IIIHIIIWHIWIINIIIIIHIHIIHHIIHIIllHHIIIIIHIIIHI TH Sr I... ‘lmilm ill lililililillil 'J 3129300672 7196 LIBRARY Michigan State University This is to certify that the thesis entitled EFFECTS OF PROSTAGLANDIN an AND GONADOTROPIN RELEASING HORMONE ON ESTRUS, OVULATION AND SERUM HORMONES IN MARES presented by Larry Charles Booth has been accepted towards fulfillment of the requirements for M.S. degree in Department of Large Animal Surgery and Medicine ZZégflZ 2g kgégéfi/fi / M profe so: Wayne D .a](3xen er Date October 24, I977 0-7639 EFFECTS OF PROSTAGLANDIN F2“ AND GONADOTROPIN RELEASING HORMONE ON ESTRUS, OVULATION AND SERUM HORMONES IN MARES BY Larry Charles Booth A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Large Animal Surgery and Medicine 1977 ABSTRACT EFFECTS OF PROSTAGLANDIN an AND GONADOTROPIN RELEASING HORMONE ON ESTRUS, OVULATION AND SERUM HORMONES IN MARES BY Larry Charles Booth This study was designed to determine if 3 injections of GnRH at 8—hour intervals with or without pretreatment with estradiol-l78 would elevate serum LH sufficiently to shorten and/or synchronize the PGFZa induced luteolysis to ovulation interval in mares. Twelve horse.mares received 1 of 3 treatments over 3 consecu- tive estrous cycles utilizing a balanced Latin square crossover (treatment A), PGF + GnRH design. Mares were injected with PGF 2a 2a (treatment B) or PGan + 32-178 + GnRH (treatment C). The mean interval in days from PGF to lst ovulation averaged 1 day less 2o (p<.lO) for treatments B (6.7 :_O.6) and C (6.2 :_0.3) than for treatment A (7.6 :_0.7). Mares treated with GnRH (treatments B and C) showed a significant (p<.Ol) 2- to 3-fold increase in LH levels when compared to controls. Pretreatment with estradiol-l78 did not appear to enhance LH release following GnRH. ACKNOWLEDGEMENTS This thesis is in reality the product of the combined efforts of several persons in addition to myself who deserve my most sincere appreciation for their time and energy. A special thank you to the following persons. Dr. Gable Conner for his counsel and guidance while serving as my academic advisor. Dr. Robert Douglas for directing the experimental project and serving as a valuable source of inspiration and knowledge for both my clinical and research interest in mare reproduction. Dr. John Gill and Dr. Roger Neitzel for their assistance and advice with the statistical analysis of the data. Dr. Patricia Noden for providing invaluable assistance and counseling in quantifying the hormones as well as doing the founda- tion work of establishing the radioimmunoassay procedures used for the mare in this laboratory. Dr. Wayne Oxender, my major professor, for providing the leadership and direction needed for me to complete this project and for exposing me to the world of research and making me aware of its importance in my training. Dr. William Riley, Chairman of Large Animal Surgery and Medicine, for providing full support for my graduate studies and for providing me with time off from my clinical responsibilities to prepare this thesis. ii Ms. Susan Wbodley for the countless number of hours that she devoted to the radioimmunoassay procedures and statistical analysis of the serum hormones. This work certainly could not have been completed without her. To my wife, Marilyn, for the encouragement and moral support she gave me during my research and for the typing and organizing of this thesis. iii TABLE OF INTRODUCTION. . . . . . . . . LITERATURE REVIEW . . . . . . Equine Estrous Cycle . Seasonal Variation. Estrus and Ovulation. Hormonal Changes During the Estrous Cycle. Luteinizing Hormone (LH). Follicle Stimulating Hormone (FSH) Progesterone. . Estrogens . . . Androgens . . . CONTENTS Prostaglandin F20 (PGFZa) . Gonadotropin Releasing Hormone (GnRH) Artificial Control of Estrus and Ovulation in Mares. Prolongation of the Luteal Phase. Termination of the Luteal Induction of Ovulation. Synchronization of Estrus METHODS AND MATERIALS . . . . Animals. . . . . . . . Experimental Design. . Estrus Detection . . . Ovulation Detection. . Blood Collection and Sample Storage. Hormone Quantification . Luteinizing Hormone (LH). Progesterone (P4) Estradiol-l78 (E2—17B) Preparation of Treatments. Statistical Analysis . RESULTS AND DISCUSSION. . . . Estrus and Ovulation after PGFZa' E2-178, and GnRH Treatments in Mares. Phase and Ovulation Serum Hormone Concentrations after Subcutaneous PGan. Serum Hormone Concentrations after PGFZG and GnRH. Serum Hormone Concentrations after PGPZa, 32-178 and Grim O O O O O O O 0 Effects of Ovarian Follicle Size on the Interval to First Ovulation. . . iv Page 22 22 22 24 26 26 26 27 28 29 3O 3O 32 32 35 39 46 50 Page Effect of Multiple Ovulations on Estrus and on the Luteolytic Response to PGan . . . . . . . . . . . . . 53 Effect of Diestrus Ovulations on the Luteolytic Response to PGFZQ Administration . . . . . . . . . . . 55 CONCLUS IONS O O O O O O O O O O O I O O C O O O O O O O O O O O 6 2 BIBLIOGWHY O O O O O I O O O O O I O O O O O O O O O O O O O O 64 Table 10 11 LIST OF TABLES Page Balanced Latin square crossover design utilizing 12 mares. O O I O O O O O O O O I O O I O O O O O O O O O O 23 Schedule of treatment events, frequency of teasing (T), bleeding (B) and rectal palpation (P) following Ovulation. O O O O O O O O O O O O O O O O O O O O O 0 0 25 Estrus and ovulation during a control cycle (a) and following treatments B and C in 12 mares . . . . . . . . 32 Serum hormone concentrations during the estrous cycle following treatment with PGFZa (10 mg so) in 12 mares (control treatment A). . . . . . . . . . . . . . . . . . 36 Serum hormone concentrations during the estrous cycle following treatment with PGFZG (10 mg so) and GnRH (200 ug/injection) in 12 mares (treatment B) . . . . . . 40 Serum hormone concentrations during the estrous cycle following treatment with PGFZQ (10 mg sc), E2-l78 (3 mg sc) and GnRH (200 ug/injection) in 12 mares (treatment C). . . . . . . . . . . . . . . . . . . . . . 46 Serum hormone concentrations in the 3 treatment groups following PGF 0 O O O O O O O O O O O I O O O O 47 2a The influence of ovarian follicle size at the time of PGan administration on the interval in days to lst ovulation in 12 mares. . . . . . . . . . . . . . . . . . 50 The number of multiple ovulations in 12 mares during 36 total estrous cycles categorized according to treatment group. . . . . . . . . . . . . . . . . . . . . 53 The influence of the number of ovulations during an estrus on the duration of estrus in 12 mares . . . . . . 54 Effect of diestrus ovulations on the luteolytic response to PGFZa administration in 4 mares. . . . . . . 61 vi Figure LIST OF FIGURES Page Serum progesterone following PGFZa (10 mg so) in mare no. 10, showing 2 ovulations within 24 hours. . . . 38 Average serum LH in mares given multiple 5 mg doses of GnRH at 24-hour intervals . . . . . . . . . . . . . . 43 Hypothetical serum LH curve expected following 3 injections of GnRH (200 ug/injection) at 8-hour intervals. . . . . . . . . . . . . . . . . . . . . . . . 45 Serum progesterone following PGFZQ (10 mg so) in mare no. 14 with 3 ovulations during an estrus . . . . . 57 Serum progesterone following PGFZa (10 mg sc) in mare no. 2, showing an early ovulation with a short estrus and a diestrus ovulation. . . . . . . . . . . . . 59 vii INTRODUCTION Ovulation control in the mare would be desirable from the standpoint of developing a preplanned breeding program for an individual mare or a group of mares. Management of the stallion would also be more efficient by reducing the number of covers or matings required per mare and allowing more efficient use of semen collected artificially. In the mare, duration of the estrous cycle is controlled by the life span of the corpus luteum (CL). Estrus occurs within 2 to 4 days after regression of the corpus luteum. This regression is termed luteolysis. Prostaglandin F a (PGFZa) has been demonstrated 2 to be an effective luteolytic agent and may be the natural luteo- lytic agent. Administration of exogenous PGF caused rapid regres- 2a sion of the mature CL (at least 5 days postovulation) and mares returned to estrus in 3 to 5 days. The ability to control the onset of estrus does not imply a control of ovulation time since there is a 3 to 7 day variation among mares in the interval between the onset of estrus and ovulation. Researchers have tried to induce ovulation during estrus in mares with various hormone treatments. Luteinizing hormone (LH) or compounds with LH activity have been used to induce ovulation in several species including mares. One such compound, human chorionic gonadotropin (HCG), has been reported to induce ovulation in mares (Loy and Hughes, 1966). However, Sullivan et al. (1973) reported 1 2 a delay in ovulation for mares treated during 3 successive estrous periods with HCG. It appeared that repetitive use of HCG may actually interfere with ovulation in mares similarly to other species. Reel et al. (1976) reported that following multiple treatment with HCG there was a progressive decline in the number of eggs shed and corpora lutea formed in rabbits, mice and cows. Experiments have suggested that the development of this ovulatory refractiveness was due to the formation of antibodies against the foreign protein (Greenwald, 1970). A new synthetic peptide containing 10 amino acids identified as gonadotropin releasing hormone (GnRH) has been used to induce endogenous LH release and thus prevent the injection of foreign proteins to induce ovulation. Mares treated with GnRH had increased LH concentrations (Garcia and Ginther, 1975; Ginther and Wentworth, 1974; Irvine et al., 1975; Oxender et al.,l977h0. Summarizing data from the above authors indicated that GnRH increased LH con- centration significantly for approximately 8 hours following treat- ment and then returned to baseline within 24 hours. The transient LH surge following a single injection of GnRH did not reduce the interval from estrus to ovulation (Ginther and Wentworth, 1974; Garcia and Ginther, 1975; Oxender et al.,l977b). However, LH con- centrations normally remain elevated 2 to 4 days prior to ovulation in mares (Whitmore et al., 1973; Noden et al., 1974; Pattison et al., 1974). Thus, it has been suggested that repeated injections of GnRH may be necessary to duplicate the normal LH response of the mare in stimulating follicular development and ovulation (Ginther and Wentworth, 1974). 