- Email: [email protected]
Intra-hypothalamic melatonin blocks photoperiodic responsiveness in the male syrian hamster
03064522/88 $3.00+ 0.00 Pergamon Press plc 0 1988 IBRO
NeuroscienceVol. 24, No. 3, pp. 987-991, 1988 Printed in Great Britain
INTRA-HYPOTHALAMIC MELATONIN BLOCKS PHOTOPERIODIC RESPONSIVENESS IN THE MALE SYRIAN HAMSTER M. H. HASTINGS, A. P. WALKER, A. C. ROBERTAand J. HERBERT Department of Anatomy, University of Cambridge, Downing St, Cambridge CB2 3DY, U.K. Ahstraet-Exposure of male Syrian hamsters to a short daylength of 8L: 16D leads to gonadal regression. This effect of photoperiod was prevented by pinealectomy or chronic exposure of the brain to exogenous melatonin delivered from in-dwelling cannulae. However, the effect of melatonin was dependent on the neural site of application. Melatonin delivered into the mid-brain, lateral hypothalamus or amygdala was ineffective. In contrast, bilateral administration of melatonin to the medial hypothalamus prevented testicular regression and maintained high circulating levels of luteinixing hormone and prolactin. These findinas sum-rest that the medial hvnothalamus contains target sites for melatonin involved in pineal-mediated phc&eriodic responses. __
The seasonal rhythms in reproductive activity displayed by photoperiodic mammals are controlled by the pineal hormone, melatonin.2~12~1*The synthesis of melatonin is rhythmic and is precisely regulated by the circadian system. i6 Exposure to shorter daylengths leads to re-entrainment of the circadian system and a consequent increase in the duration of the nocturnal peak of melatonin synthesis and secretion.2*‘2,‘4 In the Syrian hamster this leads to a
reduction in the release of gonadotrophins (luteinizing hormone and follicle stimulating hormone, LH and FSH) and prolactin by the pituitary gland and, as a consequence, gonadal involution.‘4*2’*22It is clear that the effects of melatonin are mediated through neuronal pathways. Photoperiodic control of LH secretion is not effected by changes in pituitary responsiveness to luteinizing hormone-releasing hormone (LHRH)29 but is a consequence. of changes in the frequency of the hypothalamic circhoral oscillator which drives LHRH pulsatility.‘0J8 Furthermore, although systemic immunization against melatonin has little effect upon seasonality in sheep,’ central immunization against melatonin releases photoinhibition of gonadotrophin secretion in hamsters (Walker, unpublished observations). However, the central site(s) and mechanism of action of melatonin are completely unknown. Chronic exposure of intact hamsters to high levels of exogenous melatonin, delivered from subcutaneous implants, prevents photoinhibition of the gonadal axis,” probably because the implants obscure the phasic nature of the endogenous signal. The Abbreviations: AHA, anterior hypothalamus; AMY, amygdala; FSH, follicle-stimulating hormone; LH, luteinizing LHRH, hormone; LHA, lateral hypothalamus; luteinixing hormone-releasing hormone; MBH, mediobasal hypothalamus; MID, mid-brain; FDA, pre-optic area; RIA, radioimmunooassay; SNK, StudentNewman-Keuls test.
aim of the experiments reported here was to determine whether this blockade of the photoperiodic response could also be demonstrated following chronic intra-cerebral administration of melatonin. The effects of melatonin-filled cannulae would be expected to be greatest when located in, or near, neural sites with the greatest sensitivity to melatonin. EXPERIMENTAL PROCEDURES Cannulae The lower ends’of stainless steel cannulae (15mm length, 180um o.d.. 25 urn id.: Cooners Needle Works. Aston Lank, Birmihgha’m, U.K.) were dipped into molten melatonin (Sigma) at 150°C and left in position for 30 min to allow melatonin to fill the lumen by capillary action. The cannulae were removed and cooled to allow the melatonin to solidify. The outer surface was wiped with ethanol to remove excess melatonin and the upper end was sealed with adhesive. Thin-layer chromatographic analysis revealed that heating to 150°C did not affect melatonin. In a second experiment, cannulae were prepared from glass micropipettes with similar tip dimensions to the metal cannulae (length 12 mm, id. 25 bm). The daily release rate of melatonin from a sample of cannulae was determined by incubation in saline at 37°C. The saline was changed daily and its melatonin content was determined by radioimmunoassay (RIA). The release was high (> 3 pmole/day) for the first 6 days but then stabilized at 0.81 _+0.13 pmole/day. It was assumed that this also represented the release rate in uivo. Animals and surgery Adult male Syrian hamsters (Wrights of Essex, Chelmsford, U.K.) were housed for 4 week-under a photoschedule of 16h light. 8 h darkness (LD. 16L:8D. liahts off 17.00) in individus cages with food and water avaiable ad libithm. Animals were then anaesthetized with Avertin (tribromoethanol: isopropyl alcohol) and implanted with bilateral metal cannulae directed either at the pre-optic area (POA); 0.35 cm anterior to bregma, 0.70 cm below dural surface, f 0.05 cm from midline, incisor bar 0.50 cm above ear bar, n = 6), the anterior hypothalamus (+0.24, -0.74, f0.05, n = 1l), the mediobasal hypothalamus (+0.05, -0.80, +0.05, n = 5), the lateral hypothalamus (+0.20, -0.70, f0.25,n =6),theamygdala(+0.15, -0.90, kO.35, 987
n = 6) or the mid-brain tegmental area ( -0.18, -0.60, fO.05, n = 6). The cannulae were secured in position onto the skull with dental cement and small stainless steel screws. Five other animals were pinealectomized and a further five acted as rtnoperated controls. Following surgery, ah animals were transferred to a short phot~~ri~ of 8 h light, 16 h darkness (SD, 8L: 16D* hghts off 09.00). After 13 weeks, animals were killed by cervical dislacation during the dark phase (21X30),decapitated and the pineal glands rapidly removed and stored at -70°C prior to determination of melatonin content by RIA. The brains were fixed in 10% format-saline, sectioned at 60 pm and stained with cresyi violet for verification of cannuia placement. Testes were dissected out and their weight recorded as an index of gonadal condition. To determine the effects of central melatonin on the photoperiodic regulation of pituitary secretion, animals received melatonin-filled alass cannulae placed bilaterallv into either the anterior (G = 7) or the &diobasaE hyp&thalamus (+0.05, -0.80, SO.05, n = 6), or empty cannutae were implanted into the anterior h~~a~arnus (n = 6). After exposure to SD for 13 weeks, the animals were kihed, brains and testes retained as described abave and trunk blood collected for determination of serum LH and prolactin levels by RIA. Radioimmunwssays
Pineal melatonin content was determined with an assay previously validated for Syrian hamsters using an antiserum provided by Dr 1. Arendt, University of Surrey.~~z’ Serum iH was determined with an ovine-ovine assay using antiserum 15 orovided bv Prof. G. D. Niswender, Universitv of Coloradoli and a standard preparation of rat-LH (RPZ) provided by NIADDK. This assay has baen previously validated for measurements of hamster LELZO Serum prolactin was determined using a homolcqpus kit for hamster pro&tin provided by Dr F. Talamantes, University of Cahfornia.28 Treatment effects were determined by ANOVA on logtransformed data. DiRbrences between means were tested by a post-hoc Student-N~~an-~e~s test (SNK).
