REGULATION OF MOTONEURONAL GENE - RELATED

PEPTIDE BY IN THE RAT SPINAL NUCLEUS OF THE

BULBOCAVERNOSUS: AN IN VIVO IMMUNCCTTOCHEMICAL STUDY

by

Douglas Ashley Monks

BSc McGill University, 1995

THESIS SUBMJTTED IN PARTIAL FULFILLMENT OF

THE REQUIREMENTS FOR THE DEGREE OF

MASTER OF ARTS

in the Department

of

Pçychology

O Douglas Ashley Monks, 1998

SIMON FRASER UNIVERSITY

April, 1998

All rights reserved. This work may not be reproduced in whole or in part, by photocopy or other means, without the permission of the author National tibrary Bibliothèque nationale 1*1 of Canada du Canada Acquisitions and Acquisitions et Bibliographie Services services bibliographiques 395 Wellington Street 395, nie Wellington OttawaON K1AON4 Ottawa ON KIA ON4 Canada Canada Your fi& Voire mlermce

Our I% None réfdrence

The author has granted a non- L'auteur a accordé une licence non exclusive licence allowing the exclusive permettant à la National Library of Canada to Bibliothèque nationale du Canada de reproduce, Ioan, distrïiute or sell reproduire, prêter, distn'buer ou copies of this thesis in microfom, vendre des copies de cette thèse sous paper or electronic formats. la fome de microfiche/nlm, de reproduction sur papier ou sur format électronique.

The author retains ownership of the L'auteur conserve la propriété du copyright in this thesis. Neither the droit d'auteur qui protège cette thèse. thesis nor substantial extracts from it Ni la thèse ni des extraits substantiels may be printed or otherwise de celle-ci ne doivent être imprimés reproduced without the author's ou autrement reproduits sans son permission. autorisation. ABSTRACT

The course of rnammalian sexual differentiation is largely under the control of the principal male gonadal steroid, the androgen testosterone.

Despite the importance of androgens in the rnediation of sexual differentiation, the molecular basis of androgenic activity in the central nervous system remains poorly characterized.

Calcitonin gene-related peptide (CGRP) has previously been reported to be regulated by testosterone in the Spinal Nucleus of the Bulbocavernosus

(SNB) . The locus of androgenic regulation of CGRP expression within SNB motoneurons remains unclear, as the SNE3 is a component of a highly androgen sensitive neuromuscular circuit, with multiple possible targets for androgenic action . The present research utilizes a mosaic paradigm to test the hypothesis that the intracellular androgen receptor of SNB motoneurons regulates CGRP expression directiy in this cell population .

Females heterozygous for the testicular feminization (tfm) androgen receptor mutation were androgenized perinatally to rescue the SNB and were implanted at sixty days with silastic capsules containing testosterone or nothing . Previous work has estabfished that this method yields females with -iii- approximately half of the population of SNB motoneurons expressing functional androgen receptors . The remaining SNB motoneurons express non functional androgen receptors . This cellular mosaic of androgen receptors within the

SNB allows for celiular analysis of androgen receptor activity in this neuronal population .

Mosaic females were sacrificed six to eight weeks following implantation and spinal sections containing SNB motoneurons were immunolabeled simultaneously for CGRP and androgen receptor . Testosterone was found to decrease the proportion of CGRP immunoreactive SN6 motoneurons only in cells immunolabeled for functional androgen receptor .

This finding clearly demonstrates the necessity of functional androgen receptor in SNB motoneurons for androgenic regulation of CGRP . The absence of an effect for !estosterone on cells not immunoreactive for androgen receptor challenges rnodels of androgenic regulation of CGRP within the SNB based on retrograde influence of the SNB target muscles . The current report constitutes, to the best of my knowledge, the first unambiguous in vivo denionstration of the necessity of a steroid receptor for steroid induced alterations in gene expression in a defined cell population . ACKNOWLEDGEMENTS

This work was made possible by a gant from the National Science and

Engineering Research Council of Canada held by Dr. Neil Watson.

The following people were essential for the successful cornpletion of this

wo rk.

Dr. Neil Watson, who provided endless support and invaluable advice

and who made my graduate course both rewarding and enjoyable.

Claire Vanston, who managed the breeding colony.

Spiro Getsios, who was at once critical and stimulating, providing a substantial intellectual contribution.

Melissa Holmes, who was always there.

George Hadjipavlou who provided encouragement and expert photographic assistance .

Margaret Monks and Dr. Richard C. Monks, who have always inspired and supported me. TABLE OF CONTENTS

Title Page i Approval Page ii Abstract iii Acknowledgements v Table of Contents v i List of Tables viii List of Figures ix Introduction 1 Mamrnalian Sexual Differentiation 1 Sex steroid biosynthesis 2 Steroid distribution 4 Androgens 4 The androgen receptor 7 Regulation of androgen receptor in neurons 9 Anatomy of the androgen receptor in the CNS 10 Mediation of androgenic activity by the androgen receptor II In vivo approaches to the study of androgen receptor function II Androgen receptor and antagonists 12 Tfm mutation in rats 13 Systemic androgen manipulation 14 Microimplant approaches to localization of androgenic action 15 Androgen receptor mosaicism 16 The spinal nucleus of the bulbocavernosus as a mode1 system of androgenic action Anatomy of the SNB Sexual dimorphism of the SNB Steroid sensitivity of SNB motoneurons SNB Function Hormonal regulation of the SNI3 developmentally Hormonal regulation of the SNB in adulthood Localizing androgenic effects to SNB Motoneurons Androgen receptor mosaicism in the SNB Calcitonin gene-related peptide in the SNB Anatomy of CGRP in the spinal cord Function of CGRP in rnotoneurons Regulation of CGRP expression within SNB Motoneurons Ontogeny of CGRP in the SNB Summary Methods Ani mals Androgen manipulation in adulthood Perfusion and tissue preparation -vi- Immunocytochernistry Identification of CGRP-IR and AR-IR cells Results Analyses Interrater reliability Effects of hydroxy flutarnide dose on androgen receptor identification Degree of observed mosaicism Number of SNB cells observed Effects of cell type and testosterone exposure on CGRP immunoreactivity Discussion Locus of androgenic regulation of SNB CGRP Physiological consequences of direct regulation of CGRP expression within SNB motoneurons Androgenic regulation of CGRP in the nervous system Mechanisrn of androgen receptor regulation of CG R P expression Summary List of Abbreviations Used References LIST OF TABLES

