EFFECTS OF OPIATE RECEPTOR BLOCKADE ON GONADOTROPHIN SECRETION BEFORE AND AFTER ADMINISTRATION OF THE OESTROGEN RECEPTOR BLOCKER TAMOXIFEN IN EUGONADAL MEN

Both gonadal steroids and endogenous opioid peptides (EOPs) exert an inhibitory effect on gonadotrophin secretion. It is thought that the negative feedback action of the gonadal steroids, testosterone (T) and oestradiol (E2), on the gonadotrophin secretion is mediated by EOPs. To assess the effects of EOPs and oestrogen and their interrelationship on pulsatile LH secretion we studied two groups of eugonadal men. The subjects of the first group were tested on three different occasions, firstly under basal conditions, secondly during infusion of the opiate receptor blocker naloxone (NAL) (bolus 5 mg + 2.1 mg/h for 7 h), and finally during NAL infusion after 6 weeks administration of the oestrogen receptor blocker tamoxifen (10 mg twice daily). The subjects of the second group were studied before and after 6 weeks administration of tamoxifen. NAL infusion produced a significant increase in mean serum LH levels (4.8 ± SD 1.5 to 6.2 ± 1.8 U/l) and LH pulse frequency (3.7 ± 1.6 to 5.3 ± 1.2 pulses/7 h). No change was seen in mean LH pulse amplitudes (3.5 ± 1.5 vs 3.4 ± 1.0 U/l). After tamoxifen administration alone there was a significant increase in mean LH level (from 5.7 ± 1.3 to 10.1 ± 2.4 U/l), LH pulse amplitude (from 3.8 ± 0.9 to 4.6 ± 0.9 U/l) and LH pulse frequency (from 4.2 ± 1.5 to 5.8 ± 1.7 pulses/7 h). A significant rise in mean serum LH levels was observed during NAL infusion after previous tamoxifen administration in comparison to the infusion of NAL alone (from 6.2 ± 1.8 to 10.5 ± 6.2 U/l). LH pulse frequency (5.3 ± 1.2 vs 6.3 ± 1.3 pulses/7h) and amplitude (3.4 ± 1.0 vs 3.6 ± 1.5 U/l) however, did not change. Mean serum LH level and LH pulse frequency after opiate receptor and oestrogen receptor blockade together did not differ from the results obtained after oestrogen receptor blockade alone. NAL however was expected not only to block opioid‐mediated oestrogen action blockade on LH pulse frequency and mean serum LH levels after oestrogen receptor blockade could mean that the opioid inhibition depends on oestrogen rather than androgen action. If so there could be a parallel between the lack of effect of NAL infusion on gonadotrophin secretion in tamoxifen‐treated men and the same lack of effect in gonadectomized (Shoupe et al., 1985) and postmenopausal women (Reid et al., 1983; Melis et al., 1984; Caspar et al., 1985), in that a long‐term oestrogen deprivation reduces the opioid tone. The assumption that opioid inhibition is primarily dependent on oestrogen action conflicts, however, with a number of earlier studies on the relationship of androgens and EOPs. The feedback effect of the non‐aromatizable androgen dihydrotes‐tosterone can be blocked by NAL (Veldhuis et al, 1984) and androgen receptor blockade abolishes the LH response to NAL (Balzano et al., 1987). Another possible explanation is that androgen action was impeded after tamoxifen administration and for this reason blockade of androgen mediated opioid action by NAL infusion had no additional effect on the LH pulse frequency. In an earlier study we have found that antioestrogens probably interfere with androgen action on gonadotrophin secretion (Spijkstra et al., 1988). Such an effect was apparent in the group treated with tamoxifen alone in this study. While T levels had risen 60–80% following tamoxifen administration, this substantial rise was without a suppressing effect on LH pulse frequency, which had increased upon tamoxifen administration. The results obtained in this study therefore seem to be consistent with this assumption. However some caution is warranted when interpreting the results of the study in this way. Antioestrogens are known to bind to receptors other than E2 receptors (Sutherland et al., 1980), which raises the question if antioestrogens may exert effects other than their antioestrogenic or oestrogenic properties. Indeed studies with the antioestrogen clomiphene citrate both in vivo (Kerin et al., 1985) and in vitro (Miyake et al., 1983) provided indications in support of this assumption.