.. . n. \‘Iun‘ )1 .. .-.lll‘od-J ‘ .. . ...mnmuu r. ., .. 4‘ o [AnLu . ~ I. ‘ ..It .Iv‘ . 'nll w. 1 w .5‘ A )— 83 ‘11 $4 4. _ :1. . . .. J .. .- ... .l. ...r a, ll'lviOrv‘IQ‘cl "c‘i.l. -7 1"-” .I‘ u.."-l..‘1.io112- 3.5.111- , - . fr... .f.£o.,i..io?. . . .A ovl‘ullnc' ob]?! |.....1 :0 ..O’Lyvv? -. '0’]: till! .I . 3...... I.\1....III.F.I}.IJ:. ...!y i..- .. .‘IAl’otflOf.,-lu.n‘t..vnv: ,3I‘7.‘| ll“ .l... .J...K!!t‘¢.:3£.-3‘93}- lfltJu.I..I.. 1-1.8113}. I» 1...-.4‘ ....J; -1192... --.. --Ig‘v-’ ... .17! ...! 1"}.3392 to -15") o I]. It’llfvlc'vptl debit? «'5‘; . . l . . ; ..1 1%,??? lahu'oflqtlljeil {Suffix}. 1).-.. 1! - 2! gig! .. ‘ ‘.‘ ' .. VA ~ . . . 4 fa.” o v A ..cthuhui ..I.:Ivtv?.l.e..nu. a} >\‘f.v'7\ Alli". .. 'l 13”! .‘l‘, ..11’in).§ It}? . ‘ O!J').v]cr.l I . ,I7b¢.. LU- . |\l thstf NIVERSITY Ll Iiiili’lit‘l’iliii‘lu‘illn Inmllmfll'l‘ifill‘li a (W?) 7) 3129301572 8771 This is to certify that the dissertation entitled BASAL AND ESTRADIOL-INDUCED RELEASE OF GONADOTROPINS IN DAIRY COWS WITH NATURALLY-OCCURRING OR ARTIFICIALLY-INDUCED OVARIAN CYSTS presented by KENT ROSWELL REFSAL has been accepted towards fulfillment of the requirements for Ph. D. degreein Animal Science a jor professor MSU is an Affirmative Action/Eq ual Opportunity Institution 0- 12771 LIBRARY Michigan State University PLACE lN RETURN BOX to remove We checkout from your record. To AVOID FINES return on or More date due. DATE DUE DATE DUE DATE DUE BASAL AND ESTRADIOL-INDUCED RELEASE OF GONADOTROPINS IN DAIRY COWS WITH NATURALLY-OCCURRING OR ARTIFICIALLY-INDUCED OVARIAN CYSTS BY Kent Roswell Refsal A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Animal Science 1986 ABSTRACT BASAL AND ESTRADIOL-INDUCED RELEASE OF GONADOTROPINS IN DAIRY COWS WITH NATURALLY-OCCURRING OR ARTIFICIALLY-INDUCED OVARIAN CYSTS BY Kent Roswell Refsal Two experiments were done to characterize gonadotropin release in dairy cows having ovarian cysts. In the first experiment, cows with naturally-occurring ovarian cysts were sampled to determine 1) rela- tionships between endogenous circulating estradiol or progesterone con- centrations and basal gonadotropin release and 2) the ability of the same cows to release LH and FSH after exogenous estradiol administra- tion. Cows with high endogenous estradiol had the highest basal con- centrations of LB, but there were no other significant relationships between frequency and amplitude of pulsatile gonadotropin release and endogenous sex steroid concentrations. After exogenous estradiol, a preovulatory surge of gonadotrOpins occured in only one of twelve cows. All cows released LH and FSH after being given exogenous GnRH. In the second experiment, adult nonlactating dairy cows in late diestrus were given exogenous estradiol or repeated injections of ACTH to induce formation of ovarian cysts. Changes in serum LH after exoge- nous estradiol suggested that ovarian cysts may be formed if the pre- ovulatory surge of LH occurs prematurely in relation to luteal regression. Changes in LH in ACTH-treated cows suggested that ovarian Kent Roswell Refsal cysts may also be formed when there is inhibition of the preovulatory surge of LH after luteal regression. Profiles of gonadotropin release in cows with artificially-induced ovarian cysts were similar to those of cows with the naturally-occurring condition. It was concluded from these experiments that 1) asynchrony or lack of a preovulatory surge of LH can cause formation of ovarian cysts in dairy cows, 2) the cystic condition may be maintained by refractoriness of the hypothalamo- pituitary axis to the positive feedback effects of estradiol on release of gonadotropins, and 3) dairy cows with artificially-induced ovarian cysts appear to be a suitable endocrine model for the naturally- occurring condition. ACKNOWLEDGEMENTS This dissertation could not have been a reality without the assistance of many people. My pursuit of an academic career in veteri- nary medicine was inspired by the example of Dr. Brad Seguin and the faculty of the University of Minnesota Theriogenology Section. The much appreciated efforts of Drs. E. Mather, H. Hafs, and M. Hogberg made it possible for my program to be administered in the Department of Animal Science. I extend many thanks to the members of my graduate committee, Drs. Clyde Anderson, W. Richard Dukelow, Ed Mather, and Ray Nachreiner, for counsel and positive support throughout my stay at Michigan State. Special thanks to Clyde Anderson for sharing his expertise on statisti— cal analysis and computers. Special thanks also to my major professor, Ray Nachreiner, for his great patience and confidence through times of frustration in my program. The cooperation, hospitality, and camaraderie extended by the Reproductive Physiology Group is greatly appreciated, notably the consul of Drs. Ed Convey and Roy Fogwell. Thanks also to the assistance from the Staff of the Endocrine Diagnostic Laboratory. Dr. Julio Jarrin- Maldonado and his family were of great assistance in collection of the data. Lastly, I thank my parents and especially my wife, Judy, for their concern and support throughout my program. ii TABLE OF CONTENTS LIST OF TABLES O O O O O O O O O O O O O O O O O O O O O O O O O O C O O O O O O O O I O O O O O O O O O O O O O 0 LIST OF FIGURES O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O ABBREVIATIONS O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 0 INTRODUCTION ......OOOOOOO............00............OOOOOOOOOOOOO REVIEW OF LITERATURE Interaction of the Hypothalamo-Pituitary- Ovarian Axis in the Control of Gonadotropin Release Throughout the Estrous Cycle ....................... Luteal Phase ........................................... Pre-Surge Period ....................................... Post-Surge Period ...................................... Changes in Gonadotropins and Sex Steroids in the Postpartum Period in Cattle ............................ Gonadotropins .......................................... Estradiol and Progesterone ............................. Ovarian Cysts in Bovine Reproduction ....................... Clinical Definition .................................... Etiology, Incidence and Predisposing Factors ........... Anatomy and Steroids in Fluid of Ovarian Cysts ......... Peripheral Sex Steroid and Gonadotropin Concentrations ....................................... Therapy of Ovarian Cysts ............................... EXPERIMENT I - Gonadotropin Secretion in Dairy Cows With Ovarian Cysts: Pulsatile Release Patterns and Response to Estradiol Challenge ................................. BaCkground to EXPERIMENT I 0.0.0000.........OOOOOOOOOO...... Materials and Methods for EXPERIMENT I ..................... Animals ................................................ Blood Sampling and Hormonal Treatments ................. Hormone Assays ......................................... Progesterone Radioimmunoassay .......................... Definition of Variables and Statistical Analyses ....... iii Page vi vii viii \lO‘bw a) 12 12 13 14 16 18 20 21 24 24 24 25 26 28 Results of EXPERIMENT I .................................... Ovarian Changes and Serum P4 Prior to Intensive Sampling ................................... Basal Steroids and Gonadotropins ....................... GonadotrOpin Response to Exogenous E2 .................. Gonadotropin Response to Exogenous GnRH ................ DiSC“881on0f EXPERIMENTI ......OOOOOOOOOOOOOOOOO0000...... EXPERIMENT II — Formation of Ovarian Cysts in Dairy Cows After Exogenous Estradiol or ACTH and Endocrine Profiles in Cows With Experimentally-Induced Ovarian Cysts ............... BaCkground to EXPERMENT II ......OOOOOOOOOOOOOOO0.0.0.0.... Materials and Methods for EXPERIMENT II .................... I. Induction of Ovarian Cysts ......................... Animals and Treatments ........................... Blood Collection and Rectal Examination Schedules ...................................... Hormone Assays ................................... Definition of LH Release and Statistical Analysis 0......O.........OOOOOOOOOOOOO00.0.0... II. Endocrine Profiles of Cows With Experimentally- Induced Ovarian Cysts .............................. Blood Sampling Schedule .......................... Hormone Assays ................................... Definitions of Gonadotropin Release .............. Results of EXPERIMENT II 1. Induction of Ovarian Cysts ......................... Control Group .................................... Ezlo-Treated Group ............................... ACTH-Treated Group ............................... II. Gross Ovarian Changes and Endocrine Profiles of Cows With Experimentally-Induced Ovarian Cysts .............................................. Ovarian Structures ............................... Basal Gonadotropin and Sex Steroid Secretion and Response to Exogenous E2 Discussion of EXPERIMENT II ................................ GENERAL DISCUSSION 0.0.000.........OOOOOOOOOOOOOOOOOI.....OOOOOOO iv Page 30 30 30 32 35 43 49 50 51 51 51 52 53 53 53 53 54 54 55 55 58 61 66 66 67 71 77 —‘_r SUMMARY AND CONCLUSIONS ......OOOOII.0...........OOOOOOOOOOOOOOOO APPENDIX - Validation of Commercially-Available Radioimmunoassay Kits for Measuring Estradiol-17b and Testosterone in Bovine, Canine, and Equine Serum ............ REFERENCES 0.00.00.00.00...000...........0.........OOOOOOOOOOOOOO Page 83 85 102 Table LIST OF TABLES Basal serum sex_steroid, gonadotrOpin, and cortisol data (x.1 SD) from dairy cows with naturally-occurring cystic ovarian follicles .............. Serum sex steroid and gonadotropin release in 4 adult dairy cows with experimentally-induced ovarian CYBtS ......OOOOOOOOOOOOOOOOOOOO0.00.00.00.00..0... Influence of 50% dilution of both antibody and radioligand on standard curves and serum values in a commercial radioimmunoassay for measure- ment of estradiol-17b ..................................... Effects of species and serum volume on recovery of 3H-estradiol-17a after extraction with 10 ml freshly-opened diethyl ether .............................. Effects of species and serum volume on recovery 3H-testosterone after extraction with 5 ml of diethYJ. Ether I.O......0............OOOOOOOOOIOOOOO0.0.0.00 Percent crossreactivity at 50% of total binding capacity of various steroids with antibody from commercially-available estradiol and testosterone radioimmunoassays ......................................... Accuracy of measuring unknown amounts of estradiol-17b when added to serum prior to ether extraction ............. Accuracy of measuring known amounts of testosterone when added to serum prior to ether extraction ............. Precision of estradiol and testosterone radio- immunoassays for bovine, canine and equine serum .......... vi Page 31 68 89 91 92 93 94 94 95 LIST OF FIGURES Figure Page 1. Pulsatile release of LH and FSH in adult dairy cows with ovarian cysts (H - Holstein, G - Guernsey’ ......OOOOOOOOOOOO0.00.00.........OOOOOOOOO...... 34 2. Serum LH after exogenous estradiol-17b in 12 dairy cows with ovarian cysts ............................. 37 3. Serum FSH after exogenous estradiol-17b in 12 dairy cows with ovarian cysts ............................. 39 4. Mean 1 SEM serum LH and FSH in 12 dairy cows with naturally-occurring ovarian cysts after administration of 100 ug GnRH in 0.15M NaCl ............... 42 5. Temporal changes in serum LH, progesterone and estradiol-17b in adult Holstein cows given 3.0 ml safflower seed oil on Day 0 ............................ 57 6. Temporal changes in serum LH, progesterone and estradiol-17b in adult dairy cows given 10 mg estradiol-17b in safflower seed oil on Day 0 .............. 60 7. Changes in serum cortisol in nonlactating adult dairy cows after administration of oil (controls) or estradiol-17b at 0 hr, or 100 IU ACTH gel given at 12 hr intervals ........................................ 63 8. Temporal changes in serum LH, estradiol and progesterone in nonlactating adult Holstein cows given 100 IU ACTH gel at 0 hr and repeated at 12 hr intervals for 10 days .................................. 65 9. Pulsatile release of LH and FSH and response to exogenous estradiol (E2) in 4 adult cows with experimentally-induced ovarian cysts ...................... 70 10. Inhibition curves for estradiol standard solutions and samples reconstituted after diethylether extraction of different volumes of bovine, canine and equine serum pools .................................... 97 11. Inhibition curves for testosterone standard solutions and samples reconstituted after diethyl- ether extraction of different volumes of bovine, canine and equine serum pools ............................. 99 vii cm OVX LH FSH ACTH PGFZa 1M 2210 IU RIA KEY TO ABBREVIATION centimeter(s) ovariectomized luteinizing hormone follicle-stimulating hormone gonadotropin-releasing hormone corpus luteum progesterone estradiol-17b picogram(s) milliliter(s) nanogram(s) hour(s) minute(s) millimeter(s) human chorionic gonadotrOpin degrees centigrade gravitational force milligram(s) microgram(s) molar concentration standard deviation replicate(s) probability standard error of the mean adrenocorticotropin prostaglandin FZalpha intramuscular 10 mg estradiol-17b-treatment international unit(s) radioimmunoassay viii INTRODUCTION Ovarian cysts are one of the most commonly diagnosed infertility problems in dairy cattle, detected in 5-20% of reproductive cycles. The condition is characterized by the presence of an ovarian follicular structure(s) usually 2 cm or more in diameter that may persist for days to weeks with the absence of a corpus luteum. In affected cows, repro- ductive efficiency is reduced primarily by an increase in the number of days open, despite therapeutic measures. Clinical diagnosis is based on identifying the presence of an ovarian cyst. The diagnostician usually has no further insight to the variability in histological anatomy and endocrine changes found with this condition. The cystic condition is not static, as affected cows may spontaneously resume estrous cycles or an ovarian cyst may regress with concomitant formation of another. Approximately 50% or more of ovarian cysts are detected within the first 60 days postpartum, the time when estrous cycles are reinitiated. Clinical and experimental observations suggest that a critical period for cyst formation occurs between luteal regression and ovulation in cycling cows. Hypotheses concerning a lack of or asynchrony of gona- dotropin release accompanying follicular growth have been proposed as causes of ovarian cysts. At present, data on basal gonadotropin secre- tion in cows with ovarian cysts is limited. The theme of this dissertation was to characterize gonadotropin secretion in dairy cows with ovarian cysts. Two interrelated experi- ments were conducted. The objectives of the first experiment were to determine relationships between basal sex steroid concentrations and gonadotrOpin release in cows with naturally-occurring ovarian cysts and assess gonadotropin response to estradiol challenge in the same animals. The second experiment also had two objectives. The first was to deter- mine if ovarian cysts could be artificially induced in dairy cows with exogenous estradiol or ACTH treatment and record accompanying changes in LH, estradiol, and progesterone. The second objective was to characterize relationships between sex steroid concentrations and gona- dotropin release in cows with artificially-induced ovarian cysts. REVIEW OF LITERATURE Ovarian cysts have been recognized as a cause of infertility in dairy cattle for over 50 years. Several extensive reviews have been written on the topic (Roberts, 1971; Garverick, 1982; Eyestone and Ax, 1984). This review focuses on endocrine changes associated with this condition and discusses the clinical significance of ovarian cysts in bovine reproduction. First, however, a review of gonadotropin and sex steroid release throughout the estrous cycle and postpartum period of cattle is warranted. Interaction of the Hypothalamo-Pituitary—Ovarian Axis in the Control of Gonadotropin Release Throughout the Estrous Cycle In ovariectomized (OVX) cows and ewes, LH (Forrest et al., 1980; Goodman and Karsch, 1980; Convey et al., 1983) and FSH (Convey et al., 1983) are released in pulsatile fashion with a frequency of approxi- mately 1 pulse/hour. This release of LH occurs synchronously with, and as a result of, pulsatile release of hypothalamic gonadotropin releasing hormone (GnRH) into hypOphyseal portal vessels (Clarke and Cummins, 1982; Levine et al., 1982). The frequency of GnRH release by hypothala- mic neurons has been depicted as controlled by a neural LH pulse genera- tor (Karsch, 1984). Pulsatile release of gonadotropins occurs throughout the estrous cycle, with changes in both amplitude and frequency at different stages of the cycle. Questions regarding the relationship of circulating ovarian steroids to gonadotropin release have been addressed via re- placement of ovarian steroids in OVX females and in in zitgg pituitary culture. In general, specific, homologous radioimmunoassays for bovine and ovine LH are readily available and results among laboratories have been in agreement. There has been more heterogeneity in regard to FSH radioimmunoassays used (Goodman et al., 1981; Bolt and Rollins, 1983) and as a result, pulsatile release of FSH has not been as well characterized. Hansel and Convey (1983) recently reviewed the physiology of the estrous cycle of cattle and sheep. For discussion of endocrine changes, they divided the cycle into 3 periods; the luteal phase and before and after the preovulatory surge of gonadotropins. These divisions will be used in discussion of temporal changes in sex steroid and gonadotropin concentrations and mechanisms of control. For this discussion, day 0 of the estrous cycle will be the day of the preovulatory surge of gonadotropins. Luteal Phase. After ovulation, proliferation of luteinized thecal and granulosa cells results in formation of the corpus luteum (CL). Accompanying CL growth is an increase in serum progesterone (P4) that peaks (2 to 10 ng/ml) at about day 10 and remains elevated for the next 8-9 days in cows (Robinson, 1977). Estradiol-17b (E2) remains at basal concentrations (less than 5 pg/ml) throughout the luteal phase with sporadic increases early in diestrus (Shemesh et al., 1972; Dobson and Dean, 1974) presumed to accompany growth and regression of EZ-active ovarian follicles (Ireland and Roche, 1983a). During the luteal phase of the estrous cycle, LH is released in pulsatile fashion in cows. Pulses of LH with an amplitude of 1-6 ng/ml are released every 4-6 hours (Rahe et al., 1980; Walters et al., 1984). Recently, smaller pulses of LH (less than 1 ng/ml) were detected between the larger amplitude pulses in diestrous cows (Walters et al., 1984). FSH is also released in pulsatile manner in cows, with uniform pulses (15-20 ng/ml) occurring synchronously with both large and small pulses of LH (Walters et al., 1984). There is a synergism of ovarian products that control gonadotropin release in the luteal phase. When OVX ewes were given implants con- taining P4, frequency of pulsatile LH release was reduced (1/hr to 1/6 hr), but with no change in amplitude (Goodman and Karsch, 1980). When OVX ewes were chronically exposed to luteal concentrations of E2 (3-5 pg/ml), the frequency of LH pulses was unchanged (1/hr), but amplitude was reduced (Goodman and Karsch, 1980). However, both P4 and E2 replacement in OVX ewes was needed to mimic average basal concen- trations of LH in diestrous ewes (Karsch et al., 1980). Specific effects of replacement of luteal concentrations of P4 or E2 on pulsatile release of FSH have not been examined. However, com- bined P4 and E2 replacement did not suppress basal FSH concentrations in OVX ewes to values seen in diestrous ewes (Goodman et al., 1981). A protein in ovarian follicular fluid, inhibin, may also play a role in regulation of FSH release. Administration of charcoal-extracted bovine follicular fluid suppressed FSH but not LH in ovx heifers (Ireland et al., 1983; Kiracofe et al., 1983). Pre-Surge Period. On day 18-20 of the bovine estrous cycle, luteal regression is accompanied by a rapid decline in serum P4 (Robinson, 1977). With removal of P4 influence, the frequency of LH and FSH pulses increases to 1/40-60 minutes (Rahe et al., 1980; Walters and Schallenberger, 1984). There is also a progressive increase in serum E2, reaching a peak of 10 pg/ml in cattle just prior to the preovula- tory surge of gonadotrOpins (Chenault et al., 1975; Walters and Schallenberger, 1984). This increase in E2 is mainly from one ovary (Ireland et al., 1984) and likely to be from a single, large estrogen- active ovarian follicle (Dieleman et al., 1983; Ireland and Roche, 1983b). The largest ovarian follicle at this time is the preovulatory follicle (Dufour et al., 1972). Pulsatile release of E2 was detected in caudal vena cava blood in cows, occurring synchronously with pulsatile release of LH detected in jugular samples (Walters and Schallenberger, 1984). The preovulatory surge of LH (Chenault et al., 1975; Walters and Schallenberger, 1984) and FSH (Walters and Schallenberger, 1984) begins approximately 36-48 hours after luteal regression has completed. This LH and FSH surge lasts for 6-10 hours and represents the summation of frequent pulses (1/20-30 minutes) that are also increased in amplitude (Rahe et al., 1980; Walters and Schallenberger, 1984). Serum 32 decreases rapidly during the preovulatory surge of gonadotropins (Chenault et al., 1975; Walters and Schallenberger, 1984). Both a decrease in P4 and an increase in E; are necessary prere- quisites for the preovulatory surge of gonadotropins in cycling cows or ewes. Frequency of LH pulses increased after ovariectomy of diestrous ewes, but E2 replacement was required to mimic the increased rate of pulsatile release of LH after luteal regression in cycling ewes (Karsch et al., 1983). The effect of E2 in initiating preovulatory surges of gonadotropins appears to be threefold. Initially, E2 reduces the sen- sitivity of the anterior pituitary to release LH in response to GnRH (Kesner et al., 1981; Clarke and Cummins, 1985). At the same time, however, E2 is synergistic with GnRH in priming the gonadotrophs for enhanced release of LH to subsequent GnRH stimulus (Padmanabhan and Convey, 1981). Lastly, E2 exposure sets a neural timing mechanism that schedules release of frequent pulses of hypothalamic GnRH and results in the preovulatory surge of gonadotropins (Kesner et al., 1981; Clarke and Cummins, 1985). The preovulatory surge of LH is terminated by pituitary refractoriness to GnRH (Kesner et al., 1981). P4 blocks the priming effect of E2 on gonadotrophs (Padmanabhan and Convey, 1981). Administration of exogeneous E2 did not result in a preovulatory surge of LH in diestrous cows (Short et al., 1979; Zaied et al., 1981; Stevenson et al., 1983). In OVX cows, exogenous P4 blocked the preovulatory surge of LH after replacement of physiological con- centrations of E2 (Kesner et al., 1982), but did not prevent surges of LH after pharmacological doses of E2 (Hausler and Malven, 1976; Short et al., 1979). Post-Surge Period. P4 and E2 are both low for 2-3 days after the pre-ovulatory surge of gonadotropins (Chenault et al., 1975; Walters and Schallenberger, 1984). At this time, there is divergence in LH and FSH release, where basal LH is lower after the surge than before and basal FSH is higher after the surge than before (Goodman et al., 1981; Roche and Ireland, 1981). Pulsatile release of LH was nondetectable for 6-12 hours post-surge, but FSH pulses were present (Walters and Schallen- berger, 1984). The appearance of frequent, low-amplitude pulses charac- terize the return of episodic release of LH (Rahe et al., 1980; Walters and Schallenberger, 1984). A secondary rise of FSH occurs about 12-24 hours after the preovu- latory surge of gonadotropins (Pant et al., 1977; Ireland and Roche, 1983; Walters and Schallenberger, 1984). This secondary rise of FSH is hypothesized to result from a reduction in both E2 and inhibin influence after ovulation (Padmanabhan et al., 1984). The second rise of FSH may also have a role in recruitment of the large, estrogen-active ovarian follicle that is present early in the luteal phase of the estrous cycle (Ireland and Roche, 1983a). Changes in Gonadotropins and Sex Steroids in the Postpartum Period of Cattle Gonadotropins. In nonsuckled dairy cows, serum LH concentrations are low (1 1 ng/ml) at and shortly after parturition and increase during the first 1 to 3 weeks postpartum (Echternkamp and Hansel, 1973; Carruthers and Hafs, 1980; Peters et al., 1981). This increase of LH is the net result of initiation of pulsatile release of LH (Carruthers and Hafs, 1980; Peters et al., 1981; Gitlin et al., 1983; Karg and Schallenberger, 1983). In contrast to LH, marked fluctuation of FSH was present before 1 week postpartum in dairy cows sampled at 6 hour intervals (Schams et al., 1978). Basal FSH increased in the first 5 days postpartum (Dobson, 1978; Webb et al., 1980) with no further increases throughout the next two weeks (Carruthers and Hafs, 1980; Webb et al., 1981). Pulsatile release of FSH was present on day 4 postpartum in dairy cows (every 2 hours) with no further increase in amplitude or frequency at one week later (Karg and Schallenberger, 1983). Thus, it appears that cows regain the ability to secrete FSH in pulsatile manner earlier in the postpartum period than LH. The disparity in the resumption of LH and FSH release in postpartum cows was evident in other experiments. When exogenous GnRH was given in the first and second weeks postpartum, LH release was higher in the second week (Foster et al., 1980; Gitlin et al., 1983) while FSH release did not differ between weeks but tended to be higher in the first (Poster et al., 1980; Peters and Lamming, 1984a). The post-castration rise of LH was slower in cows ovariectomized on day 4 postpartum than in cows ovariecto- mized on day 4 of the estrous cycle, but there was no difference between groups in the response of FSH (Schallenberger and Peterson, 1982). When suckled beef cows were ovariectomized on day 7 postpartum, the castration response in gonadotropin release was more evident for FSH than LH by day 14 postpartum (Convey et al., 1983). At present, it is not known whether the effect of postpartum interval on the castration response was due to differences in relative sensitivity to endogenous GnRH release, or whether pulsatile release of FSH is not dependent on pulsatile release of hypothal- amic GnRH. 10 Estradiol and Progesterone. Serum 32 is elevated at parturition in cows and decreases rapidly within 2 to 4 days postpartum (Echternkamp and Hansel, 1973; Kesler et al., 1978b; Sasser et al., 1979; Humphrey et al., 1983). Serum E2 remains low thereafter, increasing shortly before the first postpartum ovulation (Henricks et al., 1972; Echternkamp and Hansel, 1973; Kesler et al., 1979b; Humphrey et al., 1983). Although large ovarian follicles are present in the first week post- partum, exposure to pulsatile release of gonadotropins appears to be neces- sary to increase Ez concentrations in follicular fluid (Spicer, 1984). Small repeated doses of exogenous GnRH increased E2 in ovarian follicular fluid of 21-day postpartum beef cows with a numerical, but not signifi- cant, increase in serum E2 over controls (Spicer, 1984). Serum E2 was doubled (3-4 to 6-7 pg/ml) after small doses of GnRH were repeated in dairy cows at day 8 postpartum (Peters and Lamming, 1984a). P4 in peripheral serum is low at calving (Echternkamp and Hansel, 1973; Kesler et al., 1978b). Measurement of P4 in milk has been used as a means of monitoring luteal activity in dairy cows. In Europe and the United States, 80% or more of milked dairy cows had sustained increases of P4 prior to day 30 postpartum (Mather et al., 1978; Schams et al., 1978; van de Weil et al., 1979; Webb et al., 1980; Ball, 1982; Candle et al., 1982). The first increase of P4 was of shorter duration (8-12 days) and lower in magnitude than succeeding luteal phases in 50-70% of cows sampled (Mather et al., 1978; Schams et al., 1978; van de Wiel et al., 1979; Webb et al., 1980; Candle et al., 1982). Peters and Lamming (1984b) detected small ovarian structures (5 mm in diameter) by palpa- tion per rectum at the time of the shortened P4 increases and speculated 11 the presence of small anovulatory luteinized structures as described by Berardinelli et al., (1979) in pubertal heifers. The first P4 increases in 10-15% of cows sampled were longer (> 20 days) than a normal luteal phase and the first luteal phase was considered of normal length (14-20 days) in the remaining cows (Mather et al., 1978; Schams et al., 1978; van de Wiel et al., 1979). Significant peaks of LH preceeded about half of the first postpartum increases of P4, but preovulatory surges were detected prior to all second cycles (Schams et al., 1978). The hypothalamo-pituitary axis did not have the ability to release LH in response to administration of exogenous E2 shortly after parturition (Karg and Schallenberger, 1983). Although criteria regarding what consti- tuted an LH response varied among researchers, it was apparent that dairy cows regain the ability to release LH in the magnitude of a preovulatory surge after exogenous E2 between days 10 and 30 postpartum (Zaied et al., 1981; Karg and Schallenberger, 1983; Stevenson et al., 1983). In cows releasing LH after exogenous E2, the variation in magnitude of LH release and the interval from E2 treatment to peak LH tended to be inversely related to postpartum interval (Zaied et al., 1981; Stevenson et al., 1983). Ez treatment did not result in LH release in postpartum cows when endogenous P4 concentrations were indicative of luteal activity (Zaied et al., 1981; Stevenson et al., 1983). The positive feedback response of FSH release after exogenous E2 was seen as early as day 5 postpartum (Karg and Schallenberger, 1983). In postpartum cows, it thus appears that several endocrine events must take place for the initiation of estrous cycles. The hypothalamo- pituitary axis must regain both the ability to release basal LH and FSH in 12 a pulsatile manner and the ability to respond to the positive feedback effect of E2 on gonadotropin release. Also, ovarian follicles must acquire the ability to produce E2. Ovarian Cysts in Bovine Reproduction Clinical Definition. Ovarian cysts, in bovine reproduction, repre- sent a condition characterized by the development and persistence of an anovulatory ovarian follicular structure(s) (usually > 2-2.5 cm in diameter), absence of a CL and cessation of estrous cycles (Roberts, 1971; Seguin, 1975). Diagnosis of ovarian cysts is usually based on the results of a single ovarian palpation per rectum, with subjective classification that cystic structures with a thick-walled consistency are likely to be lined with luteal tissue (Zemjanis, 1970; Roberts, 1971). The presence of an ovarian cyst reflects an observation of a pathophysiological process that is not well defined. Cows with ovarian cysts may spontaneously resume estrous cycles, but with great variation in the persistence of the cystic condition (Whitmore et al., 1974). Regres- sion of an ovarian cyst may also be accompanied by formation of another (Kesler et al., 1980). Profiles of ovarian steroids in peripheral serum of cows with ovarian cysts reflect the range of values seen at any stage of the estrous cycle, but with unpredictable changes within cows across time (Kesler et al., 1980; Roy et al., 1985). Cows with ovarian cysts show considerable variation in estrous beha- vior. Most cows with ovarian cysts are anestrous (47-82% of cases) with the remaining cows showing intermittent or frequent estrous activity (nymphomania) (Bierschwal, 1966; Morrow et al., 1966; Nessan et al., 1977; Bostedt et al., 1979; Leslie and Bosu, 1983). Nessan and King (1981) 13 reported no difference in serum Ez or testosterone in cows with ovarian cysts showing nymphomania versus anestrous behavior. Saumande et al., (1979) reported that 65% of cows showing nymphomania had serum E2 concentrations similar to or above values seen for proestrous cows. Etiology, Incidence and Predisposing Factors. As detected by palpa- tion 225 rectum, ovarian cysts have been reported to occur in 5-30% of postpartum intervals in herds composed predominantly of Holstein cows (Bierschwal, 1966; Morrow at al., 1966; Whitmore et al., 1971; Britt et al., 1977; Kirk et al., 1982; Bargai, 1982; Peralta et al., 1982; Smith et al., 1985a,b). Ovarian cysts may be detected at any time, but 50-70% were diagnosed within the first 60 days postpartum (Morrow et al., 1966; Whitmore et al., 1971; Kirk et al., 1982). The incidence is higher in multiparous than in primiparous cows (Seguin, 1975; Marcek et al., 1985; Smith et al., 1985a,b; Roberts, 1971). The condition occasionally occurs in dairy heifers and beef animals (Roberts, 1971). In one study, the incidence of ovarian cysts was threefold higher in cows with other con- current postpartum disorders such as metritis, retained placenta, delayed uterine involution after dystocia, ketosis, clinical mastitis, etc. (Morrow et al., 1966). If left untreated, 50% or more of cows found to have ovarian cysts in the early postpartum period (< 60 days) spon- taneously resumed estrous cycles (Morrow et al., 1966; Whitmore et al., 1974). Rates of spontaneous recovery are lower and the intervals from diagnosis to recovery are more unpredictable when ovarian cysts are diagnosed later in the postpartum period (Whitmore et al., 1974; Kesler and Garverick, 1982). 