Table 1. Method of identifying pup karyotype by phenotypic markers. 39

Table 2. Method of identification of immunolabeling of CGRP and AR and designation of cell type. 44 LIST OF FIGURES

Figure 1 A perineal dissection including the BClW muscles of an adult fernale rat having been exposed perinatally to testosterone. 21

Figure 2 A cross section of a lumbar spinal cord including SNE3 motoneurons of an adult female rat having been exposed perinatally to testosterone. 23

Figure 3. Cells rated as imrnunolabeled for CGRP and/or AR. 46

Figure 4. Effects of testosterone treatment on CGRP-IR in SNB motoneurons mosaic for androgen insensîtivity 1

INTRODUCTION

Mammalian Sexual Differentiation

Mammalian sexual differentiation is largely a function of the hormonal milieu of the

organism through development and in adulthood (MacLusky & Naftolin 1981).

While genetic differences exist between the sexes (Maxson, 1997), their role in sexual differentiation remains unclear . The earliest event in mamrnalian sexual differentiation is genetic and to date represents the only clearly dernonstrated direct genetic contribution to sexual differentiation (Arnold, 1996). An embryonic genetic signal, the sry gene product (sometirnes referred to as the testis determining factor), initiates the formation of the male , the testis (Lovell-

Badge & Hacker, 1995).

In the absence of the srygene product the genital primordium, the genital ridge, forms fernale gonads, the (Lovell-Badge & Hacker, 1995). Sfyconstitutes the minimum genetic signal for the formation of secretory testis and the development of a male phenotype (Koopman et al. 1991). In utero, the testis are secretory with little gametogenesis, while the ovaries are essentially non-secretory and highly gametogenic (Johnson & Everitt, 1995). While the female fetus is thought not to synthesize significant ievels of endocrine , it does experience significant levels of materna1 (Weisz & Ward, 1980) and fraternal androgens (Meisel & Ward 1981). As a result, female fetuses do experience a variable degree of masculinization in utero (Gladue and Clemens, 1982). The endocrine action of the testis during perinatal critical periods initiates epigenetic deviation from the female phenotype, which occurs in the absence of the testis secretory products: the androgen testosterone and Mullerian-inhibiting

hormone (Lovell-Badge & Hacker, 1995). The developmental actions of testosterone represent a class of hormone effects referred to as "organizational"

(Phoenix et al. 1959). During adufthood, gonadal steroids have been demonstrated to influence reproductive physiology and behaviour, aithough limiied by the organism's developmental history of hormone exposure; this class of hormone actions is termed "activational" (Phoenix et al. 1959) .

While the rigid dichotomization of steroid action into organizational and activational effects has been challenged (Arnold and Breedlove 1985), it remains a useful terminology. Despite the significant role androgens play in the organization and activation of sexual behaviour, the molecular mechanisrns by which androgens exert their influence on the nervous system remain unclear .

Sex Steroid Biosvnthesis

Sex steroids fall into four major classes, the androgens, the , the progestagens and the corticosteroids (Johnson & Everitt, 1995). The gonads and adrenals are differentially steroidogenic for these four classes. Steroidogenesis is regulated primarily by the differential expression of mitochondrial enzymes

(Conley & Bird 1997, Johnson & Everitt 1995).

The gonads and the adrenals constitute the major steroidogenic organs (Johnson & Everitt 1995) . Although both the male and fernale gonads synthesize progestagens, estrogens and androgens, they are differentially secretory for them. The androgen testosterone (T4) is the major endocrine factor secreted by the testis, while the 17 P (E2) and the progestagen (P4)are the major secretory product of the ovaries . The adrenals pnmarily synthesize , mineralocorticoids and the androgens (A4) and (DHEA) .

Steroidogenisis in the gonads is regulated by tropic hormones, the gonadotropins and follicle stimulating hormone . Anterior pituitary tropic hormone secretion is in turn regulated by hypothalamic releasing hormones, which are secreted into the pituitary portal blood by hypothalamic neurons traversing the neurohyphysis (Johnson & Everitt 1995) . The production and secretion of hypothalamic releasing hormones is typically subject to negative feedback control by the endproducts , such as T4 or E2 . The cellular mechanisms underlying negative feedback control of gonadotropin releasing hormone (GnRH) by the gonadal steroids is thoug ht to be indirect, as hypothalamic GnRH containing neurons are thought not to contain steroid receptors (Sagrillo et al. 1996) . Distribution

Steroids are secreted into circulation and diffuse freely across cell membranes from the blood, but rhe rate of delivery of plasma steroids to target tissues is

regulated by steroid binding proteins present in plasma, such as albumin, orosomucoid, corticosteroid-binding globulin and sex hormone- binding globulin

(Harnmond 1997). Sex hormone-binding globulin (SHBG) binds androgens with

high affinity and SHBG bound androgen constitutes the vast majority of circulating androgen in male rats (Hammond 1997) .

