The effects of castration, testosterone replacement and photoperiod upon hypothalamic β-endorphin levels in the male syrian hamster

0306-4522/87 $3.00+ 0.00

Neuroscience Vol. 23,No. 3, pp. 10754082,1987 Printedin Great Britain

PergamonJournalsLtd 0 1987IBRO

THE EFFECTS OF CASTRATION, TESTOSTERONE REPLACEMENT AND PHOTOPERIOD UPON HYPOTHALAMIC /bENDORPHIN LEVELS IN THE MALE SYRIAN HAMSTER A. C. ROBERTS,N. D. MARTENSZ,M. H. HASTINGS and J. HWBERT* University of Cambridge, Department of Anatomy, Downing Street, Cambridge CB2 3DY, U.K. Abatraet-Syrian hamsters kept in long day-lengths have active gonads and high circulating levels of gonadal steroids. Under the influence of the pineal gland, animals exposed to short photopcriods undergo testicular regression, have low circulating levels of testosterone and gonadotrophins and elevated levels of j3endorphin within the hypothalamus. This paper describes the interaction between testosterone and photoperiod in the regulation of jendorphin levels in three regions of the hypothalamus. Hypothalamic Bendorphin levels were measured by a combination of high-performance liquid chromatography and radioimmunoassay techniques that allows separation of the r%endorphin (1-31) peptide from its metabolites and precursors. All of the fi-endorphin-like immunoreactivity in the hypothalamus of the male hamster, in both photoinhibited and photostimulated conditions, was found to represent the 31-aminoacid peptide. In photostimulated hamsters, chronic castration was associated with a significant increase of jendorphin levels in the anterior hypothalamus and mediobasal hypothalamus, which was reversed by treatment with exogenous testosterone. Castration prevented the ability of naloxone, an opiate receptor antagonist, to release luteinixing hormone, and this effect was also reversed by exogenous steroid. In photoinhibited hamsters, however, castration had no effect upon jendorphin levels in the preoptic area or mediobasal hypothalamus, and there was only a small increment in the anterior hypothalamus. Significantly, t?I-endorphinlevels in all areas of the hypothalamus of photoinhibited castrates were not decreased by testosterone treatment. In addition, administration of exogenous testosterone did not restore sensitivity to naloxone in these animals. These results show that inhibitory photoperiods uncouple the functional interaction between testosterone and the hypothalamic jendorphin system, and reinforce the proposition that a change in the activity of /?-endorphin neurons is a component of the central mechanism which mediates the environmental control of reproduction.

In photoperiodic mammals, seasonal changes in dayCentral infusions of /I-endorphin,” and antiserum length lead to a suppression of the reproductive to fl-endorphin, 39 have been used to demonstrate axis.2s Male Syrian hamsters exposed to inhibitory the role of this peptide in the regulation of pituitary photoperiods of less than 12 h light per day exhibit a luteinixing hormone (LH) secretion through modifiprecipitous decline in serum gonadotrophin levels cation of the release of luteinizing hormone-releasing followed by testicular collapse and a consequent fall hormone (LHRH) from the median eminence.**L’~27~54 in circulating levels of gonadal steroids.6*23Although There is extensive evidence of a direct interaction it is well established that these effects are mediated between gonadal steroids and neuronal /I-endorphin. by the pineal gland via changes in the nocturnal The major group of neurons synthesizing /I-endorpattern of melatonin secretion,6*23*37 the central neurophin within the hypothalamic arcuate nucleus1”‘**53 endocrine systems which translate this photoperiodic contains a small subpopulation able to concensignal into altered pituitary hormone secretion are trate exogenous steroid. 33 Conditions which modify unknown. gonadal activity and therefore circulating levels In non-photoperiodic species, endogenous opioid of gonadal steroids, for example pregnancy, repeptides, particularly /I endorphin, are involved productive suppression and menstrual cyclicity, also in the central control of a wide range of neuroalter the content of /I-endorphin within the brain”v4 endocrine and behavioural processes.22~u~31~32.u).s2 or portal circulation. ‘I Furthermore, in male and female rats, the content of fl-endorphin within the hypothalamus is increased by castration and reduced *To whom all correspondence should be addressed. Abbreuiutions: AHA, anterior hypothalamic area; HPLC, by subsequent replacement of steroid.“‘“,” However, high-performance liquid chromatography; LH, lu- it is important to note that measurement of total teinixing hormone; LHRH, luteinixing hormone- /I-endorphin-like immunoreactivity within brain releasing hormone; F’OA, preoptic area_ MBH, me diobasaI hypothalamus: IUA, radioimmunoassav: LD. tissue must be regarded with some caution. The long day-length; SD, short day-length; FQMC,’ pro: peptide can undergo a considerable amount of intracellular metabolism in both the brain and pituitary opiomelanocortin; a-MSH, a-melanocyte-stimulating hormone. gland by C-terminal proteolysis and N-terminal 1075