.&wriment
1
Bilateral implantation of intra-hypothalamic cannulae did not affect the nocturnal increase of melatonin content within the pineal &nci. Meana k3& in unoperat-ed control aniznaia 4h before- ~ghts-on were 1.30 + 0.03 pmolesfgland. Animals bearing empty implants in the anterior hypothalamus had with and 1.24 f 0.27 pmoles/gland animals melatonin-filled implants in the anterior hypothalamus had 1.25 f 0.19 pmoIes~~a~. The effect of melatonin and pineahxtomy upon the gonadal response to a short photoperiod varied be-
Table 2. Weight of paired testes, serum luteinizing hormone and serum prolactin titres of hamsters exposed to XL: 16D for 13 weeks Blank Group AHA MBW .---_. --~ .__--.l-” __-.-. 0.43 & 0.09 3.56 4 0.19 2.6L i-&33Testes (g) 9.19 &-0.02 0.67 J- 0.24 0.76 + 0.22 LH
tween the different treatment graups (ANOVA F= 9.08, d.f. 7,42, P < 0.01). Exposure of unoperated animals to SD for 13 weeks led to a marked decline in the weight of the testes whereas pinea’tectomy prevented this response (Table 1). Similarfy, bilateral metatonin implants placed into the anterior, preoptic or me&basal hypothalamus of pinealintact animals prevented or attenuated the degree of gonadal regression. The weights of the testes after 13 weeks in SD were equivalent to those seen in pinealectomized animals and sign~cant~y higher than those of unoperated controls (SNK, AHA t = 8.65 P < 0.01; POA t s9.74 P < 0.01; Mr3H t = 4.23 P c 0.05). In contrast, bilateral melatonin implants positioned in the mid-brain, amygdala or lateral hypothalamus had no effect on the photoperiodic response. The test&far weights of animals in these groups were not s~~i~~n~y different from those of unopcxated controls but were significantly lower than those of animals which received melatonin in the anterior hypothalamus (SNK, MID t = 7.41; AMY 8.54; LHA 6.73; P < 0.01).
In the second expetiment, the testicular response to SD was again blocked by intra-hypothalamic melatonin (ANOVA F= 64.8, d.f. 2,16, P x0.01). Animals bearing empty bilateral implants within the anterior hypothalamus showed gonadal atrophy whereas animals with rn~ato~n-~l~ cannulae in the anterior or mediobasal h~tha~am~s had Zarge testes after 13 weeks in SD (Table 2) In both experiments, the groups were initially larger, but over the course of the experiments a number of animals lost their melatonin-8lled carmulae. This was invariably foilowed within 4 we&s by testicular atrophy and the anrmals were exetuded from the final anaiysis.
Table I. Weight of paired testes(mean f S.E.M.) of hamsters expo& Group Testes MCZUI S.E.M.
PX
t3P
MID
AMY
0.52
*Z
i%
10.05
LHA 0.79
4::;
kO.34
to 8L: 16D for 13 weeks MBH 1.3t
kO.49
AHA 2.27
&O-25
POA 1.97
20.49
Animals were pineaIectomized (PX) or unoperated (UNOP) controls, or they received biIatcra1 melatonin-f&d catmulae p&itioned within the midbrain (MID), the amygdaht (AMY), the lateral (I&IA), mediobasal (MBH) or anterior (AHA) hypothalamus or the pm-optic area (POA). n = 5-11 per group.
