ELSEVIER

GROWTH HORMONE PRIMING AS AN ADJUNCT TREATMENT IN SUPEROVULATORY PROTOCOLS IN THE EWE ALTERS FOLLICLE DEVELOPMENT BUT HAS NO EFFECT

ON OVULATION RATE

I.M. Joyce 1, M. Khalid 2 and W. Haresign 2a

1 School of Biological Sciences, University of Nottingham, Sutton Bonington, Leics, UK

2Welsh Institute of Rural Studies, University of Wales, Aberystwyth, Ceredigion, SY23 3AL, UK

Received for publication: April 20, 1998 Accepted: July 20, 1998

ABSTRACT

This study investigated the effect of FSH alone and rGH priming followed by FSH treatment on follicle populations, follicular fluid concentrations of components of the IGF system and steroids, and the ovulation rate in sheep. Estrus was synchronized with progastagen sponges. Ewes (n = 10/group) in Group 1 served as tmtreated controls, while those in Groups 2 to 5 received a standard superovulatory treatment of 1.1 mg im oFSH twice daily for 4 d. In addition, ewes in Groups 3 and 5 were administered rGH (15 mg/d, ira) for the 7 d prior to FSH treatment. Groups 1, 2 and 3 were sacrificed just prior to the LH surge; Groups 4 and 5 were allowed to ovulate. Daily plasma samples were collected to monitor GH, IGF-1 and insulin levels. All follicles ->1.0 mm from Groups 1, 2 and 3 were counted, and follicular fluid f~m follicles >9..5 mm was assayed for estradiol, testosterone, IGF-1 and IGFBPs. Compared with the control, treatment with rGH+FSH but not FSH alone increased (P<0.001) plasma concentrations of GH, IGF-1 and insulin. The mean number of large- (_>4.5 ram) and medium-sized (2.5 to 4.0 mm) follicles was increased (P<0.01), and the mean number of small (~2.0 mm) follicles was decreased (P<0.001) by FSH treatment. The mean number of medium-sized (2.5 to 4.0 mm) foUicles was further increased (P<0.05) by rGH priming. Estradiol concentration in medium but not in large estrogenic foUicles was increased (P<0.05) by rGH priming, whereas testosterone concentration in estrogenic follicles was not altered. Components of the IGF system in medium-sized estrogenic follicles were similar in all treatment groups; however, in large estrogenic follicles rGH increased IGF-1 concentrations (P<0.05) and intensity of the 44-42 kDa IGFBP band (P<0.01). Priming with rGH did not alter superovulatory responses. These results show that rGH priming, when used as an adjunct to FSH treatment in ewes, alters components of the IGF system in large estrogenic follicles and increases the number and physiological maturity of medium-sized follicles in the ovary; it does not however alter ovulation rate responses. © 1998 by Elsevier Science Inc.

Key words: sheep, superovulation, growth hormone, IGF-1, BST

Acknowledgments: The authors thank the Ministry of Agriculture, Fisheries and Food for financial support; Elanco Animal Health Ltd., UK, for supplies of rGH; Intervet, UK, for progestagen sponges; NIADDK for oGH and oLH antisera; and Mr. M.R.P. Mitchell for technical support. ~Correspondence and reprint requests: Tel: 01970 621644; E-mall: wih@aber.ac.uk

Theriogenology 50:873-884, 1998 0093-691X/98/$19.00 © 1998 by Elsevier Science Inc. PII S0093-691X(98)00192-7

874 Theriogenology

INTRODUCTION

Between-animal variation in superovulatory responses is a major factor limiting the uptake of embryo transfer within the sheep industry. Despite considerable research effort, it is still unclear why this variability exists and how it can be reduced, although an approach that has recently shown promise in cattle is the use of recombinant bovine GH (rGH) as an adjunct treatment in superovulatory treatment protoenls (8, 10, 16). The success of this technique in cattle may well reside in the ability of rGH treatment to increase the ovarian population of small antral follicles (6, 7). Despite evidence that rGH treatment can also alter follicle populations in the ewe (9), it has not yet been possible to translate this into improved superovulatory responses (3).