1076

A. C. ROBERTSet

acetylation.2~10~14~4’~ss These processes produce peptides which show a high degree of cross-reactivity with b-endorphin antisera in ti conventional radioimmunoassay (RIA), but may possess little or no opiate activity.‘,9 It is therefore important to identify the molecular form of the peptide accurately, since an increase in the amount of metabolites, such as that reported to accompany sexual maturation in the female rat,29 may be one mechanism for reducing the degree of opioid inhibition of gonadotrophin secretion. In addition to the direct effects of steroids on brain b-endorphin levels, the responsiveness of LH secretion to the administration of exogenous opioids or to naloxone, an opiate receptor blocker, is also influenced by changes in the circulating levels of gonadal steroids. In female primates and rodents, the release of LH after naloxone administration was greatest in the phase of the cycle when steroid levels were maximal.“~3s*4s*“6Gonadectomy abolished the ability of naloxone to release LH, but this was restored by the administration of exogenous steroid 4Al93.42 All of these findings would indicate that the B-endorphin neuronal system plays a central role in the regulation of reproductive activity. Previous studies in both hamsters and sheep have demonstrated that endogenous opioids are responsive to photoperiodic stimuli7,‘2,1%%37 and exogenous melatonin.23 In photoinhibited hamsters the tissue content of fl-endorphin within the mediobasal hypothalamus (MBH) and anterior hypothalamus (AHA) was significantly increased above that in photostimulated animals,37 and the LH response to naloxone was lost.‘5*MIn the photostimulated hamster, castration blocked the response. to naloxone but this was restored by testosterone. In constrast, administration of exogenous testosterone to photoinhibited, castrated hamsters was not able to restore sensitivity to naloxone. Exposure to inhibitory photoperiods therefore altered the functional relationship between gonadal steroids and endogenous opioids in the control of LH secretion. To investigate this relationship directly, the experiments reported here endeavoured to determine the effects of castration, steroid replacement and photoinhibition upon the levels of hypothalamic fl-endorphin within the brain of male Syrian hamsters. In addition, the pattern of posttranslational processing of B-endorphin within the brain of photostimulated and photoinhibited animals was examined to determine whether reported changes in the tissue content of the endogenous opioid could be correlated with altered intraneuronal metabolism. EXPERIMENTAL PROCEDURES Experiment 1: Effects of castration and testosterone replacement upon b-ena’orphin levels and the luteirking hormone response to naioxone in photostimulated hamsters

Thirty-two male Syrian hamsters (Wrights, Essex Ltd) of 16 L : 8 D (LD; lights on

were held on a photoschedule

al

01.00 h) with water and lab chow available ad libitum. Sixteen animals were castrated under methohexitone anaesthesia, half of them receiving subcutaneous implants (Silastic, Dow-Corning, 2.0 mm i.d., 3.2 mm o.d., 20 mm length) filled with crystalline testosterone (T, Sigma). These implants have previously been shown to maintain testosterone at levels in the region of those observed in intact, photostimulated animals.36 The remaining castrates received empty implants. All animals were then left for 8 weeks. Twenty minutes before being killed at either 17.00 (intacts only) or 21 .OOh (intacts and castrates), half of the animals in each group were given naloxone (5 mg/kg intraperitoneal (i.p.)) and the others received a saline injection. Animals were killed by cervical dislocation and decapitation. Trunk blood was stored overnight at 4”C, serum harvested and stored at -20°C prior to determination of LH by RIA. The brain was quickly removed, frozen on dry ice and stored at -70°C prior to determination of /I-endorphin content. Experiment 2: Effects ojcastration and testosterone replacement upon /T-endorphin levels and the luteinizing hormone response to naloxone in photoinhibited hamsters

Thirty-two hamsters were either left intact or castrated, the castrates receiving a testosterone-filled implant, as described above. Immediately following castration and implantation, the animals were transferred to a short photoperiod (SD) of 8 L: 16 D (lights on 01 .OOh) and again left for 8 weeks. They were then treated with either saline or naloxone 20min before being killed as described above. Experiment 3: Identity of b-endorphin in the hypothalamus of photoinhibited or photostimulated, intact hamsters

Sixteen intact animals were kept for 8 weeks in LD. A second group of 16 intacts were held on SD for 8 weeks, by which time the testes had regressed. All of the animals were killed by cervical dislocation at 21 .OOh. The animals were decapitated and the brain removed and stored at -7O’C. Experiment 4: Effects of withdrawal of exogenous steroid upon Iuteinizing hormone secretion in photostimulated and photoinhibited castrates

Forty animals held on LD were castrated and given testosterone implants. Twenty animals were immediatety transferred to SD the others remained in LD. After 8 we&, serum samples were taken from both groups by cardiac puncture under avertin (tribromoethanol) anaesthesia and the testosterone implants removed. Four days later a seeoad serum sample was taken to determine the response of LH secretion to steroid withdrawal. Radioimmunoassays

