. Res. 28, 179. Rac, R, Hill, G. N., Pain, R W., and Mulhearn, C. J. (1968). Res. Vet. Sci. 9, 209. Ragan, H. A, Horstman, V. G., McClellan, R D., and Bustad, L. K (1966). Am. J. Vet. Res. 116, 161. Refetoff, S., Robin, N. 1. and Fang, V. S. (1970). Endocrino[ogtj 86, 793. Reiser, R, Sorrels, M. F., and Williams, M. C. (1959). Circ. Res. 7, 833 . Ricketts, M. H., Pohl, V., de Martynoff, G., Boyd, C. D., Bester, A J., Van Jaarsveld, P. P., and Vassart, G. (1985). EMBO J. 4, 731. Roman, S. H., Korn, F., and Davies, T. F. (1984). Clin. Chem. 30, 246. Schiller, C. A, Montali, R. J., and GroHman, E. F. (1995). Vet. Pathol. 32, 242. Scott, K G., and Reilly, W. A (1954). Metab. Clill. Exp. 3, 506.

Segal, J., Schwartz, H., and Gordon, A (1977). Endocrinology 101, 143. Siegel, E. T., and Belshaw, B. E. (1968). [n "Current Veterinary Ther­ apy" (R W. Kirk, ed.), 3rd ed., p. 545. Saunders, Philadelphia, Pennsylvania. Sims, M. H., Redding, R W., and Nachreiner, R F. (1977). J. Am. Vet. Med. Assoc. 171, 178. Sorenson, P. H. (1962). [n "Use of Radioisotopes in Animal Biology and the Medical Sciences," Vol. I, p. 455. Academic Press, New York. Sparagana, M., Phillips, G., and Kucera, L. (1969). J. Clin. Endocrinol. Metab. 29, 191. Stanbury, J. B. and Dumont, J. (1983). In "The Metabolic Basis of Inherited Disease" 0. B. Stanbury, J. B. Wyngaarden, D. S. Fred­ rickson, J. L. Goldstein, and M. S. Brown, eds.), 5th ed., p. 231. McGraw-Hill, New York. Tanabe, Y., Ishii, T., and Tamaki, Y. (1969). Gen. Compo Endocrinol.

13, 14.

Tapley, D. F. (1964). Mayo Clinic Pmc. 39, 626. Tata, J. R, Ernster, L., Lindberg, 0., Arrhenius, E., Pederson, S., and Hedman, R (1963). Biochem. J. 86, 408. Theran, P., and Thornton, G. W. (1966). J. Am. Vet. Med. Assoc. 148, 562. Thomas, C. L., and Adams, J. C. (1978). Am J. Vet. Res. 39, 239. Trum, B. F., and Wasserman, R H. (1956). Am. J. Vet. Res. 17, 271. Van Zyl, A, Schulz, K, Wilson, B., and Pansegrouw, D. (1965).

Endocrinology 76, 353.

Verschueren, C. P., and Goslings, B. M. (1992). [n "Clinico-pathologi­ cal and Endocrine Aspects of Canine Thyroid Cancer" (c. P. Verschueren, ed.), p. 95. C. P. Vershueren, Brussels. Voller, A, Bidwell, D. E., and Burek, C. L. (1980). Proc. Soc. Exp. Bioi. Med. 163, 402. Wilke, T. J. (1986). Clin. Clzell!. 32, 585. Wilke, T. J., and Eastment, H. T. (1986). Clin. Clzel1l. 32, 1746. Zieve, L., Vogel, W. c., and Schultz, A L. (1956). J. Lab. Clin. Med. 47, 663.

C H A P T E R

22 Clinical Repro ductive Endocrinology LARS-ERIK EDQ VIST AND MATS FORSBERG

1.

II.

III.

IV.

V.

ing of the complex endocrinological events that occur during normal and abnormal reproductive function, quantification of specific hormones is necessary. Ini­ tially, biological assay systems that measured the effect of a hormone on its target tissue were used; weight­ gain change was mainly used as the measure of hor­ mone concentration. These assays were relatively im­ precise, time consuming, and expensive. Later they were replaced by chemical determinations for steroid hormones. These assay systems usually required large volumes (often 1 liter) of plasma or serum, which made serial blood sampling of individual animals impossi­ ble. Some steroid hormone patterns were studied via urine analysis in 24-hour urine aliquot. Collection of such urine aliquots from domestic species was difficult and not practical under field conditions. Major progress in hormone analytical techniques occurred as the result of the development of immuno­ assay and related systems. The first of the assays, radio­ immunoassay, was developed in 1959 (Berson and Ya­ low, 1959) with the competitive protein binding assay following a few years later (Murphy, 1964). Nonradio­ metric assays, such as enzyme immunoassay, were de­ veloped in the 1970s (Engvall and Perlmann, 1971; van Weemen and Schuurs, 1971). These assay systems are sensitive, specific, and relatively inexpensive, and they require small amounts of assay material. They have been of special value for studying endocrinological reproductive function in domestic animals, in that they have made possible the study of dynamic endocrine changes through the assay of serial blood samples from the same animal. The immunoassay systems have also been useful as diagnostic aids for the identification

INTRODUCTION 589 A. Definition of Hormones 590 B. Chemical Classes of Reproductive Hormones 590 C. Hormone Receptors 592 D. Interconversion of Steroids in Target Tissues 593 E. Synthesis and Clearance of Hormones 593 ASSAY METHODS 594 A. Immunoassay 596 B. Summary 599 PHYSIOLOGY OF REPRODUCTIVE HORMONES IN THE FEMALE 600 A. Estrous Cycle 600 B. Control of the Corpus Luteum 601 C. Early Pregnancy 601 D. Pregnancy and Parturition 602 CLINICAL ASPECTS OF REPRODUCTIVE ENDOCRINOLOGY 603 A. Cattle 604 B. Sheep 606 C. Pig 606 D. Horse 607 E. Dog 610 F. Cat 611 GENERAL COMMENTS 612 A. Hormone Concentrations 612 B. Material for Analysis and Storage Effects 613 References 614

I. INTRODUCTION Clinical reproductive endocrinology includes the study of diseases of the endocrine glands involved in reproduction and their secretory products, the repro­ ductive hormones. To obtain a satisfactory understand-

CLINICAL BIOCHEMISTRY OF DOMESTIC ANIMALS, FIFTH EDITION

589

Copyright © 1997 by Academic Press All rights of reproduction in any form reserved.

590

Lars-Erik Edqvist and Mats Forsberg

and elucidation of clinical reproductive problems. In clinical practice, these methods are important from both a diagnostic and a therapy-monitoring point of view. Because this book deals with clinical biochemistry in domestic animals, the main emphasis of the chapter is on the determination of hormones and the use of the data as diagnostic aids. General reproductive endo­ crinology in domestic species is broadly covered. Read­ ers specifically interested in this subject are referred to specialized books dealing with this matter.

A. Definition of Hormones The best-understood humoral control system in the body is the endocrine system. This system uses specific messengers, termed hormones, to regulate important body functions. By the classical definition, hormones are chemical substances that are synthesized and se­ creted by ductless endocrine glands in minute quanti­ ties directly into the blood vascular system and are transported to a remote target organ where they regu­ late the rates of specific biochemical processes. The classic endocrine glands include the pituitary, thyroid, parathyroid, adrenal, pancreas, ovary, testis, placenta, and pineal glands. In the case of reproduction, the pituitary and pineal glands, gonads, and placenta play a primary role in controlling the system. Other endocrine glands such as the adrenal and thyroid glands also have some in­ fluence on reproductive function. Other organs such as the uterus and the hypothalamus, although they may not fulfill the strict definition of endocrine glands, can synthesize and secrete hormones that have a pro­ found influence on reproductive function.

B. Chemical Classes of Reproductive Hormones 1. Peptide and Protein Hormones

a. Releasing Hormones Several types of hormones are involved in the regu­ lation of reproduction. Releasing hormones are peptide hormones that are produced within the hypothalamus and transferred via the hypothalamo-hypophyseal portal veins to the adenohypophysis, where they regu­ late the synthesis and / or release of adenohypophyseal hormones. Gonadotropin-releasing hormone (GnRH), a decapeptide with the sequence pGlu-His-Trp-Ser­ Tyr-Gly-Leu-Arg-Pro-Gly-NH2, regulates the re­ lease of two important reproductive hormones, namely luteinizing hormone (LH) and follicle-stimulating hor-

mone (FSH). Thyrotropin-releasing hormone, a tripep­ tide (pGlu-His-Pro-NHz) that regulates the syntheSiS and release of thyroid-stimulating hormone (TSH), also causes the release of prolactin in several species.

b. Hypophyseal Hormones Luteinizing hormone and FSH are glycoproteins containing 13-25% carbohydrate. The molecular weight of LH in domestic animals (bovine, ovine, por­ cine, equine) is about 30,000. Ovine and equine FSH have molecular weights of about 32,000. Prolactin is a protein with a molecular weight of approximately 23,000 (bovine, ovine, porcine). The cells of the adenohypophysis can be divided into basophils (affinity for basic stains) and acidophils (affinity for acid stains). Luteinizing hormone and FSH are produced within basophilic cells; it has been dem­ onstrated that LH and FSH can be present within the same cell. Prolactin, on the other hand, has been local­ ized in acidophilic cells. As pointed out previously, the releasing hormones are important regulators of the synthesis and / or release of these hormones from the adenohypophysis.

c. Neurohypophyseal Hormones The posterior pituitary is responsible for storage and release of oxytocin, an important reproductive hor­ mone, and antidiuretic hormone (vasopressin). These two hormones are synthesized primarily in the regions of the paraventricular and supraoptic nuclei of the hypothalamus. The hormones are transported to the posterior pituitary by axoplasmic fluid. Release of these hormones occurs as a result of stimulation of the nerve cell bodies in the nuclei. Oxytocin is an octapep­ tide with a molecular weight of 1000.

d. Placental Gonadotropins A gonadotropin called pregnant mare serum gonad­ otropin (PMSG) is produced by mares during early pregnancy (days 40-140) by fetal trophoblastic cells of the chorionic girdle, which attach to, invade, and phagocytose the maternal epithelium and become em­ bedded within the uterus as specialized endometrial cups. This process begins on day 36 of pregnancy. The hormone has recently been renamed equine chorionic gonadotropin (eCG) due to its close relationship with human chorionic gonadotropin (hCG) (Farmer and Papkoff, 1979). No specific placental gonadotropins have been demonstrated in other domestic species.

e. Subunits Luteinizing hormone and FSH, as well as TSH, con­ sist of two nonidentical subunits designated ex and {3. Chemical and biological studies have indicated that the

591

Clinical Reproductive Endocrinology ACETATE

�

CHOLESTEROL

!

S-PREGNENOLONE

--

PROGESTERONE

17. - HYDROXY-

--

17. - HYDROXY -

!

P ROGESTERONE

PREGNENOLONE

CORTICOSTEROIDS

:

t

DEHYDROEPI ANDRO-

�

-

A NDROSTENED I ONE

STERONE

Jr

5-ANDROST ENED I OL

-

I

Jt

TESTOSTERONE

-

ESTRONE

-

ESTRAD I OL -17p

Jt

FIGURE 22.1 Pathway for the synthesis of biologically active steroids from acetate. The steroids secreted from the gonads and the adrenal are formed from acetate and cholesterol.

a-subunit is identical for these glycoprotein hormones within a species, whereas the f3-subunit, unique for each hormone within a species, determines the biologi­ cal activity. Subunits, by themselves, possess little or no biological activity. Individual subunits are probably not released into the circulatory system under normal physiological conditions.

2. Steroid Hormones Steroid hormones are derived from a common pre­ cursor molecule, cholesterol, via the metabolic path­ way schematically outlined in Fig. 22.1 . More than TABLE 22.1

1,500 biologically active steroids have been isolated from biological material or have been produced syn­ thetically. The molecular weight of steroid hormones is low, usually below 500 (Table 22.1). Examples of steroids that play an important role in reproductive processes are estrogens, androgens, and progestogens, with the main source being the gonads. The structures of the most important sex steroids are given in Fig. 22.2. The most common steroid hormones are usually designated by a trivial name, such as estradiol, testos­ terone, or progesterone. The International Union of Pure and Applied Chemistry (IUPAC) has recom­ mended systematic names for steroid hormones. These

Nomenclature" and Molecular Weights of Some Biologically Important Steroids and Prostaglandins

Trivial name

Systematic name

Molecular weight

Androstenedione

4-Androstene-3, 17-dione

286

Cortisol

11/3,17a,21-Trihydroxy-4-pregnene-3,20-dione

363

17f3-Estradiol

1,3,5(10)-Estratriene-3, 17f3-diol

272

Estrone

3-Hydroxy-l,3,5(1O)-estratrien-17-one

270

17a-Hydroxyprogesterone

17a-Hydroxy-4-pregnene-3,20-dione

331

Pregnenolone

3/3-Hydroxy-5-pregnen-20-one

317

Progesterone

4-Pregnene-3,20-dione

315

Testosterone

17/3-Hydroxy-4-androsten-3-one

288

PCF,"

9a,lla, 15-Trihydroxyprosta-5,13-dienoic acid

354

15-Keto-13,14-dihydro-PCF,"

9a,lla-Dihydroxy-15-ketoprost-5-enoic acid

354

" Revised Tentative Rules for Nomenclature of Steroids (1968). Biochim. Biophys. Acta 164, 453.