Changes in follicle growth due to rGH treatment may well be modulated via alterations to the intra-follicular insulin-like growth factor (IGF) system, which is believed to play a significant role during folliculogenesis (for review, see 22). Growth hormone is known to be a potent stimulator of IGF-1 production, and changes in the intra-follicular IGF system following rGH treatment have been reported in a number of species (13, 21). In addition to the effects of GH on IGF- 1 production, divergent IGFBP profiles in estrogenic and nonestrogenic follicles (15, 18) suggest that gonadotrophins may also regulate a number of components of the IGF system, an assertion which receives support from in vitro studies (11, 12). Regulation of the IGF system in follicles is therefore not simple, while subjecting the ovary to exogenous gonadotrophins adds a further layer of complexity. There is a need to improve our understanding of these complicated interactions if we are to manipulate the IGF system to advantage in superovulatory protocols.

The objectives of this study were 1) to examine the effect of FSH alone and rGH priming followed by FSH treatment on follicle growth and development, with particular regard to changes in the IGF system, follicle populations and steroid concentrations, and 2) to investigate the effect of these treatments on the ovulation rate of ewes.

MATERIALS AND METHODS

Anil~S

Mature cyclic Scottish Blackface ewes (n = 50) were housed indoors under conditions of natural daylength and temperature, and were fed a maintenance ration of hay and dried grass nuts, with water available ad libitum.

Experimental Procedure

Estrus was synchronized in all ewes using 30 nag intravaginal progestagen sponges (Chronogest, lntervet, Cambridge, UK) left in situ for 14 d. The animals were allocated at random to 5 treatment groups (n = 10/group). Ewes in Group 1 were untreated controls, while those in Groups 2 to 5 received a standard superovulatory treatment of twice-daily intramuscular injections ofoFSH (Ovagen, Immuno-cbemical Products Ltd., Auckland, NZ; 1.1 rag/injection) for 4 d, starting 3.5 d prior to progestagen sponge removal. In addition, ewes in Groups 3 and 5 received rGH priming (Somidobove, Elanco Animal Health Ltd., Basingstoke, UK; 15 mg/d, ira) over a 7-d period, administered once daily and terminated on the first day of FSH treatment. Based on preliminary

Theriogenology 875

observations of the timing of the pre-ovulatory LH surge in both the control and superovulated animals, Group 1, 2 and 3 ewes were slaughtered just prior to the LH surge (Group 1, 35 h after sponge removal; Groups 2 and 3, 20 h after sponge removal). Groups 4 and 5 were allowed to ovulate. Dally blood samples were collected by jugular venipuncture l~om ewes in Groups 1, 2 and 3 over the rGH and FSH treatment phases to monitor plasma GH, IGF-1 and insulin. In addition, blood samples were collected every 2 h via an indwelling jugular vein catheter from Groups 1, 2 and 3 for the 16 h prior to slaughter, to confirm that animals had been sacrificed before the onset of the preovuiatory LH surge. Immediately after slaughter, all follicles >1.0 mm were dissected from each ovary. Follicular fluid was individually collected fTom all follicles >_2.5 nun and divided into two portions of at least 30 pL. In the case of smaller follicles, in which the volmne of follicular fluid was inadequate to complete all the assays, follicular fluid from 2 to 3 follicles of a similar size and category of animal (preferably from the same animal) was pooled. One portion of each follicular fluid sample was stored at -20°C for the estimation of IGF-1 and IGFBPs; the other portion was diluted with 450 pL of Hank's balanced salt solution (Sigma, Dorset, UK) before storage at -20°C and subsequent assay for estradiol and testosterone. Ovulation rate in Groups 4 and 5 was determined by laparoscopy 6 d after sponge removal.

AssayMethods

Measurements of estradiol concentrations in follicular fluid were carded out without prior extraction (5). The limit of sensitivity of the assay within the present study was 21 pg/mL, and the inter- and intra-assay coefficients of variation were 13.7 and 13.5%, respectively. The assay used for the measurement of testosterone in follicular fluid was carried out without prior extraction (15, 20). The limit of sensitivity of the assay within the present study was 88 pg/mL, and the inter- and intra- assay coefficients of variation were 9.0 and 9.8%, respectively.