Coronal frozen sections (6OO~m) were cut from the brains obtained in experiments 1 and 2. h&mpu&w (1.2 mm diameter) were taken bilaterally from f area (POA), AHA and MBH as described experiment 3, the whole hypothaksmus, con from the AHA and MBH, was dissected the method of Glowinski and Iversm.zo TrrtnsvCrse cw were made through the optic chiasm r&rally aml #&m&iately in front of the ma~&&+ry bodii caueally. Ho&~~tal cuts were made at the Ievel of the anterior eommissure, and laterally through the hypothalamic s&us. All brain tissue samples were extracted in 250~1 of a mixture of methanol/l M HCI (4: 1) con&ah&g 300 @ml iodoacetamide and phenyhnethylsu@imnyl fluoride. After centrifugation (10,OOQgat room temperature for 4 mig), the supematant from each brain sample in exper&n& 1 aztd 2 was evaporated under Muurn at WC, dissolved in 900 ~1 of assay b&z (&OSM and O.fJ@Ji % containing 0.25% bovine m al-in thiomersalate) and +ndorphin cotient measured by RIA (experiments 1 and 2) as deseri&i p~~+ousiy.~~ In experiment 3, following centrifugation, extracts of four hype-

/I-Endorphin,

steroids and photoperiod

1077 W.

T

8t

INT

CAST C&ST

Treatment

INT

CAST

INT

CAST CqST T&T

Tf%T

T&T AHA

MBH

Area

C+T

POA

Fig. 1. Tissue /?-endorphin content (means f S.E.M.) of three hypothalamic areas from intact, castrated or testosterone-implanted male hamsters maintained on a stimulatory photoperiod (16 L: 8 D) for 8 weeks. thahuni were pooled and dried down under vacuum at 40°C. The residue was dissolved in 3QOpl 50% acetic acid and retained at 4°C for separation of different molecular forms of ~~ndo~~n by ~~-~~o~~~ liquid c~orna~~phy (HPLC) prior to q~tifi~tion by RIA.zp The antiserum to Jcndorphin cross-reacted on an equimolar basis with human lipotropin, human and camel b-endorphin 1-31, #I-endorphin 1-27 and l-26, and their N-acetylated forms. Serum LH was measured by IUA using reagents supplied by NIADDK for rat LH. This assay has been used previously in hamsters.36*37 Statistics

Treatment effects were determined by analysis of variance of log-transformed data and differences in means were tested post hoc by the Student-Newmann-Keuls test. RRSULTS There were no significant differences between LH concentrations or brain fl-endorphin levels from intact animals killed at 17.00 or 21 .OOand these groups were pooled for further analysis. Likewise, naioxone had no significant effect upon hypothalamic fl-endorphin levels and so the data from saline- and

naloxone-treated group.

animals were pooled within each

The tissue content of /Sendorphin within the hypothalamus of photostimulated animals showed a clear gradation with highest levels in the MBH and lowest in the POA (Fig. 1). In all three areas, fl-endorphin levels increased following castration. This effect was significant in the AHA (F = 3.33, d.f. 2,27, P < 0.05) and MBII (F = 4.71, d.f. 2,26, P < 0.05). Administration of exogenous testosterone completely reversed the effects of chronic castration upon /I-endorphin. In all three areas of the hypothalamus, fl-endorphin levels in castrates given testosterone were identical to those observed in intact animals. Manipulation of the steroidal en~ronment of photostimulated animals had the expected effect upon serum LH levels (Table la) (treatment effect, F = 19.06, d.f. 2,24, P ~0.01). Castration led to a 34fold increase in LH but this was prevented by steroid replacement. The stimulatory effect of

Table 1. Serum luteinixing hormone levels (meanfS.E.M., n&ml) of intact, castrated or castrated and testosterone-implanted male Syrian hamsters Intact (a) Long days Saline Naloxone (b) Short days Saline Naloxone

Castrate

Castrate + t~tos~rone

1.21 If:0.18 4.98 + 0.18*

5.19 f 1.09 7.34 f 0.76

1.00*0.09 4.02 f 2.01*

0.59 f 0.04 0.55 f 0.05

1.33 f0.18 2.27 j, 1.05

0.60 f 0.02 0.62 f 0.10

Injected with naloxone or saline and killed following 8 weeks on (a) long days or (b) short days. (n = 4-8 per observation). *Significantly different (P < 0.05) from respective saline group.

1078

A. C. ROEERTS et

al.

