0021-972X/78/4702-0447$02.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1978 by The Endocrine Society

Vol. 47, No. 2 Printed in U.S.A.

Effects of Prolactin and Prolactin Plus Luteinizing Hormone on Plasma Testosterone Levels in Normal Adult Men* ROBERT T. RUBIN, RUSSELL E. POLAND, IRVING SOBEL, BARBARA B. TOWER, AND WILLIAM D. ODELL Departments of Psychiatry, Pharmacology, and Medicine, UCLA School of Medicine, Harbor General Hospital Campus, Torrance, California 90509

V-

ABSTRACT. In an earlier study, normal adult men were shown to have increased plasma testosterone (T) levels over a several-hour period after haloperidol-induced increases in plasma PRL levels. The present study was designed both to replicate our first study and to examine the potential synergism of PRL and LH in influencing T levels on a short term basis in normal men. Eight volunteers received on 4 separate days an im injection of saline or 0.5 mg haloperidol at 1000 h and an iv injection of saline or 88 IU human LH (hLH) at 1100 h in a double blind randomized block design arranged to augment plasma levels of PRL, LH, and PRL and LH together on the different test days as well as to afford a saline control day. Only five of the eight subjects had prompt PRL responses to haloperidol equivalent to those of our earlier study. As the purpose of this study was to examine the effect of increased PRL on plasma T levels, these five subjects were used for the determination of

changes in plasma T. After haloperidol administration, their PRL levels rose an average of 19 ng/ml, to the high-normal range, and after the hLH infusions, their LH levels rose an average of 71 ng/ml. On the saline control day, mean T levels showed the normal diurnal decline. After 0.5 mg haloperidol, T levels were maintained for several hours, and after 88 IU hLH, T levels were increased for several hours. Increased PRL levels concomitant with hLH administration did not produce a T response greater than that caused by hLH alone. The results of this study replicate the effect of druginduced PRL augmentation on plasma T levels found in our earlier study, but they fail to demonstrate a synergistic effect of acutely increased PRL on LH-stimulated T secretion. PRL thus seems to be another pituitary hormone capable of increasing plasma T in adult men, but it clearly is a weaker stimulus than LH. (J Clin Endocrinol Metab 47: 447, 1978)

I

N A previous study (1) we reported that in normal adult men, plasma testosterone (T) levels increased in a dose-related manner after pharmacologically induced increases in plasma PRL levels. In that study, five healthy male volunteers received on different days an im injection of either normal saline, 0.25 mg haloperidol, or 0.05 mg haloperidol (a dopamine-blocking agent). After the administration of 0.25 mg and 0.05 mg haloperidol, average plasma PRL levels increased by 7 and 12 ng/ml, respectively; gonadotropin levels

were not affected. On the saline control day, average plasma T levels showed the normal diurnal decline, whereas after haloperidol, average plasma T levels were increased for several hours. PRL thus seemed to be another pituitary hormone capable of increasing plasma T in normal adult men. The present study was designed both to replicate part of our first study and to examine the potential synergism of PRL and gonadotropins on T secretion in men on a short term basis over several hours. As in the first study, haloperidol (0.5 mg im) was used to increase plasma PRL levels. In addition, human LH Received May 25, 1977. Address requests for reprints to: Dr. Robert T. Rubin, (hLH) was administered iv to increase the B-4 Neuroendocrine Laboratory, Harbor General Hospi- plasma levels of this gonadotropin. Serial tal, Torrance, California 90509. * This work was supported by NIMH Research Sci- blood sampling for several hours again was entist Development Award MH-47363 (R.T.R.), NIMH accomplished in order to assess accurately the Grant MH-28380, Office of Naval Research Contract integrated levels of plasma LH, PRL, and T N00014-77-C-0245, NIH General Clinical Research Center Grant RR-00425, and McNeil Laboratories Inc., Fort after the acute perturbations of the pituitary hormones. Washington, PA. 447

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 12:37 For personal use only. No other uses without permission. . All rights reserved.