14 Other factors have been associated with the formation of ovarian cysts in cattle, but cause-and—effect relationships have not been well established. Such factors included seasonality (higher incidence in fall and winter), nutrition and high milk production (see Seguin, 1975; Kesler and Garverick, 1982 for review). Heredity has also been shown to influence the incidence of ovarian cysts in dairy cattle (Casida and Chapman, 1951; Bane, 1964; Kirk et al., 1982). The cause of ovarian cyst formation is unknown, but has long been speculated to be a lack of LH release at estrus. Cows with ovarian cysts did not have degranulation of anterior pituitary basophils after estrus when compared to cows that ovulated (Jubb and McEntee, 1955). The LH content of pituitary glands collected post-estrus was higher in cows with ovarian cysts than in cows that ovulated (Yamauchi et al., 1954; Donaldson and Hansel, 1968). In the first 30 days postpartum, formation of ovarian cysts was associated with a lack of LH release in response to an increase of endogenous E2 concentrations (Kesler et al., 1979a). In cycling cows, clinical observation and results from experiments to induce ovarian cysts indicated that the period between luteal regression and ovulation was a critical time for ovarian cyst formation (Roberts, 1971; Wiltbank et al., 1961; Liptrap and McNally, 1976; Nadaraja and Hansel, 1976). Nadaraja and Hansel (1976) speculated that formation of ovarian cysts after exogenous E2 could be caused by release of LH that was premature in relation to CL regression. Anatomy and Steroids in Fluid of Ovarian Cysts. Most anatomical descriptions of ovarian cysts are based on data collected by abattoir survey, thus clinical history and therapeutic measures are often 15 unknown. There is considerable variation in the histological anatomy of bovine ovarian cysts. The granulosa layer may be very prominent (> 20 cell layers), present but in varying stages of degeneration, or completely absent (Yamauchi and Inui, 1954; Short, 1962; Al-Dahash and David, 1977b; Leidl et al., 1979; Brown et al., 1982). The thecal layer also varied in thickness and extent of degenerative changes (Yamauchi and Inui, 1954; Al-Dahash and David, 1977b; Leidl et al., 1979; Brown et al., 1982). Evidence of luteinization may also be present. Grossly visible luteal tissue may completely line the wall of an ovarian cyst, or may appear as a crescent-shaped layer at the base of the cyst as seen on cut surface (Roberts, 1971; Al-Dahash and David, 1977a). Luteinization of the theca may also be histologically visible as generalized in distribu- tion or in small, focal patches (Al-Dahash and David, 1977b). In general, granulosa cells were absent when there was histological evi- dence of luteinization in thecal layers (Yamauchi and Inui, 1954; Short, 1962; Al-Dahash and David, 1977b; Leidl et al., 1979). The basement membrane was absent in nearly all ovarian cysts examined (Al-Dahash and David, 1977b). Steroiogenic potential of ovarian cysts appears related to the viability of granulosa cells or luteal tissue. Yamauchi and Inui (1954) observed marked estrogen bioactivity in antral fluid only from ovarian cysts with a prominent granulosa layer. 82 concentrations were highest in antral fluid of ovarian cysts with an intact granulosa layer (Leidl et al., 1979; Bamburg et al., 1981). When the granulosa layer was dege- nerate or absent, steroidogenesis appeared to be interrupted at the step 16 for 17-alpha-hydroxylation of P4, thus preventing accumulation of andro- gens in ovarian cyst fluid (Short, 1962; Choi et al., 1982). P4 concentrations in cyst fluid were highest when the cyst wall contained luteinized tissue (Leidl et al., 1979; Bamburg et al., 1981). Peripheral Sex Steroid and Gonadotropin Concentrations. Most endocrine data from cows with ovarian cysts was obtained from single samples collected at the time of diagnosis, usually just prior to treatment. P4 in serum or milk can range from being nondetectable to values similar to those of diestrous cows (Kittock et al., 1974; Seguin, 1975; Nakao et al., 1976; Dobson et al., 1977; Nessan et al., 1977; Saumande et al., 1979; Leslie and Bosu, 1983; Vasquez et al., 1984; Marcek et al., 1985), with great variation in the distribution of low and high values among reports. Serial sampling of cows with ovarian cysts showed that increases of P4 may occur as short-interval peaks of less than 2 weeks duration, or may be similar in length to a normal luteal phase (Kesler et al., 1980; Gunzler and Schallenberger, 1980; Roy et al., 1985). Sources of P4 in these instances could include luteinized ovarian cysts or an undiagnosed CL (Al-Dahash and David, 1977a; Gunzler and Schallenberger, 1980). Recent endocrine and gross anatomical data from an individual cow with an ovarian cyst raises the possibility of anovulatory luteal structures as being the source of short-duration P4 peaks (Roy et al., 1985). 82 concentrations in peripheral serum of cows with ovarian cysts may also span the range of values seen throughout the estrous cycle (Seguin et al., 1975; Dobson et al., 1977; Kesler et al., 1981) with one 17 report suggesting that some individuals with ovarian cysts may have cir- culating E2 well above proestrual concentrations (Saumande et al., 1979). Changes in serum E2 across time have not been as well defined as for P4 in cows with ovarian cysts, but it is apparent that some cows can maintain high concentrations of E2 for at least a month (Lunaas et al., 1974; Kesler et al., 1980). In general, cows with ovarian cysts and high endogenous E2 have low P4 values (Kittock et al., 1974; Dobson et al., 1977; Leidl et al., 1979; Saumande et al., 1979). There has been no evidence for increased testosterone in peripheral sera of cows with ovarian cysts when compared to values in cycling cows (Kesler et al., 1979; Nessan and King, 1981). Basal gonadotropin secretion has not been well characterized in cows with ovarian cysts. LH concentrations in single samples collected from cows with ovarian cysts were within the range of values in cycling cows, excluding the preovulatory surge of LH (Seguin et al., 1976; Dobson et al., 1977; Zaied et al., 1981). Kesler et al. (1980) measured LH in samples collected at three-day intervals for a month in cows with ovarian cysts and saw sufficient variation within cows to suggest that pulsatile release may be present. Serum FSH concentrations in single samples from cows with ovarian cysts have been reported in one study, and were also within ranges seen for cycling cows (Dobson et al., 1977). Zaied et al. (1981) gave exogenous E2 to cows with ovarian cysts to test the ability of the hypothalamo-pituitary axis to release LH. LH release did not occur in postpartum cows or cows with ovarian cysts when serum P4 was above 1.0 ng/ml. When endogenous P4 was low, 3 of 5 cows 18 with ovarian cysts released LH, but the response was more variable and delayed when compared to LH release by 4 of 4 anestrous cows at 30-40 days postpartum. Therapy of Ovarian Cysts. The objective of therapy of ovarian cysts in cattle is to initiate fertile estrous cycles. Manual rupture of the cystic structure offered little if any improvement over rates of spontaneous recovery (Roberts, 1971; Vasquez et al., 1984). Casida et al. (1974) gave ovine pituitary extract to cows with ovarian cysts and were the first to report success with a preparation of LH bioactivity. At present, perhaps the most common treatment of ovarian cysts is admin- istration of either human chorionic gonadotropin (HCG) or GnRH. The therapeutic effect of exogenous GnRH results from the release of endoge- nous LH in cows with ovarian cysts (Kittock et al., 1973; Garverick et al., 1975; Seguin et al., 1976; Kesler et al., 1979b). A positive response to HCG or GnRH therapy occurred either with luteinization of the cystic structure(s) or ovulation of an ovarian follicle with subsequent CL development (Yamauchi, 1955; Seguin, 1975; Kesler et al., 1981; Tanabe and Brofee, 1982). This response was accom- panied by an increase in serum P4 (1 2.0 ng/ml) by 2 to 4 weeks post- treatment (Kittock et al., 1974; Seguin et al., 1976; Kesler et al., 1979b; Nakao et al., 1983). In cows with ovarian cysts and high endoge- nous P4, spontaneous luteolysis may be responsible for return to estrus and not the effects of exogenous gonadotropin therapy (Gunzler and Schallenberger, 1980; Marcek at al., 1985). In field trials, less than 30% of untreated cows with ovarian cysts resumed estrous cycles within 19 30 days of diagnosis (Bierschwal et al., 1975; Seguin et al., 1976; Stolla et al., 1980; Vasquez et al., 1984). HCG or GnRH treatment resulted in resumption of estrous cycles within 30 days in 70 to 100% of cows with ovarian cysts (Bierschwal et al., 1975; Elmore et al., 1975; Garverick et al., 1976; Seguin et al., 1976; Whitmore et al., 1979; Stolla et al., 1980; Kudlac et al., 1984; Marcek et al., 1985). In the latter-cited field trials, conception rates after HCG or GnRH ranged from 40 to 80% of treated cows with ovarian cysts, with average inter- vals from treatment to conception ranging from 30 to 87 days. Stolla et al. (1980) reported that conception rates were reduced in cows diagnosed and treated for ovarian cysts after 12 weeks postpartum. Leslie and Bosu (1983) used a prostaglandin analogue as an initial treatment for ovarian cysts and evaluated the response with respect to pretreatment P4 concentrations. Seventy-seven percent of cows with P4 above 1.0 ng/ml were in estrus within 7 days posttreatment with an 80% conception rate reported over the first two subsequent estrous periods. Only 8.3% of cows with low P4 values at the time of prostaglandin treat- ment conceived within 6 weeks posttreatment. At present, the main limi- tation of using prostaglandins as a treatment for ovarian cysts was that palpation per rectum did not provide a means of accurate identification of cows that would be likely to respond (Leslie and Bosu, 1983; Nakao et al., 1983; Marcek et al., 1985). EXPERIMENT I GonadotrOpin Secretion in Dairy Cows with Ovarian Cysts: Pulsatile Release Patterns and Response to Estradiol Challenge 20 BACKGROUND T0 EXPERIMENT I Interactions within the hypothalamo-pituitary-ovarian axis are important in the regulation of estrous cycles in domestic animals (Hansel and Convey, 1983). In ovariectomized (OVX) cows (Forrest et al. 1980) and ewes (Goodman and Karsch, 1980), there is pulsatile release of LH, resulting from secretion of hypothalmic gonadotrOpin releasing hor- mone (GnRH) (Levine et al., 1982). Replacement of luteal concentrations of either progesterone (3-5 ng/ml) or estradiol-17b (2-3 pg/ml) modified the frequency and/or amplitude of episodic pulses of LH in OVX ewes. Progesterone (P4) decreased frequency but didn't alter amplitude, and estradiol-17b (E2) decreased amplitude without changing frequency (Goodman and Karsch, 1980). Similar temporal relationships of LH secre- tion are seen in bovine estrous cycles. Episodic pulses of LH during diestrus were of lower frequency and higher amplitude than during the transitions from luteal regression to estrus or from the preovulatory LH peak to corpus luteum maturation (Rahe et al., 1980; Walters et al., 1984; Schallenberger et al., 1984; Walters and Schallenberger, 1984). The preovulatory surge of LH is also controlled by ovarian steroids. Exogenous E2 initiated a preovulatory-like release of LH in prepubertal heifers (Saba et al., 1976; Schillo et al., 1983), OVX cows (Short et al., 1979; Kesner et al., 1981), and 30-40 day postpartum dairy cows that had not resumed estrous cycles (Zaied et al., 1981). This positive feedback effect, seen when P4 is low, is thought to be due to both an increase in GnRH release and a direct effect on pituitary responsiveness to GnRH, as seen in yi££g_(Padmanabhan and Convey, 1981) and supported by data obtained in vivo (Kesner et al., 1981). 21 22 Ovarian cysts are one of the most commonly diagnosed infertility problems in dairy cows, reported to occur in 5-30% of postpartum inter- vals in dairy cows (Morrow et al., 1966; Whitmore et al., 1971; Britt et al., 1977, Bargai et al., 1982; Peralta et al., 1983; Smith et al., 1985a,b). The condition has been defined as the persistence of an ano- vulatory ovarian follicular structure in the absence of a CL (Roberts, 1971; Seguin et al., 1976), with interruption of estrous cycles. The cystic condition is not static however, as affected cows may spon- taneously resume estrous cycles or cystic structures may regress with concomittant formation of another (Morrow et al., 1966; Whitmore et al., 1974; Kesler et al., 1980). Concentrations of E2 and P4 in peripheral serum span the range of values seen throughout bovine estrous cycles (Seguin et al., 1976; Dobson et al., 1977; Saumande et al., 1979). Measurement of ovarian steroids across time show that changes within and among affected cows are extremely variable (Lunaas et al., 1974; Gunzler and Schallenberger, 1980; Kesler et al., 1980; Roy et al., 1985). Inadequate release of LH at estrus has been hypothesized as cause of ovarian cyst formation (Jubb and McEntee, 1955; Donaldson and Hansel, 1968). Kesler et al. (1979a) measured LH, E2 and P4 in early postpartum dairy cows and observed a lack of LH release in response to an increase in endogenous E2 in cows that formed ovarian cysts. Administration of exogenous E2 after 30 days postpartum resulted in release of LH in 3 of S cows with ovarian cysts that was delayed when compared to the response of 4 of 4 cows without ovarian cysts (Zaied et al., 1981). 23 Information on the nature of pulsatile gonadotropin secretion in dairy cows with ovarian cysts may provide insight into the apparent relative lack or delay of LH release after increase of endogenous or exogenous E2. At present, pulsatile release of gonadotrOpins has not been characterized in cows with ovarian cysts. An experiment was con- ducted to address two objectives. The first was to determine whether gonadotropins are released in episodic patterns in cows with ovarian cysts and identify relationships, if any, between peripheral serum sex steroid concentrations and gonadotropin secretion. The second objective was to assess gonadotropin release after exogenous E2 in the same cows. 24 MATERIALS AND METHODS FOR EXPERIMENT I Animals. Seven Holstein cows (3-10 years old) culled from dairy herds due to infertility resulting from ovarian cysts were referred by veterinary practitioners. The postpartum interval of these cows ranged from 7 to 24 months. Milking was discontinued at the farm of origin. The cows were brought to the Michigan State University Veterinary Clinical Center. Three other Holstein and two Guernsey adult nonlac- tating cows used at the Veterinary Clinical Center for teaching purposes were found to have ovarian cysts and made available for study. The first cow was admitted during August, 1980 and the last in November, 1982. Animals were held in individual tie stalls and no observations of estrous behavior were made. The cows were fed brome grass-alfalfa hay twice daily, given trace mineral salt twice weekly and had fresh water .39 libidum. Ovarian cysts were defined as follicular structures of at least 2.5 cm diameter, persisting for at least 10 days in the absence of a corpus luteum (Roberts, 1971). Ovarian palpation per rectum and measurement of coccygeal venous serum P4 at 3 to 7 day intervals were used to monitor the presence of ovarian cysts and rule out the occurrence of estrous cycles. This monitoring period ranged from 7 to 21 days. Blood Samplingyand Hormonal Treatments. An indwelling jugular catheter was inserted in each cow on the day before experimental sampling. Catheter patency was maintained with heparinized saline. 