Endocrinological Assessment of Toxic Effects on the Male Reproductive System in Rats Treated with 5-Fluorouracil for 2 Or 4 Weeks

The Journal of Toxicological Sciences, 49 Vol.27, No.1, 49-56, 2002

ENDOCRINOLOGICAL ASSESSMENT OF TOXIC EFFECTS ON THE MALE REPRODUCTIVE SYSTEM IN RATS TREATED WITH 5-FLUOROURACIL FOR 2 OR 4 WEEKS

Setsuko TAKIZAWA and Ikuo HORII

Department of Preclinical Science, Nippon Roche K. K., Research Center, 200 Kajiwara, Kamakura, Kanagawa 247-8530, Japan

(Received October 25, 2001; Accepted December 11, 2001)

ABSTRACT — Endocrinological assessment of male reproductive toxicity was carried out in SD-Slc male rats treated with 5-FU (0, 20, 30 mg/kg/day) orally for 2-week or 4-week term. Serum hormone levels including GnRH, FSH, LH, prolactin, total and free testosterone, inhibin B, pro-alpha C, and activin A were determined as well as histopathological examination of the reproductive organs. The 5-FU treated groups showed histopathological changes in the testis such as degeneration of seminifer- ous epithelium. An obvious decrease in serum testosterone level was observed with a reduced organ weight of the seminal vesicle and prostate. However, no significant changes were noted in serum LH or FSH levels, nor in the morphological examination of the Leydig cells. Decreased serum levels were noted in activin A and prolactin. An increased serum level was noted in GnRH and pro-alpha C whose synthesis is regulated by FSH. Serum inhibin B levels showed a tendency toward decreasing with morphological change (vac- uolation) in Sertoli cells. These results indicated that male reproductive toxicity induced by 5-FU would be augmented by decreased serum prolactin and testosterone levels as well as a decreased func- tion of Sertoli cell, in addition to the direct cytotoxic effects on germ cells. It is suggested that these endocrinological changes related to male reproductive toxicities can be detected even in the 2-week- treated study.

KEY WORDS: 5-FU,Rat,Testis,Testosterone,Prolactin,InhibinB

INTRODUCTION toxicity of germ cells during spermatogenesis through antimetabolic activity of 5-FU mediated by an inhibi- Themalereproductivetoxicityof5-fluorouracil tion of thymidylate synthetase and an erroneous incor- (5-FU; antimetabolites) has been investigated in exper- porationintoRNAandDNA(O’Dwyeret al., 1987; imental animals (D’Souza and Narayana, 2001; Horii Pinedo and Peters, 1988; Schuetz et al., 1984; et al., 1985; Miyazaki et al., 1974; Russell and Russell, Waxman et al., 1990). 1991), and some morphological changes such as In addition to the morphological assessment of defects of spermatogenesis, atrophy of seminiferous male reproductive organs such as the testis and seminal epithelium, and decreased organ weight of the male vesicle, endocrinological evaluation gives useful infor- reproductive organs (seminal vesicle and prostate) mation for clarifying the toxicological mechanism were demonstrated. It is known that male reproductive through an endocrinological axis. Male reproductive toxicity of 5-FU is primarily caused by the direct cyto- toxicities would often be managed by the feedback