Plasma and follicular fluid IGF- 1 concentratiom were measured using a modified version of the RIA reported previously (15). Within the present study, samples were extracted using a two-step approach. The first step, an acid-ethanol extraction was used to remove a high proportion of the IGFBPs; the second step, a competitive exclusion method (1), was used to complete the separation of IGF-1 from IGFBPs by blocking IGFBP-binding sites with an excess of non-measured IGF (IGF- 2). The concentration of IGF-2 that was used had previously been shown not to interfere in the assay (Figure 1), and yet was 10 to 30 times in excess of IGF-1 concentrations in the samples. This competitive exclusion modification was introduced to increase the accuracy of the assay and reduce variation in cold recoveries, especially with plasma samples. Briefly, acid-ethanol mixture was added to the sample at a 4:1 ratio, incubated for 15 rain and centrifuged at 2100 g and 4°C. The supematant was then decanted, mixed with 30 ng of IGF-2, neutralized with 0.855M Tris buffer, and incubated for 30 mill prior to assay. Using this approach, serial dilutions of extracted plasma fi'om untreated and rGH-treated ewes and of extracted follicular fluid from large (>3.0 nun) and small (<3.0 mm) follicles showed parallelism with the IGF-1 standard curve. Following the addition of 360 and 120 ng/mL of cold IGF- 1 to samples of plasma from untreated and rGH-treated ewes and of follicular fluid from large (>3.0 mm) and small (<3.0 mm) follicles, recovery of IGF-1 was high (mean, 94%) and variability of recovery was low (between 85 and 102%). The limit of sensitivity of the assay was 1.24 ng/mL, and the inter- and intra-assay coefficients of variation were 10.0 and 11.8%, respectively.

876 Theriogenology

1 0 0 ~ • e • _ . ~ . ~ , ,

80 •

60

40

2o] - - '1. .t

o / ~ ~ f ,

1 10 100 1000

ng per tube

Figure 1. Cross-reaction oflGF-2 with the IGF-1 first antibody. The percentage binding of serial dilutions of IGF-I~I) and IGF-2(@) are shown. At 50% binding the specificity of the IGF-1 first antibody for IGF-2 was 0.68%. The 30 ng/tube of IGF-2 used for the competitive exclusion process was below the concentration of IGF-2 that resulted in significant displacement of 125I-IGF-1 from the first antibody.

Insulin concentrations in plasma were measured using an RIA kit (Lifescreen Ltd., Watford, UK). The limit of sensitivity of the assay was 0.66 IU/mL, and the inter- and intra-assay coefficients of variation were 6.3 and 6.1%, respectively. Plasma GH levels were measured by RIA, using oGH for iodination and standards, and oGH antisera supplied by NIADDK. Within the assay, rGH (Somidobove) and oGH standards were super-imposable, and the assay showed parallelism between the standard curve and serial dilutions of sheep plasma. The limit of sensitivity of the assay was 1.6 ng/mL, and the inter- and intra-assay coefficients of variation were 7.8 and 10.7%, respectively. Measurements of LH concentrations in plasma were carried out without prior extraction (17). All samples were run in a single assay. The limit of sensitivity of the assay within the present study was 0.42 ng/mL, and the intra-assay coefficient of variation was 7.6%.

The IGFBPs were identified in follicular fluid using SDS-page/Westem fig, and blotting following a method previously used in this laboratory (15). Band intensity was quantified by scanning densitometry, and variability between different autoradiographs was ameliorated mathematic, ally using a standard sample of follicular fluid on each run.

Statistical Analysis

At dissection, follicles were classified as small (~2.0 ram), medium (2.5 mm to 4.0 mm) or large (>4.5 ram), and this classification was used to study changes in follicle populations.

