Gonadotropin-Releasing Hormone Treatment Induces Follicular Growth and Ovulation in Seasonally Anestrous Mares'

BIOLOGY OF REPRODUCTION 36,1199-1206 (1987) Gonadotropin-Releasing Hormone Treatment Induces Follicular Growth and Ovulation in Seasonally Anestrous Mares' A. L. JOHNSON Department of Animal Sciences Rutgers, The State University of New Jersey New.Brunswick, New Jersey 08903 ABSTRACT A study was conducted to evaluate the effectiveness of gonadotropin-releasing hormone (GnRH) pulse infusion to stimulate follicular development and induce ovulation in seasonally anestrous standardbred mares. Seventeen mares were selected for use in this experiment, on the basis of a previous normal reproductive history, and were housed under a photoperiod ~lf 8L:16D beginning one week prior to the start of the experiment (second week in January). Mares were in,fused with 20 pg (n=7) or 2 pg (n=6) GnRH/h, or were subjected to photoperiod treatment only (controls, n=4). Serum concentrations of luteinizing hormone (LH), folliclestimulating hormone (FSH), and progesterone did not vary, and neither significant follicular development nor ovulation was observed in any control mare throughout the experimental period 060 days). By contrast, both groups of GnRH-treated mares showed elevated serum concentrations of LH and FSH within one day after the start of infusion. Mares infused with 20 pg GnRH/h had at least one follicle greater than or equal to 25 mm in 7.4 f 1.3 (mean?r SEM) days following the start of infusion, and ovulated in 12.0 f 0.7 days. In the 2-pg- GnRH/h treatment group, a 25-mm follicle was detected in 5.7 f 0.7 days, and ovulation occurred after 10.0? 0.3 days of infusion. Ovulation in every ;instance was followed by a functional luteal phase, as indicated by the profiles of progesterone secretion. Finally, the number of follicles ovulated per mare was related to the dose of GnRH infused, in that mares infused with1 20 pg GnRH/h had 3.0 f 0.6 ovulations per mare, whereas there were 1.2 f 0.2 ovulations per mare in the 2-pgGnRH/h-infused group (p<0.05). Results indicate that the infusion of GnRH in a pulsatile pattern can induce follicular development and ovulation in seasonally anestrous mares in the absence of photoperiodic stimulation. These data provide additional evidence that the physiological state of seasonal anestrus in the mare is the resulr of a decrease in the quantity and/orpulse frequency of hypothalamic GnRH secretion. INTRODUCTION The mare is a seasonally polyestrous breeder that - * - typically undergoes a two- to four-month period of anestrus during the winter months (Ginther, 1979). Although this decline in reproductive activity is associated with a seasonally related decrease in pituitary content (Hart et al., 1984) and serum levels (Garcia et al., 1979; Turner et al., 1979) of luteinking hormone (LH), evidence currently suggests that seasonal anestrus results from a decrease in hypothalamic gonadotropin-releasing hormone (GnRH) secretion (quantity and/or pulse frequency) and not Accepted December 12, 1986. Received October 8, 1986. 'This work was supported by a grant from the Rutgers University Research Council and the New Jersey Agricultural Experiment Station (publication D-06106-03-86). This paper was presented in part at the 19th Annual Meeting of the Society for the Study of Reproduction (Abstract 150), July 14-17, 1986, Ithaca, NY. 1199 from a change in pituitary responsiveness to GnRH. For instance, hypothalamic content (Strauss et al., 1979; Hart et al., 1984) and secretion (Sharp and Grubaugh, 1986) of GnRH are decreased during anestrus compared to the breeding season, whereas the total number of pituitary GnRH receptors are unchanged (Hart et al., 1984). Furthermore, the amount of LH released in response to relatively small doses of GnRH was found to be similar in anestrous compared to cycling mares (Alexander and Irvine, 1986). By contrast, the role of follicle-stimulating hormone (FSH) in mediating reproductive seasonality is less well understood, in that there appears to be no seasonal regulation of this gonadotropin in the mare (Garcia et al., 1979). Previous investigators have found.[hat administration of GnRH to the anestrous mare induces the release of both LH (Ginther and Wentworth, 1974; Evans and Irvine, 1976; Alexander and Irvine, 1986) and FSH (Evans and Irvine, 1976), and stimulates

