ELSEVIER

DOMESTIC ANIMAL ENDOCRINOLOGY

Vol. 14(6):381-390, 1998

THE P H Y S I O L O G I C A L EFFECTS OF N A T U R A L V A R I A T I O N IN G R O W T H H O R M O N E GENE C O P Y N U M B E R IN R A M L A M B S

E. Gootwine, .1 J.M. Suttie,** J.C. McEwan,** B.A. Veenvliet,** R.P. Littlejohn,** P.F. Fennessy,** and G.W. Montgomery***

*Institute of Animal Science, ARO, The Volcani Center, Bet Dagan 50250, Israel, **AgResearch, Invermay Agricultural Centre, Private Bag 50034, Mosgiel,

New Zealand, and ***AgResearch Molecular Biology Unit, Department of Biochemistry and Centre for Gene Research, University of Otago, PO Box 56,

Dunedin, New Zealand

Received Janua~ 16,1997 Accepted June 27,1997

The effects of natural variation in the number of copies of the growth hormone (GH) gene on growth parameters, plasma GH profiles, and the response to GHRH challenge were compared in Coopworth ram lambs from selection lines differing in body composition and GH levels. Different genotypes at the GH locus carried two, three, or four copies of the GH gene and GH secretion was studied under ad libitum feeding conditions and in the fasted state. There were no significant effects of GH genotype on any parameters of growth or body composition. Basal serum GH concentration, GH pulse frequency, and GH pulse amplitude differed significantly with selection line and fasting, but did not differ significantly between the GH genotypes. Significant differences of subtle nature were found between the GH genotypes in their responsiveness to GHRH. For the ad libitum-fed Lean selection line animals, the first GHRH challenge resulted in a higher mean maximum response for GH1/GH1 than GH2/GH2 (P < 0.05). Between the first and the second challenges there was a decrease in maximum response for the GH1/GH1 genotype and an increase for the GH2/GH2 genotype (P < 0.05 for GH genotype main effect). The differences between GH genotypes in response to GHRH challenge suggest that polymorphism in the number of GH gene copies in sheep may have physiological implications for the function of the GH axis, which may be manifested in growing lambs only under specific genotype-environment combinations. © Elsevier Science Inc. 1998

I N T R O D U C T I O N

Growth hormone (GH) plays a key role in regulating tissue growth and metabolism in lambs. Growth rate or body composit ion of lambs can be manipulated by daily adminis- tration of exogenous human, bovine, or ovine recombinant GH, by immunization against hormones of the GH axis, and by the production of transgenic sheep carrying additional copies of the GH gene (1). Moreover, changes in growth rate and body composit ion in sheep by nutritional means, or after genetic selection, may be associated with alterations in the secretary pattern of GH (1,2). In Coopworth lambs, GH pulse frequency and GH response to growth hormone-releasing hormone (GHRH) differ in ram lambs from divergent lines selected either for or against fatness (3,4).

Sheep are polymorphic for the number of GH gene copies, unlike cattle and pigs, which carry only one copy of the GH gene. There are two alleles at the ovine GH locus: the GH1 allele with a single GH copy, and the GH2 allele that contains two GH copies of the GH coding region, termed GH2-N and GH2-Z (5,6). Consequently, three possible genotypes

© Elsevier Science Inc. 1997 0739-7240/98/$17.00 655 Avenue of the Americas, New York, NY 10010 PII S0739-7240(98)00043-X

382 GOOTWlNE ET AL.

TABLE 1. DISTRIBUTION OF LAMB PROGENY FOR FOUR HETERDZYGOUS GH1/GH2 RAMS CLASSIFIED BY SELECTION

LINE, SIRE, AND G H GENOTYPE

Sire Selection G H genotype

no. line GH1/GHI GHI/GH2 GH2/GH2

No. of lambs

232 Fat 7 23 9 293 Fat l0 II 12 479 Control 12 24 2 160 Lean 5 12 3 Overal l 34 70 26

exist: GH1/GH1, GHI/GH2, and GH2/GH2. In most sheep breeds, the GH1 allele has a low frequency of about 0.1., but the Coopworth breed, the GH1 allele frequency is about 0.5 (E. Gootwine, unpublished results).