-*-

r------~--1

-*l-ml

INT

CAST CAST

INT

r--

CAST CAST

Treatment T&T Area

Fig. 2. Tissue ~~~0~ or ~~~~~p~nt~

MBH

n.sl

INT

Fn.sl

CAST CAST

T&T AHA

T&T POA

content (means 2 S.E.M.) of three ~~ areas from intact, castrated mate hamsters nMntain& on an inhibitory photoperkd (8 L: 16Dj for 8 weeks.

naloxone upon LH levels (drug effect F = 22.4, d.f. 1,24, P < 0.01) was significantly altered by castration and steroid replacement (Drug-treatment interaction F = 5.28, d.f. 2,24, P < 0.05). Naloxone given to intact animals increased serum LH. Tbis response was lost in castrates but was restored by exvganous testosterone.

Exposure to short photvperiods altered the response of neuronal jkndorphin levels to changes in the steroidal environment. j?-Endorphin levels in the PCA wzre not altered by castration, but testosterone treatment produced a small but sign&ant increase over intact levels (Fig. 2) (F = 3.95, d.f. 2,27, P < 0.05). In the AHA, chronic castration increased ~-endorphin levels, but these were not reduced by and remamed with testosterone treatment signifkantly higher than in intact animaht (F = 4.64, d.f. 227, P ~0.05). ~-~~in levels in the MB3 of p~toi~i~~ hamsters were m&acted by either castration or testosterone a~i~~~tion. Manipulation of the steroidal ~~ronment had a significant effect upon serum LH levels determined after 8 weeks in short photoperiods (Treatment e&f% F= 19.79, d.f. 2,25, P eO.01) (Tahie lb). Castrated animals bad levels of LH sign&a&y bigher than those in intact males although these Ievek were lower than those observed in photostimuIatod castm%s. AS in photostimulated hamstem, the post-castmtion increase in LH levels was prevented by exogenous observed in testosterone. In contrast to the rqonses phot~~rn~at~ hamstem, naloxone had no sign&ant eikt upon serum LH k&s in any of the photoi~bit~ animals. In particular, exogenous tes-

tosterone failed to restore sensitivity to nafoxone in photoinhibited castrates. Experiment 3 Separation by HPLC of pooled extracts of hp ~~~~rn~ photoperiods showed noreactivity &u&d at no evidence for the presemz of either C-~-or N-acetylated forms of ~~~~ in the hypethalamus of the hamster. Experiment 4 Table 2 presents serum LH levels in castrated animals before, and 4 testosterone imphknt. fn steroid with&W Ied to a LH secretion. In constrnst, photoinW&ed tion. DiSCWON These axgcrimeab shvw thst levels of jkndorphin inlooalimdregionsofthe S~~~~~~ e&t of testos&rone is sen&ve to photosthmdated hams/kl&@in k@s are increased by ca&&iti and r&to& by treatment with enogenous teste. E%Mbwer, in pbto-

1079

/I-Endorphin, steroids and photoperiod Long day

fi

Pi

a

I

2r

i

Oxidized

a. cENDl-31 b.Ac-cENDl-31 c. cENDl-27 d. ENDl-26 e. AC-cENDl-27 1. AcENDl-26

a

d

bc

15 Retention

e

f

20

25

time (min)

Fig. 3. Chromatographic separations by HPLC of B-endorphin-like immunoreactivity in the pooled hypothalami from male Syrian hamsters maintained in long or short photoperiods.

pituitary to release LH in response to the opiate antagonist naloxone altered concurrently. Although testosterone restored the LH response to naloxone in photostimulated, castrated hamsters, it failed to do so in photoinhibited, castrated animals. Chromatographic identity of /?-endorphin Chromatographic separation confnmed that the /I-endorphin-like immunoreactivity in the hypothalamus of photostimulated animals with high serum testosterone titres, or photoinhibited animals with lower serum testosterone titres, was in the form of fl-endorphin 1-31. No evidence for C-terminal or N-acetylated forms was obtained. In this respect, the Table 2. Serum luteinizing hormone (mean *S.E.M., ng/ml) of photostimulated and photoinhibited castrated Syrian hamsters before (0 days) and 4 days after withdrawal of exogenous testosterone (n = 20 per group) Photostimulated Photoinhibited

0 days

4 days

1.25kO.17 0.67 + 0.08

4.08 + 0.38. 0.88 f 0.15

*Significantly different (P < 0.05) from respective 0 days measurement.

male hamster differs from the adult rat, in which about 40% of /I-endorphin-like immunoreactivity in the brain is due to C-terminal shortened peptides.14*29 C-Terminal proteolysis is also absent in the pituitary of the hamster, again contrasting with the rat, although the neurointermediate lobe of the pituitary of the hamster does exhibit N-terminal acetylation of /?-endorphin (unpublished observations), in common with other species. It is likely, therefore, that the elevation of /I-endorphin within the MBH and AHA following castration of photostimulated animals represented an increase in the content of the 1-31 peptide, rather than the content of any metabolites. Elevated levels of /I-endorphin have also been shown to correlate with reduced serum testosterone titres in the socially subordinate talapoin monkey.30 As with the Syrian hamster, HPLC separation showed that /l-endorphin 1-31 was the major endorphin peptide in the cerebrospinal fluid, although the molecular forms of /?-endorphin in neuronal tissue remain to be determined. In contrast, pubertal development in the female rate was associated with marked changes in post-translational processing of #I-endorphin.29 There is, therefore, general agreement