1200 JOHNSON follicular development (Evans and Irvine, 1977). Recently, several groups of investigators have found that twice-daily injections of GnRH or a GnRH analog (Fitzgerald et al., 1986) or continuous administration of GnRH (Hyland et al., 1986) or a GnRH analog (Sanderson et al., 1986) can initiate follicular growth and induce ovulation in acyclic mares. We have previously reported that treatment of anestrous mares with relatively high doses of GnRH in a pulsatile pattern (50 or 250 pg GnRH/h) resulted in a preovulatory-like LH surge, and induced ovulation in an average time of about 9 days (Johnson, 1986a). The present report extends these observations by having closely monitored serum LH and FSH concentrations, follicular development, and the number of ovulations following pulsatile infusion of substantially lower doses (2 and 20 pg/h) of GnRH. MATERIALS AND METHODS Standardbred mares, ages 3 to 24 years, were fed a standardized diet (adjusted to maintain body weight) of a commercial sweet grain mixture (Agway, Bordentown, NJ) and mixed hay. Water and trace mineral salts were available ad libitum. From October to the end of the study (mid-april), the ovaries of all mares were palpated (per rectum) and examined ultrasonically, using a linear-array scanner (Technicare 2 10 DX, Pitman-Moore, Inc., Washington Crossing, NJ). Seasonal anestrus was defined by the complete absence of corpus luteum function (serum progesterone levels consistently less than 1.0 ng/ml) and the failure of ovarian follicles to develop to a diameter greater than 20 mm (the majority of follicles were 10 mm or less). Behavioral estrus was assessed using a pony stallion. True behavioral estrus (versus occasional signs of estrous behavior that were sporadically noted in anestrous mares) was indicated when a mare displayed signs of winking, squatting, and urination after exposure to the stallion for a period of 3 to 5 consecutive days. All mares were housed under a photoperiod of 8L:16D from one week prior to the start of the experiment until after an ovulation had been detected. Beginning with the pre-experimental period, mares were bled at least once a day (8 ml by jugular venipuncture) for the duration of the experiment, and serum samples were subsequently assayed for LH, FSH, and progesterone. Mares were randomly allotted to one of three treatment groups: the first group (n=7) received 20 pg GnRH/h via a chronic indwelling infusion catheter (PE 50, Intramedic, Clay-Adams, Inc., Parsippany, NJ ; Johnson, 1986a,b); the second group (n=6) was infused with 2 pg GnRH/h; the third group of mares (n=4) was subjected to the photoperiod treatment only. All other variables between GnRH-infused and control mares (i.e., blood sampling, frequency of teasing, and evaluation of ovarian characteristics) were similar. Gonadotropinreleasing hormone (Peninsula Laboratories, Belmont, CA) was infused as a single 5-second pulse in 118 pl saline per h using an Autosyringe infusion pump (Model AS 2H; Autosyringe, Inc., Hooksett, NH). In all cases, the infusion was continued until one day after ovulation had been detected. Ovulation of a follicle was determined according to the criteria outlined by Ginther (1986). The total number of induced ovulations was taken as that number of follicles ovulated within a 24-h period. Serum progesterone (Malinowski et al., 1985) and LH (Johnson, 1986c) were determined by radioimmunoassay (RIA), as previously described. Sensitivity of the