PIWITARY AND GGNADAL RESPONSES TO GGNADOTRQPEN RELEASENG HORMONE EN BULLS Dissertation for the Degree of Ph. D. MECHGAN STAYE UNIVERSITY KRIT MONGKONPUNYA 1974 LI B R A R Y Michiga: State «3 Univ ersity W6 “3—1.” This is to certify that the thesis entitled Pituitary and Gonadal Responses to Gonadotropin Releasing Hormone in Bulls presented by Krit Mongkonpunya has been accepted towards fulfillment of the requirements for Ph.D. Physiology 81 Dairy Science degree in Major professor May 17, 197h Date 0-7639 ABSTRACT PITUITARY AND GONADAL RESPONSES TO GONADOTROPIN RELEASING HORMONE IN BULLS By Krit Mongkonpunya Two experiments were designed: 1) to determine the luteinizing hormone (LH) response to gonadotropin releasing hormone (GnRH) in bulls at 2, 4 or 6 months of age and the consequent testosterone response to increased endogenous LH; and 2) to determine blood levels of LH and testosterone and gonadal and epididymal sperm in pubertal bulls during and after chronic GnRH treatment. Nine bulls, three each at 2, 4 and 6 months of age received three doses (200, 400 or 800 pg; intramuscular injection (im) of GnRH, one on each of 3 consecutive days. Jugular blood was sampled at frequent intervals and serum LH, testosterone and androstenedione were determined by radioimmunoassay. Serum LH significantly increased (P<0.01) from 1 ng/ml before GnRH to a peak of about 24 ng/ml at 45 minutes, then it declined gradually to pretreatment concentrations within 6 hours. This pattern of LH response was similar in bulls at 2, 4 or 6 months of age. Peak responses after 200, 400 or 800 ug GnRH were not different. However, the duration of the LH response after 800 pg GnRH was longer than that after 200 ug GnRH. LH responses also differed significantly by days of treatment. Krit Mengkonpunya The time from GnRH injection to LH peak, magnitude of the peak and duration of serum LH response were significantly greater on day 1 than on days 2 or 3. These data indicated that the LH response of prepubertal bulls to 24-hour repeated GnRH administrations decreases. A11 doses of GnRH stimulated serum testosterone but older bulls responded more than younger bulls. In 2-month-old bulls, testosterone averaged 0.4 ng/ml before GnRH, increased slightly to 0.6 ng/ml at 2-3 hours and declined to 0.3 ng/ml at 6 hours after GnRH. In 4-month- old bulls comparable values were 0.6, 1.2 and 0.5 ng/ml. In contrast, testosterone in 6-month-old bulls averaged 1.7 ng/ml before GnRH, increased significantly (P<0.05) within 1 hour (4.6 ng/ml) and peaked (5.3 ng/ml) at about 2 hours after GnRH. Since age was without effect on LH response to GnRH, this increased capability of the testis to release testosterone between 2 and 6 months of age suggested that the testis is more directly related to sexual maturation than the anterior pituitary. Androstenedione concentrations of blood either before GnRH or after GnRH were only about 13% of those of testosterone. However, all doses of GnRH increased androstenedione significantly (P<0.05) at 2, 4 or 6 months of age, and this response was unaffected by age of the bull. On the average androstenedione increased about 3-fold from 120 pg/ml before GnRH to 300 pg/ml at 1-2 hours, and then declined to 60 pg/ml at 8 hours after GnRH. Similar to the LH response, androstenedione responses differed from the first to the third day. On day 1, androstenedione increased gradually until 4 hours after GnRH. On day 2 androstenedione peaked at 1 hour, and on day 3 it peaked at 2 hours. The general patterns of LH and androstenedione responses Krit Mengkonpunya to GnRH on each day were similar although androstenedione lagged behind LH. Since androstenedione significantly (P<0.05) increased after GnRH at all ages and the testosterone response was significant only in 6-month-old bulls, probably indicates that 2- or 4-month-old bulls possess a different mechanism for converting androstenedione to testosterone, or the processes of the conversion may be blocked or insufficiently stimulated in younger bulls. Experiment II was designed as a triple split-plot with nesting, and involved 18 bulls, six each at 2, 4 and 6 months of age. Three bulls at each age either received saline or 40 ug GnRH (im) twice daily fer 28 days. TWO months after the last injection sperm numbers were measured in the epididymis and testis. GnRH injections had no effect on growth rate or testicular dimensions measured throughout the experimental period (3 months). Testicular and body growth were significantly correlated (r = 0.82, P<0.01). Regardless of treatments, the youngest age at which I found sperm was 5 months, whereas appreciable numbers of testicular spermatozoa were found in all bulls at 9 months. The GnRH treatment failed to increase testicular or epididymal sperm counts. Serum LH in saline-treated bulls was essentially constant, averaging 1.3, 1.5 and 1.7 ng/ml for 2, 4 and 6-month-old bulls, respectively; and it did not differ from the average pretreatment concentration in bulls subsequently treated with GnRH. Serum LH significantly increased (P<0.01) 30 minutes after GnRH. The greatest response was after the first. injection (AM) on day 1; the values were 32.7, 27.4 and 14.7 ng/ml fer 2, 4 and 6-month-old bulls, respectively. The comparable data after the second injection (PM) on day l were 23.6, 12.9 and 8.4 ng/ml. Krit MOngkonpunya Although the LH response on the average was greater in younger bulls this was not significant (P>0.05). The average (AM+PM) serum LH at 30 minutes after GnRH on day 1 from all bulls (20.0:4.6 ng/ml) was significantly (P<0.05) higher than those on day 7, 14, 24 and 28 which averaged 5.5:0.8, 6.5:0.9, 5.0:0.9 and 6.210.2 ng/ml respectively. Serum LH at 30 or 60 days after the end of treatment period (28 days) did not differ significantly between saline or GnRH-treated bulls; the values in the control bulls were 1.2, 0.5 and 0.6 ng/ml for 2-, 4- and 6-month-old bulls, respectively. The respective values in GnRH-treated bulls were 2.1, 0.3 and 0.4 ng/ml. In 6—month-old bulls treated with GnRH, average testosterone concentrations (5.8:0.3 ng/ml) were significantly higher (P<0.05) than those at 2 or 4 months of age, which averaged 0.7:0.1 and 2.011 ng/ml respectively. Comparable data for the control group were 3.81.4, 2.4:.2 and 0.5:.1 ng/ml for 6, 4 and 2 months, respectively. Serum testosterone in 6-month-old bulls treated with GnRH was also greater (P<0.05) than that in saline-treated bulls (5.8 vs 3.8 ng/ml), but this difference could not be detected in younger animals. To summarize, the results indicated that LH released in response to GnRH was similar in bulls aged 2 to 6 months, but the testosterone responses increased during this interval. Unlike testosterone, andros- tenedione responses were similar in the bulls at 2, 4 or 6 months of age, but the magnitude of the responses was lower than that of testosterone. LH responses decreased with repeated injections of GnRH. Furthermore, repeated GnRH injections did not hasten puberty (sperm production) in the bull. PITUITARY AND GONADAL RESPONSES TO GONADOTROPIN RELEASING HORMONE IN BULLS BY KRIT MONGKONPUNYA A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Departments of Physiology and Dairy Science 1974 TO MY PARENTS ii BIOGRAPHICAL SKETCH The author was born in the District of Wungparng, Pacharng, Lumphoon, Thailand on August 1, 1938. He received his elementary education in Wungparng public school system and his secondary education in Prince Royal's College, a Presbyterian Missionary School in Chiengmai, Thailand. He entered the Agriculture College at Kasetsart University, Bangkok, Thailand and received a Bachelor degree of science in June 1961. His thesis was entitled "Nutritive value of paragrass". Upon his graduation, he joined the Department of Biology, Kasetsart University as an instructor. In 1965 he was awarded a scholarship from South East Asia Treaty Organization (SEATO) for graduate study at West Pakistan Agriculture University, Lyallpur, West Pakistan. He received his Masters of Science degree in physiology in July 1967. His thesis was entitled "Effects of day light and ambient temperature on reproductive perfor- mance in mice". Again, he rejoined the Department of Biology, Kasetsart University. In September 1971 he was granted a scholarship from the Rockefeller Foundation to continue his graduate study in physiology and dairy science at Michigan State University. He is completing the requirements for the Ph.D. degree in May 1974. iii ACKNOWLEDGMENTS The author thanks Dr. Harold D. Hafs, Professor of Physiology and Dairy Science, for the advice, guidance and assistance he has extended throughout his study program. Appreciation is also expressed to Dr. H. A. Tucker for his advice and help in preparing this manuscript. The author is also grateful for the advice and consent of other guidance committee members, Drs. E. P. Reineke, J. L. Gill and J. E. Wilson. The author also thanks Dr. R. Neitzel for statistical advice and computer programming. For their assistance in the laboratory and in collecting blood samples the author thanks his colleagues Dr. R. A. Bourne, Dr. L. A. Edgerton, Dr. J. A. Koprowski, Dr. Y. C. Lin, 0. W. Smith, T. M. Louis, D. T. Vines, J. N. Stellflug, R. J. Kittok, T. W. Beck, B. E. Seguin, P. A. Noden and V. G. Smith. A debt of gratitude is extended to Drs. E. M. Convey, W. D. Oxender, J. H. Britt and P. K. Kindel for their interest and advice during this study. The financial support rendered by the Kasetsart University and The Rockefeller Foundation for a scholarship is appreciated. Finally, the author thanks his wife, Aporn, whose encouragement and sacrifices proved invaluable. iv TABLE OF CONTENTS DEDICATION . BIOGRAPHICAL SKETCH ACKNOWLEDGMENTS TABLE OF CONTENTS LIST OF TABLES . LIST OF FIGURES LIST OF APPENDICES . INTRODUCTION . REVIEW OF LITERATURE . 1. 2. 3. The Hormones of the Hypothalamus Effect of GnRH on Gonadotropin Release Luteinizing Hormone Releasing Factor (LRF) in Cattle Gonadal Steroid Response to Increased LH after GnRH Administration Functional Relationship among Hypothalamus, Pituitary and Gonads . Reproductive and Endocrine Changes in Fetal Through Pubertal Stage . Current Hypothesis Concerning Mechanisms of Initiation of Puberty . MATERIALS AND METHODS l. Radioimmunoassay of Testosterone a. Testosterone Extraction b. Radioimmunoassay . c. Assay Validation . Page ii iii iv vii ix 12 16 20 20 2O 21 21 2. Radioimmunoassay of Androstenedione . a. Androstenedione Extraction . b. Radioimmunoassay . c. Assay Validation . 3. Radioimmunoassay of Luteinizing Hormone (LH) a. Radioiodination b. Radioimmunoassay . 4. Blood Sample Collection and Intravenous Administration . 5. Designs and Procedures of Experiments . a. Experiment I . b. Experiment 11 RESULTS AND DISCUSSION 1. Experiment I a. LH Response . b. Testosterone Response c. Androstenedione Response . 2. Experiment II . a. Body weight . b. Testicular Dimension . c. Testicular and Extratesticular Sperm Counts d. Serum LH . . e. Serum Testosterone . GENERAL DISCUSSION . BIBLIOGRAPHY . APPENDICES . vi Page 25 25 25 26 26 27 28 3O 33 33 35 36 36 45 SO 58 58 58 6O 62 75 87 91 102 Table 10. 11. 12. 13. 14. 15. LIST OF TABLES Comparison of two methods for assay of testosterone in bull sera Comparison of RIA values with and without steroid purification on LH-20 Sephadex Schedule for GnRH treatments to bulls in Experiment 1 Serum LH after injection (im) of GnRH into bulls at 2, 4 or 6 months of age Serum LH after injection (im) of 200, 400 or 800 pg GnRH into bulls . . . . . . . . . . . . . Summary of ANOVA for LH data after GnRH Serum LH after injection (im) of GnRH into bulls on three consecutive days . Serum testosterone after injection (im) of GnRH into bulls at 2, 4 or 6 months of age . Summary of ANOVA for testosterone data after GnRH Serum androstenedione after injection (im) of GnRH into bulls at 2, 4 or 6 months of age . Serum androstenedione after injection (im) of GnRH into bulls on three consecutive days Summary of ANOVA for androstenedione data after GnRH Average testicular dimensions of bulls , Gonadal size and sperm, and epididymidal sperm at 60 days after 28 days of GnRH treatment . Serum LH response to saline or GnRH (40 pg, im) injected twice daily for 28 days in bulls at 2, 4 or 6 months of age . vii Page 23 24 34 37 38 39 44 46 49 S3 S4 S6 S9 61 63 Table 16. 