Estradiol influences oxytocin-immunoreactive brain systems

ESTRADIOL

INFLUENCES OXYTOCIN-IMMUNOREACTIVE BRAIN SYSTEMS

G. F. JIKIKOWSKI, J. D. CALDWELL,* “Dcpartmenl

Abtcilung Anatomie of Anatomy and Biological

C. A. PEDERSFN* and W. E. STUMPF*

und Zellblologie. Universitat Ulm, F.R.G. Sciences Research Center. liniversity of North Hill. NC 2751~1. U.S.A.

Carolina,

Chapel

immunocytochcmlcally for cstrogcn-dependent changes of level. Ovarlectomircd rats were treated with \ubcutaneous silastlc Implants with estradiol. or empty implants as controls for 2 days (short term treatment). Another group of rats was injected weekly for 2 months with I mg estradiol (long term. hlph dose treatment). After perfusion fixation serial Vibratome sectlons were stained with antlhodles to oxytocin. In control animals, oxytocin immunoreactive perikarya were found in the magnocellular hypothatamlc nuclei. Accessory oxytocin neurons appeared in various hypothalamic sites: immunostained neuronat processes were visible in the preoptic region, the lateral septum, the ventromedial hypothalamus .Ind the median eminence. In short term estradlot trealed animals. additional immunoreactive perikarqa could be observed m the septohypothalamic nucleus. the lateral subcommissural area. the medial preoptlc .,rea. the perifornicat region. the zona incerta and the ansa lenticularls. An increased number 01 lmmunostained fibers was found in the lateral septum. the preoptlc region. the striatum and the amygdatu. .4nimats trcatcd with high doses of cstradiot for 2 months showed ouytocln immunostaining only m the paravcntricular and supraoptic nuclei and in the median eminence. The distrihutlon of oxytocin immunorcactlve neurons in the magnocellular nuclcl did not change with changing estradiot levels. Physiological amounts of estrogen given for 2 days increased the number of oxytocincrglc neurons visible outside the classical magnocellutar nuclei while prolonged. high dose estrogen treatment diminished lmmunostaining in these oxytocinerglc sybterns. Abstract-The

rat

brain

oxytocin immunoreacttvity

was

examined

at the light microscopical

The nonapeptide oxytocin is synthesized in hypothalamic neurons. These neurons are located in the magnocellular paraventricular and supraoptic nuclei as part of the hypothalamo-neurohypophysial system. Ovytocin is secreted into the portal and systemic circulation as a posterior pituitary peptide. The importancc of oxytocin as a peripheral hormone. inducing labor and milk ejection, is well known (for review see Ref. 9). Oxytocinergic neurons occur also outside the magnoccllular nuctci. These “accessory” neurons have been described in the anterior commissural nucleus, the per-iventricular nucleus and the perifornical region.“,‘-‘ “” These cells. as well as some of the neurons within the magnocellular nuclei have central projcctIOns.“.‘h Oxytocin immunoreactive processes appear in the lateral septum. the olfactory bulbs. the lower brainstem including the medulla oblongata and the spIna cord.“‘” Oxytocin containing synapses have been described in many parts of the limbic sbstcm ’ Estradiol has multiple effects on oxytocinergic neurons. Estradiol stimulates the release of oxytocin into the bloodstream and into the cerebrospinal fluid.” The amount of oxytocin measured in different brain arcas changes with changing estrogen levels.’ Immunoreactivity of some oxytocinergic neuron5 changes after estrogen treatment.” Estradiol also induce\ axonal sprouting of magnoccllular hypothalamlc neurons in Ritz.” Estradiol enhances the tiring I ate of oxytocincrgic neurons.’