Correspondence: Setsuko TAKIZAWA

Vol. 27 No. 1 50

S. TAKIZAWA et al. mechanism of toxicologically responsible hormones in in the 2-week study. All rats in the 2- and 4-week tox- terms of toxicological intensity and property. Although icity studies were killed at 10 weeks old. The 0-4W endocrinological assessment of 5-FU was performed group was used as control in both studies. in female rat (Kuo et al., 1975), male reproductive tox- icity has not been evaluated endocrinologically. Dose selection rationale In this study, endocrinological assessment was In a 4-week repeated oral toxicity study of 5-FU carried out in rats treated with a low dose of 5-FU for 4 in Wistar rats, mortality was observed at 42 mg/kg/day weeks or low and high doses of 5-FU for 2 weeks, and and above, and an inhibition of body weight gain and changes in endocrinological parameters were clarified. hematological changes were noted at 30 mg/kg/day A possible toxicological mechanism through the endo- and above (Miyazaki et al., 1974). A 5-week repeated crinological axis was also investigated, as well as a oral toxicity study of 5-FU in SD rats showed a mortal- correlation of dose and treatment period. ity at 40 mg/kg/day (Horii et al., 1985). An inhibited body weight gain, hematological changes, and a MATERIALS AND METHODS decreased organ weight of reproductive organs were noted at 20 mg/kg/day. Based on these results, 20 mg/ Animals kg/day in the 4-week study and 20 or 30 mg/kg/day in SD-Slc male rats (12 males at 5 weeks old for the the 2-week study were chosen as the dose levels that 4-week study and 12 males at 7 weeks old for the 2- must induce some reproductive toxicity without severe week study) were obtained from Nippon SLC Corpora- inhibition of body weight gain and hematological tion (Shizuoka, Japan). After 1 week of acclimatiza- changes. tion, 10 healthy animals (5 rats/group) were selected for the experiment. They were housed individually in Experimental testing design metal cages in a room with a temperature of 22 ± 2°C, On the day next to final dosing (13:00-15:00), a relative humidity of 55 ± 10%, and a 12-hr light/dark animals were euthanised by carbon dioxide, and blood cycle (light: 7:00-19:00). Standard diet CRF-1 samples were collected from the heart. Serum samples (Charles River Japan, Ltd.) and tap water were given were stored at –40°C until endocrinological analysis. ad libitum. The animals were classed according to their Male reproductive organs were removed, weighed, and body weights and allocated randomly to groups using a fixed in Bouin’s solution (testes and epididymides) or computerized randomization program. 15% buffered formalin (seminal vesicle and prostate). HE staining specimens were histopathologically evalu- Test article and dosing ated by microscopy. 5-Fluorouracil (5-FU) was provided by Nippon RocheK.K.5-FUwasdissolvedindistilledwater,and Measurement of endocrinological parameters administered to male rats by oral intubation for 28 or Serum follicle stimulating hormone (FSH) and 14 days at a dose level of 0, 20, or 30 mg/kg/day with luteinizing hormone (LH) levels were determined by an administration volume of 10 ml/kg body weight. radio immunoassay (RIA) kit supplied from NHPP, Volumes for administration were calculated based on NIDDK (Bethesda, MD, USA). Prolactin (Rat prolac- the most recent body weight. tin (rPRL) [125I] assay system, Amersham Life Science, Buckinghamshire, UK), total testosterone (DPC total Animal experiment - dosing design testosterone kit, Diagnostic Products Corporation, Los Two-week and 4-week treatment periods were Angeles, USA), and free testosterone (DPC free tes- chosen to investigate endocrinological properties and tosterone kit, Diagnostic Products Corporation) levels if the changes observed in a 4-week study can be were analyzed by RIA. Gonadotrophin Releasing Hor- detected even in a 2-week study that is needed essen- mone (GnRH)(Endokit Red TM LHRH, CYT Immune tially to support human clinical trials. In a 4-week Sciences, Inc., MD, USA), inhibin B (Inhibin B assay repeated toxicity study, animals were treated with 5- kit, Serotec, Oxford, UK), Activin A (Activin-A assay FU at a dose level of 0 (0-4W group) or 20 mg/kg/day kit, Serotec), and pro-alpha C (Inhibin pro-alpha C (20-4W group). In a 2-week repeated toxicity study, assay kit, Serotec) were measured by enzyme-linked dose levels of 20 (20-2W group) and 30 mg/kg/day immunosorbent assay (ELISA). All analyses were per- (30-2W group) were chosen. The age at the start of formed in duplicate. dosing was 6 weeks in the 4-week study, and 8 weeks

Vol. 27 No. 1 51

Endocrinological assessment of male reproductive toxicity induced by 5-FU.