LH assay was at least 0.6 ng/ml. Serum FSH was assayed, using equine FSH (LER-1686-2; from Dr. L. E. Reichert) as reference preparation (potency of 1.44 NIH-FSH-S1 units/mg), 12' I-labeled equine FSH (E 219B; from Dr. H. Papkoff), and an antihuman FSH serum (hfsh#5, from NIAMDD). Radioiodination of FSH was accomplished by using the chloramine-t method, as described by Greenwood et al. (1963). Briefly, the assay was conducted by combining 100 pl of serum or reference preparation with 100 pl 0.01 M sodium phosphate buffer (ph 7.2) and 200 pl of anti-fsh serum (diluted to 1: 18,000). After a 24-h incubation at 4"C, 100 p1 12' I-equine FSH were added, and 24 h later, 100 p1 anti-rabbit 7-globulin (1:20) were added. Separation of the bound and free 12' I-FSH was accomplished by adding 1.0 ml cold buffer and centrifuging the tubes at 1500 X g for 30 min. Parallelism to the equine FSH reference preparation was tested for various dilutions (1:1.25 X lo5 to 3.13 lo4) of a mare pituitary homogenate and serum (25 to 100111) from cycling and anestrous mares and a stallion. Follicle and ovulation data between treatment groups were analyzed by a one-way analysis of variance (ANOVA), and significant interactions partitioned by the Newman-Kuels multiple range test. Hormone data within a treatment group over time were analyzed by split-plot ANOVA (Snedecor and

OVULATION INDUCTION IN ANESTROUS MARES 1201 Cochran, 1967), while differences in peak concentrations, day of peak concentration, and length of luteal phase between treatment groups were analyzed by one-way ANOVA. RESULTS Displacement curves of equine FSH (efsh), a pituitary extract, and serum from a stallion and an anestrous and cycling mare were found to be parallel (Fig. 1). Cross-reaction of the assay with equine LH and prolactin was negligible (<O.OS%). Sensitivity of the assay (the amount of FSH that could be differentiated from zero hormone at the 95% confidence interval) was at least 8 ng/ml serum. The intra- and interassay coefficients of variation for the 5 assays reported herein were 6.2% and 14.7%, respectively. Pretreatment serum concentrations of LH and progesterone in all mares were less than 0.6 and 0.25 ng/ml, respectively, and serum FSH did not vary between treatment groups (overall pretreatment mean 27.2 f 3.4) (Tables 1, 2, and 3). No mare in any group had a follicle greater than 20 mm in diameter, and the majority of follicles were less than 10 mm. Following the start of GnRH infusion, LH concentrations increased to maximum levels (9-18 TABLE 1. Concentrations of serum LH in seasonallv anestrous mares serving as controls and those infused with 2 or 20 ;g GnRH/h (mean f SEM).~ Peak Pretreatment concentration Treatment n (ng/ml) (ng/ml) Day of peakb 20pgGnRH/h 7 <1.0 11.12 f 1.44 9.1 k 1.2 2pgGnRH/h 6 <1.0 12.11 f 1.63 10.6 * 0.2 Control 4 <1.0 - C - athere were no differences between groups treated with 2 pg vs. 20 pg GnRH/h (p>o.os). binfusion begun on Day 0. Values remained less than 0.6 ng/ml throughout period observed. ng/ml) by the eleventh day of treatment (Fig. 2). Although there were neither differences in the peak LH concentration nor day of peak concentration between the two GnRH-treated groups (Table l), serum LH increased at a faster rate in mares treated with 20 pg GnRH/h, as indicated by the finding that by Day 2 of infusion, LH was significantly higher than in the 2-pg-GnRH/h treatment group (Fig. 2). By contrast, serum LH levels remained less than 1.0 ng/ml throughout the sampling interval in the control treatment. +2- +I. I- 3 0-0 J -I- -2. ng Reference Preparation 25 50 75 100 1-11 Serum or pituitary homogenate FIG. 1. Displacement mwes of an equine follicle-stimulating hormone (FSH) reference preparation (efsh, LER-1686-2), equine luteinizing hormone (LH) reference preparation (elh; E98A), equine pituitary extract (PE), and serum pools from a stallion (S), cycling mare (M), and anestrous mare (MA).