Androaens

The testicular androgen testosterone is the major endocrine factor mediating male sexual differentiation. Testicular function in ontogeny has been well studied in the rat (Weisz & Ward 1980) . A spike in plasma testosterone is observed in male rats on prenatal day 18 and on postnatal day 1 (Weisz and Ward, 1980), corresponding to the perinatal critical periods for the masculinization of reproductive functions in castrated males or females by the administration of exogenous testosterone (Goy, 1970). Plasma T4 remains low until the onset of puberty at which point levels rise to and plateau at adult levels (Johnson & Everitt,

1995). Testosterone secretion by the testis is pulsatile and tonic but shows both circadian and ultradian rhythmicity (Desjardins et al. 1981). Testosterone is synthesized in the Leydig cells of the testis as a consequence of

LH receptor stimulation, which is thought to result in increased transport of cholesterol across the rnitochondrial membrane via the steroid acute regulatory protein (Stocco 1997) and complexes with a testicular SHBG in Sertoli cells subsequent to secretion into the plasma (Hammond 1997) . Circulating testosterone interacts with target tissues by activating androgen receptors directly or by being converted into active androgenic or estrogenic metabolites which then interact with their respective receptors (Martini et al. 1993) .

The ovaries also synthesize and secrete the androgens testosterone, dihydrotestosterone (D HT) , Sa-androstenediol (Sa-dio1) and 3P-androstenediol

(3P-diol) in smail amounts and variably in the estrous cycle and through development, which have marked antagonistic properties on estrus du ration and receptivity in female rats (Erskine 1983). Androgen receptor function participates significantly in both sexual behaviour and reproductive capacity in female rodents

(Gladue and Clemens, 1982, Lyon & Glenister 1980) .

Although the gonads and the adrenal glands are the major source of steroid hormones, the central nervous system (CNS) has some steroidogenic capability .

Many brain sites express 5-a reductase which forms androgenic metabolites of testosterone (Martini et al. 1993) andlor cytochrome P450 aromatase, which forms estrogenic metabolites of testosterone (Lephart 1996) . Early reports indicated, counterintuitively, that many of testosterone's

masculinizing actions can be achieved with exogenous estrogen (Levine and

Mollins Jr. 1964, Sodersten 1973, Emery & Sachs 1975), although the necessity of

estrogen receptors for some testosterone activaied masculine sexual behaviour is

questionable due to the defeminized and partially mascuiinized sexual behaviour

of alpha estrogen receptor nuIl male mice (Ogawa et al. 1997, Rissman et al.

1996), as well as the possibility that exogenous estrogen acts indirectly, for

example by altering adrenal action (Gorzalka et al. 1975).

Testosterone's estrogenic metabolite E2 has both genomic and non-genomic

actions in the brain (McEwen, 1991, Wehling, 1997) . Genomically, E2 acts through the intracellular estrogen receptor, which is found extensively in brain areas containing both arornatase activity and androgen receptors (Simerly et al

1990). Cytochrome P450 aromatase has been demonstrated to be under direct androgenic regulation in al1 areas examined having estrogen synthesizing capability with the exception of the amygdala (Abdelgadir et al. 1994) .

in the CNS, testosterone's androgenic rnetabolites include DHT, testosterone's 5- a reduced metabolite, as well as A4, which is alço synthesized in the adrenal glands. DHT activates the androgen receptor and is not readily aromatized to an estrogen and as such is useful in separating testosterone's androgenic and estrogenic actions. DHT may also be metaboiized to a number of androgens, such

as 3a-diol and 3P-diol (Martini et al. 1993).

DHT is approximately 10 times more potent an activator of the androgen receptor

(see below) than is T as measured by the mamrnilary mouse turnor virus

chloramphenicol acetyl transferase (MMTV-CAT) reporter gene in vitro (Deslypere

et al. 1992). Although the MMW-CAT gene is lacking an androgen response

element, it has a potent responçe element wiih which the androgen

receptor is thought to interact (Tan et al. 1992). DHT has been widely implicated in the virilization of the male genitalia (Johnson & Everitt 1995) . The role of DHT in the CNS remains unclear, although DHT and 3a-androstenediol are more potent inhibitors of luteinizing hormone secretion in rats than is testosterone

(Martini et al. 1993).

The Androaen Receptor

While steroid hormones are endocrine factors and thus have a systemic distribution, they have discrete action on specific ce11 populations or 'Yarget tissues". Sensitivity to steroids is conferred to target tissues by the presence of appropriate receptors in their composite cells (Beato, 1989) . Steroids may also act as autocrine and paracrine factors, as some target tissues possess steroid metabolizing molecules . For example, T4 rnay be metabolized to E2 (Lephart

1996) as well as to other androgens (Martini et al. 1993). Steroid hormones may effect neural tissue both genomically, through hormone receptors or non-

genomically, by interacting with membrane bound receptors as well as with

neurotransrnitter receptors (Wehling 1997,McEwen 1991) .

The androgen receptor belongs to the steroid hormone receptor superfamily,

which also includes such intracellular receptors as the estrogen and progesterone

receptors, as well as the hormone and retinoid receptors . Steroid hormone

receptors typically have three fu nctional domains: a carboxy (COOH) terminus

ligand binding dornain, a DNA binding domain and an amino (NH3) terminus factor

interacting domain (Evans 1988) .

lntracellular steroid receptors may exist in unbound form within the cytosol or within the cell nucleus of target cells . Unbound nuclear steroid receptors rnay exist freely within the nucleus or bind DNA without altering gene transcription

(Truss and Beato, 1993) . Upon binding ligand steroid, cytosolic receptors are believed to dissociate from inhibitory complexes, translocate to the nucleus, bind

DNA and alter gene transcription (Truss and Beato 1993) . The androgen receptor is thought to be cytoplasmic when unbound and to translocate to the nucleus upon binding ligand, based on cell fractioning studies and studies with specific agonists and antagonists of the androgen receptor (Kemppainen et al. 1992) .

