Neuroscience Letters, 146 (1992) 125-130 © 1992 Elsevier Scientific Publishers Ireland Ltd. All rights reserved 0304-3940/92/$ 05.00
125
NSL 09051
Immunohistochemical colocalization of tyrosine hydroxylase and estradiol receptors in the sheep arcuate nucleus Martine Batailler a, Dominique Blache b, Jean Thibault c and Yves Tillet a aLaboratoire de Neuroendocrinologie Sexuelle, INRA Station de Physiologie de la Reproduction des Mammifdres Domestiques, Nouzilly (France), ~Laboratoire de Comportement Animal, INRA Station de Physiologie de la Reproduction des Mammif~res Domestiques, Nouzilly (France) andCLabora toire de Biochimie Cellulaire, Coll~ge de France, Paris (France) (Received 3 June 1992; Accepted 4 August 1992)
Key words: Tyrosine hydroxylase; Oestradiol receptor; Immunohistochemistry; Hypothalamus; Arcuate nucleus; Sheep In sheep, the arcuate nucleus contains numerous tyrosine hydroxylase (TH) and estradiol receptor (rE2) immunoreactive (IR) perikarya and it has been shown previously in this species that catecholaminergic neurons can mediate the gonadal steroid action on the reproductive function. In the present study, double immunohistochemical labelling with antibodies against TH and rE2 have been used to demonstrate the presence of rE2 in TH-IR neurons in the arcuate nucleus where the distribution of TH-IR and rE2-IR neurons overlap each other. Only less than 10% of all the rE2-IR perikarya presented TH immunoreactivity. It was therefore hypothesized that either such a low number of double labelled neurons can support the effects ofestradiol in this area or that the effect of this steroid was indirect. In the latter case it might be first mediated byfl-endorphin neurons which have been previously described in this nucleus.
Central catecholamine neurons appear to mediate the effects of sexual steroids in the control of luteinizing hormone releasing hormone (LHRH) secretion in sheep (for review see ref. 30). They might constitute a direct target of the gonadal steroids since steroid treatment of castrated ewes induced modification of the catecholamine content in different brain areas [29]. Moreover, it has been demonstrated that the destruction of hypothalamic dopaminergic neurons of group A15 with 6-hydroxydopamine induced a modification of estradiol effects on luteinizing hormone (LH) pulsatility in the ewe [28]. Other studies developed in the ewe have also demonstrated the involvement of dopamine in estradiol-induced seasonal anestrus [19]. Similar involvement of catecholamines in the steroid feedback on LH secretion has also been demonstrated in rodents (for review see ref. 4). These physiological data have been confirmed by the autoradiographic evidence of the accumulation of [3H]steroids in catecholaminergic neurons in the rat brain, mainly in the hypothalamus and in the brainstem [14, 25], after injections of the steroid in the periphery.
Correspondence. Y. Tillet, Laboratoire de Neuroendocrinologie Sexuelle, INRA Station de Physiologie de la Reproduction des Mammif6res Domestiques, 37380 Nouzilly, France.
More recently, many steroid receptors have been characterized and specific monoclonal antibodies raised against these receptors have been obtained and used for immunohistochemical investigations. With this method the distribution of estradiol receptors (rE2) immunoreactive (IR) structures have been mapped first in the brain of rodents [6, 9, 18, 26, 33], opossum [10] and recently in the sheep diencephalon [5]. In all these studied species, the rE2-IR neurons were distributed in the same areas where steroid concentrating neurons have been observed. In sheep diencephalon, the most important concentration of rE2-IR neurons has been observed in the ventromedial hypothalamic area where numerous dopaminergic neurons have been previously described [31]. However, the arcuate nucleus contains not only dopamine neurons but also opiate peptides and neuropeptide Y (NPY) as demonstrated in rats [20, 21] and sheep [2] and growth hormone releasing hormone as demonstrated in rats and guinea pigs [32]. In rodents, double labelling methods have shown that rE2 might be co-localized in the same neuron with catecholamines [14, 25] or with fl-endorphin [1]. Therefore, the aim of this study was to demonstrate the putative presence of rE2 in the dopamine or non-dopamine neurons of the arcuate nucleus of sheep. Six adult (3-6 years old) 'Ile de France' ewes from the
126 laboratory flock were used. In order to control and to keep a constant level of gonadal steroid, the animals had been castrated and received an intravaginal implant of progesterone (Ovigestone, Intervet France) 15 days after the castration in order to increase the level of rE2 concentration [5]. Five days after the setting of the progesterone implant the ewes were killed by decapitation by a licensed butcher in an official slaughterhouse. Twenty minutes before sacrifice, the animals received an intravenous injection of 5,000 U of heparin. Just after decapitation the heads were perfused through both carotids with 6 liters of 1% sodium nitrite diluted in saline warmed at 37°C, and by 8 liters of fixative cooled at 4°C prepared according to Nakane [22]. However, we used 4% paraformaldehyde rather than the 2% as used by the latter author. The diencephalon was dissected and postfixed in the same fixative for 2 days and rinsed in 15% sucrose diluted in the same buffer for 3 days at +4°C. Ten-#m-thick sections were cut in the frontal plane and collected on gelatin chrome-alum treated slides [24]. They were dried at -80°C at least for 8 days before treatment. The peroxidase-antiperoxidase (PAP) method was used to visualize both tyrosine hydroxylase (TH) and estradiol receptors (rE2) on the same section. For rE2 detection, the peroxidase was revealed with a mixture of 3,3'-diaminobenzidine (DAB) and nickel ammonium sulfate according to a protocol previously described [5], while for T H detection DAB alone was used. In these conditions the former reaction was characterized by a black precipitate and the latter by a brown precipitate.