1202 JOHNSON TABLE 2. Concentrations of serum FSH in seasonally anestrous mares serving as controls and those infused with 2 or 20 pg GnRH/hr (mean i SEM). Peak Pretreatment concentration Treatment n (ng/ml) (ng/ml) Day of peaka 20 pg GnRH/h 7 34.0 i 7.5 135.2 f 22.7 7.7 i 1.1 2pgGnRH/h 6 29.2 f 4.7 72.0 t 8.7b 3.2 t 0.4' Control 4 20.1 f 1.6 - - Overall mean t S EM 27.2 i 3.4 ahfusion begun on Day 0. bp<0.05 vs. 20 pg GnRH/h group cp<o.ol vs. 20 pg GnRH/h group. Concentrations of serum FSH did not change in the control group over the interval investigated (p>0.05), and mean levels ranged from 18.7 to 33.5 ng/ml (Fig. 3). By comparison, FSH was significantly increased in both GnRH treatment groups by one day after the start of infusion, and remained elevated for a period of 5 days and 7 days in the 2-pg-GnRH/hand 20-pg-GnRH/h-inJected groups, respectively, before abruptly declining. Peak serum concentrations of FSH in individual mares were significantly higher in mares treated with 20 pg GnRH/h (135.2 f 22.7 ng/ml) than in the 2-pg-GnRH/h treatment group (72.0 f 8.7), and the day of peak concentrations occurred significantly later (7.7 f 1.1 vs. 3.2 f 0.4 days from the start of infusion, respectively) (Table 2). Mares infused with 20 pg GnRH/h were found to have at least one follicle greater than or equal to 25 mm in diameter in 7.4 f 1.3 (mean f SEM) days, and 12.0- - 10.0-5 cn 8.0- - +I E > 6.0. c v 3 4.0. 0-0 0--0.-a treatment group 0 2 4 6 8 1 0 1 2 1 4 DAYS FROM START OF INFUSION FIG. 2. Serum luteinizing hormone (LH) in seasonally anestrous mares serving as controls and those infused with 2 pg or 20 pg gonadotropin-releasing hormone (GnRH)/h. Pulsatile infusion of GnRH was continued until one day after ovulation was detected (11.0 t 0.3 days and 13.0 2 0.7 days in the 2- and 20-pg-GnRH/h-treated groups, respectively). all mares had ovulated within 15 (mean 12.0 f 0.7) days. In the 2-pg-GnRH/h-infused group, a 25-mm follicle was found in 5.7 f 0.7 (mean f SEM) days, and ovulation in all mares had occurred within 11.0 (mean, 10.0 f 0.3) days (Table 4). By comparison, control mares failed to show any detectable follicular development, nor did any ovulations occur, within the experimental period (>60 days). Finally, the number of follicles ovulated per mare was related to the dose of GnRH administered, in that mares infused with 20 pg GnRH/h had 3.0 * 0.6 (range 1-6) ovulations per mare, whereas those TABLE 3. Serum progesterone prior to treatment, and during the luteal phase following an induced ovulation.ab Treatment Progestcrone Peak Length of Pretreatment concentration luteal phase n (ng/m 1) (ng/mo Day of peakc (days)d 20 pg GnRH/h 7 0.18 i 0.11 19.97 2 2.37 9.6 i 0.7 2 pg GnRHlh 6 0.05 i 0.02 17.50 t 2 37 8.3 f 0.6 Control 0.15 + 0.06 - e - 14.4? 0.2 14.2 f 0.5 - athere were no differences for any parameter between groups of mares infused with 2 pg GnRH/h vs. 20 pg GnRH/h (p>0.05). bvalues represent mean * SEM COvulation = Day 0. dindicated by number of days serum progesterone was elevated above 1.0 ng/ml. eno ovulations occurred during experimental period; values remained less than 1.O ng/ml throughout period observed.

OVULATION INDUCTION IN ANESTROUS MARES 1203 90.C 80.C 70.( - 60.C 5 cn - +I 50.C E \ - m K 40S I cn LL 30.C 20.C 10s.. _ -............... 0 2 4 6 8 10 12 14 16 18 DAYS FROM START OF INFUSION FIG. 3. Serum follicle-stimulating hormone (FSH) in seasonally anestrous mares serving as controls and those infused with 2 pg or 20 pg gonadotropin-stimulating hormone (GnRH)/h. infused with 2 pg GnRH/h had 1.2 f 0.2 ovulations per mare (5 mares ovulated a single follicle, one mare ovulated 2 follicles) (p<o.os ; Table 3). Furthermore, although after ovulation there were no differences between GnRH-treated groups for the overall peak concentration of progesterone, day of peak con- centration, or length of the luteal phase (Table 4), the peak concentration of progesterone for individual mares, irrespective of treatment group, was found to be related to the number of follicles ovulated (regression equation, Y = 4.49X + 14.69, I = 0.728, df = 12;~ = 0.003). TABLE 4. Number of days to first 25-mm follicle, days to ovulation, and total number of follicles ovulated (within 24 h) in seasonally anestrous mares serving