17. 18. 19. 20. Summary of ANOVA for LH data during 2X40 pg GnRH daily for 28 days . . . Serum LH and testosterone in control bulls treated with saline twice daily for 28 days . Serum LH and testosterone in bulls treated with GnRH twice daily for 28 days Serum testosterone response to saline or GnRH (40 pg, im) injected twice daily for 28 days in bulls at 2, 4 or 6 months of age . . . . . . . . . . . Summary of ANOVA for testosterone data during 2X40 pg GnRH daily for 28 days viii Page 65 68 71 77 79 LIST OF FIGURES Figure Page 1. Gonadotropin releasing hormone (GnRH) and thyrotropin releasing hormone (TRH). Amino acid sequences . . S 2. Dose response curve for NIH-LH-B8 standard . . . . . . 32 3. Serum LH after injection (im) of GnRH into bulls on three consecutive days . . . . . . . . . . . . . 43 4. Serum testosterone after injection (im) of GnRH into bulls at 2, 4 or 6 months of age . . . . . . . 48 5. Serum androstenedione after injection (im) of GnRH into bulls on three consecutive days . . . . . . . 51 6. Serum LH at 0, 0.5 or 3 hours after injection (im) of GnRH into bulls on days 1, 7, 14, 24 or 28 . . . 70 7. Serum LH at 0.5 hour after injection (im) of GnRH into bulls in the morning (AM) or evening (PM) . . 74 8. Serum testosterone at 0, 0.5 or 3 hours after injection (im) of saline or GnRH into bulls at 2, 4 or 6 months of age . . . . . . . . . . . . 83 9. Serum testosterone on day 1 (AM or PM) at O, 0.5 or 3 hours after injections (im) of saline or GnRH into bulls at 6 months of age . . . . . . . . 8S ix LIST OF APPENDICES Appendix Page I. Composition of reagents used in radioimmunoassay . 103 II. Preparation of liquid scintillation fluids . 106 III. Composition of a reagent used in determination of 107 sperm number INTRODUCTION The practical goals of research in physiology of reproduction in bulls are increased fertility and sperm production to cope with the increasing demand for semen from proven sires. Low libido, low sperm production and delayed puberty are problems which delay genetic progress in the artificial insemination industry. Development of the reproductive system is under the influence of the hypothalamo-hypophyseal system. General relationships between the pituitary, hypothalamus and testis have been established in the adult. However, precisely when and how the hypothalamo—hypophyseal system participates in testicular maturation during puberty is unknown. Following the determination of the structure of hypothalamic luteinizing hormone (LH) releasing hormone (Matsuo gt_§13, 1971), evidence indicated that this decapeptide caused release of follicle stimulating hormone (FSH) as well as LH in many species, including cattle. This decapeptide has been called LH RH or LH-RH/FSH-RH, but because of its ability to release both LH and FSH the term gonado- tr0pin releasing hormone (GnRH) will be used in this thesis. The purpose of this thesis was to study the effects of adminis- tration of GnRH on pituitary function and secondarily on gonadal function. Two experiments were designed to: 1) determine the LH and androgen responses to GnRH in bulls at 2, 4 or 6 months of age; and 2) determine blood levels of LH and testosterone, and gonadal and epididymal sperm in pubertal bulls during and after chronic GnRH treatment . REVIEW OF LITERATURE 1. The Hormones of the Hypothalamus In 1947, it was postulated that the anterior pituitary gland, which controls the peripheral endocrine glands, is regulated by factors or hormones originating in the hypothalamus (Green and Harris, 1947). Evidence accumulated that at least nine hormones from the hypothalamus control the biosynthesis and release of pituitary hormones. Among these hormones, two of them (thyrotropin releasing hormone (TRH) and gonado- tropin releasing hormone (GnRH), Figure l) have been isolated and synthesized and are available for investigation. The record of investigational development concerning the chemistry, physiological function and search for their chemical structure has been reviewed (McCann and Porter, 1969; Meites, 1970; Martini and Ganong, 1971; Schally gt_§1,, 1972; Guillemin and Burgus, 1972; Bogdanove, 1972; Schally 33.213’ 1973; and Convey, 1973). Following determination of the structure and synthesis of GnRH by Matsuo 23.21: (1971), it was demonstrated that GnRH preparations with the structure shown in Figure l have the same activity as the native substance purified from the hypothalamus. The view that one hypothalamic hormone (GnRH) could be responsible for stimulating the release of both FSH and LH is supported by considerable physiological and biochemical evidence. For example this decapeptide caused release of FSH and LH in rats (Schally et_§l:, 1971c), sheep (Reeves gt 31., 1972), humans (Schally gt_al,, 1971b; Rebar §t_§13, 1973) and cattle Figure 1.--Gonadotropin releasing hormone (GnRH) and thyrotropin releasing hormone (TRH). Amino acid sequences. Gn-RH M.W. ' 118! 10 Gly TRH 2 2 "UH c c _ _ o 2 2 H H g H N oflc — 2 H H H N OIC z _ H H _ in 2 CH O—C\ Pro-(Amide) His (Pyro) Glu (Zolman and Convey, 1973; Kaltenbach gt_al,, 1973). Small doses of GnRH stimulated the release of FSH and LH in rat pituitaries cultured in_vit£g_(Schallyy§£_§1:, 1972). Furthermore, chemical and enzymatic inactivation of GnRH (Schally 23.31:: 1971a) was accompanied by loss of both FSH and LH releasing activities, and chromatographic and electrophoretic separational techniques failed to separate the two activities (Schally gt_al:, 1971b). Evidence against the concept of one hypothalamic hormone controlling the discharge of both LH and FSH was demonstrated by Johansson gt_§13 (1972) and Currie §t_§13 (1971). They reported that GnRH caused release only of LH. Moreover, another hypothalamic preparation essentially free of the GnRH moiety caused release of LH, and even greater amounts of FSH (Bowers, 1973). At low doses,this preparation stimulated only FSH release. Thus, Bowers (1973) proposed that the decapeptide (Figure l) is LH-RH; and that FSH-RH is a separate hypothalamic releasing hormone of unknown structure. In contrast to the effects of GnRH on LH and FSH, 25 pg of GnRH given intravenously to men had no significant effect on thyroid stimulating hormone (TSH), growth hormone (GH), adrenocorticotropin hormone (ACTH), insulin or blood glucose (Wagner et_a13, 1972). Recently, using a radioimmunoassay (RIA) Nett et 31. (1973) determined GnRH in serum of normal rams (7111.4 Pg/ml) and in serum of castrate ewes (12816.1 pg/ml). The time required for one-half of GnRH activity to disappear from blood was about 6 minutes in sheep (Nett et_al,, 1973), and rats (Redding and Schally, 1973). Thus, it appears that GnRH normally reaches the peripheral circulation, but it is rapidly inactivated. 2. Effect of GnRH on Gonadotropin Release The magnitude and time course of increased blood LH after GnRH administration vary among species, doses, routes of administration and physiological states of the experimental animals. In cattle, the LH response is proportional to the dose of GnRH (Zolman gt 31,, 1973), especially when given on the basis of body weight (Golter §t_§13, 1973). There is also interaction between doses and routes of administration. Peck e£_a13 (1973) found that intravenous and intrauterine administrations were equally effective; while Tarnavsky and Reeves (1973) found that intrapulmonary administration (intratracheal) was less effective than intramuscular. Golter e£_§13 (1973) found that 1.3 pg GnRH given in acidified saline or in % carboxymethyl cellulose per kilogram of body weight caused similar LH responses, but a subcutaneous silastic implant of GnRH was less effective. They concluded that the rate and duration of release of GnRH are more important than the dose of GnRH in regulating LH release. Convey (1973) concluded that "GnRH causes release of LH in_viyg_and ifl.!i££9 and pituitary responsiveness to exogenous GnRH varies with reproductive state probably as a result of variation in the steroid environment". Chakraborty §t_al, (1973a) reported that the serum LH response in prepubertal female pigs subjected to repetitive treatment with GnRH (16 injections of 25 pg over 96 hours) was gradually decreased between the first and third injections. Thereafter, the LH response was consistent but at a lower level. Chakraborty et_al. (1973b) also reported that infusion of GnRH at a rate of 2.3 pg per hour into anestrous ewes failed to maintain elevated serum LH concentrations after 3 hours of infusion. \ 3. Luteinizing Hormone Releasing Factor (LRF) in Cattle To my knowledge, there are only two reports of hypothalamic content of LRF activity in cattle. Both used a partial purified extract of the hypothalamus in a rather insensitive and indirect bioassay. The LRF potency in the hypothalami of Holstein bulls from birth to 12 months of age was determined by the ovarian ascorbic acid depletion assay in rats (Macmillan and Hafs, 1968b). LRF activity was not detectable in the hypothalami from bulls less than 5 months of age; it increased between 6 and 10 months and the increase was associated with increased LH in the plasma. In heifers during the estrous cycle Hackett and Hafs (1969), using the same assay technique found that hypothalamic LRF was increased from day 20 to day 7 and decreased on days 11 and 18. The elevated LRF was associated with release of LH, ovulation, and luteal growth; the reduced LRF occurred during the period of maximal progestin secretion. 4. Gonadal Steroid Response to Increased LH After GnRH Administration McCormack and Spies (1972) administered an extract from ovine stalk-median eminence into mature rhesus males and measured serum LH and testosterone. Testosterone began to rise later than LH, and increased 2- to lZ-fold above baseline within 15 to 90 minutes after the beginning of treatment. In women experiencing abnormally infrequent or absence of menses, Newton and Collins (1972) found that GnRH increased plasma LH, FSH and estradiol 17-8. Kastin e£_al. (1972) found that men and prepubertal boys showed significant increases in estradiol after injection of GnRH. Similarly, blood LH and testosterone increased in response to GnRH in children at puberty (Roth 33 31:, 1972 and 1973). 