Intra-hypothalamic melatonin, hamsters and photoperiod
989
ministered sub-cutaneously.24 In contrast to these stimulatory effects upon gonadal function, in the white-footed mouse, Peromyscus leucopus, intracerebral pellets which released about 400 pmoles of melatonin daily mimicked the effects of short daylengths and terminated ovarian cyclicity or caused testicular collapse. 6*7,9Intra-cerebral melatonin can therefore stimulate or suppress photoperiodically sensitive functions, depending on the species studied, indicating that it has a specific effect upon neuroendocrine systems rather than a generally disruptive influence on hypothalamo-pituitary function. In the experiments reported here, melatonin implants positioned in the lateral hypothalamus, amygdala or mid-brain, areas intimately involved in neuroendocrine integration, were without’effect upon the photoperiodic response. However, application of melatonin throughout the medial hypothalamus inDISCUSSION hibited the gonadal response to short photoperiods. Central administration of melatonin to P. leucopus, Chronic bilateral exposure of the hypothalamus to delivered as a pellet or as an injection, was also most exogenous melatonin did not disrupt the normal effective when applied to the medial hypothalamus, nocturnal increase in pineal melatonin content. However, the same cannulae did prevent gonadal re- particularly the suprachiasmatic and retrochiasmatic gression, a pineal-dependent phenomenon, in animals regions.” In this species, melatonin has been shown to have an effect upon gonadotrophin-releasing exposed to a short photoperiod. This effect of melneurons? but further atonin was apparent only with implants directed at hormone-immunoreactive studies are needed to define precisely the location of the medial hypothalamus but not the lateral hypomelatonin-sensitive elements within the hypothalamus, amygdala or mid-brain. Intrathalamus. If the anterior hypothalamus is a site of hypothalamic melatonin blocked the photoperiodic response of neuroendocrine systems controlling the action of melatonin, local diffusion of melatonin release of both LH and prolactin. from cannulae positioned in adjacent tissue in the The blockade of photoperiodic responsiveness in pre-optic and mediobasal hypothalamus may have pineal-intact animals receiving chronic systemic melbeen sufficiently high to obscure the detection of the atonin has been widely documented in rodents’5*24 endogenous melatonin signal by neurons within this and ungulates.” Under physiological conditions, the area. If diffusion did occur, it must have been limited brain is exposed to a phasic, circadian melatonin because cannulae placed in the lateral hypothalamus, signal.16 Reciprocal changes in the duration of the 2.00 mm away from the sensitive medial regions, were diurnal phase of low melatonin secretion and the not able to influence the photoperiodic response. nocturnal phase of high secretion provide a highNeurotoxic lesions of neurons within the medial fidelity representation of the seasonal changes in anterior hypothalamus of male and female Syrian daylength.2*3s’2Artificial exposure to chronically high hamsters were reported to block gonadal responses to levels of exogenous melatonin obscures the rhythmic photoperiod, although it is unclear whether this was nature of the endogenous pineal melatonin signal and an effect upon the generation of, or the detection of, as a result the animal becomes insensitive to ambient the pineal melatonin signal.13 In fact, the lesions may photoperiod. The neuroendocrine consequences of have affected both processes. In the present study, the this insensitivity, in terms of gonadal collapse or intra-hypothalamic implants of exogenous melatonin maintenance, vary between species and within a clearly impaired the detection of a normal endospecies at different seasons.4s7*‘5In the Syrian hamster genous melatonin signal. The content of melatonin and several other species, the neuroendocrine state within the pineal of cannulated animals was equivproduced by chronic exposure to melatonin is similar alent to that reported previously in intact animals’4~2s to that produced by pinealectomy, a treatment which and much higher than levels reported in animals with also destroys photoperiodic sensitivity.4J2J7~24 lesions of the melatonin-rhythm generating sysThe total daily secretion of melatonin in the hamtem.‘2.‘6 In pinealectomized animals in which injecster has been estimated to be approximately tions of exogenous melatonin were given to mimic the 86pmoles.26 In the present study, maintained daily melatonin profile normally experienced under short exposure of the medial hypothalamus to less than 1 photoperiods, knife cuts between the anterior and pmole was sufficient to obscure detection of this mediobasal hypothalamus made the animals insensignal. This is to be compared to the much larger sitive to melatonin and so prevented testicular colamounts (15.5 nmoles) required to block photolapse.23 Together with the present findings, the availperiodic responses in the Syrian hamster when adable data therefore support the hypothesis that a
Serum levels of LH and prolactin were also affected by melatonin treatment (LH, ANOVA F = 4.95, d.f. 2,16, P ~0.05; prolactin, ANOVA F = 13.37, d.f. 2,15, P < 0.01, Table 2). Animals with melatonin implants in the AHA had higher LH (SNK t = 4.03 P < 0.05) and prolactin titres (SNK t = 6.95 P < 0.01) than animals with empty implants. Similarly melatonin implants in the MBH maintained LH (SNK t = 3.57 P < 0.05) and prolactin (SNK t = 4.82 P < 0.01) at levels above those seen in the animals with empty implants. Although there was a tendency for melatonin implants to maintain a higher level of prolactin when implanted into the AHA rather than MBH, overall there were no significant differences in hormone titres between the two groups receiving melatonin.
990
M. H. HASTINGS et al.
population of neurons in the medial hypothalamus, probably in the anterior region, are sensitive to pineal melatonin and participate in the photoperiodic regulation of reproductive activity. In the present study, both LH and prolactin failed to respond to a short photoperiod in the presence of chronic intra-hypothalamic melatonin, an effect previously demonstrated with sub-cutaneous implants. 22yl Pituitary secretion of these two hormones is regulated by neuroendocrine systems which are to some degree separable within the hypothalamus. In both the Djungarian (Phodopus sungorus) and the Syrian hamster, long photoperiods stimulate, and short photoperiods suppress, the secretion of both hormones. However, in the Djungarian hamster photoinhibition of these two hormones is triggered at different critical daylengths Similarly, secretion of prolactin and LH in the Syrian hamster may become dissociated. In animals previously exposed to photoperiods of 8L: 16D, transfer to 12L: 12D stimulates the secretion of LH but has no effect upon pituitary release of prolactin.” It is unclear whether the neuro-
endocrine systems which control the release of LH and prolactin respond to the melatonin signal independently or whether a single neural locus detects the signal and the photoperiodic information it contains is then relayed along independent pathways controlling the different hormonal end-points. In the present study, melatonin implants applied to either the anterior or mediobasal hypothalamus were equally effective in controlling both hormonal systems. More precisely circumscrilxd application of melatonin to individual sites within the hypothalamus will be used to determine if there are anatomically separable systems within the hypomelatonin-sensitive thalamus, differentially controlling the photoperiodic response of gonadotrophins and prolactin.
work was supported by the A.F.R.C. and M.R.C. Dr J. Arendt, University of Surrey, kindly supplied the anti-melatonin serum, Dr F. Talrunantes provided the hamster prolactin RlA kit, Dr G. Niswen&r supplied anti-ovine LH serum and NlADDK provided the rat LH standard. Acknowledgements-This
EEFERJCNCES
1. Arendt J. (1986) Role of the pineal gland and melatonin in seasonal reproductive function in mammals. Oxford Rev. reprod. B&l. 8,266320. 2. Bittman E. (1984) Melatonin and photoperiodic time measurement: evidence from rodents and ruminants. In The Pineal Gland (4. Reiter R. J.). DD. 155-192. Raven Press. New York. 3. Carter ‘D. S. and Gold&& B. D. (1983) Anti-go&al e&et8 of timed melatonian infirsions in pioaalcctmniand male
Djungarian hamsters (Pkodopus sluylorrrssungorus): duration is the critical parameter. En&c&efo8y 1%3,1261-1267. 4. Carter D. S.. Hall V. D.. Tam&in L. and Goldman B. D. (1982) Pineal is mouired for testicular maintenance in the * Turkish ha&& (Mero&cetus bra&i). Eiulocrinology 111,‘8631871. 5. Duncan M. J., Goldman B. D., DiPinto M. N. and Stetson M. H. (1985) Testicular function and pelage color have different critical daylengths in the Djungarian hamster, Pho&pm smgonu snngoms. Eru&Mm&gy 116,424-W. 6. Glass J. D. and Lynch G. R. (1981) Melatonin: identification of sites of anti-gonaclal action in mouse brain. Science 214, 821-823. 7.