592

Lars-Erik Edqvist and Mats Forsberg

PROGESTERONE OH

OH

ESTRADIOL - 1 7/3

OH

ESTRONE

OH

,

'

" 0

1 5-KETO-13.1 4-DIHYDRO-PGF2<x

FIGURE 22.2 The number sequence for the carbon atoms of the steroid skeleton and lettering sequence for the four rings are shown for testosterone. The structures of three other important sex steroid hormones, estrone, estradiol-17/3, and progesterone, as well as the structures of prostaglandin F�, and its blood plasma metabolite 15-keto-13,14dihydroprostaglandin F2M are also depicted.

systematic names describe the chemical and steroiso­ metric characteristics of the particular steroid hormone (Table 22. 1).

3. Prostaglandins Prostaglandins constitute a group of 20-carbon un­ saturated fatty acids with molecular weights usually between 300 and 400 (Table 22. 1 ). Prostaglandins are not hormones in the strictest sense, and the expressions "parahormones" or "local hormones" have been used to describe these substances. This is because prosta­ glandins are not secreted from any particular gland, and the biological half-life of prostaglandins is usually extremely short, allowing, in most cases, only a local action. Several different prostaglandins are found in a number of types of mammalian tissues. One prosta­ glandin released from the uterus, prostaglandin F2a (PGF2a ), plays an important role in regulating repro­ ductive cycles in domestic species through the con­ trol of luteal activity in nonpregnant animals and the initiation of delivery in pregnant animals. The struc­ tures of prostaglandin Fm and its main metabolite, 15-keto-13,14-dihydroprostaglandin F2a, are given in Fig. 22.2.

C. Hormone Receptors Steroid hormones are fat soluble, so they are able to enter all cells of the body because the lipid cellular boundaries present no barrier to the steroid. The ques­ tion thus arises as to how specific actions occur in this physiological situation. Steroid hormone concentra­ tions in plasma are very low compared to those of many compounds, including their important precur­ sor, cholesterol. For example, plasma estrogens in nonpregnant domestic animals range from as low as 10 pmol / liter to as high as 150 pmol / liter. In this situa­ tion, most cells within the body have very low concen­ trations of estrogen. The specificity of tissue response occurs because cells of tissues that have need of estro­ gen stimulation should have receptors that enable those particular cells to concentrate the hormone within the cell and, more importantly, to elicit particu­ lar cellular responses. Hence the important generaliza­ tion that specific tissue responses require specific re­ ceptors to be present within the cell, in this case, for a particular steroid hormone. Binding of a hormone to a receptor in a target cell can be considered to be the primary event in the action of that hormone. Such a hormone-receptor interaction

Clinical Reproductive Endocrinology

will cause a measurable and distinct biological re­ sponse for each hormone. Receptors are defined by the criteria of having a limited binding capacity (receptors are saturable, which limits the number of hormone molecules that can enter a target cell), of binding spe­ cific hormones (e.g., estrogen receptors are specific for estrogenic compounds), and of creating a biological response upon binding. Steroid hormones enter cells by passive diffusion and bind to receptors inside the cell. It is assumed that only non-protein-bound or free hormone can enter target cells. The protein-bound steroid is thus virtually biologically inactive, and only the minute quantity of free (non-protein-bound) steroid can enter the cell. The protein binding is, however, reversible, and as free hormone leaves the vascular system into target cells, protein-bound hormone will be released to replace the deficit. On entry, the steroid interacts with its receptor in the target cell. When the steroid interacts with its receptor, a steroid-receptor complex is formed. The hormone-receptor complex is then activated" and alters gene expression. The target cell responds by in­ creased RNA synthesis with the transcription of spe­ cific mRNA that enters cytoplasm and stimulates pro­ tein synthesis. The specific effect of steroid hormones on target cells is altered cell function related to a change in the pattern of protein synthesis. Protein hormones such as LH and FSH do not enter the target cell to exert their effects, but interact with their receptors, which are located on the plasma mem­ branes of the cell. The binding of the hormone to the cell surface receptor activates one or more second mes­ sengers, such as cyclic AMP (3',S'-AMP). The second messenger is the intracellular mediator of many actions of LH and FSH in the ovary and the testis. Second messenger is thought to activate another intracellular enzyme, protein kinase, which will influence the trans­ port of cholesterol into the mitochondrion and the con­ version of cholesterol to pregnenolone, which is the rate-limiting step in the biosynthetic pathway for the steroids that play a significant role in reproductive pro­ cesses. II

D. Interconversion of Steroids in Target Tissues The effects of steroid hormones on cells can be accen­ tuated or modulated by the conversion of the entering hormone to another form. For example, many of the tissues that are particularly responsive to androgens have the enzyme Sa-reductase, which converts testos­ terone to 5a-dihydrotestosterone (5a-DHT). 5a-DHT has a much higher affinity for the androgen receptor within the target cell, which makes 5a-DHT more bio-

593

logically active than testosterone in terms of its andro­ genic effects. The androgenic potency of 5a-DHT is twice that of testosterone. Another important interconversion of steroid hor­ mones is the one resulting from the increase in circulat­ ing cortisol concentrations that occurs in the fetal lamb prior to parturition. The elevated fetal cortisol concen­ trations stimulates 1 7a-hydroxylase, C17-C20 lyase ac­ tivity, and probably aromatase activity in the placenta. These enzymes make it possible for progesterone to be converted to estrogens in the placenta. Estrogens then affect the synthesis of prostaglandin F2a, which precipitates delivery. The interconversion of progester­ one to estrogen is well documented in the sheep and probably occurs in a similar way in the goat and the cow.

E. Synthesis and Clearance of Hormones The determination of concentrations of hormones in biological fluids, including plasma, urine, saliva, and feces, has been useful in determining the reproductive status of animals. Although there are a number of fac­ tors that can influence hormone concentrations, the overriding factors are synthesis and clearance. Of greatest interest is the rate of synthesis of a hormone from a particular endocrine gland, because factors that govern clearance are usually stable; therefore, the con­ centration of a hormone usually reflects its rate of syn­ thesis or secretion. The synthesis of steroid hormones of the reproduc­ tive system is under the control of gonadotropins that are released in pulsatile fashion. This mechanism has a profound influence on the secretion of testosterone in the male, in that changes in pulsatile rate can occur a number of times a day with increases in pulse rate resulting in greatly increased concentrations of testos­ terone. For example, in males of many domestic spe­ cies, testosterone values can vary from 3.5 to 20 nmol/ liter within a period of a few hours, with the extremes still representing normal production of testosterone by the testes. The usual judgement as to normalcy is based on an animal having at least the minimal or basal con­ centration. In the female, estrogen and progesterone synthesis by the ovary is also under the control of a pulsatile mode of gonadotropin secretion. The pulse rate usually remains relatively stable over limited peri­ ods of time so that fluctuations in concentration of these hormones are not as acute as for androgens. Thus, gonadotropin pulsatility is not a consideration in inter­ preting the significance of progesterone and estrogen concentrations in biological fluids. In the female, synthesis rates for ovarian steroid hormones are obviously related to ovarian function.

594

Lars-Erik Edqvist and Mats Forsberg

Progesterone concentrations, relatively stable during the luteal phase of the estrous cycle, decline rapidly over a 24- to 36-hour period during luteolysis. Estrogen values continually increase during the follicular phase of the cycle, declining with the onset of the gonadotro­ pin preovulatory surge as the granulosa is converted from estrogen to progesterone production. Even though secretion rates can change for both progester­ one and estrogen, analysis of these hormones usually provides useful information as to luteal or follicular activity, respectively. If one wishes to use hormone values (in blood, for example) as an indication for se­ cretory activity of an endocrine organ, one other factor must be considered: the conversion of steroid hor­ mones by peripheral tissues. For example, in primates, estrone concentrations are derived mainly from the conversion of ovarian estradiol-1 7/3 and adrenal andro­ stenedione by tissues such as the liver. Steroids are eliminated via conjugation with gluc­ uronic acid and / or sulfates to form inactive mono- or diglucuronides or sulfates. These conjugates are all water soluble, with excretion occurring via urine or bile (feces). Conjugation occurs mainly in the liver, and the conjugates lack steroidal activity. Steroids are also rendered inactive by their metabolism to compounds that have greatly reduced biological activity. In this way, steroids are rapidly cleared from the bloodstream. Clearance is defined as the volume of blood that would be totally cleared of a particular steroid per unit time. Clearance can thus be expressed as liters / minute, and the clearance for most steroid hormones is around 1 liter / minute. In most situations, the clearance rate of steroids is relatively constant, so that blood concentra­ tions are a relatively good measure of fluctuations in production rates. Placental gonadotropic hormones such as hCG and eCG are produced in high concentrations and have much longer half-lives than the pituitary gonadotro­ pins and prolactin. The latter have half-lives around 10-30 minutes, whereas the corresponding figures for the placental hormones are from 1 .5 days for hCG to 6 days for eCG. One exception is equine LH, which has structural similarities to eCG and also has a much longer half-life (days) than LH from other species. The half-lives are increased because the molecules contain a larger carbohydrate moiety than do FSH and LH of most other species. In the blood, prostaglandins are rapidly metabo­ lized to their respective 15-keto-13,14-dihydro com­ pounds (Fig. 22.2). Primary prostaglandins such as PGF2a have a half-life in the peripheral circulation that is less than 20 seconds, whereas 15-keto-13,14-dihydro­ PGF2a has a somewhat longer half-life of about 8 minutes. Ninety percent or more of PGF2a is metabo­ lized during one passage through the lungs. The 15-

keto metabolites are biologically inactive and are de­ graded into short dicarboxylic acids before being excreted into urine. An example of the rapid metabo­ lism of prostaglandins is the fact that more than 90% of a 25-mg intramuscular injection of PGF2" in the cow is excreted in the urine and feces (2 : 1 ratio) over a 48-hour period (Neff et al., 1981).

II. ASSAY METHODS The radioimmunoassay technique was originally in­ troduced for the measurement of plasma insulin (Ber­ son and Yalow, 1959), and the enzyme immunoassay techniques for the quantitative determination of immu­ noglobulin G (Engvall and Pedmann, 1971 ). The tech­ niques are competitive and utilize the same basic prin­ ciple, which is based on the ability of nonlabeled hormone to compete with a fixed amount of isotopi­ cally or enzymatically labeled hormone for the binding sites on a fixed amount of protein. The nonlabeled hormone reduces the number of free binding sites on the protein, thus decreasing the availability of the bind­ ing sites to the labeled hormone. At equilibrium, the free hormone is separated from the protein-bound hor­ mone, and the reaction is quantified by the determina­ tion of the amount of labeled hormone that is antibody­ bound or free (Fig. 22.3). The degree of inhibition of binding of the labeled hormone to the binding protein is a function of the concentration of nonlabeled hor­ mone present in the solution. As a basis for the quanti­ fication, a standard curve is developed with fixed amounts of labeled hormone and binding protein incu­ bated together in the presence of a known and graded concentration of unlabeled hormone (Fig. 22.3). Certain disadvantages exist to the use of radioiso­ topes as labels in immunoassays, including the limited shelf life and stability of radiolabeled compounds, the need for relatively expensive counting equipment (es­ pecially for tritium-labeled compounds), the need for well-trained personnel and specialized laboratory equipment, and problems in disposal of radioactive waste. Consequently, attempts have been made to de­ velop assays that use nonisotopic labels such as en­ zymes, fluorogens, and chemiluminescent precursors. Enzyme immunoassays can be as sensitive, accurate, and precise as radioimmunoassays, and specificity de­ pends on the quality of the antibody, as is true for radioimmunoassay systems (Munro and Stabenfeldt, 1984). Of particular interest in domestic animals has been the use of these techniques in the determination of progesterone in blood (Munro and Stabenfeldt, 1984; Meyers et al., 1988; Lopate and Threlfall, 1991) and in milk (Amstadt and Cleere, 1981; Sauer et al., 1981; Allen and Foote, 1988; Etherington et al., 1991). Enzyme

595

Clinical Reproductive Endocrinology B I N D I NG PROTEI N

� � �

� � �

� � �

� � � "0 c

:::l a .c

� .:;; :u C1l a '6 � "if!.