Briefly, sections were thawed at +4°C for 5 rain and rinsed twice in phosphate buffered saline (PBS) for 20 rain. Then the sections were pretreated according to a method previously described [6]. After this pretreatment, the sections were rinsed in PBS, incubated simultaneously in a mixture of rat monoclonal antiserum raised against rat rE2 (Abbott) and a rabbit polyclonal antiserum raised against rat T H [27] diluted 1/10 and 1/1,000 respectively in a mixture of 0.1% gelatin, 0.5% Triton X-100, 0.1% sodium azide in PBS, for 72 h at +4°C. After 3 rinses (20 min) in PBS, the sections were successively incubated in a mixture of sheep globulin anti-rabbit gamma globulin (1/100) and goat globulin anti-rat gamma globulin (Abbott) in PBS for 30 min at +4°C, rinsed 3 times in PBS, and incubated for 30 min in a mixture of rat PAP (Abbott). The incubation with the second antibody and the PAP was repeated twice successively. After 3 rinses in PBS and Tris-HC1 buffer (0.05 M pH 7.6) the peroxidase was revealed in a mixture of 0.02% DAB, 0.5% nickel ammonium sulfate and 0.003% H202 in Tris-HCl for 3-6 min. After the reaction of the rat PAP, the sections were rinsed twice in Tris-HC1 and twice in PBS, pretreated in 1% H202 for 10 min, and rinsed again twice in PBS before the incubation in rabbit PAP diluted 1/100 in PBS. After rinsing in PBS and TrisHC1, the peroxidase was revealed as mentioned above in the same mixture without nickel ammonium sulfate. rE2-IR neurons were numbered by computer image analysis (Biocom) on a contiguous section only processed for rE2 immunohistochemistry (3 animals) and the double labelled neurons were observed by direct exami-
VMH
o
g~ ~
o
Fig. 1. Schematic drawing of frontal sections through the ventromedial hypothalamic area from rostral (A) to caudal (F) level, illustrating the distribution ofTH-IR ('~) and rE2-1R(O) neuronson the right side and TH/rE2-1Rdouble labelledneurons (0) on the left side. AN, arcuate nucleus: EM, eminencia medialis; F, fornix; OT, optic tract: PT, pars tuberalis: V, third ventricle:VMH, ventromedialhypothalamicnucleus.
Fig. 2. Double immunohistochemical labelling in the sheep arcuate nucleus. A,B: TH-IR neurons were characterized by a brown precipitate distributed throughout the cytoplasm (thick arrow) and rE2-1R neurons were characterized by a black precipitate in the nucleus (double arrow). The double labelled neurons appeared with a black nucleus and a brown cytoplasm (arrowhead). C: low magnification of the arcuate nucleus, note that the TH-IR neurons (brown) were mainly observed in the lateral part of the nucleus whereas the rE2-IR neurons (dark) were mainly observed near the third ventricle (V). D: higher magnification of C, note that as observed on A and B, only few neurons were doubly labelled (arrowhead). Bar = 20/tm (A,B,D); 100/.tm (C).
128
nation of the treated slides through a light illuminated microscope. T H immunoreactivity (IR) appeared as a brown precipitate distributed throughout the cytoplasm of perikarya and processes, the nucleus was never stained (Fig. 2A,B). These neurons were observed in the arcuate nucleus, the periventricular area and in the retrochiasmatic area. In the arcuate nucleus T H - I R neurons were close to the third ventricle in the rostral part and more lateral in the caudal part of the nucleus (Fig. 1). The staining observed in neurons immunoreactive to anti-rE2 appeared as a black precipitate and exhibited a different labelling pattern. It was distributed mainly in the nucleus which was intensely stained (Fig. 2A,B), but it could also be observed in the cytoplasm, in this case the staining intensity was less important than in the nucleus. These neurons were found in the arcuate nucleus, in the periarcuate area and in the ventromedial part of the ventromedial nucleus (Fig. 1). The arcuate nucleus contained a high concentration of rE2-IR perikarya (5001,500 per section). Because both labellings were clearly observed and distinct from each other, double labelled neurons were easily identified as neurons exhibiting a black nucleus and a brown cytoplasm (Fig. 2). They were observed in the area where T H - I R neuron distribution overlapped that of rE2-IR neurons: the arcuate nucleus (Figs. 1 and 2C,D). However, double labelled neurons were not very numerous and were mainly distributed in the caudal half of the arcuate nucleus, in its lateral part. N o more than 10-50 double labelled neurons were observed on each section showing the largest number of double labelling, which represented less than 10% of the rE2-IR neuron population of the arcuate nucleus. Similar values were observed for the T H - I R neuron population. The distribution of rE2-IR and T H - I R neurons respectively described in our study is quite similar to previous observations in sheep [5, 31]. The distribution of double labelled neurons corresponded to the part of the arcuate nucleus which presented the most extensive overlapping between T H - I R and rE2-IR neurons. However, the percentage of double labelled neurons appeared to be low when compared to the population of T H - I R or rE2IR neurons. A possible hypothesis might be that the amount of T H or rE2 was below the threshold of immunohistochemical detection. This hypothesis might be verified by using colchicine pretreatment of the animals to increase T H immunoreactivity. However, our results are very similar to the observation of Sar [25] who has shown the same proportion of double labelling in the rat arcuate nucleus using T H immunohistochemistry and [3H]estradiol autoradiography. This low rate of double labelling could be compared to the 25% of double labelled neurons