The ovine GH gene is about 1.7 kb in length (7). Sequence analysis has shown that the GH2-N and the GH2-Z are almost identical in their nucleotide sequences (8), but only the GH2-N copy of the gene seems to be expressed in the pituitary (9). However, if the gene duplication influences GH gene expression or transcription, then GH genotypes may differ in their GH plasma levels, growth, or body composition.

The present study was undertaken to examine the physiological effects of the duplica- tion at the ovine GH locus. Specifically, our aim was to compare growth rate, body composition, and GH secretion in Coopworth lambs with GH1/GH1, GH1/GH2, or GH2/GH2 genotypes born in half-sib families from Coopworth rams that were heterozy- gous for the GH gene duplication. We used families derived from selection lines that differed in body composition and GH levels (10).

MATERIALS AND METHODS

Experiments were performed in accordance with the 1987 Animal Protection (Codes of Ethical Conduct) Regulations of New Zealand, after approval had been granted by the Invermay Agricultural Centre Animal Ethics Committee.

1 2 3 4 5 6

12.2

6.7

Figure 1. Geno typ ing at the ovine G H locus after EcoRl restriction analys is o f genomic DNA. Lanes 1, 3, and 5 GHI/GH2. Lanes 2 and 6 GH1/GHI. Lane 4 GH2/GH2. Size o f restr ict ion f ragments in kb is indicated.

GH GENE DUPLICATION AND GH RELEASE IN LAMBS 383

TABLE 2. L E A S T SQUARES MEANS AND STANDARD ERRORS OF DIFFERENCE (SED) FOR LIVEWEIGHTS AND BACKFAT

DEPTH OF LAMBS CLASSIHED BY GH GENOTYPE AND SELECTION LINE, WITH SIGNIFICANCE OF GH GENOTYPE BY

SELECTION LINE INTERACTION (NS = NOT SIGNIFICANT).

Selection GH genotype Average Parameter line G H 1 / G H 1 GH1/GH2 GH2/GH2 SED sig

Birth weight (kg) F 4.06 3.86 3.71 C 4.49 4.48 4.70 0.28 ns L 5.62 4.76 5.16

Weaning weight (kg) F 22.6 22.1 22.2 C 23.6 22.6 23.7 1.45 ns L 28.3 24.0 21.4

Six-month weight (kg) F 39.1 38.9 39.2 C 42.4 42.1 43.2 2.22 ns L 51.0 46.3 40.1

Fat depth C (mm) F 4.1 4.6 4. l C 3.0 2.7 2.9 0.63 ns L 1.5 1.9 1.6

Tissue depth GR (mm) F 18.0 19.1 18.0 C 12.9 13.6 12.6 1.85 ns L 8.3 8.5 7.9

Animal Selection and Management. Coopworth lambs (n = 401) born in 1991 at the Invermay Agricultural Research Centre, were from 12 sires used in lines selected for (Fat line) or against (Lean line) weight-adjusted ultrasonic backfat thickness (10), or from an unselected control line. In a prel iminary survey, four of these sires were found to be heterozygous GH1/GH2, two of which were from the Fat line, one from the Lean line and one from the Control line. All 130 male and female progeny of those four sires from their respective selection lines were genotyped at the GH locus (Table 1).

The lambs were grazed together from birth to weaning at approximately 10 wk of age. Subsequently, the lambs were allocated into three feeding groups: males or females grazed on pasture, and males fed indoors on a pelleted concentrate ration. Parentage, date of birth, birth and rearing rank, birth weight, weaning weight, and live weight at 6 mos were recorded for all lambs. In addition, ultrasonic measurements were recorded at 6 mos of age: backfat depth over the middle of the exterior surface of the L. dorsi muscle at the 12 rib (C) and the tissue depth (GR) at the point 11.0 cm from the backbone over the 12 rib (11).