5. Functional Relationship Among Hypothalamus, Pituitary and Gonads The integration of physiological functions of the hypothalamus, pituitary and gonads are not yet known with certainty. Keeping in mind that there might be species differences, few studies have been done on farm animals. Most of the evidence given in this section is from laboratory animals. The general relationship between the hypothalamus and pituitary- gonadal axis has been actively investigated. Results may be summarized as fellows. As described in the previous section, a hypothalamic hormone (GnRH) regulates the synthesis and release of pituitary gonadotropins (LH and FSH). Gonadal functions are stimulated due to the increase of gonadotropin concentrations in the blood. The positive effects of gonadotropins on the male reproductive tract are mostly a consequence of androgens produced by the testes (Donovan, 1970). Sampling every hour Katongole et_§1, (1971) and Smith §t_al, (1973) demonstrated episodic increases in serumllL hiindividual bulls, about four LH peaks occurred during every 24-hour period. Most of the LH spikes were accompanied or followed within 1 hour by increases in serum testosterone. Gonadotropins can also act directly on the seminiferous tubules, since LH administration to hypophysectomized lambs caused proliferation of sertoli cells in the seminiferous tubules, while large doses of testosterone failed to show this effect (Courot, 1970). However, FSH is believed to stimulate the germinal epithelium of the testes, while the principal action of LH is to activate the Leydig cells for the production of androgens (Li, 1972). Possibly the actions of these two hormones are synergistic (Courot, 10 1970), and altering the FSH-LH ratio also may be involved in their physiological function (Charro et_§1,, 1973). At the molecular leve1,gonadotropin stimulation of androgen secretion evidently involves activation of Leydig cell adenyl cyclase activity (Dufau et_al,, 1973). However, there is evidence that the initial stimulation of steroidogenesis by gonadotropins may operate by pathways other than that directly involving adenyl cyclase and cyclic AMP formation (Catt and Dufau, 1973) because low gonadotropin levels (0.1 ng/ml), which stimulate testosterone release in isolated rat testis, failed to enhance the formation of cyclic AMP. Gonadal steroids influence the release and synthesis of hormones from the hypothalamus and pituitary. Administration of estrogen or testosterone decreased serum LH concentration in rats (Bogdanove, 1963; Jackson, 1972; Ramirez and McCann, 1963; and Smith and Davidson, 1968), rams (Pelletier, 1970) and cattle (Odell and Kiddy, 1969). This is the so-called negative feedback control of steroids on gonadotropin secretion. The anatomical site of action of the negative feedback control is the anterior hypothalamus and pituitary. Thus, after injection of radioactive sex steroids into rats (Touhimaa and Niemi, 1972; Stern and Eisenfeld, 1971; Sar and Stumpf, 1973a,b), there was more radioactivity in the anterior hypothalamus and anterior pituitary than other parts of the brain. Sar and Stumpf (1973b) found that about 10-15% of cells in the anterior pituitary of rats retained radioactivity in their nuclei within 1 hour after injection of 1,2-3H testosterone, while no accumulation of radioactivity was found in the other lobes of the pituitary. The cells retaining radioactivity were identified as gonadotrophs. The major component of the ll radioactivity retained in the pituitary after the injection of 3H- testosterone was identified as dihydrotestosterone (Stern and Eisenfeld, 1971). Karsch §t_§13 (1973) concluded that the negative feedback regulation of LH release is a finely tuned system with a set—point which appears to change as a function of time after castration in monkeys. Further- more, progesterone can synergize with estradiol in suppressing LH release. Possibly, LH secretion is reduced by the combined inhibitory effect of progesterone and estradiol since castrated rats were less sensitive to progesterone than intact rats (Ramirez and McCann, 1963). Increased blood LH after castration differs with age and sex (Yamatomo at 31., 1970). For example, in adult male rats serum LH increased sharply by 8 hours after castration and reached a plateau by 18 hours. In adult females the increment in LH per unit time was less than in males, finally reached concentrations comparable to those of the males 10 days after castration. Similar patterns of the increment in serum LH were also observed in immature males and females, but the concentra- tions were less than in the adults. Though it has been demonstrated that the negative feedback of steroids can have effects at the pituitary or hypothalamus, there is no evidence that similar direct effects occur normally. Besides, we do not know the mechanism of action of the steroid feedback. It has been reported that aromatization of the steroids in the hypothalamus may be one mechanism essential for the negative feedback (Beyer gt al., 1970; Reddy 35 31., 1973). Reddy gt_al, (1973) observed that the intact male hypothalamus was more active than that of intact female in the aromatization of androstenedione to estrone; castration increased this activity in both sexes. However, in rats, Swerdloff $3.31: (1972) 12 administered testosterone and dihydrotestosterone (an A-ring reduced steroid which is not convertible to estrogen) over a broad dose range and found that dihydrotestosterone was more potent than testosterone in suppressing serum LH, but they were equally potent in suppressing FSH. They concluded that conversion of androgens to estrogens was not required for inhibition of the hypothalamo-pituitary axis. Consequently, the mechanism of the negative feedback control system is still open. 6. Reproductive and Endocrine Changes in Fetal Through Pubertal Stage A functional relationship may exist between the pituitary and gonads in the bovine fetus as early as 90 days of gestation (Oxender gt_§l,, 1972a; Kim 33.32:, 1972; Mongkonpunya §t_§l:, 1974). In male fetuses, serum LH and testosterone were highest at 90 days of gestation and decreased as fetal age increased (Mongkonpunya 33 31" 1974). Nakai gt_al, (1972) demonstrated that castration of male fetal rats resulted in an increased percentage of basophilic cells in the pituitary, suggesting early existence of a feedback mechanism between the testes and the hypothalamic-pituitary axis. The gonads of the fetal horse at about 4 months were composed mainly of interstitial cells (Raeside 33 31., 1973). During fetal development Reyes et_§l, (1973) found the highest testosterone concentration in human testes at 8-12 weeks, but negligible quantities in ovaries or adrenals. They concluded that there was high androgen production in the testis at the time of male genital differentiation. In bovine neonates during the first 7 days of life, the concentra- tion of serum LH was relatively lower than in fetuses regardless of stage of gestation or sex (Oxender et_al., 1972b). However, 13 testosterone secretion in bovine neonates has not been studied. In humans at birth, Forest gt 31. (1973) demonstrated that testosterone in umbilical cord blood was significantly higher in males than in females. In male infants, testosterone concentration increased from birth to-2-3 months, then gradually declined until 7 months when its concentrations stabilized until puberty. Reproductive and endocrine development in Holstein and Hereford bulls from birth through 1 year of age were investigated in our laboratory (Macmillan and Hafs, 1968a,b; Macmillan and Hafs, 1969; Purchas, Macmillan and Hafs, 1970; Rawlings, Hafs and Swanson, 1972; Swanson gt 31., 1971a). The results from these studies can be summarized as follows: a. Sperm Number.--Testicular sperm were found in some bulls at 5 to 7 months and in all bulls at 8 months. Average testicular sperm concentration increased about 14-f01d between 7 to 12 months, and sperm concentration at 12 months was similar to that for mature bulls. In contrast, the sperm content of caudae epididymides at 1 year is only about 20% of that in mature bulls. It was concluded that Holstein bulls attain their mature rate of spermatogenesis by 1 year of age. The subsequent increases in sperm production with advancing age were primarily due to increases in testicular weight. b. Pituitary and hypothalamic endocrine changes.--Pituitary weight increased linearly from birth to 1 year of age. While the concentration of pituitary LH was highest at 1 month, the total gland content was greatest at 6 months of age. Estimated total serum LH did not signifi- cantly change from birth to 2 months, increased to 4 months and increased again between 6 and 10 months of age; the two increases 14 in LH were associated with increases in hypothalamic GnRH. In contrast to these results, Odell and Kiddy (1969) and Odell gt a1. (1970) reported that there were no consistent changes in blood LH concentration during sexual maturation in heifers or bulls. Concentration of pituitary FSH was greatest at 2 months, but pituitary content was highest at 5 and 6 months and both declined after 6 months of age. The data suggested that puberty in bulls commences at 2 months and is qualitatively completed by 10 months of age. The greatest changes occur between 6 and 10 months. c. Development of reproductive tract.--Thirteen criteria related to androgen secretion were studied. The results indicated that the period of accelerated reproductive development associated with puberty was terminated by 9 to 10 months of age. It was also revealed that pubertal development may be biphasic; one phase occurred between 2 to 4 months and the other from 6 to 9 months of age - the periods of rapid increases in plasma LH. The second period of development was typified by appearance of more obvious external manifestations of puberty and by rapid onset of spermatogenesis. d. Testicular and blood plasma androgens.--Using a gas-liquid chromatographic assay, Rawlings et_al, (1972) found that testicular concentration of testosterone increased from birth to 5 months, decreased at 6 months and then generally increased to 11 months. Testicular androstenedione fell from 4 months to 7 months and remained stable at a low level thereafter. After 4 months of age, testosterone predominated. Plasma testosterone generally increased from undetectable concentrations at birth to about 4 ng/ml at 11 months of age. After 15 8 months of age, testicular and serum testosterone increased with the rising titers of plasma LH. Mann gt_§1, (1949 and 1960) reported a high correlation between androgen and secretory function of accessory organs which could be observed long before any sign of active spermatogenesis in bulls. Lindner (1959) and Lindner and Mann (1960) chemically identified endocrine constituents of bull testes from 28 days to 17 1/2 years of age and found testosterone and androstenedione could be detected as early as 39 days. The combined content of the two steroids per unit weight of tissue was not very different in the testes of very young calves or of mature bulls. However, androstenedione predominated in testes of young bulls whereas in mature bulls testosterone was predominant. Between the 2nd and 6th month of life the rate of development of seminal vesicles increased at an exponential rate and this was closely associated with testosterone content of the testes. e. Histological development of bovine testes.-—Hooker (1944 and 1970) reported that Leydig cells differentiated from interstitial mesenchymal cells in bulls by 4 months of age, increased in number until 2 years of age. Thereafter there was a gradual increase in size of the Leydig cells mainly due to vacuolation. More recently Galina (1972) studied micromorphological changes in the immature equine, bovine, porcine and ovine testes from birth through 365 days of age. It was found that: 1) In immature testes, the seminiferous tubules were mostly irregular with two types of intratubular cells: supporting cells and gonocytes. The interstitial cell population showed important differences among these four species. 16 The interstitium of horse and pig neonates were populated by cells with large round nuclei, similar to mature Leydig cells, whereas in the immature bull and ram testes, only mesenchymal cells and fibroblasts were observed. 2) In the prepuberal phase, tubules with lumina were present with primary spermatocytes in their four meiotic stages. The number of primary spermatocytes increased throughout this stage, indicating a constant increase in the activity of the germ cells. However, the interstitium during this prepubertal phase did not show any noticeable difference from the early neonatal stage. 3) Characteristic of the pubertal phase was the presence of sperm in the seminiferous tubules, with enormous variation among individuals. The interstitium, which were differentiated from the mesenchymal cells, contained a juvenile form of Leydig cell. Collectively, this evidence strongly indicates that steroidogenic activity in bovine testes is high in the first trimester of fetal life and gradually declines as gestation advances. Thereafter, synthesis is considerably lower until onset of puberty, at approximately 5-6 months of age. These changes are closely associated with secretion of LH and FSH from the anterior pituitary. 7. Current Hypotheses Concerning Mechanisms of Initiation of Puberty "Puberty is that stage in development during which maturation of the sexual apparatus occurs and reproductive capability is attained" (Root, 1973). Puberty is a gradual sequential phenomenon influenced by many factors such as genetics, nutrition, environment, and hormonal conditions (Cole and Cupps, 1969). Studies of the testicular development in cattle have been limited. Based mostly on work in laboratory animals, three hypotheses of sexual maturation have been proposed. 