1080

A. C. ROBERTS

from these studies that /I-endorphin (1-31) is raised in a variety of conditions associated with reproductive inhibition, but is should bc stressed that different neurochemical mechanisms may be involved in such changes of opioid function. Because peptides are amenable to various biochemical modifications, neuropeptidergic transmitter molecules present the neuron with a highly flexible range of responses to match a variety of stimuli. These responses may involve differential processing of larger precursor molecules as well as metabolic alterations to the primary structure of individual peptides. For example, /I-endorphin is cleaved from a larger precursor molcule, pro-opiomelanocortin (POMC), which also contains the amino-acid sequence of a-melanocyte-stimulating hormone (a-MSH). This peptide has been shown to have the potential to interact with /I-endorphin in the regulation of LH release.26 It therefore remains to be determined whether environmental regulation of reproduction involves changes in intraneuronal metabolism of the entire POMC molecule in addition to modification of the individual peptides. /?-Endorphin and steroids

Accumulating evidence links increased activity in the endogenous opioid systems with the ability of gonadal steroids to inhibit LH release. In male rats and photostimulated hamsters, castration removed gonadal inhibition of LH release and increased hypethalamic levels of /I-endorphin.” This may arise because of a reduction in release of the peptide in the absence of a steroidal activation of the neuron. In both male and female rats, administration of exogenous gonadal steroids restored activation of the endorphin neuron, presumably increasing the rate of neuropeptide release and thereby reducing both the content of the peptide in neural tissue and also circulating LH levels. 4749*50 This effect occurred in photostimulated hamsters, most markedly in the MBH, an area containing the ceil bodies of the arcuate nucleus, and less prominently in the BOA, an area containing and predominantly fibres terminalsl”‘* (J. Herbert, unprtbtiszrod observations). The site and mechanism of action of testosterone upon the /.I-endorphin-containing neu.ronaI system is unclear. In female monkeys and rats, levels of the peptide in the pituitary portal blood vary during the menstrual or oestrous cycles and are increased by the administration of ovarian steroids,W’ sugg6sting that steroids enhance terminal release of 8-endotpbin, although other influences, for example m&&ion of genomic expression within the cell body shot&l not be excluded.*’ In the photoinhibited hamster, the relationship between testosterone and hypothaiumic B-endotphin is very different from that aecn in Photost&&te!d animals. Castration resulted in 0nIy a amaIl incrcaae in p-endorphin levels in the AHA, and none in the other areas of the hypothalamus. In addition, ex-

et al.

ogenous testosterone did not reduce fi-endorphin levels in any of the areas studied. These results show that some process associated with inhibitory photoperiods desensitizes the hypothalamic /?-endorphin system to the effects of testosterone. Since treatment of photoinhibited animals with testosterone did not reduce p-endorphin, the increased levels of the peptide observed previously in photoinhibited or melatonin-treated animals23 probably occurred independently of the decline in serum titres of steroid which accompanied gonadal involution. This effect upon /I-endorphin neuronal activity is, therefore, qualitatively different from that, seen in the photostimulated, castrated hamster, in which changes in /I-endorphin content were clearly steroid-dependent. These data suggest that the photoperiodic timemeasuring apparatus within the brain has a direct access to the fi-endorphin neuronal system. However, it remains to be determined whether the increased levels of B-endorphin in the photoinhibited animal represent an invactivation or an increase in activity within the /I-endorphin system. Although LH levels were lower in castrates given testosterone than in untreated castrates after 8 weeks in short photoperiods, castration of intact, photoinhibited hamsters’3*43,” or removal of the testosterone implants from photoinhibited castrates did not cause LH levels to rise. Gonadal steroids do not, therefore, suppress LH release in the hamster once photoinhibition has been established, although they may have an important role in the suppression of LH release in the initial stages of photoinhibition. The absence of detectable steroid negative feedback after 8 weeks in short photoperiods is consistent with inactivation of the fl-endorphin system which is a major mediator of such feedback. The eflect of naloxone