Glass J. D. and Lynch G. R. (1982) Evidence for a brain site of action of melatonin in the white-footed mouse, Peromyseus leucopus. Neuroena&rinology 34, 1-6
8. Glass J. D. and Lynch G. R. (1983) Diurnal rhythm of response to chronic intra-hypothafarnic melatonin in the white-footed mouse, Peromysctu leucopus. Neuroendocrino~y 35, 117-122. 9. Glass J. D. and Knotts L. K. (1987) A brain site of action for the anti-gonadal action of melatonin in the whita-footed mouse (Peromyscm leucopus): involvement of the immunoreactive G&H neuronal system. Neuromdecrinoiogy 46, 48-55. 10. Goodman R. L. and Karsch F. J. (1981) The hypothalamic pulse generator: a key determinant of nproductiveeycles in sheep In Biological Clocks in Seam& Repro&rive Cycles (eds Follett B. K. and Folktt D. E.), pp. 223-236. Wright, Bristol. 11. Hastings M. H., Herbert J., Hut&son J., Powers J. B., Steel E. A. and Walker A. P. (1987) Photopededic history controls the neuroendocrine interpretation of the nineal melatonin sianal in the male Svrian hamster. Sot. Nwosci. 17th A. Meetbzg. Abatr. 242.7. 12. Hastings M. H., Herbert J., Martensa N. D. and Roberts A. C. (1985) Annual reproductive rhythms in mammals: mechanisms of light svnchronization. Ann. N.Y. Acad Sci. 453. 182-204. 13. Hastings M. H.,ko&ts A. C. and Herbert J. (1985) Neurotc& lesions of the anterior hyP&mlamus disrupt the photoperiodic but not the circadian system of the Syrian hamster. Neuruendocrin&gy 49, 316-324. 14. Hastings M. H., Walker A. P. and Herbert J. (1987). The effect of asymm&e reductions of photoperiod on pineal mclatonin, locomotor activity and gonadal condition of male Syrian hamsters. J. tir. 114, 221-229. 15. Ho&an K. (1981)The role of the pineal gland in the Photoperiodic control of seasonal cycles in hamsters. In Biological Clocks in Seasonal Reoroductive Cycles &ds Follett B. K. and Follett D. E.). on 237-250. Wright. Bristol. 16. Klein D. C. (1979) Cir&lian rhythms in the pineal gland. In Endocrtie Rhyf/& &l. Krieger D. T.)T& 203-223. Raven Press, New York. 17. Lincoln G. A. and Ebling F. J. P. (1985) Effect of constant-release implants of melatonin on seasonal cycles in reproduction, prolaetln secretion and moulting in rams. J. Reprod. Fert. 73, 241-253. 18. Lincoln G. A. and Short R. (1980) Seasex& breeding: Nature’s contraceptive. Rec. Prog. Horn Res. 36, l-52. 19. Niswender G. D., Relchert L. E., MidgeBy A. R. and Nalbandov A. V. (1969) Radioimmunoassay for hovine and ovine luteinizing hormone. En&crino~ogy 84, 1166-I 173. 20. Pickard G. E. and Silverman A. J. (1979) Effects of photoperiod on hypothalamic luteinizing hormone-releasing hormone in the male hamster. J. Endocr. 84, I1661 173.
Intra-hypothalamic
melatonin, hamsters and photoperiod
991
21. Reiter R. J. (1983) Seasonal reproductive events related to the pineal gland. In The Pineal Glund und its Endocrine Role (eds Axelrod J., Fraschini F. and Velo J. P.), pp. 303-316. Plenum Press, New York. 22. Reiter R. J. (1983) Melatonin as the hormone which mediates the effects of the pineal gland on neuroendocrinereproductive axis of the Syrian hamster. In The Pineal Gland and its Endocrine Role (eds Axelrod J., Fraschini F. and Velo J. P.), pp. 317-33O.Plenun Press, New York. 23. Reiter R. J.. Dinh D. H.. de 10s Santos R. and Guerra J. C. (1981) Hvnothalamic cuts suaaest a brain site for the anti-gonadotrophic action of melatonin in the Syrian hamster.‘Neu~o~c~.~L~tt. 23, 315-318.24. Reiter R. J., Vaughan M. K. and Waring P. J. (1975) Studies on the minimal dosage of melatonin required to inhibit pineal antigonadotrophic activity in male golden hamsters. Horm. Res. 6, 258-267. 25. 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 beta-endorphin content and the luteinixing hormone response to naloxone in the male Syrian hamster: Endocrinology 117, 141-147. 26. Rollag M. D., Panke E. S., Trakulrungsi W., Trakulrungsi C. and Reiter R. J. (1980) Quantification of daily melatonin synthesis in the hamster pineal gland. Endocrinology 106, 231-236. 27. Skene D. J., Smith I. and Arendt J. (1986) Radio-immunoassay of pineal 5-methoxytryptophol in different species: comparison with pineal melatonin content. J. Endocr. 110, 177-184: 28. Talamantes F.. Marr G.. DiPinto M. N. and Stetson M. H. (1984) Prolactin orogles durine estrous cvcle and oreenancv in the hamster as measured by homologous RIA. Am. J. ‘Physiol. (En&& Metab. IO)-247, El28-129. s 29. Turek F. W., Alvis J. D. and Menaker M. (1977) Pituitary responsiveness to LRF in castrated male hamsters exposed to different photoperiodic conditions. Neuroendocrinology 24, 140-146.