Genotyping at the GH Locus. Blood samples were collected and DNA extracted for Southern blotting analysis (12). DNA samples were digested with EcoRI (Boehringer Mannheim, Germany), separated on 0.8% (W/V) agarose gel, transferred to Hybond-N + membranes (Amersham, Bucks, UK), and hybridized with a goat GH cDNA probe (13) labeled with [a32P]-dCTP. The genotype at the GH locus was determined for each individual according to the EcoRI restriction pattern (Figure 1).

Blood Sampling for Plasma GH Level . For the GH profile analysis, three ram lambs were randomly selected from each GH genotype for each of the for heterozygous sires, except for GH2/GH2 progeny of ram 479, from which only one lamb was available. Thus, 34 of the 130 GH-genotyped lambs were sampled. The lambs were 8 mos of age at the start of blood sampling. One week before sampling commenced, the lambs were brought indoors, kept in group pens and offered a pelleted diet ad libitum. On the day of the measurement, blood samples (4 ml) were withdrawn at 12-min intervals for 9 hr, via an indwelling jugular cannula, as described previously (3). At 6 and 7.5 hr after blood sampling, an intravenous bolus of synthetic human GHRH (1-29 fragment) (Sigma Chemical Company, St. Louis, MI, USA) was given to each lamb at a dose of 0.1 mg/kg

384 GOOTWINE ET AL.

15

- t - 5

0

15

10 .So "T"

5

0

15

" i "

(5 5

0

15 E

T (5 5

0

Fat:GH1/GH1

_ Fa t : GH2/GH2

i L i i i

~ ~ L e a n : GH1/GH ;

Lean: GH,2/GH2

I I 1 I I I

0 1 2 3 4 5 6

Time (hours)

Figure 2. GH profiles over the first 6 hr of sampling for one typical animal in the fed state from each of (a) the GHI/GH1 genotype from the Fat selection line, (b) the GH2/GH2 genotype from the Fat selection line, (c) the GH1/GH1 genotype from the Lean selection line, (d) the GH2/GH2 genotype from the Lean selection line. Samples detected as peaks by Pulsar are denoted by open circles.

body weight. Five days after the first blood sampling, the lambs were fasted for 48 hr and the blood sampling regime was repeated.

G H Assay. Plasma was separated from the blood samples, frozen, and stored at - 2 0 ° C until assay. Ovine GH was measured in plasma samples by a homologous radio- immunoassay as described previously (3). The inter- and intra- assay coefficients of variation were 10.3% and 12.0%, respectively, based on plasma control pools averaging 4.5, 9.3, and 21.9 ng/ml. The minimum detectable dose was 0.5 ng/ml.

S ta t i s t ica l Analys is . Live weights and fat depth records were analyzed by A N O V A (14). The treatment terms fitted were selection line (Fat, Control, and Lean,), sire nested within selection line, GH genotype (GH1/GH1, GH1/GH2, GH2/GH2), and the interac- tion of these terms. As the Lean and Control lines were represented by progeny from only one sire, the selection line, and the sire effects were confounded; thus genotype differ- ences were assessed within sires rather than between sires.

GH profiles were analyzed by means of the pulse detection routine PULSAR (15), with standard deviation as a function of dose given by (5.48 x dose + 11.30), the peak- splitting criterion set at 2.70 and a smoothing window of 3 hr. Remaining inputs were the default values suggested by Merriam and Wachter (15). The pulse parameters (log mean GH, mean basal GH, pulse frequency, and log pulse amplitude) derived from the PULSAR

GH GENE DUPLICATION AND GH RELEASE IN LAMBS 385

TABLE 3. LEAST SQUARE MEANS AND STANDARD ERRORS OF DIFFERENCE (SED, FOR ARITHMETIC MEANS) AND

STANDARD ERRORS OF THE RATIO (SER, FOR GEOMETRIC MEANS) FOR G H PROFILE PARAMETERS CLASSIFIED BY

G H GENOTYPE AND SELECTION LINE (AVERAGED OVER NUTRITION TREATMENT), WITH SIGNIFICANCE OF G H

GENOTYPE BY SELECTION LINE INTERACTIONS (NS = NOT SIGNIFICANT)