3 Estradiol concentrations in peripheral blood reached maximal values 1 to 2 days prior to ovulation and appeared to influence pituitary LH release. Data reported by Garcia and Ginther (1975) indicated that pretreatment with estradiol-178 (Hz-17B) enhanced the LH response to a dose of GnRH. Therefore, it was decided to test the effects of estradiol and repeated GnRH treatments on the interval from luteolysis to ovulation. The objective of this research was to determine if repeated GnRH treatments with or without estradiol pretreatment would main- tain elevated LH concentrations, stimulate follicular deve10pment and synchronize ovulation following PGF induced luteolysis. 2o LITERATURE REVIEW Equine Estrous Cycle Seasonal Variation An important feature of reproduction in the mare is that she is classified a seasonal breeder. Changes in the duration of day- light have influenced the mare's reproductive pattern while environ- mental temperature and nutrition have been shown to modify but not change the basic effect of light (Quinlan et al., 1951). Mares appear to recognize and respond to the changing duration of daylight between 4 important days in the light year. One is the longest day of the year (June 21), the summer solstice, and another is the shortest day of the year (December 21), the winter solstice. The other 2 days are the spring (March 21) and fall (September 21) equinoxes, points halfway between the solstices when the duration of daylight and darkness is equal (Greenhoff and Kenney, 1975). The major population of mares reached its peak of fertility before, around and after the summer solstice and its low point of fertility around the winter solstice. The fertility was either decreasing or increasing the remainder of the year (Hutton and Meacham, 1968). The peak incidence of ovulation occurred around the summer solstice in both the northern and southern hemispheres (Kenney et al., 1975). Individual mares may exhibit regular estrous cycles the year round, while others exhibit a variable period of anestrus during 5 the winter (Day, 1940). The ovaries of mares in deep anestrus are described as being small, firm and containing follicles less than 5 mm in diameter but no corpora lutea were found (Van Niekerk et al., 1973). During late winter and early spring when estrous periods begin, follicles tend to persist for prolonged periods with irregular estrous periods often lasting more than 10 days and not always accompanied by ovulation (Greenhoff and Kenney, 1975). Because follicles often become atretic rather than ovulate, and because the mares are in estrus, Greenhoff and Kenney (1975) term this the period of "anovulatory receptivity." By April, ovulations occur in over 50% of the estrous periods (Arthur, 1958; Osborn, 1966) and the duration of estrus shortens (Andrews and McKenzie, 1941; Trum, 1950; Van Niekerk, 1967), probably because ovulations occur sooner. In summary, associated with increasing daylight, mares showed an increased incidence of psychic estrus, an increased incidence of ovulation and a decreased length of estrus. Estrus and Ovulation During the period of ovulatory receptivity or natural breeding season, the estrous cycle ranged from 19 to 23 days, diestrus from 14 to 16 days, and estrus from 4 to 9 days (Andrews and McKenzie, 1941; Trum, 1950; Quinlan et al., 1951; Van Niekerk, 1967; Ginther et al., 1972; Stabenfeldt et al., 1972; Ginther, 1974). Most researchers indicated that ovulation occurred 24 to 48 hours before the end of estrus (Day, 1940; Andrews and McKenzie, 1941; Trum, 1950; Ginther et al., 1972; Plotka et al.. 1972; Stabenfeldt et al., 1972; Whitmore et al., 1973; Ginther et al., 6 1974). The size of the follicle just prior to ovulation has been reported to be 4.5 to 6.5 cm in Thoroughbreds and 2.5 to 3.5 cm in ponies (Day, 1940). These values are within the range reported by subsequent researchers (Andrews and McKenzie, 1941; Witherspoon and Talbot, 1970; Ginther et al., 1972). Most mares are reported to ovulate only 1 follicle during a given estrous period. The frequency of multiple ovulations in horse mares has been reported as low as 3.8% (Andrews and McKenzie. 1941), up to 25% (Stabenfeldt et al., 1972), to a maximum of 42% (Warszawsky et al., 1972). Stabenfeldt et a1. (1972) reported that the interval between double ovulations was about 24 hours. Ovulations were also reported to occur after the end of estrus (Stabenfeldt et al., 1972; Ginther, 1974). Hormonal Changes During the Estrous Cycle Luteinizing Hormone (LH) Serum LH has been quantified in the peripheral blood of mares by several investigators (Anand et al., 1973; Whitmore et al., 1973; Noden et al., 1975; Evans and Irvine, 1976; Garcia and Ginther, 1976a). The reported LH concentrations by the above authors differ greatly; however, several different assay procedures and LH standards were used. There was general agreement that LH concentrations are low during mid-diestrus and begin to increase a few days prior to the onset of estrus, reaching maximum concentrations 1 to 2 days postovulation. A gradual decrease in LH occurred during the next 4 to 6 days. The pattern of an extended period of increasing LH concentration during estrus and peaking beyond ovulation was in sharp contrast to the short duration of LH release prior to 7 ovulation in cows, ewes and sows (Swanson et al., 1972; Hansel and Ecternkamp, 1972). The prolonged LH surge in the mare may be related to the long periods of follicular growth and estrus in mares compared with other domestic animals. The total biological role of LH is not yet known. It has been suggested by Evans and Irvine (1975) that LH may play an important role in the maturation oftjuaovulatory follicle as well as have a luteotropic effect in the mare. The findings of Pineda et al. (1972) and Pineda et a1. (1973) utilizing an antiserum against an equine pituitary fraction having follicle stimulating hormone (FSH) and LH activity supported these views. They showed that the largest follicle in antiserum treated mares was reduced in volume, weight and diameter versus controls, and the weights of corpora lutea were significantly less for antiserum treated mares. The persistence of high LH concentrations into the postovulatory period has been proposed as the reason for double ovulations often reported (Geschwind et al., 1975) . Garcia and Ginther (1976a) used ovariectomized mares to study seasonal variation in LH. The LH con- centrations were low in these mares during the anovulatory season and high during the ovulatory season, indicating an environmental influence affecting basal LH levels in the absence of ovarian steroids. Follicle Stimulating Hormone (FSH) It was only recently that a method of measuring serum concentra- tions of equine FSH was developed (Evans and Irvine, 1975, 1976). The authors quantified circulating concentrations of FSH during the estrous cycle and early pregnancy in the mare. They showed that 8 serum FSH levels were elevated during late estrus and early diestrous period and again at mid-diestrus, 10 to 13 days before the next ovulation. These reported surges of FSH seemed to coincide with periods of follicular development as demonstrated by Pineda and Ginther (1972) and Pineda et a1. (1973). Evans and Irvine (1975) suggested that the mid—diestrus surge of FSH was important in trig- gering the maturation of the next ovulatory follicle. Work in pony mares (Miller et al., 1977) also showed a definite surge of FSH 9 to 11 days prior to ovulation which would agree with the work in horse mares. However, the data from the pony mares did not show a rise in FSH serum levels prior to ovulation as was found in horse mares. No explanation can be given for these differences at this time. Initiation of corpus luteum development and progesterone pro- duction occurred rapidly following ovulation. Follicle stimulating hormone levels have been shown to be increased during this time and LH levels were still quite high, gradually decreasing to baseline levels by 5 days postovulation. The presence of high levels of gonadotropins at this stage has been postulated as being important for early luteal development. The dependency of corpora lutea on pituitary hormones has been tested utilizing antibodies to equine pituitary fractions with FSH and LH activity (Pineda et al., 1972, 1973). The mean weight of the corpus luteum was significantly less for antiserum-treated mares than for control mares, indicating that gonadotropins were necessary to maintain the corpus luteum. Garcia and Ginther (1974) administered pituitary hormones to determine if they would extend luteal life span or prevent the luteolytic effect of other treatments. In one trial, 9 pony mares were treated with 2,000 units of HCG (shown to have LH and FSH activity) each day from days 11 to 16 of diestrus and this treatment failed to prevent luteolysis. In a second experiment, HCG or an equine pituitary extract was given daily on days 10 to 17. There was a significant reduction in progesterone concentrations by day 13 in control mares and day 17 in HCG-treated mares. However, progesterone did not decrease in the pituitary extract-treated mares, suggesting maintenance of the corpus luteum. Although research to date has failed to identify the active luteotropic substance in mares, the studies certainly indicated that maintenance of the equine corpus luteum was dependent upon circulating gonadotropins. Progesterone Progesterone is a steroid produced by the corpus luteum. Recently, numerous studies have reported progesterone concentrations during the estrous cycle in mares. The earlier work was done with competitive protein binding assay techniques (Smith et al., 1970; Stabenfeldt et al., 1972; Plotka et al., 1972; Sharp and Black, 1973; Allen and Hadley, 1974; Van Niekerk et al., 1975), while later studies utilized radioimmunoassay (RIA) techniques (Squires et al., 1974; Noden et al., 1975; Palmer and Jousset, 1975a). Although there was variation among