100

• • • • • • • • • • • •

• • • •

• • • •

• • • •

• • • •

• • • •

• • • •

SO 25 2 4

8

.,. RADIOAC T I V E HOR�ONE BOUND TO BINDING PROTEIN

NONRADIOACTIVE HOR�ONE

-+

o

• • • • • • • • • • • •

"0 c

75

0

RADIOACTIVE HOR�ONE

2

� � �

t;] G

--

� � �

• • • •

66%

pmol

t;] G /!J /!J G G /!J /!J G /!J I!l I!l 4

-+

pmol

� � � 50%

� -- � � �

I!l G G G G G G [;] G /!J [!J [!J /!J /!J /!J /!J I!l I!l G /!J t:l t;] 8 p mol

•

33%

100

:::l 75 a .c � :;; o.£ 50 C1l a '6 2 5 � "if!. 0

• • • •• • •

1 00%

pmol

t;] t;] I!l G

FREE HORMONE

• G • I!l • I!l • I!l

• • • • • [;] • • I!l • • /!J

I!l [;] I!l [;]

I!l G

• • • • • • • •

I!l G G [!J

I!l [!J [!J G

• • G [;]

I!l G G G

G G G [!J G G

2

'5,

~ 2

,

.Q

-1 -2

8

CONC ENTRATION OF HORMONE

0

2

4

8

pmol

FIGURE 22.3 The principle of an immunoassay technique is based upon the ability of the binding protein to bind the labeled hormone. Excess labeled hormone is added to ensure saturation of the hormone-binding sites on the binding protein. The addition of increasing amounts of nonlabeled hormone (2, 4, and 8 pmol) results in a proportional decrease in the quantity of labeled hormone bound to the binding protein. Separation of labeled hormone bound to the binding protein from free hormone must be achieved before quantification can be done. In the lower part of the figure, this reaction is depicted in three different ways. In the panel to the left and in the middle, percent labeled hormone bound to the binding protein is on the ordinate, and the amount of hormone on a linear scale (left) and log scale (middle) is on the abscissa. In the panel to the right, the logit of the response variable is on the ordinate, and the amount of hormone is plotted on a log scale. The method depicted in the middle and to the right can be used for determination of parallelism. The 10gitl 1og transformation (right) is frequently used when immunoassay data are analyzed by computer. immunoassay in the laboratory setting is very efficient, particularly in the time required for the assay reaction (2 hours or less), in ease of separation of bound from free hormone when using the microtiter plate system (30 seconds to wash 96 wells on a microtiter plate), and in speed of end-point analysis (optical densities can be determined in 1 minute for a 96-well microti­ ter plate). Among the various approaches to enzyme immuno­ assay, the double antibody sandwich method for deter­ mination of hormones is frequently used. In this assay system, the plastic wells are coated with antibody. The sample to be processed is then added, and its hormone binds to the antibody-coated plastic well. A second

enzyme-labeled antibody directed against another epi­ tope of the hormone is then added. The amount of enzyme-labeled antibody bound is directly propor­ tional to the amount of hormone in the sample. There are two advantages with this methodology: (1) the hormone does not need to be isolated for labeling, and (2) the same general technique can be used to label different antibodies. The method can only be used to assay hormones with at least two binding sites; thus, it is unsuitable for measurement of low-molecular­ weight hormones such as steroid hormones. The analytical equipment used for enzyme immuno­ assay can be used for a variety of determinations, in­ cluding those involved in disease surveillance and

596

Lars-Erik Edqvist and Mats Forsberg

drug analysis. This flexibility allows for the sharing of one specialized spectrophotometer among several disciplines at a great reduction in cost. Another benefit is that the analytical equipment needs little mainte­ nance. The elimination of the problems engendered by use of radioisotopes is a major advantage for enzyme immunoassay. This advantage and the fact that color change is fundamental to enzyme immunoassay means that the assay can be used visually to determine the presence or absence of a corpus luteum (CL) in domes­ tic and other species. Today, it is possible to use enzyme-immunoassay analytical systems for proges­ terone in blood or milk on the farm beside the animal to get a direct assessment of the functional ovarian status of an animal (Nebel et al., 1989; Herrier et al., 1990; Matsas et al., 1992; Romagnolo and Nebel, 1993). Chemiluminescent-labeled immunoassays are com­ mercially available. All cases of chemiluminescence assay systems are based on a reaction producing light in the visible spectrum. Such an analytical system is used in the authors' laboratory for the determination of several different hormones in domestic species (Forsberg et al., 1993a). A microtiter plate with 96 wells can be read and processed in less than 2 minutes. It is interesting to note from a historical basis that radioimmunoassay originally developed by endocri­ nologists was later utilized by immunologists. Con­ versely, enzyme immunoassay developed by immu­ nologists for studying mechanisms of disease was later adopted by endocrinologists (Dray et al., 1975). For the foreseeable future, immunoassay techniques, espe­ cially enzyme immunoassays, will be the laboratory methods of choice in reproductive diagnostic endocri­ nology of domestic species.

A. Immunoassay 1. Production of Antibodies Immunoassay techniques utilize antibodies as bind­ ing protein. Hormones such as LH and FSH, which are glycoproteins with molecular weights of around 30,000, are antigenic because of their size and chemical composition. In general, a lower level of purity is re­ quired of the polypeptide hormones for antibody pro­ duction as compared to the hormone used in the label­ ing procedure. If an assay for bovine LH utilizes an antibody to bovine LH or radio- or enzyme-labeled bovine LH as tracer, and bovine LH as the standard, the assay system is completely species specific and is said to be of the homologous type. Such a system represents the ideal immunoassay system for measuring a polypeptide hormone. Because of limited availability and / or lack of

suitable purity of polypeptide hormone preparations, heterologous assay systems have been developed. In these cases, an antiserum to a polypeptide hormone of one species has been used for the determination of the same polypeptide hormone in another species. The standard hormone used for quantification of the assay should, however, originate from the same species for which the measurements are performed. There are polypeptide hormone antisera available that show a high degree of cross reactivity. One such antiserum of special interest in the field of reproductive hormones in domestic species is a polyclonal LH anti­ serum raised against ovine LH (Niswender et al., 1969). This antiserum reacts specifically with LH from other species and has been used for the determination of LH in approximately 45 species including the cow, sheep, pig, cat, and dog (Millar and Aehnelt, 1977; Madej and Linde-Forsberg, 1991). A monoclonal antibody gener­ ated against bovine LH has been reported to have high cross reactivity among species (Matteri et al., 1987; Bravo et al., 1992; Forsberg et al., 1993b). Steroid hormones and prostaglandins have consid­ erably lower molecular weights and thus are not im­ munogenic per se. However, these structures can be rendered immunogenic if covalently linked to large carrier molecules such as bovine serum albumin, and specific antibodies can be elicited in this fashion (Land­ steiner and Van der Scherr, 1936). In order for such a hormone-protein conjugate to be immunogenic, ap­ proximately 10-20 hormone molecules should be pres­ ent per molecule of protein. In the case of bovine serum albumin, about 30% of the sites available for conjuga­ tion should be occupied. Most naturally occurring steroid hormones and prostaglandins contain hydroxyl or ketone groups that are used to prepare derivatives containing active groups such as carboxyl or amino groups. These groups are then activated so that they react with amino or carboxyl groups of the protein molecule. The speci­ ficity of the antisera obtained by immunization with a steroid-protein conjugate is dependent on the site used for conjugating the steroid to the protein. More specific antisera are obtained if the hapten (steroid) is attached to the protein at a site remote from the characteristic functional groups of the hormone (Lindner et al., 1970). The most frequent species used for production of polyclonal antibodies are sheep and rabbits. One of the most popular and efficient schedules for immunization involves multiple injections in the back and neck of the animal of the antigen emulsified in complete Freund's adjuvant (Vaitakaitus et al., 1971). During immuniza­ tion, the developing antibody titer is monitored, and a relative large number of milliliters of serum can be obtained when a suitable titer has been achieved. A

Clinical Reproductive Endocrinology

few milliliters of a high-titer antiserum are usually sufficient for millions of immunoassay determina­ tions. Antisera seem to be quite stable when stored at -200 C, although the usual preferred temperature is - 700 C. A new development in the technology of antibody production was the discovery that hybridomas could be used to produce an endless supply of antibodies with certain specificity (Kohler and Milstein, 1975). The procedure involves the fusion of two cell lines: B lymphocytes selected for the production of a specific antibody and myeloma cells that have the capacity for permanent growth. Antibody production occurs by injection of the cell lines into mice; permanency is assured by maintaining a supply of cells in the frozen state. Monoclonal antibodies can be used for the quan­ titative immunoassay of hormones. They have the ad­ vantages of specificity, unlimited supply over time, and the possibility of standardizing assay methods be­ tween laboratories. Disadvantages are that they have lower affinities than do polyclonal antibodies, and they do not always form precipitates with antigens. Mixing of monoclonal antibodies may result in an increase of affinity (Ehrlich et al., 1982).

2. Labeled Hormone In radioimmunoassay techniques for polypeptide hormones, the antigen (hormone) is most commonly used for preparing the radioactive tracer. Usually, ra­ dioactive iodine, 1251, is used for radioiodination of the antigen. The two most frequently used techniques for iodination are free or immobilized chloramine-T (Iodo­ beads) (Hunter and Greenwood, 1962; Markwell, 1982) and the lactoperoxidase procedure (Thorell and Johansson, 1971 ). Peptide hormones containing tyrosyl or histidyl residues can also be iodinated with these techniques . Many RIA systems for steroid hormones and prosta­ glandins utilize tritiated forms of these molecules that are available commercially. Because tritium has a con­ Siderably longer half-life than iodine, tritium tracers can be used in many cases over several years, whereas the iodinated tracers often have to be prepared monthly. There are, however, certain advantages in using iodinated tracers for steroid hormones and pros­ taglandins: Simpler and cheaper counting systems can be used, that is, gamma counting as opposed to liquid scintillation counting. Another advantage of radioio­ dine over tritium is its higher specific activity, which increases the sensitivity of the assay. Direct incorpora­ tion of iodine in the skeleton of steroid hormones re­ sults in a loss of the immunoreactivity. Thus, the ap­ proach taken for radioiodination of steroid hormones

597

has been to link a tyrosyl or histidyl molecule to the steroid molecule, making direct radioiodination possi­ ble (Niswender, 1973), or to iodinate a compound such as tyramine and conjugate the iodinated compound to the steroid molecule (Lindberg and Edqvist, 1974). Most EIA systems for steroid hormones use horse­ radish peroxidase, alkaline phosphatase, or f3-galacto­ sidase as labels. The enzyme-labeled steroid is pro­ duced the same way as has been described for the synthesis of steroid-protein conjugates for the produc­ tion of antibodies.

3. Separation of Antibody-Bound and Free Hormone An essential part of any immunoassay system is an efficient procedure for the separation of antibody­ bound and free hormone. Several different approaches have been taken to achieve a rapid and efficient separa­ tion. Currently, the most frequent separation proce­ dures in use are based on (1) antibodies coupled to an insoluble polymer, (2) precipitation of antibody-bound hormone, or (3) adsorption of free hormone. Antibodies coupled to an insoluble polymer have been used for separating antibody-bound and free hor­ mone in RIA procedures for both protein hormones (Wide and Porath, 1966) and steroid hormones (Abra­ ham, 1969). One procedure involves decanting from polystyrene tubes in which antibodies have been ad­ sorbed to the surface of the tube, which is followed by determination of radioactivity in the antibody-bound (contained in the tube) or free (contained in the eluent) form. In . most EIA systems designed for low­ molecular-weight hormones, antibody-bound hor­ mone is measured after free hormone has been re­ moved by washing the wells of the microtiter plate that have been previously coated with antibody. Another common procedure for RIA utilizes anti­ bodies covalently coupled to an insoluble polymer granule (Wide and Porath, 1966). In this case, free and antibody-bound hormone are separated through cen­ trifugation. After removal of the supernatant contain­ ing the free hormone, the antibody-bound radioactivity can be determined. Antibody-coated glass beads have been used in both RIA and EIA (Schmidt et al., 1993). Separation is achieved by washing the bead, and then the radioactivity or enzyme activity is determined. Precipitation of antibody-bound hormone has been achieved through the addition of ammonium sulfate (Mayes and Nugent, 1970) or polyethylene glycol (Des­ buquois and Aurbach, 1971), leaving the free hormone in the solution. The latter precipitation procedure has been found advantageous for the precipitation of prostaglandin-antibody complexes (Van Orden and Farley, 1973). Precipitation of the antibody-bound hor-

598

Lars-Erik Edqvist and Mats Forsberg

mone complexes can also be achieved through the ad­ dition of a second antibody prepared against the first antibody. Thus, in an RIA technique for bovine LH that utilizes an antiserum to bovine LH raised in a rabbit (first antibody), the second antibody will be an antibody prepared against rabbit gamma globulin. The addition of the second antibody will result in a precipi­ tate containing the antibody-bound LH, which can be separated from the supernatant by centrifugation. The time required for the separation can be decreased by the addition of polyethylene glycol (Eisenman and Chew, 1983). Systems using a second antibody coupled to insoluble particles are efficient and commonly used (Dericks-Tan and Taubert, 1975; Forsberg et al., 1993b). For steroid hormones, a traditional separation pro­ cedure is the adsorption of free hormone to dextran­ coated charcoal. After the addition of the charcoal, the separation of free and antibody-bound steroid is achieved through centrifugation. This method is rapid and efficient in separating free and bound steroid hor­ mones. However, the charcoal can also adsorb some of the antibody-bound steroid, which is called "strip­ ping." To control this, timing of the reaction is impor­ tant. Carrying out the reaction at 4°C has been found to limit this dissociation (Abraham, 1974).