17 These three theories, described in greater detail below, are open to question, and none has been proven. Most likely, sexual maturation is multifaceted, involving several mechanisms which are turned on sequentially. a. A decrease in sensitivity of the hypothalamus and pituitary to sex steroid feedback (Smith and Davidson, 1968). The hypothalamic- pituitary system is more sensitive to the inhibitory effects of gonadal steroids in immature than in mature animals (Byrnes and Myer, 1951; Ramirez and McCann, 1963, 1965). Thus, several authors have suggested that the onset of puberty, at least in rats, might be associated with a decrease in the sensitivity of the higher nervous centers to sex steroid feedback resulting in increased gonadotropin release which precipitates puberty. Moreover, the gonadotropin response of the pituitary to the stimulation of hypothalamic hormone also changes during the course of sexual maturation. Spona and Luger (1973) demonstrated that the male rat pituitary ig_yitrg_released greater amounts of LH in response to GnRH during prepubertal ages (10-40 days) than after puberty. In contrast Roth gt a}, (1972 and 1973) found that pituitaries of boys at prepubertal ages released amounts of LH which were significantly less than those at puberty. Recently August gt a}, (1972) demonstrated that early enlargement of human testes was associated with gradual increases in blood LH, testosterone and FSH, suggesting an upward shift in the threshold of the feedback mechanism (i.e., decreased sensitivity of the pituitary to testosterone feedback). Touhimaa and Niemi (1973) 3 injected labelled H-testosterone or 3H-estradiol into fetal, neonatal or adult rats and found that the retention of radioactivity in the hypothalamus was higher in the neonate than in the adult. This 18 suggesting that a more avid steroid receptor mechanism existed in the neonate, and may contribute to the increased inhibitory effects of steroids on gonadotropin release before puberty. As previously mentioned castration stimulates LH release. If steroids lose their ability to suppress gonadotropin release as animals age, one might expect castration to increase LH more at an early age. However, Crim and Geshwind (1972) found that the LH response after castration of rams was equivalent at 30 or 150 days of age. Furthermore, Foster §t_§13 (1972) reported that pituitary LH did not differ in intact and castrated male or female lambs within 14 days after castration. b. FSH induction of sensitivity to LH in Leydig cells. Based upon results from a series of experiments, Odell §t_al, (1973) postulated that a change in sensitivity of the higher nervous centers to inhibition by gonadal steroids could not be the sole explanation for the process of sexual maturation. For example, serum LH in male rats between birth and sexual maturation increased only slightly while FSH increased markedly between 12 and 21 days of age, then declined as sexual maturation proceeded. Moreover, hypophysectomized sexually immature males showed no increase in testicular weight after treatment with LH for 5 days, but testicular weight increased after FSH. Treatment of immature hypophysectomized rats with FSH for varying periods followed by FSH or FSH plus LH produced a significant increase in testicular weight while LH alone produced no effect. However, FSH plus LH produced greater effect than FSH alone and the difference became progressively greater with the time of FSH pretreatment. Thus, they postulated that FSH induces sensitivity of the testes to LH in male rats, and that sexual maturation may occur without alteration of steroid feedback sensitivity. 19 c. A decrease of Sa-reduction of androgens in Leydig cells. Folman et_al, (1972) demonstrated that testes of immature rats formed larger quantities of dihydrotestosterone (35%) and other Sa-reduced metabolites of testosterone than the testes of mature rats (2-3%). It has been shown (Swerdloff et 31., 1972) that dihydrotestosterone was more potent than testosterone in suppressing gonadotropin release., Besides, Baulieu gt a}: (1968) found that dihydrotestosterone was more effective in stimulation of cell division in prostate gland epithelium while testosterone was more effective in stimulation of prostatic secretions. Thus, possibly the mechanism of sexual maturation is associated with metabolic changes in the testes itself. In other words, during sexual maturation there is diminishing Sa-reduction or a decrease in the amount of the more effective inhibitors from the testes. Testicular enzyme systems besides hydroxysteroid dehydrogenase complex and Sa-reductase, may be involved in sexual maturation. Monn 33 El. (1972) reported that two separate isoenzymes of cyclic nucleotide phosphodiesterase named "c" and "f" were found in adult rat testes while only "c" component was present in 20—day old rats. The increase in phosphodiesterase activity was paralleled by an identical increase in testicular adenyl cyclase activity between 20 and 50 days of age. Macindoe and Turkington (1973) reported, in rats, a close association between sexual maturation and ornithine decarboxylase and S-adenoxyl-L-methonine decarboxylase. In my judgement, all of the factors and mechanisms mentioned above probably are involved one way or another in sexual maturation. MATERIALS AND METHODS 1. Radioimmunoassay of Testosterone Radioimmunoassay (RIA) of testosterone recently has been developed and validated by Smith (1972) in our laboratory. The procedures were described in detail elsewhere (Smith and Hafs, 1973). As compared with gas-liquid chromatography (GLC) or competitive protein binding assays, RIA has relatively higher sensitivity; testosterone can be measured in as little as 0.1 m1 of bull's blood serum. However, the RIA procedure described by Smith was still laborious; only about 40 unknowns could be completed daily. A more efficient procedure was desirable because of the large numbers of samples in this research. Thus, the following procedure for RIA of testosterone was developed whereby more than 100 unknowns could be assayed within 1 day. a. Testosterone Extraction.--Aliquants (100 p1) of each unknown were placed in three 15X85 mm disposable culture tubes and about 3000 dpm 3 H,l-Z-testosterone (New England Nuclear, Boston, Mass.) was added to one of the triplicate tubes to estimate procedural losses. Each tube was vortexed with 2 ml nanograde benzenezhexane (1:2) for 30 sec, then stored at -20°C for at least 1 hour to freeze the aqueous phase. The organic solvent from the tube with 3H- testosterone was decanted into a scintillation vial, and the solvent in each of the other two extraction tubes was decanted into two 10x75 mm disposable culture tubes for RIA. 20 21 b. Radioimmunoassay (RIA).--The organic solvent was evaporated and 200 p1 antibody1 (diluted 1:3000 in 0.1% Knox gelatin in 0.1 M phosphate-buffered saline (PBS), pH 7.1; Appendix I.B.4) was added. Two sets of standard tubes containing 0, 20, 50, 100, 200, 500, 1000, 1500 and 2000 pg testosterone were included in each assay and treated similarly to the unknowns. After addition of antibody, each tube was vortexed 10 sec and incubated 30 min at room temperature. Then about 30,000 dpm 3H-testosterone (3H-1,2,6,7-testosterone; New England Nuclear; 91 c/mM diluted in 200 pl 0.1% Knox gelatin PBS, pH 7.1) was added to each tube, and the tubes were vortexed 5 sec and incubated at 4° for 12-24 hours. To separate free from antibody-bound testos- terone, 1.0 m1 of dextran-coated charcoal (0.25 g carbon decolorizing neutral Norit and 0.025 g Dextran 150 in 100 ml distilled water) was added at 4°, and each tube was vortexed for 5 sec, equilibrated at 40 for 10 min and centrifuged (2500 g) for 10 min at 5°. Antibody— bound 3H-testosterone in 0.5 ml of the supernatant fluid was measured in a liquid scintillation spectrometer. For comparison among assays, standard sera with high and low testosterone, and extracts from blank extraction tubes were assayed with each set of unknown serum samples. c. Assay Validation.--Similar to progesterone RIA (Louis 35 31,, 1973), preliminary studies revealed extraction efficiency of 3H-l,2- testosterone was relatively constant (8310.2%, n=240). The largest source of variance was among assays; thus the mass of testosterone determined in each set of unknowns was corrected for the average loss 1Dr. G. D. Niswender furnished rabbit antiserum to testosterone-3- oxime-bovine serum albumin; antiserum code number 667. 22 of tracer determined within that set. The sensitivity of this assay was less than 20 pg testosterone, an amount which displaced about 8% (P<0.01) of the 3H-testosterone bound to antibody. I determined testosterone in aliquants of sera from 28 bulls by the modified assay and by the RIA described by Smith and Hafs (1973); results (Table 1) from the two methods were highly correlated (r=0.81) with the new method resulting in slightly (=12%) higher testosterone values than the old method (P<0.05). The within-sample coefficient of variation for the 28 testosterone determinations by my method was 10%, less than the 31% for the other assay. The specificity of the antibody was described by Ismail, Niswender and Midgley (1972) and by Smith and Hafs (1973). As another means of validating the testosterone RIA's, I purified eight extracts of unknown fetal serum samples and one steer serum sample on LH-20 Sephadex columns with two solvent systems. Firstly, the benzenezhexane extracts of steroids were eluted with chloroform: ethanol (98:2) which brought testosterone and androstenedione off the column together between 15 and 22.5 ml of effluent. Secondly, the elution from the first system was evaporated, dissolved in choloroform: heptanezethanol (200:200:l, saturated with water), layered on an LH-20 Column and again eluted with chloroform:heptane:ethanol, to separate androstenedione from testosterone. By the second solvent system, androstenedione eluted between 20-25 ml and testosterone between 40-45 ml of effluent. The tubes containing the highest radioactivity representing androstenedione and testosterone were quantified by the present RIA to compare (Table 2) with the results of RIA directly applied to benzenezhexane extracts of sera without chromatography. 23 TABLE l.--Comparison of two methods for assay of testosterone in bull sera. Old methoda Determination New method Determination 1 Number of sample (ng/ml) 7276844202942866120005646741 6423243751232160433312330341 064.2618452126442002153640744 6423352431437.141523312236151 1633725047859892637218835672 6334343451533152634312245232 0512254852409762552170424877 5334443442433152634313344231 12345678901234567890 11111111112 1 2 23456 22222 7 2 8 2 3.301.32 27 3.261.27 77:. 22 3. 3.661. Avg 1 SE aSmith and Hafs (1973). 23 TABLE l.--Comparison of two methods for assay of testosterone in bull sera. Old methoda Determination New method Determination 1 Number of sample (Hg/ml) 7276844202942866120005646741 6423243751232160433312330341 064261845212644.2002153640744 6423352431432141523312236151 1.6337250470059892637218835672 6334343451533152634312245232 O512254852409762552170424877 5334443442433152634313344231 1234567890 1 3.30:.32 3.261.27 22 3.771.27 3.66:. Avg 1 SE 3Smith and Hafs (1973). 24 TABLE 2.--Comparison of RIA values with and without steroid purification on LH-20 Sephadex. Fetal Testosterone Androstenedione serum Solvent Sephadex Solvent Sephadex sample extract LH-20 extract LH-20 (ng/ml) 1 0.20 0.50 0.21 0.14 2 0.14 0.34 0.03 0.00 3 4.11 2.82 0.12 0.03 4 3.92 2.33 0.13 0.03 S 0.45 0.46 0.04 0.13 6 0.69 0.68 0.02 0.03 7 0.23 0.37 0.07 0.00 8 0.60 0.32 0.07 0.04 9 (steer serum) 30.02 0.26 0.00 0.04 Avg 1 S.E. 1.15:0.5 0.90:0.3 0.0810.02 0.05:0.01 25 The correlation coefficient for paired estimates of testosterone in the nine samples was 0.99. Average testosterone determined by RIA on solvent extracts of serum without chromatography did not differ signi- ficantly from those obtained by RIA after LH-20 chromatography, indicating that substances separated from testosterone by the chromatography did not compete appreciably with testosterone. 