25 Blood samples were collected and treatments administered by the following schedule: On Day 1, starting at 0800 hr, blood (10 ml) was collected at 15 minute intervals for 4 hours, with an additional 10 ml drawn at each hourly interval. Immediately after the last 15-minute sample at 1200 hr, 1.0 mg estradiol was given intramuscularly. The first two cows sampled received estradiol in the form of estradiol-17b- cyclopentylproprionate (ECPO, Upjohn Co., Kalamazoo, MI). The other 10 cows received 1.0 mg E2 dissolved in 3 ml safflower oil (Sigma Chemical Co., St. Louis, MD), a preparation shown to initiate predictable LH surges in OVX cows (Kesner et al., 1981). After Ez challenge, blood (20 ml) was collected at 3 hr intervals for 30 hr (until 1800 hr, Day 2) to determine whether or not preovulatory-like surges of gonadotropins occurred. Immediately after the 1800 hr, Day 2 sample was taken, all cows received a subcutaneous injection of 100 ug GnRH (Sigma Chemical Co.) dissolved in 2 ml of 0.15M NaCl. Blood (10 ml) was collected at 30 min intervals for 2 hr. Because the results of the E; challenge would not be known until after hormonal assay, GnRH was given to show whether releasable pituitary gonadotropins were present in the advent of no gonadotropin release after exogenous E2. After collection, blood was allowed to clot for 4-8 hr at 20 C and then centrifuged at 10009 for 15 min. Serum was aspirated and stored at -20 C until thawed for assay. Hormone Assays. Validated radioimmunoassays were used to quantify LH (Convey et al., 1976) and FSH (Carruthers et al., 1980) in all sera collected. E2 (Appendix), P4 and testosterone (Appendix) were measured in the first and last hours of the 15 min interval collection period. 26 Because stress-associated cortisol elevation had been shown to suppress pulsatile LH in OVX beef cows (Echternkamp, 1984) cortisol (Roth et al., 1983) was measured in all samples drawn at 15 min intervals. Progesterone Radioimmunoassay. Serum P4 concentrations were measured by radioimmunoassay, using a commercially available kit (Coat-a-Count® No Extraction Progesterone, Diagnostic Products Corporation, Los Angeles, CA). Preliminary validation data showed that dilution curves of unextracted bovine sera did not parallel binding inhibition curves made with P4 standards in human plasma supplied by the manufacturer. To remove binding differences probably caused by species' differen- ces in serum proteins, P4 was removed from serum by solvent extraction and new standards were made. One ml of serum was extracted twice with 6 ml petroleum ether (Mallinckrodt, Inc., Paris, KY). Each extraction step included mixing by horizontal shaking for 10 min, centrifugation at 5209 for 5 min, freezing of the aqueous layer, and decanting of the solvent layer into 16 x 100 mm glass culture tubes. The petroleum ether was then evaporated under vacuum in a prewarmed (40'C) vortex evaporator (Haake-Buchler Instruments Inc., Saddle Brook, NJ). One ml of phosphate buffered saline-0.25% gelatin (PBSG) was added to each tube after solvent evaporation and the tubes were incubated in a 37 C water bath for 1 hr. Eighty-six percent of 3H-P4 added to bovine sera (New England Nuclear, Boston, MA) was recovered with the extraction and reconstitu— tion steps. Standards were prepared by dissolving P4 (Sigma Chemical Co., St. Louis, MO) in methanol at a concentration of 1 mg/ml and then further dissolving aliquots of this solution in PBSG to a final 27 concentration of 20 ng/ml. Final PBSG dilutions used in standard curves were 0.1, 0.2, 0.5, 1.0, 2.5, 5.0, 10.0 and 20 ng/ml. Assays were run in 12 x 75 mm anti-progesterone antibody-coated polyproPylene tubes supplied by the manufacturer. Total count tubes received 1.0 ml of 125I-progesterone, also supplied by the manufacturer (approximately 40-50 x 103 cpm). One hundred ul of the appropriate stan- dard or sample was mixed with 1.0 ml of tracer in the antibody-coated tubes. A 12-20 hr room temperature (20°C) incubation was allowed for antigen-antibody equilibration. The contents of all except total count tubes were poured off and the tubes allowed to stand inverted for 5-10 min to maximize drainage. The tubes were placed in a gamma radiation counter (Micromedic, Horsham, PA) and antibody-bound counts were deter- mined. Standard curves were calculated with a log-logit transformation, with mean 1 SD regression coefficient of -0.899 1 0.054 and y-intercept value of 0.964 I 0.216 for 6 assays. In the absence of unlabeled hor- mone, 58.8 :_2.9% of 1251-P4 was bound to antibody. Antibody specifi- city data against 15 other related steroids were provided by the manu- facturer, and the highest cross reactivity at 50% decrease of total binding was 2.4% for 11-desoxycortisol. Extrapolation of the standard curve at two standard deviations below the mean total binding value indicated an assay sensitivity of 25 pg/tube. Intra- and interassay coefficients of variation were 8.1 and 11.9%, respectively. When 0.25 or 1.25 ng P4 were added to bovine serum, 98.5 and 103% were measured in the assay, respectively. Dilution curves of extracted bovine sera paralleled displacement curves of progesterone in PBSG. 28 Definition of Variables and Statistical Analyses. The 12 cows were retrospectively classified by basal serum E2 and P4 concentrations into estrogenic (1 10 pg/ml E2 and < 0.5 ng/ml P4, Group 1), low steroid (< 10 pg/ml E2 and < 0.5 ng/ml P4, Group 2), and luteinized (< 10 pg/ml E2 and_l 0.5 ng/ml P4, Group 3) groups. These classifications were based on correlations of histological anatomy of ovarian cysts with steroid content in cyst fluid (Leidl et al., 1979; Choi et al., 1982). The 95% confidence intervals of the mean LH or FSH concentrations in a bovine serum pool were calculated (r-17). In 15 min samples, LB or FSH peaks were defined as values exceeding the previous or second pre- vious sample by the 95% confidence limit of the assay (Convey at al., 1983). The critical increases in this study were 1.0 ng/ml for LH and 10 ng/ml for FSH. A nadir was the lowest point between 2 peaks. A gonadotropin baseline was the mean of all points within one standard deviation of the mean of the nadirs. Average overall LH, FSH or cor- tisol was the mean of all 15 min samples from an individual cow. A preovulatory surge of LH was thought to have occurred post 32 when a peak above 10 ng/ml was flanked by values showing ascending and descending trends (Swanson and Hafs, 1971). Similar temporal changes were used to identify a surge of FSH, with expected increases of at least 50 ng/ml above basal values. Basal concentrations of sex steroid and gonadotropin and cortisol variables described above were contrasted among the three groups by one- way analysis of variance with differences identified with Scheffe's interval (Gill, 1978). Chi-square analysis was used to determine 29 whether the synchrony of pulsatile secretion of LH and FSH differed among groups. Differences among groups in LH or FSH after exogenous E2 or GnRH were tested with split-plot analysis of variance. 30 RESULTS OF EXPERIMENT I Ovarian Changes and Serum P4 Prior to Intensive Sampling. Ovarian cysts ranged in size from 2.5 to 5.0 cm in diameter. In the pre-sampling period for monitoring the cystic condition, two cows showed disappearance of a cystic structure accompanied by deveIOpment of another on the contra- lateral ovary. Subsequently, one of these cows was assigned to Group 1, the other to Group 2. Cystic structures were initially detected on both ovaries of a third cow later assigned to Group 2, with one ovarian cyst disappearing during the documentation period. The three cows had each been examined 4 times throughout a 14 day period. No palpable ovarian changes were detected in the other 9 cows. There were no P4 concentrations above 1.0 ng/ml throughout the docu- mentation period. Only one cow (later assigned to Group 3) had P4 values consistently between 0.5 ng/ml and 1.0 ng/ml. The mean 1 SD pre-sampling. P4 concentrations for cows subsequently assigned to Groups 1, 2, or 3 were 0.06 1 0.05 (r-14), 0.30 i 0.30 (r-10), and 0.5 :_0.3 (r-4). respectively. Basal Steroids and Gonadotropins. Classification Groups 1, 2, and 3 contained 6, 4, and 2 cows, respectively. Steroid and gonadotropin means for each group on the day of frequent blood collection are shown in Table 1. As expected by classification criteria, basal serum E2 was higher (p a 0.002) in Group 1 than in the other groups and serum P4 was higher (p-0.001) in Group 3 cows than in the other groups. Baseline and average overall LH concentrations were higher (p < 0.05) in Group 1 than 31 in Groups 2 and 3. There were no differences among groups in testos- terone, cortisol, or other gonadotrOpin variables. TABLE 1. Basal serum sex steroid, gonadotrOpin, and cortisol data (31-: SD) from dairy cows with naturally-occurring cystic ovarian follicles.a Cows were assigned to groups based on estradiol and progesterone concentrations. Ovarian Cyst Classification Variable Group 1 Group 2 Group 3 Estradiol-17b (pg/m1) 18.5 i .6b 3.5 i 1.5 3.5 i 0.5 progesterone (ng/ml) 0.03 i 0 08 0.18 i 0.17 0.70 i 0.11.b Testosterone (ng/ml) 0.12 i 0 06 0 10 -_1-_ 0.0 0.13 i 0.05 Overall LH (ng/ml) 2.4 : 0.2° 1.7 i 0.5 1.7 i 0.3 No. LH peaks/4hr 2.2 i 1.5 2.3 i 1.7 1.5 :1; 0.7 L1! peak amplitude (ng/ml) 1.4 i 1.0 1.4 i 1.1 1.4 i 0.5 Baseline LH (ng/ml) 1.9 :1; .4° 1.3 i 0.3 1.2 i 0.4 Overall FSH (ng/ml) 74 i 27 60 i 18 78 i 6 No. FSH peaks/4hr 1.8 i 1.2 2.5 i 1.7 1.5 i 0.7 FSH peak amplitude (ng/ml) 14.5 i 8.8 14.7 i 10.3 23.8 i 6.7 Baseline FSH (ng/ml) 69 i 27 54 i 14 71 i 5.6 Overall Cortisol (ng/ml) 14.1 i 8.3 15.8 i 4.6 21.3 i 6.4 8Note that the precision of estimation is variable among hormone assays bDiffers from other values in a row by 9.: 0.002 cDiffers from other values in a row by p < 0.05 There was considerable variation in frequency of LH and FSH pulsa- tile release within groups, probably accounting for the lack of dif- ferences among groups. The frequency of LB or FSH pulses/4hr ranged from 0 to 4 in Groups 1 and 2 and from 1 to 2 in Group 3. Gonadotropin release patterns of six cows are shown on Figure 1. These individuals were selected as examples of low and high pulse numbers from Groups 1 and 2 and one example from Group 3. 32 Seventy-two percent of all LH pulses were accompanied by concomit- tant FSH pulses and 75% of FSH pulses were accompanied by LH pulses. This relationship of synchronous versus asynchronous pulses did not differ among groups (p > 0.3) for LH or FSH. An additional 16% and 21% of gonadotropin pulses occurred when cows had only LH or only FSH pulses during the 4 hr sampling period, respectively. Gonadotropin Response to Exogenous E2. The ability to detect dif- ferences among groups in gonadotropin secretion after exogenous E2 was confounded with the E2 preparations used. Use of ECP was discontinued because of lack of data on the ability of ECP to initiate preovulatory surges in cows. There were no differences among groups in LH or FSH release after exogenous E2 (p > 0.05). Gonadotropin secretion in response to exogenous E2 was then assessed on a within-cow basis, by identification of LH or FSH values after E2 treatment that were more than three standard deviations above the basal overall LH or FSH mean for each individual. Changes in LH and FSH after exogenous E2 for each cow are shown on Figures 2 and 3, respectively. The two cows that received ECP were from Group 1. Neither cow had post treatment LB or FSH values that were considered as outliers from pre-treatment basal LH or FSH values. Ten cows received E2 in safflower oil. Four of these cows were from Group 1, with no change in LH or FSH after treatment. Four were from Group 2, of which 2 individuals showed increases in gonadotropins after E2. One individual, Cow HM, showed the only preovulatory-like LH surge recorded in the experiment, with increased LH values at 15, 18, 33 Figure 1. Pulsatile release of LH and FSH in adult dairy cows with ovarian cysts (H - Holstein, G = Guernsey). LH or FSH pulses are iden- tified by an *. Corresponding serum concentrations of estra- diol-17b (E2) and progesterone (P4) are also listed. 12. 10. 12.0 SERUM LH m on PSHxlO'l (0) IN nglml a 10. .O‘ G4(GROUP 1) 34 52-126 palml P4 -O.2 ng/ml t 8 9 10 11 12 H78(GROUP 1) o. _ E2-16.3 pglml 0- P4-0 nglml .- t It ww 1 9 10 l J 11 12 H15(GROUP 1) I T 82-290 pg/ml P4=0.0 ng/ml TIME OF HM(GROUP 2) 12.0 C 62:4.3 pglml 10.0, P4=0.4 nglml 8.0 6.0 4.0- F ‘ ‘ ‘ 2.0- ‘ 1 8 9 10 11 12 HO(GROUP 2) 12.0- » £224.7pglml 10.0»- P4=O nglml b‘ . Q B'OIW 6.0- 4.0L 2.01 10.0 8.0 ‘ 5-0* 62:13.1 polml P4=0.8 nglml 4.0» .- t . 2.0M 8 9 1O 11 12 DAY (AM) 35 and 21 hr post-E2 that peaked above 20 ng/ml (see Figure 2). FSH values from HM were also markedly elevated at 15, 18, and 21 hr post-£2 (see Figure 3). Prior to E2 treatment, this cow had four synchronous episo- dic pulses of LH and FSH/4 hr and was one of the animals that had a changeover of cystic structures detected by palpation per rectum. Cow H0, another member of Group 2, had a transient increase in LH at 21 through 30 hr post E2, with values ranging from 3.0 to 5.7 ng/ml (see Figure 2). This cow had 0 and 3 episodic pulses of LH and FSH respec- tively, prior to exogenous £2, and was the individual showing disap- pearance of an ovarian cyst prior to sampling. One of the two Group 3 members also showed transient increase in LH. Cow HS had increased LH values from 18 through 30 hr post E2 with values ranging from 3.0 to 6.3 ng/ml (see Figure 2). This cow had one pulse/4 hr of LH and FSH. Gonadotropin Response to Exogenous GnRH. All cows released both LH and FSH after GnRH, with no differences observed among Groups (p > 0.05). Mean values of LH and FSH for each sample taken after GnRH admin- istration are shown on Figure 4. The cow having the preovulatory LH surge after exogenous 82 had the lowest LH values for post GnRH samples, with concentrations of 1.0, 5.9, 7.9 and 9.7 ng/ml at 0.5, 1.0, 1.5 and 2.0 hr after GnRH, respectively. Three of these values were below and one just at the 99% confidence interval for the mean LH value for each time period. Corresponding FSH values for the same cow were also below 99% confidence intervals of mean response after GnRH, with concentrations 36 Figure 2. Serum LH after exogenous estradiol-11b in 12 dairy cows with ovarian cysts. Prior to treatment, X': SD endogenous estradiol-17b for Groups 1, 2 and 3 were 18.6 I 6.6, 3.5 1 1. and 3.5 1 0.5 pg/ml, respectively. Also prior to treatment, .1 SD serum progesterone for Groups 1, 2 and 3 were 0.03 i 0.08, 0.18 i 0.17, and 0.70 1 0.14 ng/ml, respectively. Values marked with an * are 3 SD above the mean of 17 samples collected at 15 min intervals from -4 to 0 hr within each individual (data not shown). 2 x Serum LH in ng/ml E 5 5 h) -b 0) (D N-bmm mem 37 r- COWS GROUPI on 0H9 7 064 0H78 on IHR " GROUP 2 _ * s * _ ' .* s * A - 3.. see/do ~— 1 1 1 1 1 1 1 1 1 1 1 Jr h'GROUPES * ' * HS _ * * s P— . 0 0 H56 1 1 1 1 1 1 L 1 ’ 1 1 1_ '3 O 3 6 9 12 l5 IS 21 24 27 30 Hours after Exogenous Estradiol (1.0 mg, IM) 38 Figure 3. Serum FSH after exogenous estradiol-17b in 12 dairy cows with ovarian cysts. Prior to treatment, x‘: SD endogenous estradiol-17b for Groups 1, 2 and 3 were 18.6 i 6.6, 3.5 1 122 and 3.5 i 0.5 pg/ml, respectively. Also prior to treatment, X .1 SD serum progesterone for Groups 1, 2 and 3 were 0.03 1 0.08, 0.18 1 0.17, and 0.70 i 0.14 ng/ml, respectively. Values marked with an * are 3 SD above the mean of 17 samples collected at 15 min intervals from -4 to 0 hr within each individual (data not shown). 120 100 80 60 40 O 120 I00 Serum FSH in ng/ml 0) (I) O O 39 GROUP l ‘\ H5 ° H56 i l L l 4 l J l l I l l 1 '3 O 3 6 9 |2|5 |82l242730 Hours After Exogenous Estradiol (1.0 mg, IM) 40 of 59, 92, 105, and 116 ng/ml measured at 0.5, 1.0, 1.5, and 2.0 hr post GnRH, respectively. 41 Figure 4. Mean i SEM serum LH and FSH in 12 dairy cows with naturally- occurring ovarian cysts after administration of 100 ug GnRH in 0.15M NaCl. GnRH was given 30 hr after administration of 1.0 mg estradiol-17b. 42 2:9. 5 Iwu Eacom 5 o 5 0 7 5 2 0 5 0 1 q - _ a d L TIT///////.