4. Reliability Criteria The reliability of immunoassay analyses depends on specificity, sensitivity, accuracy, and precision.

a. Specificity The specificity of an immunoassay, or its freedom from interference by substances other than the one to be measured, is dependent on several different factors, the most important being the specificity of the antise­ rum used. Demonstration of specificity for the immu­ noassay of large protein hormones such as LH and FSH is relatively difficult and relies upon indirect criteria. Because it is not possible to synthesize these hormones, they must be isolated and purified from biological ma­ terial. The purity of such preparations is variable, and the most common cause of nonspecificity for these hormones is impurity of the immunizing material. A relatively common finding is the cross reaction of TSH with antibodies to LH, and vice versa. Antibody specificity is usually demonstrated by testing the bind­ ing of hormones other than the one intended to be measured to the antibody. If, for example, bovine TSH significantly inhibits the binding of bovine LH to an antibody to bovine LH, this indicates that the antise­ rum used is nonspecific or that the TSH preparation contains LH. If the inhibition curves are parallel, the latter explanation is likely because the parallelism indi-

cates the same binding kinetics. It should be noted that parallelism, in itself, is not adequate proof of specific­ ity. As indicated previously, both LH and TSH are composed of two subunits, an a-subunit that is identi­ cal for the two hormones and a ,B-subunit that is unique for each hormone. It is possible that an antiserum could contain binding sites that will react only with the a­ subunit. In such a case, the dose-response curve of LH and TSH utilizing such an antisera will be parallel, the assay system will not be hormone specific, and thus the system will be invalid for the measurement of LH (Niswender and Nett, 1977). Double antibody sandwich methods utilizing monoclonal and / or poly­ clonal antisera can partly reduce the problem. In the case of immunoassay techniques for steroid hormones and prostaglandins, the same proof of speci­ ficity has to be undertaken. Here the situation is sim­ pler, because lower-molecular-weight hormones can easily be purified and, in most cases, produced synthet­ ically. Furthermore, comparison of immunoassay re­ sults with results obtained by mass fragmentography gives very valuable information, because the latter technique can be considered as an absolute proof of structure and thus specificity. Some idea as to the specificity of an antiserum to a steroid hormone can be gained from the position of the steroid molecule that is used as the anchoring point to the protein (Fig. 22.2). If an antiserum to estradiol1713 is produced through the use of an antigen conju­ gated via the hydroxyl group at carbon 17 of the ste­ roid, the resulting antiserum will react almost equally well with estrone and thus will have relatively poor specificity. This is because the only structural differ­ ence between the estradiol-1 7,B and estrone molecules is the configuration at position 17 (Fig. 22.2). In general, steroid antibodies are more specific for the portion of the steroid molecule that protrudes from the carrier protein and less specific for the portion of the steroid used for linkage to the protein. Thus, in the case of estradiol-1 713, highly specific antibodies have been de­ veloped after immunization with conjugates when car­ bon 6 of the B ring has been used as the site of attach­ ment to the protein (Exley et ai., 1971). By use of the same fundamental principle concerning the selection of appropriate sites for conjugation, specific antibodies have been produced against progesterone conjugated to the protein through carbon 11 and to testosterone through carbons 1 and 3 (Niswender and Midgley, 1970). Protein hormone determination by immunoassay is often performed on blood serum or plasma. The influence of serum or plasma on the binding of the tracer to the antibody must be investigated. The assay of different amounts of serum or plasma should result

599

Clinical Reproductive Endocrinology

in curves parallel to those obtained with protein hor­ mone standard. In some steroid hormone immunoas­ says, the hormone is extracted by an organic solvent from a plasma sample. Organic solvents can also be used as a means to selectively remove steroids from biological fluids. For example, most immunoassay pro­ cedures designed to measure progesterone utilize anti­ sera developed against a progesterone-I I-protein con­ jugate, which results in a minor cross reaction with corticosteroids (Thorneycroft and Stone, 1972). 1£ a non­ polar solvent such as petroleum ether is used for the extraction of progesterone from serum or plasma sam­ ples, about 80-90% of progesterone is extracted, leav­ ing the more polar corticosteroids in the plasma ( J0hansson, 1969). The use of such a selective extraction system increases the overall assay specificity. Direct analytical systems for steroid hormones omitting the extraction step are also employed. Because steroid hor­ mones in a blood sample to be analyzed are bound to carrier proteins, direct assay systems have to secure that all steroid molecules, both free and protein-bound, in the sample are given equal opportunities to inter­ act with the antibody used in the assay. The syn­ thetic steroid danazol (17a,2,4-pregnanedien-20yno(2,3-d)isoxazol-17-01) can be used to displace progesterone from the binding proteins (Carriere and Lee, 1994). Likewise, 8-ANS (8-anilino-1-naphthalene­ sulfonic acid) and levonorgestrel (D( )-norgestrel) can be used to enhance the displacement of testosterone from its binding proteins (Hoyle and Ebert, 1990). Cer­ tain immunoassay systems may require purification of the plasma extract to achieve an acceptable specificity. The main problem in the immunoassay of primary prostaglandins is that they can continue to be formed in large amounts by platelets after the blood sample has been obtained (Samuelsson et al., 1975). Therefore, the concentrations of PGF2a reported in blood serum or plasma appear, in most cases, to be 100-1000 times higher than the actual values (Granstrom and Kindahl, 1976). The primary prostaglandins have a very short half-life in the circulation (Hamberg and Samuelsson, 1971) and are rapidly converted to their corresponding I5-keto-13,I4-dihydro derivatives. The latter have con­ siderably longer half-lives and occur in higher concen­ trations than the parent compounds (Beguin et al., 1972). Analysis of metabolites of prostaglandin F2a avoids the problem of the overestimation observed for the parent compound in that the metabolites are formed only within the body and values thus remain stable once a blood sample has been obtained. Radio­ immunoassay systems utilizing antibodies to 9a, l la­ dihydroxy-15-ketoprost-5-enoic acid and 5a,7a­ dihydroxy-l l-ketotetranorprosta-1,16-dioic acid have been developed (Granstrom and Samuelsson, 1972). -

Most problems involved in the determination of the primary prostaglandins are avoided if their main me­ tabolites, the 15-keto-13,14-dihydro compounds, are measured.

b. Sensitivity The sensitivity of an immunoassay is defined as the smallest quantity of hormone that the assay can detect reliably. Usually, two kinds of sensitivity are evalu­ ated. The sensitivity of the standard curve is defined as the smallest amount of hormone that is significantly different from zero at the 95% confidence limit. How­ ever, the most meaningful sensitivity to establish is the smallest amount of hormone that can be measured per unit of biological fluid, for example, per milliliter of plasma.

c. Accuracy The accuracy of an assay is defined as the extent to which the measurement of a hormone agrees with the exact amount of the hormone. Accuracy is often deter­ mined by comparing immunoassay data with values determineq by other procedures such as gravimetry, gas-liquid chromatography, and mass spectrometry. For steroid hormones and prostaglandins, accuracy is also often determined by recovery experiments in which different amounts of hormones are added to a biological fluid, such as plasma, that contains low concentrations of the hormone. The amount of hor­ mone measured in the assay is then compared with the amount of hormone added.

d. Precision Two types of precision are usually evaluated. The within-assay precision is determined from duplicate measurements of the same sample within the same assay. The between-assay precision is determined from replicate analyses of the same sample in different assays. Usually the between-assay variance is greater than the within-assay variance. Assay variance should be checked continuously with each assay of a certain hormone by use of plasma pools containing set amounts of the hormone. Usually three different plasma sets containing low, medium, and high hor­ mone concentrations are used. Within- and between­ assay variations in immunoassay procedures are usu­ ally greater than for most other routine procedures used in clinical chemistry.

B. Summary The use of immunoassay techniques has dramati­ cally increased our knowledge and understanding of reproductive endocrinology in domestic species. These

600

Lars-Erik Edqvist and Mats Forsberg

procedures are relatively simple to perform, inexpen­ sive, sensitive, and specific. Data gained from these techniques have resulted in characterization of endo­ crine changes throughout the estrous cycle and preg­ nancy in domestic animals.

III. PHYSIOLOGY OF REPRODUCTIVE HORMONES IN THE FEMALE This section is relatively brief in its coverage of endo­ crinological events during the reproductive cycle. Readers interested in a more complete presentation of hormonal events involved in reproduction are referred to texts dealing specifically with the subject. Also, some hormones with some influence on reproductive pro­ cesses, such as those of the thyroid and adrenal glands, are not covered in this section.

A. Estrous Cycle The major endocrine events that precede ovulation have been well documented in the cow (Chenault et al., 1975), ewe (Nett et ai., 1974), sow (Shearer et al., 1972), mare (Evans and Irvine, 1975; Palmer and Jous­ set, 1975; Stabenfeldt et al., 1975), dog (Concannon et al., 1975), and cat (ShiUe et al., 1979b). In large domestic animals (cattle, horse, pig, sheep, and goat), follicle growth occurs during the luteal phase in spite of the inhibitory nature of progesterone, the main secretory product of the CL. Although follicles are usually not ovulated during the luteal phase in most species, the mare occasionally ovulates during the luteal phase (Hughes et at., 1973). With regression of the CL, follicles grow rapidly prior to ovulation because of gonadotropin stimula­ tion. The follicles secrete increasing amounts of estro-

gen during development, which is important for onset of sexual receptivity as well as for the initiation of the surge release of gonadotropins that is essential for the ovulatory process (Fig. 22.4). Estrogens initiate the re­ lease of LH and FSH through the release of GnRH (Moenter et al., 1990). In most species, the preovulatory surge of gonadotropins begins approximately 24 hours before ovulation and is usually of short duration, for example, 8-10 hours in the cow. The mare is an excep­ tion in that large amounts of LH are released during an 8- to 9-day period with ovulation occurring on the third day (Geschwind et al., 1975). Another important function of GnRH is to elicit sexual receptivity. Thus, the onset of the preovulatory LH surge and sexual receptivity are coordinated via GnRH synthesis and re­ lease. Following ovulation, a CL is formed under the in­ fluence of pituitary gonadotropins. In most species, LH is the major luteotropin; prolactin is thought to play a role in sheep and in rodents and may be the predominant luteotropin in these species. If pregnancy does not ensue, the CL regresses, which permits the estrous cycle to be repeated. This well-timed sequence occurs repetitively at set intervals if not interrupted by pregnancy. A summary of estrous-cycle activity in common do­ mestic animal species is given in Fig. 22.5. Seasonal breeders, such as the mare, ewe, doe, and queen, un­ dergo cyclic ovarian activity only during the breeding season, whereas the cow, sow, and bitch are affected little, if any, by photoperiod and can have cyclic ovar­ ian activity throughout the year. Estrous-cycle length is approximately 21 days in the cow, doe, mare, and sow and 17 days in the ewe. The bitch has a much longer estrous cycle, the luteal phase often being be­ tween 50 and 80 days in duration. The interval between cyclic ovarian activity in the bitch is extended even LH

prostaglandin

progesterone

c

.2

) e

_

estradiol

.

-4

. .. ...

-3

-2

-1

o

Days before and after LH peak

2

FIGURE 22.4 Schematic presentation of the preovulatory events in general applicable to the cow, doe, ewe, and sow. These endocrine events also occurs in the mare, but in this species, the release of LH occurs over a considerably longer time period.

601

Clinical Reproductive Endocrinology

Cow Doe

Ewe

Sow

Mare

CL life span Estrus Ovulation Day of cyde CL life span Estrus Ovulation Day of cyde CL life span Estrus Ovulation Day of cycle CL life span Estrus Ovulation Oa of cle

It

It

It

1 3 5 7 9 1 1 1 3 1 5 1 7 1 9 21 1 3 5 7 9 1 1 1 3 1 5 1 7 1 9 21 1 3 5 . . . . . .