Selection G H genotype Average

Paramete r line GH1/GH1 GH1/GH2 GH2/GH2 SED/SER sig

G H concent ra t ion F 2.4 2.0 2.0 (geometr ic mean ng/ml) C 3.8 3.6 5.6 1.26 ns

L 7.6 5.1 4.9 Basal G H concentra t ion (ng/ml) F 1.8 2.0 1.7

C 3.0 2.9 3.4 0.73 ns L 4.9 3.7 3.4

G H pulse f requency (peaks/6 hr) F 4.5 3.4 3.8 C 4.0 3.5 5.5 0.71 ns L 5.5 5.2 4.7

G H pulse ampl i tude F 3.5 2.4 2.7 (geometr ic mean, ng/ml) C 4.4 4.3 5.3 1.24 ns

L 8.6 6.1 5.7

analysis were analyzed by ANOVA (14). The main effects fitted were selection line (Lean, Fat, or Control), sire-nested within the selection line, GH genotype (GH1/GH1, GH1/GH2, or GH2/GH2), nutritional level (fed, fasting), and all interactions involving these terms. Terms involving nutritional treatment were tested in the within-animal stratum and other terms were tested in the between-animal stratum. Statistical significance was assessed at the 5% level unless otherwise stated.

To analyze the GH response to the GHRH challenges, each response profile was modeled by the critical exponential curve

G H = a + (b + ct)r ' + e

where t is time, a, b, and c are linear parameters of the response curve, r is a shape parameter of the response curve, and e is normally distributed error. The parameter c expresses the response at high value of t, and is included to prevent biases in estimating the other parameters. For both challenges, comparison of curve parameters between selection lines and GH genotypes was performed with time nested within the ANOVA framework described for the GH pulse analysis, according to Butler and Brain (16). In addition, the maximum postchallenge GH concentration for each challenge was obtained for ANOVA by means of the model described for the GH pulse analysis.

RESULTS

GH Genotype and Allele Frequency. The distribution of the GH genotypes among the four half-sib families derived from the heterozygous GH1/GH2 sires is given in Table 1. Lambs from all the three possible genotypes were present in the four families and the frequencies of the GH1/GH1, GH1/GH2, and GH2/GH2 genotypes among the lambs were 0.26, 0.54, and 0.20, respectively. From these data, the dams' allele frequencies, which did not differ significantly between selection lines, were estimated to be 0.57 and 0.43 for the GH1 and the GH2 alleles, respectively.

Growth and Fat Deposition. Mean growth and body composition classified by se- lection line and GH genotype are presented in Table 2. These results did not differ significantly from results for all 401 lambs born in 1991 (data not shown). There was a significant effect of selection line on live weights, C, and GR, but there were no significant effects of GH genotype on any parameters of growth or body composition.

GI-I Profiles. GH was released in a pulsatile manner in all lambs, and typical GH profiles for the different treatment groups in the fed state are presented in Figure 2.

386 GOOTWINE ET AL.

TABLE 4. MEANS AND STANDARD ERRORS OF DIFFERENCES (SED, ARITHMETIC MEANS) OR STANDARD ERRORS OF

RATIOS (SER, GEOMETRIC MEANS) FOR GH PROFILES CLASSIFIED BY NUTRITIONAL TREATMENT, WITH

SIGNIFICANCE OF NUTRITIONAL TREATMENT (NS = NOT SIGNIFICANT; ** = P '~ 0 . 0 1 )

Parameter Fed Fasted SED/SER sig

Mean GH concentration 2.8 3.8 1.10 ** (geometric mean, ng/ml)