(autoradiography for [3H]estradiol and formaldehyde induced fluorescence for monoamines) observed in the locus coeruleus complex [14]. In this latter case, the double labelled neurons are not numerous. Even if the number of TH/rE2-IR neurons is low, it seems sufficient to support the role of estradiol on catecholamine. In castrated ewes treated with an estradiol implant, physiological studies have demonstrated that the level o f homovanillic acid, a metabolite of d,opamine, is larger than in castrated ewes without implants [28]. In ovariectomized rats, large amounts of estradiol have been found to increase the turnover of dopamine in the median eminence [11] which receives dopaminergic fibers from the arcuate nucleus [12]. More recently, it has been shown that estradiol valerate treatment of ovariectomized rats induces a decrease of T H and monoamine oxydase activity in the tuberoinfundibular nucleus [15]. Such results strongly suggest that rE2-IR dopaminergic neurons of the arcuate nucleus mediate the effect of the steroid. An alternative hypothesis might be that the majority of rE2-IR neurons are short interneurons directly connected to dopaminergic neurons. Observations similar to our results have been recorded in the guinea pig arcuate nucleus where the presence of progesterone receptors (rPg) has been detected only in a few numbers of T H - I R neurons [7]. In rats, the arcuate nucleus contains not only dopamine neurons but also NPY [21], adrenocorticotropin and opiates such as fl-endorphin [20]. In sheep, the presence of fl-endorphin and NPY has also been demonstrated [2]. All these neurons are putative targets for estradiol. Moreover, double immunohistochemical labelling with anti-rPg and anti-rE2 have demonstrated the presence of both receptors in the same neurons of the arcuate nucleus of guinea pigs [33]. In the same species, a part of the rPg-IR neurons contained fl-endorphin [23]; however, the colocalization of this peptide in the rE2-IR neurons of the arcuate nucleus has not yet been demonstrated, while it has been observed in the hypothalamic ventromedial nucleus [1]. Moreover, in sheep, the control of reproduction is tightly dependent on a successive action of progesterone and estradiol [17]. According to these results the fl-endorphin neurons described in the sheep arcuate nucleus might contain rE2. This hypothesis is supported by the demonstration of a modulation of fl-endorphin concentration after estradiol treatment of female rats [13]. In the ewe such a role of estradiol on flendorphin content has been demonstrated during the preovulatory period [8] and could support our hypothesis for fl-endorphin content of T H immunonegative/rE2IR neurons in the sheep arcuate nucleus. In rodents, NPY is involved in the steroid control of L H R H secretion [16] but the demonstration of colocali-
129
zation of rE2 and NPY has never been recorded. In sheep, to the best of our knowledge, such interactions between rE2 and NPY have not been studied. Our results indicate that only less than 10% of the rE2IR neurons of the sheep arcuate nucleus are dopaminergic. Low rates of double labellings have likewise been observed in rat, i.e. 10% of rE2-TH double labelled neurons in the arcuate nucleus [25] and 25% neurotensin/rE2 double labelled neurons in the rostral preoptic area [3]. However, several data in rodents as well as in sheep indicate that the activity of dopaminergic neurons could be controlled by estradiol. Both observations might indicate that such a low number of neurons could support a physiological regulation, by a direct action of the steroid or through another neuronal system such as fl-endorphin neurons. Although functional data relevant to this hypothesis are missing, these observations provide new information on the morphological relationships between dopamine and estradiol in the sheep hypothalamus. The authors wish to thank Dr. R. Porter for the English revision of the manuscript.
1 Akesson, T.R. and Micevych, RE., Endogenous opioid-immunoreactive neurons of the ventromedial hypothalamic nucleus concentrate estrogen in male and female rats, J. Neurosci. Res., 28 (1991) 359-366. 2 Antonopoulos, J., Papadopoulos, G.C., Karamanlidis, A.N. and Michaloudi, H., Distribution of neuropeptides in the infundibular nucleus of the sheep, Neuropeptides, 14 (1989) 121 128. 3 Axelson, J.F., Shannon, W. and Van Leeuwen, F.W., Immunocytochemical localization of estrogen receptors within neurotensin cells in the rostral preoptic area of the rat hypothalamus, Neurosci. Lett., 136 (1992) 5 9. 4 Barraclough, C.A. and Wise, RM., The role of catecholamines in the regulation of pituitary luteinizing hormone and follicle stimulating hormone secretion, Endocrine Res., 3 (1982) 91-119. 5 Blache, D., Etude Neurobiologique du Comportement Sexuel Femelle et de la S6cr6tion de l'Hormone Lut6inisante chez la Brebis Ovariectomis6e: M6canismes Centraux d'Action des St6roides Ovariens, Th6se de l'Universit6 de Bordeaux II, 1991, pp. 142. 6 Blaustein, J. and Turcotte, J.C., Estrogen receptor-immunostaining of neuronal cytoplasmic processes as well as cell nuclei in guinea pig brain, Brain Res., 495 (1989) 75 82. 7 Blaustein, J.D. and Turcotte, J.C., A small population of tyrosine hydroxylase-immunoreactive neurons in the guinea-pig arcuate nucleus contains progestin receptor-immunoreactivity, J. Neuroendocrinol., 1 (1989)333-338. 8 Domanski, E., Chomicka, L.K., Ostrowska, A., Gajewska, A. and Mateusiak, K., Release of luteinizing hormone-releasing hormone, fl-endorphin and noradrenaline by the nucleus infundibularis/median eminence during periovulatory period in the sheep, Neuroendocrinology, 54 (1991) 151 158. 9 DonCarlos, L.L., Monroy, E. and Morrell, J.I., Distribution of estrogen receptor-immunoreactive cells in the forebrain of the female guinea pig, J. Comp. Neurol., 305 (1991) 591-612.
