Department of Medicine Harbor-UCLA Medical Center Torrance, California ABSTRACT

Transcription

1 BOLOGY OF REPRODUCTON 37, (1987) Testosterone Selectively ncreases Serum Follicle-Stimulating Hormonal (FSH) But Not Luteinizing Hormone (LH) in Gonadotropin-Releasing Hormone Antagonist-Treated Male Rats: Evidence for Differential Regulation of LH and FSH Secretion1 S. BHASN, T. J. FELDER, and R. S. SWERDLOFF Department of Medicine Harbor-UCLA Medical Center Torrance, California ABSTRACT Both testosterone (T) and gonadotropin-releasing hormone (GnRH)-antagonist (GnRH-A) when given alone lower serum luteinizing hormone (LH) and follicle-stimulating hormone (FSH) in intact and castrated rats. However, when graded doses of testosterone enanthate (T.E.) were given to GnRH-A -treated intact male rats, a paradoxical dose-dependent increase in serum FSH occurred; whereas serum LH remained suppressed. This surprising finding led us to ask whether the paradoxical increase in serum FSH in GnRH-A -suppressed animals was a direct stimulatory effect of Ton the hypothalamic-pituitary axis or the result of a T effect on a testicular regulator of FSH. To test these hypotheses, we treated adult male castrated rats with GnRH-A and graded doses of T.E. in both intact and castrated rats, serum LH remained undetectable in GnRH-A-treated rats with or without T.E. However, addition of T.E. to GnRH-A led to a dose-dependent increase in serum FSH in castrated animals as well, thus pointing against mediation by a selective testicular regulator of FSH. These data provide evidence that pituitary LH and FSH responses may be differentially regulated under certain conditions. When the action of GnRHis blocked (such as in GnRH-A -treated animals), T directly and selectively increases pituitary FSH secretion. NTRODUCTON The effects of steroid hormones on pituitary luteinizing hormone (LH) and follicle-stimulating hormone (FSH) synthesis and secretion are complex and involve interactions with the hypothalamic releasing hormone gonadotropin-releasing hormone (GnRH) and other regulatory neurotransmitters. n general, in intact adult males, testosterone (T) is known to cause a dose-dependent inhibition of LH and FSH secretion through effects at the hypothalamic and pituitary levels (Sherins and Loriaux, 1973; Swerdloff and Walsh, 1973, Stewart-Bentley et al., 1974; Santen, 1975; Plant et a!., 1978). Under certain circumstances, however, steroid hormones have been shown to stimulate gonadotropin release (Swerdloff and Walsh, 1973). The best example of such positive feedback is the stimulatory effect of Accepted January 26, Received November 10, This work is supported in part by the training Grant #2-T32-AM GnRH antagonist for these studies was provided by Marvin J. Karten, Ph.D. of the Contraceptive Development Branch of the NCHHD. 2Reprint requests. estrogen on LH release at proestrus in the female rat (Kalra and Kalra, 1983). The positive feedback effects of the gonadal steroids on gonadotropin secretion are much more pronounced in the female. n the course of our contraceptive studies with a GnRH antagonist (GnRH-A), we noted that T paradoxically increases serum FSH, but not LH, in GnRH-A-treated male rats, as well. Similar observations demonstrating a T-induced increase in FSH in the male rat were recently reported (Rhea et al., 1986). This paper extends these observations by demonstrating dose-dependence of this effect and also elucidates some potential mechanisms involved in a T-induced rise in serum FSH. Studies in ntact Animals MATERALS AND METHODS Adult male Wistar rats were divided into 6 groups of 6 animals each as follows: Group -vehicle alone; Group -GnRH-A (250 zg/day); Group -GnRH-A (250.zg/day) + testosterone enanthate (T.E.) 0.05 mg/day; Group V-GnRH-A (250 iig/ day) + T.E mg/day; Group V-GnRH-A (250 55

2 56 BHASN ET AL. jig/day) + T.E mg/day; Group V-GnRH-A (250 pg/day) + T.E mg/day. The GnRH-A used in these studies (Ac-D [2] Na!1, 4 Cl D Phe2, D Trp3, D Arg6, D Ala #{176})GnRH-HoAc was provided by the Contraceptive Development Branch of the NCHHD. The dose of the GnRH-A (250 pg/day) was shown in previous experiments to induce medical hypophysectomy in intact male rats. The animals were treated for 70 days. At the end of the treatment period, the animals were killed by decapitation, testes and accessory sex organs weighed, and sera frozen for hormone measurements. Studies in Castrated Animals To elucidate whether the T-induced paradoxical increase in FSH was the result of a T effect on a selective