Basal GH concentration (ng/ml) 2.5 2.8 0.24 ns GH pulse frequency (pulses/6 hr) 4.0 4.6 0.43 ns GH pulse amplitude 3.4 4.6 1.11 **

(geometric mean, ng/ml)

Summary statistics for the GH secretion parameters, classified by GH genotype and selection line, are given in Tables 3 and 4. All GH profile parameters differed significantly with selection line, generally with higher values for the lean line, lower values for the fat line, and intermediate values for the control line. Geometric mean GH level and pulse amplitude were higher (P < 0.01) by about 35% during fasting compared with those of the ad libitum fed lambs (Table 4). There were no significant interactions among selection line, GH genotype, and nutrition treatment for any of the variables investigated.

G H R H Challenge. The shape parameter for the critical exponential curve showed no evidence of significant variation with GH genotype or selection line, for either challenge, in either the fed or fasted state. Therefore, the response to the GHRH challenge can be described adequately through the maximum GH response, which did vary (Table 5), with significant GH genotype by selection line interactions for both challenges and significant GH genotype main effect for the difference between challenges. The maximum GH response did not differ significantly between the GH genotypes for the Fat and the Control selection lines, whether fed or fasted. However, for the ad libitum fed Lean selection line animals, the first GHRH challenge resulted in significantly higher mean maximal response for GH1/GH1 than for GH2/GH2 animals (Figure 3). The GH1/GH2 animals were intermediate. In contrast, in the fasted state, higher responses to both challenges were observed for the GH2/GH2 genotype than for the GH1/GH1 genotype, with GH1/GH2

animals intermediate (Figure 3). This contrast was significant for the second challenge.

TABLE 5. MEAN MAXIMUM POSTCHALLENGE GH CONCENTRATION (NG/ML) FOR FIRST AND SECOND GHRH CHALLENGE IN FED AND FASTED STATE, CLASSIFIED BY GH GENOTYPE AND SELECTION LINE.

GH Genotype

Selection line GHI/GH1 GH1/GH2 GH2/GH2 SED sig

Challenge l--Fed Fat 33.1 50.4 41.2 Control 33.0 29.8 68.8 19.5 * Lean 94.5 51.9 25.6

Challenge 2--Fed Fat 43.6 39.7 42.2 Control 26.9 41.0 71.6 15.8 ns Lean 5 t.9 52.8 43.4

Challenge 1 Fasted Fat 49.7 28.3 16.0 Control 37.0 31.5 29.5 19.5 ns Lean 49.1 50.9 72.9

Challenge 2--Fasted Fat 19.9 20.9 18.7 Control 24.9 37.5 14.3 15.8 * Lean 57.3 80.4 122.5

SEDs refer to comparisons between Fat and Lean selection lines and the significance of GH genotype by selection line interaction (ns = not significant; * = P < 0.05) is given.

GH GENE DUPLICATION AND GH RELEASE IN LAMBS 387

100

8O

E 60 ¢ -

--r 4o

2o

0

100

8o

~ 6o & 212 40

2o

0

Fed

~ \ Fasted

li". \

l i '.. \ ,7 x. ' , . 0 20 40 60 80 100 120 140 160 180

Time from First Challenge (rain)

Figure 3. Mean GH concentration after two successive GHRH challenges for Lean selection line animals for each GH genotype in (a) the fed state and (b) the fasted state; G H I / G H I ( ); G H I / G H 2 ( . . . . . . . . . . . . ); G H 2 / G H 2 ( . . . . . . ).

Averaged over other treatments, maximum GH response decreased between challenges by 12.7 ng/ml for the GH1/GH1 genotype and increased by 7.8 (SED 6.9) ng/ml for the GH2/GH2 genotype, with an intermediate of 0.9 ng/ml for GH1/GH2 animals.