10 Fox, C.A., Ross, L.R., Handa, R.J. and Jacobson, C.D., Localization of cells containing estrogen receptor-like immunoreactivity in the Brazilian opossum brain, Brain Res., 546 (1991) 96-105. 11 Fuxe, K., H6kfelt, T. and Nilsson, O., Castration, sex hormones, and tubero-infundibular dopamine neurons, Neuroendocrinology, 5 (1969) 107 120. 12 Fuxe, K., Agnati, L.F., Kalia, M., Goldstein, M., Anderson, K. and Harfstrand, A., Dopaminergic systems in the brain and pituitary. In: Springer-Sandoz Advanced Text: Basic and Clininal Aspects of Neurosciences The Dopaminergic System, Springer, Berlin, 1985, pp. 1 25. 13 Genazzani, A.R., Trentini, G.P., Petraglia, F., De Gaetani, C.F., Criscuolo, M., Ficarra, G., De Ramundo, B.M. and Cleva, M., Estrogens modulate the circadian rhythm of hypothalamic betaendorphin contents in female rats, Neuroendocrinology, 52 (1990) 221 224. 14 Heritage, A.S., Grant, L.D. and Stumpf, W.E., 3H estradiol in catecholamine neurons of rat brainstem: combined localization by autoradiography and formaldehyde-induced fluorescence, J. Comp. Neurol., 176 (1977) 607 630. 15 Jones, E.E. and Naftolin, F., Estrogen effects on the tuberoinfundibular dopaminergic system in the female rat brain, Brain Res., 510 (1990) 84-91. 16 Kalra, S.P., Kalra, P.S., Sahu, A. and Crowley, W.R., Gonadal steroids and neurosecretion: facilitatory influence on LHRH and Neuropeptide Y, J. Steroid Biochem., 27 (1987) 677-681. 17 Karsch, F.J., Legan, S.J., Ryan, K D . and Foster, D.L., Importance of estradiol and progesterone in regulation on LH secretion and estrous behavior during the sheep estrous cycle, Biol. Reprod., 23 (1980) 404-413. 18 Koch, M. and Ehret, G., Immunocytochemical localization and quantitation of estrogen-binding cells in the male and female (virgin, pregnant, lactating) mouse brain, Brain Res., 489 (1989) 101 112. 19 Meyer, S.L. and Goodman, R.L., Neurotransmitters involved in mediating the steroid-dependent suppression of pulsatile luteinizing hormone secretion in anestrous ewes: effects of receptor antagonists, Endocrinology, 116 (1985) 2054-2061. 20 Mezey, E., Kiss, J.Z., Mueller, G.P., Eskay, R., O'Donohue, T.L. and Palkovits, M., Distribution of the pro-opiomelanocortin derived peptides, adrenocorticotrope hormone, ~-melanocyte-stimulating hormone and fl-endorphin (ACTH, ~-MSH, fl-END) in the rat hypothalamus, Brain Res., 328 (1985) 341 347. 21 Nakagawa, Y., Shiosaka, S., Emson, P.C. and Tohyama, M., Distribution of neuropeptide Y in the forebrain and diencephalon: an immunohistochemical analysis, Brain Res., 361 (1985) 52-60. 22 Nakane, P.K., Application of peroxidase-labelled antibodies to the intracellular localization of hormones, Acta Endocrinol., 67 (1971) 190 202. 23 Olster, D.H. and Blaustein, J.D., Immunocytochemical colocalization of progestin receptors and fl-endorphin or enkephalin in the hypothalamus of female guinea pigs, J. Neurobiol., 21 (1990) 768780. 24 Pappas, P.W., The use of chrome alum-gelatin (subbing) solution as a general adhesive for paraffin sections, Stain Technol., 46 (1971) 121-124. 25 Sar, M., Estradiol is concentrated in tyrosine hydroxylase-containing neurons of the hypothalamus, Science, 223 (1983) 938-940. 26 Sar, M. and Parik, I., Immunohistochemical localization of estrogen receptor in rat brain, pituitary and uterus with monoclonal antibodies, J. Steroid Biochem., 2 (1986) 497 503. 27 Thibault, J., Vidal, D. and Gros, F., In vitro translation of mRNA
130 from rat pheochromocytoma tumors; characterization of tyrosine hydroxylase, Biochem. Biophys, Res. Commun., 99 (1981) 960-968. 28 Thi6ry, J.C., Martin, G.B., Tillet, Y., Caldani, M., Quentin, M., Jamain, J. and Ravault, J.P., Role of hypothalamic catecholamines in the regulation of LH and prolactin secretions in the ewe during seasonal anestrus, Neuroendocrinology, 49 (1989) 80-87. 29 Thiery, J.C., Monoamine content of the stalk-median eminence and hypothalamus in adult female sheep as affected by daylength, J. Neuroendocrinol., 3 (1991 ) 407-4 11. 30 Thi6ry, J.C. and Martin, G.B., Neurohypophysiological control of the secretion of gonadotrophin-releasing hormone and luteinizing hormone in the sheep a review, Reprod. Fertil. Dev., 3 (1991) 137 173.
31 Tillet, Y. and Thibault, J., Catecholamine-containing neurons in the sheep brainstem and diencephalon: immunohistochemical study with tyrosine hydroxylase (TH) and dopamine-fl-hydroxylase (DBH) antibodies, J. Comp. Neurol., 290 (1989) 69-104. 32 Tramu, G., Beauvillain, J.C., Pillez, A. and Mazzuca, M., Pr6sence d'une substance immunologiquement apparent6e /~ la somatolib6rine extraite d'une tumeur pancr6atique humaine (hpGRF) dans des neurones de l'aire hypophysiotrope du cobaye et du rat, C.R. Acad. Sci. Paris, 297 (1983)435-440. 33 Warembourg, M., Jolivet, A. and Milgrom, E., lmmunohistochemical evidence of the presence of estrogen and progesterone receptors in the same neurons of the guinea pig hypothalamus and preoptic area, Brain Res., 480 (1989) 1-15.