,

,

,

1 3 5 7 9 1 1 1 3 1 5 1 7 1 3 5 7 9 1 1 1 3 1 5 1 7 1 3 5. . . . ..

,

,

,

1 3 5 7 9 1 1 1 3 1 5 1 7 1 9 21 1 3 5 7 9 11 1 3 1 5 1 7 1 9 2 1 1 3 5 . . . . . .

.,

.,

.,

1 3 5 7 9 1 1 1 3 1 5 1 7 1 9 21 1 3 5 7 9 1 1 1 3 1 5 1 7 1 9 21 1 3 5 . . . . . .

.., Coitus - nonpregnant ..,

Queen

CL life span Estrus Ovulation Oay of cycle

1 3 5 7 9 . . . . ...... 2 1 1 3 5 7

Bitch

CL life span Estrus Ovulation Day of cycle

1 3 5 7 9 1 1 . . ...... . . .. . . . .. . . ... .... .. ...... .. .... .. ... . . . . ..... .. ..... .. .. . . . .. .. . . . . 50 - 80

_ No coitus -

1 3 5 7 9 ............... 35 37 39 41 1 3 5 7 . . .

FIGURE 22.5 Comparative data on the duration of estrus, time of ovulation, and duration of CL function in the cow, ewe, sow, mare, queen, and bitch. (Modified from Stabenfeldt, 1974; Edqvist and Stabenfeldt, 1989.)

further by the occurrence of a 4 to 5-month anestrum period following regression of the CL. The cat is an induced ovulator, requiring coital stimulation for ovu­ lation and thus for CL formation. In the absence of coitus, follicles develop every 15-20 days, with follicu­ lar growth and regression occupying 5-7 days of each period. An ovulatory, nonfertile mating results in the formation of CL that persist for approximately 35 days and a syndrome termed pseudopregnancy in the cat.

B. Control of the Corpus Luteum The regression of the CL (luteolysis) is a key event that is responsible for the well-timed estrous cyclicity seen in most domestic species. The importance of the uterus in the control of the lifespan of the CL has been documented through hysterectomy in the cow, ewe, sow, and mare (du Mesnil du Buisson and Dauzier, 1 959; Wiltbank and Casida, 1956; Stabenfeldt et al., 1974a). Removal of the uterus from these species dur­ ing the luteal phase results in prolongation of luteal activity. It is now well established that the uterus in these species syntheSizes and releases PGF2a, which causes the CL to regress (McCracken et ai., 1972). The temporal release patterns of PGF2a (Fig. 22.4), usually in a pulsatile mode lasting a few hours, have been described in the ewe (Harrison et al., 1972; Barci­ kowski et al., 1974), sow (Gleeson et al., 1974), doe (Fredriksson et al., 1 984), and cow (Nancarrow et al., 1973). Some of the problems involved in determining PGF2cv such as its short half-life and formation by plate-

lets at collection, can be avoided if the main plasma metabolite, 15-keto-13,14-dihydro-PGF2a1 is deter­ mined. Data are available on the patterns of the metab­ olite during luteolysis in the cow (Kindahl et ai., 1976a,b), ewe (Peterson et ai., 1976), mare (Neely et ai., 1979), doe (Fredriksson et al., 1984), and sow (Shille et al., 1979a). Regression of CL is usually accomplished within 48 hours following the onset of the prostaglandin re­ lease. It is likely that estrogens, presumably of ovarian follicle origin, initiate PGF2a release. Estrogen also ini­ tiates the formation of endometrial oxytocin receptors (McCracken et al., 1984), which in sheep are important for pulsatile synthesis and release of PGF2a in that oxy­ tocin can initiate the release of PGF2a (Sharma and Fitzpatrick, 1974). In ruminants, oxytocin is synthe­ sized and released from the CL in response to PGF2w which in turn initiates the synthesis and release of PGF2a by the uterus. This mechanism is the basis for the pulsatile secretion of PGF2a (for reviews, see Flint et al., 1992; Whates and Denning-Kendall, 1992). In the dog and cat, PGF2a does not appear to be involved in luteolysis, although its precise role is still uncertain.

C. Early Pregnancy Modification of PGF2a release is essential for the establishment of pregnancy in the species (cow, ewe, mare, and sow) in which this compound serves as the luteolysin (Kindahl et ai., 1976a; Nett et ai., 1976; Shille et al., 1979a). The rapid elongation of fetal membranes,

602

Lars-Erik Edqvist and Mats Forsberg

which precedes the critical time of the initiation of luteal regression by about 3 days in the nonpregnant animal, appears to be important for the modification of prostaglandin release. Fredriksson et al. (1984) for goats and Zarco et ai. (1988) for sheep have shown that the main change in PGF2a synthesis and release in nonpregnant vs pregnant animals involves a continu­ ous, not pulsatile, mode of secretion. In fact, PGF2a concentrations increase at the onset of pregnancy, but the release is continuous, which allows for the prolon­ gation of luteal activity. Maternal recognition of pregnancy in the cow, ewe, doe, and sow involves mechanisms that alter the pros­ taglandin release to protect the CL from luteolysis. In the cow, ewe, and doe, conceptuses secrete trophoblast proteins between days 10 and 21 -24 of pregnancy. These substances are produced before the blastocyst attaches to the endometrium, are secreted into the uter­ ine lumen, and bind to uterine oxytocin receptors, blocking the episodic uterine PGF2a secretion (for a review, see Flint et ai., 1992). In the sow, in which the maternal recognition of pregnancy is controlled by conceptus-derived estrogens (Perry et ai., 1976; Fig. 22.6), similar trophoblast proteins have been isolated from preimplantation blastocysts (La Bonnardil�re, 1993). Significant increases in intraluminal PGF2a have been observed in pigs (Frank et al., 1978) and cattle (Thatcher et al., 1979), but not in horses (Zavy, 1979), beginning about two weeks postconception. Possible uterine antiluteolytic factors include PGE (Hoyer et al., 1978), and PGE2 (Magness et al., 1978), with increased production of PGE2 by endometrial tissue having been demonstrated in ewes during early pregnancy (Ellin­ wood et al., 1979). Modification of the release pattern of PGF2a (pulsa­ tile to continuous) by the luteotropic products from the conceptus and uterus is probably the most important factor that allows luteal activity to continue. The net result is that luteal activity is extended in the cow, ewe, mare, and sow beginning at about 14 days following ovulation. Modification of PGF2a release does not ap­ pear to be important for the establishment of preg­ nancy in the dog and cat.

D. Pregnancy and Parturition The presence of a CL is necessary for the mainte­ nance of pregnancy in the vast majority of cows (Ester­ green et ai., 1967). The pig also requires luteal support throughout gestation (du Mesnil du Bussion and Dau­ zier, 1957). In the ewe, the presence of a CL is required for the first 50-60 days of gestation (Linzell and Heap, 1968). After this time period, the fetoplacental unit

MARE

w

t:(

0

u: ...J :::l

10

. .. ... . ---

-

(f) W Z 0 a: t(f) W

DOE

SO

30

EWE

.

(OW

•.... . ..

o

sow

10

SO

�",,'

. • •... . .. ..

.. .. .. ..

.. .. ..... . " " ..

, /, / ' .. " .

. ....

0

70

90

.. ... .. .. .' ,

/

�" ,'

.. ..

.

100

30

50

150

70

200

90

DAYS OF PREGNANCY FIGURE 22.6 Blood levels of conjugated estrogens (mainly es­ trone sulfate) during early pregnancy in the mare, doe, ewe, cow and sow. Data derived from Kindahl et al., 1982 (mare); Chaplin and Holdsworth (1982) (doe); Tsang (1978) (ewe); Gaiani et al., (1982) (cow); and Robertson and King (1974) (sow).

secretes significant amounts of progesterone. The dog requires the presence of a CL for all of gestation. Soko­ lowski (1971 ) found that ovariectomy even as late as day 56 postbreeding resulted in premature delivery. The necessity for secondary CL, formed in the mare between days 40 and 60 of gestation, has been dis­ cussed over the years. The secondary CL are the result of eCG secretion by the endometrial cups, which are formed from a circular band of cells of placental origin (chorionic girdle cells) that invade the endometrium and form isolated endocrine organs of temporary func­ tion (Allen, 1969). Although secondary CL are not es­ sential for the maintenance of pregnancy in the mare in that the primary CL continues to function for up to 150 days, they do add extra progestational support for the pregnancy during the time placental production of progesterone is being established. Progesterone sup­ port of pregnancy in the mare begins to be assumed by the placenta as early as day 50 of gestation but is not complete for all mares until approximately 100 days or later, a time that coincides with the beginning

Clinical Reproductive Endocrinology

demise of both primary and secondary CL (Holtan et al., 1979). It has been demonstrated that eCG has a close im­ munological relationship with equine LH (Farmer and Papkoff, 1979); further, a luteotropic effect of eCG has been shown by incubation studies of CL (Squires et al., 1979). If fetal loss occurs after the formation of endometrial cups, continuing eCG production sup­ ports luteal activity, even to the point of making lysis of these CL difficult in conjunction with the pharmaco­ logical administration of prostaglandin FZa" The endocrine activity of the fetoplacental unit can be monitored through the measurement of conjugated estrogens, especially estrone sulfate, in peripheral plasma or urine (Fig. 22.6). In the pregnant sow, con­ centrations of estrone sulfate become detectable at day 17 of pregnancy, increase until about day 28, then de­ cline to low or undetectable values, and then increase again beginning around days 75-80 of pregnancy, re­ maining high until parturition (Robertson and King, 1974). In the mare, estrone sulfate values begin to in­ crease around days 35-40 of pregnancy with accentu­ ated production at 80-90 days and with the highest values obtained from day 150 to parturition (Terqui and Palmer, 1979; Kindahl et al., 1982). In the cow (Gaiani et al., 1982) and ewe (Tsang, 1978), estrone sulfate levels can be detected from around day 70-80 after conception, and in the goat, concentrations start to increase around day 50 of pregnancy (Chaplin and Holdsworth, 1982). In the pregnant cow, estrone sul­ phate concentrations in milk start to increase between days 100 and 120 (Hatzidakis et al., 1993; Henderson et al., 1994). In all species, high values of estrone sulfate in the peripheral blood, milk, or urine are strong evi­ dence for the presence of a viable fetus. The mare produces two estrogens during pregnancy that are unique to equids, equilin and equilenin, both of which have unsaturated B rings (Girard et al., 1932). By fetal gonadectomy in the horse, Pashen and Allen (1979) have shown the importance of fetal gonads in the production of estrogen in cooperation with the placenta. Little evidence of estrogen production during pregnancy has been reported for the dog (Hadley, 1975). Estrogen production by the fetoplacental unit in the cat also appears to be minimal during gestation (Verhage et al., 1976). An important endocrine change that occurs prior to parturition in the cow, ewe, and sow involves an increase in the synthesis and release of unconjugated estrogens by the fetoplacental unit. This increased es­ trogen synthesis is reflected in elevated plasma estrone concentrations in the pregnant cow beginning 20-30 days prepartum (Edqvist et al., 1973), in the ewe about 2 days before parturition (Challis et al., 1971 ), and in

603

the pig about 1 week before delivery (Robertson and King, 1974). In the cow (Stabenfeldt et al., 1970; Edqvist et al., 1973) and bitch (Smith and McDonald, 1974), parturition is preceded by an abrupt fall in progester­ one concentrations 24-48 hours prior to delivery. In the ewe (Stabenfeldt et al., 1972), mare (Noden et al., 1978), and sow (Baldwin and Stabenfeldt, 1975), partial withdrawal of progesterone occurs prior to delivery. In the cow (Fairclough et al., 1975; Edqvist et al., 1976, 1978), ewe (Liggins et al., 1972), and bitch (Concannon et al., 1988), it has been demonstrated that prostaglan­ din release initiates regression of the CL and thus is responsible for the withdrawal of progesterone. High estrogen and prostaglandin concentrations combined with low progesterone concentrations in­ crease the contractile state of the uterus. Prostaglandins may also initiate cellular changes within the cervix, in addition to the effect of relaxin, which result in cervical softening and dilation. Cervical stimulation, the result of the initial entry of the fetus into the pelvic canal, causes the reflex release of oxytocin from the posterior pituitary. This increases the intensity of uterine con­ tractions and thus aids the final delivery process.