(Accepted 8 September
1987)
NeuroscienceVol. 24, No. 3, pp. 987-991, 1988 Printed in Great Britain
INTRA-HYPOTHALAMIC MELATONIN BLOCKS PHOTOPERIODIC RESPONSIVENESS IN THE MALE SYRIAN HAMSTER M. H. HASTINGS, A. P. WALKER, A. C. ROBERTAand J. HERBERT Department of Anatomy, University of Cambridge, Downing St, Cambridge CB2 3DY, U.K. Ahstraet-Exposure of male Syrian hamsters to a short daylength of 8L: 16D leads to gonadal regression. This effect of photoperiod was prevented by pinealectomy or chronic exposure of the brain to exogenous melatonin delivered from in-dwelling cannulae. However, the effect of melatonin was dependent on the neural site of application. Melatonin delivered into the mid-brain, lateral hypothalamus or amygdala was ineffective. In contrast, bilateral administration of melatonin to the medial hypothalamus prevented testicular regression and maintained high circulating levels of luteinixing hormone and prolactin. These findinas sum-rest that the medial hvnothalamus contains target sites for melatonin involved in pineal-mediated phc&eriodic responses. __
The seasonal rhythms in reproductive activity displayed by photoperiodic mammals are controlled by the pineal hormone, melatonin.2~12~1*The synthesis of melatonin is rhythmic and is precisely regulated by the circadian system. i6 Exposure to shorter daylengths leads to re-entrainment of the circadian system and a consequent increase in the duration of the nocturnal peak of melatonin synthesis and secretion.2*‘2,‘4 In the Syrian hamster this leads to a
reduction in the release of gonadotrophins (luteinizing hormone and follicle stimulating hormone, LH and FSH) and prolactin by the pituitary gland and, as a consequence, gonadal involution.‘4*2’*22It is clear that the effects of melatonin are mediated through neuronal pathways. Photoperiodic control of LH secretion is not effected by changes in pituitary responsiveness to luteinizing hormone-releasing hormone (LHRH)29 but is a consequence. of changes in the frequency of the hypothalamic circhoral oscillator which drives LHRH pulsatility.‘0J8 Furthermore, although systemic immunization against melatonin has little effect upon seasonality in sheep,’ central immunization against melatonin releases photoinhibition of gonadotrophin secretion in hamsters (Walker, unpublished observations). However, the central site(s) and mechanism of action of melatonin are completely unknown. Chronic exposure of intact hamsters to high levels of exogenous melatonin, delivered from subcutaneous implants, prevents photoinhibition of the gonadal axis,” probably because the implants obscure the phasic nature of the endogenous signal. The Abbreviations: AHA, anterior hypothalamus; AMY, amygdala; FSH, follicle-stimulating hormone; LH, luteinizing LHRH, hormone; LHA, lateral hypothalamus; luteinixing hormone-releasing hormone; MBH, mediobasal hypothalamus; MID, mid-brain; FDA, pre-optic area; RIA, radioimmunooassay; SNK, StudentNewman-Keuls test.
aim of the experiments reported here was to determine whether this blockade of the photoperiodic response could also be demonstrated following chronic intra-cerebral administration of melatonin. The effects of melatonin-filled cannulae would be expected to be greatest when located in, or near, neural sites with the greatest sensitivity to melatonin. EXPERIMENTAL PROCEDURES Cannulae The lower ends’of stainless steel cannulae (15mm length, 180um o.d.. 25 urn id.: Cooners Needle Works. Aston Lank, Birmihgha’m, U.K.) were dipped into molten melatonin (Sigma) at 150°C and left in position for 30 min to allow melatonin to fill the lumen by capillary action. The cannulae were removed and cooled to allow the melatonin to solidify. The outer surface was wiped with ethanol to remove excess melatonin and the upper end was sealed with adhesive. Thin-layer chromatographic analysis revealed that heating to 150°C did not affect melatonin. In a second experiment, cannulae were prepared from glass micropipettes with similar tip dimensions to the metal cannulae (length 12 mm, id. 25 bm). The daily release rate of melatonin from a sample of cannulae was determined by incubation in saline at 37°C. The saline was changed daily and its melatonin content was determined by radioimmunoassay (RIA). The release was high (> 3 pmole/day) for the first 6 days but then stabilized at 0.81 _+0.13 pmole/day. It was assumed that this also represented the release rate in uivo. Animals and surgery Adult male Syrian hamsters (Wrights of Essex, Chelmsford, U.K.) were housed for 4 week-under a photoschedule of 16h light. 8 h darkness (LD. 16L:8D. liahts off 17.00) in individus cages with food and water avaiable ad libithm. Animals were then anaesthetized with Avertin (tribromoethanol: isopropyl alcohol) and implanted with bilateral metal cannulae directed either at the pre-optic area (POA); 0.35 cm anterior to bregma, 0.70 cm below dural surface, f 0.05 cm from midline, incisor bar 0.50 cm above ear bar, n = 6), the anterior hypothalamus (+0.24, -0.74, f0.05, n = 1l), the mediobasal hypothalamus (+0.05, -0.80, +0.05, n = 5), the lateral hypothalamus (+0.20, -0.70, f0.25,n =6),theamygdala(+0.15, -0.90, kO.35, 987
n = 6) or the mid-brain tegmental area ( -0.18, -0.60, fO.05, n = 6). The cannulae were secured in position onto the skull with dental cement and small stainless steel screws. Five other animals were pinealectomized and a further five acted as rtnoperated controls. Following surgery, ah animals were transferred to a short phot~~ri~ of 8 h light, 16 h darkness (SD, 8L: 16D* hghts off 09.00). After 13 weeks, animals were killed by cervical dislacation during the dark phase (21X30),decapitated and the pineal glands rapidly removed and stored at -70°C prior to determination of melatonin content by RIA. The brains were fixed in 10% format-saline, sectioned at 60 pm and stained with cresyi violet for verification of cannuia placement. Testes were dissected out and their weight recorded as an index of gonadal condition. To determine the effects of central melatonin on the photoperiodic regulation of pituitary secretion, animals received melatonin-filled alass cannulae placed bilaterallv into either the anterior (G = 7) or the &diobasaE hyp&thalamus (+0.05, -0.80, SO.05, n = 6), or empty cannutae were implanted into the anterior h~~a~arnus (n = 6). After exposure to SD for 13 weeks, the animals were kihed, brains and testes retained as described abave and trunk blood collected for determination of serum LH and prolactin levels by RIA. Radioimmunwssays
Pineal melatonin content was determined with an assay previously validated for Syrian hamsters using an antiserum provided by Dr 1. Arendt, University of Surrey.~~z’ Serum iH was determined with an ovine-ovine assay using antiserum 15 orovided bv Prof. G. D. Niswender, Universitv of Coloradoli and a standard preparation of rat-LH (RPZ) provided by NIADDK. This assay has baen previously validated for measurements of hamster LELZO Serum prolactin was determined using a homolcqpus kit for hamster pro&tin provided by Dr F. Talamantes, University of Cahfornia.28 Treatment effects were determined by ANOVA on logtransformed data. DiRbrences between means were tested by a post-hoc Student-N~~an-~e~s test (SNK).