2. Radioimmunoassay (RIA) of Androstenedione Regarding RIA of androstenedione, the procedure was similar to that for testosterone. However, there are some minor differences which are described below. a. Androstenedione Extraction.--Aliquants of 200 pl of each serum sample were extracted with 2 m1 of nanograde benzenezhexane (1:2). This volume of the serum samples was twice as much as used in testos- terone extraction, because serum androstenedione level was lower than that Of testosterone (Lindner, 1959; Lindner and Mann, 1960) and results from preliminary studies indicated that the androstenedione assay was not as sensitive as the testosterone assay (50 vs 20 pg). However, the ratio between serum and the organic solvent volumes should not be narrower than 1:5 because the serum often became emulsified during extraction, and this lowered extraction efficiency. The other steps for extraction were the same as for testosterone. b. Radioimmunoassay (RIA).--Androstenedione antiserum2 was diluted to 1:1000 in 0.1% Knox gelatin in PBS, and 200 pl of this dilution was used per assay tube. The specificity of the antibody was described by Mongkonpunya gt_al, (1974). 2Dr. G. D. Niswender furnished androstenedione antiserum code number 866. 26 c. Assay Validation.--Similar to testosterone RIA, the extraction efficiency of 3H-1,2-androstenedione was relatively constant (84.2:4.4%, n=28). The sensitivity of this assay was less than 50 pg and results obtained from the eight fetal serum samples and one steer serum (Table 2) indicated that the within-sample coefficient of variation was greater than for testosterone assay (52 vs 15%). As a means of validating the androstenedione assay, similar to that described for the testosterone RIA, the results from RIA directly applied to benzenezhexane extracts without chromatography were compared with the results from RIA after androstenedione purification on LH-20 Sephadex (Table 2). The correlation coefficient for estimates with and without chromatography of the nine samples was 0.44. Average androstenedione determined by RIA on solvent extracts of serum without chromatography did not differ significantly from those obtained by RIA after LH-20 chromatography. Thus, substances separated from androstenedione by the chromatography did not compete appreciably with androstenedione. 3. Radioimmunoassay (RIA) of Luteinizing Hormone (LH) The bovine LH assay has been described elsewhere (Oxender, 1971; Oxender et_al,, l9ZRO. The details of the procedure, including some minor changes, are as follows: LH antiserum was developed by repeated injections of NIH-LH-BS into guinea pigs (Appendix I.C.l). Purified bovine LH (LER-1072—2) used for iodination was supplied by Dr. L. E. Reichert Jr. (Emory University, Atlanta, Georgia). This preparation had an LH potency of 1.66 NIH-LH- Si units/mg and showed no FSH activity when tested at 3600 pg in the 27 Steelman-Pohley assay. It had a thyroid stimulating hormone (TSH) contamination estimated at 0.021 USP units/mg. a. Radioiodination.--Purified bovine LH (LER-1072-2) had been previously dispensed into l-ml vials (2.5 p1 of a 1 pg/pl Solution in glass distilled water) and stored at -20°C. These vials were thawed immediately before iodination and iodination was performed at room temperature. Twenty-five p1 of 0.5 M sodium phosphate buffer at pH 7.5 (Appendix I.A.l) was added to the hormone and mixed. One mCi of 1251 (80-140 mCi/m1, Amersham/Searle Corporation, Arlington Heights, Illinois 60005) was added, and the contents gently mixed. Forty pg chloramine-T (Appendix I.A.3, Eastman Organic Chemicals, Rochester, New York) was added to the vial, the vial was stoppered, and the contents were gently mixed by finger tapping. The reaction was stopped at exactly 2 minutes by adding 125 pg sodium metabisulfite (Appendix I.A.4). After thorough mixing, 25 pl of 2.5% bovine serum albumin (BSA, Nutritional Biochemicals, Inc., Cleveland, Ohio) in 0.01 M phosphate buffered saline (PBS) pH 7.0 (PBS-2.5% BSA) was added to diminish the loss of hormone adhering to the glass vial. A 1X12 cm glass column packed with Bio Gel P-60 (Bio Rad Labs., Richmond, California) was equilibrated previously by passing 0.05 M sodium phosphate buffer pH 7.5 (Appendix I.A.2) through, and then 2 ml PBS-2.5% BSA was added and eluted with buffer to reduce non-specific binding of the protein hormone to glass. One hundred pl of transfer solution (Appendix I.A.S) was added to the vial with iodinated LH and the contents of the vial were layered beneath the buffer on the surface of the column. Seventy pl of rinse solution (Appendix I.A.6) was added to the hormone vial, recovered, and also layered beneath the buffer on 28 the column. The iodinated LH was eluted from the column under gravity with 0.05 M sodium phosphate buffer and 15 ml were collected in l-ml aliquots from the column in 12X75 mm disposable glass tubes containing 1 m1 of 2% Knox gelatin in PBS. The elution profile was determined by quantifying the radioactivity of 10 pl from each of the 15 tubes in an automatic gamma counter (Nuclear Chicago Corp., Des Plaines, Illinois). The first peak represented iodinated LH and the second peak represented 125 125I-LH tube was used in the LH RIA. The iodinated free I. The peak LH was quite stable; stored at 4°C, it could be used up to 2 weeks. b. Radioimmunoassay.--Each unknown was assayed in dilution duplicate. Two selected dilutions made in PBS-.l% Knox gelatin (Appendix I.B.4) of each unknown were added to disposable glass culture tubes (12x75 mm) with a Hamilton microliter syringe (Hamilton Co., Whittier, California). PBS with 0.1% Knox gelatin was then added to give a total of 500 p1. As discussed by Hunter (1967), use of two dilutions provided evidence of the specificity of the assay which is not provided with duplicate determinations of the same dilution. Each lot of 100 tubes included 10 tubes containing 0, 0.08, 0.16, 0.32, 0.64, 1.28, 2.56, 5.12, 10.24, and 20.48 ng of standard LH (NIH-LH-BB, National Institutes of Health, Endocrinology Study Section, Bethesda, Maryland; Appendix I.B.S). Two hundred p1 of LH antibody (Appendix I.B.7), hereafter referred to as first antibody, was added at a dilution of l:200,000 to each of the culture tubes and the tubes were incubated at 4°C for 24 hours. Solutions of 125I-LH for RIA were prepared by diluting the stock 125 I-LH with PBS containing 0.1% Knox gelatin so that 100 pl contained about 20,000 CPM. One hundred pl of 125I-LH solution was then added to each tube. Incubation was continued at 4°C for 24 hours. 29 Sheep anti-guinea pig gamma globulin'(SAGPGG, Appendix I.C.2), later referred to as the second antibody, was diluted to a titer which would optimally precipitate the guinea pig gamma globulin. The second antibody formed an antigen—antibody-antibody complex large enough to be precipitated by centrifugation. Two hundred p1 of SAGPGG was added to each tube and incubation continued for 72 hours. After each addition, the tubes were vortexed gently and covered during the incubation to retard evaporation. Following the final incubation, 3 m1 of cold PBS (Appendix I.B.l) 25I-LH. The bound 125 was added to each tube to dilute the unbound 1 I—LH was precipitated by centrifugation at 2500 g for 30 minutes in a refrig- erated centrifuge with a swinging bucket rotor (Sorvall Model RC-3, Ivan Sorvall, Inc., Norwalk, Connecticut). The supernatant fluid was decanted and the tubes allowed to drain for 30 minutes and remaining fluid adherent to the neck and lip of the tube was removed with absorbent tissue. The bound 125 I-LH of the precipitate was then quantified in an automatic gamma counter. Samples were usually counted for 10 minutes or for a total of 4000 or 10,000 counts, whichever accumulated first. This information was punched automatically on paper tape by a Teletypewriter (Teletype Corp., Skokie, Illinois). The standard curve was calculated by multiple regression analysis on a CDC 3600 computer. The values for standard LH assay fit linear, quadratic and cubic components of the regression equationzcorrelation coefficients were consistently 0.99 to 1.00. These regression coefficients were entered manually into an Olivetti computer (Programma 101, Olivetti Underwood, New York, New York). The counting time for each unknown was entered into the computer from the punched tape through a 3O Punched Tape Editor (Beckman Model 6912 Tape Editor, Beckman Instruments, Inc., Fullerton, California), and LH concentrations in the unknowns were computed automatically. Control tubes were included in each assay to determine background radioactivity (tube containing 1:400 control guinea pig serum in place of the first antibody), total counts added (tube containing only 125 I-LH) and counts in the precipitate (tube containing no unknown or standard). Values for the duplicate standards were averaged and plotted as the percent of 125 I-LH precipitated at each dose of LH standard (Figure 2). Oxender £3.21: (1972b) reported that the assay detected as little as 0.1 ng LH (NIH-LH-BS). Mean recoveries of 0.1, 0.5, 2.0 and 8.2 ng of bovine LH added to 100 p1 of serum were 0.1, 0.4, 1.9 and 8.4 ng, although precision was reduced at the high level. 4. Blood Sample Collection and Intravenous Administration One day before each experiment, jugular cannulation (Vikim Vinyl Tubing, Size V10, Bo Lab, Derry, N.H.) was performed in each animal. Approximately 20 cm of the 60 cm tubing was inserted into a jugular vein and affixed to the neck with tag cement (Nasco, Fort Atkinson, Wisconsin) on 7.6X12.7 cm adhesive tape. Each cannula was flushed with 50% glucose in 3.5% sodium citrate and sealed until used for blood collection. The blood sampling procedure included the following steps; 1) 5 ml of blood and citrate were withdrawn and discarded, 2) 10 ml of blood sample was taken and transferred into a 18X85 mm test tube, and 3) the cannula was flushed with 3.5% sodium citrate. When the cannula needed to be sealed for long periods or between long sampling intervals, 50% glucose in 3.5% sodium citrate was used to 31 .vwmwcmum wmuzqquz pom o>w=o omcommow omomsu.m onswfim 32 O L 1 1 V l ‘ O O O O O 9 (D to <1- cu aqnl 0192 01 Gimme) bugpugq % I00 IO Standard NIH-LH-BB (ng) O.| 33 flush and fill the cannula. The frequency of blood sampling in each experiment are shown in the particular experiment and also with results of that experiment. Regarding intravenous administration of GnRH through the cannula, the appropriate volume of normal saline containing GnRH was injected into the cannula and then the cannula was flushed with 5 to 10 ml 3.5% sodium citrate. 5. Design and Procedure of Experiments a. Experiment I. This experiment was designed to study the responses of serum LH, testosterone and androstenedione in pubertal bulls after treatment with GnRH. Three 3X3 Latin squares were used (Table 3). Nine bulls, three each at 2, 4 and 6 months of age (average 75, 93 and 137 kilograms body weight, respectively), and three doses of GnRH (200, 400 or 800 pg) were used. Each bull received the three doses of GnRH by intramuscular injection, one on each of 3 consecutive days. Jugular blood was sampled by indwelling cannula at intervals of 0.5, 0.25 and 0 hour before GnRH, and,then at intervals of 0.25, 0.5, 0.75, 1.0, 1.5, 2.0, 3.0, 4.0, 6.0, 8.0 and 12 hours after GnRH. Blood samples were left at room temperature for 2 to 4 hours, and then kept over night at 4°C before sera were separated by centrifugation at 2500 g for 15 min. The sera were kept at -20°C until assayed for LH, testosterone and androstenedione. Since animal number 9 in Table 3, a 6-month-old bull, was found to be a cryptorchid, the results obtained from this animal were excluded from the analysis. Exclusion of the ninth bull could cause minor bias in estimates of differences by dose or by day. However, the sample average differences by dose or day were rather small. The significant interactions of day or dose with sampling interval for LH and androstenedione could be slightly biased. 34 TABLE 3.