COMMENTS

448

Materials and Methods Eight healthy male volunteers, ages 19-25 yr, on no drugs or medications, were studied on 4 separate days at least a week apart. They reported to the hospital Clinical Studies Center at 0730 h, at which time an indwelling venous catheter was placed and filled with heparinized saline (20 U/ml). Blood samples were taken every 20 min from 0900-1700 h. Coded multiple dose vials containing saline alone, haloperidol (0.5 mg/0.5 ml saline), and hLH (LER 1549; 88 IU/5 ml saline)1 were prepared by the hospital pharmacy. After four baseline blood samples taken between 0900-1000 h, the subjects were given at 1000 h an 0.5-ml im injection containing either normal saline or 0.5 mg haloperidol. Three more blood samples were taken, and then at 1100 h, the subjects were given a 5-ml bolus iv injection containing either normal saline or 88 IU hLH into the antecubital fossa of the arm not used for blood sampling. The 4 test days, arranged in a double blind randomized block design, were as follows: 1) control—saline im at 1000 h, saline iv at 1100; 2) increased PRL—haloperidol im at 1000 h, saline iv at 1100; 3) increased LH—saline im at 1000 h, hLH iv at 1100; and 4) increased PRL and LH— haloperidol im at 1000 h, hLH iv at 1100. The 0.5mg dose of haloperidol had been shown in our first study to increase peak plasma PEL levels by an average of 23 ng/ml, to the high normal range (35-40 ng/ml), which also had resulted in a statistically significant increase in plasma T compared to saline control (1). The 88-IU dose of hLH has been shown by others to increase plasma LH levels to the high normal range (40 mlU/ml), as determined by RIA (2). Peak PRL levels occurred in our earlier study at 1100 h, which was the time chosen for the bolus iv administration of hLH, to result in a temporal coincidence of augmented plasma levels of the two hormones. The blood sample handling, hormone RIAs, and statistical analyses were identical to those previously reported (1), except that PRL was iodinated by the lactoperoxidase method (3). All samples from the same subject were run in the same hormone assay. The statistical analyses were performed on the average change from baseline (0900-1000 h) each hormone showed during the 7 h after the first injection (1020-1700 h). Individual data are not presented but are available from the 1

Kindly supplied by the Ng+ional Pituitary Agency, USPHS. Potency = 2200 IU/mg, based on the OAAD bioassay (National Pituitary Agency, personal communication).

JCE & M • 1978 Vol47 • No 2

>

National Auxiliary Publications Service.2 Two-way analysis of variance, with the Greenhouse-Geisser correction for degrees of freedom, and Kendall's coefficient of concordance (4), was used to compare the hormone changes among the subjects and -^ across the 4 experimental days. Student's t test for paired data and Wilcoxon's signed rank test were , used for multiple comparisons of hormone changes between individual days. The nominal significance levels of the multiple comparisons were corrected for the number of comparisons done, and the true significance levels are given in this report.

Results Neither the haloperidol nor the hLH injections produced any subjective or observable behavioral effects in the volunteers, so that ± the experiment remained double blind until the drug codes were broken. Of the eight vol- ~i unteers studied, three had very small PRL responses on both days during which 0.5 mg , haloperidol was administered. Their inte- -\ grated plasma PRL levels increased an average of only 5 ng/ml compared to the saline control day. In contrast, the other five subjects had clear PRL responses to haloperidol, showing an average increase of 19 ng/ml (Fig. 1A), which was equivalent to the average PRL response to 0.5 mg haloperidol in our earlier study (1). As the purpose of the present study was to examine the effect of increased PRL on plasma T levels, only the five subjects who had large PRL responses were used in the \ calculations of the results of the present study. ^ The reason for the lack of a substantial PRL rise after haloperidol in three of the eight subjects is not clear, but might have been related to individual differences in drug absorption after im administration (5), resulting in lower blood levels of the drug. Figure 1, A-D illustrates the plasma PRL, 2

See NAPS document 03226 for four pages of supple- ^ mentary material. Order from NAPS c/o Microfiche Publications, P.O. Box 3513, Grand Central Station, New V York, New York 10017. Remit in advance for each NAPS x accession number. Institutions and organizations may use purchase orders when ordering; however, there is a billing charge of $5 for this service. Make checks payable to Microfiche Publications. Photocopies are $5. Microfiche are $3 each. Outside the U.S. and Canada postage is $3 4 for a photocopy and $1 for a fiche. •4

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 12:37 For personal use only. No other uses without permission. . All rights reserved.

COMMENTS FSH, LH, and T levels, averaged across the five subjects, for the four experimental conditions. Table 1 lists, for each subject and each treatment day, the average levels of the four hormones during the hours postinjection (1020-1700 h) minus the corresponding average pretreatment baseline levels (0900-1000 h) for that day. Negative values in Table 1

449

indicate postinjection decreases from preinjection baseline, and positive values indicate postinjection increases. Also listed in Table 1 are the mean changes from baseline across the five subjects and the results of the statistical analyses. Both LH (on the non-hLH administration days) and FSH were secreted in their usual B

PROLACTIN ra)/ml (AVERAGE VALUES)

0900

1000

1100

FOLLICLE STIMULATING HORMONE (AVERAGE VALUES) N-S

1200

1300 1400 1500 TIME OF DAY

1600

1700 1800

0900

ng/nil

> HALOPERIIXX. + LH

1000 1100 1200 1300 1400 1500 1600 1700 1800 TIME OF DAY

D 600-.