Theriogenology 877

For analysis of follicular fluid hormone concentrations, follicles were first classified as estrogenic or non-estrogenic. Classification of follicles in terms of their estrogenicity has previously been undertaken in our group by plotting follicle numbers against estradiol concentration (15). In untreated animals this results in a bimodal distribution, with few follicles having intermediate estradiol concentrations. Plotting follicle numbers against estradiol:testosterone ratio instead of estradiol concentration within the present study had no effect on the classification of individual follicles but reduced the number of follicles in the intermediate region between the 2 subpopulations, allowing for clearer identification of a threshold value for classification. The bimodal distribution of follicles fi'om untreated controls when follicle numbers were plotted against either estradiol or estradiol:testosterone ratio was not apparent in the FSH-treated animals, with large numbers of follicles having intermediate estradiol concentrations, and very few with either low (typical of non-estrogenic follicles) or high (typical of estrogenic follicles fxom untreated animals) estradiol concentrations. Therefore, the characterization of estrogenic and non-estrogenic follicle subpopulations was undertaken using the threshold values established in the control group. On this basis, follicles were classified as estrogenic if they had an estradiol:testosterone ratio >5.0, and non- estrogenic if they had an estradiol to testosterone ratio <5.0.

Analysis of follicular fluid data was undertaken using each animal as a degree of freedom; for animals with more than 1 observation in each category the mean of these observations was used. There were too few non-estrogenic follicles in the FSH-treated groups to allow for further analysis of this category; in addition, insufficient medium-sized estrogenic follicles in the control group meant that 2 ANOVA analyses were needed. Analysis 1 compared large follicles for all treatment groups while Analysis 2 compared medium and large follicles for the FSH alone treatment and for the rGH+FSH treatment group. Statistical differences between treatment groups for the comparison of plasma hormone concentrations were also analyzed by ANOVA. Ovulation rates in Groups 4 and 5 were compared using a t-test.

RESULTS

Peripheral Plasma Hormone Concentmtiom

The IGF-1 concentrations in plasma (Figure 2a) significantly increased (1)<0.001) within 24 h of the start of rGH priming. Despite declining following the end of treatment, concentrations still remained significantly higher than pretreatment values at slaughter. By contrast, control ewes and those treated with FSH alone maintained basal peripheral plasma IGF-1 profiles for the duration of the experiment. Circulating insulin profiles (Figure 2b) showed a similar pattern to IGF-1, with significantly higher (P<0.001) concentrations in the rGH+FSH animals during the period of rGH treatment. In contrast to IGF-1, insulin concentrations fell to pretreatment levels within 48 h of the end of rGH treatment, and were not different fTom pretreatment values at the time of slaughter. There was no difference in insulin concentrations between control animals and those treated with FSH alone for the duration of the experiment. Plasma GH concentrations also showed a similar pattern to IGF-1, with a significant (P<0.001) increase in concentrations in the rGH+FSH group during rGH treatment (pretreatment: 9.6 ng/mL; during treatment: 26.9 ng/mL). As with plasma insulin, concentrations had declined to pretreatment values by the time of slaughter. Control animals and those treated with FSH alone had similar plasma GH profiles

878 Theriogenology

for the duration of the experiment (6.8 and 7.5 ng/mL, respectively). All animals had typicai (<5.0 ng/mL) presurge follicular phase LH concentrations for the 16 h prior to slaughter.

(a) 350

300

250

200

150

L~ 100

50

0

FSH rGH ~ ~

1 ! 1 ! I ! [ ! ! I !

(b)

° ~

30

25

20

15

10

5

0

1 SED

f-

I I I I ! I I I I I I

3 4 5 6 7 8 9 10 11 12 13 14

Days from sponge insertion

Figure 2. Mean circulating IGF-1 (Figure 2a) and insulin (Figure 2b) concentrations in control (0), FSH alone (A) and rGH+FSH-treated groups (Ig). All ewes were synchronized with progestagen sponges inserted on Day 0 and removed on Day 13. Both the FSH alone and rGH+FSH-treated groups received a standard superovuiatory treatment of oFSH on Days 11 to 14. Daily rGH injections were administered to the rGH+FSH- treated group on Days 5 to 11.

Follicular Fluid Hormone Concentrations

The data comparing follicular fluid hormone concentrations in large estrogenic follicles in the 3 treatment groups (Analysis 1) are presented in Table 1, and the data comparing follicular fluid hormone concentrations in medium- and large-sized estrogenic follicles in the FSH alone and the GH+FSH treatment groups (Analysis 2) are presented in Table 2.