//////////////// L ..ll..7/////////////////////////A l M 111% E mum”. a .. ... VI; - _ "D— _ r _ _ a O o o 0 0 O _E\mc c_ I1. EDEmw 1.0 1.5 2.0 Hours after GnRH 0.5 O 43 DISCUSSION OF EXPERIMENT I In an abbatoir survey, 30% of bovine reproductive tracts with ovarian cysts also had a corpus luteum (Al-Dahash and David, 1977a). These may have represented situations where cystic structures remained after a cow spontaneously resumed estrous cycles or responded to gona— dotropin therapy. P4 concentrations recorded in the present study indi- cate that the occurrence of undetected corpora lutea was unlikely. Classification of ovarian cysts was based on the assumption that high E2 concentrations were accompanied by active granulosa tissue and P4 was positively related to the extent of thecal lutenization. In this study, the two cows classified as having luteal tissue (Group 3) had circulating P4 concentration between 0.5 and 1.0 ng/ml. Others (Leslie and Bosu, 1983; Nakao et al., 1983) did not consider ovarian cysts as having significant luteal activity until serum or skim milk con- centrations of P4 were above 1.0 ng/ml. Within the capabilities of the present assay used, further work is necessary to determine minimal P4 concentrations that would be considered as having P4-mediated influence on gonadotropin release. Correlation of histological anatomy of ovarian cysts and sex steroid content in cyst fluid showed that when granulosa cells were pre- sent, E2 usually accumulated. If not, steroidogenesis was interrupted at the conversion of P4 to 17-alpha-hydroxyprogesterone (Choi et al., 1982/1983). Androgens of thecal origin thus should not accumulate. This is supported by the lack of differences in testosterone among 44 groups in this study and other reports showing no difference in serum testosterone between cycling cows and cows with ovarian cysts (Kesler et al., 1979; Nessan and King, 1981). Our data depict circulating sex steroids in one discrete point in time. Kesler et al. (1980) showed great variation in E2 and P4 among and within untreated cystic cows in a 30 day period. Inspection of their hormone data across time suggested that significant F; concentra- tions (> 6 pg/ml) were maintained in some cows while increases in P4 (> 1.0 ng/ml) were much shorter-lived. Further work is needed to determine how long the high circulating E2 concentrations seen in Group 1 cows could be maintained. The follicular basement membrane was absent in almost all ovarian cysts examined histologically (Al-Dahash and David, 1977b) and this may allow for prolonged granulosa cell viability. Peripheral sex steroid concentrations measured in this study resemble values seen in the periovulatory period of the bovine estrous cycle. Apart from the preovulatory surge, LH secretion at that time was characterized as having frequent, low amplitude pulsatile patterns (Rahe et al., 1980; Schallenberger et al., 1984; Walters and Schallenberger, 1984). A similar secretion pattern was seen in postpartum cows just prior to the initiation of estrous cycles (Gitlin et al., 1983). LH release in many of the 12 cows sampled in this study qualitatively resembled these patterns. Sampling intervals of 4-5 min were necessary to delineate small amplitude, short-lived LH pulses in ewes (Karsh et al., 1983) and cows (Walters et al., 1984) and therefore the 15 min ~interval employed in this study may not have detected small pulses in 45 some individuals. This may have contributed to the elevated baseline LH in Group 1 cows. Episodic FSH secretion patterns have not been well characterized in cattle. This is partly due to controversy among assay results regarding relationships between LH and FSH release (Bolt and Rollins, 1983). Within the confines of our ability to detect pulsatile release of gona- dotropins, release of LH and FSH generally occurred simultaneously. This is in agreement with data from cycling cows (Schallenberger et al., 1984; Walters et al., 1984; Walters and Schallenberger, 1984). There was considerable variation among the cows with ovarian cysts in baseline FSH, but the range of values was not different than those seen in post- partum or cycling cows (Carruthers et al., 1980; Schallenberger et al., 1984; Walters et al., 1984; Walters and Schallenberger, 1984). Two mem- bers of Group 1 had FSH values consistently above 90 ng/ml, which could also be consistent with the elevated values seen in OVX heifers (Ireland at al., 1983). There was no evidence for stress-related increase of cortisol causing suppression of gonadotropin release. Simultaneous collection of jugular and caudal vena cava (Walters and Schallenberger, 1984) or utero-ovarian venous blood (Ireland at al., 1984) has provided information on interactions in gonadotropin and E2 secretion. In proestrus, pulsatile release of E2 was synchronous with pulsatile release of LH (Walters and Schallenberger, 1984). The cows in Group 1 had high serum E2, low serum P4, and LH or FSH pulse frequencies ranging from 0 to 4/4 hr. Further work is needed to first 46 confirm whether the ovarian cyst is the site of E2 production and if so, whether E2 secretion is independent of pulsatile gonadotropin release. None of the Group 1 cows released LB or FSH above basal con- centrations after exogenous 82. These results suggest that the high endogenous E2 concentrations may have provided negative feedback on hypothalamic centers involved in the positive feedback response to E2. In rats, 2-hydroxyestrone, an E2 metabolite, may interfere with the interaction of catecholaminergic and GnRH-producing neurons and prevent release of the preovulatory surge of LH (Okatani and Fishman, 1984). A similar mechanism may be in effect in the Group 1 cows. Gonadotropin release after exogenous E2 was extremely varied in the cows in Groups 2 and 3. The responses could be classified as no increase, transient release (LH increase, but less than 7 ng/ml), or a preovulatory—like surge of LH. A similar variety of responses was seen after E2 was given to postpartum dairy cows, where the interval to LH peak and magnitude of peak LH were inversely and positively related to postpartum interval, respectively (Karg and Schallenberger, 1983; Stevenson et al., 1983). P4 inhibits the positive feedback effect of E2 on LH release, so lack of or a transient release of LH after exoge- nous Ez would be an expected response in Group 3 cows (Zaied et al., 1981; Kesner et al, 1982). If the formation or maintenance of ovarian cysts is caused by a reduced sensitivity to the positive feedback effect of E2, then the graded increases in LH could be indicative of stages of recovery of hypothalamo-pituitary sensitivity to E2. Turnover of cystic 47 structures was detected in the present study, in support of data from Kesler et a1. (1980). All cows released both LH and FSH after exogenous GnRH, as expected from the results of other studies (Kittok et al., 1973; Seguin et al., 1976; Kesler et al., 1979c). The uniformly low LH response by the cow having the post E2 surge of gonadotropins is suggestive of depletion of releasable pituitary gonadotrOpin (Convey et al., 1977) and/or refrac- toriness to GnRH (Kesner et al., 1981; Halters and Schallenberger, 1984). Although the exogenous GnRH was in essence a treatment for ovarian cysts, the cows were not monitored for the return of estrous cycles since the potential influence of exogenous Hz was not controlled. In summary, cows with ovarian cysts were grouped according to E2 and P4 profiles. The classification groups represented individuals with high endogenous E2 (Group 1), low E2 and P4 (Group 2), and P4 concentrations suggestive of luteal activity (Group 3). The fre- quency and amplitude of pulsatile LH or FSH release were variable, with no characteristics unique to any group. Baseline LH was higher in Group 1 cows (p < 0.05). Given the present knowledge on pulsatile release of LH and FSH, patterns seen in the cows with ovarian cysts were similar to what would be expected in postpartum or cycling cows, apart from the preovulatory surge of gonadotropins. When given exogenous E2, none of the six cows in Group 1 responded with an increase in LH and FSH over basal concentrations. But all mem- bers of the group released LH and FSH after GnRH. This suggested that the hypothalamic centers involved in initiation of the preovulatory 48 gonadotropin surges were either not responsive to or under inhibition by the high endogenous E2 concentrations. Of the four cows in Group 2, one had a preovulatory gonadotropin surge, one had transient release of LH, and the others showed no increase in baseline gonadotropin concentrations after exogenous E2. This group, with its low endogenous E2 and P4 concentrations were expected to be the best candidates to show a positive feedback response to exogenous 82. The lack of a preovulatory surge of gonadotropins in 3 of the 4 cows supports an earlier hypothesis of a reduced hypothalamic sensitivity to E2 having a role in ovarian cyst formation (Kesler et al., 1979a; Kesler et al., 1980; Zaied at al., 1981). As previously discussed, the endogenous P4 concentrations in Group 3 cows may have been sufficient to suppress the positive feedback response to exogenous 132. Because the incidence of ovarian cysts in dairy cows is highest in the first 60 days postpartum, the cows in this study represent a less common subgroup at six months or more postpartum. Further work is needed to determine if similar relationships between sex steroid produc- tion and gonadotropin release are seen in early postpartum cows with ovarian cysts. Future therapeutic studies should attempt to classify ovarian cysts by E2 and P4 profiles and address efficacy of treatment for each category. EXPERIMENT II Formation of Ovarian Cysts in Dairy Cows After Exogenous Estradiol or ACTH and Endocrine Profiles in Cows With Experimentally-Induced Ovarian Cysts 49 50 BACKGROUND T0 EXPERIMENT II Steroidogenic capabilities of bovine ovarian cysts are probably determined by several factors. The development and viability of granu- losa or luteinized thecal tissue are directly related to concentrations of estradiol-17b (E2) or progesterone (P4) in cyst fluid, respectively (Choi et al., 1982/1983). Episodic release of LH and FSH varies greatly among cows with ovarian cysts (see Experiment 1) and the contribution of pulsatile gonadotropin release toward maintenance of steroidogenesis by ovarian cysts is unknown. A recent review (Eyestone and Ax, 1984) emphasized that most endocrine data from cows with ovarian cysts was collected at unknown intervals after cyst formation. It is not known if the morphological or endocrine variation seen with ovarian cysts repre- sent stages of a single or multiple pathophysiologic mechanisms. Ovarian cysts have been induced in cattle, intentionally or not, by a variety of treatments given in late diestrus or proestrus. Such treatments have included exogenous E2 (Wiltbank et al., 1961; Nadaraja and Hansel, 1976), exogenous ACTH (Liptrap and McNally, 1976), or anti- serum against bovine LH (Nadaraja and Hansel, 1976). Exogenous E2 given concurrently with manual CL enucleation resulted in formation of ovarian cysts lined with grossly-visible luteal tissue (Whitmore et al., 1972). In one study where ovarian cysts were formed after exogenous E2, changes in peripheral serum E2 and P4 were extremely variable among affected heifers (Nadaraja and Hansel, 1976). After exogenous ACTH, cyst for- mation was characterized by a rise of serum E2 to proestrual 51 concentrations that were then maintained for several days until ovula- tion after spontaneous cyst regression or treatment with human chorionic gonadotrOpin (Liptrap and McNally, 1976). The mechanisms whereby ovarian cysts are formed after exogenous E2 or ACTH are unknown. Nadaraja and Hansel (1976) speculated that a pharmacologic dose of E2 would cause luteal regression with a premature surge of LH that could alter follicular growth or maturation. ACTH treatment blocked the preovulatory surge of LH in synchronized heifers (Stroebel and Moberg, 1982), probably through a cortisol-mediated inhi- bition of LH release and increased P4 production by the adrenals (Stoebel and Moberg, 1982; Li and Wagner, 1983; Echternkamp, 1984; Watson and Munro, 1984). An experiment was conducted with two objectives. The first was to induce ovarian cysts in dairy cows with exogenous E2 or ACTH and measure LH and sex steroid changes accompanying cyst formation. The second was to characterize basal gonadotropin release and the response to exogenous E2 in cows with artificially-induced ovarian cysts. MATERIALS AND METHODS FOR EXPERIMENT II I. Induction of Ovarian Cysts Animals and Treatments. Fifteen adult (14 Holstein and 1 Guernsey) parous cows were brought to the Michigan State University Veterinary Clinical Center in August, 1980. The cows were purchased at an auction market and age, parity, and previous reproductive history were unknown. .Criteria for selection included complete uterine involution and evidence of estrous cyclicity as determined by palpation per rectum. 52 The cows were kept in tie stalls throughout the course of the experiment. Water was provided 22 libidum. Bromegrass-alfalfa hay was fed twice daily and trace-mineral salt was given to each cow twice weekly. On Day -28 of the study, 7 cows were diagnosed by palpation peg rectum as being in diestrus and were given 25 mg prostaglandin F2a (PGFZa, Lutalyse, Upjohn, Kalamazoo, MI) intramuscularly (IM). On Day -18, all cows were diagnosed as having a corpus luteum (CL) and were given 35 mg PGFZa, IM. The intent of this synchronization schedule was to have all cows in mid- to late diestrus at the time of treatment. On Day -1, an indwelling jugular cannula was inserted in each cow. Cannulae were kept patent with heparinized saline. Rectal examination results indicated that 3 cows had begun CL regression and were in proestrus or estrus at this time. These individuals were blocked among each of three treatment groups. The other cows were randomly assigned to the 3 treatment groups: 1) controls, receiving an IM injection of safflower oil, 2) 10 mg E2 (E210) in safflower oil (Sigma Chemical Co., St. Louis, MO), 3) 100 IU ACTH gel (AdrenomoneR, Burns-Biotech, Omaha, NE) given subcutaneously and repeated at 12 hr intervals for 10 con- secutive days. Oil, E210 and the first ACTH injection were given at 12 noon on Day 0. Blood Collection and Rectal Examination Schedules. Twenty ml of blood was drawn just prior to treatment on Day 0. Ten or 20 ml blood samples were then collected at 4 hr intervals from Day 0 through 12 noon on Day 8. Blood was allowed to clot for 4-8 hr at 20 C and then kept at 53 4 C for 12-16 hr before centrifugation at 1000 g for 15 min. Sera were poured into 12 x 75 mm glass tubes and stored at -20 C before assay. Reproductive tracts were examined by palpation pe£_rectum (Zemjanis, 1970) on Day -1 and then daily from Days 2 through 20 of the experiment. Description of ovarian structures and uterine consistency were recorded. Hormone Assays. Hormones were quantified by radioimmunoassay tech- niques. LH (Convey et al., 1976) was measured in all samples collected at 4 hr intervals. E2 (Appendix) was measured at 8 hr intervals from Days 0 through 8. Serum P4 (see EXPERIMENT I) was measured at 8 hr intervals from Days 0 through 4 and then once daily from Days 5 through 8. Cortisol (Roth et al., 1983) was measured at 4 hr intervals for the first 36 hr of the experiment and then at 4 hr intervals throughout Days 3 and 5. Definition of LH Release and Statistical Analysis. A preovulatory surge of LH was defined as a peak of at least 10 ng/ml magnitude lasting for 2 or more consecutive 4 hr interval samples. Cortisol changes among treatment groups were analyzed by split-plot analysis of variance with differences among groups identified with Scheffe's interval (Gill, 1978). II. Endocrine Profiles of Cows With Experimentally-Induced Ovarian Cysts Blood Sampling Schedule. On Day 15 of the cyst induction experi- ment, cows that were identified as having ovarian cysts were submitted to the sampling and treatment regimen described in EXPERIMENT I. In 54 short, blood samples were collected at 15 min intervals from 8 am through 12 noon. One mg E2 in safflower oil was then injected IM and additional blood samples were drawn every 3 hr for 30 hr to determine if a preovulatory surge of gonadotropins occurred. Hormone Assays. LH and FSH (Carruthers et al., 1980) were measured in all samples collected. E2 and P4 were measured in the first and last hours of the 15 min sampling period. Definitions of Gonadotropin Release. In 15-min samples, pulsatile release of LH or FSH was identified as values exceeding the previous or second previous sample by the 95% confidence interval of the assay (Convey et al., 1983). These intervals were 1.0 ng/ml for LH and 10 ng/ml for FSH. Average overall LH or FSH was the mean of all samples drawn at 15 min intervals. Criteria for identification of a preovulatory surge of LH were described in Materials and Methods for cyst induction. A preovulatory surge of FSH was defined as an increase of more than 50 ng/ml above average FSH, lasting for at least two consecutive 3 hr interval samples. Evidence for an increase in basal LH or FSH secretion was assessed on a within-cow basis, by identification of values more than three standard deviations above the average overall mean. 55 RESULTS OF EXPERIMENT II 1. Induction of Ovarian Cysts Control Group. All control cows showed palpation and hormonal changes indicative of CL regression, ovulation, and new CL formation. Ovarian cysts were not diagnosed at any time. Despite prior estrus synchronization measures, naturally-occurring estrus was estimated to occur as early as Day -3 and as late as Day 9 in control animals. As a result, preovulatory surges of LH were detected in 3 of 5 individuals. Changes in serum LH, P4, and E2 for the control cows that underwent luteal regression after Day 0 of the experiment are shown on Figure 5. LH surges were preceeded by a decline in serum P4, with concentrations remaining below 0.3 ng/ml for 20-24 hr before the surge of LH. The three preovulatory surges of LH ranged from 18.5 to 28.5 ng/ml in magni- tude and were of at least 8-10 hr duration. Serum E2 increased in a 24 hr period prior to the LH surge, with peak values ranging from 7 to 16 pg/ml. Serum E2 decreased abruptly during the LH surges, to con- centrations below 2 pg/ml. The LH surges occurred early enough in the sampling period in two individuals to record the rise in P4 accompanying CL growth. Serum P4 was above 0.8 ng/ml by 4-5 days after the LH surge in these individuals. Serum P4 increased progressively from 0.1 to 1.5 ng/ml from Days 0 to 3 in the cow that ovulated prior to Day -1 (data not shown). Serum P4 remained between 1.7 to 2.9 ng/ml from Days 3 through 8 in this indi- vidual. Palpation results indicated that this latter cow was returning (to estrus on Day 19 and 20 of the experiment. 