Activated steroid recepto rs interact wit h specific regions of DNA, termed "hormone response elements" (HRE's), and interact with the transcription initiation complexes (TIC'S), enhancing or repressing gene expression. The cascades of

gene expression resulting from steroid hormone receptor activation rnay have

phenotypic consequences of extremely long duration (McEwen, 1991) . The

specificity of HREs is not exact, and several receptor types may interact with the

same HRE's . Further, the HRE's and the TIC'S which are active Vary according to

ce11 type . The effects of activation of different steroid hormone receptors may

therefore be comparable within certain cell types and the effects of activation of

the sarne steroid hormone receptor may differ between ceIl types (Beato and

Sanchez-Pacheco, 1996).

Although some steroid receptors rnay act as monomers (Glass, 1994), the

androgen receptor binds HREs exclusively as a homodimer, requiring two ligand

activated androgen receptor molecules to cooperatively induce DNA interaction

(Wong et al. 1994). Unlike other steroid receptors, which seern to have high

affinity for a unique HRE, the androgen receptor binds both glucocorticoid

receptor response elements and progesterone receptor response elements in vitro

(Kemppainen et al. 1992) . PREs and GREs are commonly found in genes known to be under androgenic control (Tan et al. 1992) .

Recrulation of the androcien rece~torin neurons.

In hamster motoneurons, and rat brain homogenates, androgen receptor mRNA levels have been reported to be regulated through the androgen receptor although an endogenous source of estrogen appears necessary for this effect (Quarmby et al. 1990, Drengler et al. 1996b) . In motoneurons of the spinal nucleus of the bulbocavernosus (see below), an extremely androgen sensitive motoneuron pool, androgen receptor irnmunoreactivity has been reported to be upregulated by testosterone (Matsumoto et al. 1996) . Target innervation as measured by axotomy or grafting, has also been implicated in regulation of motoneuronal androgen receptor levels, with intact connections to target muscles serving to rnaintain androgen receptor levels (Yu & McGinnis 1986, Al-Shamma & Arnold

1995, Lubischer & Arnold 1995).

Anatornv of the androaen receptor in the CNS.

The anatomical distribution of androgen receptor has been cornprehensively described in the rat brain using androgen autoradiography (Sturnpf & Sar 1976) and in situ hybridization (Simerly et al. 1990) . The highest levels of androgen receptor mRNA are reported in the , notably in the preoptic area and ventromedial nucleus, as well as in interconnected structures, such as the septum, the hippocampus and the amygdala (Simerly et al. 1990) . Moderate levels of androgen receptor mRNA are found throughout the nervous system, particularly in the brainstem . The distribution of androgen receptors have also been described in the spinal cord, notably in motoneurons, which are variably sensitive to androgens (Lumbroso et al. 1996) . Typically, females exhibit a similar distribution but a less intense labelling of androgen receptor, which has been demonstrated to be due to low circulating androgens (Drengler et al. 1996a, Matsumoto et al 1996) . However, sex differences in cytosolic androgen binding in brain homogenates have been reported in long term gonadectornized rats (McGinnis & Katz 1996) .

Mediation of androaenic aciivitv bv the androaen receptor,

Androgenic actions reported in the mammalian nervous system to date have been strictly genomic in nature . Some evidence has been presented (Frye et al. 1996) that the androgen 3-a androstenediol (3a diol), which binds weakly to the androgen receptor and has behavioural effects not blocked by flutamide administration, may interact with the GABAA receptor . It seems likely that such rnechanisrns exist , although demonstrating such effects for testosterone, which binds strongly to the androgen receptor, have been elusive in neural populations . Androgen receptor mediated alterations in gene expression are thought to be cell type specific, which is to Say that it is expected that different types of androgen sensitive neurons will express different genes in response to androgen receptor activation . ln vivo approaches to the studv of androaen rece~torfunction

The systemic nature of steroid hormone action has made demonstrations of receptor function difficult. Studies of androgenic activity in vivo have been largely activational and systemic in nature, observing the alterations in gene expression associated with androgen removal through castration or administration of specific antagonists and the subsequent replacement of androgen through administration of exogenous T4 or DHT. An inherent difficulty with these approaches is identifying the site of steroid action. The CNS is vastly interconnected with a great number of target cell populations. Several demonstrations of indirect trophic actions of steroid hormones have been made, for example in the rat spinal nucleus of the bulbocavernosus (Freernan et al. 1996) . Consequently, interpretations of the locus of action of a hormone cannot be made simply by correlating the hormonal state of the organism and an alteration in a ceIl population .

Androqen receptor aqonists and antaaonists.

Both flutamide and hydroxy-flutamide (OH-F) have been extensively used as specific antagonists of the androgen receptor. Hydroxy-flutamide has been reported to bind the ligand binding domain and induce translocation of the androgen receptor into the cell nucleus without altering gene transcription

(Kemppainen et al. 1992). Flutarnide has no specific antagonist activity in vitro, but is thought to be activated in vivo by rnetabolism to OH-F (Kemppainen et al.

1992). Cyproterone acetate has also been used as an anti androgen in developmental studies ( e.g. Neumann 1966) although it has agonistic properties at the androgen receptor (Kemppainen et al. 1992). The principle androgenic metabolite of T4, DHT, has been widely used as a

specific of the androgen receptor, which it activates strongly in vitro

(Deslypere, 1992) and which has powerful virilizing effects in vivo (Martini et al.

1993). Further, DHT is not aromatized into an estrogenic compound, facilitating

interpretation of its effects (Martini et al. 1993).