Statistical analysis B level. Two rats which showed vacuolation of Sertoli Differences between the control and drug-treated cells had the lowest level of Inhibin B. Activin A was groups were evaluated by Student’s t-test (Gad and distinctly decreased in all the drug-treated groups (Fig. Weil, 1982). p<0.05 was considered statistically sig- 2). nificant. DISCUSSION RESULTS In this study, histopathological examination Body weight revealed degeneration of seminiferous epithelium and The 20-4W group showed an inhibition of body vacuolation of Sertoli cells in the testis, and desqua- weight gain, and the 20-2W and 30-2W groups demon- mated cell debris in the ducts of epididymis in rats strated a decrease of body weight during the experi- given 5-FU. 5-FU caused no histopathological changes mental period. in the seminal vesicle or prostate in spite of a decreased relative organ weight. Similar lesions in the male Organ weight of reproductive system reproductive system, referring to reduced organ Regarding absolute organ weight, 5-FU caused a weight, defects of spermatogenesis, and so on, were significant decrease in weight of the testes, epidid- noted in previous studies on 5-FU (Miyazaki et al., ymides, seminal vesicle, and prostate in correspon- 1974; Horii et al., 1985; Russell and Russell, 1991; dence to a decrease in body weight. On the other hand, D’Souza and Narayana, 2001). abrupt body weight loss led to feigned results in rela- It has been known that the pharmacological tive organ weight: the testis and epididymis. In con- effects of 5-FU were mediated through the following trast, the relative organ weights of the seminal vesicle mechanisms for cytotoxicity: (1) Inhibition of and prostate were decreased in the drug-treated groups thymidylate synthetase (mutilation of deoxyuridylic (Fig. 1). acid and catalysis in the synthesis of thymidylic acid) resulting in a decrease of thymidine and inhibition of Histopathological examination of reproductive sys- DNA synthesis (Waxman et al., 1990); (2) Direct tem incorporation into nuclear RNA, resulting in process- Histopathological examination revealed degener- ing errors during formation of cytoplasmic RNA ation of seminiferous epithelium in testis and desqua- (Pinendo and Peters, 1988) and (3) Incorporation of mated cell debris in ducts of epididymis in the drug- residues into DNA, resulting in DNA damage treated groups. While vacuolation of Sertoli cells was (O’Dwyer et al., 1987; Schuetz et al., 1984). These observed in two animals of the 20-2W group, no sig- effects of 5-FU on DNA and RNA suggest a lead to nificant changes were noted in Leydig cells. No histo- toxicological changes (such as myelosuppression and pathological changes were noted in seminal vesicle gastrointestinal toxicity) in germ cells through disor- and prostate in spite of a decreased relative organ ders in proliferation of cells and expression of genes, weight (Table 1). followed by lesions in reproductive organs. Cells in S phase are most sensitive to the 5-FU toxicity (Elstein et Endocrinological examination al., 1993). As proliferation of germ cells is active in the The drug-treated groups showed no significant testicular organ, numerous germ cells pass the S-phase, changes in serum FSH and LH levels compared with indicating severer adverse effects on the testis by treat- the control group. The 20-2W group demonstrated a ment with 5-FU through the direct cytotoxic mecha- statistically significant increase of serum GnRH level, nism. and the other drug-treated groups also showed a trend Both the present 2- and 4-week repeated oral tox- toward increase. Serum levels of prolactin were icity studies showed more remarkable reduced organ decreased in all the drug-treated groups. Testosterone weight of the seminal vesicle and prostate rather than (total and free) was apparently decreased in the 20-2W that of the testis. In organ weight, a lesion in the testis and 30-2W groups without significant changes of the accompanied by histopathological changes through serum LH level. Pro-alpha C was increased in the 20- direct cytotoxic effects of 5-FU was not severer than 4W group, and a trend toward increase was shown in that in the seminal vesicle and prostate without any his- the other drug-treated groups. The drug-treated groups topathological changes. As the growth of seminal ves- showed a trend toward a decrease of the serum inhibin icle and prostate is regulated by androgen (testosterone

Vol. 27 No. 1 52

S. TAKIZAWA et al. and dihydrotestosterone (DHT)), decreased serum tes- 1998; Steiner et al., 1982). Pulsatile secretion of GnRH tosterone level could be involved in severely reduced is involved in pulsatile secretions such as those of FSH organ weight of the seminal vesicle and prostate. and LH (Weiss et al., 1993). Because serum hormone While no significant changes were observed in levels secreted from the pituitary are largely variable serum FSH and LH levels, an increased serum GnRH by pulsatile secretion, it would be difficult to detect level was noted in the drug-treated groups. Pro-alpha slight changes induced by the drug treatment. Serum C, whose mRNA level is regulated by FSH (Kaipia et FSH and LH levels in which no changes were exhib- al., 1994), showed a trend toward increase in the drug- ited in the present study might be increased substan- treated groups. It has been documented that secretion tially, and functionally corresponded to increased of glycoprotein hormones such as FSH and LH synthe- serum GnRH and pro-alpha C levels. sized in the pituitary shows pulsatile secretion Inhibin is a heterodimer consisting of an alpha (Bonavera et al., 1997; Dong et al., 1993, 1994; Ellis subunit and a beta subunit. The major inhibin in male and Desjardins, 1982; Ping et al., 1995; Refojo et al., is inhibin B that includes an alpha subunit (pro-alpha

Fig. 1. Relative organ weight of male reproductive organs in rats treated with 5-FU. Data are expressed as the mean ± S.E. from five animals per group. *: Significantly different from the control group (p<0.05). **: Significantly different from the control group (p<0.01). Control: 0-4W group.