Blockade of opiate receptors by naloxone released LH in intact rats’.’ and photo&imuIated hamsters i5.36,37indicating the presence of en&genous opioid tone as a regulator of LH release. Castration of animals of either species stimulated LH levels, in part by increasing the frequency of the circhoral oscillator driving the release of luteinizing hormonereleasing hormone (LHRH). NaIexone was then ineffective in promoting further LH discharge, leading to the suggestion that endogenous opioids do not exert any inhibitory influence over LH release in the castrate. This inactivation of the /I-endorpbin system would lead to a reduction in release of the peptide and consequently an increase in tissue content as described. Naloxone did not release LH in intact or testosterone-treated photoinbibited hanbster, even though LH levels and the frequency of the LHRH &choral osciIlator were both low. By an&g$ with the castrate, the failure of the LH response to naloxone, which has been reported mo&y in p@otoinhibited hamstersis*“” and sheep,‘*‘*might~be taken

j-Endorphin,

steroids and photoperiod

as evidence that endogenous opioids do not exert a suppressive effect upon LH release during reproductive inhibition. If opioids were responsible for the chronic inhibition of LH secretion, then release of opioid control by receptor blockade would be expected to induce a discharge of LH. However, a recent pharmacological study in the rat has demonstrated that chronic treatment with the exogenous opiate, morphine, prevented or even reversed that normal ability of naloxone to release LH,3 indicating that the functional relationship between opiate activity and the operation of the LHRH pulse generator is not constant and that the naloxone test may not always be a reliable indicator of opioid function. If within the photoinhibition enhanced activity Jendorphin system there would not necessarily be an enhanced responsiveness to naloxone. Indeed, chronic inactivation of the LHRH pulse generator by endogenous opioids may alter its operating characteristics and make it insensitive to acute blockade of

1081

opioid activity. Final resolution of the apparent paradox between tissue fl-endoxphin levels and the response to naloxone awaits further direct analysis of the effects of photoperiod upon the function of the B-endorphin neuron. Nevertheless, these data indicate that photoperiodic stimuli alter the pattern of the interaction between steroids and endogenous opioid peptides. It is now important to understand more about the way in which the pineal signal, which seems to be transmitted to the hypothalamus as the nocturnal duration of the melatonin peak,6J3 is translated into altered fl-endorphin activity, and how this, in turn, may link the internal reproductive state of the animal to the external environment.

Acknowledgements-This work was supported by the MRC. The authors are grateful to NIADDK for supplying the RIA reagents, to Tim Crane and Sue Insole for the diagrams, and to Jane Rowe11 for her help in preparing the

manuscript.

REFERENCES 1. Akil H., Young E. and Watson S. J. (1981) Opiate binding properties of N- and C-terminus modified j?-endorphins. Peptides 2, 284-292. 2. Akil H., Ueda Y., Lin H. L. and Watson S. J. (1981) A sensitive coupled HPLC/RIA technique for separation of endorphins: multiple forms of B-endorphin in rat pituitary intermediate v. anterior lobe. Neuropeprides 1, 4294%. 3. Almcida 0. F. X., Schulz R. and Herz A. (1986) Paradoxical LH and prolactin responses to naloxone after chronic treatment with morphine. J. Endocr. 108, 181-189. 4. Bhanot R. and Wilkinson M. (1983) Opiatergic control of luteinizing hormone is eliminated by gonadectomy. Endocrinology 112, 399-40 1. 5. Bhanot R. and Wilkinson M. (1984) The inhibitory effect of opiates on gonadotrophin secretion is dependent upon gonadal steroids. J. babcr. 102, 133-141. 6. Bittman E. L. (1984) Melatonin and photoperiodic time measurement: evidence from rodents and ruminants. In The f%eu/ Gland (ed. Reiter R. J.), pp. 155-192. Raven Press, New York. 7. Brooks A. N., Lamming G. E., Lees P. D. and Haynes N. B. (1986) Opioid modulation of LH secretion in the ewe. J. Reprod. Fert. 76, 693-708. 8. Cicero T. J., Schainker B. A. and Meyer E. R. (1979) Endogenous opioids participate in the regulation of the hypothalamic-pituitary-luteinizing hormone axis and testosterone’s negative feedback control of luteinizing hormone. Endocrinology 104,1286129 1. 9. Deakin F. J. W., Dostrovsky J. 0. and Smyth D. G. (1980) Influence of N-terminal acetylation and C-terminal proteolysis on the analgesic activity of B-endorphin. Biochem. J. 189. 501-508. of B-endorphin immunoreactive peptides 10. -&r&-M., Lazum C., !%iah N. G: and.ChretiA M. (1983) Character&on in rat pituitary and brain by coupled gel permutations and reversed-phase high-performance liquid chromatography. J. Chromat. 266, 163172. 11. Drouva S. V., Eppelbaum J., Tapia-Aranciba L., Laplante L. and Kordon C. (1981) Opiate receptors modulate LHRH and SRIF release from mediobasal hypothalamic neurons. Neuroendocrinology 32, 163-167. 12. Ebling F. J. P. and Lincoln G. A. (1985) Endogenous opioids and the control of seasonal LH secretion in Soay rams. J. Endocr. 187, 341-353. 13. Ellis G. B. and Turek F. W. (1980) Photoperiodic regulation of serum luteinizing hormone and follicle-stimulating hormone in castrated-adrenalectomized male hamsters. Endocrinology 106, 1338-1344. peptides in rat hypothalamus. 14. Emeson R. B. and Eipper B. A. (1986) Characterization of pro-ACTH/endorphin-derived J. Neurosci. 6, 837-849. 15. Eskes G. A., Wilkinson M. and Bhanot R. (1984) Short-day exposure eliminates the LH response to naloxone in golden hamsters. Neuroendocrinology 39, 281-283. 16. Eve&t B. J., Meister B., Hijkfelt T., Melander T., Terenius L., Rohaeus A., Theodorsson-Norheim E., Dockray G., Edwardson J., Cue110C., Elde R., Goldstein M., Hemmings H., Ouimet C., Walaas I., Greengard P., Vale W., Weber E., Wu J.-Y. and Chang K.-J. (1986) The hypothalamic arcuate nucleus-median eminence complex: immunohistochemistry of transmitters, peptides and DARPP-32 with special reference to coexistence in dopamine neurones. Bruin Res. Rev. 11, 97-155. 17. Finley J. C. W., Lindstrom P. and Petrusz P. (1981) Immunocytochemical localization of b-endorphincontaining neurons in the rat brain. Neuroendocrinology 33, 28-42. 18. Gee C. E., Ching-Ling C. C., Roberts J. L., Thompson R. and Watson S. J. (1983) Identification of proopiomelanocortin neurones in rat hypothalamus by in siru cDNA-mRNA hybridization. Nature 306, 374-376. 19. Gabriel S. M., Simpkins J. W. and Kalra S. P. (1983) Modulation of endogenous opioid influence of luteinizing hormone secretion by estradiol and progesterone. Endocrinology 113, 1806-1811. 20. Glowinski J. and Iversen L. L. (1966) Regional studies of catecholamines in the rat brain-I. J. Neurochem. l&65=9.