testicular regulator of FSH secretion or the result of a direct T effect on the hypothalamicpituitary axis, we treated adult castrated male Wistar rats as follows: Group -vehicle alone; Group - GnR-A (250 pg/day); Group 1l-GnRH-A + T.E. (0.05 mg/day); Group 1V-GnRH-A + T.E. (0.15 mg/day); Group V-GnRH-A + T.E. (0.50 mg/day); Group V-GnR=-A + T.E. (1.5 mg/day). Another group of castrated male rats was treated only with T.E. (0.5 mg/day). All animals were treated for 70 days. At the end of the treatment period, the animals were killed by decapitation, accessory sex organs weighed, and sera frozen for hormone measu rements. Hormone Measurements Serum Lii and FS= concentrations were measured by reagents supplied by NADDM. The minimal detectable concentration in the serum L assay was 0.05 ng/m! of rat LH RP-2 and for FSH was 0.75 ng/ml of rat FSH RP-2. Serum T concentrations were measured by modification of an iodinated T assay (Fudenburgh et a!., 1983). Modification included extraction of the serum samples with 10 volumes of a mixture of ethyl acetate and hexane (3:2) prior to radioimmunoassay. Extraction recoveries were consistently over 90%. The intra- and interassay coefficients of variation for the rat LH, FSH, and T assays have been characterized in our laboratory as follows: rlh ± 3.1 and ± 11%; rfsh ± 2.4 and ± 12%; T ± 3.8 and ± 8.3%. Statistical Analyses The data were averaged across animals within each treatment group. The group means were compared by analysis of variance using BMDP statistical software (Dixon, 1985). - Studies in ntact Animals RESULTS Testes and accessory organ weights. GnRH-A treatment markedly decreased testes weight to levels seen after hypophysectomy (p<0.0001), Addition of T.E, led to a dose-dependent increase in testes weights. However, even at the highest dose of.e., testes weights were still significantly less than the intact controls (Table 1). Prostate and seminal vesicle weights were markedly decreased by GnRH-A treatment alone. Graded doses of.e. led to a dose-dependent increase in prostate and seminal vesicle weights. Maximal increase in TAB.E 1. Effects of testosterone (T) in gonadotropin.releasing hormone.antagoniat (GnRH.A).treated intact mak rats. GroUp V V V Testes (g) Prostate (g) Seminal vcsk k (g) Luteini#.ing hormone (ng/m) Serum T (ng/rnl) , ,59 t 0.02k s 0003* * ab 0.89 t t 0.03k s 0.019* ,020* <005ac 0.65 ± O ± * <0,05ac 2.68 s i 0.09* ± s <005nc d 2.97 ± O.l ± <005ac ± 1.34k - Significantly different from Group, p - < Some values below the limit of detection (<0.05 ng/ml). - All values below the limit of detection (<0,05 ng/ml). d - Significantly different from Group, p #{149} <0.05. #{149} h animals were grouped as follows:. vehicle alone;. GnRH-A alone;. GnRH-A + testosterone enanthate (T.E., 0.05 ng/day); V. GnR- A + T.E. (0.15 mg/day); V. GnR#{149}A + T.E. (0.50 mg/day); V. GnRH-A + T.E. (1.50 mg/day). Data are means S SEM n.6 n each group.

3 TESTOSTERONE PARADOXCALLY NCREASES FSH c J irl, 4 u- 5+ * Group u. w - 2- U GnRH-A (250pq/doy) - TE (mg/doy) - nnn *4-5+ ifi V V Vi ± FG. 1. Effects of testosterone enanthate (T.E.) on serum folliclestimulating hormone (GnRH-A)-treated intact rats. Data are means ± SEM n=6 in each group. = Significantly different from Group p<o.0o5; + = significantly different from Group p< prostate and seminal vesicle weights was seen at the 0.50-mg dose. A further increase of the T.E. dose to 1.50 mg did not lead to a significant increase in prostate or seminal vesicle weights (Table 1). Serum T. Serum T concentrations were decreased to castrate levels by GnRH-A treatment. A progressive increase in the T.E. dose led to a dose-dependent increase in serum T concentrations (Table 1). Serum LH. Treatment with GnRH-A alone decreased serum L to below the limit of detection (<0.05 ng/ml, p<0.0001). Serum LH concentrations remained undetectable in all other groups treated with GnRH-A and T.E., regardless of the T.E. dose (Table 1). Serum FSH. GnRH-A treatment alone markedly decreased serum FSH concentrations (p<0.0001) (Fig. 1). Addition of graded doses of T.E. led to progressive increases in serum FSH concentrations. However, serum FSH concentrations, even at the highest dose of T.E., were significantly lower than non-gnr-a-treated controls. Group GnRH-A T.E. (mg/d) E 0 C U, U- Ui U, (250.ug/d) 10- nnnr11 U ll V V Vi - + ± FG. 2. Effects of testosterone enanthate (T.E.) on serum folliclestimulating hormone (GnRH-A)-treated castrated rats. Data are means ± SEM n=6 in each group. * = Significantly different