A subpopulation of hypothalamic ncurophysin neurons has nuclear receptor sites for cstradiol:“.” Stimulating effects of oxytocin on the onset of matcrnal behavior”,“.“’ and on female sexual behavior‘ arc estrogen dcpcndent. Thcsc multiple estrogen effects on oxytocin ncurons indicate that estradiol influcnccs the morphological appearance of oxytocinergic pcrikarya and processes. The aim of the present study is to invcstigatc the effects of two different estrogen conditions on the topographical distribution of immunoreactive oxyrocin neurons. F.YPF:RIMENT.ZI. PROCEI)I RES Twelve female rats (Sprague Dawteq, 200 g) ucrc ovartectomired and implanted subcutaneously uith IO mm SIIastic implants containing 0.5 mg of I7/Lestradiot propmnate (Sigma). Ovarlectomlred control animals rcccIvcc1 empty implants. Silastic tubing was obtained from DowCorning Cat. No. 603.235. Implants were soaked for I h 111 100°io ethanol and 4X h in phosphate-bulfcred saline (0.01 M phosphate butler pH 7.2 containing 0.9”~, NaCl: PBS) prior to admimstratlon. Implants of this length and description provide a blood estradiol lecet of approximatetk 100 pg ml ’ ‘The survival tnne of these ammals was 7 days (short term treatment). Another group of ovarlcctoml7cd r~ls recclvcd weekly subcutaneous injections of I mg I7[Lcstrad1ol propionate in Isotonic saline for 2 months (long term. high close treatment). An,mals were ancathctizcd with chloralhydratc dnd killed tq cardiac perfusIon with I “II paral)mmatdehyde m PBS for I5 min. Bralns were removrd and postlixcd h! tmmerslon in a”,,, paraformaldehyde overnight. Fift, jr m &al coronal or horlrontal Cc‘ctions wcrc cut \Illh :I

238

C;. F.

Vibratome (Lancer). Complete rostrocdudal series of tions were immunostained. Sections were taken from beginning of the anterior commissure till the end of median eminence, according to the atlas of Paxinos Watson.”

JIKIKOWSKI

secthe the and

Antisera to synthetic oxytocin (Sigma) were raised in rabbits by intradermal immunization with complete Freunds adjuvant and oxytocin, coupled to thyroglobulin (Sigma) with formaldehyde. Booster injections were performed with oxytocin and complete Freunds adjuvant only. Sera of the immunized rabbits were characterized with a with the best immunosolid phase assay. 29 The antiserum cytochemical properties (K23) had no measurable cross-reactivity with [Arglvasopressin, D-endorphin, [Metlenkephalin, [Leulenkephalin and substance P. The cross-reactivity with [Arglvasotocin was less than 0.05%, with isotocin and mesotocin less than 0.01% (all peptides obtained from Sigma). Antisera to oxytocin were also obtained from commercial suppliers (Biermann Comp. and Miles). Anti-oxytocin was diluted 1:2000 in PBS, containing 0.05% Triton X-100. Sections were incubated overnight at 4°C. After incubation with anti-rabbit IgG (DACO) diluted 1: 100 in PBS for 30 min at room temperature, sections were incubated with rabbit peroxidase--antiperoxidase complex (Sternberger and Meyer) diluted 1: 50 in PBS for 30 min at room temperature. Immunoprecipitates were stained with 3’,3’-diaminobenzidine and hydrogen peroxide. Immunocytochemical controls were carried out with antisera, preabsorbed with oxytocin or preimmune serum instead of anti-oxytocin. Sections were mounted onto slides, dehydrated through ascending ethanol series, coverslipped with Entellan and examined with a Zeiss photomicroscope.

RESULTS

Ovariectomized animals treated with silastic implants for 2 days survived surgery well. Animals injected with estradiol for 2 months appeared sick and lost weight. This latter group developed large pituitary tumors. All antisera to oxytocin used in the present study gave identical immunostaining. In control incubations antisera preabsorbed with oxytocin and preimmune sera showed no immunoreactivity. Oxytocin immunoreactive perikarya and processes could be found in all animals; however, there were considerable topographical differences among the

PI d.

different neurons

experimental observed

groups. were

about

All the

oxytocinergic same

size.

The

was between 15 and 20 ILrn. A distinction between magnocellular and non-magnocellular oxytocinergic cell groups could not be made. Within each experimental group, the immunocytochemical findings were consistent. The topographical distribution of oxytocinergic neurons and processes with and without estradiol treatment is given in charts of serial brain sections (Figs l&l 3). mean