Theriogenology 879

Table 1. Analysis 1 : follicular fluid hormone and IGFBP concentrations in large estrogenic follicles in the Control, FSH alone and rGH+FSH-treated groups

Control FSH alone rGH+ FSH SED

Estradiol (ng/mL) 175.32 a 47.11 b 43.29 b 26.1

Testosterone (ng/mL) 8.91 a 2.22 b 2.04 b 1.81

Estradiol :testosterone ratio 26.91 a 24.76 a 24.36 a 5.19

IGF-1 (ng/mL) 141.6 a 170.4 ab 184.8 b 17.6

44-42kDa IGFBP (U/lxL) 0.97 a 1.28 a 1.66 b 0.19

35kDa IGFBP (U//aL) 0.11 a 0.21 a 0.16 a 0.10

a'bwithin rows, means with different superscript letters are significantly different (P<0.05 at least).

Table 2. Analysis 2: follicular fluid hormone and IGFBP concentrations in medium- and large-sized estrogenic follicles in the FSH alone and rGH+FSH-treated groups

FSH alone rGH+FSH Medium Large Medium Larse SED

Estradiol (ng/mL) 23.73 a 47.11 b 34.17 c 43.29 bc 5.02

Testosterone (ng/mL) 2.45 a 2.22 b 2.13 a 2.04 b 0.40

Estradiol:testosterone ratio 11.32 a 24.76 b 19.45 b 24.36 b 3.66

IGF-1 (ng/mL) 179.2 170.4 195.2 184.8 21.6

44-42kDa IGFBP CLI/gL) 1.15 a 1.28 a 1.28 a 1.66 b 0.18

35kDa IGFBP CLI/pL) 0.22 a 0.21 a 0-07 a 0.16 a 0.10

a'bwithin rows, means with different superscript letters are significantly different (P<0.05 at least).

Estradiol concentrations in large estrogenic follicles from the Control group were significantly higher (P<0.001) than in the FSH and rGH+FSH-treated groups, which did not differ. Estmdiol concentrations in meditun-sized follicles were significantly higher (P<0.05) in the rGH+FSH-treated group than the FSH alone group. Testosterone concentrations in the follicular fluid of large estrogenic follicles were significantly higher (P<0.001) in the control group compared to the 2 FSH- treated groups, with no difference between FSH alone and rGH+FSH groups in either large or medium-sized follicles.

IGF-1 concentrations in large estrogenic follicles were higher (P<0.05) in the rGH+FSH group but not in the FSH alone group compared with the control, while IGF-1 concentratiom in medium- sized estrogenic follicles were similar in the FSH alone and rGH+FSH-treated groups. Western ligand blot analysis of IGFBPs in follicular fluid from large estrogenic follicles revealed 2 bands,

880 Theriogenology

one at 44-42 kDa (presumed IGFBP3) and the other at 35 kDa (presumed IGFBP2). While the 44- 42 kDa band was clearly present in all samples, the band at 35 kDa was much weaker, being undetectable in many follicles. Mean intensity of the 44-42 kDa band in large estrogenic follicles was significantly higher (P<0.05) in the rGH+FSH group compared with both the controls and the FSH alone group. However, there was no difference in 44-~.2 kDa band intensity in medium-sized follicles between the FSH alone and rGH+FSH-treated groups. Mean intensity of the 35 kDa band did not differ across treatments in either size category of follicles.

30

o := 20 ~o

0

,.0

10 Z

b b

<2.0 2.5-4.0 >4.5 Follicle size (mm)

I SED

g g

Figure 3. Follicle numbers in three size categories in Control ~ ) , rGH+FSH ([~) and FSH alone

( ~ ) groups, a'bwithin follicle size categories, columns with different superscripts vary significantly (P<0.05, at least).

Follicle Populations and Ovulation Rate

Priming with rGH had no effect on superovulatory responses (meaniSEM: FSH alone, 18.0+_.3.06; rGH+FSH, 18.0~.63). In terms of follicle populations there were two treatment effects (Figure 3). First, there was a significant difference between the follicle populations of the control group and the two FSH-treated groups, with the FSH-treatedgroups having fewer (P<O.001) small, and more medium- (P<0.05) and large-sized (P<0.01) follicles than the control ~nim~ls. Second, there was an effect of rGH priming, with significantly more (P<0.05) medium-sized follicles in the rGH+FSH group compared to those animals treated with FSH alone. Although no statistical analysis was possible, comparison of the follicle population data l~om the FSH alone group slaughtered just prior to the LH surge, with the ovulation rate data from the FSH alone group that was allowed to ovulate revealed that the mean, and variability about the mean, of follicles >4.0 mm in the ovary just

Theriogenology 881

prior to the LH surge, was more closely matched with the mean ovulation rate and its variability than other follicle categories (Table 3). This was also the case in the rGH+FSH treated group.