DISCUSSION

The objective of this research was to investigate the effects of natural variation in the number of copies of the GH gene in sheep on growth parameters, plasma GH profiles, and the response to GHRH challenge. There were interesting subtle effects of GH gene copy number on GH secretion. When response to the first GHRH injection was compared with the response to the second challenge given 1.5 hr later, lambs of the GH1/GH1 genotype had a higher response to the first challenge, while lambs of the GH2/GH2 genotype had a higher response to the second challenge. Heterozygous (GH1/GH2) lambs were inter- mediate. The effects of GHRH challenge were also influenced by the number of copies of the GH gene, together with selection for body composition and nutritional state. The highest response to the GHRH challenge occurred in Lean lambs and this response varied significantly with GH genotype and nutrition treatment. The effects of the number of copies of the GH gene on growth rate, body composition, or GH secretion profiles were not significant. However, Lean lambs from the GH1/GH1 genotype were heaviest at weaning and 6 mos of age and had the highest GH pulse amplitudes.

The possible effects of the polymorphism at the oGH locus were compared within sire groups, in progeny of heterozygous GH1/GH2 Coopworth rams out of selection lines known to differ markedly in body composition and GH secretion (3). In most sheep breeds, the frequency of the GH1 allele is low, making it difficult to obtain GH1/GH1 lambs for a comparative study. We took advantage of the moderate frequency of the GH1

388 GOOTWINE ET AL.

allele in the Coopworth breed and chose families from GH1/GH2 rams with progeny of all three genotypes.

The lambs examined here were from only one or two sires per selection line. Thus, it has to be taken into consideration that sire and selection line effects are confounded in the preset study. Nevertheless, the effects of selection line on growth rate, body composition, and GH secretion patterns were similar to results from all the lambs born in the selection line in that year and to previously published results reporting higher birth weight and weaning weight, and lower backfat depths in lambs from the Lean selection line (3). The GH profiles were similar to patterns published previously for lambs from these selection lines (3,4). GH release in fed and fasted lambs also agreed with previous reports on the effects of fasting in lambs (17,18). Taken together these results demonstrate that the GH duplication influenced the response to GHRH, but had little effect on body composition and plasma profiles of GH, particularly when compared with the effects of selection in the Lean and Fat lines.

Fed Lean lambs that lacked the duplication (GH1/GH1) had the highest GH response at the first GHRH challenge. Lean ram lambs have significantly higher GH pulse amplitude and frequency (3), and higher responses to GHRH (4) than ram lambs from the fat line. Ram lambs from the Lean line also have larger pituitary glands (independent of body weight) and a higher pituitary GH content (19). Similar increases in pituitary weight and serum GH concentrations were recorded in pigs selected for reduced backfat (20). Thus, it seems that any effects of variation in GH gene copy number on increased GH response may only be apparent when pituitary GH content and GH release are high, such as in the Lean selection line. In support of this suggestion, the pituitary content of GH was significantly lower in animals from the Lean selection line that were homozygous for the duplication (GH2/GH2) compared with the other genotypes (19).

In contrast, responses to GHRH challenge in fed lambs were low and similar for all GH genotypes in the Fat line, whereas the highest responses in the Control line were recorded in GH2/GH2 animals. In the fasted state, GH response to exogenous GHRH was higher in Lean animals homozygous for the duplication (GH2/GH2) at both the first and second challenge, with a higher response at the second challenge. Fasting increased GH pulse frequency and amplitude. Effects of selection on response to GHRH challenge, but not in basal GH secretion parameters has been reported previously (21,22). In a study with calves selected for increased milk fat or milk fat plus protein yield, basal GH concentra- tion was unaffected by the selection, but GH release after administration of GHRH or thyrotropin-releasing hormone was found to increase (21).

The reasons for variation in responses to GHRH between the GH genotypes in the present experiment are not known. It can be suggested that the presence of the GH2-Z gene copy could influence transcription or GH mRNA stability. However, there is no evidence so far that the presence of GH duplication effects in growing lambs the size or amount of pituitary GH mRNA measured by northern blotting (19). In addition, a recent study using a polymorphism to differentiate between the mRNA of the two gene copies (9), demonstrated that the GH2-Z gene copy was not expressed in detectable levels in the pituitary of a single nonlactating Awassi ewe, homozygous for the GH duplication. Yet, the possibility that presence of the GH2-Z gene copy could influence transcription or GH mRNA stability in physiological states such as fasting (as in the present experiment) or during lactation have to be considered. The differences observed may also result from effects of the duplication on the pituitary gland during development.