56 Figure 5. Temporal changes in serum LH, progesterone and estradiol-17b in adult Holstein cows given 3.0 m1 safflower seed oil on Day 0. All cows showed changes of the reproductive tract, as detected by palpation per rectum, indicative of spontaneous luteal regression and ovulation. 57 EoEtooxm so >60 he: -8 a: lw/Bd u! (v-v) logpensa JO lwlfiu u! (o—o) euommefimd 10 (H) H‘I 1.1111169 58 E210 Treated Group. Peak serum E2 concentrations ranged from 200-600 pg/ml at 8-16 hr post injection and then progressively declined to values below 3 pg/ml by Day 3 (see Figure 6). Marked vulvar swelling and hyperemia was noted for a 1-2 day period in all cows given E210. Four cows had a palpable CL with serum P4 ranging from 3.5 to 5.4 ng/ml just prior to treatment. Serum P4 declined rapidly after E210 treatment, with all cows showing palpation changes indicative of luteal regression on Day 2. As P4 was decreasing, surges of LH were recorded in these four cows, beginning at 16-20 hr post-injection (Figure 6). These surges of LH were of 12-24 hr duration with peak values ranging from 12 to 44 ng/ml. Ovarian cysts were diagnosed in 2 cows after E210-induced luteoly- sis. In cow H3, the ovaries remained inactive after CL regression until a 10mm diameter follicle was identified on the left ovary on Day 7. This follicle increased to 25 mm in diameter by the next day and was diagnosed as an ovarian cyst. An increase in serum E2 accompanied devel- Opment of the cyst, with a concentration of 9 pg/ml measured on Day 8 (Figure 6). Cow H10 was diagnosed as having a 35mm diameter follicular cyst on the right ovary on Day 7. This structure was thought to origi- nate from a 15mm follicle first detected on Day 5. An increase of serum P4 accompanied formation of this ovarian cyst, with concentrations above 2 ng/ml by Day 7 (Figure 6). Two other cows with E210-induced luteal regression did not form ovarian cysts. However, neither animal had palpable evidence of CL .growth or an increase in P4 by Day 8. Instead, one cow had a spontaneous Figure 6. 59 Temporal changes in serum LH, progesterone and estradiol-17b in adult dairy cows given 10 mg estradiol-17b in safflower seed oil on Day 0. Luteal regression, as determined by ovarian palpation peg rectum, occurred in all cows shortly after treatment. An ovarian cyst was diagnosed in each of cows H3 and H10 on Day 7. Cows H8 and H24 did not form ovarian cysts, but ovarian structures were not identified by palpation until detection of luteal tissue on Days 12-13. 60 EoEtooxm so can b q «0 d0 0 l1 _ _ _ _ __ _ _ _ _ _ _ _ .. _ _ /\«L VNI M 18~ OFI a 68 L00» eewmwmmflw 100. c 400. 1 OON ’ 4 SN m: , MI , a 1ooo « Jco... 1w/0d ug (v-v) |ogpens3 JO 1111/6u u! (o—o) eu0181seBOJd 1o (H) H'l wnles 61 preovulatory surge of LH occurring on Day 5, the other on Day 8, (Figure 6). These surges were of 8-12 hr duration and reached peak values of 22 and 42 ng/ml. Maximum serum E2 values were 4-5 pg/ml just prior to the LH surges, with a decline to 1 pg/ml occurring while the surges were in progress (Figure 6). Moderate uterine tone was detected on the day of the LH surge for both cows, but ovarian follicles of preovulatory size were not detected. Both cows showed evidence of CL growth as detected by palpation, by Days 12 or 13 of the experiment. One additional cow ovulated prior to E210 injection. Palpation observations and serum P4 changes were indicative of CL growth (not shown). This cow showed palpation changes characteristic of CL regression and estrus on Days 18-20. ACTH-Treated Group. Mean i_SEM concentrations of cortisol in serum samples collected on Days 1, 3 and 5 of the experiment are shown on Figure 7. Cortisol was higher in ACTH-treated cows at all times sampled except 12 and 24 hr after the start of the experiment (p < 0.05). Cortisol means from control and E210-treated cows did not differ. Four of S cows had a palpable CL with serum P4 concentrations ranging from 1.9 to 5.4 ng/ml just prior to the start of ACTH treatment. Ovarian cysts formed in 2 of these individuals. These cows (H12 and H22) had palpation evidence of CL regression on Days 2 and 3, respec- tively. CL regression was followed by the appearance of a single 20-25mm diameter follicular cyst one day later in each animal. CL regression was not accompanied by a typical decline in serum P4, where Figure 7. 62 Changes in serum cortisol in nonlactating adult dairy cows after administration of oil (controls) or estradiol-17b at 0 hr, or 100 IU ACTH gel given at 12 hr intervals. ACTH treat- ment was continued throughout the periods where cortisol was not measured. Means marked with * are higher than other means within the same time (p < 0.05). EoEtooxw Lo .950: 8.8.8.8. 93:11.8 om on N» 881% mmmm em ON 6. N. m e — q _ — _ 41\_ u — — _ _1)d - _ _ _ _ _ fi _ 1— II E luv 1H! \3 .- . 411a 14x ., 2 1 H . I ¥\H\ . [H so .6 .285... sec. 1.. .. 6 see so 6 :54 3.00. .1. .5528 1 .. ON Om OOOOO mNCOlDV 00 gm (9:1) lw/fiu u! 10911109 wnleg as '1' x Figure 8. 64 Temporal changes in serum LH, estradiol and progesterone in nonlactating adult Holstein cows given 100 IU ACTH gel at 0 hr and repeated at 12 hr intervals for 10 days. Cows H12 and H22 each formed an ovarian cyst, first diagnosed by palpation peg rectum on Days 2 and 3, respectively. Cow H20 underwent luteal regression and became anestrus. Cow H1 did not undergo luteal regression during the sampling period. «cmEtooxm so .30 \O'o \OIOIO V ‘10 1 O. ONI ...? -.r. < 4 IO 1 N NPI 00 ~- 0 1n «1.1/Gd u! (v—v) logpensa JO 1111/5u u! (0..) eu0191s9601d JO (Hun wnles 66 concentrations remained between 0.7-2 ng/ml in cow H12 and 0.3-1.5 ng/ml in Cow H22 just before and after cyst diagnosis (Figure 8). Cow H12 had 2 short-lived peaks of serum £2 above 10 pg/ml prior to cyst formation, while transient increases of serum E2 did not rise above 6 pg/ml in cow H22 (Figure 8). Preovulatory surges of LH did not occur. The two other cows with a CL present at the onset of ACTH treatment did not form ovarian cysts, but had altered estrous cycles. Cow H1 did not show palpation evidence of CL regression until Day 20. Serum P4 remained above 2.0 ng/ml in all sera assayed and an LH surge was not observed. Palpation results showed that CL regression had occurred after the Day -18 PGFZa treatment in this cow and post-estrual bleeding was observed on Day -15. Thus, Cow H1 apparently had an interestrous interval of approximately 35 days. Cow H20 showed palpation changes and a decrease in P4 on Days 2-3 that were indicative of CL regression (Figure 8). By Day 6, this cow had clinical mastitis and maintained a rectal temperature above 40 C for several days despite systemic anti- biotic therapy. After CL regression, there was no surge of LH and the ovaries remained inactive through the course of the experiment. The remaining cow that received ACTH was thought to have ovulated on Days 0 or 1. This cow had palpation changes and an increase in serum P4 (not shown) indicative of CL growth. This individual subsequently had palpation characteristics of CL regression and estrus on Days 19-20. 11. Gross Ovarian Changes and Endocrine Profiles of Cows with Induced Ovarian Cysts Ovarian Structures. On Day 15 of the experiment, ovarian cysts that were initially formed after E210 or ACTH treatment were still 67 present as determined by palpation per rectum. Since the time of cyst diagnosis, no changes in ovarian structures were detected in Cow H10. A second 25 mm follicular structure was detected on the ovary ipsilateral to the original ovarian cyst on Days 9 and 14 in cows H22 and H3, respectively. The two ovarian cysts were still present by Day 15 in H22, but the second cystic structure had disappeared one day later in H3. An additional 10-15 mm follicle was detected on each of Days 8 and 14 in Cow 12 and both follicles were still present on Day 15. Basal GonadotrOpin and Sex Steroid Secretion and Response to Exogenous 82. Serum E2 and P4 concentrations and patterns of basal LH and FSH release are shown on Figure 9. Sex steroid concentrations, average overall LH and FSH and numbers of episodic LH and FSH pulses/4 hr are also summarized on Table 2. Although there were few animals and considerable variation, there was suggestion of some relationships bet- ween E2! P4 and LH release. Basal E2 and P4 concentrations appeared inversely related. There also appeared to be an inverse relationship between both numbers of episodic pulses of LH and overall average serum LH with serum concentrations of P4. The increases in serum E2 or P4 that were associated with cyst for- mation in Cows H3 and H10, respectively, were still evident on Day 15. Serum E2 in Cow H3 was 18.6 pg/ml, a concentration similar to those seen in control cows prior to preovulatory surges of LH. Pulsatile release of LH and FSH occurred in a synchronous manner in H3. Sex steroid con- centrations in H10 were indicative of the presence of functional luteal 68 Table 2.--Serum sex steroid and gonadotropin release in 4 adult dairy cows with experimentally-induced ovarian cysts. Individual Cowa Item H3 H10 H12 H22 Serum Estradiol (pg/m1) 18.6 3.0 2.0 6.6 Serum Progesterone (ng/ml) 0.2 1.5 0.6 0.2 LH pulses/4 hr 3 0 0 2 )5: SD Average LH (ng/ml) 1.7:0.6 0.7:0.l 1.01:0.3 1.73:0.9 FSH pulses/4 hr 1 1 5 2 31' i 80 Average FSH (ng/ml) 52:7 40:6 58:10 49:7 aFrom samples drawn 7-12 days after ovarian cyst formation. tissue, with no LH pulses and one FSH pulse/4 hr. Neither H3 or H10 had increased LH or FSH after exogenous E2 (Figure 9). Both cows forming ovarian cysts after ACTH treatment also showed evi- dence of ovarian steroidogenesis. P4 concentrations in Cow H12 suggested the presence of some active luteal tissue (0.6 ng/ml). This individual had 5 pulses of FSH in the 4 hr sampling period, but no pulses of LH. After 1.0 mg E2 challenge, Cow H12 had an increase in serum LH and FSH over basal values at 15-21 hr, but not of the magnitude of preovulatory surges. At the time of sampling, Cow H22 had evidence of significant E2 production (7 pg/ml), with pulsatile release of both LH and FSH were seen in 15 min samples. Cow H22 had a preovulatory surge-like release of both LH and FSH. at 9-15 hr after exogenous Ez (Figure 9). Figure 9. 69 Pulsatile release of LH and FSH and estradiol (E2) in 4 adult cows with ovarian cysts. Corresponding basal concentrations are also listed. LH with an *. Post-E2 increases of LH are marked with a *. response to exogenous experimentally-induced E2 and progesterone (P4) or FSH pulses are marked or FSH above basal values 7O _oEszmm asocoooxm Lots 2:0: .36 so 65.... OmnmvN _N m. m. N. m w m «wocfi Eu: 609 gm Eom . — - - d1 u ‘1‘ 4 - _\‘- q _ _Eoewo" we :23 1 _ssemouum 22mm 1 .320 Eu... 1 «u... 260 on NNVN _N m. 9 N_ m w mrwoocN_ ED: EGO. gm F50 _ q _ s — q a . 1 _ q _ _1 . .53.. mo. .m. 22.8 - mmhwnvnw _exoeoahm gram - .580 1.5.... 1 we... 260 Q 9 _o_om.:mm aaocouoxm cote 3:0: >06 Co 66:. On New mm. m. N. m 0 names. 62. 68. 68 ram a .1 _ u a — a — — uV — _ _ q {bl/\Xi 9101;111:0104 1 N . m use. .5331“. 23mm - .Eoe onuwm 28mm 1 ozone 2.9566; mm 102.. Boo On NNVN _Nm_ 0. N. m w m cWWcQ F6: :60. chm 60m a a q . . _ _L—___ 2111. . \ _ v mmkwmvnm a 1 mm 9:0. . 1 _esemouve 23mm 1 2‘86.me 55m % nsoew «c6530.; Nm 1 m: 300 QN‘DWVMN 9 Iw/Bu U1 (o—o)01/Hs:1 J0 (H) H‘I wmes 71 DISCUSSION OF EXPERIMENT II Temporal changes in serum LH, E2, and P4 concentrations occurring with CL regression and the preovulatory surge of LH in control cows were similar to previous description of normal bovine estrous cycles (Hansel and Convey, 1983). Control cows also showed palpable changes in ovarian structures and uterine consistency that were indicative of normal CL regression, ovulation, and subsequent CL development (Zemjanis, 1970). Three cows were blocked among the treatment groups because they had either just ovulated or begun estrus prior to the initiation of respec- tive treatments. None of these cows formed ovarian cysts. Palpation data indicated that all of these three cows had a normal interestrous interval over the course of the experiment. Wiltbank et al. (1961) reported a shortened interestrous interval in 6 of 10 cycling heifers when given exogenous E2 on days 3 or 4 of the estrous cycle. Heifers receiving continuous infusion of ACTH, beginning on day 2 of the estrous cycle had no preovulatory surges of LH 18-20 days later (Li and Wagner, 1983). When given early in the estrous cycle, the ability of exogenous E2 or ACTH to alter estrous cycle length may depend on the day of cycle when treated or duration and dosage of treatment regimen. Four cows had a CL present at the time of E210 treatment. With pre-treatment estrous synchronization procedures, these animals would have been in mid- to late diestrus. Ten mg E2 was a pharmacological dose and its effects appeared to be both luteolytic and directly stimu- latory to LH release. The interval from E210 to the onset of the surge 72 of LH was 16-20 hr, an interval similar to the positive feedback response of LH after exogenous E2 was given to OVX cows (Short et al., 1973; Kesner et al., 1981). However in the present study, these induced surges of LH occurred before luteal regression was complete. These were likely to be dose-related effects of E2, since LH release did not occur in diestrous cows after 1 or 2 mg exogenous E2 (Short et al., 1979; Zaied et al., 1981). Luteal regression and LH release after E210 treatment apparently did not directly result in ovulation. Instead, 2 cows each either formed an ovarian cyst or had a spontaneous preovulatory surge of LH 5-8 days later. At present, it is not known what factors may have deter- mined whether an ovarian cyst formed or a preovulatory surge of LH occurred. Although the number of cows in this study was small, the percentage forming ovarian cysts after E2 treatment (50%) was between the pre- viously reported 44% (Whitmore et al., 1972) and 70% (Wiltbank et al., 1961) occurrence rates. Nadaraja and Hansel (1976) described formation of ovarian cysts in 5 heifers given exogenous E2 on day 15 of the estrous cycle, but it was not clear if that was the total number treated. The diversity of sex steroid changes accompanying ovarian cyst for- mation after E210 treatment suggest the possibility of different pathophysiologic mechanisms. The lack of a preovulatory surge of LH in response to the rise in serum E2 (Cow H3) supports a previous hypothesis of reduced hypothalamic sensitivity to the positive feedback effects of 73 E2 on LH release (Kesler et al., 1979b; Kesler and Garverick, 1982). Marked release of LH occurred before CL regression was complete, as pro- posed by Nadaraja and Hansel (1976). This might result in premature luteinization of a growing follicle and account for the rise in serum P4 that accompanied cyst development in Cow H10. At one week after ovarian cyst formation, the endocrine profile of Cow H3 was characterized as having high endogenous E2, episodic basal release of LH and FSH, and no increase in gonadotropin after exogenous E2. Those traits of gonadotropin release closely resembled LH and FSH secretion characteristics of the cows with naturally-occurring ovarian cysts and high endogenous E2 that were sampled in EXPERIMENT I. If the elevated E2 concentrations remained stable between Days 8-15 of the experiment in Cow 83, the lack of gonadotropin release after 1.0 mg E2 may be due to negative feedback of the hypothalamo-pituitary axis from chronic E2 exposure. In Cow H10, the increase in serum P4 that occurred with cyst for- mation would also be consistent with normal CL growth (Hansel and Convey, 1983). Whitmore et al. (1972) showed that the weight of luteal tissue present in an ovarian cyst could equal the weight of a normal CL. Such ovarian cysts could theoretically be susceptible to natural luteolytic mechanisms and spontaneous recovery. The low number of epi- sodic LH or FSH pulses/4 hr (Rahe et al., 1980; Walters et al., 1984) and lack of gonadotropin release after 1.0 mg E2 were expected results for a cow having functional luteal tissue (Short et al., 1979; Zaied et al., 1981). 