Table 3. Ovulation rate and the number of follicles >3.0, 4.0, 4.5 and 5.0 mm (+SEM) in the FSH alone and rGH+FSH-treated groups

FSH alone rGH+FSH

Ovulation rate 18.0-J=3.0 18.0+_2.6

Follicles >3.0 mm 20.7Y.2.5 25.2_+_2.5

Follicles >4.0 mm 17.2=1:2.5 19.9_+.2.4

Follicles >5.0 mm 10.4+ 1.9 8.0+ 1.7

DISCUSSION

The objectives of this study were twofold: 1) to investigate the effect of rGH priming on ovulation rate in sheep superovulated with oFSH, and, 2) to compare the effect of FSH alone and rGH priming followed by FSH treatment on follicle growth and development.

In order to undertake the second of these objectives it was necessary to compare follicles at similar stages of development and of a similar health status. Preliminary experiments confirmed the work of other authors (14) in finding that the interval between sponge removal and the LH surge is shorter in superovulated animals compared to untreated controls. To harvest follicles in the control and FSH-treated groups just prior to the LH surge (and therefore with near-maximal estradiol concentrations in potentially ovulatory follicles), animals in the FSH alone and rGH+FSH groups were sacrificed correspondingly sooner after sponge removal than animals in the control group. Plasma LH concentrations in samples taken every 2 h for the 16 h prior to slaughter confirmed that none of these animals had undergone the LH surge.

In terms of the health status of the follicles that were collected, a high level of estradiol production is widely accepted to indicate that a follicle is healthy. Because of the atypical estradiol concentrations observed in the FSH-treated groups it was necessary to refine the estrogenic/non-estrogenic classification procedure used previously in our group (15) by including a measure of follicular androgen status (testosterone) in the criteria. A number of factors suggest that the use of the estradiol:testostemne ratio (and the threshold criteria chosen) to determine the health status of follicles in this experiment was valid. First, in the control group, no follicles that were classified as estrogenic on the basis of their estradiol:testosterone ratio would have been classified as non-estrogenic using criteria previously reported by our group (15). Second, the high proportion of follicles classified as estrogenic in the FSH-treated groups is consistent with the high ovulation rate observed in those animals that were allowed to ovulate. Third, the estradiol:testosterone ratio of follicles classified as estrogenic did not differ across the 3 treatment groups despite the differing estradiol concentrations, suggesting that the follicles classified as

882 Theriogenology

estrogenic in the FSH-treated groups did indeed have an active aromatase enzyme system. Fourth, the 35 kDa IGFBP (presumed IGFBP2) band intensity in the follicles classified as estrogenic in the 3 treatment groups did not differ. Since IGFBP2 is known to be altered in non-estrogenic follicles compared with estrogenic follicles (15, 18), it would be expected that if the estradiol:testosterone ratio criterion were incorrect then IGFBP2 concentrations across the 3 treatment groups would differ.

Apart from dramatically altering the size distribution of follicles, FSH treatment resulted in significantly lower estradiol and testosterone concentrations in estrogenic follicles, compared to untreated controls, but had no effect on the estradiol:testosterone ratio of estrogenic follicles in the 2 groups. This suggests that FSH treatment reduced the absolute quantity of androgen substrate available for aromatization, but probably not the activity of the aromatase enzyme system itself. Such a reduction in androgen production might result from a comparative imbalance in the FSH and LH stimulation of follicles in ewes treated with the highly purified form of ovine FSH used in this experiment. Indeed, a gonadotrophin preparation (PMSG) with a high LH-like activity has been shown to stimulate estradiol production more than a gonadotrophin preparation with a low LH content (FSH-p) (14). There was, however, no effect of FSH treatment on components of the IGF system within the pool of estrogenic follicles, providing strong support for the concept that follicle health status (15, 18) rather than FSH stimulation levels, per se, are of over-riding importance in the regulation of follicular IGF status.