Tfm mutation in rats.

The sequelae of androgen receptor failure in target tissues have been studied in ani mals expressing an androgen receptor mutation, the testicu lar feminization mutation (ffm) . Early genetic studies established the tfm mutation to be X-linked (Lyon 1970) . Homozygotes for the tfrn mutation occur only in the male offspring of carriers of the tfm mutation as these males are rendered infertile due to androgen insensitivity . Carriers for the tfm mutation have partial androgen sensitivity, with individual cells of target tissues being either completely androgen sensitive or tfrn affected (see below).

Karyotypic males expressing the tfm mutation (having the X~Ychromosornal complernent) develop secretory testis and high levels of circulating testosterone and LH (indicative of a failure of androgen receptor mediated HPG axis negative feedback control), yet express female secondary sex characteristics . The pattern of gonadotropin release (Goldman et al. 1975, Shapiro et al 1974) as well as sexuai behaviour (Olsen 1979a, 1979b) of tfm affected males is, however, defeminized and partially masculinized, an effect which is thoug ht to be mediated by central metabolism of testosterone to estrogen, to which ffm affected males are thought to be fully sensitive (Olsen and Whaien, 1982) .

The rat tfm androgen receptor results from a point mutation in the ligand binding domain (Yarborough et al. 1990) resulting in only residual androgenic activity in affected target tissues . One quantitative estimate of tfm androgen receptor activity has been made based on [3H]-T4 and [3H]-DHT binding assays of cytosolic fractions, which yielded androgenic binding 10-15% of normal (Naess et al. 1976). The extent to which this translates into functionality of tfm androgen receptors is somewhat ambiguous, although pharmacological doses of T4P or

DHT (20nglml) induced inhibition of gonadotropin secretion in tfm affected male rats (Naess et al. 1976). The reduced functionality of the rat tfm androgen receptor is attested to by the absence of peripheral masculinization of tfm affected males.

Systemic androaen mani~ulation.

Androgenized females have often been used to study the sufficiency of androgens in various aspects of sexual differentiation (Goy, 1970), including normal patterning of male-typical mounting and intromission behaviour (Sachs et al.

1973), penile formation and ejaculation (Ward, 1969) . Several proteins have been demonstrated to be regulated at the transcriptional level through action on the androgen receptor (e-g. Berger, 1989, Argente et al. 1990, Zorrilla et al. 1990) .

These studies have relied on systemic delivery of DHT, leading to ambiguity concerning androgen's site of action . There have been no unambiguous in vivo demonstrations of the genomic effects of the androgen receptor within individual neurons, in the SNB or elsewhere in the central nervous system .

Microimplant ap~roachesto localization of androqenic action.

The local adminstration of hormone through stereotaxically placed microimplants has been practiced for soma time (Smith et al. 1977) . This rnethodology relies on spatial restriction of hormone delivery without resolving the multiplicity of androgen sensitive cell types present, complicating interpretation. For example, one such study has demonstrated androgen receptor mediation of y - aminobutyric acid decarboxylase 67 expression due to microimplantation of DHT in the male rat anterior preoptic area (Grattan et al. 1996). While this provides strong evidence for a local effect of androgen receptor activation in the anterior preoptic area resulting in alteration in GABA synthesis, it does not constitute a demonstration that androgen receptors within GABAergic neurons are mediating this effect, as cellular resolution of androg en receptors and GABA synthesis is not possible employing this methodology. Androqen receptor mosaicism.

Mammalian females (XX sex chromosome complement) undergo a process of

random X inactivation embryonically (Lyon, 1961). During the embryonic

preimplantation blastocyst stage, the trophectoderm selectively inactivates

paternal X chromosmes while the inner cell mass retains activity in both X

chromosomes (Monk & Harper, 1979). The inner cell mass undergoes random X

chromosome inactivation (Lyonization) with the onset of cellular differentiation .

Lyonization is thought to be complete by the onset of gastrulation (Monk &

Harper, 1979) .

The consequence of Lyonization is a phenotypic mosaic of cells, with approximately half of al1 cells expressing materna1 genes and approximately half of al1 cells expressing paternal genes residing on the X chromosome . As the androgen receptor gene is carried on the X chromosome, fernales heterozygous for the tfm mutation are genetically mosaic for functional androgen receptors in target tissues .

Genetically mosaic tissues are useful as discriminators of locus of androgenic influence on gene expression, a problem which is particularly evident in studies of the central nervous system, where structures are vastly interconnected and effects of systemic testosterone manipulation could be attributed to androgenic influence on a number of androgen sensitive structures . A mosaic cell population allows in

vivo comparisons within individual animals, between neighboring cells differing only in activity of the androgen receptor .

Androgen receptor mosaicism has been used previously as a tool for studying androgen receptor mediated events by generating Xm X sxr mice heterozygous for both the sex reversal mutation (sxr) and the tfm mutation (Drews et al. 1988,

Breedlove 1986) . The sex reversal mutation is a translocation of the distal short arm of the Y chromosome onto an X chromosome. as a result, the genes necessary for testicular development and spermatogenesis (including sr')are expressed, and a masculine phenotype results (Cattanach et al. 1971) . These animais have secretory testis associated with varyi ng degrees of masculinization and mosaic androgen receptor expression . Breedlove (1986) used these mosaics to localize the trophic action of testosterone on SN6 motoneurons in the course of development . Drews and others (1988) used Xhm Xsxr mice to demonstrate androgen receptor mediated differences in epididirnal cell morphology as well as the presence of estrogen receptors in both wild type and ffm cells of the epididimis .