from Group p< Studies in Castrated Animals Treatment with GnRH-A led to a marked decrease in serum L and FSH. However, when graded concentrations of T.E. were added to GnRH-A, there was a progressive increase in serum FSH (Fig. 2), whereas serum L remained suppressed below the limit of detection (Table 2). The addition of graded concentrations of T.E. to GnR-A also led to dose-dependent increases in serum T concentrations (Table 2). Prostate and seminal vesicle weights were markedly lower in castrated controls. The addition of T.E. led to a dose-dependent increase in prostate and seminal vesicle weights. Treatment with T.E. alone lowered serum LH to TABLE 2. Effects of testosterone (T) on gonadotropin-releasing hormone-antagonist (GnRH-A)-treated castrated male rats. Group V V V Prostate weight (g) ± ± ± 0.027k ± 0043t ± 0,0136b ± 0112b Seminal vesicle weight (g) ± ± ± 0033a ± 0026b ± 0054b ± 0019b Luteinizing hormone (ng/ml) 6.20 ± 0.81 <O.05 <o.osbc <o.o5bc <005bc <o.o5bc T 0.50 ± ± ± ± ± ± 1.34 ava Group, p = < b5 Group, p = < c = All values below the limit of detection. The animals were grouped as follows:. vehicle only; 11. GnRH-A alone;. GnRH-A mg testosterone enanthate (T.E.)day; V. GnRH-A mg T.E.; V. GnRH-A + T.E mg/day; V. GnRH-A + T.E mg/day. Data are means ± SEM n=6 in each group.

4 58 BHASN ET AL. below the limit of detection (<0.05 ng/ml, p<0.0001). Serum FSH concentrations in this group were significantly lower than in the corresponding vehicle-treated castrated controls (4.27 ± 0.13 ng/ml vs ± 4.04). DSCUSSON These data confirm previous observations that GnRH-A effectively suppresses serum LH and FSH in intact and castrated male rats (Asch et al., 1981; Heber et al., 1982; Kenningsberg et al., 1984; Weinbauer et al., 1984). These data demonstrate that in GnR-A-treated intact as well as castrated male rats, T exerts differential effects on serum LH and FSH. When graded doses of T.E. are added to GnRH-A, serum LH concentrations remain suppressed while serum FSH concentrations demonstrate a dosedependent paradoxical increase. There are several potential explanations for this T induced paradoxical increase in serum FSH. First, T could alter the secretion of a selective testicular regulator (inhibitor or stimulator) of FSH secretion. Alternatively, T could act directly on the hypothalamic-pituitary axis. The data from the castrated male rats point against mediation by a testicular regulator of FSH secretion and suggest a direct T effect on the hypothalamic-pituitary axis. One could postulate that pituitary LH and FSH secretion are under the influence of separate hypothalamic releasing hormones and that T differentially affects the hypothalamic release of the two releasing hormones. n opposition to this hypothesis are data demonstrating that T increases FSH in pituitary monolayers not exposed to any hypothalamic hormones (Steinberger and Chowdhury, 1974; 1977; Drouin and Labrie, 1976; Denef et al., 1980; Miller and Wu, 1981; Leveque and Grotjan, 1982). Further, in intact and castrated male rats, in the absence of GnRH-A, T leads only to a dose-dependent decrease in both serum LH and FSH (Swerdloff and Walsh, 1973). We therefore propose that the action of T on FSH in determined by whether or not the pituitary has been primed by GnR. Under conditions when the pituitary action of GnR is blocked by GnRH-A, T leads to an increase in FSH. This hypothesis is consistent with observations that T increases serum FSH in prepubertal animals (Naqui and Johnson, 1969) and that T directly increases FSH secretion and cell content of pituitary cells in culture (Steinberger and Chowdhury, 1974, 1977; Diouin and Labrie, 1976; Denef et al., 1980; Miller and Wu, 1981; Leveque and Grotjan, 1982). The stimulation of both FSH release and total FS- content of rat anterior pituitary cells by a number of gonadal steroids is well documented (Steinberger and Chowdhury, 1973; 1977 Drouin and Labrie, 1976; Denef et al., 1980; Miller and Wu, 1981; Leveque and Grotjan, 1982). Progesterone, estradiol-17 beta, and dihydrotestosterone have also been shown to stimulate FSH release and cell content of pituitary cells in culture (Miller and Wu, 1981; Leveque and Grotjan, 1982). n fact, these four steroids have similar effects, cause approximately the same degree of stimulation, and differ only slightly in their potency in stimulating basal FSH release and total FSH. Thus, it would appear that all four of these important gonadal steroids act on the pituitary to stimulate FSH secretion. n cultures, the increase