The GH treatment had 3 main effects on the follicular parameters measured: 1) an increase in the number of medium-sized follicles, 2) an increase in the physiological maturity of medium- sized follicles (as measured by estradiol concentration and estradiol:testosterone ratio); and 3) an increase in IGF-I concentrations (compared with controls) and 44-42 kDa IGFBP band intensity (compared with the control and FSH treatment alone groups) in large-sized follicles, The major effects of rGH priming on follicle development, therefore, appear to have been on medium-sized follicles, while the effects of rGH priming on the IGF system were only apparent in larger follicles. This suggests that the effects of rGH treatment on the development of medium-sized follicles are not modulated via the IGF system. Unlike IGF type 1 receptor mRNA (19) and GH receptor mRNA (4), insulin receptor mRNA has (to our knowledge) yet to be confirmed in sheep follicles. However, insulin has been implicated as an ovarian regulator in sheep (2) and shows affinity for the IGF type 1 receptor. Since circulating concentrations of both insulin and GH were increased by rGH priming in this experiment, it is possible that the effects of rGH priming on the development of medium-sized follicles were mediated by a direct effect of GH or were indirectly affected through increased insulin concentrations.

A previous study on the effect of rGH treatment alone on follicle development in ewes (9) reported that rGH increased the number of follicles between 2.1 and 3.0 nun while tending to increase the number of follicles between 3.1 and 4.0 mm, the combination of both categories being very similar to that of the medium-sized category used in this experiment. The similarity in the results between this previous study and the current experiment is perhaps difficult to reconcile with the additional FSH treatment imposed in the present study; however, it may suggest that there is a specific stage in follicle development during which follicles are responsive to the effects of rGH treatment.

Theriogenology 883

It is of interest to speculate why rGH priming does not increase superovulatory responses in sheep, while a number of studies (8, 10, 16), though not all (13), have reported that rGH priming increases superovulatory responses in cattle. Assuming that there is a specific stage in follicle development during which follicles are responsive to the effects of rGH treatment in both sheep and cattle, it seems likely that in cattle this stage of follicle development overlaps with the pool of follicles selected for ovulation in superovulatory protocols, while in sheep this is not the case. Indeed, In our present experiment, comparison of ovulation rate with follicle profiles in those animals sacrificed prior to the LH surge revealed that there was a strong similarity between ovulation rate and the number of estrogenic follicles _>4.0 mm (in terms of both the mean and the SEM) in both the FSH alone and rGH+FSH-treated animals, suggesting that follicles _>4.0 mm immediately prior to the LH surge formed the ovulatory pool. If this is the case, it is likely that there was no ovulation rate response to rGH priming despite an increase in the number and physiological maturity of medium-sized follicles because this category of follicles (2.5 to 4.0 mm) overlapped only minimally with the pool of ovulatory follicles (>_4.0 mm).

In conclusion, rGH priming, when used as an adjunct to FSH treatment in superovulatory protocols, increased the number and the physiological maturity (measured by estradiol and the estradiol:testosterone ratio) of medium-sized follicles in the ovary of the ewe but had no effect on the ovulation rate. In addition, follicles stimulated by exogenous FSH treatment had reduced estradiol and testosterone concentrations and exhibited a low incidence of atresia. Further work is needed to establish the mechanism of action of rGH treatment and to assess whether or not follicular changes brought about by rGH priming can be manipulated to advantage in superovulatory protocols in the sheep.

REFERENCES

1. Blum WF, Breier BH. Radioimmunoassays for IGFs and IGFBPs. Growth Reg 1994;4(Suppl 1):11-19.

2. Downing JA, Scaramuzzi ILl. Nutrient effects on ovulation rate, ovarian function and the secretion of gonadotrophic and metabolic hormones in sheep. J Reprod Fertil 1991;43(Suppl):209-227.

3. Eckery DC, Moeller CL, Nett TM, Sawyer HR. Recombinant bovine somatotropin does not improve superovulatory response in sheep. J Anim Sci 1994:72:2425-2430.