One salient difficulty with the tfm sex reversal mutation cross is the variability in endogenous androgens . The interindividual variation in testicular function complicates both qualitative and quantitative analysis . The spinal nucleus of the buIbocavernosus as a mode1 svstem of

androgenic action

The developmental actions of testosterone, being the primary gonadal androgen,

result in several central nervous system sexual dimorphisms, or morphological

differences between males and females (Breedlove 1992) . One such dimorphism

is the spinal nucleus of the bulbocavernosus (SNB), which is present in adult

males and virtually absent in adult females (Breedlove & Arnold 1980) . Sexual dimorphisms homologous to the SNI3 are present in rodents and primates, including hurnans (Onuf 1899) .

The Spinal Nucleus of the Bulbocavernosus (SNB) has a nurnber of advantages as a rnodel system for the molecular basis of androgenic action on the central nervous system : it is easily accessible, is comprised of a srnall number of large motoneurons and it has a relatively simple and well delineated function essential to sexual behaviour. Further, androgenic effects on the SNB are thought to be mediated directly through the androgen receptor as 80 - 90 % of SNB motoneurons are androgen sensitive as estimated by autoradiography (Breedlove and Arnold, 1983b) and immunocytochemistry (Freernan et al. 1995) but none accumulate radiolabeled 17-B estradiol (Breedlove and Arnold, 1983b), or express a-estrogen receptor mRNA (Simerly et al. 1990) . This pattern of steroid sensitivity (Le. very androgen sensitive but not estrogen sensitive) facilitates 19 interpretation of testosterone's mechanism of action . Figure 1.

A perineal dissection includin~the bulbocavernosus muscles of an adult fernale rat havina been exposed perinatallv to testosterone .

penis

medial bulbocavernosus

lateral bulbocavernosus

CO~OB

(levator ani muscle hidden underneath colon)

Figure 2.

Lumbar s~inalcord cross section includina SNB (arrows) from a female rat androaenized ~erinatallv. brighffield photomicrograph 20X magnification (scale bar = 500 microns)

Anatornv of the SNB.

The SN8 is a pool of motoneurons spanning lumbar segments 5 and 6 as well as

occasionally extending into sacral segment 1 of the rat spinal cord . The SNB

innervates the perineal muscles, including the base of the penis . The perineal

muscles innervated by SN6 motoneurons are: the sexually dimorphic medial and

lateral bulbocavernosus (BC) and levator ani (LA) muscles as well as the sexually monomorphic externaf anal sphincter. The BC and LA muscles are considered functionally and anatornically related and will thus be referred to as the BC/LA muscle complex (see figure 1, figure 2).

Sexual dimorphism of the SNB.

Several estimates of SNB motoneuronal number and size have been made using various rnethodulogies and with varying results (Breedlove and Arnold 1981

Sachs et al. 1985 Sengelaub et al 1989) . Generally, estimates employing stereology, an unbiased system for cell number estimation, have arrived at mean values of 160 -200 motoneurons in males, 50 motoneurons in females and 30 in tfm affected males (Breedlove & Arnold 1981 Sengelaub et al. 1989a) . As the BC muscle is absent in females and the presence of LA muscle in females is a matter of controversy- being virtually or entirely absent- SNB motoneurons in these groups are thought tu innervate the external anal sphincter, which is present in both sexes . Steroid sensitivity of SNB rnotoneurons.

Autoradiographic estimates of SNB steroid binding indicate that in the aduR male rat, roughly 95% of SNB rnotoneurons accumulate testosterone and virtually none

(5%) accumulate estrogen (Breedlove and Arnold 1980), whereas in a non dimorphic motoneuron pool, the retrodorsolateral nucleus, fewer (80%) accumulate testosterone . Androgen receptor immunoreactivity, a method of estimation which includes unbound androgen receptors, likewise finds a greater proportion of SNB motoneurons (82%) than RDLN motoneurons (59%) to be androgen sensitive (Freeman et al. 1995) .

SN6 function.

The SNB has been directly implicated in the sexual behaviour of the male rat, mediating the efficacy of ejaculation (Hart 1983) . The bulbocavernosus (BC) and levator ani (LA) have been implicated in both the formation and removal of the seminal plugs which male rats adhere to the cervix following ejaculation, as BC or

LA lesions eliminate cups (intense erections) and significantly increase the number of intromissions necessary to remove seminal plugs and significantly decrease the strength of plug adherence (Sachs 1982, Wallace and Hart 1983) .

These functions are critical to effective sperm transport, intermale sperm cornpetition, and thus male reproductive success (Wallace and Hari 1983, Hart 1983) .

Hormonal regulation of the SN9 developrnentallv.

Surprisingly, androgen does not appear to act directly on SNB motoneurons to

rescue them from prograrnmed cell death, or spoptosis, which occurs in the

absence of perinatal testosterone (Nordeen et al. 1985) . The inference that testosterone does not act directly on SNB motoneurons to rescue them from

apoptosis is drawn from a number of lines of evidence .

SN6 motoneurons do not express androgen receptors until after the critical period for this trophic action has passed (see below), whereas the BCILA muscles are androgen sensitive perinatally, excluding a direct action of the androgen receptor within SNB motoneurons, and supporting a roie for the BClLA muscles in this process (Jordan et al. 1997, Fishman et al. 1990) . Removal of supraspinal SNB afferents does not influence SNB development (Fishman & Breedlove 1985) .

Elimination of SNB motoneurons has no effect on BClLA development (Fishman &

Breedlove 1988) but removal of BClLA muscles (Kurtz et al. 1992), or local delivery of flutamide to the BC/LA (Fishman & Breedlove 1992) prevents SNB suwival . Androgenic influence on the SNB resulting indirectly from androgenic activity on the BCllA muscle is clearly demonstrated developmentally, as the SNB is spared from apoptosis by the trophic actions of testosterone on the bulbocavernosus and

levator mi muscles during a critical period bounded by embryonic day 18 and postnatal day 7 in the rat (Nordeen et al. 1985, Ward et al. 1996, Forger et al

1992, Freeman et al. 1996).