Vol. 27 No. 1 53

Endocrinological assessment of male reproductive toxicity induced by 5-FU.

C) and beta B subunit and it is synthesized in Sertoli groups without significant changes of serum LH level. cells. Because the serum inhibin A level is quite low in No histopathological changes were noted in Leydig male animals (Illingworth et al., 1996), it was not cells where testosterone is synthesized. Serum activin determined in the present study. While the synthesis of A secreted from Leydig cells was also decreased in the pro-alpha C is stimulated by FSH, that of the beta B- drug-treated group, suggesting some functional lesion subunit is related to protein kinase C and/or the cellular in Leydig cells without morphological changes. Pro- calcium level (Klaij et al., 1992). Although the serum lactin that affects each stage of spermatogenesis inhibin B level was decreased in the drug-treated (Hondo et al., 1995) and accelerated the synthesis of group, the alpha subunit (serum pro-alpha C level) testosterone by LH in Leydig cells (Bartke, 1971) was showed a trend toward increase, suggesting that a decreased in the drug-treated groups. Decreased serum decreased synthesis of beta B-subunit could be prolactin level could cause a decrease of serum testos- involved in the decreased serum inhibin B level. terone level. Decreased serum levels of testosterone In the present study, histopathological examina- are obvious and show not only a possible decreased tion revealed a lesion (vacuolation) of Sertoli cells in synthesis of testosterone but also that other mecha- two animals of the 20-2W group. These animals also nisms may be involved in the decreased serum testos- showed the lowest value in serum inhibin B level. It is terone level. It was reported that 5-FU (Rat; 24 mg/kg/ known that lesion of Sertoli cells reflects a decreased day; 5 days) led to a drastic fall of plasma testosterone ratio of inhibin/FSH in humans (Tsatoulis et al., 1990). levels with some changes in P450 activity related to A trend toward decreasing serum inhibin B levels (no testosterone metabolism (Stupans et al., 1995). Such change in serum FSH level) was noted in some drug- changes of enzyme activity in the liver may lead to a treated animals without histopathological changes in promotion of testosterone metabolism, followed by a Sertoli cells, suggesting some functional lesion in Ser- distinct decrease of the serum testosterone level. toli cells. The endocrinological changes including Prolactin that affects each stage of spermatogen- decreased levels of testosterone, prolactin, activin and esis and accelerates the synthesis of testosterone by LH inhibin B, and increased levels of pro-alpha C and was decreased in the drug-treated groups. GnRHcanbedetectedevenat20mg/kg/dayafter2- In serum testosterone (total and free), whose syn- week treatment. A 2-week study showed severer thesis and secretion are regulated by LH, those levels results than a 4-week study even at the same dose. were distinctly decreased in the 20-2W and 30-2W Older age at the start of dosing leads to severer toxicity

Table 1. Histopathological examination of male reproductive organs in rats treated with 5-FU. Group Control (0-4W) 20-4W 20-2W 30-2W Number of animals examined 5555 Grade 1 2 3 1 2 3 1 2 3 1 2 3 Testis Degeneration of seminiferous 0 0 0 1 0 0 3 1 0 3 0 0 epithelium Exfoliation of seminiferous 0 0 0 0 0 0 1 0 0 0 0 0 epithelium Vacuolation of Sertoli cells 0 0 0 0 0 0 2 0 0 0 0 0 Formation of multinucleated 0 0 0 0 0 0 2 0 0 0 0 0 giant cell in seminiferous tubule Epididymis Desquamated cell debris in 0 0 0 1 0 0 0 0 2 3 0 0 ducts Decrease of sperms in ducts 0 0 0 0 0 0 0 0 2 0 0 0

Grade 1: Slight, Grade 2: Moderate, Grade 3: Severe.