125
NSL 09051
Immunohistochemical colocalization of tyrosine hydroxylase and estradiol receptors in the sheep arcuate nucleus Martine Batailler a, Dominique Blache b, Jean Thibault c and Yves Tillet a aLaboratoire de Neuroendocrinologie Sexuelle, INRA Station de Physiologie de la Reproduction des Mammifdres Domestiques, Nouzilly (France), ~Laboratoire de Comportement Animal, INRA Station de Physiologie de la Reproduction des Mammif~res Domestiques, Nouzilly (France) andCLabora toire de Biochimie Cellulaire, Coll~ge de France, Paris (France) (Received 3 June 1992; Accepted 4 August 1992)
Key words: Tyrosine hydroxylase; Oestradiol receptor; Immunohistochemistry; Hypothalamus; Arcuate nucleus; Sheep In sheep, the arcuate nucleus contains numerous tyrosine hydroxylase (TH) and estradiol receptor (rE2) immunoreactive (IR) perikarya and it has been shown previously in this species that catecholaminergic neurons can mediate the gonadal steroid action on the reproductive function. In the present study, double immunohistochemical labelling with antibodies against TH and rE2 have been used to demonstrate the presence of rE2 in TH-IR neurons in the arcuate nucleus where the distribution of TH-IR and rE2-IR neurons overlap each other. Only less than 10% of all the rE2-IR perikarya presented TH immunoreactivity. It was therefore hypothesized that either such a low number of double labelled neurons can support the effects ofestradiol in this area or that the effect of this steroid was indirect. In the latter case it might be first mediated byfl-endorphin neurons which have been previously described in this nucleus.
Central catecholamine neurons appear to mediate the effects of sexual steroids in the control of luteinizing hormone releasing hormone (LHRH) secretion in sheep (for review see ref. 30). They might constitute a direct target of the gonadal steroids since steroid treatment of castrated ewes induced modification of the catecholamine content in different brain areas [29]. Moreover, it has been demonstrated that the destruction of hypothalamic dopaminergic neurons of group A15 with 6-hydroxydopamine induced a modification of estradiol effects on luteinizing hormone (LH) pulsatility in the ewe [28]. Other studies developed in the ewe have also demonstrated the involvement of dopamine in estradiol-induced seasonal anestrus [19]. Similar involvement of catecholamines in the steroid feedback on LH secretion has also been demonstrated in rodents (for review see ref. 4). These physiological data have been confirmed by the autoradiographic evidence of the accumulation of [3H]steroids in catecholaminergic neurons in the rat brain, mainly in the hypothalamus and in the brainstem [14, 25], after injections of the steroid in the periphery.
Correspondence. Y. Tillet, Laboratoire de Neuroendocrinologie Sexuelle, INRA Station de Physiologie de la Reproduction des Mammif6res Domestiques, 37380 Nouzilly, France.
More recently, many steroid receptors have been characterized and specific monoclonal antibodies raised against these receptors have been obtained and used for immunohistochemical investigations. With this method the distribution of estradiol receptors (rE2) immunoreactive (IR) structures have been mapped first in the brain of rodents [6, 9, 18, 26, 33], opossum [10] and recently in the sheep diencephalon [5]. In all these studied species, the rE2-IR neurons were distributed in the same areas where steroid concentrating neurons have been observed. In sheep diencephalon, the most important concentration of rE2-IR neurons has been observed in the ventromedial hypothalamic area where numerous dopaminergic neurons have been previously described [31]. However, the arcuate nucleus contains not only dopamine neurons but also opiate peptides and neuropeptide Y (NPY) as demonstrated in rats [20, 21] and sheep [2] and growth hormone releasing hormone as demonstrated in rats and guinea pigs [32]. In rodents, double labelling methods have shown that rE2 might be co-localized in the same neuron with catecholamines [14, 25] or with fl-endorphin [1]. Therefore, the aim of this study was to demonstrate the putative presence of rE2 in the dopamine or non-dopamine neurons of the arcuate nucleus of sheep. Six adult (3-6 years old) 'Ile de France' ewes from the
126 laboratory flock were used. In order to control and to keep a constant level of gonadal steroid, the animals had been castrated and received an intravaginal implant of progesterone (Ovigestone, Intervet France) 15 days after the castration in order to increase the level of rE2 concentration [5]. Five days after the setting of the progesterone implant the ewes were killed by decapitation by a licensed butcher in an official slaughterhouse. Twenty minutes before sacrifice, the animals received an intravenous injection of 5,000 U of heparin. Just after decapitation the heads were perfused through both carotids with 6 liters of 1% sodium nitrite diluted in saline warmed at 37°C, and by 8 liters of fixative cooled at 4°C prepared according to Nakane [22]. However, we used 4% paraformaldehyde rather than the 2% as used by the latter author. The diencephalon was dissected and postfixed in the same fixative for 2 days and rinsed in 15% sucrose diluted in the same buffer for 3 days at +4°C. Ten-#m-thick sections were cut in the frontal plane and collected on gelatin chrome-alum treated slides [24]. They were dried at -80°C at least for 8 days before treatment. The peroxidase-antiperoxidase (PAP) method was used to visualize both tyrosine hydroxylase (TH) and estradiol receptors (rE2) on the same section. For rE2 detection, the peroxidase was revealed with a mixture of 3,3'-diaminobenzidine (DAB) and nickel ammonium sulfate according to a protocol previously described [5], while for T H detection DAB alone was used. In these conditions the former reaction was characterized by a black precipitate and the latter by a brown precipitate.