IV. CLINICAL ASPECTS OF REPRODUCTIVE ENDOCRINOLOGY The development of immunoassay techniques for hormone determinations in domestic species has cre­ ated laboratory procedures that are useful as diagnos­ tic aids in clinical work. For some reproductive hor­ mones, relatively well defined indications for their clinical use are known at present and more are likely to be identified in the future. Important differences exist between humans and animals with respect to endocrine analysis. For example, steroid hormone con­ centrations are much lower (10-fold less in the case of estrogens) in animals, which produces a requirement for more rigorous assay systems. In animals, the most useful information comes from assessing gonadal or fetoplacental activity (vs pituitary activity), and thus, the determination of steroid hormones is emphasized. Besides the fact that gonadotropin values are less use­ ful for the assessment of clinical situations, variations in amino acid composition of specific hormones among species mean that multiple systems have to be devel­ oped to determine the content of one protein hormone across species lines. Antibodies such as the one devel­ oped for LH by Niswender et al. (1969), which is able to detect LH in many species, are the exception. For hormone analyses to be useful as a diagnostic tool, certain criterion have to be fulfilled. The concen­ tration of the hormone at the sampling site (usually

604

Lars-Erik Edqvist and Mats Forsberg

a peripheral vein) should closely correlate with the amount of the hormone being released from the endo­ crine gland. It is preferable that the release pattern of the hormone be steady, which allows valid information to be obtained on the secretory status of the endocrine gland from one sample. Several reproductive hor­ mones do not fulfill the latter criterion, and thus their determination is not useful from a routine diagnostic view. For example, the duration of the LH surge ob­ served in conjunction with ovulation is short in most domestic species except the horse. In the cow, the pre­ ovulatory LH peak has a duration of 8-10 hours, re­ quiring samples to be obtained every 4 hours to detect the peak (Karg et al., 1976). In the mare, the duration of the peak is considerably longer, 8-9 days, which allows a less frequent sampling interval (Geschwind et al., 1975). However, the long duration of LH peak in the mare prevents the determination from being useful in predicting ovulation. One reproductive hormone, progesterone, has been found to be of significant cliYlical value in females of most domestic species, and OVerall, its analysis gives the most useful information on the reproductive status of animals. Other hormones with established or poten­ tial clinical use will be discussed further for each spe­ cies. As a percentage of total number of clinical assays, the highest number have been utilized as pregnancy tests. It is worth keeping in mind that the analysis of hormones as a diagnostic aid in solving clinical prob­ lems supplements, but does not replace, the informa­ tion that should be gained by a car�ful clinical exami­ nation. Although blood has been the usual medium for hor­ mone analysis, milk, urine, saliva, and even feces are also sources for gaining useful endocrine information. The latter substances have the advantage in certain situations of being easier to collect and, at the very least, they avoid the use of venipuncture. It has been shown that the determination of estrogen conjugates in urine is much more effective in revealing ovarian follicle production of estrogens than the analysis of either free or conjugated estrogen in plasma.

A. Cattle 1. Progesterone Several reports are available on progesterone con­ centrations in blood during early pregnancy in cattle (Pope et al., 1969). A finding of importance is that pro­ gesterone can be measured in the milk of lactating cows and, further, that its concentration accurately re­ flects the concurrent plasma concentration of proges­ terone (Laing and Heap, 1971).

The difference that exists in both plasma and milk progesterone concentrations 19-24 days after a fertile breeding as compared to a nonfertile breeding has been used as an early pregnancy test (Shemesh et al., 1968; Robertson and Sarda, 1971; Fig. 22.7). The plasma pro­ gesterone concentration in blood of pregnant cows at 21 days postbreeding is almost always at least 2 ng / ml (6 nmol/ liter) and usually 4-8 ng/ ml (13-26 nmol/ liter), as compared to less than 0.5 ng/ ml (1.6 nmol/ liter) in the nonpregnant animal at the same time. Ele­ vated progesterone concentrations, however, only re­ flect the presence of luteal tissue and are not directly indicative of the presence of a fetus in u tero. Further­ more, a slight prolongation of luteal activity in a non­ pregnant animal can occur that results in elevated pro­ gesterone concentrations at day 21, a situation in which the analytical progesterone result would be positive and the animal would be falsely considered pregnant. The accuracy of the forecast for pregnancy (positive forecast) thus is often lower than desirable, in most cases ranging between 75 and 90%. The negative fore­ cast, however, is more accurate; cows having low pro­ gesterone concentrations in milk or blood 21 days post­ breeding will almost always not be pregnant. The accuracy of the positive forecast can be increased if progesterone analyses are also carried out on samples obtained at the time of insemination to eliminate ani­ mals inseminated during the luteal phase of the estrous cycle. Cows inseminated during the luteal phase will be in the luteal phase 21 days later and the animal will be falsely considered pregnant. It should be recognized that the main focus of early pregnancy diagnosis is for more efficient management of breeding. In most cases involving early pregnancy diagnosis, pregnancy status needs to be verified again at about 40 days postbreeding. In spite of its limita­ tions, progesterone analysis for pregnancy diagnosis is the most common clinical use of any of the reproduc­ tive hormones. Progesterone analysis can also be used for the retrospective confirmation of the absence (or presence) of a CL at the time of insemination in cattle. Since inadequate detection of estrus is the most com- · mon cause of l ow fertility in herds utilizing artificial insemination, the determination of progesterone con­ centrations at the time of insemination can be a useful tool when herds with fertility problems are encoun­ tered. Karg et al. (1976) indicated that improper timing of insemination, as judged from milk progesterone de­ terminations, occurred in 15% of the cases in a con­ trolled field test; the figure rose to 26% under practical field conditions. Lower figures have been reported by others, namely 0 to 3% in the Netherlands (Van de Wiel et al., 1978), and 4% in Sweden (Oltner and Edqvist, 1981). The use of progesterone determina­ tions in milk obtained at the time of insemination in

Clinical Reproductive Endocrinology

605

PREGNANCY TEST

cow 8

� «

7 -

� a..

6

�w

4

�': 5 w z

3

ttJ 2 f­ en o rr: a..

O

5 ------�3� O 2� 1 5--------�2�O------�� 5---------1�O--------J �L------J DAYS AFTER BREEDING

FIGURE 22.7 Time after breeding for utilizing progesterone analysis as a means of pregnancy diagnosis. Progesterone content in nonpregnant (e) cows compared with progesterone content of pregnant animals (0) (Stabenfeldt et ai., 1969b).

cows with questionable heat signs and inconclusive genital-tract findings can serve as a valuable tool for educating the staff responsible for insemination (Gar­ cia and Edqvist, 1990). A potential important area for the diagnostic use of progesterone analysis is the elucidation of clinical syndromes in the postpartum period, in which cows fail to show sexual receptivity for extended periods of time. Animals that have reestablished ovarian activity can often be distinguished from those that have not. Elevated progesterone values indicate that significant ovarian activity is present. Cows that have luteal activ­ ity can be manipulated through PGF2" treatment with a reasonable expectation of initiation of a new cycle. Progesterone determinations can also be used to verify the presence of luteal tissue in conjunction with endo­ metritis / pyometra (Pepper and Dobson, 1987) and ovarian cysts (Booth, 1988; Sprecher et al., 1990). Both conditions should respond to prostaglandin therapy. Differentiation of luteal vs follicular ovarian cysts in cattle is not usually done because it is technically diffi­ cult by palpation per rectum and because the same go­ nadotropin treatment can be used in both situations. Luteinization of the structure with a luteotropin is fol­ lowed in 10-14 days with prostaglandin treatment. In both cases though, the use of progesterone analysis is useful in establishing the therapeutic response. 2. Estrone Sulfate High concentrations of estrone sulfate are found in blood beginning at about day 80 of pregnancy and in milk from about day 100 (Fig. 22.6). In studies using

estrone sulfate concentrations in milk samples col­ lected at 120 days or more of gestation as an indicator for pregnancy status, an overall accuracy rate of 96% was found (Power et al., 1985). A similar figure has also been reported by McCaughey et al. (1982). These authors suggested that estrus and mastitis may influ­ ence the accuracy of the test. When comparing periph­ eral blood concentrations of estrone sulfate among three breeds of cattle, it was found that the breed giving birth to the lightest calves had lower estrone sulfate concentrations in the interval 101-200 days of gestation (Abdo et al., 1991).

3. Other Substances A protein of placental origin (bovine pregnancy­ specific protein B; bPSBP) has been observed in the blood of pregnant cows beginning between days 1 6 and 21 o f gestation (Sasser e t al., 1983; Sasser and Ruder, 1987). This discovery of a protein that is ob­ served only if an embryo is present opens the way for an early and definitive pregnancy diagnosis in cattle by blood analysis. Humblot et al. (1988) compared diag­ nosis of pregnancy in Friesian cattle by determination of progesterone and bPSBP in blood. The study re­ vealed the accuracy (ratio of cows positive and preg­ nant to total cows with positive) of the positive forecast to be 67.2% (82 / 122) for progesterone on day 24 after insemination. The accuracy of the negative forecast for progesterone was 98% (52 / 53). The accuracy of the positive forecast for bPSPB increased with gestation age from 86.2% (50 / 58) on day 24 to 98.8% (83 / 84) on

606

Lars-Erik Edqvist and Mats Forsberg

day 70. The accuracy of the negative diagnoses by bPSPB increased from 71.8% (84 / 11 7) on day 24 to 100% (83 / 83) on days 30 to 35. The authors concluded measurement of bPSPB 30 days after insemination is an efficient test both for positive and negative pregnancy diagnoses. A cardinal principle of pregnancy detection is that the test should be effective by the time the estrous cycle would end if the animal was not preg­ nant. This reemphasizes the main point of early preg­ nancy detection: It is done so the animal can be rebred if not pregnant. The test fulfilling this criterion has yet to be found.

B. Sheep

1. Progesterone Progesterone analysis as an early test for pregnancy has been used in sheep (Robertson and Sarda, 1971). In the ewe, the progesterone analysis has to be carried out on blood samples, because most breeds of sheep are not lactating at the time of breeding. Maximal luteal phase progesterone concentrations in the ewe are ap­ proximately 2-4 ng / ml (6-13 nmol / liter), whereas the concentrations at estrus range from 0.15 to 0.25 ng / ml (0.5-0.8 nmol / liter) (Stabenfeldt et ai., 1969c; Dickie and Holzmann, 1992). Using an amplified enzyme im­ munoassay technique for plasma progesterone, an ac­ curacy of 100% was reported for the diagnosis of preg­ nancy using samples taken between days 15 and 16 from a flock of 130 ewes (106 diagnosed pregnant and 24 diagnosed nonpregnant) (McPhee and Tiberghien, 1987). A relatively marked increase in progesterone values from 2-4 ng / ml (6-13 nmol / liter) to 12-20 ng / ml (38-64 nmol/ liter) occurs between days 60 and 125 of pregnancy (Stabenfeldt et al., 1972). This increase is due to increased progesterone production from the fetoplacental unit. The contribution of the CL to pro­ gesterone concentrations remains constant throughout pregnancy. Thus, the relatively large difference in pro­ gesterone concentrations between nonpregnant ewes with a CL present and ewes with fetus(es) present could be used to minimize the relatively high false­ positive forecasts observed for sheep at 17 days post­ breeding. In a trial that determined plasma progester­ one concentrations in 46 ewes on days 18 and 70 after mating, correct pregnancy diagnoses were made for 80% and 93% of ewes, respectively (Weigl et al., 1975). This approach might be useful for intensive sheep op­ erations for the identification of nonpregnant animals for early culling. The limitations of progesterone analy-

sis as a pregnancy test discussed for cattle apply to sheep as well.

2. Estrone Sulfate In the pregnant ewe, estrone sulfate produced from the fetoplacental unit is detected in elevated concentra­ tions beginning at around day 70 after conception (Tsang, 1978; Fig. 22.6). Worsfold et ai. (1986) deter­ mined estrone sulfate concentrations in blood from ewes bred 85 days previously and found the accuracy of the nonpregnant vs pregnant interpretations was 44% and 88%, respectively. A considerable overlap in estrone sulfate values between ewes with single and multiple fetuses was found, and even at day 1 1 6 of pregnancy, estrone sulfate concentrations overlapped between groups of ewes with single and multiple fe­ tuses. Because of this finding, the authors concluded that estrone sulfate concentrations could not be used as a predictive tool for litter size (Fletcher and Wors­ fold, 1988).

3. Other Substances Circulating pregnancy-associated proteins have been reported in sheep. A heat-labile protein with a molecular weight of around 8000 was found in sera of sheep as early as day 6 of gestation (Cerini et ai., 1976). It has been demonstrated that the bovine pregnancy­ specific protein B also can be used for early detection of pregnancy in the ewe because pregnant ewes have a blood antigen that cross reacts with antibodies to bPSPB (Ruder et ai., 1988). In a study comparing the accuracy of detecting pregnancy with an ultrasonic device, a real-time scanning instrument, and RIA for bPSPB in ewes, it was concluded that the RIA for bPSPB detected pregnancy earlier and more accurately than the ultrasonic device and was as accurate as the real­ time scanning instrument (Ruder et al., 1988). Ovine placental lactogen (oPL) has been demon­ strated in the peripheral blood of pregnant ewes (Kelly et ai., 1974). Increased maternal oPL concentrations oc­ cur between 40 and 50 days of pregnancy, reach maxi­ mum concentrations between days 120 and 140, and decline as parturition approaches (Chan et al., 1978). The determination of oPL concentrations might be used as a basis for a specific pregnancy test in the ewe.