the reports, there was general agreement that serum progesterone concentrations were low during estrus. Most reports indicated less than 1 ng/ml serum progesterone during estrus. Progesterone concentrations increased within 24 to 36 hours following ovulation and proceeded to maximal values by 5 to 7 days after ovulation. Although there were fluctuations in progesterone concentrations during diestrus, they generally remained elevated 10 until luteolysis on day 13 or 14. Following luteolysis, concentra- tions decreased rapidly until the low estral levels were reached 2 to 3 days later. Of interest, Stabenfeldt et al. (1972) noted no significant difference in the progesterone levels between mares with l or 2 corpora lutea. When 2 ovulations occurred, the values did not increase until 24 hours after the 2nd ovulation. It was prOposed that the presence of a large preovulatory follicle was probably inhibiting the steroidogenic capabilities of the corpus hemorrhagicum formed by the lst ovulation. Estrogens The determination of the circulating concentrations of estrogens has been more difficult than progesterone due to much lower circulat- ing concentrations and the presence of several estrogen compounds. Thus, these estrogens cross react with estrogen antisera and compli- cate quantification procedures and interpretation of the results. Reported levels in plasma or serum vary greatly (Pattison et al., 1974; Noden et al., 1975; Plotka et al., 1975). However, the relative changes reported between estrus and diestrus were similar. Serum estradiol concentrations reported by Pattison et al. (1974) for estrous and diestrous mares were 141 pg/ml and 20 pg/ml, respec— tively, while values reported by Noden et al. (1975) were 11.5 pg/ml and 2.2 pg/ml. Plasma estrone as reported by Noden et a1. (1975) ranged from 9.2 to 12.5 pg/ml throughout the estrous cycle, showing no significant difference among the days. The estrogen values obtained by Plotka et a1. (1975) combined estradiol and estrone; as a result, they failed to show the changes in estrogen levels around the time of ovulation as reported by the others. 11 In general, near the beginning of estrus the blood concentrations of estradiol—178 began increasing, peaking approximately 1 to 2 days before ovulation. This increase in estradiol occurred parallel with an increase in follicular size. Estradiol concentrations appeared to decrease by the time ovulation had occurred and reached diestrus levels within a day or two after ovulation. Androgens The physiological role of the androgens as they relate to the mare's estrous cycle is not known. Recent studies in rats (Schreiber and Ross, 1976) suggest the possibility that androgens may be linked with control of atresia and preantral follicular maturation. Andro- stenedione, an estrogen precursor, has been shown to increase prior to ovulation (Noden et al., 1975; Ganjam and Kenney, 1976) at a time when the large nonovulatory follicles are regressing. Recently, a preovulatory rise in dehydroepiandrosterone levels has been reported in mares (Rance et al., 1976). The significance or source is not yet known. ' P Prostaglandin F2c ( GFZGL Although PGF a is usually not thought of as being a reproductive 2 hormone, it does fit the classical definition of a hormone. It is secreted by an organ, the uterus, and travels through the blood to exert its effect on a distant target organ, the ovary. The concen- tration of the F-series of prostaglandins has been measured in the uterine veins during the estrous cycle of the mare by Douglas and Ginther (1976). The PGF concentration increased significantly between days 10-14 of diestrus near the time luteolysis occurs. It was assumed by the authors that the primary prostaglandin measured 12 was PGF The implications of the findings add an important piece 2a' of evidence to the growing body of knowledge indicating that prosta- glandin may be the natural uterine luteolysin in the mare. During the 19705, investigators had demonstrated that prosta- glandins were potent luteolysins in many species. Work in the mare (Douglas and Ginther, 1972) showed that a single subcutaneous injection of PGan caused mares to return to estrus in 3 days. Subsequent work confirmed the luteolytic effect of exogenous PGF in the cycling 2a mare (Noden et al., 1974), pseudopregnant and pregnant mares (Kooistra and Ginther, 1976) and hysterectomized mares (Douglas et al., 1975). The post-treatment effects on the behavioral, hormonal and ovulatory events of the ensuing estrus were similar to controls (Noden et al., 1974). It is of interest that the newly developing corpus luteum (lst 4 days after ovulation) appeared resistant to the luteolytic effects of PGF (Douglas and Ginther, 1975a; Oxender et al., 2a 1975a). The reason has not been determined. The minimal effect dose of exogenous PGFZa in horse and pony mares averaged approximately 8 Ug/kg of body weight (Douglas and Ginther, 1972; Oxender et al., 1975a). This is in marked contrast to the sheep, in which the minimal effective dose is 144 ug/kg (Douglas and Ginther, 1973b). The unusual sensitivity of mares to PGan may relate to the greater affinity of the membrane of the corpus luteum from mares for binding PGF when compared with the 2d cow (Kimball, 1976). Gonadotropin Releasing Hormone (GnRH) Gonadotropin releasing hormone is a peptide containing 10 amino acids which is produced by the hypothalamus and is transported to 13 the adenohypophysis through a system of portal vessels. Natural and synthetic GnRH has been shown to cause pituitary release of both LH and FSH in several species (Schally et al., 1976). Ginther and wentworth (1974) demonstrated that single intravenous injections of GnRH in anestrous pony mares caused an immediate increase in plasma LH concentration that gradually decreased within 2.5 hours to pre- treatment levels. Administration of GnRH subcutaneously on day 2 of estrus caused a more prolonged rise in LH values which reached maximum 1 hour post-treatment and then decreased to pretreatment levels by 8 hours. In these experiments there was no effect on the GnRH treatment on follicular development, time of ovulation or length of estrus. Evans and Irvine (1976) further characterized the effects of GnRH in the mare by measuring both FSH and LH follow- ing injection of GnRH into 7 anestrous mares. They reported signifi- cant increases in both FSH and LH following GnRH treatment. In fact, FSH release was significantly greater than that for LH. This is of interest because it had been suggested that GnRH was primarily LH releasing in the human, rat, sheep and cow when the FSH and LH changes were compared by expressing the increases as a fraction of the baseline concentration (Evans and Irvine, 1976). The authors suggested that the difference in FSH and LH response of the mare to GnRH in comparison to other species was due to a difference in the pituitary response. Pursuing this difference in pituitary response between species still further, experiments on women seem to indicate that GnRH caused the release of stored gonadotropins rather than active synthesis and that the pituitary could be depleted of gonadotropin (Newton et al., 1973). 14 A refractory period in LH release following repetitive injec- tions of GnRH has been demonstrated in the ewe and the steer (Rippel et al., 1974; Tannen and Convey, 1977). Rippel et al. (1974) con- cluded from their data that both the anestrous and ovariectomized ewe had a refractory period in the mechanism controlling GnRH induced LH release. Successive injections of GnRH at 96-hour intervals resulted in LH release similar to the initial dose. How- ever, a decline in LH release occurred at 72-hour injection intervals. Pituitary concentrations of LH and FSH were not depleted, which sug- gested that pituitary secretion was not necessarily related to hormone content of the gland. In recent work by Bailey and Douglas (1977) where anestrous mares were injected with GnRH every 8 hours for 14 days, a period of refractivity was not apparent. Artificial Control of Estrus and Ovulation in Mares Understanding how the previously mentioned reproductive hormones interact with one another to control the mare's estrous cycle is, of necessity, very important if we are to try to artificially manipu- late this dynamic process. Studies on cause and effect relationships are difficult to devise. So far, studies involving only 2 or 3 hormones have been done, and conclusions drawn from the information pieced together may not be valid. Studies thus far have been on the effects of one hormone upon the circulating concentrations of another or simply comparing the interrelationships in concentrations over time; for example, one hormone increasing in concentration when another is decreasing. Thus, one should proceed with caution to make generalizations about what is known at this time. 15 The estradiol and LH curves are roughly synchronized, except that the estradiol peak occurred approximately 2 days before the LH peak and the estradiol concentration decreased sooner following ovulation than LH (Pattison et al., 1974; Noden et al., 1975). In contrast, estradiol and progesterone fluctuated reciprocally, as would be expected; progesterone characterized the diestrous phase and estrogen characterized the estrous phase. The progesterone decrease at the end of diestrus preceded the increase in LH (Noden et al., 1974), while the progesterone increase after ovulation preceded the reduction in LH concentrations (Anand et al., 1973). The above mentioned relationships between estradiol, progesterone and LH suggested that progesterone had a negative effect on LH output while estrogen had a positive effect on pituitary LH release. As was noted earlier, LH concentrations followed a seasonal pattern in ovariectomized mares. In a subsequent study, Garcia and Ginther (1976b) examined the effects of exogenous estradiol and progesterone on LH in ovariectomized mares during the ovulatory season. Their conclusions were that estradiol-178 treatment caused a transient increase in plasma LH levels while progesterone treatment decreased plasma LH. However, the concurrent administration of estradiol-178 and progesterone caused