--Schedule for GnRH treatments to bulls in experiment I. Day Age 4 Bull 1 2 3 (m0) us 2 1 200a 400 800 2 800 200 400 3 400 800 200 4 4 200 400 800 5 400 800 200 6 800 200 400 6 7 800 200 400 8 200 400 800 9 400 800 200 aValues are pg GnRH given intramuscularly. 35 b. Experiment II. This experiment was designed as a triple split—plot with nesting, to study the chronic effect of GnRH on serum LH, testosterone and testicular development in pubertal bulls. Nine bulls, three each at 2, 4 and 6 months of age, were each given intramuscularly 40 pg of GnRH twice daily (once between 0800 to 0900 hours and the other between 1600 to 1700 hours) for 28 days. Another three bulls at each age were injected with saline as controls. Blood samples were collected by vena puncture immediately before injections and then at 0.5 and 3 hours after injections on day l, 7, 14, 24 and 28 of treatment. It should be mentioned that originally we planned to collect blood samples at 7-day intervals i.e. day l, 7, 14, 21 and 28. Due to a severe snow storm on day 21, we sampled blood on day 24 instead. Blood and serum samples were handled as described before until LH and testosterone were quantified. At 2 months after conclusion of the treatment, the approximate period from beginning of spermatogenesis until sperm are available for ejaculation (Almquist and Amann, 1961), the bulls were castrated. Gonadal and extragonadal sperm were determined by the methods described by Macmillan and Hafs (1968a). Body weight and testicular dimensions (Foote §t_al,, 1970; Hahn 35.31:» 1969) were also recorded during the experimental period. RESULTS AND DISCUSSION Experiment 1: Serum LH, Testosterone and Androstenedione After 200, 400 or 800 pg GnRH (im) in 2-, 4- or 6-Month-Old Bulls a. LH Response: Serum LH responses to GnRH are listed by age of bulls and by doses of GnRH in Tables 4 and 5, respectively. The analysis of variance of these data, but excluding those obtained before GnRH injections, is in Table 6. Serum LH responses to GnRH did not differ significantly among ages (2, 4 or 6 months, Table 4) or among doses (200, 400 or 800 pg, Table 5) of GnRH. On the average serum LH significantly increased (P<0.01) from 1 ng/ml before GnRH to 19 ng/ml within 15 minutes after GnRH. The peak of 24 ng/ml was at 45 minutes, then LH declined progressively to 1 ng/ml within 6 hours after GnRH. This result resembles that reported by Miyachi et_al, (1973). They found that LH and FSH increased to about the same extent at all ages after injection of GnRH in rats. In humans, Rebar g£_al, (1973) found that LH responses to 50, 150 and 450 pg GnRH iv were statistically indistinguishable. Our data suggest that all three of the GnRH doses used in bulls were above that which gives a maximal LH response. Although serum LH after 200, 400 or 800 pg GnRH (Table 5) did not differ significantly, the duration of the LH response to intramuscular injection of 800 pg GnRH was longer than that after 200 pg GnRH (4 vs 36 37 TABLE 4.-—Serum LH after injection (im) of GnRH into bulls at 2, 4 or 6 months of age. Sampling interval Age (months) of bull after GnRH 2 4 6 (hr) (ng/ml) Before 1:1 1:1 1:1 0.25 19138 21:2 15:1 0.50 20:2 24:2 19:3 0.75 23:2 25:2 22:4 1.0 20:2 22:2 20:5 1.5 17:3 19:2 19:7 2.0 18:5 20:2 18:7 3.0 10:3 13:2 14:7 4.0 4:1 5:1 5:3 6.0 1:1 1:1 1:1 8.0 1:1 1:1 1:1 12.0 1:1 2:1 1:1 aEach value is the mean : standard error of LH responses to 200, 400 and 800 pg GnRH. 38 TABLE 5.--Serum LH after injection (im) of 200, 400 or 800 pg GnRH into bulls. Dose of GnRH (pg) Sampling interval after GnRH 200 400 800 (hr) (Hg/m1) Before 1:1 1:1 1:1 0.25 20:4a 17:2 19:2 0.50 20:2 21:3 23:2 0 75 22+2 22+3 27+3 1.0 18:2 19:3 25:3 1.5 13:1 16:2 26:5 2.0 13:2 18:4 26:6 3.0 8:1 9:2 19:5 4 0 2:1 3:1 8:2 6 0 1:1 1:1 1+1 8 0 1:1 1:1 1:1 12.0 2:1 1:1 1:1 aEach value is the mean : standard error of observations on three bulls at 2 and 4 months, and on two bulls at 6 months. TABLE 6.--Summary of ANOVA for LH data after GnRH. 39 Source df SS MS PI Age 2 153.40 76.70 .73 N.S. Day 2 424.18 212.09 2.03 N.S. Animal/Age 5 1,572.97 314.59 3.01 <.10 Dose-Direct 2 565.44 282.72 2.70 N.S Error "a" 12 1,253.18 104.43 Sampling Interval 10 19,572.49 1,957.24 77.54 <.01 Age X Interval 20 271.78 13.58 .54 . N.S Day X Interval 20 1,709.69 85.48 3.38 <.01 Dose X Interval 20 1,238.48 61.92 2.43 <.01 Error "b" 170 4,291.14 25.24 40 3 hours, respectively). This explains the significant interaction in LH response between the dose‘bf GnRH and the sampling interval after GnRH injection (Table 6). This result is in agreement with the observation (Golter 35 31., 1973) that the durations of LH responses to 0.03, 0.3 and 3.0 pg GnRH per kilogram of body weight of Hereford bulls were 55, 220 and 366 minutes, respectively. Zolman §t_§1, (1973) reported that peak and duration of serum LH responses in mature Holstein bulls varied with the dose of GnRH up to 160 pg when given intravenously. In my study the differences in LH responses among animals within age only approached significance (Table 6; P<0.10). Thus, the marked increase in serum LH which occurred after each dose of GnRH was rather consistent among animals. The magnitude of LH peaks after GnRH in my prepubertal bulls was only about 60% of those reported by Golter et_al: (1973) and Zolman gt_al, (1973), suggesting that adults may respond to GnRH with greater release of LH. Route of administration of the GnRH also could have affected this result. Examples of the influences of how steroid environment can affect the LH response to GnRH have been published. For example, Krey §t_al, (1972) demonstrated that in female monkeys, the LH responses to 10 pg GnRH at mid-luteal phase (day 19-20) was greater (P<0.05) than at other phases of the menstrual cycle. Similarly in women (Yen gt_al., 1972) the response was greatest during day 21-23 of the cycle. Convey (1973) concluded in his review that "the pituitary responsiveness to exogenous GnRH varies with reproductive state, probably as a result of variation in the steroid environment". Moreover, it has been hypothesized that the pituitary 41 of immature animals is more sensitive to negative steroid feedback as compared with that of mature animals (Smith and Davidson, 1968). Mode of administration also can affect the LH response to GnRH. Peck £5 31, (1973) studied the release of LH in heifers during the late luteal phase of the estrous cycle after intravenous or intra- uterine administration of GnRH. After 1 mg of GnRH, serum LH at 2 hours was 38.3 and 28.7 ng/ml for intravenous and intrauterine administration, respectively. Golter EE.§l: (1973) studied different vehicles for GnRH in bulls, and found that 1.3 pg GnRH (sc) per kilogram of body weight in acidified saline or in 2% carboxymethyl cellulose showed comparable peaks of LH responses (41:10 vs 55:10 ng/ml), but the later vehicle prolonged (P<0.05) the duration of LH response. Though there was no significant (P>0.10) difference between LH responses among days, a significant (P<0.05) interaction between day and interval indicated that peak and duration of LH responses differed by day of treatment (Figure 3 and Table 7). That is, serum LH responses (Midaysl, 2 and 3 rose sharply within 15 minutes after GnRH, but the magnitude of the peak on Day 1 was significantly greater (P<0.05) than those on days 2 and 3. While serum LH onwmmno uzmflo mo Howwo oumwcmum a same can mucomowamw ucflom comm .mxmw o>fiu=oomcoo conga :o mfiasn one“ Imam mo fieflv cowuomncfi wmumm :4 Edwmm--.m owsmflm 43 Imco cote mesa: w m N N (O (luv/511) H'I wmes m TABLE 7. 44 --Serum LH after injection (im) of GnRH into bulls on three consecutive days. Sampling interval Day of administration after GnRH 1 2 3 (hr) (ng/m1) Before 2:13 1:1 1:1 0.25 20:3 18:2 19:3 0.50 22:2 21:3 20:2 0.75 28:3 23:2 21:1* 1.0 28:4 18:2* 16:1* 1.5 27:5 16+2* 12:2* 2.0 29:5 16:2* 11:3* 3.0 18:5 9:2 9:3 4.0 8:3 3:2 3:1 6.0 1:1 1:2 1:1 8.0 1:1 1:1 1:1 12.0 2:1 1:1 1:1 aMean : standard errors for eight bulls each given 200, 400 or 800 pg GnRH im. *Significantly less than value on day 1. 45 maintain the LH response after 3 hours, because GnRH decreased the LH concentration and content in the pituitary. b. Testosterone Response: Since the serum LH responses to GnRH were similar at 2, 4 and 6 months of age regardless of the dose of GnRH, I decided to determine the testosterone response to increased endogenous LH. Serum testosterone, determined in the same samples in which I determined LH, are shown in Table 8 and Figure 4. The summary of the analysis of variance of the data, omitting samples taken before GnRH, is listed in Table 9. In Z-month-old bulls, testosterone averaged 0.4 ng/ml before GnRH, increased slightly to 0.6 ng/ml at 2-3 hours and declined to 0.3 ng/ml at 6 hours after GnRH (P>0.05). The comparable values were 0.6, 1.2 and 0.5 ng/ml in 4—month-old bulls (P>0.05). In contrast, in 6-month- old bulls, testosterone averaged 1.7 ng/ml before GnRH, increased significantly (P<0.05) within 1 hour (4.6 ng/ml) and peaked (5.3 ng/ml) at about 2 hours after GnRH injection, and then returned to 1 ng/ml 6 to 8 hours after GnRH. The lack of a significant increase in serum testosterone in 2- or 4-month-old bulls was not caused by a lack of LH in the blood as shown in Table 4. Hooker (1970) reported that differentiation of mesenchymal cells or fibroblastic cells of the testicular interstitium in bull calves began about 3 1/2 months of age and cytochemically they became fully developed as secretory cells during the next 6 1/2 months. These results clearly demonstrate a difference in sensitivity of the Leydig cells to increased endogenous LH in bulls at different ages, and suggests that this testicular change from 2 to 6 months of age is related with sexual maturation. 46 TABLE 8.-—Serum testosterone after injection (im) of GnRH into bulls at 2, 4 or 6 months of age. Age (months) of bull Sampling interval after GnRH 2 4 6 (hr) (rig/m1) Before 0.4:.1 0.6:.1 .7: 0.5 0.4:.1a 1.0:.1 .1: 1.0 0.5:.1 0.9:.2 .6:. l 5 0.5: 1 1.0:.2 .8: 2 0 0.6: 1 1.2:.1 .3: 3 0 0.7+ 1 1.0:.2 .1: 4 0 0.6: l 0.9:.2 .2: 6 O 0.3: l 0.6:.1 .1: 8 0 0.2+ 1 0.4:.1 .6:. 3Each value is the mean : standard error of testosterone responses to 200, 400 and 800 pg GnRH. 47 Figure 4.--Serum testosterone after injection (im) of GnRH into bulls at 2, 4 or 6 months of age. Each point represents the mean : standard error of nine or six observations, three in each bull. Serum Testosterone (rig/ml) 01 I A 04 N O 48 6 mo; n=2 1- 1 \O \ \ 1 P ¥ .\ ¥ 4mo, n= 3 \, "gm ”he- i\- 8-": .13 i’ i i 1 211101; n= 3 3 1 36\9 0 2 ‘4 Hours after GnRH 49 TABLE 9.--Summary of ANOVA for testosterone data after GnRH_ Source df SS MS F PI Age 2 287.48 143.74 2,356.39 <.01 Day 2 .06 .03 .49 N.S Animal/Age 5 9.48 1.89 30.98 <.01 Dose-Direct 2 .21 .10 1.64 N.S Error "a" 12 .73 .061 Sampling Interval 7 89.88 12.84 61.14 <.01 Age X Interval 14 94.10 6.72 32.00 <.01 Day X Interval 14 3.27 .23 1.09 N.S. Dose X Interval 14 2.40 .17 .81 N.S. Error "b" 119 25.32 .21 50 Our results showing increased release of testosterone in 6-month- old bulls is in agreement with the results of Zolman and Convey (1973) who reported that 10, 40 and 160 pg of GnRH given to mature bulls caused significant increases in serum testosterone and androstenedione (184 and 75%, respectively). In contrast to the LH response to GnRH, the significant day x interval interaction did not occur for testosterone (Table 9). Evidently, the reduced LH response on day 2 and 3 (Table 7) was still high enough to give a maximal testosterone surge. Smith §t_§l, (1973) reported that, on the average, an increase in LH of only 2.4 ng/ml resulted in increased testosterone. Moreover, unlike LH, there was a significant difference in testosterone among animals within age. The coefficients of variation were 100, 80 and 73% in 2-, 4— and 6-month-old bulls, respectively; indicating that there was more variation in serum testosterone in younger than in older bulls. c. Androstenedione Response: Blood androstenedione at various intervals after GnRH is listed by ages of bulls in Table 10 and by days of treatments in Table 11. A graphic illustration of an interaction between day and interval is in Figure 5. The concentrations of blood androstenedione were about 68, 83 and 93% lower than testosterone in 2-, 4- and 6-month-old bulls, respectively. In contrast to testosterone, blood androstenedione, on the average, after GnRH (200, 400 or 800 pg) increased significantly (P<0.05) at 2, 4 or 6 months of age, and on each day of treatment (Table 10 and 11). On the average androstenedione increased about three-fold from 119 pg/ml before GnRH to an average of 297 pg/ml at 51 .HH3n some cw oco .mCOwum>womno ucwflo mo came on» mucomowmou ucfiom comm .mxmo o>flpzoomcoo omega :0 mfiasn oucfi :mco mo meflv :ofluooncfl Houmm ocoflwocoumOchm Edwomun.m enamflm 52 Imco coco mono: iOmN .Oom (IuJ/bd) auogpauaismpuv Lumas 53 TABLE 10.