perikaryal

diameter

Medial preoptic area In the medial preoptic area, scattered oxytocinergic neurons could be found in control animals. The number of these cells and their staining intensity was elevated in rats treated with estradiol implants for 2 days. In these animals neuronal processes were found extending rostrally to the olfactory bulbs and dorsally to the lateral septum. These oxytocinergic perikarya and processes could not be observed after long term estradiol treatment. Lateral subcommissural area In the dorsolateral preoptic region, ventrolateral to the anterior commissure, a group of scattered neuronal pcrikarya contained oxytocin immunoreactivity in control animals. The number of these neurons was increased after short term estradiol treatment to form a nucleus-like group of cells (Fig. la,b). Additional immunostained perikarya appeared in these animals in the septohypothalamic nucleus. In horizontal sections it became apparent that these cell groups form a continuous group of neurons (Fig. 2) projecting dorsally to the lateral septum (Fig. 7) and laterally to the striatum. The lateral subcommissural nucleus (Fig. Ic) and the septohypothalamic nucleus were devoid of oxytocin immunoreactivity after long term estradiol treatment.

Bed nucleus of the stria terminalis Only in animals treated with estradiol for 2 days did scattered oxytocinergic perikarya appear in the dorsomedial portion of the bed nucleus of the stria

Fig. 1. The lateral subcommissural nucleus (acn) of control animals (a) and short term estradiol-treated rats (b) shows differential distribution of oxytocin immunoreactive neurons and axons. The lateral subcommissural nucleus of long term estradiol treated animals (c) is free of oxytocin immunostaining (CA. commissura anterior). I : 250. Fig. 2. Horizontal section of the rat hypothalamus after 2 days of estradiol treatment: neurons of the lateral subcommissural nucleus (acn), the septohypothalamic nucleus and the perifomical region form a continuous group of oxytocinergic neurons. Oxytocinergic nerve fibers are visible around the third ventricle (arrow). 1:250. Fig. 3. Neurons of the intersupraoptico-paraventricular islands (ispi) and the supraoptic nucleus showed oxytocin like immunoreactivity in all experimental groups (CO, optic chiasma). I : 150. Fig. 4. Oxytocinergic neurons occur after short term treatment with estradiol in the lateral part of the stria terminalis (St) and rostrally of the paraventricular nucleus (arrow). I : 150. Fig. 5. Immunoreactive perikarya occurred in all animals in the paraventricular visible in the ventromedial hypothalamus. After short estradiol treatment, neurons appeared as a dorsal paraventricular cell group (arrow) and dorsally

nucleus. Processes were additional oxytocinergic of the fornix (F). 1:250.

Estradiol

influences

oxytocin-IR

brain

systems

-.*

d

_ 2

*

I

Figs

I 5

f

3

G. F

240

terminalis. These cells projections (Fig. 4).

had

dorsal

and

JIKIKOWSKI

lateral

fornix projecting laterally (Fig. 5). Only few of these in control animals. These neurons could not be found after long term high dose estradiol treatment.

cells occurred

Periventricular region of the hypothalamus

Oxytocinergic neurons could be found in all experimental groups in the anterior part of the periventricular nucleus. Immunostained cells occurred in close vicinity to the third ventricle. Their number decreased in rostrocaudal direction. In short term estradiol-treated rats (Fig. 6b) these periventricular oxytocin cells showed more cytoplasmic immunoreactivity than controls (Fig. 6a) and had intensely stained processes extending towards and along the ventricular wall (Fig. 6b). Large oxytocin-containing varicosities could be visualized in the subependymal space in these animals (Fig. 2). Rats injected with estradiol for 2 months had only weakly stained periventricular oxytocin neurons comparable to those visible in controls (Fig. 6~). Paraventricular nucleus

The distribution of the oxytocin-immunoreactive neurons in the paraventricular nucleus was similar in estradiol-treated animals and in controls. Processes of these neurons were visible in the ventromedial hypothalamus, the zona incerta, and the median eminence. In horizontal sections processes of paraventricular neurons could be observed extending caudally to the midbrain and the brainstem. The staining intensity and the number and appearance of perikarya and processes were similar in all experimental groups. Two additional clusters of oxytocinergic neurons, one dorsal and the other one rostra1 to the paraventricular nucleus could be seen only in animals treated with estradiol for 2 days. Processes of these neurons extended towards the lateral hypothalamus and to the striatum. Intersupraoptico-paraventricular

6’1 trl.

islands

Clusters of oxytocinergic neurons were found in the vicinity of blood vessels connecting the paraventricular and supraoptic nuclei. Most of their processes seemed to extend to the median eminence. The number and staining intensity of these cells was similar in all experimental groups (Fig. 3). Supraoptic nucleus