as controls and those infused with 2 pg or 20 pg GnRH/h (mean f SEM).a Days to:b Number of First 25-mm follicles ovulated Treatment n follicle Ovulation (range among mares) 20 pg GnRH/h 7 7.4 f 1.3' 12.0 f 0.7' 3.0 f Ohe (1 to 6) 2 Mg GnRH/h 6 5.7 f 0.7' 10.0 f 0.3' 1.2 f 0.2d (1 or 2) Control 4 >60d >60d OC ~~ ~~~ ~ ~ ~ ameans within a column not followed by a common letter are significantly different (p <0.05). bfrom start of infusion (Day 0). DISCUSSION Serum concentrations of LH were significantly increased above pretreatment values within 48 h after the start of infusion in groups treated with 2 and 20 pg GnRH/h, and this finding is consistent with previous reports that the pituitary of the anestrous mare is competent to secrete LH in response to GnRH (Ginther and Wentworth, 1974; Alexander and Irvine, 1986). Although the rate of increase was more rapid in mares infused with 20 pg GnRH/h (Fig. 2), peak concentrations obtained during the infusion period in individual mares were not different between the GnRH-treated groups (Table 1). These results are similar to those from a previous study, in which infusion doses of 50 and 250 pg GnRH/h were used

1204 JOHNSON (Johnson, 1986a), in that the rate of the serum LH increase was dose-dependent. Alexander and Irvine (1986) reported that the endogenous LH pulse frequency from mares during the follicular phase was 1.14 pulses per h, while the frequency during the luteal phase was considerably lower (0.09 pulses/h), and during anestrus was essentially nondetectable. In addition, they determined that LH responses were induced in all three reproductive states with a dose as low as 3.3 pg GnRH per mare. They concluded that the LH secretion in response to these relatively small doses of GnRH was similar to spontaneous, endogenous LH pulses. Data reported herein show that continued administration of relatively low doses (2 and 20 pg) of GnRH in hourly pulses results in a prolonged increase in serum LH, a profile that is typical of a normal, seasonal preovulatory increase (Whitmore et al., 1973 ;Johnson, 1986c). Furthermore, these data, and those provided by Alexander and Irvine (1986), support the proposal that the concentrations of circulating LH are controlled primarily by the GnRH pulse frequency, and not by seasonal differences in pituitary responsiveness to GnRH. Pretreatment serum concentrations of FSH from anestrous mares reported herein are similar to those previously determined by Thompson et al. (1983) and Hines et al. (1986), but are substantially higher than those reported by Evans and Irvine (1976) and Miller et al. (1980). The quantitative discrepancies among studies are probably due to differences among assay systems and reference preparations; nevertheless, qualitative changes relative to reproductive state would appear to be consistent. For instance, in the present study, pulsatile GnRH treatment caused a significant increase in serum FSH within one day in both treatment groups compared to the controls, and levels continued to increase for 3.2 2 0.4 days (to a level 246.2% of the pretreatment value) and 7.7 f 1.1 days (to 397.4% compared to the pretreatment value) in the 2-pg-GnRH/h and 20-pg-GnRH/h treatment groups, respectively. By comparison, Evans and Irvine (1976) reported that a single 1-mg injection of GnRH increased FSH to a maximum of 349.3%, compared to the pretreatment level, within 30 min of injection. It is of significance to note that the decline in serum FSH in each mare preceded that of LH, and that this decline occurred prior to the termination of the GnRH pulse infusion (the mean dav for the termination of pulse infusion was 11.0 days and 13.0 days in the 2-pg-GnRH/h and 2O-pg-GnRH/h treatment groups, respectively). In fact, when serum concentrations of FSH from individual mares are normalized relative to the day of ovulation, values one day prior to ovulation were not different from preinfusion values (34.1? 6.0 versus 34.0 f 2.5 ng/ml, respectively, in the 20-pg-GnRH/h-infused group, and 28.6? 