.&wriment
1
Bilateral implantation of intra-hypothalamic cannulae did not affect the nocturnal increase of melatonin content within the pineal &nci. Meana k3& in unoperat-ed control aniznaia 4h before- ~ghts-on were 1.30 + 0.03 pmolesfgland. Animals bearing empty implants in the anterior hypothalamus had with and 1.24 f 0.27 pmoles/gland animals melatonin-filled implants in the anterior hypothalamus had 1.25 f 0.19 pmoIes~~a~. The effect of melatonin and pineahxtomy upon the gonadal response to a short photoperiod varied be-
Table 2. Weight of paired testes, serum luteinizing hormone and serum prolactin titres of hamsters exposed to XL: 16D for 13 weeks Blank Group AHA MBW .---_. --~ .__--.l-” __-.-. 0.43 & 0.09 3.56 4 0.19 2.6L i-&33Testes (g) 9.19 &-0.02 0.67 J- 0.24 0.76 + 0.22 LH
tween the different treatment graups (ANOVA F= 9.08, d.f. 7,42, P < 0.01). Exposure of unoperated animals to SD for 13 weeks led to a marked decline in the weight of the testes whereas pinea’tectomy prevented this response (Table 1). Similarfy, bilateral metatonin implants placed into the anterior, preoptic or me&basal hypothalamus of pinealintact animals prevented or attenuated the degree of gonadal regression. The weights of the testes after 13 weeks in SD were equivalent to those seen in pinealectomized animals and sign~cant~y higher than those of unoperated controls (SNK, AHA t = 8.65 P < 0.01; POA t s9.74 P < 0.01; Mr3H t = 4.23 P c 0.05). In contrast, bilateral melatonin implants positioned in the mid-brain, amygdala or lateral hypothalamus had no effect on the photoperiodic response. The test&far weights of animals in these groups were not s~~i~~n~y different from those of unopcxated controls but were significantly lower than those of animals which received melatonin in the anterior hypothalamus (SNK, MID t = 7.41; AMY 8.54; LHA 6.73; P < 0.01).
In the second expetiment, the testicular response to SD was again blocked by intra-hypothalamic melatonin (ANOVA F= 64.8, d.f. 2,16, P x0.01). Animals bearing empty bilateral implants within the anterior hypothalamus showed gonadal atrophy whereas animals with rn~ato~n-~l~ cannulae in the anterior or mediobasal h~tha~am~s had Zarge testes after 13 weeks in SD (Table 2) In both experiments, the groups were initially larger, but over the course of the experiments a number of animals lost their melatonin-8lled carmulae. This was invariably foilowed within 4 we&s by testicular atrophy and the anrmals were exetuded from the final anaiysis.
Table I. Weight of paired testes(mean f S.E.M.) of hamsters expo& Group Testes MCZUI S.E.M.
PX
t3P
MID
AMY
0.52
*Z
i%
10.05
LHA 0.79
4::;
kO.34
to 8L: 16D for 13 weeks MBH 1.3t
kO.49
AHA 2.27
&O-25
POA 1.97
20.49
Animals were pineaIectomized (PX) or unoperated (UNOP) controls, or they received biIatcra1 melatonin-f&d catmulae p&itioned within the midbrain (MID), the amygdaht (AMY), the lateral (I&IA), mediobasal (MBH) or anterior (AHA) hypothalamus or the pm-optic area (POA). n = 5-11 per group.