C. Pig

1. Progesterone Luteal-phase progesterone concentrations in the pig are considerably higher than in cattle and sheep, namely, 20-50 ng / ml (64-159 nmol/ liter), whereas

607

Clinical Reproductive Endocrinology

concentrations at estrus are below 0.5 ng / ml (1.6 nrnol / liter) (Stabenfeldt et aI., 1969a). The difference in progesterone values between nonpregnant and pregnant animals 19-24 days after service has been used as an early pregnancy test (Robertson and Sarda, 1971; Edqvist et aI., 1972; Saiz et aI., 1988). In the pig, progesterone determination is usually performed on blood. Because of the sensitivity of the assay systems and the concentration of progesterone in blood in pigs during the luteal phase, analyses can be performed on a small volume of blood (about 10 drops), allowing the sample to be obtained through a small incision in an ear vein. The limitations associated with using progesterone determinations as a pregnancy test in the pig are similar to those previously discussed for cattle. Progesterone analyses have also been used to deter­ mine ovarian activity in clinically anestrous gilts, as well as to establish the stage of the estrous cycle in gilts and subsequent response to treatment (King et aI., 1985). It is also possible to monitor luteal-phase activity in the sow through measurement of fecal gestagens (Hulten et aI., 1994).

2. Estrone Sulfate Previous studies of estrogen concentrations in urine revealed a marked increase in estrogen between day 20 and 30 of pregnancy in pigs (VelIe, 1958). Studies of blood concentrations of estrone sulfate in pigs during pregnancy showed patterns similar to those deter­ mined in urine (Robertson and King, 1974; Fig. 22.6). In early pregnancy in the pig, it has been hypothesized that estrogen synthesized by the early preimplantation embryo may be the messenger for the maternal recog­ nition of pregnancy (Perry et al., 1976) and that the elevated estrone sulfate concentrations in the maternal circulation during early pregnancy reflect fetal synthe­ sis. The determination of estrone sulfate in early preg­ nancy in the pig is thus a specific pregnancy test (Edq­ vist et aI., 1980, Cunningham et aI., 1983). The index of discrimination between estrone sulfate concentrations in blood of pregnant vs nonpregnant pigs around 25 days after breeding is very high. As a pregnancy test, Sugiyama et aI. (1985) reported an accuracy rate of 98% in pigs between days 20 and 26 of pregnancy. Horne et al. (1983) found litter size on days 20-26 of pregnancy to be positively correlated with estrone sulfate values, but such a relationship was not found later in preg­ nancy. During the latter part of pregnancy, the mater­ nal blood concentrations of estrogens (estrone sulfate, estrone, and estradiol-17f3) are very high and can thus be used to confirm pregnancy at this stage of gestation (Fig. 22.6).

In the pregnant sow, the concentration of estrogen in urine and feces follows the same pattern as in blood (Choi et aI., 1987). Szenci et aI. (1993) compared ultraso­ nography and the determination of unconjugated es­ trogen in feces at 25 to 30 days after insemination for the diagnosis of pregnancy in pigs. Based on farrowing data, the positive and negative predictive values for the ultrasound were 93.2 and 100%, respectively. For the determination of unconjugated estrogens, the cor­ responding predictive values were 93.8 and 55.1 %, re­ spectively.

D. Horse 1. eCG (PMSG) Although pregnancy diagnosis has been done in mares through the measurement of equine chorionic gonadotropin (eCG), formerly named pregnant mare serum gonadotropin (PMSG), the main drawback to its use is that the presence of eCG does not guarantee the presence of a fetus, but indicates that a viable fetus was present at the time of endometrial cup formation. This situation exists because endometrial cups have autonomy of function and continue to secrete PMSG for a period of time in spite of loss of the fetus (Allen, 1969). This means that both mares with normal preg­ nancies and mares that experience embryonic mortal­ ity after day 40 of gestation will have elevated eCG concentrations in blood. The use of eCG determina­ tions as a positive pregnancy diagnosis test will cause some mares to be diagnosed as pregnant that will not deliver a foal (Mitchell, 1971; Jeffcott et aI., 1987; Fig. 22.8).

2. Progesterone Progesterone analysis is useful for establishing the presence or absence of ovarian activity in animals with puzzling behavioral patterns. Agitated or aggressive behavior, often interpreted as sexual in orientation, occurs without regard to luteal status. Progesterone analysis can be helpful because elevated values directly indicate the presence of a CL and, additionally, are evidence that folliculogenesis and ovulation are nor­ mal. Relatively low luteal-phase values for progester­ one (1-3 ng / ml; 3-10 nmol / liter) vs normal luteal­ phase values (>3 ng / ml; >10 nmol / liter) are often associated with the presence of a persistent CL (Staben­ feldt et aI., 1974b). Progesterone values of mares with persistent luteal activity are low because some PGF2a synthesis and release often occurs about 14 days post­ ovulation, albeit insufficient to cause complete luteo­ lysis (Neely et aI., 1979).

608

Lars-Erik Edqvist and Mats Forsberg 1 00000

. 1 0000

� C!l

Q)

()

1 000

1 00 50

0

10

30

50

70

90

110

1 30

1 50

DAYS AFTER OVULATION 1 000 c:::

'0 E

>250

c

!;i: w

I a.. ...J :::> en w z 0 a: fen w

1 00

10 5

o

10 20 30 40 50 60 70 80 90 1 00 1 1 0 1 20 1 30 1 40 1 50

DAY OF P REGNANCY

FIGURE 22.8 (Upper) Blood plasma concentrations of eCG in five normal pregnant mares (0) and in one mare in which the fetus died around day 50 of pregnancy (e). (Lower) Geometric means of blood plasma concentrations of estrone sulfate in the five normal mares (shaded area) and the estrone sulfate concentration in the mare in which the fetus died around day 50 and was aborted on day 113 of pregnancy (Darenius et al., 1988).

Progesterone analyses can also be useful in mares that fail to manifest sexual receptivity, yet have cyclic ovarian activity (Hughes et al., 1973). Progesterone analysis at 5-day intervals over 20 days (approximately one estrous cycle length) can verify the presence or absence of cyclic ovarian activity. The time of ovulation can also be predicted within a 2- to 3-day interval, a prediction that can be helpful to the veterinary prac­ titioner anticipating the next time of ovulation. Breed­ ing may have to occur by artificial means in these

situations. Hinrichs et al. (1988) compared the accuracy of determining day of ovulation ::!:: 1 day using three different methods: (1) an immediate, qualitative ELISA for progesterone in blood, (2) a quantitative ELISA for progesterone in blood, and (3) daily teasing to detect estrus. Ovulation was detected by ultrasound examina­ tion per rectum. The accuracy in determining day of ovulation ::!:: 1 day using the three methods was 72% for the qualitative progesterone assay, 88% for the quantitative progesterone assay, and 86% for teasing.

Clinical Reproductive Endocrinology

609

3. Conjugated Estrogens

4. Testosterone

Estrone in its unconjugated form (free estrogen) re­ flects important physiological events in the mare begin­ ning at about day 75 of gestation when the fetoplacen­ tal unit begins to produce estrone in rapidly increasing amounts (Nett et al., 1975). More importantly, it has been shown that estrone is rapidly conjugated after secretion to water-soluble estrogen conjugates and the ratio between free and conjugated forms is 1 : 100 (Ter­ qui and Palmer, 1979). Terqui and Palmer (1979) and Kindahl et al. (1982) have shown that significant in­ creases in estrone conjugate concentrations occur be­ tween days 35 and 40 of gestation (Fig. 22.8). Kindahl et al. (1982) showed the increase to be 10- to 20-fold between days 35 and 40. Concentrations then decline slightly, with a further increase noted at the same time that free estrone concentrations begin to increase. The initial source of increased estrone production during gestation days 20 and 70 is the ovaries (Daels et al., 1990). Later on, its occurrence is likely driven by the attachment of the embryo and the production of eCG (PMSG). Sist et al. (1987) found a majority of pregnant mares to have significantly higher serum estrone sul­ fate concentrations by day 60 after breeding. They also determined estrone sulfate in milk from pregnant mares and reported concentrations to be lower than in serum but following the same pattern. In the authors' experience, mares should be pregnant for more than 90 days before analyses of estrone sulfate in plasma can serve as a reliable indicator of pregnancy (Fig. 22.8). Estrone conjugate concentrations in urine can be used to document pregnancy in the mare and, in fact, may be more accurate than plasma analysis because of the concentrating aspects associated with urine for­ mation (Daels et al., 1991 ). Increased amounts of fecal estrogens and gestagens have also been reported in the pregnant mare (Schwarzenberger et al., 1991 ). In addition to being a means of confirmation of pregnancy, estrone conjugate analysis also allows the soundness of the pregnancy to be assessed because estrogen concentrations reflect the dynamics of a grow­ ing fetus. Thus, it is possible not only to indicate that pregnancy is in progress by estrone conjugate analysis, but also to indicate whether the pregnancy is proceed­ ing well or is in some state of compromise. Analysis of estrone conjugates is an important aid for the verifi­ cation of pregnancy in the mare and does not suffer the drawback of false positive diagnosis as is the case for eCG (Fig. 22.8). Estrone conjugate analysis also has been used to assess follicle growth patterns in nonpreg­ nant mares, a test that was difficult to do by determin­ ing free estradiol-17f3 (Makawiti et al., 1983; Daels et al., 1991).

Testosterone values vary in the mare according to the reproductive state. Values, usually less than 15 pg / ml (52 pmol / liter) during anestrum, range between 20 and 40 pg / ml (69-139 pmol / liter) during cyclic ovarian activity, with the higher values being observed during the follicular phase of the cycle immediately before ovulation. Testosterone determinations have been used to aid the diagnosis of granulosa-theca cell tumors in the mare (Stabenfeldt et al., 1979) and to differentiate granulosa-theca cell tumors from ovarian teratoma (Panciera et al., 1991). Leydig-like cells in the theca appear to be the source of testosterone. In cases of granulosa-theca cell tumors, testosterone values vary with values ranging from 40 pg to over 100 pg/ ml (139-347 pmol / liter). These tumors generally develop slowly, and it is not known whether this slow develop­ ment also reflects a slowly developing capacity for testosterone production or whether there is variability in the number of testosterone-secreting cells among tumors. Aggressive stallion-like behavior is associated with values less than 150 pg/ ml (520 pmol/ liter) (Meinecke and Gips, 1987). In the authors' experience, this is the only clinical situation involving abnormal behavior in the mare in which a direct relationship to gonadal steroids has been established. Testosterone determinations in the male horse have been used as an aid in the diagnosis of cryptorchidism. Cox (1975) reported that horses with less than 40 pg / ml (139 pmol / liter) plasma should be considered cas­ trated, whereas animals with concentrations less than 100 pg/ ml (347 pmol / liter) should be considered as having testicular tissue present. Although some cryp­ torchid animals have testosterone concentrations less than 100 pg / ml (347 pmol l liter), most have values range from 200 to 1000 pg / ml (693-3467 pmoll liter) (Cox, 1975). Testosterone concentrations in intact males usually range from 1000 to 2000 pg / ml (3467-6934 pmol l liter) (Cox et al., 1973; Berndtson et al., 1974). hCG administration for the purpose of stimulating tes­ tosterone production by the testes has been suggested as a means of resolving cases in which values are be­ tween 40 and 100 pg/ml (139-347 pmol / liter). The dosage of injected hCG has been 6000-12,000 LU. (Cox et al., 1986; Arighi and Bosu, 1989; Silberzahn et al., 1989), and the second sample for analysis of testoster­ one was obtained after 30-120 minutes (Cox et al., 1986), after 24 hours (Arighi and Bosu, 1989), or after 3 days (Silberzahn et al., 1989). For routine diagnosis, the authors do not recommend the use of hCG because most single baseline determinations are adequate for diagnosis. The main value of a second determination is for substantiation of the first values, and thus the

610

Lars-Erik Edqvist and Mats Forsberg

time and expense required for the second sample is not commensurate with the information to be gained (Stabenfeldt and Hughes, 1980). Analysis for estrone sulfate conjugates for the diag­ nosis of cryptorchidism may be preferred over that for testosterone because it requires only one analysis and the accuracy is slightly improved (Cox et ai., 1986). Arighi and Bosu (1989) recommended both resting tes­ tosterone and estrone sulfate to diagnose accurately the presence of testicular tissue. It should be noted, however, that estrone sulfate analysis for cryptorchi­ dism cannot be used in horses less than 3 years of age or in donkeys because little estrone sulfate is produced in either of these situations. It is the authors' experience that the presence of sexual behavior in suspected cases of equine cryptor­ chidism is not necessarily dependent upon elevated testosterone concentrations in that many patients (about one-third) referred because of behavioral prob­ lems are, in fact, castrated. In essence, some animals can maintain normal libido with low circulating con­ centrations of testosterone. In this situation, testoster­ one analysis is very helpful either to eliminate unneces­ sary surgery or to indicate to the surgeon that testicular tissue is present and should be found on surgical entry.