Oxytocinergic neurons were numerous in the anterior and posterior (retrochiasmatic) supraoptic nucleus in short and long term estradiol-treated animals as well as in controls (Fig. 3). Many supraoptic nerve fibers with varicosities seemed to project towards the median eminence. Only in rats treated with estradiol for 2 days did additional oxytocinergic neurons occur medially to the posterior supraoptic nucleus. Fornix

Short term estradiol-treated rats showed groups of intensely stained oxytocin neurons dorsolateral of the

Zona incerta

Numerous oxytocinergic neurons appeared in the ventral part of the zona incerta in rats treated with silastic implants of estradiol for 2 days (Fig. 8). Neuronal processes connected this group with the dorsocaudal part of the paraventricular nucleus. Immunostained fibers of these neurons extended to the amygdala. the thalamus and the nucleus ventromedialis hypothalami. Only a few of these oxytocin neurons were visible in control animals. The zona incerta of long term estradiol-treated rats did not contain oxytocin immunostaining. Ventral nucleus of the ansa lenticularis

Only in rats that had received silastic implants of estradiol for 2 days could oxytocinergic neurons be found dorsally and laterally of the optic tract in the area of the ansa lenticularis. Some cells appeared at the medial margin of the nucleus medialis amygdalae (Fig. 9). Projections extended laterally to the capsula externa and the claustrum. Many oxytocinergic nerve fibers were visible in these animals throughout the amygdala, the piriform cortex and the claustrum. The exact origin of most these processes could not be determined. DISCUSSION

The immunocytochemical examination of serial Vibratome sections, performed in the present paper, is a valuable tool for studying the topography of peptidergic brain sites. Because of the thickness of these sections it is possible to trace the site of origin and projection of immunostained processes without the use of special tracing techniques. The use of Triton X- 100 to improve penetration of the antisera provided a “Golgi-like” immunostaining.15 There was a variability in the staining intensity of immunoreactive neurons. We did not perform a densitometrical quantitation of the oxytocinergic reaction product since it is doubtful that the staining intensity of the peroxidaseantiperoxidase reaction varies in proportion to the amount of antigen present in a cell.3 Furthermore Vibratome sections may contain whole cells or just parts of cells, therefore the staining intensity of neurons may vary due to the amount of cytoplasm present in the sections. The topography of oxytocinergic neurons found in ovariectomized control animals confirms the location of oxytocin cells described in the 1iterature.22,‘a We found that estradiol treatment influences the topographical distribution of immunostained oxytocinergic neurons and processes. Oxytocin immunoreactive neurons and projections have been described before in male and cycling

Estradiol

influences

oxytocin-IR

brain

systems

Fig. 6. Neurons of the periventricular nucleus contain oxytocin immunostaining in control animals (a). In animals treated with estradiol for 2 days (b) or for 2 months (c). The intensity of immunostaining of perikarya and processes is elevated after short estradiol treatment. Oxytocinergic processes extend to the third ventricle (III). I :420. f’ig. 7. The number

of oxytocin

t’ig. 8. After short estradiol Fig. 9. Oxytocin

containing neuronal processes term estradiol treatment.

treatment extending

oxytocinergic neurons to the paraventricular

in the lateral septum 1: 500.

is increased

appear in the zona mcerta nucleus. I : 250.

immunostained neurons appear in the medial amygdala tract (TO) after 2 days of estradiol treatment.

after short

(zi) wtth processes

(am) and dorsally I :320.

of the optic

G

242

Controls

1..

JIKIKOWSKI

(‘I cd

2 days estrodiol treatment Fig. IO.

2 months estrodiol treatment

Estradiol

Controls

influences oxytocrn-IR

2 days

estradiot Fig.

I I.

brain systems

treatment

2 months

estradiol

treatment

Con-trots

2 days estradiol treatment

2 months estmdioC treatment

Estradiol

Controls

influences

oxytocin-IR

2 days estradiol

brain

treatment

systems

2 months

estradiol

treatment

Figs 10~13. Series of transverse sections of rat brains. Triangles and open triangles represent groups ol neurons and weakly stained groups of neurons immunoreactive for oxytocin. Nerve fibres are given as dotted lines. The halves of the frontal planes show immunoreactive neurons in control animals. animals treated for 2 days with silastic implants of estradiol. and rats after 2 months treatment with injections of estradiol.