in total FSH response (FSH released in the medium and the FSH content of the cell) to T suggests a stimulation of net FSH synthesis by this steroid. Five-alpha reduction of T does not appear to be critical in mediating its effects on FSH synthesis and release (Leveque and Grotjan, 1982). The significance of this T effect on FSH remains unclear. Rhea et al. (1986) have suggested that this increase in FSH may, at least in part, be responsible for the T effects on spermatogenesis. This seems unlikely because T is able to support spermatogenesis even in surgically hypophysectomized male rats, suggesting that the effects of T on spermatogenesis are exerted directly at the testicular level (Boccabella, 1963; Clermont and Harvey, 1967; Steinberger and Duckett, 1971). These data point out that pituitary L and FSH secretion may be differentially regulated under certain circumstances. Although LH and FSH responses are concordent under most circumstances (GnRH stimulation, castration, etc.), they may diverge under certain circumstances. When the action of GnR is blocked, such as in prepubertal animals or under GnRH-A treatment, T selectively increases FSH, but not L, secretion. REFERENCES Asch RH, Balamaceda JP, Eddy CA, Sider-Khodr TM, Coy DH, Schally AV, nhibition of post-castration rise of LH and FSH in female rhesus monkeys by administration of an LHRH inhibitory analog. Fertil Steril Boccabella AV, Reinitiation and restoration of spermatogenesis with testosterone proprionate and other hormones after a long term post-hypophysectomy regression period. Endocrinology

5 TESTOSTERONE PARADOXCALLY NCREASES FSH 59 72: Clermont V. Harvey SC, Effects of hormones on spermatogenesis in the rat. n: Wolstenholme GEW, O Connor M (eds.), Little, Brown & Co. Boston. CBA Found Colloq Endocrinol 16: Denef C, Hautekeete E, Dewals R, DeWof A, Differential control of LH and FSH secretion by androgens in rat pituitary cells in culture: functional diversity of sub-populations separated by unit gravity sedimentation. Endocrinology 106: Dixon WJ, BMDP Statistical Software. Berkeley, CA: University of California Press. Drouin J, Labrie F, Selective effects of androgens on LH and FSH release in anterior pituitary cells in culture. Endocrinology 98: Fudenburgh U, Zits WB, SoLos JF, Direct measurement of testosterone in a pediatric center with the use of a radioimmunoassay kit and unextracted serum. Clin Chem 29: Heber D, Dodson R, Swerdloff RS, Pituitary receptor site blockade by a GnRH antagonist in vivo. Science 216: Kaira SP, Kalra PS, Neural regulation of LH secretion in the rat. Endocr Rev 4: Kenningsberg D, Littman BA, Hodget GD, Medical hypophysectomy using a GnRH antagonist. Fertil Steril 42: Leveque NW, Grotjan HF Jr nteraction of progesterone with testosterone and dihydrotestosterone on FSH release by cultures of rat anterior pituitary cells. Bio Reprod 27: Miller L, Wu J, Estrogen regulation of FSH production in vitro species variation. Endocrinology 108: Naqui RH, Johnson DC, ncreased production of FSH in immature female rats by a single injection of androgen or estrogen. J Endocrinol 45:29-35 Plant TM, Hess DL, Hotchkiss J, Knobil E, Testosterone and the control of gonadotropin secretion in the male rhesus monkey. Endocrinology 103: RheaMA, Marshall GR, Weinbauer GF, Nieschlag E, Testosterone maintains pituitary and serum FSH and spermatogenesis in GnRHantagonist suppressed rats. J. Endocrinol Santen RJ, s aromatization of testosterone to estradiol required for inhibition of LH in mean. J Clin nvest 56: Sherins RJ, Loriaux DL, Studies on the role of sex steroids in the feedback control of FSH concentrations in men. J Clin Endocrinol Metab 36: Steinberger E, Chowdhury M, The effects of testosterone and estradiol on the basal and LRF-stimulated secretion of gonadotropins in the pituitary cell culture. Endocr Res Commun 1: Steinberger E, Chowdhury M, The effects of testosterone proprionate and estradiol benzoate on the in vitro synthesis of FSH. Biol Reprod 16: Steinberger E, Duckett GE, Hormonal control of mammalian spermatogenesis. Physiol Rev 51:1-22 Stewart-Bentley M, Odell W, Horton R, The feedback control of LH in normal adult men. J Clin EndocrinolMetab 38: Swerdloff RS, Walsh PC, Testosterone and estradiol suppression of LH and FSH in adult male rats: duration of treatment and combined treatment. Acts Endocrinol 73:11-21 Weinbauer GF, Surmann FJ, Akhtar FB, Shah FB, Vickery B, Nieschlag E, Reversible inhibition of testicular function by a GnRH antagonist in monkeys. Fertil Steril 42:906-14