Vol. 27 No. 1 54

S. TAKIZAWA et al. in 1,3-dinitrobenzene, and indicates that one of the rea- REFERENCES sons is a different ability of drug metabolism (Brown et al., 1994). In the present study, the age at the start of Bartke, A. (1971): Effects of prolactin on spermatoge- dosing was 6 weeks in the 4-week study, and 8 weeks nesis in hypophysectomized mice. J. Endocr., in the 2-week study. Older age at the start of dosing in 49, 311-316. the 2-week study may be involved in severer effects. Bonavera, J.J., Swerdloff, R.S., Leung, A., Lue, Y.H., In conclusion, it was indicated that male repro- Baravarian, S., Superlano, L., Sinha-Hikim, ductive toxicity induced by 5-FU would be augmented A.P. and Wang, C. (1997): In the male Brown- by decreased serum prolactin and testosterone levels as Norway (BN) male rat, reproductive aging is well as a decreased function of Sertoli cells, in addition associated with decreased LH-pulse amplitude to the direct cytotoxic effects on germ cells. It is sug- and area. J. Androl., 18, 359-365. gested that these endocrinological changes related to Brown, C.D., Forman, C.L., McEuen, S.F. and Miller, male reproductive toxicities can be detected even in M.G. (1994): Metabolism and testicular toxic- the 2-week-treated study. ity of 1,3-dinitrobenzene in rats of different ages. Fundam. Appl. Toxicol., 23, 439-446. ACKNOWLEDGMENT Dong,Q.,Li,B.,Rintala,H.,Blair,S.,Spaliviero,J. and Handelsman, D.J. (1993): LH pulsatility, The present study was carried out as a part of a biopotency and clearance during undernutrition testicular toxicity study of 5-FU (Inomata et al., 2000) in orchidectomized mature rats. Am. J. Phys- in order to assess the endocrinological effects. We iol., 265, E304-E313. thank Dr. H. Matsumoto and A. Inomata for supplying Dong, Q., Bergendahl, M., Huhtaniemi, I. and serum samples from a testicular toxicity study of 5-FU. Handelsman, D.J. (1994): Effect of undernutri-

Fig. 2. Serum GnRH, FSH, LH, prolactin, total testosterone, free testosterone, inhibin B, pro-alpha C, and activin A lev- els after treatment of 5-FU. Data are expressed as the mean ± SE from five animals per group. *: Significantly dif- ferent from the control group (p<0.05). **: Significantly different from the control group (p<0.01).

Vol. 27 No. 1 55

Endocrinological assessment of male reproductive toxicity induced by 5-FU.