246

(i. I’.

.IIRIK~WSKI

female rats in the magnocellular hypothalamic nuclei and in accessory groups.‘2~‘“~“.‘h.2R In the present paper we show new central sites of oxytocin immunostained perikarya and processes in female ovariectomized rats treated with low amounts of estradiol for 2 days. Oxytocin neurons in the medial preoptic area, the medial amygdala, the septohypothalamic nucleus, and the bed nucleus of the stria terminalis have not yet been described. In the lateral subcommissural area, the perifornical region and the zona incerta, short term estradiol treatment increased the number of visible oxytocinergic perikarya. It became evident in horizontal sections that the groups of neurons in the lateral subcommissural nucleus and in the septohypothalamic nucleus form a continuous, nucleus-like group of cells after short estrogen treatment. The immunostained oxytocin neurons influenced by estradiol were the “accessory neurons”.‘* The site of projection of some of these accessory neurons has been determined. Neurons of the commissural region project to the lateral septum,*’ magnocellular neurons of the zona incerta project to the medial forebrain bundle, the amygdala, the thalamus and to the fornix.26 Oxytocinergic synapses have been described in parts of the limbic system.’ Oxytocin neurons in the preoptic area, are in part responsible for the mediation of estrogen-dependent maternal or sexual behavior.‘.19 The observation that intraventricular injections of antiserum to oxytocin or oxytocin antagonists delayed the onset of estrogen-induced maternal behavior”.” suggests that estrogen activates some of the central oxytocinergic pathways. The estradiol implants used in the present paper provide a plasma estradiol level of 100 pg/ml plasma in rats4 which exceeds the plasma estradiol levels measured at proestrou@ and peripartum.4 The short term treatment with this amount of estradiol increased the number of oxytocin immunoreactive neurons and processes outside the magnocellular nuclei. Our findings that these neurons disappear after long term treatment with estradiol is consistent with the data of Rhodes,*’ who used silastic implants of estradiol with a plasma estrogen level of 45 pg/ml for 1 month. Long term treatment with high doses of estradiol is known to cause tumors of the anterior lobe of the pituitary. Our long term estradiol treated group of rats developed these tumors. This latter group of animals showed no oxytocin immunostaining in the lateral subcommissural nucleus, the septohypothalamic nucleus, the perifornical region, the zona incerta and the ansa lenticularis while oxytocin immunostained neurons remained unchanged in the paraventricular and supraoptic nuclei and the intersupraoptico-paraventricular islands. The finding of estradiol-dependent changes of oxytocinergic immunostaining parallels the effects of testosterone described for vasopressinergic cells.‘” It was found

(‘I td

that vasopressinergic neurons and processes III the bed nucleus of the stria terminalis disappear after castration and reappear with testosterone treatment. The oxytocin-immunoreactive neurons in the anterior commissural-septohypothalamic region may be a parallel system in female rats which is sensitive to estradiol. In the periventricular nucleus the number of neuronal processes and the intensity of perikaryal immunostaining was increased after short term estrddiot treatment as compared with controls. Long term estradiol treatment diminished oxytocin immunostaining in this region. In short term estrogen-treated rats oxytocin-immunoreactive processes were visible extending towards the third ventricle. Long stained processes and varicosities appeared in the subependymal space. It is possible that periventricular neurons are the source of oxytocin detected in the cerebrospinal fluid. Cerebrospinal fluid oxytocin levels change in parallel with endogenous estrogen levels.“~” Oxytocin neurons outside the paraventricular and supraoptic nuclei have not yet been shown to contain nuclear estrogen receptors.” Accessory neurons, however, change their immunoreactivity for oxytocin due to estrogen treatment.*’ The mechanisms underlying these estrogen effects have to remain the subject of speculation. It is possible that non-genomic estrogen actions (like membrane effects) or indirect estrogen effects (through neighboring estrogen targets) are involved. Immunocytochemistry cannot determine whether the secretory activity of neurons is stimulated or inhibited. Therefore the question has to remain open whether the increased immunostaining after short term estradiol treatment is due to an inhibition of pcptide secretion or increased synthesis of oxytocin. Similarly the disappearance of immunoreactive oxytocin after long term estradiol treatment can be due to excessive secretion as well as stopped synthesis. These questions have to be addressed to in sitzc hybridization. Oxytocinergic neurons in the paraventricular and supraoptic nuclei as well as neurons in the intersupraoptico-paraventricular islands are part of the hypothalamo-neurohypophysial system.“.“.‘” A subpopulation of these cells has nuclear receptor sites for estrddiol.‘6,24 Oxytocin secretion into the portal and systemic circulation is stimulated by estradiol.“-” It is likely that estradiol controls the hypothalamosystem oxytocinergic genoneurohypophysial Our study shows that neurons of this oxytocinergic system do not change their immunomorphological appearance after short term or long term estradiol treatment. The distribution of oxywithin the paraventricular tocinergic neurons subnuclei’ does not seem to be affected by changing estradiol levels. The exact amounts of oxytocin within neuronal perikarya cannot be quantified with the peroxidase-antiperoxidase method used in the mica]]y,lh.?l.24