Intra-hypothalamic melatonin, hamsters and photoperiod
989
ministered sub-cutaneously.24 In contrast to these stimulatory effects upon gonadal function, in the white-footed mouse, Peromyscus leucopus, intracerebral pellets which released about 400 pmoles of melatonin daily mimicked the effects of short daylengths and terminated ovarian cyclicity or caused testicular collapse. 6*7,9Intra-cerebral melatonin can therefore stimulate or suppress photoperiodically sensitive functions, depending on the species studied, indicating that it has a specific effect upon neuroendocrine systems rather than a generally disruptive influence on hypothalamo-pituitary function. In the experiments reported here, melatonin implants positioned in the lateral hypothalamus, amygdala or mid-brain, areas intimately involved in neuroendocrine integration, were without’effect upon the photoperiodic response. However, application of melatonin throughout the medial hypothalamus inDISCUSSION hibited the gonadal response to short photoperiods. Central administration of melatonin to P. leucopus, Chronic bilateral exposure of the hypothalamus to delivered as a pellet or as an injection, was also most exogenous melatonin did not disrupt the normal effective when applied to the medial hypothalamus, nocturnal increase in pineal melatonin content. However, the same cannulae did prevent gonadal re- particularly the suprachiasmatic and retrochiasmatic gression, a pineal-dependent phenomenon, in animals regions.” In this species, melatonin has been shown to have an effect upon gonadotrophin-releasing exposed to a short photoperiod. This effect of melneurons? but further atonin was apparent only with implants directed at hormone-immunoreactive studies are needed to define precisely the location of the medial hypothalamus but not the lateral hypomelatonin-sensitive elements within the hypothalamus, amygdala or mid-brain. Intrathalamus. If the anterior hypothalamus is a site of hypothalamic melatonin blocked the photoperiodic response of neuroendocrine systems controlling the action of melatonin, local diffusion of melatonin release of both LH and prolactin. from cannulae positioned in adjacent tissue in the The blockade of photoperiodic responsiveness in pre-optic and mediobasal hypothalamus may have pineal-intact animals receiving chronic systemic melbeen sufficiently high to obscure the detection of the atonin has been widely documented in rodents’5*24 endogenous melatonin signal by neurons within this and ungulates.” Under physiological conditions, the area. If diffusion did occur, it must have been limited brain is exposed to a phasic, circadian melatonin because cannulae placed in the lateral hypothalamus, signal.16 Reciprocal changes in the duration of the 2.00 mm away from the sensitive medial regions, were diurnal phase of low melatonin secretion and the not able to influence the photoperiodic response. nocturnal phase of high secretion provide a highNeurotoxic lesions of neurons within the medial fidelity representation of the seasonal changes in anterior hypothalamus of male and female Syrian daylength.2*3s’2Artificial exposure to chronically high hamsters were reported to block gonadal responses to levels of exogenous melatonin obscures the rhythmic photoperiod, although it is unclear whether this was nature of the endogenous pineal melatonin signal and an effect upon the generation of, or the detection of, as a result the animal becomes insensitive to ambient the pineal melatonin signal.13 In fact, the lesions may photoperiod. The neuroendocrine consequences of have affected both processes. In the present study, the this insensitivity, in terms of gonadal collapse or intra-hypothalamic implants of exogenous melatonin maintenance, vary between species and within a clearly impaired the detection of a normal endospecies at different seasons.4s7*‘5In the Syrian hamster genous melatonin signal. The content of melatonin and several other species, the neuroendocrine state within the pineal of cannulated animals was equivproduced by chronic exposure to melatonin is similar alent to that reported previously in intact animals’4~2s to that produced by pinealectomy, a treatment which and much higher than levels reported in animals with also destroys photoperiodic sensitivity.4J2J7~24 lesions of the melatonin-rhythm generating sysThe total daily secretion of melatonin in the hamtem.‘2.‘6 In pinealectomized animals in which injecster has been estimated to be approximately tions of exogenous melatonin were given to mimic the 86pmoles.26 In the present study, maintained daily melatonin profile normally experienced under short exposure of the medial hypothalamus to less than 1 photoperiods, knife cuts between the anterior and pmole was sufficient to obscure detection of this mediobasal hypothalamus made the animals insensignal. This is to be compared to the much larger sitive to melatonin and so prevented testicular colamounts (15.5 nmoles) required to block photolapse.23 Together with the present findings, the availperiodic responses in the Syrian hamster when adable data therefore support the hypothesis that a
Serum levels of LH and prolactin were also affected by melatonin treatment (LH, ANOVA F = 4.95, d.f. 2,16, P ~0.05; prolactin, ANOVA F = 13.37, d.f. 2,15, P < 0.01, Table 2). Animals with melatonin implants in the AHA had higher LH (SNK t = 4.03 P < 0.05) and prolactin titres (SNK t = 6.95 P < 0.01) than animals with empty implants. Similarly melatonin implants in the MBH maintained LH (SNK t = 3.57 P < 0.05) and prolactin (SNK t = 4.82 P < 0.01) at levels above those seen in the animals with empty implants. Although there was a tendency for melatonin implants to maintain a higher level of prolactin when implanted into the AHA rather than MBH, overall there were no significant differences in hormone titres between the two groups receiving melatonin.
990
M. H. HASTINGS et al.
population of neurons in the medial hypothalamus, probably in the anterior region, are sensitive to pineal melatonin and participate in the photoperiodic regulation of reproductive activity. In the present study, both LH and prolactin failed to respond to a short photoperiod in the presence of chronic intra-hypothalamic melatonin, an effect previously demonstrated with sub-cutaneous implants. 22yl Pituitary secretion of these two hormones is regulated by neuroendocrine systems which are to some degree separable within the hypothalamus. In both the Djungarian (Phodopus sungorus) and the Syrian hamster, long photoperiods stimulate, and short photoperiods suppress, the secretion of both hormones. However, in the Djungarian hamster photoinhibition of these two hormones is triggered at different critical daylengths Similarly, secretion of prolactin and LH in the Syrian hamster may become dissociated. In animals previously exposed to photoperiods of 8L: 16D, transfer to 12L: 12D stimulates the secretion of LH but has no effect upon pituitary release of prolactin.” It is unclear whether the neuro-
endocrine systems which control the release of LH and prolactin respond to the melatonin signal independently or whether a single neural locus detects the signal and the photoperiodic information it contains is then relayed along independent pathways controlling the different hormonal end-points. In the present study, melatonin implants applied to either the anterior or mediobasal hypothalamus were equally effective in controlling both hormonal systems. More precisely circumscrilxd application of melatonin to individual sites within the hypothalamus will be used to determine if there are anatomically separable systems within the hypomelatonin-sensitive thalamus, differentially controlling the photoperiodic response of gonadotrophins and prolactin.
work was supported by the A.F.R.C. and M.R.C. Dr J. Arendt, University of Surrey, kindly supplied the anti-melatonin serum, Dr F. Talrunantes provided the hamster prolactin RlA kit, Dr G. Niswen&r supplied anti-ovine LH serum and NlADDK provided the rat LH standard. Acknowledgements-This
EEFERJCNCES
1. Arendt J. (1986) Role of the pineal gland and melatonin in seasonal reproductive function in mammals. Oxford Rev. reprod. B&l. 8,266320. 2. Bittman E. (1984) Melatonin and photoperiodic time measurement: evidence from rodents and ruminants. In The Pineal Gland (4. Reiter R. J.). DD. 155-192. Raven Press. New York. 3. Carter ‘D. S. and Gold&& B. D. (1983) Anti-go&al e&et8 of timed melatonian infirsions in pioaalcctmniand male
Djungarian hamsters (Pkodopus sluylorrrssungorus): duration is the critical parameter. En&c&efo8y 1%3,1261-1267. 4. Carter D. S.. Hall V. D.. Tam&in L. and Goldman B. D. (1982) Pineal is mouired for testicular maintenance in the * Turkish ha&& (Mero&cetus bra&i). Eiulocrinology 111,‘8631871. 5. Duncan M. J., Goldman B. D., DiPinto M. N. and Stetson M. H. (1985) Testicular function and pelage color have different critical daylengths in the Djungarian hamster, Pho&pm smgonu snngoms. Eru&Mm&gy 116,424-W. 6. Glass J. D. and Lynch G. R. (1981) Melatonin: identification of sites of anti-gonaclal action in mouse brain. Science 214, 821-823. 7.