a greater decrease in plasma LH levels than progesterone alone. In summary, it appeared that the ovarian steroids were exerting their influence upon the hypothalamus, altering the release of GnRH and the subsequent release of LH by the anterior pituitary, or that they had a direct effect on the pituitary or perhaps the ovarian steroids were controlling both the hypothalamus and pituitary. l6 Reeves et al. (1971a) showed that LH release in response to GnRH in ewes was greatest during an 8-hour period on day 1 of the estrous cycle, providing evidence to indicate elevated serum concen- trations of estradiol increased LH release after GnRH treatment. In addition, pretreatment of ewes with 250 to 500 mg of estradiol benzoate 20 hours before GnRH treatment increased LH release com- pared to control ewes (Reeves et al., 1971b). Zolman et al. (1973), however, failed to detect a significant variation in the magnitude of LH release in heifers treated with GnRH on different days of the estrous cycle. Estradiol-l78 pretreatment enhanced the LH response in mares to a dose of GnRH, which was ineffective in eliciting a response when given without estradiol (Garcia and Ginther, 1975). The data indi- cated that estradiol treatment alone might also induce an LH response. While it is evident that the ovarian steroids modify the hypo- thalamic pituitary response, the mechanisms involved are not known. It appeared that the role of estradiol and progesterone on LH release was complex and may involve modulation at both the hypothalamic and pituitary level. While the control mechanisms of estrus and ovula- tion appear complex, several researchers have successfully changed the functional life of the corpus luteum. In addition, there has been limited success with induction of ovulation in mares. Following is a brief review of the methods used to control estrus and ovulation in mares. Prolongation of the Luteal Phase Progestins, a term used to identify several compounds with pro- gesterone activity, are the synchronizing agents used most extensively 17 in many species. The theory is to maintain progestin blood concen- trations and keep the animals in a diestrous state artificially until the corpora lutea in all animals of the group have regressed. When the progestin is withdrawn, estrus and ovulation occur in a synchronized fashion. The original studies utilizing exogenous progesterone in the mare were conducted by Loy and Swan (1966). Injections of 100 mg of progesterone per day and higher amounts beginning in mid-diestrus blocked estrus and ovulation in all mares. Follicular development occurred prior to the end of treatment when the treatment period extended beyond 20 days after ovulation. Fifty milligrams per day prevented estrus but not ovulation. In contrast, treatment with 100 mg of progesterone per day starting on the lst day of estrus failed to prevent estrus and ovulation. Synthetic progestins were also studied by Loy and Swan (1966). One of these, 6a-methyl—l7a-acetoxyprogesterone (MAP), administered orally to mares in midcycle or to mares showing erratic estrous behavior, failed to inhibit estrus or ovulation at levels of from 400 to 1,782 mg per day. Melengestrol acetate (MGA) also was inef- fective in cycling mares when given orally at 10 to 20 mg per day. Another orally administered synthetic progestin (17—a-allyl-estra- triene-4-9-ll, l7-B-ol-3-one) was reported to be effective in cycling mares (Webel, 1975) . This new progestin in a dose of 15 or 20 mg daily for 18 days resulted in a withdrawal-to-estrus interval of 5.4 I.3°1 and 5.8 :_2.7 days, respectively. Another method of prolonging the luteal phase or delaying follicular development has been to block pituitary gonadotropin release. Methallibure, a potent inhibitor of gonadotropin secretion, was reasonably successful in synchronizing estrus in mares, but 18 induced inappetence and other side effects made it unsuitable (First, 1973). Thus, exogenous progestins appeared to be more satisfactory than methalibure in delaying estrus and ovulation in mares. Termination of the Luteal Phase One of the analogues of prostaglandin F20 or PGFZa itself caused rapid regression of the mature corpus luteum. Several reviews on the applied uses of PGF are available (Oxender et al., 1974; 2a Allen and C00per, 1975; Lauderdale and Miller, 1975; Nelson, 1976). Prostaglandin F administration caused a predictable return to 2d estrus in 2 to 4 days if the mare had a mature corpus luteum. As had been mentioned previously, the corpus luteum was refractory to PGFZa treatment for 4 to 5 days following ovulation (Allen and Rowson, 1973; Douglas and Ginther, 1973a; Oxender et al., 1975a). In addi- tion to the lst 4 days postovulation, mares already in estrus could not be synchronized with a group treated with PGF Therefore, only 2o' about 60% of a group of mares treated at any one time with PGFZa would be in estrus 2 to 4 days later. Two PGFZo treatments 12 to 14 days apart have been used to synchronize estrus in group treated mares (Palmer and Jousset, 1975b). Although PGFZa treatment induced luteolysis in most mares on day 5 to 14 postovulation, and the interval from PGan treatment to ovulation averaged 8.5 to 9.0 days, the variation of response for individual mares ranged from 3 to 12 days. This lack of precision for predicting the time of ovulation in advance has caused researchers to pursue additional methods to control the time of ovulation in mares. 19 Induction of Ovulation Several studies have been done on the effects of HCG on estrus and ovulation (Loy and Hughes, 1966; Ginther, 1971; Sullivan et al., 1973; Voss et al., 1975). Data from the above work consistently showed that the duration of estrus was reduced in mares treated with HCG. The variation in length of estrus was also reduced (Loy and Hughes, 1966). A large percentage, 88.9% (Loy and Hughes, 1966) and 82.4% (Sullivan et al., 1973), ovulated within 48 hours post- treatment. The work by Sullivan and his co-workers showed that, although the treatment was effective during the lst cycle, treatment during the 2nd and 3rd cycles was ineffective in reducing the estrus to ovulation interval or the length of estrus. In fact, the 3rd consecutive treatment resulted in a significant lengthening of estrus. Although there appears to be some controversy over the effects of repeated HCG treatment in mares, HCG administration has not been shown to have adverse effect on conception rates (Loy and Hughes, 1966; Voss et al., 1975). Gonadotropin releasing hormone has also been used to induce ovulation in mares (Ginther and Wentworth, 1974; Irvine et al., 1975; Heinze and Klug, 1975). It has been demonstrated to cause a signifi- cant increase in serum LH levels after a single administration. Irvine et al. (1975) showed that 2 mg GnRH given on day 2 of estrus significantly shortened estrus, but the time of ovulation was not changed. However, when 2 mg GnRH was administered daily, the duration of estrus and the time of ovulation were significantly shorter than for the controls. In clinical tests demonstrating the efficacy Of GnRH, Heinze and Klug (1975) showed that all but 1 of 39 mares having a mature 20 follicle ovulated within 48 hours of an injection of 4 mg GnRH when the estrus lasted greater than 9 days. Only 11 of 16 control mares ovulated within the same period. This indicates that at least in some mares GnRH treatment may be successful in inducing ovulation of a mature follicle. Synchronization of Estrus and Ovulation The research results reported above led some researchers to combine progestin and luteolytic treatments with gonadotropin treat- ment in an attempt to increase the synch rony of ovulation in mares. Holtan et al. (1974, 1977) administered 50 mg progesterone per day for 19 days followed by HCG 6 days after withdrawal to 21 pony mares. Fifteen of the 21 mares ovulated between days 6 to 9 after withdrawal. A comparative study by the same authors utilized PGF and HCG. The 2a PGFZa was given on days 0 and 18 and the HCG was given 6 days follow- ing the 2nd injection of prostaglandin. The use of 2 injections of PGan 18 days apart was designed to induce luteolysis if a mature corpus luteum was present at the lst injection and hOpefully all corpora lutea in all mares would be approximately 10 to 19 days old by the time of the 2nd injection. Results indicated that better synchronization of the day of ovulation occurred following the progesterone-HCG treatment than with the PGF a-HCG treatment. 2 However, a French research group used the PGF a-HCG combination 2 on a different schedule and obtained favorable results. Palmer and Jousset (1975b) tested a regime as follows: day 0, prostaglandin; day 6, HCG; day 14, prostaglandin; day 20, HCG. Treatment was started without regard to the stage of the estrous cycle. Because the onset of estrus is more synchronized than the day of ovulation 21 using PGF an ovulating hormone such as HCG is useful to reduce Za' this variability. About 50% of the 48 mares treated ovulated within 48 hours of the 2nd HCG injection. Although these treatments can be used to obtain synchrony of ovulation, there remains the question about the fertility of con- trolled ovulations. In an experiment Holtan, Douglas and Ginther (1977) used a combination of progesterone, PGF and HCG together 2a followed by single or multiple breedings to determine pregnancy rate. The treated pony mares received 75 mg progesterone for 10 days in addition to PGFZa (1.25 mg) on day 7 during progesterone treatment. The prostaglandin was used to prevent survival of any corpora lutea beyond the progesterone withdrawal period. Human chorionic gonado- trOpin (2,000 IU) was given 5 days following progesterone withdrawal. There was no significant difference in the pregnancy rate between the control mares or the treated mares that received either single or multiple breedings. Recent work by Oxender et al.