E. Dog 1. Progesterone In the bitch, there is little difference between the progesterone patterns of the pregnant and nonpreg­ nant luteal phase. Plasma concentrations of progester­ one are elevated throughout the luteal phase. How­ ever, although mean concentrations of progesterone are higher during the latter part of the pregnant luteal phase than during the nonpregnant luteal phase, these differences are not significant enough to allow proges­ terone determinations to be used as a pregnancy diag­ nosis test. Bitches usually ovulate at the onset of sexual recep­ tivity, remain sexually receptive for about the first week of luteal activity, and, in fact, are fertile during this period of time (Holst and Phemister, 1974). Proges­ terone analysis can be used to confirm the occurrence of ovulation. Concentrations usually increase slightly above baseline in the periovulatory period, reaching about 15 nmol / liter at ovulation, followed by a sus­ tained increase beginning 24 hours following ovulation (Concannon et ai., 1975; Fig. 22.9). This approach may be used in a retrospective analysis of a breeding cycle in an attempt to correlate the time of ovulation with other criteria such as vaginal cytological changes. The determination of progesterone in daily samples of

blood obtained during the peri ovulatory period gives a more precise timing of ovulation than can be obtained by vaginal cytology (Linde and Karlsson, 1984). In situations wherein a particular pairing of animals does not result in a mutual sexual attraction and where artificial insemination must be used, progesterone analysis is a useful tool to verify ovulation. Artificial insemination with frozen-thawed semen usually re­ sults in lower pregnancy rates than artificial insemina­ tion using fresh semen. Using progesterone determina­ tions to pinpoint ovulation, it was found that the pregnancy rate following insemination with frozen­ thawed semen could be improved when the insemina­ tion was performed at progesterone concentrations less than 30 nmol / liter (Linde-Forsberg and Forsberg, 1989, 1993; Fig. 22.9). Progesterone analysis can also be useful in cases of short estrous-cycle intervals to determine if ovulatory failure has occurred, a situation in which progesterone concentrations are low following the termination of estrus. Progesterone analysis is also useful in cases when there is some question as to the completeness of removal of ovarian tissue during an ovariohysterec­ tomy. An important distinction concerning the dog is that, unlike other domestic species, the bitch is keyed into sexual receptivity at the end of the follicular phase by progesterone. Thus, dogs may be sexually attractive because of odors arising from the vagina that are due to factors such as infection. However, bitches will only accept males in the presence of increased progesterone concentrations, a situation which is almost always as­ sociated with luteal tissue as part of a remaining ovar­ ian remnant.

2. Testosterone The most common endocrine test in male dogs is testosterone for the purpose of checking the secretory status of the Leydig cells. Testosterone values in nor­ mal dogs range from about 1 ng/ ml (3.5 nmol / liter) to about 10 ng / ml (35 nmol / liter) because of the pulsatile release pattern of testosterone. Testosterone concentra­ tions in castrated dogs are less than 0.5 nmol / liter. Assay of basal testosterone concentration may allow the diagnosis of the absence of testicular tissue in thE same manner as described previously for the malE horse. Sometimes the positive confirmation of presenCE of testicular tissue may require the use of hCG or GnRH to stimulate testosterone production. In this situation, a resting plasma sample is collected immediately be· fore the administration of the stimulating hormone, and a second blood sample is collected 1 hour later. A significant increase in plasma testosterone concen· tration is diagnostic of the presence of testicular tissue

611

Clinical Reproductive Endocrinology

Estradiol

90 s o

75

c

60

�

E

"*CD

e Cl

45

e Co 30 g iii 1 5

"0

o

Progesterone Proestrus 9 days (range 3-21 days)

I Estrus 9 days

(range 3 - 21 days)

Metestrus

FIGURE 22.9 Schematic representation of the temporal relationships among estradiol, progesterone, peak of luteinzing hormone (LH), and ovulation in the bitch. The recommended time of mating or artificial insemination is based on the progesterone concentration.

Testosterone analysis is done in conjunction with fertility examinations, often in stud animals that are presented as infertility cases following a prolonged show tour. Most of these animals have testosterone concentrations that are compatible with normal sper­ matogenesis, although sperm counts are often very low. The management of this syndrome is still un­ certain. The authors have observed feminizing syndromes in intact dogs in which concentrations of testosterone are greatly decreased below normal (to 100 pg / ml or less; :5350 pmol / liter), and estradiol values are 2-3 times normal (30-45 pg / ml vs the normal 15 pg/ ml; 110-165 pmol / liter and 55 pmol / liter). It is thought that a majority of these cases involve Sertoli cell tu­ mors. Dogs with confirmed Sertoli cell tumors have high peripheral plasma levels of inhibin and sup­ pressed levels of LH and testosterone (Grootenhuis et aI., 1990). These authors were, however, unable to de­ tect differences in blood concentrations of estradiol between dogs with Sertoli cell tumors and control dogs.

3. Relaxin Relaxin is produced in the pregnant dog beginning between days 20 and 25 of gestation but is undetectable during anestrus, throughout nonpregnant ovarian cy­ cles, and in male dogs (Steinetz et aI., 1987). Maximal concentrations are attained by days 40-50 of preg­ nancy and are followed by slight declines before partu­ rition. Relaxin is produced predominantly by the pla­ centa, and it is thus possible to use relaxin analysis to confirm pregnancy in the dog.

F. Cat

1. Progesterone Progesterone analysis can be used to verify the oc­ currence of ovulation in the cat following coitus. Ovu­ lation usually occurs 24-36 hours after coital contact with a male at the appropriate time of the follicular phase of the estrous cycle (Shille et al., 1983). If the breeding schedule is very limited in time, it is possible for a cat to be bred too early in the follicular phase with coitus failing to elicit LH release. More rarely, females may allow copulation at times other than the follicular phase. In both these situations, ovulatory fail­ ure would be documented by the finding of low pro­ gesterone values 10 days postbreeding. In one study, progesterone concentrations greater than 1 .87 ng / ml (6 nmol / liter) were consistently associated with luteal­ phase ovaries, and values less than 0.15 ng / ml (0.5 nmol / liter) were associated with follicular-phase ovaries (Lawler et aI., 1991).

2. Estrogen As the cat is an induced ovulator (requires coitus), estradiol analysis can be used to assess the presence of ovarian follicle activity with values ranging from 10 pg/ ml (37 pmol/ liter) in the interfollicular phase to 60 pg/ml (220 pmol/ liter) during folliculogenesis. As indicated previously, queens have ovarian follicle growth patterns that last 5-7 days, followed by a slightly longer interval prior to the next growth phase (Shille et al., 1979b). Estrogen analysis could document the presence or absence of ovarian follicular activity in animals that fail to manifest sexual activity. It also

612

Lars-Erik Edqvist and Mats Forsberg

could be used to assess the completeness of an ovario­ hysterectomy in cats that are spayed, but that present signs suggestive of sexual receptivity. The usual find­ ing is that of low estrogen concentrations, which indi­ cates that the behavior is not sexual in orientation. In the domestic cat, estrogen metabolites are primarily excreted in the feces (Shille et ai., 1990; Most! et ai., 1993). Analyses of fecal steroids is a useful noninvasive approach for monitoring ovarian function in exotic felines (Graham et ai., 1993).

3. Testosterone Testosterone analyses can be used to evaluate Leydig cell function in the testis of the male cat. The range of values is usually between 1 and 10 ng / ml (3.5-35 nmo i l liter).

4. Relaxin Relaxin is produced by the fetoplacental unit begin­ . mng about day 20 of gestation. Maximal concentrations are achieved by day 30-35 (Addiego et al., 1987). Re­ laxin concentrations thus can be used to assess preg­ nancy status in the cat and even to assess its normalcy, based on the fact that fetoplacental units in jeopardy produce less relaxin.

V. GENERAL COMMENTS A. Hormone Concentrations In the foregoing presentation, concentrations of hor­ mones have not been emphasized. This is because there is still some variability in the values reported by vari­ ous laboratories. It is important that clinical endocri­ nology laboratories understand and have experience with their assay systems in relation to particular clini­ cal syndromes. For example, the actual concentration of progesterone during the follicular phase of the estrous cycle of domestic animals is approximately 100 pg/ml (318 pmol / liter), certainly no greater than 200 pg / ml (636 pmoil liter) plasma. Some laborato­ ries, however, report basal values of 1-2 ng / ml (3.26.4 nmoil liter) for progesterone. The authors have used 1 ng / ml (3.2 nm01 l liter) as the lowest concentra­ tion of progesterone that is compatible with an actively secreting CL, especially for the cow, ewe, mare, and sow, and pay particular attention to the accuracy of values in this range. Actually, even values between 0.5 and 1.0 ng / ml (1.6-3.2 nmoi l liter) are viewed as indicating luteal activity. It is possible to produce use­ ful information with higher basal concentrations; how­ ever, the line that divides the presence and absence of

.... ..I- - -L-.. .. 1 ..L---L - - - --- - I l-r 1

1 20

..

1 00 BO

DOG

..

- - - -

-

----�--�--��, �J-'�--�----�

L -- -L :I� I t

1 20

.. .. ..

LU 1 00

�d]

� �

BO

, � . ---�--�---'-

1 20

LU z o a: 1 00 LU I(f)

�

o a: a.. z 12 mg / dl) and anal-sac adenocarcino­ mas (CAC), lymphoma, or miscellaneous tumors (MISC TUMOR); and dogs with normocalcemia « 12 mg/dl) and anal sac adenocarcinomas, lymphoma, or miscellaneous tumors. From Rosol et al. (1992b).

increased in hypercalcemic dogs and was significantly less than that in dogs with primary hyperparathyroid­ ism (Fig. 23.35). Surgical removal or radiation therapy of the adenocarcinoma results in a rapid return to nor-

1 00

�

80

S, 60 a.. 'I

I- 40 a..

.. ..

20 0 12

..

.. .. ..

.. ..

..

..

.. .. ..

.. .. ..

.. ..

.. ..

, .. ,

.. .. .. ,

.. ,

.. .. ..

,

, ,,

.. ..

,,

.. .. ..

Total Calcium (mg/d\) 14

16

18

, ,,

*.

.. .. 20

FIGURE 23.33 Regression of PTHrP and total calcium in dogs with anal sac Ca. There was a significant linear correlation (0.87, P < 0.01) between serum calcium and N-terminal PTHrP concentra­ tions in dogs with humoral hypercalcemia of malignancy and adeno­ carcinomas derived from apocrine glands of the anal sac.

mal of serum calcium and phosphorus, increased se­ rum PTH, and decreased 1,25-dihydroxyvitamin 0 (Rosol et al., 1992b). Postsurgical survival in dogs with adenocarcinoma and hypercalcemia ranged from 2 to 21 months, with a mean of 8.8 months. Sublumbar metastases occurred in a high percentage (94%) of the dogs and were associated with a recrudescence of the biochemical alterations in serum and urine.

ii. Nude-Mouse Model of the Canine Anal Sac Adenocarcinoma A model of HHM has been devel­ oped in nude mice utilizing a serially transplantable anal sac adenocarcinoma (CAC-8) derived from a hy­ percalcemic dog (Rosol et al., 1986). Nude mice with transplanted CAC-8 developed severe hypercalcemia and hypophosphatemia. Serum calcium concentration returned to the normal range in two days after surgical removal of the tumors. Serum 1,25-dihydroxyvitamin o levels were significantly increased and were corre­ lated with serum calcium levels in CAC-8-bearing nude mice. Histomorphometric analysis of lumbar ver­ tebrae revealed increased bone formation and resorp­ tion. The adenocarcinoma (CAC-8) produces PTHrP,

Thomas J. Rosol and Charles C. Capen

658 90 80

E 0, .9:

0 z

�

�

:>

>x 0 a:: 0 >I

(5 .n "!.

•

70

50 �

30 20

"

,

•

60

40

•

L

_ _ _ _ _ _ _ _ _

... ·

� • r

:

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

...

� ;

0

g-

r

,

' 6

10 mg/ dl (4.1 mmol l liter) is adequate; 2.0-10 mg / dl (0.8-4.1 mmoi l liter) is marginal; and