4.9 versus 20.2 f 4.7 ng/ml, respectively, in the 2-pg-GnRH/h-infused group). One possible explanation for this observation may be that this decrease in serum FSH, despite the continued exposure to GnRH, may occur in response to the secretion of follicular inhibin. For instance, several groups of investigators have reported that inhibin is produced by granulosa cells of the follicle (Erickson and Hsueh, 1978; Henderson and Franchimont, 1983), and in the bovine, increasing concentrations of inhibin-like activity are associated with an increase in the size of the preovulatory follicle (Welschen et al., 1977; Henderson et al., 1984). Inhibin has been shown to block selectively FSH secretion from the pituitary (Marder et al., 1977; de Jong and Robertson, 1985). Miller et al. (1979) have previously reported the presence of an inhibin-like substance in equine follicular fluid, and have shown that injection of this substance, in vivo, decreases serum FSH in ovariectomized mares. In the present study, the growth of follicles was stimulated soon after the initiation of GnRH pulse infusion, as indicated by the presence of a 25-mm follicle at 7.4 and 5.7 days in the 20-pg- GnRH/h- and 2-pg-GnRH/h-infused groups, respectively, and these preovulatory follicles may be a source of inhibin production. Alternatively, there is some evidence to indicate that secretion of follicular estrogens may play a role in the negative feedback on FSH secretion. Garza et al. (1986) reported that treatment of ovariectomized mares with estradiol benzoate decreased serum FSH. Whether or not estrogens play a physiological role in the regulation of FSH release in the intact mare remains to be determined. The length of infusion treatment required to induce ovulation was not different between GnRH treatment groups (Table 4), and the overall mean number of days to ovulation for the two groups (11.1 k 0.5 days) and range in days among mares (9-15 days) are comparable to values previously found (8.8 * 0.7 days, and 7-11 days, respectively) using considerablv hipher doses of GnRH (50 and 250 UP L O

OVULATION INDUCTION IN ANESTROUS MARES 1205 GnRH/h; Johnson, 1986a). By comparison, in recent studies in which GnRH or a GnRH analog was administered in a continuous or twice-daily fashion, ovulation was observed to occur after 3 to 19 days of treatment (Hyland et al., 1986; Fitzgerald et al., 1986; Sanderson et al., 1986). There was in the present study, however, an apparent dose-related influence on the number of follicles ovulated (Table 4), as indicated by the number of corpora lutea found on one or both ovaries within a 24-h period. Infusion of 20 pg GnRH/h resulted in the ovulation of 3.0 f 0.6 follicles per mare, compared to 1.2 f 0.2 ovulations per mare in the 2-pg-GnRH/h group. Previous studies in women have shown that the incidence of multiple follicular development and multiple ovulations can be enhanced by increasing the dose of GnRH administered (Liu et al., 1983; Santoro et al., 1986). For example, Liu et al. (1983) reported that 6 of 6 patients treated with a pharmacologic dose of GnRH (10 pg at 60- to 120-min intervals) responded with the development of 2-5 preovulatory-sized follicles. By contrast, in the present stu dy neither ovulation nor any significant follicular development were found at any time throughout the duration of the experiment in the control mares that were subjected only to the 8L:16D photoperiod. The finding that a nonstimulatory photoperiod will prolong acyclicity in seasonally anestrous mares is in agreement with previously published data from our laboratory (Malinowski et al., 1985). It is also of significance that in every instance an induced ovulation was followed by a functional luteal phase. There were no differences in the lengths of luteal phase between GnRH-infused groups (combined mean, 14.3 f 0.3 days), and the duration was not different from that following a spontaneous, seasonal ovulation (Johnson, 1986b). In this regard, the mare differs from at least one other domestic animal; repeated injections of GnRH to seasonally anestrous ewes will induce ovulation within 48 h after the initiation of treatment, but this ovulation is often followed by abnormal luteal function (McLeod et al., 1982). One explanation proposed for the lack of functional corpora lutea following ovulations induced with GnRH in the ewe is that the period of exposure of developing follicles to episodic LH secretion before the LH surge is too short to result in a fully matured preovulatory follicle (Hunter