(1977tn was done to determine the dose of GnRH required to cause increased serum LH in mares and to see if this increase would synchronize ovulation following induction of luteolysis with PGF Results showed a significant rise in LH 2a' levels using GnRH (1 to 5 mg sc). However, the interval between PGan and ovulation was not significantly reduced, nor was the time of ovulation synchronized. Although much has been learned about the controls of estrus and ovulation in mares, it appears much more research will be necessary before we can accurately predict in advance the day of ovulation for mares. METHODS AND MATERIALS Animals Twelve mares of mixed breeding, 4 to 21 years of age and weighing 300 to 500 kg, were used in this experiment, which was conducted during June and July of 1976. The mares were kept on pasture through- out the experiment. Experimental Design Twelve mares were teased daily to determine estrous behavior and time of ovulation in late May 1976. On the day of ovulation (day 0), which occurred during estrus of a control cycle, each mare was assigned to l of 4 squares in a balanced Latin square crossover design (Table 1). Each of the 12 mares received 1 of 3 treatments in each of 3 consecutive estrous cycles such that each mare received all 3 treatments (A, B and C) during the experimental period. The order of treatments for each mare was fixed. The design insured that each treatment preceded and followed the other 2 treatments an equal number of times. This design allowed.for detection of period effects as well as residual effects of one treatment on subsequent treatments. The 3 experimental treatments were: 1) A, Controls, Prosta- glandin F (PGF ); 2) B, PGF 2a 2a (GnRH); and 3) C, PGF 2a and Gonadotropin releasing hormone , GnRH and Estradiol-17B (E2-17B). The PGF 2a 2a (10 mg so) was injected on day 7 postovulation; peanut oil (P0) or 22 23 < m o m o < o m N om0mmouo mumzvm sauna poocmHmm .H dance 24 E2-l78 (3 mg sc) was injected on days 8 and 9; and saline or GnRH (0.2 mg sc) was injected at 1630 and 2400 hours on day 10 and at 0800 hours on day 11 (Table 2). In the case of mares having 2 or more ovulations during estrus, PGF was administered 7 days follow- 2a ing the last ovulation. Diestrus ovulations were not considered when calculating the day for PGan administration. Blood sampling, estrus detection and rectal palpation were per- formed according to plan throughout the 3 consecutive estrous periods stopping 15 days postovulation following the final treatment regime. Estrus Detection Mares were teased daily each afternoon at 1630 hours with a mature stallion. A teasing criterion was established and behavioral signs were noted as follows: T = Raising tail prior to being mounted or after being dismounted U = Urinating anytime during teasing W = Winking vulva anytime during teasing Sq = Squatting 50 ll Raising tail while being mounted , S = Switching of tail anytime during teasing K = Kicking (includes strikes) during teasing B = Attempts to bite the teaser M = Movement to prevent mounting by the teaser N = Failure to stand firmly while being mounted D = Standing firmly while being mounted but failing to elevate tail X = Failure of teaser to mount 25 Table 2. Schedule of treatment events, frequency of teasing (T), bleeding (B) and rectal palpation (P) following ovulation Days post- Treatmenta ovulation A B C 7, 1630 hours T'B'P-PGFZQ T'B'P-PGFZG T,B,P-PGFZa 8, 1630 T,B-PO T,B-PO T,B-E2 9, 1630 T,B-PO T,B-PO T,B-E2 10, 1630 T,B,P-S T,B,P-GnRH T,B,P-GnRH 2400 B-S B-GnRH B-GnRH 11, 0800 B—S B-GnRH B-GnRH 1630 T,B T,B T,B 12, 1630 T,B T,B T,B 13, 1630 T,B,P T,B,P T,B,P 14, 1630 T,B T,B T,B 15, 1630 T,B T,B T,B aPGF = 10 mg sc 2a Estradiol-17B (E2) = 3 mg in peanut oil so GnRH = 0.2 mg in saline sc Peanut oil (P0) = 3 m1 sc Saline (S) = 1 m1 sc 26 For purposes of determining onset of estrus and estrus duration, it was decided that a mare would not be considered in estrus until she stood firmly to be mounted with her tail raised along with a majority of positive signs, including winking, squatting and urinating. Cessation of estrus was noted when the mare no longer stood to be mounted. Ovulation Detection Mares were palpated rectally to determine ovarian follicular changes at least every 3rd day unless a mare was in estrus or an ovarian follicle greater than 35 mm in diameter was detected. Mares in the latter category were palpated daily until ovulation or follicu- lar atresia occurred. Blood Collection and Sample Storage Each day at 1630 hours, at 1630 and 2400 on day 10 and 0800 on day 11, blood (20 ml) was collected by jugular venipuncture utilizing a 20 ml vacutainer tube with a 20 gauge 1 1/2 inch needle. The samples were placed on ice until transferred to the laboratory, where they were allowed to stand under refrigeration for 48 to 72 hours to allow sufficient clot retraction. After centrifugation, the decanted serum was placed in plastic vials and stored at -20 C for quantifi- cation of hormones. Hormone Quantification Serum concentrations of luteinizing hormone (LH), progesterone (P4) and estradiol-178 (E2—17B) were determined using radioimmuno- assay procedures. 27 Luteinizing Hormone (LH) Plasma LH concentrations were assayed using a double antibody technique developed by Niswender et al. (1968) for rat LH and modi- fied for equine LH by Noden et al. (1974). Dilution duplicates of equine serum were used, 200 pl for sample unknowns thought to be low in LH and 50 ul and 100 pl for sample unknowns thought to be high in LH concentration. All samples were diluted to 500 ul in phosphate buffered saline (PBS) with 0.1% Knox gelatin. Four sets of standard equine LH sera were placed at equal intervals throughout the assay. The standard equine LH (LER 1138-1, 0.27 U NIH LH-Sl/mg) was supplied by Dr. L. E. Reichert of Emory University, Atlanta, Georgia. Each standard set consisted of 12 tubes containing a range from 2 to 100 ng of equine LH in 500 pl of PBS-0.1% Knox. Two hundred microliters of lst antibody (rabbit antiovine LH #15 from Gordon Niswender, Colorado State University, Fort Collins, Colorado) was added to each culture tube at a dilution of l:32,000 in 1:400 NRS (normal rabbit serum) in PBS:EDTA. Incubation of the tubes at 4 C for 24 hours allowed equilibration between the free antigen (LH), free antibody and bound antigen-antibody complex. Purified ovine LH (LER-1056C2) was then radioiodinated using 1251 in a method similar to that described by Niswender et al. (1968). The stock solution of 1251 was diluted with PBS-0.1% Knox so that 100 ul contained approximately 20,000 counts per minute (CPM). One hundred microliters of 125I-LH solution was then added to each culture tube and the tubes were again incubated at 4 C for 24 hours. The labeled 125I-LH competes with the unlabeled antigen (LH) for the available antibody binding sites during the incubation period. 28 A 2nd antibody (sheep anti-rabbit gamma globulin) diluted 1:10 in PBS:EDTA was then added to form an antigen-antibody-antibody complex large enough to be precipitated by centrifugation. Two hundred microliters of the 2nd antibody was added to teach tube, followed by incubation for 72 hours. After incubation, 3 ml of cold 125I PBS was added to each tube to dilute the unbound -LH. Centrifu- gation at 3,000 rpm for 30 minutes in a refrigerated.centrifuge pre- cipitated the bound 125I-LH and the supernatant fluid was discarded. The bound 125I-LH in the precipitate was then quantified in an auto- matic gamma counter. A standard curve was calculated from counts derived from the sets of standard sera using a computer program and multiple regression analysis procedures. The LH concentrations in the unknowns were computed by multiple regression analysis comparison with the standard curve . Progesterone (PA) The assay technique used for P has been described (Convey et 4 al., 1977). Briefly, the assay utilized mare's serum in duplicate. Progesterone was extracted from 100 ul serum diluted with 100 ul of buffer by vortexing for 30 seconds with 2 ml of redistilled benzene- hexane solvent (1:2). This produced a benzene-hexane layer containing the P4 at the top of the tube and an aqueous lower layer. The solvent layer has a lower freezing point than the aqueous layer, allowing separation by placing the tubes in a freezer for 1 hour and then pouring off the solvent layer containing the P into assay tubes. 4 Three sets of standards (.01 to 1.0 ng) were included in each assay. 29 Solvent was evaporated from the unknowns and standards in a vacuum oven. Progesterone antibody (MSU anti-progesterone #74) was diluted 1:5,000 in PBS-0.1% Knox gelatin containing 1:100 normal rabbit serum. TWO hundred microliters of the diluted antibody was added to each 3 assay tube and vortexed. Then 200 pl of H-P (5,000 cpm/ZOO Ul) 4 was added to each tube. After 24 hours of incubation at 5 C, 400 pl of 2nd antibody (sheep anti-rabbit gamma globulin) in PBS-0.1% Knox was added to each tube. After 48 hours of incubation the assay tubes were centrifuged at 3,000 rpm for 30 minutes and 0.5 m1 of the supernatant fluid was diluted with scintillation fluid (4.5 ml) for radioactive quantification. Estradiol-178 (En-17B) The technique used for assaying E2-17B has been described (Oxender et al.,1977a). The assay utilized mare's serum in tripli- cate. Estradiol was extracted from 0.5 ml serum by vortexing for 1 minute with 5 m1 of redistilled benzene. This produced a benzene layer containing the estradiol at the top of the tube. The lower serum layer was frozen in dry ice-methanol bath and the benzene layer was decanted into assay tubes. Three sets of standards (0 to 50 pg estradiol-178) were included in the assay. Benzene was evaporated from the unknowns and standards in a vacuum oven. Estradiol antibody (MSU anti-estradiol #74) was diluted 1:5,000 in PBS-0.1% Knox gelatin and 1:100 normal rabbit serum and vortexed. After 24 hours of incubation at 5 C, 400 pl of 2nd antibody (sheep anti-rabbit gamma globulin) in PBS-0.1% Knox was added to each tube. After 48 hours of incubation, the assay tubes were centrifuged at 30 3,000 rpm for 30 minutes and 0.5 ml of the supernatant fluid was diluted with scintillation fluid for radioactive quantification. Preparation of Treatments Dr. J. W. Lauderdale of the Upjohn Company provided the PGan- Tham salt. The salt was reconstituted and diluted in 0.85% saline to the desired concentration of PGF a free acid (10 mg/3 ml) for 2 individual mare treatment. Estradiol-178 was purchased as a desiccated powder from Sigma Laboratories. Peanut oil was used as a diluent to obtain a final concentration of 3 mg estradiol-178 per 3 ml peanut oil for treatments. The gonadotropin releasing hormone (GnRH) used was an analogue (D-Leu26) supplied by Dr. S. Webel of Abbott Laboratories. The desiccated powder was dissolved in saline to provide 200 pg of GnRH per 3 ml saline. Statistical Analysis Four separate 3 x 3 Latin squares were utilized with a one-way analysis of variance to test for significant effects of periods, treatments and residual effects of treatments. With this technique, any residual effects of treatments and/or period effects could be compensated for by using adjusted treatment means when testing treat- ments. Then, to qualify significant effects of treatments on the intervals from PGFZu to estrus and PGFZa to lst ovulation, orthogonal contrasts were performed using a one-sided t-statistic. A Latin square design with repeat measure was used to test the significance of changes in serum hormones utilizing a split-plot analysis of variance followed by orthogonal contrasts among treatments and contrasts over time when indicated. 