LUTEINIZING HORMONE ng/ml (AVERAGE VALUES)

- •

-o -O

1000

IIOO

1200

1300 I4OO ISOO TIME OF DAY

I6OO

sa i u

LH

HALWCRIOIX

•*

0900

TESTOSTERONE ng/ml (AVERAGE VALUES) N-S

MALOPERIOOL • LM

0»mg

o — o HALOPCMOOt. OSOmg * • • • » SALINE CONTROL

S»LINt CONTROL

s

1700 I8OO

INJtCTIONS

0900

I0OO

IIOO

1200 1300 MOO 1900 TIME OF DAY

1600 1700 1800

FIG. 1. A-D, Daytime plasma PRL, FSH, LH, and T levels, averaged across five subjects, for the four experimental conditions. > TABLE 1. Effect of haloperidol (HPD) and hLH, alone and combined, on plasma FSH, LH, PRL, and T infivenormal adult men LH b

FSH" Subject

HPD + Saline HPD hLH

Saline HPD

hLH

-13 -21 -2 27 -7

-12 -21 3 -12 -20

-7 -6 -19 -8 -8

2 12

4 5

-8 -7 -29 4 -11

0 3 6

Mean

-10

-3

-12

-10

5

1

2 3

10

hLH

HPD Saline HPD + hLH

14 -15

85 116 38 79 75

73 80 23 74 67

1.7 2.2 -4.8 4.9

7.5

18.0 18.8 12.2 17.0 22.3

6

79

63

2.3

17.7

20 2

Td

PRL C

hLH

+ hLH

Saline HPD

hLH

HPD + hLH

-9.5 -3.1 0.6 5.0 2.7

40.6 7.2 7.1 18.6 29.3

-1.14 -0.70 0.54 -0.61 0.56 2.06 0.57 0.70 -0.83 -0.84 -0.25 0.41 -2.13 -0.69 1.64

0.74 1.73 1.53 0.59 0.44

-0.9

20.6

-0.83 -0.08

1.01

HPD

0.76

Values shown are average posttreatment levels (1020-1700 h) minus average pretreatment baseline levels (0900-1000 h), in nanograms per ml. Statistical tests include two-way analysis of variance (F ratios for subjects and treatments) and Kendall's coefficient of concordance (W) for each hormone. " Fsuhjod = 1.24; df = 4,12; P = NS. FT™.,™, = 0.55; df = 3, 12; P = NS. Kendall's, W = 0.08; P = NS. 6 Fsubjoci = 4.65; df = 4, 12; P < 0.025. F-iwmem = 37.3; df = 3, 12; P < 0.0005. Kendall's, W = 0.94; P < 0.01. c F.subjm = 1.48; df = 4, 12; P = NS. F-iwm.ni = 9.36; df = 3, 12; P < 0.005. Kendall's, W = 0.81; P < 0.01. " Fsuhjec = 1.34; df = 4, 12; P = NS. F-,wmtnt = 5.02; df = 3, 12; P < 0.025. Kendall's, W = 0.65; P < 0.05.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 12 November 2015. at 12:37 For personal use only. No other uses without permission. . All rights reserved.

450

COMMENTS

asynchronous, episodic patterns. Average FSH levels among the five subjects were less variable than those of our earlier study and showed no significant change across the 4 experimental days (Fig. IB and Table 1). Average LH levels were not affected by the haloperidol injections, but they were increased sharply and significantly to peak plasma levels of 400 ng/ml on the 2 hLH administration days (Fig. 1C and Table 1). Multiple comparisons revealed no difference in average LH levels between the saline control and the haloperidol-alone days, but the increase in mean LH levels was significantly higher by 16 ng/ml (P < 0.06) on the hLH-alone day compared to the haloperidol plus hLH day. This was not the result of a greater peak level of plasma LH on the hLH-alone day, but rather because of a somewhat longer disappearance curve (Fig. 1C). The hLH administrations on both days, however, produced an "episode" of increased plasma LH concentration that was about twice the usual magnitude of normal endogenous LH secretory episodes. Average PRL levels among the five subjects showed no consistent change on either the saline control day or the hLH administration day (Fig. 1A). The administration of 0.5 mg haloperidol resulted in prompt, statistically significant increases in PRL levels, an average of 18 ng/ml for haloperidol alone and 21 ng/ml for haloperidol plus hLH (Table 1). By 1600 h, PRL levels were almost back to baseline, as in our first study. Multiple comparisons revealed the PRL increases to be significant for both the haloperidol-alone day and the haloperidol plus hLH day compared to saline control (P < 0.09); the PRL increases on the two haloperidol administration days were not significantly different from each other. For T (Fig. ID), there was the usual decline in plasma concentrations on the saline control day, as also described in our earlier study. In contrast to this normal decline, plasma T levels after haloperidol alone remained fairly constant throughout the hours of the study, beginning about 60 min after the corresponding drug-induced increase in plasma PRL levels (Fig. 1A). On both the hLH-alone and the haloperidol plus hLH days, plasma T levels

JCE & M • 1978 Vol 47 • No 2

rose within an hour after hLH administration and remained elevated throughout the rest of the sampling period (Fig. ID). The changes t» between mean preinjection and postinjection T values for the saline control, haloperidolalone, hLH-alone, and haloperidol plus hLH days were -0.8, -0.1, +0.8, and +1.0 ng/ml, respectively. These changes were significantly different across the 4 test days considered r together, as indicated by the F ratio and Kendall's W (Table 1). Again, as in our earlier study, when the conservative GreenhouseGeisser correction was applied, the acrossdays F ratio fell just short of significance (df = 1,4; P < 0.10), but, as before, the significant