As adults, tfm affected males have little or no detectable SNB (Breedlove & Arnold

1981, Sengelaub et al. 1989a), similar to adult wild type males administered the specific androgen flutamide perinatally (Breedlove and Arnold,

1983a) . A direct action of the androgen receptor is necessary, therefore, for the sexual dimorphism present in the SNB . The sufficiency of androgen for the formation of the SNB dimorphism is dernonstrated by administering exogenous androgen to females during this critical period . Karyotypic females (XX chromosomal complement) receiving exogenous testosterone during this period develop a perineal musculature and concomitant motoneuronal innervation resembling that of males (Sacns and Thomas 1985, Ward et al. 1996, Freeman et al. 1996) .

Although perinatal administration of testosterone is sufficient for the formation of a male typical SNB, androgen receptor activation alone is insufficient for this to occur . DHT treatment alone results in an atypical pattern of organization of motoneurons innervating the BC and LA muscles, a result which may be related to the pharmacological doses of DHT used (Breedlove 1985, Sengelaub et al.,

1989b) . Similarly, perinatal administration of exogenous testosterone serves to

decrease SNB motoneuron number in intact male rats (Sachs & Thomas. 1985) .

Hormonal reaulation of the SN6 in adulthood.

Systemic alterations in plasma T4 (e-g. castration and T4 replacement) result in

alterations in SN6 motoneuronal size (Breedlove and Arnold, 1980) . Somal size

in motoneurons is thought to be an index of protein synthesis (Sato, 1994),

leading to the inference that T4 acts to increase protein synthesis in SNB

motoneurons .

Several alterations in mRNA levels resuiting from castration and reversed by

testosterone replacement have been documented in the SNB . Specifically,

m RNA levek of calcitonin gene-related peptide (Popper & Micevych 1990)

(Popper, Abelson, Micevych, 1992), 6-tubulin (Matsumoto et al.

1993) and connexin 32 (Matsumoto et al. 1991) have been identified as being

regulated by systemic testosterone manipulations . Further, calcitonin gene-

related peptide synthesis has also been reported to be regulated by systemic testosterone manipulations (Popper & Micevych 1989) . Localizing these effects of androgen to the androgen receptor within SN6 motoneurons has remained problernatic, however (Popper, Abelson, Micevych 1992, Popper, Abelson,

Micevych 1992, Popper, Ulibarri & Abelson 1992) . Localizina androaenic effects to SNE3 motoneurons.

The primary site of action of androgen's effects on SNB motoneurons cannot be

inferred with systemic manipulations of androgen activity . The SNB is but one

target for androgenic action . 60th efferent (Jordan et al, 1 997, Fishman et al.

1990, Dube et al. 1975 Popper, Ulibarri and Micevych 1992) and afferent

(Monahan & Breedlove 1991) structures are androgen sensitive and retrograde

androgenic influences frorn peripheral structures on SNB motoneurons have been well documented (Freeman et al 1996) .

Andro~enreceptor rnosaicism in the SNB.

Androgenized fernale rats heterozygous for the tfm mutation (Xlm7 Xwt ) have been generated in order to study androgen receptor mediated events in the SNB

(Freeman et al. 7996, Watson et al. 1996) . Females were androgenized with exogenous T4 during the perinatal critical periods . The substitution of exogenous

T4 for the sxrmutation employed by Breedlove (1 986) has made the rnosaic less subject to interindividual variability in circulating androgen, facilitating quantitative analysis of the SNB androgen receptor rnosaic . Androgenized carriers of the tfm mutation were found to have significantly fewer SN6 motoneurons than androgenized wild type female littermates . Further, mosaic anirnals had approxirnately half as many androgen sensitive SNB motoneurons as androgenized wild type fernale littermates (Freeman et al. 1996) .

The fact that tfm motoneurons persist in the SNB in adulthood provides a direct demonstration that androgenic regulation of SNB motoneuron apoptosis occurs outside of the SN8 . When taken in the context of evidence implicating the muscle but not descending afferents ta the SNB in the regulation of SN8 motoneuron apoptosis, this observation provides compeHing evidence limiting apoptotic regulation of the SNB to the BC/LA muscles.

In another study of mosaicism of SNB motoneurons, Watson and others (1996) dernonstrated that androgenic regulation of SN6 motoneuronal somal size is regulated directly within the SNB motoneurons . As somal size is thought to be an index of translational activity within motoneurons (Sato 1994), this result is suggestive of a direct alteration of protein synthesis due to the androgen receptor within SNB motoneurons (Sato, 1994) . It is yet undetermined which proteins are being regulated directly by testosterone in SNB motoneurons .

Calcitonin aene-related peptide in the SNB

Calcitonin Gene-Related Peptide is a member of the calcitonin related peptide multigene family, which includes those peptides encoded on the calcitonin genes

(Zaida et al. 1990) . CGRP is a 37 amino acid protein whose existence was predicted from the cloning of the calcitonin gene, which has a sequencing suggestive of alternative splice patterns (Arnara et al. 1982) and was later discovered in neural tissue (Rosenfeid, 1983) . CGRP is encoded on the same gene as calcitonin and arises through alternative splicing of the calcitonin gene in neurons (Cote et al. 1990) . Two distinct forms of CGRP are encoded on the two calcitonin genes, alpha CGRP and beta CGRP. The nucleotide sequencing of the two isoforms is sufficiently different ?O show low hybridization of mRNA probes

(Popper & Micevych 1990) . Antisera directed against synthetic CGRP fragments is variably reactive towards alpha and beta CGRP isoforms . CGRP 1, the antibody used in this study, shows 100% reactivity with alpha CGRP and 78% reactivity with beta CGRP (Penninsula). An un related peptide, also shows a 46% sequence homology with CGRP isoforms, although neither mRNA probes or antibodies cross react significantly with this peptide or its mRNA transcript.