74 Serum cortisol values after exogenous ACTH treatment were similar to those previously reported where preovulatory surges of LH (Stoebel and Moberg, 1981; Li and Wagner, 1983) or pulsatile LH release (Echternkamp, 1984) were suppressed in cattle. Also, serum cortisol means in control or E210-treated cows were similar to previous reports of normal cows (Roth et al., 1983; Li and Wagner 1983; Thompsen et al., 1984). Liptrap and McNally (1976) did not report percentages of cows forming ovarian cysts after ACTH treatment, so it is not known if the variety of responses seen in the present study were encountered. Three cows underwent luteal regression after ACTH treatments were begun. No preovulatory surges of LH were detected. It was likely that the per- sistent fever resulted in one cow becoming anestrus. The appearance of ovarian cysts shortly after luteal regression in the other two cows suggested that these follicles may have been recruited as preovulatory follicles. Formation of these cysts was accompanied by periodic fluc- tuations of serum E2 and P4, not the sustained increases in plasma E2 as reported by Liptrap and McNally (1976). In the present study, the rela- tive contribution of the adrenal and ovary as sources of circulating P4 was unknown. At best, serum E2 and P4 changes accompanying and following ovarian cyst formation after ACTH treatment could be called transient in nature. Based on previously discussed reports (Stroebel and Moberg, 1982; Li and Wagner, 1983), it was expected that preovulatory surges of LH would be suppressed by ACTH-mediated cortisol release. In both cows that formed 75 ovarian cysts after ACTH-treatment, ovarian follicles newly identified after Day 9 of the experiment may have been the source of circulating E2 and P4 measured on Day 15. The presence of pulsatile gonadotropin release and increase in gonadotrOpins after exogenous E2 suggested that these cows may have been in the process of reinitiating estrous cycles. One cow given ACTH had a prolonged estrous cycle. Administration of a synthetic glucocortacoid on day 10 of the cycle prolonged luteal function for 10 days in heifers (Kanchev et al., 1976). However, luteal regression was not prevented in heifers receiving exogenous ACTH infu- sion beginning on day 16 of the cycle (Li and Wagner, 1983). Morrow et al. (1966) reported a higher incidence of ovarian cysts in cows with clinical problems such as metritis, milk fever, ketosis, or clinical mastitis in the early postpartum period. Further work is needed to determine whether metabolic and environmental stresses could interfere with LH release and result in the formation of ovarian cysts, especially in the early postpartum period. In summary, ovarian cysts formed in cows undergoing luteal regression after receiving E210 or ACTH (4 of 6 cases, excluding the individual who became anestrus). Spontaneous surges of LH did not occur at the time of cyst formation. However, increases in E2 and P4 accompanying cyst formation after E210 treatment suggested the pre- sence of gonadotropin stimulus. The relative lack of E2 or P4 release in cows forming ovarian cysts while receiving ACTH was suggestive of suppression of gonadotropin release. Thus, ovarian cysts may be formed when either significant gonadotropin release occurs, but is asynchronous 76 with luteal regression and follicular maturation or gonadotropin release is inhibited. After cyst formation, a variety of sex steroid and gonadotropin secretion traits were seen in the 4 cows with artificially-induced ovarian cysts. In the respective cows with highest endogenous E2 or P4 concentrations (H3 and H10), the lack of gonadotropin release after exogenous E2 suggested that the positive feedback response of the hypothalamo-pituitary axis can be inhibited not only by P4, but E2. The appearance of additional ovarian follicles, presence of pulsatile basal LH and/or FSH release, and gonadotropin release after exogenous E2 in cows H12 and H22 may have represented transition toward resumption of estrous cycles. Further work is needed to determine whether these artificially- induced ovarian cysts are appropriate models for the naturally-occurring condition. Information needed would include diagnostic confirmation by direct visualization of ovarian structures, monitoring endogenous endocrine changes over a longer observation period, and histological evaluation of artificially-induced ovarian cysts. 77 GENERAL DISCUSSION The overall theme of this research was to study relationships bet- ween circulating sex steroid concentrations and gonadotropin release in dairy cows with ovarian cysts. This was done in cows with naturally- occurring or artificially-induced ovarian cysts. Gonadotropin secretion was measured in terms of characteristics of basal release and the abi- lity of the hypothalamo-pituitary axis to release gonadotropins in response to exogenous E2. Assessment of basal gonadotropin secretion was based on reports by others of characteristics of gonadotropin changes throughout estrous cycles, the postpartum period, and after ovariectomy. Assessment of gonadotropin release after exogenous E2 was based on the documented positive feedback response in ovariectomized cattle, prepubertal heifers, and dairy cows that are at least two weeks postpartum, but without a functional CL (Saba et al., 1976; Short et al., 1979; Zaied et al., 1981; Kesner et al., 1982; Schillo et al., 1983; Stevenson et al., 1983). The basic question was whether gona- dotropin release in cows with ovarian cysts was similar to that of cows with comparable sex steroid concentrations in other normal physiological states. All ovarian cysts diagnosed in the present experiments were of a thin-walled, follicular consistency (Roberts, 1971). No differences were noted in naturally-occurring or artificially-induced cystic struc- tures. Physical characteristics of the ovarian cysts and uterine con- sistency, as detected by palpation per rectum, did not provide insight as to the endocrine status of the cows that were studied. There was 78 considerable variation in E2 and P4 concentrations, but the cows with ovarian cysts could be grouped into three basic categories. These cate- gories reflected physiological states where cows were under high E2 influence (proestrus-like), low steroid influence (early postpartum or after preovulatory surges of gonadotrOpins), or under luteal influence (presence of a growing, mature, or regressing CL). Seven cows with ovarian cysts (6 naturally-occurring, l artificially-induced) had basal E2 concentrations (10-30 pg/ml) that were at or above peak proestrual values in control cows in EXPERIMENT 11 (7-16 pg/ml). In proestrous cows, pulsatile gonadotropin release was present at a frequency of approximately 1/hr (Rahe et al., 1980; Walters and Schallenberger, 1984). Thus, cows with ovarian cysts with 2 or more LH pulses/4 hr are probably not divergent from expected results in cycling cows (Figure 1). Some individuals with high endogenous E2 had no obvious pulsatile gonadotrOpin release (See Figure l) and further documentation would be needed to determine if this may be of special clinical significance in regard to prognosis for recovery. None of the 7 high E2 cows with ovarian cysts had gonadotrOpin surges after exogenous E2. This result is similar to the loss of the positive feedback release of LH that occurred in old female rats that were in a constant-estrus state, an anovulatory condition with high endogenous E2 (Lu et al., 1980). In OVX rats, replacement of E2 for 10 or 17 days resulted in a progressive inhibition of LH surges that was thought to be mediated via alterations of hypothalamic norepinephrine and serotonin metabolism (Walker, 1983). In the present experiments, 79 the duration of exposure to high endogenous E2 in cows with naturally- occurring cysts was unknown. The cow with the experimentally-induced ovarian cyst in this category (Cow H3, Figures 6 and 9) could possibly have had E2 concentrations at or above 10 pg/ml for a week. It is not known if experimental E2 replacement will result in inhibition of the positive feedback response in cattle. Five cows with ovarian cysts (4 naturally-occurring, l artificially-induced) had serum steroid concentrations that suggested a lesser likelihood of being under E2 (§_7 pg/ml) or P4 (§_0.5 ng/ml) influence. These individuals also had extensive variation in pulsatile gonadotropin release (Figures 1 and 9), however patterns were similar to those of postpartum cows, or cycling cows after the preovulatory surge of gonadotrOpins (Rahe et al., 1980; Peters et al., 1981; Gitlin et al., 1983; Karg and Schallenberger, 1983; Walters and Schallenberger, 1984). GonadotrOpin release after exogenous E2 was not consistent in this group, where two individuals had preovulatory surges, one showed a significant rise in LH above baseline, and the remaining two showed no increase in LH or FSH (Figures 2, 3, and 9). The two individuals having the preovulatory surges (Cows HM and H22) both had evidence of ovarian follicular growth when monitored by palpation peg rectum, and both had pulsatile release of LH and FSH (Figures 1 and 9) prior to E2 challenge. The variety of gonadotrOpin responses after exogenous E2 in this group were suggestive of those of early postpartum cows prior to initiation of estrous cycles (Zaied et al., 1981, Karg and Schallenberger, 1983; Stevenson et al., 1983). 80 Four cows with ovarian cysts (2 naturally-occurring and 2 artifi- cially induced) had evidence of functional luteal tissue. The indivi- duals (Cow H56, Figures 1-3; Cow H10, Figure 6 and 9) with the highest P4 concentrations had basal gonadotrOpin secretion that was similar in character to diestrous cows (Rahe et al., 1980; Walters and Schallenberger, 1984). These cows also showed no increase in LH or FSH after exogenous E2, an expected response in cows with functional luteal tissue (Short et al., 1979; Zaied et al., 1981; Stevenson et al., 1983). The remaining cows (HS, Figures 2 and 3; H12, Figure 9) had slightly lower serum P4 values and more variable patterns of pulsatile gonadotro- pin release. After exogenous E2, one individual showed an increase in LH over baseline (Figure 2) and the other had an increase in both LH and FSH (Figure 9). It is possible that the endogenous P4 concentrations in these latter individuals may have negatively influenced Ez-induced gona- dotropin release. The range of basal FSH concentrations in cows with naturally- occurring ovarian cysts was considerable (35-120 ng/ml, Figure 3), while concentrations were more uniform among cows with artificially-induced ovarian cysts (40-60 ng/ml, Table 2). Leidl et al. (1983) reported a lack of inhibin activity in bovine ovarian cysts, where injection of ovarian cyst fluid did not depress FSH in cows but injection of bovine ovarian follicular fluid resulted in a significant decrease of FSH. Perhaps the high basal FSH concentrations in some cows with ovarian cysts are the result of a lack of develOpment of other ovarian follicles. 81 In EXPERIMENT II, 10 mg E2 administration resulted in consistent luteolysis and LH release in four cows, but only two formed ovarian cysts. One explanation for this result may be the status of the ovarian follicle population at the time of E2 treatment, where cows that formed ovarian cysts had follicles that were affected by the LH surge. Convey et al. (1976) gave exogenous GnRH at selected intervals (10, 30 and 50 hr) after prostaglandin treatment in Holstein heifers. Heifers that received GnRH at 30 hr did not show estrus and 1 of 5 formed an ovarian cyst. Those results suggested that the timing of GnRH therapy may have resulted in premature luteinization of ovarian follicles and may have predisposed formation of the ovarian cyst. Further work needs to be done to determine if there is a critical day of the estrous cycle effect or ovarian follicular status that will result in a predictable high incidence of ovarian cyst formation after exogenous E2. Liptrap and McNally (1976) did not measure LH in cows forming ovarian cysts after ACTH administration, but reported significant increases in E2 that accompanied and were maintained after cyst for- mation. Similar increases in E2 were not seen with ovarian cyst for- mation after ACTH in EXPERIMENT II, where preovulatory surges of LH were not detected. Recently, Peters and Liptrap (1985) induced formation of ovarian cysts in sows with ACTH given in late-diestrus and proestrus. They reported no rise of E2 when gonadotropin surges were completely inhibited, but increased E2 with partial suppression of gonadotropin surges and ovarian cyst formation. Perhaps the extent of suppression of 82 LH release may account for the differences in E2 changes seen in the results of Liptrap and McNally (1976) and EXPERIMENT II. Echternkamp (1984) showed that stress in the form of acute restraint in previously unhandled beef cows resulted in increases in serum cortisol comparable to concentrations seen in ACTH-treated cows in EXPERIMENT II. Relocation of a dairy herd resulted in approximate doubling (9 to 20 ng/ml) of cortisol for 2-3 days, but it is questionable whether that would constitute sufficient stress to inhibit preovulatory surges of gonadotrOpins. The more frequent occurrence of ovarian cysts in cows with other early postpartum metabolic or disease problems (Morrow et al., 1966) may reflect instances of stress-related suppression of LH release. Although direct contrasts were not made, it appeared that rela- tionships between sex steroid profiles and gonadotropin release was similar in cows with naturally-occurring and artificially-induced ovarian cysts. This suggests that experimentally-induced ovarian cysts are a reasonable endocrine model for the naturally-occurring condition. Use of such a model would provide a means for further characterization of the endocrinology of ovarian cysts. This information could have cli- nical application in prevention or improvement of therapeutic measures for ovarian cysts in cattle. 