31 A one-way analysis of variance was used to test for significant differences between ovulation intervals within treatment groups when looking at the effects of follicle size at the time of prostaglandin F2“ administration. A one-way analysis of variance was also used to compare the mean intervals for estrus duration within treatment groups for mares having 1 or 2 ovulations. Distribution of multiple ovulations among the 3 different treat- ment groups A, B and C was evaluated utilizing Chi square analysis. RESULTS AND DISCUSSION Estrus and Ovulation after PGF a1 E2-17B, and GnRH Treatments in Mares The interval from PGFZc (10 mg sc) treatment to estrus averaged 3.4 i 0.2 days for the control mares (treatment A). Mares in treat- ment B had a similar interval of 3.2 i 0.3 days; however, the interval from PGFZa to estrus for treatment C mares, 2.3 :_0.1 days, was sig- nificantly shorter (p<0.5) than for the other 2 groups (Table 3). Table 3. Estrus and ovulation during a control cycle (A) and following treatments B and C in 12 maresa Treatment Interval Ab Bc C --—- — (days) ----------------- From PGFZa to estrus 3.4 :_O.2 3.2 :_O.3 2.3 i 0.1 From PGF a to lst 7.6 :_0.7 6.7 :_0.6 6.2 :_0.3 ovulation From onset of estrus 4.6 :_O.6 3.5 :_O.6 3.9 i 0.4 to lst ovulation From last ovulation 1.8 i 0.3 1.9 :_O.2 2.3 :_O.2 to end of estrus Duration of estrus 6.6 + 0.4 5.8 + 0.7 6.3 + 1.0 aValues are means :_SEM. bn = 11 Cn = 11 32 33 Treatment C mares received 3 mg of estradiol-178 on 2 consecutive days following PGF a and it seems likely the exogenous estradiol-173 2 was responsible for the earlier detection of behavioral estrus in these mares. The behavioral response to exogenous estrogens appears to vary with the reproductive state of the animal. The administration of 5 mg of estradiol-178 to mares with an active corpus luteum will not induce estrus. However, ovariectomized mares or anestrous mares respond to 1 mg of estradiol-17B and show behavioral estrus within 3 to 6 hours (Hughes and Stabenfeldt, 1977). The response observed in E2-17B treated mares in this experiment following induced luteolysis may be similar to mares in anestrus or ovariectomized mares, since low progesterone levels have also been reported for ovariectomized and anestrous mares. The lst ovulation in the treatment A mares was detected on the average 7.6 :_O.7 days after PGF administration or 4.6 i 0.6 days Zn from the onset of estrus. The signs of behavioral estrus ended 1.8 i 0.3 days following the last ovulation and the mean duration of estrus was 6.6 :_0.4 days for control mares (Table 3). Data for these control mares compare favorably with results reported by pre- vious researchers for mares receiving prostaglandin F (Douglas and 2a Ginther, 1972; Allen and Rowson, 1973; Noden et al., 1973, 1974; Oxender et al., l975a,b). The interval from PGFZa to let ovulation in treatment B mares averaged 6.7 :_O.6 days and, in treatment C, 6.2 :_0.3 days (Table 3). Although these intervals averaged nearly 1 day shorter than the controls, the differences were not statistically significant. There were no significant differences in the interval from last ovulation 34 to the end of estrus, or duration of estrus among the 3 different treatment groups. Previous work reported by Ginther and Wentworth (1974) showed that single injections (400 mg) of synthetic GnRH in pony mares on day 2 of estrus did not significantly alter follicular development or time of ovulation. Subsequent work by Garcia and Ginther (1975) tested the effects of several different doses and treatment regimes of GnRH including pretreatment with estradiol. In one experiment continuous intravenous infusion of GnRH sufficient to maintain an ovulatory LH concentration for 24 hours on day 2 of estrus did not alter ovulation time in the mare. Although pretreatment with estradiol-178 enhanced the LH response to a dose of GnRH, this treatment also failed to influence the time of ovulation. Oxender et al. (1977b) conducted 2 experiments utilizing PGFZa and GnRH in an attempt to predict or synchronize ovulation in mares. The lst experiment demonstrated that mares treated subcutaneously with 2 or 5 mg of GnRH 96 hours after PGFZa treatment had a 2.5-fold increase in serum LH; however, time of ovulation and duration of estrus were not significantly altered from controls. In a 2nd experiment, mares received GnRH (5 mg sc) daily (starting 96 hours after PGan) until ovulation occurred. Again the interval from PGFZa treatment to ovulation and duration of estrus were not significantly shortened nor synchronized. In contrast to the experiments with GnRH which failed to change the time of ovulation in mares, Irvine et a1. (1975) reported a shorter duration of estrus for GnRH treated mares. Both intravenous and subcutaneous treatments with GnRH on day 2 of estrus signifi- cantly shortened duration of estrus and time to ovulation; the 35 subcutaneous treatments were repeated daily until ovulation. The success of Irvine and his co-workers is not easily explained, since several authors have reported that GnRH treatment failed to change the time of ovulation in mares. However, 1 or 2 factors may account for the differences. One series of GnRH treatments was completed during April when the mean duration of estrus for the control mares was reported as 11.8 :_2.1 days, and this is, it appears, signifi- cantly longer than the mean duration of estrus reported for mares in June and July of 5.2 i 0.5 days (Noden et al., 1975). Secondly, Irvine et al. (1975) treated mares on day 2 of estrus rather than 96 hours after PGFZa and not all PGFZa-treated mares were in estrus at the time of GnRH treatment in the latter experiments. Thus, it appears the time that GnRH is used during estrus may also be a factor. A few researchers have treated estrous mares with GnRH to induce ovulation. Heinze and Klug (1975), in a clinical trial, reported that 38 of 39 mares ovulated a mature follicle within 48 hours following an injection of 4 mg GnRH during a period of estrus lasting more than 9 days. In contrast, only 11 of 16 nontreated control mares ovulated within the same period. Thus, GnRH appears to have the ability to induce ovulation in mares under certain circumstances. Serum Hormone Concentrations after Subcutaneous PGF 20—- Serum progesterone averaged 5.4 :_0.7 ng/ml in treatment A mares when 10 mg PGF -Tham salt was injected sc 7 days after the previous 2a ovulation (Tables 4 and 7). PrOgesterone decreased rapidly (p<.01) to 1.3 :_0.3 ng/ml within 24 hours and, by 48 hours, pro- gesterone concentrations averaged below 1 ng/ml, where they remained 36 Table 4. Serum hormone concentrations during the estrous cycle following treatment with PGFZG (10 mg sc) in 12 mares (control treatment A)a Serum Hormone Day Progesterone LH Estradiol ---------- (ng/m1)----------- -- (pg/m1) -- Day of PGF2(1 treatment 5.4 :_0.7 141 :_ 36 6.8 i 1.6 Two days after PGan 0.4 :_0.0 197 :_ 63 8.0 :_1.5 Onset of estrusb 0.2 i 0.0 298 :_ 61 8.6 :_l.4 One day before ovulation 0.2 :_0.0 1035 i 216 15.1 :_2.3 Ovulation detected 0.3 :_0.1 875 :_l64 6.3 :_0.1 End of estrus 2.5 + 0.3 1072 + 224 3.7 + 0.5 a Values are means :_SEM. bMares stood firmly when mounted by teaser stallion. until ovulation occurred (Figure 1). Following ovulation, progesterone increased (p<.05) to 2.5 :_0.3 ng/ml within 48 hours, indicating rapid luteinization of the corpus hemorrhagicum and signaling termination of estrus behavior. Progesterone levels continued to rise (p<.05) to a peak diestrus concentration of 6.2 :_0.7 ng/ml 7 days later. The relative changes in serum progesterone following the induced luteolysis were comparable to those reported by Allen and Rowson (1973), Noden et al. (1974) and Oxender et al. (1975a,b). The LH changes during estrus in these mares were similar to those previously reported by Pattison et al. (1974), Whitmore et al. (1973), Noden et a1. (1974) and Oxender et al. (l975b). Serum LH averaged 141 :_36 ng/ml when PGF was given during mid-diestrus 2a 37 mcazonm .OH .0: meme ca Aom m6 oav 8N .muson vm canvas mcoflumas>o m mom mafizoaaom wcouoummmoum Esnmw .H musmem 38 out: 20m... $3 N. o. m w v N o _ n _ . _ a F 1L _ _ malailaxx /. slam Pu: owe me SE mm 96 \~\\o 6:358 \. i \ o\o o yo Name‘s 1" N (D (um/On) anoaalsaeoaa Q 2 Figure l 39 (Tables 4 and 7). The LH concentration increased to 298 :_61 ng/ml by the onset of estrus and continued to increase throughout estrus to 875 :_164 ng/ml by the time of ovulation, with a peak of 1072 i 224 ng/ml by the end of estrus. Estradiol concentrations for mares during the control estrus increased to a peak prior to ovulation. On the day of PGFZa treat- ment, estradiol averaged 6.8 :_1.6 pg/ml and had increased slightly to 8.6 :_1.4 pg/ml at estrus onset (Tables 4 and 7). During estrus. estradiol levels increased (p<.05) to 15.1 :_2.3 pg/ml 1 day prior to ovulation and then rapidly decreased by the day of ovulation to 6.3 i_0.l pg/ml (p<.05). The levels continued to decrease, averaging 3.7 :_0.5 pg/ml by the end of estrus (p<.05). These results are in agreement with estradiol levels reported from the same laboratory by Noden et al. (1975) for mares during normal estrous cycles and Oxender et al.