Estradiol

influences

oxytocln-JR

present paper.’ Therefore we cannot determine whether the secretory activity of hypothalamoneurohypophysial oxytocin neurons was affected in our experiments. The cytoplasmatic contents of oxytocin. however, apparently remained within the range of immunocytochcmical detectability, even after long trcatmcnt with high estradiol doses. The amount of pcrikaryal oxytocin is determined by the rates ol synthesis and secretion. Therefore it is possible that a neuron responds to estrogen treatment with increased transcription and translation without an immunocytochemically detectable change of the gene product m its cytoplasm. The additional rostra1 and dorsal paraventricular oxytocincrgic cell groups. visible after short term cstradiol treatment. arc perhaps also accessory

brain

system\

not neurons. neurohypophysial

241

belonging system.

to

the

hypothalamo-

CONCI.I’SION

Our data indicate that the oxytocinergic brain topography is to a certain extent dynamic. Pcrrphural are capable of passing the steroid hormones bloodPbrain barrier and are therefore predcstinod to control this plasticity. It is possible that stcrotddependent changes of immunostaining, similar to the changes found among oxytocinergic neurons also exist for other peptidcrgic and amine@ systems.

REFKRENCES I.

2. 3. 4. 5 6. 7. 8.

9. IO.

II 12. 13. 14. IS I6 17. IX. I9

20. 21.

23.

23. 34.