83 SUMMARY AND CONCLUSIONS In EXPERIMENT I, 12 adult dairy cows with naturally-occurring ovarian cysts were grouped according to endogenous serum concentrations of E2 and P4. Assignment to groups reflected endocrine states of high E2 influence, low E2 and P4 influence, and luteal influence. The fre- quency of pulsatile release of gonadotropins varied considerably within high E2 and low steroid groups ranging from 0 to 4 per 4 hr. The fre- quency of pulsatile release of gonadotropins in cows with luteal influence tended to be more consistent (l to 2 per 4 hr). Overall, there were no differences among groups in frequency of pulsatile LH or FSH release, baseline FSH, serum testosterone or serum cortisol. Baseline LH was higher in the high E2 group. Pulsatile release of gona- dotrOpins in cows with ovarian cysts was variable, but characteristics were similar in general to patterns reported by others in postpartum or cycling cows of corresponding E2 and P4 status. Only 1 of 12 cows with naturally-occurring ovarian cysts had a preovulatory surge-like release of gonadotrOpins after exogenous E2 challenge. The low incidence of preovulatory surges after exogenous E2, especially in cows in high E2 and low steroid groups, suggested a refractoriness of the hypothalamo-pituitary axis to the positive feedback effects of E2. In EXPERIMENT II, non-lactating dairy cows in mid- to late-diestrus were given a placebo, 10 mg E2, or repeated injections of 100 IU ACTH to induce formation of ovarian cysts. Serum cortisol was higher in ACTH-treated cows than in cows receiving the other treatments. All cows 84 given placebo treatment underwent spontaneous luteal regression and ovu- lation. Administration of 10 mg E2 initiated synchronous luteal regression with preovulatory surges of LH occurring before luteal regression was complete. Two of 4 cows with Ez-induced luteal regression formed ovarian cysts, with one showing a progressive increase of endogenous E2 accompanying cyst formation and the other showing a progressive increase of P4. These results suggested that formation of ovarian cysts after exogenous E2 occurred with gonadotrOpin-supported ovarian steroidogenesis. Ovarian cysts were formed in 2 of 3 cows that underwent spontaneous luteal regression while receiving ACTH. The cow not forming an ovarian cyst became anestrus. Preovulatory surges of LH did not occur after luteal regression. Changes in serum E2 or P4 in cows forming ovarian cysts after ACTH were transient in nature. These results indicated that ovarian cyst formation may occur when the preovulatory surge of LH is suppressed. Serum E2 and P4 concentrations in the 4 cows with artificially- induced ovarian cysts indicated the presence of examples to fit each of the high E2, low steroid, and luteal-influence categories used to group cows with naturally-occurring ovarian cysts. Characteristics of both pulsatile basal gonadotrOpin release and gonadotr0pin release after 1.0 mg E2 in cows with artificially-induced ovarian cysts resembled those of cows with naturally-occurring ovarian cysts. Cows with artificially- induced ovarian cysts appear to be a suitable endocrine model for the naturally-occurring condition. APPENDIX Validation of Commercially-Available Radioimmunoassay Kits for Measuring Estradiol-17b and Testosterone in Bovine, Canine and Equine Serum APPENDIX Validation of Commercially-Available Radioimmunoassay Kits for Measuring Estradiol-17b and Testosterone in Bovine, Canine and Equine Serum INTRODUCTION The widespread application of radioimmunoassay (RIA) techniques has prompted the development and commercial marketing of RIA kits. Several classes of hormones, such as thyroid or adrenal and sex steroids are structurally similar among mammalian species, thus these commercial RIA kits may provide a readily-available source of assay materials for the researcher or diagnostician in animal science or veterinary medicine. Midgley et al. (1969) and Hafs et al. (1977) recommended that cri- teria for RIA performance include data on assay specificity, sen- sitivity, accuracy, precision, and parallelism. In the adaptation of a human diagnostic RIA kit for measurement of the structurally similar hormone in another species, the criteria of sensitivity and parallelism should receive particular attention. For instance, animal samples may not be accurately quantified if the normal physiological concentrations do not fall in the effective range of standard curves designed to opti- mally measure concentrations in human samples (Reimers et al., 1981). In cases where hormones are measured in serum, the equilibrium of the antibody-radioligand reaction may be affected by interactions with endo- genous serum proteins (Malvano, 1983). These non-specific effects may differ where standards are prepared in the serum of one species and unknowns are from a different species (Gostein and VanHaelst, 1973). Estradiol-17b and testosterone have been measured by RIA in a variety of biological fluids and variation due to reproductive status or pathology 85 86 within a species may be as diverse as variation among species. The pur- pose of this paper is to present assay validation data for the measurement of estradiol-17b and testosterone in bovine, canine, and equine serum using commercially-available RIA kits. MATERIALS AND METHODS RIA kits for the measurement of estradiol-17b or testosterone in diethylether-extracted serum were obtained from Serono Laboratories, 11 Brooks Drive, Braintree, MA (ESTRADIOL 1251 Kit and TESTO® RIA Kit, respectively). Each kit contained a phosphate buffer (pH-7.3) with 0.1% Na azide (PB), standards, antibody, and 125I-labeled radioligand (tracer). These reagents were in lyophilized form and were reconsti- tuted with double-distilled water, as per manufacturer's instructions. A solution of 20% w/v polyethylene glycol (PEG) for precipitation of antibody-bound tracer was also provided in each kit. Extraction Procedure - The serum extraction method was similar for both assays. Serum or reagent volumes for the estradiol validation will be noted in the text, and the appropriate volumes used in the testosterone extraction will be noted in brackets. Two [0.5] ml of double distilled water was extracted to determine background (extraction blank). Three— tenths to 3 ml [.3 to 1.5 ml] of serum was pipetted into 16x125 mm pyrex screw-topped culture tubes (Corning Glass Works, Corning, NY). Extraction tubes used in the estradiol RIA had been baked at 400 C for 4 hours and cooled to room temperature prior to extraction, in an attempt to remove residual contamination that would increase assay background (England at al., 1974). Ten [5] ml of freshly-opened anhydrous 87 diethylether (Mallinckrodt, Inc. Paris, KY) was added to each tube and the tubes capped with teflon-lined screw caps. Use of freshly-opened ether was critical to the estradiol assay. Extraction blank values from ether that had been exposed to air for at least 12 hr ranged from 20-30 pg/ml, whereas extraction blanks from freshly-opened ether were below 0.5 pg/ml. The tubes were laid horizontally in a shaker (Eberbach Corp., Ann Arbor, MI) and the contents mixed by shaking for 30 min. The extraction tubes were stood upright and the aqueous and organic layers were allowed a 10 min period to fully separate. The caps were removed and the aqueous layer was aspirated off the bottom of each tube with a glass Pasteur pipette. One hundred ul of 0.1N NaOH was pipetted into each tube, the caps replaced, and the shaking step was repeated for 10 min. The caps were removed and 100 ul of 0.1N glacial acetic acid was added to each tube followed by a final 10 min shaking period. This washing procedure was done to remove neutral lipids in the solvent phase via saponification with a basic solution and pH readjustment with the addition of acid. The tubes were placed in a -20 C freezer until the aqueous portion was frozen. The ether fraction was then decanted into 16x100 mm glass tubes. These latter tubes had also been baked prior to use in the estradiol RIA. The tubes containing the solvent fraction were placed in a 40 C warmed vortex-evaporator (Haake Buchler Inc, Saddle Brook, NJ) and the ether evaporated under vacuum. One ml of PB was added to each dried tube and the tubes were placed in a 37 C water bath for 1 hr for dissolution of the extracted hormone. The contents of each warmed tube was then mixed by vortexing for 15 sec. Aliquots of reconstituted samples were then added to assay tubes. Extraction 88 efficiency was assessed as the percentage of 3H-estradiol-17a (Carruthers and Hafs, 1980) or 3H-testosterone (both from New England Nuclear, Boston, MA) present in the reconstituted sample versus the amount added to serum prior to extraction. Tritiated hormone was added to 1, 2 or 3 ml of serum prior to extraction for the estradiol assay or to 0.5 or 1.5 ml serum prior to extraction for the testosterone assay. Effects of species and serum volume on extraction efficiency were tested by two-way analysis of variance. .Aggay Procedure - In an attempt to improve the estradiol RIA sen- sitivity and decrease expenses for a large number of research samples, the reconstituted estradiol tracer and antibody were further diluted in half with PB. Reagent dilution resulted in a shift of the standard curve with a tendency for the assay to be more sensitive (see Table 3). When bovine sera were assayed with diluted and undiluted reagents, there was virtually no difference in results (Table 3). Thus, further estra- diol validation data was obtained with the additional reagent dilution. The estradiol standard curves were modified with the addition of 1.0 and 2.5 pg/ml to the 0, 5, 10, 25, 50, 100 and 200 pg/ml standards supplied by the manufacturer. The 1.0 and 2.5 pg/ml standards were prepared with appropriate combinations of 0 and 5 or 10 pg/ml standards, respectively. Effects of further dilution of testosterone tracer and antibody have not been investigated. The testosterone standard curve was not modified; the 0, 0.05, 0.1, 0.2, 0.5, 1.5, 3.0 and 6.0 ng/ml standards supplied with the kit were all included in the assay. The following assay procedures were identical for both estradiol and testosterone RIA. Reagents were pipetted into 12x75 mm polystyrene 89 culture tubes (Walter Starstedt, Princeton, NJ). Total count tubes con- tained 100 ul tracer only. Non-specific binding tubes received 200 ul Table 3.--Influence of 50% dilution of both antibody and radioligand on standard curves and serum values in a commercial radioimmuno- assay for measurement of estradiol-17b. Estradiol Concentrations (pg/ml, r-2) Extrapolated From Standard Curves Level of binding inhibition Reagents 90% 75% 50% 35% Undiluted 4.3 14.0 45.6 89.3 Diluteda 2.2 8.0 29.0 60.3 Serum Estradiol Measured (pg/ml) Bovine Serum Samples (r-2) Reagents 1 2 3 4 5 Undiluted 18.8 3.0 12.1 20.4 11.0 Diluted 19.5 2.8 11.9 21.4 13.6 aReconstituted antibody and radioligand were further diluted in assay buffer. of 0 standard and 100 ul of tracer. The remaining assay tubes received 100 ul of the appropriate standard or sample, 100 ul of respective anti- body and 100 ul of tracer. The contents of the tubes were mixed by vor- texing for 5 sec. The assay tubes were covered and an 8—16 hour incuba- 90 tion period at room temperature (20 C) was allowed for equilibration of the antigen-antibody reaction. One ml of the 4 C PEG solution was added to all tubes except total count tubes. After PEG addition, the contents of the assay tubes were mixed by vortexing for 5-10 seconds and the tubes were kept at room temperature for 20 minutes for precipitation of protein (antibody-bound fraction). The tubes were then centrifugated at 1500 g at 4 C for 20 minutes. The supernatant was poured from all except total count tubes. The inverted tubes were allowed to stand for 5-10 minutes on absorbant paper for further drainage of supernatant. The tracer remaining in each tube was determined with a 1-2 minute counting time on a gamma-counter (Micromedic, Horsham, PA). Standard curves were calculated by fitting a linear regression of the log of the standard with the logit of the % radioactive counts bound to antibody. Concentrations of unknowns were extrapolated from standard curves. Assessment 2f Validation Criteria - Specificity data was provided by the manufacturer, expressed as percent relative crossreactivity at 50% supression of total binding. Assay sensitivity was the estimated hor- mone concentration corresponding to the binding percentages at the 95 and 99% confidence intervals (CI) below the mean specific binding at the 0 standard. Accuracy was estimated by measurement of known amounts of hormone added to serum samples prior to other extraction. Parallelism was assessed by subjective comparison of standard curves with binding inhibition curves obtained with extraction of different volumes of the same serum sample. Precision was estimated by deter- mining intra- and interassay coefficients of variation for each species. 91 RESULTS Extraction efficiency results for estradiol and testosterone are shown on Tables 4 and 5, respectively. The species from which serum was obtained did not affect recovery of either 3H-estradiol-17a or 3H- testosterone. Less 3H-estradiol-17a was recovered after extraction of 3 ml serum (86.4%) than from 1 ml serum (101.1%) with 10 ml ether (p < 0.05). There was no difference in 3H-testosterone recovery after extraction of 0.5 or 1.5 ml serum with 5 ml ether. Table 4.--Effects of species and serum volume on recovery of 3H-estradiol-17a after extraction with 10 ml freshly—opened diethyl ether. Values are expressed as % of 2334 cpm. Overall species Species of Origin 1.0 ml 2.0 ml 3.0 ml mean Canine 109 88 81 92.7 Equine 102 93 92 95.7 Bovine 92 93 86 90.3 Overall volume mean 101.08 91.3 86.3a 92.9 avalues in a row differ at p < 0.05 92 Table 5.--Effects of species and serum volume on recovery 3H- testosterone after extraction with 5 ml of diethyl ether. Values are expressed as % of 8275 cpm. Volume of serum extracted (r - 3) Overall species Species of Origin 0.5 ml 1.5 ml mean Canine 84 80 82.0 Equine 86 83 84.5 Bovine 84 82 83.0 Overall volume mean 84.3 81.6 83.0 Specificity data provided by the manufacturer for the relative crossreaction of various steroids is shown on Table 6. For estradiol, the corresponding hormone concentrations at 95% and 99% CI from the mean total specific binding were 1.7 and 2.8 Pg/tube, respec- tively. For the testosterone RIA, 0.02 and 0.04 ng/tube were the respec- tive values at the 95% and 99% CI from the mean total specific binding. These values are estimates of assay sensitivity. Accuracy of measuring different amounts of exogenous estradiol or testosterone added to serum are shown on Tables 7 and 8, respectively. When corrected for extraction efficiency, 98% and 107% of exogenous estradiol and testosterone was measured, respectively. 93 TABLE 6. Percent crossreactivity at 50% of total binding capacity of various steroids with antibody from commercially-available estradiol and testosterone radioimmunoassays.a % RELATIVE CROSSREACTIVITY Testosterone Compound Estradiol antibody antibody Estradiol 17b 100 .058 Estrone 1.3 .013 Testosterone 0.016 100 Cortisol 0.006 .006 Aldosterone <0.0009 NG Androstenedione 0.0001 NG Progesterone 0.0005 .031 17-Hydroxyprogesterone <0.0001 NG Desoxycorticosterone 0.0001 NG Dehydroepiandrosterone (0.0001 NG Pregnenolone 0.0006 NG Dihydrotestosterone NG 22.5 Estriol NC .063 a Values were obtained from kit instructions supplied by Serono Laboratories, Braintree, MA. NG - Not given 94 Table 7.--Accuracy of measuring known amounts of estradiol-17b when added to serum prior to ether extraction. Values are expressed as [amount measured/amount added] x 100% (r - 2). Estradiol added (pg/tube) Species Species Serum 2.5 5.0 10.0 20.0 Avg. Bovine 80 81 87 121 92.3 Canine 74 122 91 110 99.3 Equine 132 107 89 88 104.0 Estradiol Dose Avg. 95.3 103.3 89.0 106.3 98.4 Table 8.--Accuracy of measuring known amounts of testosterone when added to serum prior to other extraction. Values are expressed as [amount measured/amount added] x 100% (r - 2). Testosterone added (pg/tube) Species Species Serum .05 .15 .60 Avg. Bovine 80 107 87 91.3 Canine 140 107 112 119.7 Equine 120 107 103 110.0 Testosterone Dose 113.3 107 100.7 107 Average 95 Table 9.—-Precision of estradiol and testosterone radioimmunoassays for bovine, canine and equine serum. Coefficients of Variation (%) Hormone 1X Conc. interassay intraassay Assayed Species of Sample (no. assays) (r-S) Bovine 9.0 pg/ml 20.4 (14) 9.2 Estradiol Canine 3.1 pg/ml 12.6 (3) 13.4 Equine 3.5 pg/ml 7.4 (3) 17.1 Bovine 8.6 ng/ml 14.4 (16) 3.9 Testosterone Canine 0.56 ng/ml 27.7 (13) 4.5 Equine 0.50 ng/ml 14.1 (4) 11.4 96 Figure 10. Inhibition curves for estradiol standard solutions and samples reconstituted after diethylether extraction of different volumes of bovine, canine and equine serum pools. 97 Ao_oom mo: :59. E 6.86:3...” onTBEmbmw com 00. 0.8 0mm 00. on mm mm. 4 _ _ _ _ A _ _ ¢._muoz_oz_m o_1.=omam 1_<._.O._. o\o N.vuoz_oz_m o_n=omamuzoz o\o oummuz_2\w._.zaoo 43.0% P . _ On QN 0.. n. no Siam 001: owho