1082

A. C. ROBERTSet al.

21. Greenstein B. D. (1986) Steroid hormone receptors in the brain. In Neuroendocrinology (eds Lightman S. L. and Everitt B. J.), pp. 3248. Blackwell, Oxford. 22. Grossman A. and Rees L. H. (1983) The neuroendocrinology of opioid peptides. Br. med. Bull. 39, 83-88. 23. Hastings M. H., Herbert J., Martens2 N. D. and Roberts A. C. (1985) Melatonin and the brain in photoperiodic mammals. In Phofoperiodism, Melutonin and the Pineal, CIBA Foundation Symposium 117. pp. 57-78. Pitman. London. 24. Kalra S. P. and Leadem C. A. (1984) Control of luteinizing hormone secretion by endogenous opioid peptides. In Opioid Modulation of Endocrine Function (eds Delitala G. et al.), pp. 171-184. Raven Press, New York. 25. Kinoshita F., Nakai Y., Katakami H., Kato Y., Yajima H. and Imura H. (1980) Effect of /?-endorphin on pulsatile luteinizing hormone release in conscious castrated rats. Life Sci. 27, 843-846. 26. Khorram 0. and McCann S. M. (1986) Interaction of a-melanocyte-stimulating hormone with /I-endorphin to influence anterior pituitary hormone secretion in the female rat. Endocrinology 119, 1071-1075. 27. Leadem C. A., Crowley W. R., Simpkins J. W. and Kalra S. P. (1985) Effects of naloxone on catecholamine and LHRH-release from the perifused hypothalamus of the steroid-primed rat. Neuroendocrinology 4@, 497-500. 28. Lincoln G. A. and Short R. V. (1980) Seasonal breeding: Nature’s contraceptive. Rec. Prog. Harm. Res. 36, l-52. 29. Martens2 N. D. (1985) Changes in the processing of B-endorphin in the hypothalamus and pituitary gland of female rats during sexual maturation. Neuroscience 16, 625-640. 30. Martens2 N. D., Vellucci S. V., Keveme E. B. and Herbert J. (1986) B-endorphin levels in the cerebrospinal fluid of male talapoin monkeys related to dominance status and the luteinizing hormone response to naloxone. Neuroscience 18, 651458. 31. Meyersen B. and Terenius L. (1977) /3-endorphin and male sexual behaviour. Eur. J. Phurmuc. 42, 191-192. 32. Millan M. H. and Herz A. (1985) The endocrinology of the opioids. In?. Rev. Neurobiol. 26, l-83. 33. Morrell J., McGinty J. F. and Pfaff D. W. (1986) A subset of B-endorphin or dynorphincontaining neurons in the medial basal hypothalamus accumulates estradiol. Neuroendocrinology 41, 417426. 34. Petraglia F., Locatelli V., Penalva A., Cocchi D., Genazzani A. R. and Muller E. E. (1984) Gonadal steroid modulation of naloxone-induced LH secretion. J. Endocr. 101, 33-39. 35. Quigley M. E. and Yen S. S. C. (1980) The role of endogenous opiates in LH secretion during the menstrual cycle. J. clin. Endo. Merab. 51, 179-181. 36. Roberts A. C., Hastings M. H., Martens2 N. D. and Herbert J. (1985) Naloxone-induced secretion of LH in the male Syrian hamster: modulation by photoperiod and gonadal steroids. J. Ena’ocr. W, 243-248. 37. Roberts A. C., Martensz N. D., Hastings M. H. and Herbert J. (1985) Changes in photoperiod alter the daily rhythms of pineal melatonin content and hypothalamic /Iendorphin content and the luteinizing hormone response to naloxone in the male Syrian hamster. Endocrinology 117, 141-148. 38. Sarkar D. J. and Yen S. S. C. (1985) Changes in #i-endorphin-like immunoreactivity in pituitary portal blood during the estrous cycle and after ovariectomy in rats. Endocrinology 116, 2075-2079. 39. Schulz R., Wilhelm A., Pirke K. M., Gramsch C. and Herz A. (1981) b-endorphin and dynorphin control serum luteinizing hormone level in immature female rats. Nature 294, 757-759. 40. Sirinathsinghji D. J. S. (1985) Modulation of lordosis behaviour of female rats by naloxone, B-endorphin and its anti-serum in the mesencephalic central grey: possible modulation by LHRH. Neuroe&crinology Z@,222-230. 41. Smyth D. G., Massey D. E., Zakarian S. and Finnie M. D. A. (1979) Endorphins are stored in bioIogicaiiy active and inactive forms: isolation of alpha-N-a&y1 peptides. Nuture 279, 252-254. 42. Sylvester P. W., Van Vugt D. A., Aylesworth C. A., Hanson E. A. and Meites J. (1982) Effects of morphine and naloxone on inhibition by ovarian hormones of pulsatile release of LH in ovariectomized rats. Kwvendocrinulogy 34, 269-272. 43. Tate-Ostroff B. and Stetson M. H. (1981) Correlative changes in the response to castration and the onset of refractoriness in male golden hamsters. Endocrinology 32, 325-329. 44. Urbanski H. F., Simpson S. M., Ellis D. H. and Follett B. K. (1983) Secretion of follicle-stimulating hormone and luteinizing hormone in castrated golden hamsters during exposure to various photoperiods and to natural dq>bs. J. Endow. 99,