4. Eckery DC, Moeller CL, Nett TM, Sawyer HR. Localization and quantification of binding sites for FSH, LH, GH and IGF-1 in sheep ovarian follicles. Biol Reprod 1997;57:507-513.

5. FoxcroR GIL Shaw ILl, Hunter MG, Booth P J, Lancaster RT. Relationships between LH, FSH and prolactin secretion and ovarian follicular development in the weaned sow. Biol Reprod 1987;36:175-191.

6. Gong JG, Baxter G, Bramley TA, Webb R. Enhancement of ovarian follicle development in heifers by treatment with rBST: a dose response study. J Reprod Fertil 1997;110:91- 97.

7. Gong JG, Bramley TA, Webb R. The effect of recombinant bovine somatotropin on ovarian follicular growth and development in heifers. J Reprod Fertil 1993;97:247-254.

8. Gong JG, Bramley TA, Wilmut I, Webb R. Effect ofrBST on the superovulatory response to pregnant mare serum gonadotropin in heifers. Biol Reprod 1993;48:1141-1149.

884 Theriogenology

9. Gong JG, Campbell BK, Bramley TA, Webb R. Treatment with rBST enhances ovarian follicle development and increases the secretion of IGF-I by ovarian follicles in ewes. Anita Reprod Sci 1996;41:13-26.

10.Gong JG, Wilmut I, Bramley TA, Webb R. Pretreatment with rBST enhances the superovulatory response to FSH in heifers. Theriogenology 1996;45:611-622.

11. Grimes RW, Barber JA, Shimasaki S, Ling N, Hammond JM. Porcine ovarian granulosa ceils secrete IGFBP-4 and -5 and express their mRNAs: regulation by FSH and IGF-1. Biol Reprod 1994;50:695-701.

12. Grimes RW, Samaras SE, Barber JA, Shimasaki S, Ling N, Hammond JM. Gonadotropin and cAMP modulation oflGF binding protein production in ovarian granulosa cells. Am J Physiol 1992;262:E497-E503.

13. Herrier A, Einspanier R, Schams D, Niemann H. Effect ofrBST on follicular IGF-1 contents and the ovarian response following superovulatory treatment in dairy cows: a preliminary study. Theriogenology 1994;41:601-611.

14.Jabbour I-IN, Evans G. Ovarian and endocrine responses of Merino ewes to treatment with PMSG and/or FSH-p. Anim Reprod Sci 1991;26:93-106.

15. Khalid M, I-Iaresign W. Relationships between concentrations of ovarian steroids, IGF- 1 and IGF-binding proteins during follicular development in the ewe. Anim Reprod Sci 1996;41:119-129.

16. Kuehner LF, Rieger D, Walton JS, Zhao X, Johnson WH. Effect of a depot injection of rbST on follicular development and embryo yield in superovulated Holstein heifers. Tberiogenology 1993 ;40:1003 - 1013.

17.Mcleod BJ, Haresign W, Lamming GE. The induction of ovulation and luteal function in seasonally anoestrous ewes treated with small dose multiple injections of GnRH. J Reprod Fertil 1982;65:215-221.

18. Monget P, Monniaux D, Pisselet C, Durand P. Changes in IGF-1, IGF-2, and their binding proteins during growth and atresia of ovine ovarian follicles. Endocrinology 1993;132:1438- 1446.

19. Perks CM, Denning Kendall PA, Gilmour RS, Wathes DC. Localisation of mRNAs for IGF-1, IGF-2, and the type-1 IGF receptor in the ovine ovary throughout the estrous cycle. Endocrinology 1995; 136:5266-5273.

20. Purvis K, IUius AW, Haynes NB. Plasma testosterone concentrations in the ram. J Eodocrinol 1974;61:243-253.

21. Samaras SE, Hagen DR, Bryan KA, Mondschein JS, Canning SF, Hammond JM. Effects of GH and gonadotropin on the IGF system in the porcine ovary. Biol Reprod 1994;50:178-186.

22.Spicer LJ, Echternkamp SE. The ovarian insulin and IGF system with an emphasis on domestic animals. Dom Anim Endocrinol 1995;12:223-245.