--Serum androstenedione after injection (im) of GnRH into bulls at 2, 4 or 6 months of age. Sampling interval Age (months) 0f bUII after GnRH 2 4 6 (hr) (pg/m1) Before 130:50a 100:50 120:80 0.5 110:40 310:80 190:120 1.0 260:50 330:80 220:50 2.0 380:90 260:70 310:110 4.0 200:70 270:70 160:90 8.0 50:20 100:50 10:7 3Each value is the mean : standard error of androstenedione response to 200, 400 and 800 pg GnRH. “or Q ‘3.“ Slum? LI. ".‘a .5571"... ' 54 TABLE ll.--Serum androstenedione after injection (im) of GnRH into bulls on three consecutive days. Sampling interval Day after GnRH l 2 3 (hr) (pg/m1) Before 63:543 64:30 231:61 0.5 105:60 139:65 370:88 1.0 261:92 316:36 248:55 2.0 394:96 94:30 456:65 4.0 418:50 0:0 227:63 8.0 34:23 64:60 76:31 3Each value is the mean : standard error of eight bulls each given 200, 400 or 800 pg GnRH im. i‘ JtFfivu . ..'l y no“: W.“ W; 55 l to 2 hours, and then declined to 58 pg/ml at 8 hours after GnRH. Neither the differences among ages nor those among days approached statistical significance. Doses of GnRH had essentially equal effects on stimulating serum androstenedione concentrations, although androstenedione differed significantly (P<0.05) among animals within age (Table 12). A significant (P<0.01) interaction in the androstenedione responses associated with day and interval (Table 12) is illustrated in Figure 5. On day l, androstenedione increased gradually until 4 hours after GnRH. On day 2, androstenedione peaked at 1 hour, and on day 3 it peaked at 2 hours. Serum LH also exhibited a significant day x interval interaction. Thus, the pattern of androstenedione response to GnRH was generally similar to that for LH although it lagged behind the LH response by 15 min to 2 hours. These data may suggest that LH affects serum androstenedione. Thus, the testes of 2- or 4-month-old bulls are as capable of androstenedione synthesis as the testes of 6-month-old bulls, while testosterone synthesis was not evident in large quantity until 6 months. In the testes of 6-month-old bulls the peak of testosterone response after GnRH was about 20-fold greater than that for androstenedione (5.3 vs 0.3 ng/ml). Others (Lindner, 1959; Lindner and Mann, 1960) have reported that androstenedione is the predominant androgen before puberty in bulls. But this is the first report that androstenedione is capable of responding to LH stimulation before puberty. My data suggest that LH stimulation of testicular testosterone synthesis is probably blocked before 6 months of age possibly in part at the conversion of androstenedione to testosterone, since serum TABLE 12.--Summary of ANOVA for androstenedione data after GnRH. S6 Source df SS MS F PI Age 2 86,022 43,011 .69 N.S Day 2 19,813 9,906 .39 N.S Animal/Age 5 531,674 106,334 .17 <.05 Dose-Direct 2 93,259 46,629 .83 N.S Error "a" 12 305,820 25,485 Sampling Interval 5 1,058,110 211,622 .31 <.01 Age X Interval 10 291,543 29,154 .87 N.S Day X Interval 10 1,190,282 119,028 .55 <.01 Dose X Interval 10 141,697 14,169 .42 N.S. Error "b" 85 2,849,286 33,521 ‘rh-wfi-‘mujmmg. o '.- . 57 androstenedione in young bulls (2 and 4 months of age) was increased significantly whereas testosterone was not increased. It would be interesting to test whether blockade of the activity of 17-8- hydroxysteroid dehydrogenase would cause the accumulation of androstenedione rather than testosterone after LH stimulation in postpubertal bulls. In other words, deficiency of l7-B-hydroxysteroid dehydrogenase may account for low testosterone in 2- and 4-month-old bulls relative to 6-month—old bulls, and activation of this enzyme may be associated with onset of testosterone synthesis at puberty. In summary, results from this experiment indicated that serum LH increased about 24-fold within 45 minutes after GnRH injection, and declined to pretreatment concentrations within 6 hours after GnRH. This pattern of LH response to GnRH was similar at all ages of bulls and at all doses of GnRH, except that the higher dose (800 pg) of GnRH showed a longer duration of LH response. In addition, there was a decreased responsiveness in LH release to repeated administrations of GnRH. In response to the increased LH release, serum testosterone in 6—month-old bulls increased about 3-fold within 2 hours after GnRH, but in 2- or 4-month-old bulls serum testosterone was not significantly altered. Unlike testosterone, androstenedione significantly increased at all ages and the pattern of the androstenedione response to GnRH was generally similar to that for LH. .__....... ,. ,-.~ _ v- “use. T , . _ i- “ “J, 58 Experiment 11: Effect of GnRH Injections Twice Daily for 28 Days on Serum LH and Testosterone, and on Reproductive Development in Pubertal Bulls a. Body Weight: Within the same age, the average body weights of bulls at the beginning of the experiment were not statistically different. The averages were 73, 117 and 173 kilograms for bulls at 2, 4 and 6 months of age, respectively. Average daily gain, measured during the experimental period, between control and GnRH-treated groups did not differ significantly. However, daily rate of gain during the experimental period increased with advancing age (0.50, 0.88 and 1.27 kilograms for 2, 4 and 6 months of age, respectively). Macmillan and Hafs (1968b) reported that the body weight of bulls at 2, 4 and 6 months were 73, 138 and 179 kilograms, respectively, and the increase in body weight with increased age was linear. They concluded that the period of puberty was not associated with marked changes in rate of growth. b. Testicular Dimension: The average testicular circumference, width and length are shown in Table 13. Statistical comparisons of these parameters between bulls given saline and bulls given GnRH were not significantly different (P>0.05). Chakraborty §t_313 (1973b) infUSed GnRH at 2.3 pg/hour into anestrous ewes for 20 hours and found no effect on ovarian follicular growth. Though their investigation cannot be directly compared with the present study in prepubertal bulls, experimental animals in both studies were sexually inactive. Macmillan and Hafs (1968a,b and 1969) demonstrated that testicular growth (weight, width and length) was 59 TABLE 13.-—Average testicular dimensions of bulls. . . . a Test1cular d1mens1on Age GnRH at beginning (mg/day) Circumference Width Length (mo) (cm) 2 0 _ 15.0 3.9 5.7 40X2 14.5 3.8 5.8 4 0 19.5 4 4 7 4 40X2 19.1 4.3 7.2 6 0 25.5 6 2 10 2 40X2 26.5 6.1 9.8 aThe averages of 4 measurements during the experimental period. 60 associated with increasing concentrations of hypothalamic LH releasing factor, plasma LH and body weight. Thus, failure of GnRH to stimulate testis growth could be attributed to the gradual decrease of the LH response to the repeated treatment of GnRH as demonstrated in Experiment I and in this experiment (Figures 6 and 7). c. Testicular and Extratesticular Sperm Counts: Testicular spermatozoa were found at 60 days after the last GnRH injection in all bulls except two which received GnRH treatment starting at 2 months of age (Table 14). At 9 months I found appreciable numbers of testicular spermatozoa, and this is in agreement with the report by Macmillan and Hafs (19683); the youngest age at which they found sperm was 5 months. However, they reported nearly 3-fold more total testicular sperm than I found (average 0.16 vs 0.5 and 6 vs 15 billion at 7 and 9 months, respectively). These differences in sperm numbers are principally due to reduced (34 and 45% at 7 and 9 months, respec- tively) testicular weight and also sperm concentrations in the present study (63 and 52% for 7 and 9 months, respectively). In my bulls, total epididymal sperm numbers, at 9 months of age were 3.28X109, similar to the 4.00X109 reported by Macmillan and Hafs (1968a). In general, epididymidal sperm can be enumerated more precisely than testicular sperm because more debris is present in testicular homogenates. Thus, although the rate of spermatogenesis in my bulls may have been somewhat less than those reported by Macmillan and Hafs (1968a), the general changes from 5 to 9 months of age were similar. .oowzu coca mmofi mfi .mHHSn mo poneszn .moNoumeomm can flHHH> cam HH> .H> mowmpmv mowumeuomm moosaoch 61 m.amwhm.moefi m.~emha.omefi e.~ho.e H.3om.- m.ooe.om e.ANho.eeN Nxoe m.emmnm.meeH e.oeehm.umwfi m.oo~.e o.mhw.mm n.3hm.em m.emhm.mem o a N.Hhe.m Asv oo.e o.oomo.o o.ohm.o m.ohm.om H.6hm.ea Nxoe Aug o.e . may 00.3 N.oom.o N.Nhe.m v.3hw.om m.wh~.ooH o A Aug o.m 0 “so Ho.o one: m.o m.~om.eu o.NHhm.me Nxoe ”my m.~ o o.oouo.o H.ohm.o m.Hhe.mH H.whm.mm o m Aeono fieonv fiuase\mo~xv AM\oonU Agog hm: axoe\wnv mos: xvoniumo: Hawk Hmuob cowumwwcoocoo Eduonom mo unmfimz xmco owe Ewomw Hmexofiuwmo HmpOH ewomm uwflsowumoh coconomasouwu moumou Hooch Hazm w M .ucoEHHouu :dcu mo mxmo mm wopmm mxmv co um Enomm Hwowsxofivflmo cam .euomm ecu euwm Hmvmcouuu.vH mam0.1) between control and treated animals. At this age, the total sperm counts were greater in the tail (1,786 million) than in the head-body (1,434 million) of the epididymis. This observation is in agreement with reports by Almquist and Amann (1961) and Macmillan and Hafs (1968a); the tail of epididymis is the major storage organ for sperm awaiting ejaculation. However, at 5 or 7 months of age, sperm numbers were higher in the head-body epididymis. In the younger bulls, spermatogenesis had begun only recently and the sperm probably had not yet accumulated in the tail of epididymis. d. Serum LH: From Experiment I, intramuscular administration of GnRH into 6- month-old bulls increased serum LH and serum testosterone within 45 minutes. After injection of LH into mature bulls (Smith §£_§1,, 1973), testosterone doubled in blood within 1 hour, peaked at Its to 2 hours and remained high for at least 4 hours. Thus, in the present experiment, blood was sampled at 0, 0.5 and 3 hours after injections of GnRH beginning between 0800-0900 hours and again between 1600-1700 hours on days 1, 7, 14, 24 and 28. Serum LH values in these samples are shown in Table 15. The summary of analyses of variance of the same data are in Table 16. Administration of GnRH (40 pg im twice daily for 28 days) signifi- cantly (P<0.01) increased serum LH (Table 15) in all bulls. This effect was similar in bulls at all ages and between AM or PM injections. However, serum LH responses were significantly affected by day (1, 7, 14, 24 and 28) of the experimental period (P<0.05), and by the sampling ted twice daily fer 28 days in bulls 1njec TABLE 15.-Serum LH response to saline or GnRH (40 U8, im) Age of bull 6 months 4 months 2 months Interval relative 63 uzosoa ~4u>c> d-u1uz waosq- '¢1\ 5 OH .HNO OHH OHQ .4 +l +l +l +1 +l +l +l +1 +l +l +l +l +l +1 5 VON MQO OLD? HOV \OO’: NVN NwH Nlhl") HMH NN H . O mmm VVID NVM mnm NV) 1: OOH OOH OON OOO OO "-4 +| +l +l +1 +l +l H +l +1 +| +l +| +I +l '3 NNl‘s OON 000000 asp-4.... 03‘. U) HHN HON OOV HHH OH asaau: c>h-os c>m>c> 61h.q- fine a OOH OVO HNH ONO OH +l +1 +l +I +l +1 +| +l +1 +l +| +| +l +| 5 MVH OOO [\HO \th ON NEW ONN 'HON OOH HID N H A H E 0 20 “31.05 \OHO mNm OHM ON I: U OOO OOO OOH OOO CO -H +l+l+l +l+l+l +l+l+l +l+l+| +1“ -; ox—aw: u:h«b« «>c>r4 \o~o.~ .«cn U) OHH HOH HHM OON HO N M No-N Oa;H NMN Ocom \DH 5 CNN OHH OHN HOO MM +I+l+l +|+l+l +l+l+l +l+l+l H H 5 aorxco mono OON Q'BH [\B ONO OMH Hsov HNN V0 M N 03 0:.) MNN NNO NO OON V? "-4 OOO OOO OOH HOH HO H +1 +1 +1 +1 +1 +I +1 +I +I +I +I +I +I +1 (can (\MLD OOH Ohm ONLO nae OOO OOH OOQ’ MHH HO 1’: O -H p U thO LOO LDO IDO m $ 00 o o o o o o o o a o o . ~21 c>v> c>v> c>N> c>N> c> "-1 O p E Z 2 3 a 1 a .3. a >‘ <- m H h H O O O O 4.6:3.8 1.6:O.6 l.S:0.4 3.7:1.7 2.2:1.6 1.0:0.4 3.0 1.9:0.7 7.2:2 1.8:0.6 l.2:0.1 1.6:O.9 8.5:0.6 12.5:10.3 1.2:O.4 0.7:O.2 0.9:O.2 8.0:3.9 5.1:2.9 0.4 0.2 1.1 PM .5 0.7 0.5 1.1:O.5 2.7:1.2 2:1.2 3. 0.7 3.0 Table 15.-(Cont'd). 1.0:0.4 4 0:2.1 0.3 1.4:0.2 : 0.6:0.1 1. 2 1.0:0.3 8.5:2. 1 1.0:0.4 1.9:0.6 .7:1.2 1:2.1 1 4. 6:0.7 0.1 1. 0.5 1 AM D24 .8 8:0 1. .2:0.4 .6:O.9 l 2.0:O.8 l.1:0.4 3.0 .4 5:0 2.4:1.0 1.0:0.3 1. .4 2:0 2.6:1.S 0.9:0.0 1. 6.8:4.0 lO.2:3.0 6:0.9 .1:0.3 1. 1 l .7 6:0 1. 0.7:0.2 PM 7:1.4 4. 0.2 0. 1 0.5 .5 2.5:0 0.2 1:1.1 2. 0 .4:1. 3.0 64 \OOM t\ OOO O +I+I+I +I OOO \O NNH H \OHH m OOO O +l+l+l +| £00100 m HOO O MOH O: OHN O +l+l+l +1 OOOH O HOO H H NMLD O OOO H +I+I+I +1 HOLD M HHH N H010 O OOOO O +l+|+l +1 HBO oo HNN O H #50: N OHO O +l+|+l +| NBM 00 HNN :5 D28 .8:0.5 1 5.0:3.9 2. 7.5:1.2 l.4:0.1 1.2:0.3 2. 0.2 1:0.9 0. 0.5 5:0.2 5:1.5 1. 5 1 1.0 5.1:l.8 2:0. 2. 3.0 1.7:0.2 .5:0.2 1 3:0.2 1. Mean 3Each value is the mean (: SE) of three bulls. 65 . . . 4. .. . . u; .. 1.4.3552!!! utilisi- oo.v No.NmN we mom .m.z mwm.H vm.m Nm.ee w zaoz< mo Numeasm--.ofi mgm