tion on pulsatile luteininzing hormone (LH) tive inhibin. Mol. Cell. Endocrinol., 85,237- secretion in castrate and intact male rats using 246. an ultrasensitive immunofluorometric LH assay. Kuo, E.Y., Esber, H.J., Taylor, D.J. and Bogden, A.E. Endocrinology, 135, 745-750. (1975): Effects of cancer chemotherapeutic D’Souza, U.J. and Narayana, K. (2001): Induction of agents on endocrine organs and serum levels of seminiferous tubular atrophy by single dose of estrogens, progesterone, prolactin, and luteiniz- 5-fluorouracil (5-FU) in Wistar rats. Indian J. ing hormone. Cancer Res., 35, 1975-1980. Physiol. Pharmacol., 45, 87-94. Miyazaki, H., Imamura, S., Koyama, K., Hara, T., Ellis, G.B. and Desjardins, C. (1982): Male rats Nishikawa, S., Shiramizu, K., Ohguro, Y. and secrete luteinizing hormone and testosterone Shimizu, M. (1974): Safety evaluation of oral 5- episodically. Endocrinology, 110, 1618-1627. fluorouracil (acute, subacute, chronic toxicity Elstein, K.H., Zucker, R.M., Shuey, D.L., Lau, C., and teratological studies). Clinical Report, 8, Chernoff, N. and Rogers, J.M. (1993): Utility of 2603-2640. the murine erythroleukemic cell (MELC) in O’Dwyer, P.J., King, S.A., Hoth, D.F. and Leyland- assessing mechanisms of action of DNA-active Jones, B. (1987): Role of thymidine in biochem- developmental toxicants: Application to 5-Fluo- ical modulation: A review. Cancer Res., 47, rouracil. Teratology, 48, 75-87. 3911-3919. Gad, S. and Weil, C.S. (1982): Statistics for toxicolo- Pinedo, H.M. and Peters, G.F. (1988): Fluorouracil: gist. In Principles and Methods of Toxicology Biochemistryandpharmacology.J.Clin. (Wallance, H.A., ed.), pp.273-320, Raven Press, Oncol., 6, 1653-1664. New York. Ping, L., Mahesh, V.B. and Brann, D.W. (1995): Hondo, E., Kurohmaru, M., Sakai, S., Ogawa, K. and Effect of NMDA and non-NMDA receptor Hayashi, Y. (1995): Prolactin receptor expres- antagonists on pulsatile luteinizing hormone sion in rat spermatogenic cells. Biol. Reprod., secretion in the adult male rat. Neuroendocrinol- 52, 1284-1290. ogy, 61, 226-234. Horii, I., Yagishita, M., Okamoto, S., Mori, N. and Refojo, D., Arias, P., Moguilevsky, J.A. and Feleder, Udaka, K. (1985): Toxicity study of doxifluri- C. (1998): Effect of bacterial endotoxin on in dine (4) -comparative 5 weeks oral toxicity vivo pulsatile gonadotropin secretion in adult studywithFT-207and5-FUinrats-.Jpn.Phar- macol. Ther., 13 (suppl. 2), 293-335. male rats. Neuroendocrinology, 67, 275-281. Inomata, A., Matsumoto, H. and Horii, I. (2000): Col- Russell, L.D. and Russell, J.A. (1991): Short-term laborative work to evaluate toxicity on male morphological response of the rat testis to reproductive organs by repeated dose studies in administration of five chemotherapeutic agents. rats 18) comparative 4 and 2 weeks oral Am.J.Anat.,192, 142-168. repeated dosing studies on male reproductive Schuetz, J.D., Wallace, H.J. and Diasio, R.B. (1984): organs in rats treated with 5-fluorouracil. J. 5-Fluorouracil incorporation into DNA of CF-1 Toxicol. Sci., 25, Special Issue, 179-186. mouse bone marrow cells as a possible mecha- Illingworth, P.J., Groome, N.P., Byrd, W., Rainey, nism of toxicity. Cancer Res., 44, 1358-1363. W.E., McNeilly, A.S., Mather, J.P. and Steiner, R.A., Bremner, W.J. and Clifton, D.K. (1982): Bremner, W.J. (1996): Inhibin-B: A likely can- Regulation of luteinizing hormone pulse fre- didate for the physiologically important form of quency and amplitude by testosterone in the inhibin in men. J. Clin. Endocrinol. Metab., 81, adult male rat. Endocrinology, 111, 2055-2061. 1321-1325. Stupans, I., Richards, D.A. and McClure, M.T. (1995): Kaipia, A., Penttila, T.-L. and Toppari, J. (1994): Fol- Effects of 5-fluorouracil treatment on rat liver licle-stimulating hormone regulation of inhibin microsomal enzymes. Xenobiotica, 25,1-8. alpha-subunit mRNA in staged rat seminiferous Tsatsoulis, A., Shalet, S.M., Moriis, I.D. and de tubules. Eur. J. Endocrinol., 131, 323-329. Kretser, D.M. (1990): Immunoactive inhibin as Klaij, I.A., Timmerman, M.A., Block, L.J., a marker of Sertoli cell function following cyto- Grootegoed, J.A. and de Jong, F.H. (1992): toxic damage to the human testis. Horm. Res., Regulation of inhibin beta B-subunit mRNA 34, 254-259. expression in rat Sertoli cells: Consequences for Waxman, S., Scher, B.M., Hellinger, N. and Scher, W. the production of bioactive and immunoreac- (1990): Combination cytotoxic-differentiation

Vol. 27 No. 1 56

S. TAKIZAWA et al.

therapy of mouse erythroleukemia cells with 5- fluorouracil and hexamethylene bisacetamide. Cancer Res., 50, 3878-3887. Weiss, J., Crowely, W.F., Jr., Halvorson, L.M. and Jameson, J.L. (1993): Perifusion of rat pituitary cells with gonadotropin-releasing hormone, activin, and inhibin reveals distinct effects on gonadotropin gene expression and secretion. Endocrinology, 132, 2307-2311.

Vol. 27 No. 1