(1975b) for mares treated with prostaglandin F2“. Serum Hormone Concentrations after PGFZG and GnRH Serum progesterone values in treatment B mares followed a similar pattern to that of the control mares (Tables 5 and 7). At the time of PGF20L administration, progesterone concentration averaged 5.3 i 0.5 ng/ml and decreased to 0.3 i 0.0 ng/ml within 48 hours. The serum concentration remained less than 1 ng/ml throughout estrus, increasing to 2.0 i_0.3 ng/ml by the end of estrus. Diestrus values averaged 4.5 :_0.6 ng/ml by the 7th day postovulation. Serum LH averaged 176 i_46 ng/ml on the day of PGF treatment 2a and had increased slightly to 249 :_44 ng/ml 2 days after PGFZa (Tables 5 and 7). These values were not different than those in the control mares, nor would they be expected to be different since GnRH 40 Table 5. Serum hormone concentrations during the estrous cycle following treatment with PGan (10 mg sc) and GnRH (200 ug/injection) in 12 mares (treatment B)a Serum Hormone Day Progesterone LH Estradiol ---------- (ng/ml)-—-------- --(pg/ml)-- Day of PGFZa treatment 5.3 :_0.5 176 :_ 46 8.6 :_1.7 Two days after PGFZa 0.3 :_0.0 249 :_ 44 10.7 :_2.7 Onset of estrus 0.3 :_0.0 389 :_ 71 12.0 :_2.8 One day before ovulation 0.5 :_0.3 563 :_104 16.8 :_3.2 Ovulation detected 0.3 :_0.1 549 :_ 92 6.0 :_0.9 End of estrus 2.0 + 0.3 832 + 195 4.6 + 2.7 a Values are means :_SEM. treatment was not begun until 3 days (72 hours) after PGF injection. 20 Onset of estrus varied in its time relation to PGFZa treatment but averaged 3.2 :_0.3 days after PGFZa (Table 3). Serum LH at estrus onset averaged 389 :_71 ng/ml, which was not significantly different than the 298 :.61 ng/ml for control mares. The treatment effect of GnRH on serum LH at estrus onset was not evident when comparing data from Tables 4 and 5, since several mares would not yet have received their GnRH treatments. However, in comparing treatments A and B in Table 7, the GnRH treated mares showed a significant (p<.01) 2- to 3-fold increase in LH levels versus little or no change in the con- trols. An average concentration of 729 i 80 ng/ml was obtained 8 hours following the let GnRH injection, and levels remained elevated throughout the GnRH treatment period. 41 It had not been anticipated that 3 injections of GnRH (200 pg/ injection) at 8-hour intervals would cause elevated serum LH for longer than 24 to 48 hours following treatment. And indeed, the serum LH was 563 :_104 ng/ml 1 day before ovulation and 549 :_92 ng/ml on the day of ovulation (which was significantly less than control values [p<.01]). Preliminary work done by Oxender et al. (l977b)lunishown that serum LH was increased greater than 3-fold within 60 minutes after 5 mg of GnRH was given as a single dose or multiple doses repeated at 24—hour intervals. The serum LH had decreased within 8 hours after each GnRH injection and was similar to the control group by 24 hours (Figure 2). In those experiments ovulation was not hastened by the GnRH treatments. Based on this work, the present experiment was designed to test the hypothesis that repeated injections of GnRH given before serum LH levels had fallen back to baseline levels might produce an additive effect similar to that presented in Figure 3. Creating a sustained high serum LH level over a 24-hour period coupled with increasing endogenous LH release during estrus might have provided the stimulus for an earlier ovulation and a shortened period of estrus. Success was achieved in elevating serum LH 2- to 3-fold over the 24-hour duration of treatment and the interval from PGFZa to let ovulation was approximately 1 day shorter for mares in this group (Table 3). The pattern of estradiol-178 serum levels in treatment B mares during estrus closely resembled that of the control mares. Serum levels averaging 8.6 :_l.7 pg/ml at the time of PGFZa treatment. increased to a peak of 16.8 :_3.2 pg/ml 24 hours prior to ovulation and then fell rapidly to 6.0 i 0.9 pg/ml by the time ovulation 42 .Abnemav .Hm um Hmosmxo Eoum ooummmo< .mam>uouca usoenvm um mmco mo mmmoo mE m mHmHuHsE cm>am manna ca m4 Essen mmoum>m .m masons GnRH GnRH GnRH 43 l050 - (IW/l-BEII 831 bu)H'l Figure 2 ar:II::-____ -> "“' ----- a I 1 l 1 I 1 8 S 8 8 8 8 m h to <- n - 96 24 HOURS 44 .mHm>kucH Moonlm um Acoauooflsfl\mn OONV mmco mo mcofluooflCH m OCHBOHHOM omuommxo o>uso ma Essen Hecauonuom>m .m ousmflm 45 CON 00w 00m 000. CON. 1,25 “.0 20:.0u33 NAACJD! 92.30.39; 11.. Inmmm 410....”1hgr 2. 00¢. (Hull-82H 19") 6“) H'I Figure 3 46 occurred. The estradiol level had decreased to 4.6 :_2.7 pq/ml by the end of estrus (Table 5). Serum Hormone Concentrations after PGcm’ Eq-l78 and GnRH Serum concentrations of progesterone for mares in treatment C did not differ significantly from mares in treatments A and B (Tables 6 and 7). The LH curve followed the trend of a gradual increase, Table 6. Serum hormone concentrations during the estrous cycle following treatment with PGFZG (10 mg sc), E2-178 (3 mg sc) and GnRH (200 pg/injection) in 12 mares (treatment C)a Serum Hormone Day Progesterone LH Estradiol ---------- (ng/ml)-----—---- --(pg/ml)-- Day of PGFZa treatment 4.7 :_0.4 141 :_ 25 5.2 :_1.0 Two days after PGFZa 0.4 :_0.1 189 :_ 28 7.2 i_l.5 Onset of estrus 0.3 :_0.1 223 :_ 30 7.6 :_l.6 One day before ovulation 0.1 :_0.0 268 :_ 41 13.4 :_l.9 Ovulation detected 0.2 :_0.0 369 i_ 74 6.5 :_1.4 End of estrus 2.4 + 0.1 569 + 112 5.9 + 0.7 aValues are means : SEM. peaking near the time of ovulation, similar to treatments A and B (Table 6). Concentrations of LH on the day of PGFZc treatment averaged 141 i_25 ng/ml, which was not different from treatments A and B, which had levels of 141 :_36 ng/ml and 176 :_46 ng/ml, respec- tively. Three injections of GnRH following pretreatment with E2-17B induced an approximate 2-fold increase in serum LH levels in the 47 Table 7. Serum hormone concentrations in the 3 treatment groups following PGF a 2a Serum Hormone Treatment Day Progesterone LH Estradiol --------- (ng/m1)------—--- --(pg/ml)-- Treatment A 7 5.4 i 0.7 141 i. 36 6.8 i 1.6 8 _+_ 3 119 i 22 6.1 i 2.0 9 0.4 :_ . 197 i. 63 8.0 i. . 10 (1630 hrs) 0.3 + 0.1 183 :_ 29 8.5 i 0.9 10 (2400) 0.3 :_0.0 239 i. 35 8.2 i 11 (0800) 0.5 :_0. 291 :_101 13.4 :_4. 11 (1630) 0.3 i_0.1 272 :_ 63 14.1 I. . Treatment B 7 4.2 i_0.4 176 :_ 46 8.6 :_1.7 8 1.3 :_0.3 182 :_ 29 7.9 :_l. 9 0.3 :_ . 249 :_ 44 10.7 i 2.7 10 (1630) 0.2 :_0.1 352 :_ 72 6.7 :_l.l 10 (2400) 0.2 :_0.1 729 :_ 80 7.6 :_l.5 11 (0800) 0.3 :_ . 868 :_112 10.2 t 11 (1630) 0.2 i 0.1 615 :_127 9.2 :_2.0 Treatment C 7 6.0 :_0.6 141 :_ 25 4.8 i 8 1.3 :_0 128 :_ 20 4.8 :_ .0 9 0.3 :_0.1 189 :_ 28 6.1 :_ .6 10 (1630) 0.5 :_0.3 286 :_ 42 5.8 :_l. 10 (2400) 0.4 :_0.1 462 i_ 97 12.2 :_2.3 11 (0800) :_0.1 417 :_ 74 9.1 :_1.4 11 (1630) 0.4 :_ . 355 :_ 79 9.6 :_ . a Values are means :_SEM. 48 treatment C mares compared to control (treatment A) mares (Table 7). The LH response to GnRH appeared less in treatment C versus treat— ment B. One day before ovulation, the serum LH level was 268 :_41 ng/ml, which was significantly less (p<.01) than treatment A (1035 :_216 ng/ml) or treatment B (563 :_104 ng/ml). The lower LH concen- trations continued for this group through ovulation, averaging 569 i 112 ng/ml at the end of estrus compared to 1,072 i_224 ng/ml for control mares and 832 i 195 ng/ml for mares in treatment B. Estradiol levels were not different in the treatment C mares compared with the other 2 treatment groups (Tables 6 and 7). A level of 5.2 :_l.0 pg/ml at the time of PGan administration increased to 13.4 pg/ml 1 day prior to ovulation, then decreased to 6.5 :_1.4 pg/ml when ovulation was detected. The serum concentrations of estradiol recorded in Tables 6 and 7 do not reflect the effects of the exogenous administration of 3 mg E -l7B on 2 consecutive days 2 following PGF injection. Perhaps this is due to rapid transforma- 2a tion by the liver or excretion. The rationale for pretreating the mares with E2-17B is based on data suggesting that estrogen plays a significant role in regulating gonadotropin secretion by the pituitary in several species. Reeves et al. (1971b) reported that pretreatment of anestrous ewes with estradiol facilitated an increased response of LH release induced by GnRH. It has also been demonstrated in the cow and the ewe that exogenous administration of estradiol would elicit a significant rise in plasma LH in the absence of high levels of progesterone. Progesterone has been shown to have a negative feedback effect on plasma LH in these species (Hansel and Echternkamp, 1972). 