379-386.

45. Van Vugt D. A., Baskt G., Dyrenfurth I. A. and Ferin M. (1983) Naloxone stimulation of luteinizing hormone secretion in the female monkey: influence of endocrine and experimental conditions. Er&crino/ogy 113, 185%1&X 46. Van Vugt D. A., Sam N. Y. and Ferin M. (1985) Reduced frequency of latelnitig hormone secretion in the luteal phase of the rhesus monkey. Involvement of endogenous opiates. EnBocrinelog ir5, 1095-l 101. 47. Wardlaw S. L. (1986) Regulation of /&endorphin, corticotropin-like intermediate lobe peptide, and alpha melanotropinstimulating hormone in the hypothalamus by testosterone. Endocrinology 119, 19-24. 48. Wardlaw S. L. and Frantz A. G. (19ft3) Brain b-endorphin during pregnancy, parturition, and the post-parturn period. Endocrinology

113, 1664-1668.

Brain Res. US, 327-331. 49. Wardlaw S. L. Thoron L. and Frantz A. G. (1982) Effects of sex steroids on brain fkmdophin. 50. Wardlaw S. L., Wang P. J. and Frantz A. G. (1985) Regulation of /I-endorphin and ACTH in brain by estradiol. Lif Sci. 37, 1941-1947. 51. Wehrenberg W. G., Wardlaw S. L., Frantz A. G. and Ferin M. (1982) /?-Endorphin in hypophyseal portal blood: variations throughout the menstrual cycle. Eiuiucrinology 111, 879-881. 52. Wiesner J. B. and Moss R. L. (1!%6) suppression of receptive and proceptive behaviour in ovariectomized, oestrogen-primed rats by intraventricular fi-endorphin: studies of bchaviouml spec&ity. N~oen&ri*ofeg_v43,57-62. 53. Wilkes M. M., Watkins W. B., Stewart R. D. and Yen S. S. C. (1980) Localiartion and qua&t&on of Bandorphin in human brain and pituitary. Neuroendocrinology 3@, 113-121. medial basal 54. Wilkes M. M. and Yen S. S. C. (1981) Augmentation by naloxone of efflux of LHRH from supetiti hypothalamus. Life Sci. 28, 2355-2359. 55. Zakarian S. and Smyth D. G. (1982) /?-Endorphin is processed differently in specific rcgbns of rat pituitary and brain. Nature 296, 250-252. (Accepted

13 May 1987)