et al., 1986). By contrast, in the present study, follicles were exposed to elevated LH over a prolonged period of time (9--15 days) before ovulation, and this time is apparently sufficient for follicular maturation to occur. Finally, although there was no difference in the peak concentration of progesterone between the two GnRH-infused groups, there was a positive correlation between the peak concentration of progesterone for individual mares, irrespective of treatment group, and the number of follicles ovulated. These data are consistent with the observation by Squires et al. (1 986) that progesterone concentrations were greater during diestrus in mares ovulating two follicles than in those ovulating a single follicle. In addition, this findings is consistent with that of Liu et al. (1983), who reported significantly higher concentrations of progesterone during the luteal phase of GnRHtreated patients who ovulated multiple follicles compared to control subjects. In summary, pulsatile infusion of relatively low doses of GnRH (2 and 20 pg/h) induces follicular development and ovulation in seasonally anestrous mares in the absence of photoperiodic stimulation. These data provide further evidence that the winterrelated decline in reproductive activity of the mare can be predominantly attributed to a decrease in the quantity and/or pulse frequency of GnRH secretion from the hypothalamus. ACKNOWLEDGMENTS The author is indebted to Ms. S. Becker and Ms. C. Brown for excellent technical assistance, Dr. H. Papkoff for equine LH and FSH, Dr. W. R. Allen for anti-pregnant mare s serum gonadotropin (MISC), Dr. L. E. Reichert for the ovine LH and equine FSH, the National Hormone and Pituitary Program (NIAMDD) for FSH antiserum, and Ms. C. J. Uckele for preparation of the manuscript. REFERENCES Alexander SL, Irvine CHG, 1986. Effect of graded doses of gonadotrophin-releasing hormone on serum LH concentrations in mares in various reproductive states: comparison with endogenously generated LH pulses. J Endocrinol 110: 19-26 de Jong FH, Robertson DM, 1985. Inhibin: 1985 update on action and purification. Mol Cell Endocr 42:95-103 Erickson GF, Hsueh AJW, 1978. Secretion of inhibin by rat granulosa cells in vitro. Endocrinology 103: 1960-63 Evans MJ, Irvine CHG, 1976. Measurement of equine follicle stimulating hormone and luteinizing hormone: response of anestrous mares to gonadotropin releasing hormone. Biol Reprod 15: 477-84 Evans MJ, Irvine CHG, 1977. Induction of follicular development, maturation and ovulation by gonadotropin releasing hormone administration to acyclic mares. Biol Reprod 16:452-62 Fitzgerald BP, Affleck KY, Pemstein R, Loy RG, 1986. An investigation of the potential of LHRH or an agonist to induce ovulation in seasonally anestrous mares with observations on the use of the agonist in problem acyclic mares. Fourth Intern Symp on Equine Reprod, University of Calgary: Abstr. #19

1206 JOHNSON Garcia MC, Freedman LH, Ginther OJ, 1979. Interaction of seasonal and ovarian factors in the regulation of LH and FSH secretion in the mare. J Reprod Fertil 27 (Suppl):103-11 Garza F, Thompson DL Jr, St George RL, French DD, 1986. Androgen and estradiol effects on gonadotropin secretion and response to GnRH in ovariectomized pony mares. J Anim Sci 62:1654-53 Ginther OJ, 1979. Reproductive Biology of the Mare: Basic and Applied Aspects. Ann Arbor, MI: McNaughton and Gunn Inc. Ginther OJ, 1986. Ultrasonic Imaging and Reproductive Events in the Mare. Cross Plains, WI: Equiservices, p. 148 Ginther OJ, Wentworth BC, 1974. Effect of a synthetic gonadotropin releasing hormone on plasma concentrations of luteinizing hormone in ponies. Am J Vet Res 55:79-81 Greenwood FC, Hunter WM, Glover GS, 1963. The preparation of I-labeled human growth hormone of high specific activity. Biochem J 89:114-23 Hart PJ, Squires EL, Imel KJ. Nett TM, 1984. Seasonal variation in hypothalamic content of gonadotropin-releasing hormone (GnRH), pituitary receptors for GnRH, and pituitary content of luteinizing hormone and follicle-stimulating hormone in the mare. Biol Reprod 30: 1055-62 Henderson KM, Franchimont P, 1983. Inhibin production by bovine ovarian tissue in vitro and its regulation by androgens. J Reprod Fertil 67:291-98 Henderson KM, Franchimont