Briefly, sections were thawed at +4°C for 5 rain and rinsed twice in phosphate buffered saline (PBS) for 20 rain. Then the sections were pretreated according to a method previously described [6]. After this pretreatment, the sections were rinsed in PBS, incubated simultaneously in a mixture of rat monoclonal antiserum raised against rat rE2 (Abbott) and a rabbit polyclonal antiserum raised against rat T H [27] diluted 1/10 and 1/1,000 respectively in a mixture of 0.1% gelatin, 0.5% Triton X-100, 0.1% sodium azide in PBS, for 72 h at +4°C. After 3 rinses (20 min) in PBS, the sections were successively incubated in a mixture of sheep globulin anti-rabbit gamma globulin (1/100) and goat globulin anti-rat gamma globulin (Abbott) in PBS for 30 min at +4°C, rinsed 3 times in PBS, and incubated for 30 min in a mixture of rat PAP (Abbott). The incubation with the second antibody and the PAP was repeated twice successively. After 3 rinses in PBS and Tris-HC1 buffer (0.05 M pH 7.6) the peroxidase was revealed in a mixture of 0.02% DAB, 0.5% nickel ammonium sulfate and 0.003% H202 in Tris-HCl for 3-6 min. After the reaction of the rat PAP, the sections were rinsed twice in Tris-HC1 and twice in PBS, pretreated in 1% H202 for 10 min, and rinsed again twice in PBS before the incubation in rabbit PAP diluted 1/100 in PBS. After rinsing in PBS and TrisHC1, the peroxidase was revealed as mentioned above in the same mixture without nickel ammonium sulfate. rE2-IR neurons were numbered by computer image analysis (Biocom) on a contiguous section only processed for rE2 immunohistochemistry (3 animals) and the double labelled neurons were observed by direct exami-
VMH
o
g~ ~
o
Fig. 1. Schematic drawing of frontal sections through the ventromedial hypothalamic area from rostral (A) to caudal (F) level, illustrating the distribution ofTH-IR ('~) and rE2-1R(O) neuronson the right side and TH/rE2-1Rdouble labelledneurons (0) on the left side. AN, arcuate nucleus: EM, eminencia medialis; F, fornix; OT, optic tract: PT, pars tuberalis: V, third ventricle:VMH, ventromedialhypothalamicnucleus.
Fig. 2. Double immunohistochemical labelling in the sheep arcuate nucleus. A,B: TH-IR neurons were characterized by a brown precipitate distributed throughout the cytoplasm (thick arrow) and rE2-1R neurons were characterized by a black precipitate in the nucleus (double arrow). The double labelled neurons appeared with a black nucleus and a brown cytoplasm (arrowhead). C: low magnification of the arcuate nucleus, note that the TH-IR neurons (brown) were mainly observed in the lateral part of the nucleus whereas the rE2-IR neurons (dark) were mainly observed near the third ventricle (V). D: higher magnification of C, note that as observed on A and B, only few neurons were doubly labelled (arrowhead). Bar = 20/tm (A,B,D); 100/.tm (C).
128
nation of the treated slides through a light illuminated microscope. T H immunoreactivity (IR) appeared as a brown precipitate distributed throughout the cytoplasm of perikarya and processes, the nucleus was never stained (Fig. 2A,B). These neurons were observed in the arcuate nucleus, the periventricular area and in the retrochiasmatic area. In the arcuate nucleus T H - I R neurons were close to the third ventricle in the rostral part and more lateral in the caudal part of the nucleus (Fig. 1). The staining observed in neurons immunoreactive to anti-rE2 appeared as a black precipitate and exhibited a different labelling pattern. It was distributed mainly in the nucleus which was intensely stained (Fig. 2A,B), but it could also be observed in the cytoplasm, in this case the staining intensity was less important than in the nucleus. These neurons were found in the arcuate nucleus, in the periarcuate area and in the ventromedial part of the ventromedial nucleus (Fig. 1). The arcuate nucleus contained a high concentration of rE2-IR perikarya (5001,500 per section). Because both labellings were clearly observed and distinct from each other, double labelled neurons were easily identified as neurons exhibiting a black nucleus and a brown cytoplasm (Fig. 2). They were observed in the area where T H - I R neuron distribution overlapped that of rE2-IR neurons: the arcuate nucleus (Figs. 1 and 2C,D). However, double labelled neurons were not very numerous and were mainly distributed in the caudal half of the arcuate nucleus, in its lateral part. N o more than 10-50 double labelled neurons were observed on each section showing the largest number of double labelling, which represented less than 10% of the rE2-IR neuron population of the arcuate nucleus. Similar values were observed for the T H - I R neuron population. The distribution of rE2-IR and T H - I R neurons respectively described in our study is quite similar to previous observations in sheep [5, 31]. The distribution of double labelled neurons corresponded to the part of the arcuate nucleus which presented the most extensive overlapping between T H - I R and rE2-IR neurons. However, the percentage of double labelled neurons appeared to be low when compared to the population of T H - I R or rE2IR neurons. A possible hypothesis might be that the amount of T H or rE2 was below the threshold of immunohistochemical detection. This hypothesis might be verified by using colchicine pretreatment of the animals to increase T H immunoreactivity. However, our results are very similar to the observation of Sar [25] who has shown the same proportion of double labelling in the rat arcuate nucleus using T H immunohistochemistry and [3H]estradiol autoradiography. This low rate of double labelling could be compared to the 25% of double labelled neurons
