Vol.

FERTILITY AND STERILITY

55, No.2, February 1991

Printed on acid-free paper in U.S.A.

Copyright © 1991 The American Fertility Society

Effects of chronic bromocriptine-induced hypoprolactinemia on plasma testosterone responses to human chorionic gonadotropin stimulation in normal men

Fumimaro Oseko, M.D.*t Akinobu Nakano, M.D.* Keiko Morikawa, M.D.*

Jiro Endo, M.D.:\: Ataru Taniguchi, M.D.§ Tuguru Usui, M.D.II

Shimane Medical University, Shimane, Japan

To study the role played by normal levels of plasma prolactin (PRL) in the secretion oftestosterone (T) in the testes, we induced hypoprolactinemia with a daily dose of 5 mg bromocriptine administered orally in five normal men 20 to 35 years of age for 8 weeks. The basal PRL, T, luteinizing hormone, follicle-stimulating hormone, and maximum responses of plasma T to human chorionic gonadotropin (hCG) stimulation were measured every 2 weeks. Basal levels of plasma T were reduced in the 1st 2-week-Iong period of hypoprolactinemia. In the 4-week-Iong period of hypoprolactinemia, the maximal response of plasma T to hCG stimulation was significantly reduced. The findings suggest that normal levels of plasma PRL may play an important role in the secretion of T in the human testes in vivo. Fertil Steril55:355, 1991

The hypothesis that prolactin (PRL) influences the testes is controversial because, on the one hand, for example, Rubin et aU and Faglia et al. 2 suggested that PRL may have a stimulatory influence on the testes; on the other hand Besser and Thorner3 indicated that long-term hyperprolactinemia has inhibitory effects on testicular functions. Normal levels of plasma PRL are considered to potentiate the effect of luteinizing hormone (LH) on Leydig cells. 4 Bromocriptine-induced hypoprolactinemia in normal men for 13 days has no effects on plasma levels of testosterone (T) and on T responses to human chorionic gonadotropin (hCG).5

Considering these results, this study was performed to determine the roles of normal levels of plasma PRL in the testes of men by observing the effects of long-term (~ weeks) bromocriptine-induced hypoprolactinemia on plasma T and its responses to hCG.

MATERIALS AND METHODS

cine, Shimane Medical University. t Reprint requests: Fumimaro Oseko, M.D., First Division of Internal Medicine, Department of Medicine, Shimane Medical University, Shimane 693, Japan. :\: Department of Clinical Laboratories, Department of Medicine, Shimane Medical University. § Present address: Department of Internal Medicine, Kansaidenryoku Hospital, Osaka Japan. II Department of Urology, Hiroshima University School of Medicine, Hiroshima, Japan.

Five normal men, 20 to 35 years of age, volunteered for this study after thorough explanation of the experimental protocol. Oral bromocriptine (5 mg) was administered daily to each subject for 8 weeks to induce hypoprolactinemia. Human chorionic gonadotropin stimulation (hCG test) was performed before bromocriptine administration and at 2-week intervals up to 8 weeks. In the hCG test,6 4,000 units of hCG were administered intramuscularly for 3 consecutive days to each subject fasting overnight after 30 minutes bed rest at 8 to 9 A.M. Blood for measurement of T, PRL, LH, and follicle-stimulating hormone (FSH) was withdrawn from the antecubital vein one time in the morning before hCG injection for 4 days. Plasma was sepa-

Vol. 55, No.2, February 1991

Oseko et al.

Received May 7,1990; revised and accepted October 4,1990.

* First Division of Internal Medicine, Department of Medi-

Hypoprolactinemia's effects on androgen

355

Table 1 Basal Levels of Plasma T and Maximum Responses of Plasma T to RCG Stimulation in Each of Five Normal Men in a Control Period

Control

~j

Plasma T Subjects

Age

Basal

Peak

E

20j 10 0

1 2 3 4 5

I

20j 10 0

.,c

ng/mL

34 22 21 20 35

6.3 8.8 10.3 9.5 7.5

14.5 13.1 17.7 20.5 28.0

8.5 ± 1.6"

18.8 ± 5.9"

"Values are means ± SD.

BromocriPtine 5 mg daily orally

~

~

~

/[ii

~

::::::n6i

[J][]

~

~

---mi

::::::Iifii

..-----

..----

~

~

--fiiii

~ [ill]

--

IE!J

~

~

/ii;;

~ ~ ~ ~ 1TT4

Case 1

~ ~

Case 3

[]lij[]

I-

.~

a:'"

20~

10 0

20j 10 0

rated by centrifugation as soon as possible after each drawal and kept at -20°C until assayed. Plasma PRL, T, LH, and FSH were determined by radioimmunoassay.6 All samples from each subject were run in the same assay. Mean plasma concentrations of basal PRL of the five subjects before bromocriptine treatment were compared with those 2-week intervals after administration of the drug up to 8 weeks. The same comparison was performed relative to basal T and a maximum T referring to a highest value among the responses of plasma T in each hCG test. In this comparison, a basal T and a maximum T of each subject in the

~

Case 2

Iii

.5 y

.llL

ID

[jgJ

[EI]

~

[J][]

r-r-r---,

r-r-r---,

r-r-r---,

r-r-r---,

I 2 3 4

1 2 3 4

1 2 3 4

1 2 3 4

[]IT]

[]IT]

IJIT]

5.1 ,

6.3

Davs Mean ± SO

1100.01 0.0 L-..--

OOJ 3.1 * '

!