49 Garcia and Ginther (1976a) used ovariectomized mares to demon- strate that injecting 1 mg estradiol daily for 15 days caused plasma LH to also increase on a daily basis. Daily injections of 100 mg progesterone for 15 days caused a daily decrease in plasma LH, sug- gesting a similar regulatory role of estradiol and progesterone on gonadotropin secretion in mares as reported for the cow and ewe. In addition, previous work by Garcia and Ginther (1975) had shown that treatment of pony mares on day 2 of estrus with E2-17B alone increased plasma LH concentration after 24 hours and that the LH response to a dose of GnRH given on day 3 was enhanced when compared to mares not pretreated with estradiol-178. In contrast, the results of this experiment do not demonstrate an augmented LH release following E -178 pretreatment. In fact, 2 there may have been a negative effect on serum LH by the time of ovulation. The results might be related to the fact that the mares in this study were not treated on days 2 and 3 of estrus as in the above study, but rather l to 2 days prior to estrus. Another study by Keye and Jaffe (1974) suggests that the positive feedback effect of estradiol may be both dose and time related. They showed a decreased gonadotropin response to administered GnRH in women in the presence of midcycle levels of estradiol which was not consistent with their hypothesis that the midcycle surge of gonadotropins was due to increased sensitivity of the pituitary to GnRH brought about by increasing levels of estradiol. In conclusion, it appears reasonable to assume that a time and dose relationship exists in the regulation of gonadotropin secretion by estradiol. What that relationship is awaits further elucidation in the mare. 50 Effects of Ovarian Follicle Size on the Interval to First Ovulation During this study, rectal palpation of mares in diestrus often detected large follicles (320 mm diameter) that seemed to persist from the preceding estrus or in other mares follicles appeared to develop rapidly during the early diestrus period. It appears that large follicles present at the time of PGF treatment may ovulate 20!. within 48 hours (Table 8). Table 8. The influence of ovarian follicle size at the time of PGF administration on the interval in days to lst ovulation in 12 mares Size of largest ovarian Treatment follicle (mm) A B C ------------------- (days)-------------------- 10-19 8.8 i 0.7a (5)b 5.8 i 0.5 (8) 6.7 i 0.8 (7) 20-29 6.7 i_0.9 (3) 5.0 :_0.0 (2) 4.5 i_0.5 (2) 230 4.5 i 0.5 (2) 6.0 + 0.0 (l) 3.3 i 1.9 (3) aMean interval in days :_SEM. bNumber of mares in each category. Preliminary observation of the palpation records caused concern that mares having a large palpable follicle when given PGF 0 had a 2 shorter interval to ovulation regardless of the treatment group. Four mares ovulated within 48 hours of PGFZu treatment, and in these cases the interval to ovulation had to be discarded in favor of estimated values determined from the Latin square for purposes of statistical analysis of data. 51 When mares were grouped according to follicle size within each treatment group, mares in control treatment A showed a significantly shorter interval to ovulation (p<.05) when they had a follicle in the category 20-29 mm or 30 mm or greater (Table 8) at the time of PGFZa treatment. The ovulation interval in days was 6.7 :_0.9 and 4.5 i 0.5, respectively, for the above follicular size groups, in contrast to the 8.8 :_0.7-day interval for mares having follicles 10-19 mm in diameter. No significant differences between follicular size categories were observed in treatment B mares. Although not statis- tically significant because of small numbers of animals, mares in treatment C having a follicle 30 mm or larger had an ovulation interval (3.3 :_1.9 days) approximately one-half as long as the 6.7 i_0.8-day interval recorded for mares having follicles 10-19 mm in diameter. Because the number of mares in each category was small, a meaningful interpretation of the results was not possible. Further study of the influence of follicular size at the time of treatment on the interval to ovulation is needed because this is an area of clinical importance for purposes of predicting the time of ovulation in mares. Exogenous PGF has been reported to cause a surge of LH in 2a sheep (Carlson et al., 1973), cows (Hafs et al., 1975) and bulls (Kiser et al., 1976). The mechanism of action of the PGFZa is not known. Recent work by Noden et al. (1978) demonstrated a transient 1.5- to 2.5-fold increase in serum LH 30 to 60 minutes following PGFZu injection. The levels then declined toward pretreatment values during the next 3 hours. The reported transient increases in blood LH after PGan treatment in mares, ewes, cows and bulls does not appear to be related to the endogenous LH surge occurring near 52 the time of ovulation. Whether PGan is exerting a direct or indirect (Drouin et al., 1976) effect on pituitary LH release is not known. How- ever, exogenous progesterone prevented the LH increase observed at 3 to 9 hours after PGFZa treatment in cows (Hafs et al., 1975), suggesting that the increase may be due to decreased progesterone and not to a direct effect on the hypothalamic—pituitary axis. Additional experimental evidence in laboratory animals indicates the prostaglandins may be involved in the ovulation mechanism. Labhsetwar (1973) reported the occurrence of ovulation after injec- tion of PGFZa in pregnant hamsters, and Graafian follicles in rabbits were found to contain PCB and PGF, whose levels significantly increased as ovulation approached (Yang et al., 1974). Data also demonstrated that levels of PGF in the rat ovary significantly Za increased before ovulation (Iseaka’et al., 1975). In addition, Grinwich et al. (1972) blocked ovulation in rabbits using indo- methacin, a prostaglandin inhibitor. Thus, it appears that prosta- glandins in plasma and the ovary may be involved with follicular maturation and the process of ovulation in some species. In mares, exogenous PGFZa treatment caused a transient surge in serum LH and a limited number of mares appeared to ovulate (based on rectal palpa— . This ovulation tion) within 48 hours following injection of PGFZa may be triggered by the direct effect of exogenous PGFZa on the follicle or the transient LH surge. However, ovulation does not occur within 48 hours in mares with small follicles at the time of PGF treatment. 28 53 Effect of Multiple Ovulations on Estrus and on the Luteolytic Response to PGF2a_ There appeared to be a frequent occurrence of multiple ovulations during estrous cycles in this study. In 17 of the 36 (47%) estrous cycles studied, multiple ovulations were observed. Two ovulations were observed in most cases; however, 3 mares had 3 ovulations during an estrous cycle. There was no significant difference in the number of multiple ovulations among the 3 treatment groups (Table 9). Table 9. The number of multiple ovulations in 12 mares during 36 total estrous cycles categorized according to treatment group Number of Mares with Treatment 2 or More Ovulations A 4 B 7 C 6 It is difficult to compare results on frequency of ovulation with the previously reported literature because of the different experi- mental treatments. However, multiple ovulations are known to occur, and reported frequencies were 3.8% (Andrews and McKenzie, 1941), 25% (Stabenfeldt et al., 1972) and 42% (Warszawsky et al., 1972). Stabenfeldt et al. (1972) reported that mares with 2 ovulations averaged 1 day longer in estrus (6.0 :_1.2 days) than mares with 1 ovulation (5.1 :_1.9 days), although the results were not statistically significant. A similar pattern was noted in this study when ovula- tions and length of estrus were compared within treatment groups 54 (Table 10). Mares having 2 ovulations during estrus in treatment B had an estrus duration of 7.4 :_1.3 days, which was longer than mares having only 1 ovulation (4.7 :_0.5 days), and this difference approached (p<0.10) statistical significance. Likewise, mares in Table 10. The influence of the number of ovulations during an estrus on the duration of estrus in 12 mares No. of ovulations/ Treatment estrus period A B C ------------ (length of estrus [days])------------ 1 6.7 i 0.5“1 (7)10 4.7 i 0.5 (7) 5.1 i 0.8 (8) 2 7.0 i 0.7 (5) 7.4 i 1.3 (5) 8.8 i 2.3 (4) aMean 1 SEM. bNumber of mares in each treatment group. treatment C with double ovulations also had a longer estrus (8.8 i 2.3 days) than mares with single ovulations (5.1 :_0.8 days). The influence of multiple ovulations on the length of estrus should not be overlooked when comparing treatment effects among groups. The effects of double or multiple ovulations may also pro- vide a partial explanation for the variation in estrus duration among mares. Although not analyzed statistically, the interval between ovulations occurring during estrus seemed to correlate with the length of estrus (compare Figures 1 and 4). Mare no. 10 (Figure 1) shows the typical progesterone curve for a mare having 2 ovulations during estrus within a 24-hour interval. Progesterone decreased (<1 ng/ml) approximately 24 hours following PGFZa induced 55 luteolysis and remained below 1 ng/ml throughout estrus, then increased again 24 hours after the 2nd ovulation. Stabenfeldt et al. (1972) also reported that mares ovulating 2 follicles usually did not have a progesterone rise until 24 hours after the 2nd ovula- tion. Mare no. 14 (Figure 4) received the same treatment as mare no. 10; however, the progesterone pattern was unusual because this .mare ovulated a 45 mm follicle shortly after PGan treatment. Following the lst ovulation the progesterone level increased (>1 ng/ml) 3 days later, then decreased again. During this time the mare was not observed in estrus. Following the 2nd progesterone decrease to less than 1 ng/ml, the mare began showing estrus behavior which lasted 8 days. During estrus 2 more ovulations were detected in this mare, with progesterone finally increasing 24 hours after the last ovulation. After the lst ovulation, functional luteal tissue apparently formed for a short period causing the transient rise in the progesterone concentration and the suppression of estrus behavior. A similar phenomenon was also reported by Stabenfeldt et al. (1972). Effect of Diestrus Ovulations on the Luteolytic Response to PGFZG Administration Ovulations were also detected during the diestrus period in mares on this study. However, following these ovulations no increase in progesterone levels was observed. A typical progesterone curve for a mare having a diestrus ovulation is depicted in Figure 5. Whether the diestrus ovulations are true ovulations or a form of follicular atresia remains debatable. Follicular atresia may be occurring without formation of luteal tissue, or they may be soft palpable corpora lutea that rapidly regress in size. 56 .msuumm cm mafiuso chfiumHs>o m nuflz va .0: meme cw Aom ms OHV ammom mcw3oHHom wconmummmoum Esuwm .v musmfim 57 ammo... :2: 93 N ¢ o. m. s u. o. o o e N o 1 r . _ Tulqldll elm). . _ . . _ \T . // \o\«/¢ . i first. . 7 anEma .. 35mm macaw o 283:3 25mm 29536 . /fl owkel 8E mm ozm CD (nun/6") anoaalsaeoad 0 «35$. 24 25:5 mzo_._.<.5>o m It; was). < z. .3 .25: yo “8.. 92.26.39. mzommtwmooma 22mm -- 0. Figure 4 58 .coflumHs>o manummfic m can monumm uuonm m zuflz coflumHs>o xaumm cm mcfl3ocm .m .oc wumE cH now me oav ammom mafi3oHHom mcoumumwmoum Edumm .m mhsmflm 59 out on. 22.... 9:3 m. E N. o. m m c N M o .11._._q_.4.l_\o\.7 l 5:53 matmm btm\.w\. caomQ—k .1 \ boauoaoa \0 $2230 1 o\0 o o o\ 1 lo Ilololo\» lolo\ u mm on placate 32335 J zo_._.<.5>o mamhmma < 024 mar—bu Pmozm .zo_._.<..=..>o >4m