(autoradiography for [3H]estradiol and formaldehyde induced fluorescence for monoamines) observed in the locus coeruleus complex [14]. In this latter case, the double labelled neurons are not numerous. Even if the number of TH/rE2-IR neurons is low, it seems sufficient to support the role of estradiol on catecholamine. In castrated ewes treated with an estradiol implant, physiological studies have demonstrated that the level o f homovanillic acid, a metabolite of d,opamine, is larger than in castrated ewes without implants [28]. In ovariectomized rats, large amounts of estradiol have been found to increase the turnover of dopamine in the median eminence [11] which receives dopaminergic fibers from the arcuate nucleus [12]. More recently, it has been shown that estradiol valerate treatment of ovariectomized rats induces a decrease of T H and monoamine oxydase activity in the tuberoinfundibular nucleus [15]. Such results strongly suggest that rE2-IR dopaminergic neurons of the arcuate nucleus mediate the effect of the steroid. An alternative hypothesis might be that the majority of rE2-IR neurons are short interneurons directly connected to dopaminergic neurons. Observations similar to our results have been recorded in the guinea pig arcuate nucleus where the presence of progesterone receptors (rPg) has been detected only in a few numbers of T H - I R neurons [7]. In rats, the arcuate nucleus contains not only dopamine neurons but also NPY [21], adrenocorticotropin and opiates such as fl-endorphin [20]. In sheep, the presence of fl-endorphin and NPY has also been demonstrated [2]. All these neurons are putative targets for estradiol. Moreover, double immunohistochemical labelling with anti-rPg and anti-rE2 have demonstrated the presence of both receptors in the same neurons of the arcuate nucleus of guinea pigs [33]. In the same species, a part of the rPg-IR neurons contained fl-endorphin [23]; however, the colocalization of this peptide in the rE2-IR neurons of the arcuate nucleus has not yet been demonstrated, while it has been observed in the hypothalamic ventromedial nucleus [1]. Moreover, in sheep, the control of reproduction is tightly dependent on a successive action of progesterone and estradiol [17]. According to these results the fl-endorphin neurons described in the sheep arcuate nucleus might contain rE2. This hypothesis is supported by the demonstration of a modulation of fl-endorphin concentration after estradiol treatment of female rats [13]. In the ewe such a role of estradiol on flendorphin content has been demonstrated during the preovulatory period [8] and could support our hypothesis for fl-endorphin content of T H immunonegative/rE2IR neurons in the sheep arcuate nucleus. In rodents, NPY is involved in the steroid control of L H R H secretion [16] but the demonstration of colocali-
129
zation of rE2 and NPY has never been recorded. In sheep, to the best of our knowledge, such interactions between rE2 and NPY have not been studied. Our results indicate that only less than 10% of the rE2IR neurons of the sheep arcuate nucleus are dopaminergic. Low rates of double labellings have likewise been observed in rat, i.e. 10% of rE2-TH double labelled neurons in the arcuate nucleus [25] and 25% neurotensin/rE2 double labelled neurons in the rostral preoptic area [3]. However, several data in rodents as well as in sheep indicate that the activity of dopaminergic neurons could be controlled by estradiol. Both observations might indicate that such a low number of neurons could support a physiological regulation, by a direct action of the steroid or through another neuronal system such as fl-endorphin neurons. Although functional data relevant to this hypothesis are missing, these observations provide new information on the morphological relationships between dopamine and estradiol in the sheep hypothalamus. The authors wish to thank Dr. R. Porter for the English revision of the manuscript.
1 Akesson, T.R. and Micevych, RE., Endogenous opioid-immunoreactive neurons of the ventromedial hypothalamic nucleus concentrate estrogen in male and female rats, J. Neurosci. Res., 28 (1991) 359-366. 2 Antonopoulos, J., Papadopoulos, G.C., Karamanlidis, A.N. and Michaloudi, H., Distribution of neuropeptides in the infundibular nucleus of the sheep, Neuropeptides, 14 (1989) 121 128. 3 Axelson, J.F., Shannon, W. and Van Leeuwen, F.W., Immunocytochemical localization of estrogen receptors within neurotensin cells in the rostral preoptic area of the rat hypothalamus, Neurosci. Lett., 136 (1992) 5 9. 4 Barraclough, C.A. and Wise, RM., The role of catecholamines in the regulation of pituitary luteinizing hormone and follicle stimulating hormone secretion, Endocrine Res., 3 (1982) 91-119. 5 Blache, D., Etude Neurobiologique du Comportement Sexuel Femelle et de la S6cr6tion de l'Hormone Lut6inisante chez la Brebis Ovariectomis6e: M6canismes Centraux d'Action des St6roides Ovariens, Th6se de l'Universit6 de Bordeaux II, 1991, pp. 142. 6 Blaustein, J. and Turcotte, J.C., Estrogen receptor-immunostaining of neuronal cytoplasmic processes as well as cell nuclei in guinea pig brain, Brain Res., 495 (1989) 75 82. 7 Blaustein, J.D. and Turcotte, J.C., A small population of tyrosine hydroxylase-immunoreactive neurons in the guinea-pig arcuate nucleus contains progestin receptor-immunoreactivity, J. Neuroendocrinol., 1 (1989)333-338. 8 Domanski, E., Chomicka, L.K., Ostrowska, A., Gajewska, A. and Mateusiak, K., Release of luteinizing hormone-releasing hormone, fl-endorphin and noradrenaline by the nucleus infundibularis/median eminence during periovulatory period in the sheep, Neuroendocrinology, 54 (1991) 151 158. 9 DonCarlos, L.L., Monroy, E. and Morrell, J.I., Distribution of estrogen receptor-immunoreactive cells in the forebrain of the female guinea pig, J. Comp. Neurol., 305 (1991) 591-612.