Akaishi T. and Sakuma Y. (1985) Estrogen excites oxytocinergic but not vasopressincrgic cells in the paravcntrrcular nucleus of female rat hypothalamus. Bruin Res. 335, 302 305. Armstrong W. E., Warach S.. Hatton G. I. and McNeil1 T. H. (1980) Subnuclet rn the rat hypothalamrc para\cntrlcular nucleus: a cytoarchitectural, HRP. and immunocytochemical analysis. N~wo.vc;cw~c 5, I931 195X. Bigbee J. W., Kosek J. C. and Eng L. F. (1977) Effects of primary antiserum dilutwn on stainrng of “antigen-rrch“ tissues with the peroxidawantiperoxidasc technque. J. Histoc~henr. C:)~ro&w. 25, 443 447. Bridges R. S. (1984) A quantitative analysis of the roles ofdosagc. sequence and duration ofcstradrol and proge\tcrc~ne exposure in the regulation of maternal behavior in the rat. Ernlocrinol~~~r 114, 930 940. Buijs R. M. and Swaab D. F. (1979) lmmunoelectron mrcroscopical demonstration of vasopressrn and oxytocin synapses in the limbic system of the rat. Cc,// TI.\s. Rr.c. 204, 355 365 Butcher R. L.. Collins W. E. and Fugo N. W. (1974) Plasma concentration of Ll1. FSH. prolactrn. propestcrone and estradiol throughout the 4-day estrous cycle of the rat. E,l~k,c,ri,lol~,~~, 94, I704 1708. Caldwell J. D., Pederscn C. A. and Prange A. J. Jr (1986) Oxytocin facilitates the sexual receptivrty of cstrogcn treated female rats. Nrur0ppc~~tic~e.r7, I75 189. Caldwell J. D.. Greer E. R.. Johnson M. F.. Prangc A. J. Jr and Pedersen C. A. (19X6) Oxytocrn and vawprcssin ,V,wr(l~ rmmunoreactivity in hypothalamic and extrahypothalamic sates in late pregnant and postpartum rats cwlol~rinolog,r 46, 39 48. Dawood M. Y. (1984) Oxytocin. In Annuul R~wurdr Rerw~~.v (cd. Dawood M. Y ), Vol. 2. pp. I5 36. Eden I’rc\s. Montreal. DeVries G. J.. Buijs R. M and Sluiter A. A. (1984) Gonadal hormone actron\ on the morphology of vasopre\stnergrc Innervation of the adult rat brain. Brtrin Ra. 298, 141 145. Fahrbach S.. Morrell J. I. and PfalT D. W. (1985) Possrble role for endogenous owytocin rn estrogen litcrhtutcd maternal behavior in rats. ,~eurocn~locrino/~~~~ 40, 526 532. Fisher A. W. F.. Price P. G.. Burford G. D. and Lederis K. (1979) A 3-dimcnsronal rcconstructron of the hypothalamo-neurohypophysial system of the rat. Cc,// 7%~. Res. 204, 343 354. Fuchs A. R. (1985) Oxytocin in animal parturition. In O\x/ocin. Clinrd rrrrcl Idwrtrtor~ .S/rrdw.\ (ds Amoco J. .A. and Robinson A. G.). pp. 207 235. Elsevier, Amsterdam. Gerber G. B. (1979) Dcvclopment of estrddio). progesterone. prostaglandins E and F’ alpha lcvcls during pregnancy in mtcc. C‘.r. SPwr(~. Sot,. Hiol. 173, 644-649. Grzanna R., Molliver M. E. and Coyle J. T. (197X) Visualization of central norddrenergic neurons in thick section\ by the unlabeled antrbody method: a transmitter-specific Golgi image. Proc,. w(n. Ac,oc/. .%r. 1 .S..4. 75, 2502 206 Jirikowski G. F.. Caldwcll J. D.. Stumpf W. E. and Pedersen C. A. (1986) EfTccts of cstradiol on hypothalamrc oxytocmergic neurons. /6/h Mw~in~ Sot,. Ncwwrcr. 12, 376. I5 Paxrnos G and Watson C. (1982) The Ru/ Brtrfn in S/rrrorusx C‘oordintr/c\. Academic Press. Sydney Pedcrsen C. A.. Asher J. A.. Monroe Y. L. and Prange A. J. Jr (1985) Oxytocin mduces maternal behavior III wrfin female rats. S~~iwrw 216, 64X-649. Pedersen C. A. and Prangc A. J. Jr (1985) Oxytocin and mothering behavior m the rat. Phormw. 7‘11~. 28, 2X7 7Xx. Peterson R. P (1966) Magnoccllular centers in the rat hypothalamus. .I. conrp. Nrurol. 128, IX1 190. Rhodes C. H.. Morrell J. I. and Pfaff D. W. (1982) Estrogen concentrating ncurophysin containing hypothalamrc magnocellular neurons in the vasopressin deficient (Brdttleboro) rat: a study combimng steroid autoradtography :md immunohistochemistry. J. Ncwro.wi. 2, 1718 1724. Rhodes C. H.. Morrell J. I. and Pfaff D. W. (1981) lmmunohistochemical analysis of magnocchutar elcmcnts 111rat hypothalamus: distribution and numbers of cells containing neurophysin. oxytocin and vasopressin. .I. r’~~nr/~..L’cwd 198, 45 -46. Rhodes C. 11.. Morrell J. I. and Pfaff D. W. (1981) C‘hanges in oxytocrn content in the magnoccllular neurons 01. the rat hypothalamus following water deprivation or estrogen treatment. Cell Tr.\.v. Km. 216, 47 55. Sar M. and Stumpf W. E. (1980) Simultaneous locahzation of ‘H estradiol and neurophysrn I or argmin vasoprcssrn in hypothalamic neurons demonstrated by a combined technique of dry-mount autoradropraphy and rmmunocytochemistry. Neuro.vc~i Left. 17, 179 184.

248 25.

26. 27.

28. 29. 30. 31. 32.


22 Aqusr

1987)