P. Charlet-Renard Ch, McNatty KP, 1984. Effect of follicular atresia on inhibin production by bovine granulosa cells in vitro and inhibin concentrations in the follicular fluid. J Reprod Fertil 72: 1-8 Hines KK. Affleck KJ, Barrows SP, Murdoch WL, Barker KB, Loy RG, Fitzgerald BP, 1986. Evidence that a decrease in follicle stimulating hormone (FSH) pulse amplitude accompanies the onset of the breeding season in the mare. Biol Reprod 34(Suppl 1):184 Hunter MG. Southee JA, McLeod BJ, Haresign W, 1986. Progesterone pretreatment has a direct effect on GnRH-induced preovulatory follicles to determine their ability to develop into normal corpora lutea in anoestrous ewes. J Reprod Fertil 76:349-63 Hyland JH, Wright PJ, Clarke IJ, Carson RS, Langsford DA, Jeffcoat LB, 1986. Infusion of gonadotropin releasing hormone (GnRH) induces ovulation and fertile oestrus in mares during seasonal anoestrus. Fourth Intern Symp on Equine Reprod, University of Calgary: Abstr. #29 Johnson AL, 1986a. Induction of ovulation in anestrous mares with pulsatile administration of gonadotropin-releasing hormone. Am J Vet Res 47~983-86 Johnson AL, 1986b. Pulsatile administration of gonadotropin-releasing hormone advances ovulation in cycling mares. Biol Reprod 35: 1123-30 Johnson AL, 1986c. Serum concentrations of prolactin, thyroxine and triiodothyronine relative to season and the estrous cycle in the mare. J Anim Sci 62:1012-20 Liu JH, Durfee R, Muse K, Yen SSC, 1983. Induction of multiple ovulation by pulsatile administration of gonadotropin-releasing hormone. Fertil Steril 40:18-22 Malinowski K, Johnson AL, Scanes CG, 1985. Effects of interrupted photoperiods on the induction of ovulation in anestrous mares. J Anim Sci 61:951-55 Marder ML, Channing CP, Schwartz NB, 1977. Suppression of follicle stimulating hormone in intact and acutely ovariectomized rats by porcine follicular fluid. Endocrinology 101 : 1639-42 McLeod BJ, Haresign W, Lamming GE, 1982. Response of seasonally anoestrous ewes to small-dose multiple injections of GnRH with and without progesterone pretreatment. J Reprod Fertil 65: 223-30 Miller KF, Berg SL, Sharp DC, Ginther, OJ, 1980. Concentrations of circulating gonadotropins during various reproductive states in mares, Biol Reprod 22:744-50 Miller KF, Wesson JA, Ginther OJ, 1979. Changes in concentrations of circulating gonadotropins following administration of equine follicular fluid to ovariectomized mares. Biol Reprod 21:867-72 Sanderson MW, Allen WR, Greenwood RES, Ellis DR, Crowhurst JS, Simpson DJ, Rossdale PD, 1986. Induction of ovulation in anestrous mares with a slow-release implant of GnRH analogue (ICI 118630). Fourth Intern Symp on Equine Reprod, University of Calgary: Abstr. #28 Santoro N, Wierman ME, Filiconi M. Waldstreicher J, Cowley WF Jr, 1986. Intravenous administration of pulsatile gonadotropin-releasing hormone in hypothalamic amenorrhea: effects of dosage. J Clin Endocrinol 62:109-16 Sharp DC, Grubaugh W, 1986. Seasonal patterns of GnRH secretion in the horse assessed by push-pull perfusion. Biol Reprod 34(Suppl. 1): 143 Snedecor GW, Cochran WG, 1967. Statistical Methods. Ames, Iowa: Iowa State University Press, p. 369 Squires EL, Morris RP, Carnevale EM, McKinnon AO, Nett TM, 1986. Reproductive characteristics of spontaneous single and double ovulating mares and superovulated mares. Fourth Intern Symp on Equine Reprod, University of Calgary: Abstr. #52 Strauss SS, Chen CL, Kalra SP, Sharp DC, 1979. Localization of gonadotropin-releasing hormone (GnRH) in the hypothalamus of ovariectomized pony mares by season. J Reprod Fertil 27(Suppl.): 123-29 Thompson DL Jr, Godke RA, Nett TM, 1983. Effects of melatonin and thyrotropin releasing hormone on mares during the nonbreeding season. J Ankn Sci 56:668-77 Turner DD, Garcia MC, Ginther OJ, 1979. Follicular and gonadotropic changes throughout the year in pony mares. Am J Vet Res 40: 1694-1700 Welschen R, Hermans WP, Dullaart J. dejong FH, 1977. Effects of an inhibin-like factor present in bovine and porcine follicular fluid on gonadotropin levels in ovariectomized rats. J Reprod Fertil 50: 129-3 1 Whitmore HL, Wentworth BC, Ginther OJ, 1973. Circulating concentrations of luteinizing hormone during estrous cycle of mares as determined by radioimmunoassay. Am J Vet Res 34:631-36