10 Fox, C.A., Ross, L.R., Handa, R.J. and Jacobson, C.D., Localization of cells containing estrogen receptor-like immunoreactivity in the Brazilian opossum brain, Brain Res., 546 (1991) 96-105. 11 Fuxe, K., H6kfelt, T. and Nilsson, O., Castration, sex hormones, and tubero-infundibular dopamine neurons, Neuroendocrinology, 5 (1969) 107 120. 12 Fuxe, K., Agnati, L.F., Kalia, M., Goldstein, M., Anderson, K. and Harfstrand, A., Dopaminergic systems in the brain and pituitary. In: Springer-Sandoz Advanced Text: Basic and Clininal Aspects of Neurosciences The Dopaminergic System, Springer, Berlin, 1985, pp. 1 25. 13 Genazzani, A.R., Trentini, G.P., Petraglia, F., De Gaetani, C.F., Criscuolo, M., Ficarra, G., De Ramundo, B.M. and Cleva, M., Estrogens modulate the circadian rhythm of hypothalamic betaendorphin contents in female rats, Neuroendocrinology, 52 (1990) 221 224. 14 Heritage, A.S., Grant, L.D. and Stumpf, W.E., 3H estradiol in catecholamine neurons of rat brainstem: combined localization by autoradiography and formaldehyde-induced fluorescence, J. Comp. Neurol., 176 (1977) 607 630. 15 Jones, E.E. and Naftolin, F., Estrogen effects on the tuberoinfundibular dopaminergic system in the female rat brain, Brain Res., 510 (1990) 84-91. 16 Kalra, S.P., Kalra, P.S., Sahu, A. and Crowley, W.R., Gonadal steroids and neurosecretion: facilitatory influence on LHRH and Neuropeptide Y, J. Steroid Biochem., 27 (1987) 677-681. 17 Karsch, F.J., Legan, S.J., Ryan, K D . and Foster, D.L., Importance of estradiol and progesterone in regulation on LH secretion and estrous behavior during the sheep estrous cycle, Biol. Reprod., 23 (1980) 404-413. 18 Koch, M. and Ehret, G., Immunocytochemical localization and quantitation of estrogen-binding cells in the male and female (virgin, pregnant, lactating) mouse brain, Brain Res., 489 (1989) 101 112. 19 Meyer, S.L. and Goodman, R.L., Neurotransmitters involved in mediating the steroid-dependent suppression of pulsatile luteinizing hormone secretion in anestrous ewes: effects of receptor antagonists, Endocrinology, 116 (1985) 2054-2061. 20 Mezey, E., Kiss, J.Z., Mueller, G.P., Eskay, R., O'Donohue, T.L. and Palkovits, M., Distribution of the pro-opiomelanocortin derived peptides, adrenocorticotrope hormone, ~-melanocyte-stimulating hormone and fl-endorphin (ACTH, ~-MSH, fl-END) in the rat hypothalamus, Brain Res., 328 (1985) 341 347. 21 Nakagawa, Y., Shiosaka, S., Emson, P.C. and Tohyama, M., Distribution of neuropeptide Y in the forebrain and diencephalon: an immunohistochemical analysis, Brain Res., 361 (1985) 52-60. 22 Nakane, P.K., Application of peroxidase-labelled antibodies to the intracellular localization of hormones, Acta Endocrinol., 67 (1971) 190 202. 23 Olster, D.H. and Blaustein, J.D., Immunocytochemical colocalization of progestin receptors and fl-endorphin or enkephalin in the hypothalamus of female guinea pigs, J. Neurobiol., 21 (1990) 768780. 24 Pappas, P.W., The use of chrome alum-gelatin (subbing) solution as a general adhesive for paraffin sections, Stain Technol., 46 (1971) 121-124. 25 Sar, M., Estradiol is concentrated in tyrosine hydroxylase-containing neurons of the hypothalamus, Science, 223 (1983) 938-940. 26 Sar, M. and Parik, I., Immunohistochemical localization of estrogen receptor in rat brain, pituitary and uterus with monoclonal antibodies, J. Steroid Biochem., 2 (1986) 497 503. 27 Thibault, J., Vidal, D. and Gros, F., In vitro translation of mRNA
130 from rat pheochromocytoma tumors; characterization of tyrosine hydroxylase, Biochem. Biophys, Res. Commun., 99 (1981) 960-968. 28 Thi6ry, J.C., Martin, G.B., Tillet, Y., Caldani, M., Quentin, M., Jamain, J. and Ravault, J.P., Role of hypothalamic catecholamines in the regulation of LH and prolactin secretions in the ewe during seasonal anestrus, Neuroendocrinology, 49 (1989) 80-87. 29 Thiery, J.C., Monoamine content of the stalk-median eminence and hypothalamus in adult female sheep as affected by daylength, J. Neuroendocrinol., 3 (1991 ) 407-4 11. 30 Thi6ry, J.C. and Martin, G.B., Neurohypophysiological control of the secretion of gonadotrophin-releasing hormone and luteinizing hormone in the sheep a review, Reprod. Fertil. Dev., 3 (1991) 137 173.
31 Tillet, Y. and Thibault, J., Catecholamine-containing neurons in the sheep brainstem and diencephalon: immunohistochemical study with tyrosine hydroxylase (TH) and dopamine-fl-hydroxylase (DBH) antibodies, J. Comp. Neurol., 290 (1989) 69-104. 32 Tramu, G., Beauvillain, J.C., Pillez, A. and Mazzuca, M., Pr6sence d'une substance immunologiquement apparent6e /~ la somatolib6rine extraite d'une tumeur pancr6atique humaine (hpGRF) dans des neurones de l'aire hypophysiotrope du cobaye et du rat, C.R. Acad. Sci. Paris, 297 (1983)435-440. 33 Warembourg, M., Jolivet, A. and Milgrom, E., lmmunohistochemical evidence of the presence of estrogen and progesterone receptors in the same neurons of the guinea pig hypothalamus and preoptic area, Brain Res., 480 (1989) 1-15.