*

9.4 ---J

Figure 2 Responses of plasma T to hCG stimulation in each of five normal men before bromocriptine treatment (5 mg/d orally) and at weeks 2, 4, 6, and 8 after it. The maximum T response of each subject in the control period was defined as 100%, which is shown in a rectangle. The maximum T at an interval of 2 weeks during the 8-week-Iong treatment was expressed as a ratio to the respective control value in each rectangle. *, P < 0.05 (compared with a value at 0 week as a control).

control period were expressed as 100% and those after the drug administration as a ratio to the respective control value. Analysis of variance and paired Wilcoxon's test were used for statistical analysis.

RESULTS

T Weeks

Figure 1 Percent changes of mean (±SD) basal T of 5 normal men obtained at an interval of 2 weeks during 8-week-Iong treatment of bromocriptine (5 mg/d orally). The mean of each subject in the control period was defined as 100%. *, P < 0.05 (compared with a value of 0 week as a control). 356

Oseko et al.

Hypoprolactinemia's effects on androgen

A daily dose of 5 mg bromocriptine reduced significantly (P < 0.05) the mean level of plasma PRL of the five subjects in the control period (15.6 ± 4.6 [SD] ng/mL) to 8.5 ± 1.0, which was the mean level at 2 weeks after initiation of the bromocriptine treatment. The mean values at 4, 6, and 8 weeks were 7.6 ± 1.4,8.6 ± 0.8, and 7.2 ± 0.6, respectively. Each of those levels was significantly (P < 0.05) lower than the mean ofthe control. Mean basal levels of plasma LH and FSH during 8 weeks of bromocriptine administration did not change significantly compared with those in the control period. Basal plasma T and T maximum responses to hCG stimulation in the control period were shown in Table 1. The mean levels of basal plasma T in the control period decreased significantly (P < 0.05) at 4 weeks after bromocriptine treatment, and the Fertility and Sterility

tendency of significantly decreasing was recognized at 6 and 8 weeks (Fig. 1). A mean maximum response of plasma T to hCG stimulation in the control group began to decrease significantly (P < 0.05) at 2 weeks after bromocriptine administration, and this tendency (P < 0.05) lasted up to 8 weeks (Fig. 2).

nism may exist in man to explain the basis for our findings. In summary, the induction of a 4-week-long hypoprolactinemia by bromocriptine in five normal men reduced the secretion of T from the testes, as well as their T reserves. This finding may indicate a positive action by normal levels of plasma PRL on production of T in the human testes.

DISCUSSION

A daily dose of 5 mg bromocriptine administered orally to a group of five normal subjects resulted in a significant reduction (about 50%) ofthe basal plasma PRL level to that of the control level. This level was slightly lower than that (65%) obtained by Martikainen and Vihko. 5 Our findings that mean levels of basal plasma T decreased from 4 weeks after bromocriptine (Fig. 1) and the means of T maximum responses to hCG stimulation were reduced from 2 weeks after bromocriptine administration (Fig. 2) up to 8 weeks suggest that a decrease of plasma PRL by bromocriptine reduces secretion ofT from, and T reserves in, the testes. Although we were unable to exclude the effects of bromocriptine (a dopamine agonist) on the Leydig cells per se, our conclusion may be valid in view of the report by Vermes et al. 7 that dopamine has no effects on Leydig cells in rats. Furthermore, the plasma levels of LH and FSH did not change significantly throughout the experiment. In addition, Catt et al. B stated that bromocriptine-induced hypoprolactinemia decreased LH receptors in the testes of rodents and administration of PRL increased the numbers of LH receptors. The same mecha-

Vol. 55, No.2, February 1991

REFERENCES 1. Rubin RT, Goin PR, Lubin A, Poland RE, Pirke KM: Noc-

2.

3. 4.

5.

6.

7.

8.

turnal increase of plasma testosterone in man: relation to gonadotropin and prolactin. J Clin Endocrinol Metab 40: 1027,1975 Faglia G, Beck-Peccoz P, Travaglini P, Ambrosi B, Rondena M, Paracchi A, Spada A, Weber G, Bara R, Bouzin A: Functional studies in hyperprolactinemia states. In Prolactin and Human Reproduction, Edited by PG Crosignani, C Robyn. London, Academic Press, 1977, p 225 Besser GM, Thorner MO: Prolactin and gonadal function. Pathol BioI (Paris) 23:779, 1975 Bardin CW, Paulsen CA: The testes. In Textbook of Endocrinology, Edited by RH Williams. Philadelphia, W. B. Saunders Company, 1981, p 293 Martikainen H, Vihko R: hCG-stimulation of testicular steroidogenesis during induced hyper- and hypoprolactinemia in man. Clin Endocrinol (OxO 16:227,1982 Oseko F, Note S, Morikawa K, Endo J, Taniguchi A, Imura H: Influence of chronic hyperprolactinemia induced by sulpiride on the hypothalamo-pituitary-testicular axis in normal men. Fertil Steril44:106, 1985 Vermes I, Varszegi M, Toth EK, Telegdy G: Study of the effects of neurotransmitters on the hypothalamuspituitary-testis function in in vitro cell suspension system. Arch AndroI3:127, 1979 Catt KJ, Harwood JP, Clayton RN, Davies TF, Chan V, Katikeneni M, Nozu K, Dufau ML: Regulation of peptide hormone receptors and gonadal steroidogenesis. Recent Prog Horm Res 36:557, 1980

Oseko et al.

Hypoprolactinemia's effects on androgen

357