Glass J. D. and Lynch G. R. (1982) Evidence for a brain site of action of melatonin in the white-footed mouse, Peromyseus leucopus. Neuroena&rinology 34, 1-6
8. Glass J. D. and Lynch G. R. (1983) Diurnal rhythm of response to chronic intra-hypothafarnic melatonin in the white-footed mouse, Peromysctu leucopus. Neuroendocrino~y 35, 117-122. 9. Glass J. D. and Knotts L. K. (1987) A brain site of action for the anti-gonadal action of melatonin in the whita-footed mouse (Peromyscm leucopus): involvement of the immunoreactive G&H neuronal system. Neuromdecrinoiogy 46, 48-55. 10. Goodman R. L. and Karsch F. J. (1981) The hypothalamic pulse generator: a key determinant of nproductiveeycles in sheep In Biological Clocks in Seam& Repro&rive Cycles (eds Follett B. K. and Folktt D. E.), pp. 223-236. Wright, Bristol. 11. Hastings M. H., Herbert J., Hut&son J., Powers J. B., Steel E. A. and Walker A. P. (1987) Photopededic history controls the neuroendocrine interpretation of the nineal melatonin sianal in the male Svrian hamster. Sot. Nwosci. 17th A. Meetbzg. Abatr. 242.7. 12. Hastings M. H., Herbert J., Martensa N. D. and Roberts A. C. (1985) Annual reproductive rhythms in mammals: mechanisms of light svnchronization. Ann. N.Y. Acad Sci. 453. 182-204. 13. Hastings M. H.,ko&ts A. C. and Herbert J. (1985) Neurotc& lesions of the anterior hyP&mlamus disrupt the photoperiodic but not the circadian system of the Syrian hamster. Neuruendocrin&gy 49, 316-324. 14. Hastings M. H., Walker A. P. and Herbert J. (1987). The effect of asymm&e reductions of photoperiod on pineal mclatonin, locomotor activity and gonadal condition of male Syrian hamsters. J. tir. 114, 221-229. 15. Ho&an K. (1981)The role of the pineal gland in the Photoperiodic control of seasonal cycles in hamsters. In Biological Clocks in Seasonal Reoroductive Cycles &ds Follett B. K. and Follett D. E.). on 237-250. Wright. Bristol. 16. Klein D. C. (1979) Cir&lian rhythms in the pineal gland. In Endocrtie Rhyf/& &l. Krieger D. T.)T& 203-223. Raven Press, New York. 17. Lincoln G. A. and Ebling F. J. P. (1985) Effect of constant-release implants of melatonin on seasonal cycles in reproduction, prolaetln secretion and moulting in rams. J. Reprod. Fert. 73, 241-253. 18. Lincoln G. A. and Short R. (1980) Seasex& breeding: Nature’s contraceptive. Rec. Prog. Horn Res. 36, l-52. 19. Niswender G. D., Relchert L. E., MidgeBy A. R. and Nalbandov A. V. (1969) Radioimmunoassay for hovine and ovine luteinizing hormone. En&crino~ogy 84, 1166-I 173. 20. Pickard G. E. and Silverman A. J. (1979) Effects of photoperiod on hypothalamic luteinizing hormone-releasing hormone in the male hamster. J. Endocr. 84, I1661 173.
Intra-hypothalamic
melatonin, hamsters and photoperiod
991
21. Reiter R. J. (1983) Seasonal reproductive events related to the pineal gland. In The Pineal Glund und its Endocrine Role (eds Axelrod J., Fraschini F. and Velo J. P.), pp. 303-316. Plenum Press, New York. 22. Reiter R. J. (1983) Melatonin as the hormone which mediates the effects of the pineal gland on neuroendocrinereproductive axis of the Syrian hamster. In The Pineal Gland and its Endocrine Role (eds Axelrod J., Fraschini F. and Velo J. P.), pp. 317-33O.Plenun Press, New York. 23. Reiter R. J.. Dinh D. H.. de 10s Santos R. and Guerra J. C. (1981) Hvnothalamic cuts suaaest a brain site for the anti-gonadotrophic action of melatonin in the Syrian hamster.‘Neu~o~c~.~L~tt. 23, 315-318.24. Reiter R. J., Vaughan M. K. and Waring P. J. (1975) Studies on the minimal dosage of melatonin required to inhibit pineal antigonadotrophic activity in male golden hamsters. Horm. Res. 6, 258-267. 25. 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 beta-endorphin content and the luteinixing hormone response to naloxone in the male Syrian hamster: Endocrinology 117, 141-147. 26. Rollag M. D., Panke E. S., Trakulrungsi W., Trakulrungsi C. and Reiter R. J. (1980) Quantification of daily melatonin synthesis in the hamster pineal gland. Endocrinology 106, 231-236. 27. Skene D. J., Smith I. and Arendt J. (1986) Radio-immunoassay of pineal 5-methoxytryptophol in different species: comparison with pineal melatonin content. J. Endocr. 110, 177-184: 28. Talamantes F.. Marr G.. DiPinto M. N. and Stetson M. H. (1984) Prolactin orogles durine estrous cvcle and oreenancv in the hamster as measured by homologous RIA. Am. J. ‘Physiol. (En&& Metab. IO)-247, El28-129. s 29. Turek F. W., Alvis J. D. and Menaker M. (1977) Pituitary responsiveness to LRF in castrated male hamsters exposed to different photoperiodic conditions. Neuroendocrinology 24, 140-146.
(Accepted 8 September
1987)