If the GH2-Z copy of the GH gene is not transcribed or translated in the pituitary gland through life, the effects observed must related to an interaction between the two GH gene copies at the level of transcription of the GIt2-N copy of the gene. Indeed, there was

GH GENE DUPLICATION AND GH RELEASE IN LAMBS 389

e v i d e n c e for an e f fec t o f the G H gene dup l i ca t i on on G H m R N A c o n c e n t r a t i o n s in r ams

f r o m the L e a n se lec t ion l ine (19). T h e p r e s e n c e o f the dup l i ca t ed gene appea red to

dec rea se the c o n c e n t r a t i o n o f the m R N A , c o n s i s t e n t w i th the p re sen t resul ts . D N A b i n d i n g

d o m a i n s in the p r o m o t o r o f the G H 2 - Z copy m a y e f fec t the r e s p o n s e to G H R H and /o r

o the r e f fec tors o f G H t r ansc r ip t ion by c o m p e t i n g for t r ansc r ip t ion factors.

T h e h i g h f r e q u e n c y o f the G H 2 al le le in m a n y sheep b reeds (5) sugges t s tha t the G H

gene dup l i ca t i on m a y c o n f e r a se lec t ive a d v a n t a g e u n d e r ce r ta in c i r cums tances . The

c o n c l u s i o n of our s tudy is tha t G H gene copy n u m b e r in sheep does no t h a v e a m a j o r

e f fec t on basa l G H secre t ion pat terns , g r o w t h rate, or fa t depos i t i on in l ambs . The

d i f f e rences tha t were f o u n d b e t w e e n G H g e n o t y p e s in r e sponse to G H R H cha l l enge ,

w h i c h were se lec t ion l ine and feed ing s ta tus dependen t , sugges t tha t the p o l y m o r p h i s m in

the n u m b e r o f G H gene cop ies in sheep has phys io log i ca l imp l i ca t i ons for the func t ion o f

the G H axis. D i f f e r ences in p i tu i ta ry r e s p o n s i v e n e s s to G H R H m a y be m a n i f e s t e d in

g r o w i n g l a m b s on ly u n d e r speci f ic g e n o t y p e - e n v i r o n m e n t c o m b i n a t i o n s .

A C K N O W L E D G M E N T S / F O O T N O T E

We thank W.E. Bain, G.J. Greer and A. Findlay for stock management and collection of production data, Dr. K.G. Dodds for statistical advice, and Dr J.S. Fleming for comments on the manuscript.

Address correspondence to Dr. E. Gootwine, Institute of Animal Science, ARO, The Volcani Center, POB 6, Bet Dagan 50250, Israel.

REFERENCES

1. Murray JD, Oberbuer AM. Growth hormone manipulation and growth promotants in sheep. In: Progress in Sheep and Goat Research, Speedy AW, (ed). C.A.B. International, pp. 217-234, 1992.

2. Beermann DH, DeVol DL. Effect of somatotropin, somatotropin releasing factor and somatostatin on growth. In: Growth Regulation in Farm Animals. Pearson AM and Dutson TR (eds). Elsevier Applied Science, London, pp. 373-426, 1991.

3. Suttie JM, Veenvliet BA, Littlejohn RP, Gluckman PD, Corson ID, Fennessy PF. Growth hormone pulsatility in ram lambs of genotypes selected for fatness or leanness. Anim. Prod 57:119-125, 1993.

4. Suttie JM, Lord EA, Gluckmann PD, Fennessy PF, Littlejohn RP. Genetically lean and fat sheep differ in their growth hormone response to growth hormone-releasing factor. Domest Anim Endocrinol 8:323-329, 1991.