umucH wcwfimemm .Au.u=ouv--.ofi oflnme 67 intervals relative to time of GnRH injection (P<0.01). Several two- and three-way interactions were statistically significant (Table 16), but their physiological significance is obscure. Considering only the saline control group (Table 15), average serum LH during the treatment period did not differ significantly among age groups (1.3, 1.5 and 1.7 ng/ml for 2, 4 and 6 months, respectively). Serum LH levels in control bulls between 2 and 9 months of age were essentially within the same range (Table 17). The small increase in LH concentration between 4 and 5 months of age was not significant (P>0.05). This finding is in agreement with that reported by Odell 25 21° (1970), who found no marked differences in mean LH concentrations in bulls and heifers between 1 and 5 months of age. However, Macmillan and Hafs (1968b) found LH levels increased between 2 and 4 months and again between 6 and 10 months of age. In contrast to serum LH, serum testosterone increased dramatically between 2 and 9 months of age. These data suggest two possibilities: I) only minute amounts of LH are required fbr stimulation of testosterone synthesis in Leydig cells; and 2) other hormonal factors may be involved in the stimulation of testosterone synthesis. Before GnRH was given to these animals serum LH averaged 1.8:0.2 ng/ml; this was comparable to that in the control group (1.3:O.1 ng/ml). Serum LH markedly increased at 30 minutes after GnRH injection (Table 15 and Figure 6) and declined to or approached pretreatment values within 3 hours after GnRH. It was observed that serum LH concentrations at 30 minutes and 3 hours after GnRH were correlated (r = 0.57). This relationship is probably only a reflection of the time required for degradation of LH in blood. In contrast, serum LH levels befbre GnRH 68 TABLE l7.--Serum LH and testosterone in control bulls treated with saline twice daily for 28 days. Age at Age Day the beginning at sampling of expt. LH Testosterone (mo) -——— (ng/ml) 2 1 0.7:.1a 0.7:.2 2 3 28 1.6:.3 0.6:.1 4 58 0.7:.2 0.1:.0 S 88 1.8:.3 1.7:.6 4 l 1.2:.2 1.7:.7 4 S 28 2.0:.6 2.6:.5 6 58 0.9:.4 2.2:.4 7 88 0.1:.1 2.1:.7 6 l 1.6:.4 3.2:.8 6 7 28 1.9:.7 3.8: 8 S8 0.6:.4 3.6:1.l 9 88 0.6:.2 2.8:1.1 aEach value is mean : standard error of six observations from each of three bulls. Blood samples were taken at O, 0.5 and 3 hours after saline injection at 8 AM and 5 PM on each day; except days 58 and 88. 69 Ill-I‘ll .xapcmommflcmNm Become no: cap zone omsmoon mason m vcw m.o .0 mo mowmuo>m one: mosfiw> Hoyucou .HHSQ come a“ 03¢ .m:oau¢>H0mno :ooucwfio mo Honpo vamvcmum H :moE may mucomopmoa Han comm .wm no em .vH .n .H mxmw co mHHDD can“ race mo AEMO :ofiuoomcfi Houwm maze; m 90 m.o .o um 24 Espomuu.o oaswfim ‘\\\\\\ s\\\\\\\ O N Q Control (saline) 70 E] 3.0 hr. after GnRH injection C] Before GnRH injection (Ohr.) a 0.5 hr. after GnRH injection 5? (luv/5U) H'l mo: .\‘§\\\\\\V :- was 1995 5‘\\\\\\\ =- {E $0906! 3.\\\\\\\\V 9. wmas :- 3% mac .‘C.\\\\\:\: {m \\\\\\\\\\\\\\\\\\ :- ‘38 D24» 028 DH D7 Days of GnRH treatment 71 TABLE 18.--Serum LH and testosterone in bulls treated with GnRH twice daily for 28 days. Age at Age Day the beginning at sampling of expt. LH Testosterone (m0) ——-—- (Hg/m1) 2 1 11.0:s.0a 1:0.2 2 3 28 3.3:l.6 0.4:0.l 4 58 2.1:0.8 0.3:0.l S 88 --- --- 4 l 8.2:2.6 1.8:0.3 4 S 28 S.O:O.9 2.5:0.4 6 S8 0.1:0.l 2.8:0.8 7 88 0.5:0.2 4.7:1.4 6 1 S.3:l.3 S.0:0.9 6 7 28 l.8:0.2 6.2:0.8 8 58 O.l:0.l 0.8:O.3 9 88 O.7:O.2 7.3:2.2 3Each value is mean : standard error of six observations from each of three bulls. Blood samples were taken at O, 0.5 and 3 hours after GnRH at about 8 AM and 5 PM on each day; except days 58 and 88. 72 was not associated with LH concentrations at 30 minutes (r = 0.02). The serum LH response to GnRH at 30 minutes after GnRH was greatest following the first injection (AM) on day 1 (Figure 7); the values were 32.7, 27.4 and 14.7 ng/ml for 2-, 4- and 6-month-old bulls, respectively (Table 15). Comparable data after the second injection (PM) on day l were 23.6, 12.9 and 8.4 ng/ml, respectively (Table 15). The average (AM+PM) serum LH concentrations at 30 minutes after GnRH on day 1 from all bulls (20.0:4.6 ng/ml) were significantly (P<0.05) higher than those on days 7, 14, 24 and 28 which averaged 5.5:.8, 6.5:.9, 5.0:.9 and 6.2:.2 ng/ml, respectively (Figure 6). Notice that the values from day 7 to 28 are consistent; the principle decrease in the LH response to GnRH occurred before day 7. Although the younger bulls apparently had a greater LH response, this was statistically insignificant (P>0.05). Moreover, two—way interactions between age and other factors (treatment x age, day x age, AM—PM x age and interval x age; Table 16) were not significant (P>0.05) suggesting that the pituitaries of animals at different ages responded equally to the GnRH treatment. This finding together with that of Macmillan and Hafs (1968b) who found that GnRH was undetectable in the hypothalami from bulls before 5 months of age suggests that probably GnRH in the hypothalamus is a limiting factor for LH release before puberty. This result is in accord with those of Experiment I and Miyachi gt al. (1973) who found that, in rats, LH and FSH increased to about the same extent at all ages after injection of GnRH. Although there was no significant difference between the LH after AM or PM injections of GnRH, significant interactions between day and AM-PM, and between sampling interval and AM—PM (Table 16) were due to 73 .xaucmuflmwcmflm Hommwv no: vac xosu omsmoon :a cam z< mo mommuo>m one: mosfim> Houucou .HHDQ :umo cfi oco .m:0wum>nomno on“: mo gonna unmvcmum H came ecu mucomoamoh awn comm .AZQV mcfico>o Ho AZ2 ng/ml). This seeming failure of endogenous LH to stimulate testosterone is most likely associated with non-responsiveness of the Leydig cells of bulls at 2 and 4 months. For example the analysis of variance (Table 20) showed that testosterone was significantly affected by age. In the 6-month—old bulls treated with GnRH, average testos- terone (5.8:O.3 ng/ml) was significantly (P<0.05) higher than those in 2- or 4-month—old bulls treated with GnRH which averaged O.7:0.l and 2.0:0.l ng/ml (Table 19) respectively. The comparable data for control TABLE 19.-Serum testosterone response to saline or GnRH (40 pg, im) injected twice daily for 28 days in bulls at 2, 4 or 6 months of age. Age of bull 6 months 4 months 2 months Interval relative GnRH Saline GnRH Saline GnRH Saline to injection AM-PM Day (mg/m1) .8 4.5:3.4 3.4:0 6:1.1 l.9:l.3 1. 1 .4 0.4:0.1 0.1:0 0.5:0 0.4:0.1a 0.4:0. AM :3.0 1:1.8 5.5 2.1:O.S .1 0.4 5:0 0. l 0.5 5. 3:0.4 1. .6:0.8 0.5:0.3 .6 l. 0.1 0. 3.0 5.4:3.9 3.6:2.6 5:0.3 1.8:0.6 2.6:1.0 l. 4.0:1.0 2:0.9 1:0.9 1. 2:0.0 1. PM 7.6:2.8 3.8:1.S 6:3.5 1.4:0.4 5. 6:1.8 6:0.9 3. 1. 0.5 77 1.6:0.4 1. 0.1 0. 3.0 7.8:1.8 7.4:l.6 .1 .3 1.4:0 .9 3:0 1.1:0.7 l.7:l.0 1. .6:1.1 1 0.5:0.2 0.2:0.0 0.6:0.1 3:0.2 0. 0.9:0 0.3:0.2 0.3:0.3 0.5 0.7 2. 6.8:3.0 3:0.7 1. 2:0.7 1. 3.0 3 7.3:1. 7.3:1.7 8.5:3.2 2.5:1. .6 7:0 3.1:0.7 l. 1 1:0.3 0.9:0.2 l. 0.8:0.3 0. PM S.4:1.4 4.1:l.2 0.6 2.4:1.4 2.3:o.3 4.4:o.s 0.2 0.5 5 2:0.5 1. 0.4:0.2 3.0 7.1:1.7 5.4:1. 6.4:2.0 4.6:l.l 2.6:0.8 2.1:1.2 2. l.9:0.9 0.1:0.1 0.1:0.l 0.2:0.1 0.2:0.1 0.3:0.2 0.3:0.3 D14 5 1:1.2 0.6 2.6:1.2 1 0.5 S.0:2.2 0.9:0.3 3.0 1.8:0.9 1.7:0.5 6.7:2.6 6.5:3.4 2.0:1.S 4.3:0.S 0.8:0.2 .3 0.4:0 PM 8:3.1 S.S:2.3 5. 6:0.5 1. 1:0.4 1. 1.3 0.5 7:1. 3. :0.5 :0.8 2.2 0.5:0.3 3.0 '.H.. :‘A..t" '1'.WM’LS|.. 1‘: on Table 19.-(Cont'd). Hoor’) Ln NNH N +I+I+I +I NOV 00 0005 \O Han m HOH H +I+I+I +I ONM \D HH-a' N [\Hm N) OHO H +l+l+| +I r0503 \0 HHO M whm co HOO O +l+l+t +| \omm O NHH N NHH w OOO O +l+l+l +I MVH v OOO H NNM Ln OOO O +|+I+I +| vv-zr O OOO H I324 4.8:1.5 3.3:1.3 2.1:1.0 0.2 0.8:0.2 2. 1.1 3.9:l.7 0.5:0.1 2. 0. 0.5 .5 3:0 1. 2:0.8 4.6:0.5 0.5:0.3 3.0 78 mcov O MHN H +l+|+l +l oomoo O com I\ \OOLO In OOH H +I+l+l +l OOO N m-a-N Vt HOOK O HOO H +i+l+t +1 N00 5 NNN N \Dl\\o N) No... o' +l+l+l +l NLnO O MH H HOH N OOO O +l+i+l +| HOH \0 OOO O NNM N OOO O +|+I+l +1 MM? \0 OOO O D28 7.2:1.S 3.1:1.4 0.7 3. 1. 3.4:2 3.9:1.1 3.1:l.3 1.0:0.3 0.4 0. 0.5 7 6.1:2. .1 1:1 .1 .2 7:0 0.9:0.3 3.0 5.8:0.3 3.8:0.4 .1 2.010 2.4:0.2 :0.1 l 0.7 5:0. 0. Mean the mean (: SE) of three bulls. is Each value a was). All" i? t. 79 NN.N NN.NNN N4 INN .N.z NNN.N om.N HN.NN N zoHfimu :mcu m: ovxm wcwusv «new ocououmoumou pom <>oz< mo xymEE§m--.o~ mamhmucH mafiamsmm .Hw_u:ouvui.o~ ofinmb 81 animals were 3.8:O.4, 2.4:0.2 and 0.5:0.1 ng/ml for 6-, 4— and 2-month— old bulls, respectively. Testosterone in 6-month-old bulls treated with GnRH was significantly greater (P<0.05) than that of saline- treated bulls of the same age (Figure 8). This significant difference was not observed in 2- or 4-month-old bulls (Figure 8). . into the serum I expected to find an acute testosterone release /after GnRH treatment of 6-month-old bulls. However, this pattern of response was not observed (Figure 8). Considering only the first two injections of GnRH on day l in 6—month-old bulls, I observed a trend toward this pattern (Figure 9). However, the difference between GnRH and saline was not significant (P>0.05). Infrequent sampling between 0.5 and 3 hours may account for our failure to observe a sharp release of testosterone. The data in Table 19 showed marked variation in serum testosterone in AM and PM samples. Average serum testosterone of all samples taken in the AM (2.2:O.2 ng/ml) was significantly (P<0.05) lower than that in PM samples (2.8:O.2 ng/ml). However this AM vs PM difference (P<0.05) in average testosterone occurred in the saline-treated group (1.7:0.2 and 2.7:O.2, respectively), and not in the GnRH-treated bulls (2.7:0.3 vs 3.0:O.3 ng/ml). Lack of a significant difference between AM and PM serum testosterone, in GnRH—treated animals was probably due to the fact that the LH and testosterone release patterns were regularly controlled with exogenous GnRH. Testosterone levels in GnRH-treated animals slightly increased (P>0.05) during the treatment (2.6, 2.7, 2.8, 3.0 and 3.1 ng/ml for day 1, 7, 14, 24 and 28, respectively), whereas saline-treated animals did not change (1.9, 2.5, 2.7, 1.8 and 2.3 ng/ml for day 1, 7, 14, 24 and 28, respectively). 82 Figure 8.--Serum testosterone at O, 0.5 and 3 hours after injection (im) of saline or GnRH into bulls at 2, 4 or 6 months of age. Each point represents the mean : standard error of thirty observations, ten in each bull. Values at 0 hour were averages of serum testosterone in blood collected prior to each repeated injection of GnRH. Serum testosterone (ng/ml) 0' A 01 N 83 6 mo; GnRH ~ ~~~ ~ \ L \ \ \ Q ~ ~ ~ ~6 mo; Saline l 4mo;$aline ’T g‘ 55.- -' “’ ”ImoanRH v i’ 2mo; GnRH 2'5 £\§ 4 Q—-—- r—— ——— — ---——-B i L 2 mo;Sallne J o 0.5 3-0 Hours relative to the injection 1. {a 84 .ann game an ace .NCOnum>nomno ooncp mo nonno wnmucmum n came mucmmmnmon panom comm .owm mo mancoe e um manna oucn zmcu no mannmm mo nEnO m:0nuoomcn noumm mnso: m no m.o .o um H25 no 2