5. Valinsky A, Shani M, Gootwine E. Restriction fragment length polymorphism in sheep at the growth hormone locus is the result of variation in gene number. Anim Biotechnol 1:135-144, 1990.

6. Gootwine E, Sise JA, Penty JM, Montgomery GW. The duplicated gene copy of the ovine GH gene contains a PvuII polymorphism in the second intron. Anim Gen 24:319-321, 1993.

7. Byrne CR, Wilson BW, Ward K. The isolation and characterization of the Ovine growth hormone gene. Aust J Biol Sci 40:459-468, 1987.

8. Ofir R, Gootwine E, sequence analysis of the GH1, GH2-N and GH2-Z gene copies of the growth hormone locus. Proc. Int. Soc. Anim Genet, Tours, France, Abs. D034, 1996.

9. Gootwine E, Ofir R, Yossefi S. Characterization of PvuII polymorphisms between the ovine growth hormone GH2-N and GH2-Z gene copies. Anim Biotechnol 7:135-143, 1996.

10. Morris CA, McEwan JC, Fennessy PF, Bain WE, Greer GJ, Hickey SM. Selection for high or low backfat depth in Coopworth sheep: Juvenile traits. Anim Sci 65:93-103, 1997.

11. McEwan JC, Clarke J, Knowler M, Wheeler M. Ultrasonic fat depths in Romney lambs and hoggets from lines selected for different production traits. Proc NZ Soc Anim Prod 49:113-119, 1989.

12. Montgomery GW, Sise JA. Extraction of DNA from sheep white blood cells. NZ J Agric Res 33:437-441, 1990.

13. Yato M, Yamano Y, Oyabayashi K, Okuno M, Kioka N, Manaba E, Hashi H, Sakai H, Komano T, Utsumi K, Iritani A. Nucleotide sequence of the growth hormone gene cDNA from goat Capra hircus L. (Tokara). Nucl Acids Res 16:3574, 1988.

14. Genstat, Genstat Reference Manual. Clarendon Press, Oxford, 1987. 15. Merriam GR, Wachter KW. Algorithms for the study of episodic hormone secretion. Am J Physiol

243:E310-E318, 1982. 16. Butler RC and Brain P. Nonlinear contrasts in ANOVA. Genstat Newsletter 29, 1993. 17. Thomas GB, Mercer JE, Karalis T, Rao A, Cummins JT, Clarke IJ. Effect of restricted feeding on the

concentration of growth hormone (GH), gonadotropins, and prolactin (PRL) in plasma, and on the amount

390 GOOTWlNE ET AL.

of Messenger Ribonucleic Acid for GH, gonadotropin subunits, and PRL in the pituitary glands of adult ovariectomized ewes. Endocrinology 126:1361-1367, 1990.

18. Foster DL, Ebling FJP, Micka AF, Vannerson LA, Bucholtz DC, Wood RI, Suttie JM, Fenner DE. Metabolic interfaces between growth and reproduction. I. Nutrition modulation of gonadotropin, prolactin, and growth hormone secretion in the growth-limited female lamb. Endocrinology 125:342-350, 1989.

19. Fleming JS, Suttie JM, Montgomery GW, Gunn J, Stuart SK, Littlejohn RP, Gootwine E. The effects of a duplication in the ovine growth hormone (GH) gene on GH expression in the pituitaries of ram lambs from lean and fat selected sheep lines. Domest Anim Endocrinol 14:17-24, 1997.

20. Althen TG, Gerrits RJ. Pituitary and serum growth hormone levels in Duroc and Yorkshire swine genetically selected for high and low backfat. J Anim Sci 42:1490-1497, 1976.

21. Woolliams JA, Angus KD, Wilson SB. Endogenous pulsing and stimulated release of growth hormone in dairy calves of high and low genetic merit. Anim Prod 56:1-8, 1993.

22. Lovendahl P, Sejrsen K. Growth hormone release in calves selected for milk fat yield. Anim Prod 56:285-291, 1993.