Anatomy of CGRP in the s~inalcord.

CGRP immunoreactivity has been described in brain (Rosenfeld et al.. 1983) and spinal cord (Gibson et al. 1984) for a number of species, including rat . CGRP immunoreactivity iç found in sensory neurons and motoneurons, and has considerable overlap in distribution with Substance P (Gibson et al. 1984) . In the lumbar spinal cord, CGRP immunoreactive fibers are observed in the dorsal horns and dorsal to the central canal, presumably being sensory afferents, and many motoneurons are immunoreactive in the ventral horn (Gibson et al. 1984, Popper

and Micevych, 1989a).

CGRP is thought to be universally but differentially expressed in motoneurons, as

colchicine pretreatment or neuromuscular blockade, two techniques for enhancing visualization by substrate accumulation, result in virtually all somata or

neuromuscluar junctions, respectively, being irnmunoreactive (Csillik et al. 1993) .

The accumulation of CGRP at the endplate by muscular blockade may provoke increased CGRP transcription, however (Popper, Abelson & Micevych 1992) .

CGRP immunoreactivity has been described in SNB rnotoneurons as well as in neuromuscular junctions in the bulbocavernosus muscle in the male rat (Popper &

Micevych 1989a) .

Function of CGRP in motoneurons.

CGRP has been implicated as a neurotransmitter and myotrophic factor in motoneurons (Zaidi et al. 1990) . In vitro, CGRP potentiates numerous functions in striated muscle relevant to cholinergie synaptic transmission and plasticity . For example, CGRP induces postsynaptic nicotinic receptor îubunit synthesis, potentiates presynaptic quantal acetylcholine release, induces postsynaptic acetylcholinesterase activity, encourages spontaneous embryonic rnyotube electrical activity and prevents disuse induced sprouting in adulthood (New & Mudge, 1986, Tsujimoto & Kuno 1988, Changeux 1991, Lu et al. 1993, Liou & Fu

1995, Hodges-Savola & Fernandez 1995) .

Re~ulationof CGR P expression within SNB motoneurons.

Castration of male rats results in an increase in the proportion of SN6

motoneurons which are CGRP immunoreactive and testosterone administration

decreases this proportion to control levels (Popper & Micevych 1989b) . This

androgenic regulation of CGRP expression in SN6 motoneurons has been

determined to be transcriptional in nature (Popper et al. 1990) .

ldentifying the primary location where androgen acts to regulate CGRP in SNB

motoneurons bas proven problematic . Injection of BCILA muscle extracts from

both castrated male rats and intact male rats into intact male rats results in an

increase in CGRP immunoreactivity in SNB motoneurons . Injection of BC/LA

muscle extracts from castrated male rats but not intact male rats into intact male

rats results in an increase in CGRP mRNA , a result which persists after protease

treatment and heat-inactivating of the extracts (Popper, Abelson & Micevych 1992)

. While these results are suggestive of indirect androgenic action on SNB CGRP

expression through effects of the SNB target muscles, it is difficult to determine what bioactive constituents of the extract could be mediating this effect, as similar

results were obtained for mRNA levels with an inactivated extract and with axotomy of the pudenal nerve . Further, the experirnental design of that study does not preclude the possibility of androgenic regulation of CGRP within SNB motoneurons .

Ontoqenv of CGRP in the SNB.

SNB motoneurons have a developmental course distinct from other motoneuron pools within the lumbar spinal cord (see 3.3 above) . CGRP immunoreactivity is also delayed and sexually dimorphic in the SNB motoneurons (Forger et al. 1993).

Forger and others (1993) conclude that the ontogeny of CGRP production makes a role for CGRP in the regulation of apoptosis in the SNB unlikely . This conclusion is puzzling considering the results obtained . Concomitant with a two to threefold reduction in SNB rnotoneuron number during the first week postnatally

(Nordeen et al. 1985), there is a 4 fold increase in the proportion of CGRP immunoreactive SNB rnotoneurons . The developmental pattern of CGRP immunoreactivity in SNB, therefore, corresponds to the predicted pattern if CG RP is presumed to influence apoptosis in SNE3 motoneurons .

A prominent sexual dimorphism exists in the rat SNB, being present in males and virtually absent in females . The SNE3 innervates the perineal BC/LA muscles which play an essential role in male copulation . Several lines of evidence suggest that this sexual dimorphism is established indirectly in the course of development, with testosterone acting on the BClLA muscles perinatally, which then rescue SNB motoneurons from apoptosis .

ln adults exposed perinatally to testosterone (intact males or androgenized fernales), 80-90 % of SN8 motoneurons are androgen sensitive . The function of the intracelluiar androgen receptors found in adult SN6 motoneurons is unknown .

Somata of SNB motoneurons enlarge in response to adult exposure to testosterone, suggesting an increase in gene expression .

Several candidate proteins have been identified as being regulated by testosterone in SNB motoneurons at the transcriptional level . It is not known whether these proteins are being regulated by the androgen receptor within SNB motoneurons or whether the observed alterations in mRNA levels resuIt from actions of testosterone at the BCLA muscles or elsewhere in the nervous system . METHODS

The specific hypothesis tested was that the androgen receptor within SN6

motoneurons regulates CGRP expression . The following experiment assessed the effects of systemic testosterone on SNB motoneuronal expression of CGRP which is attributable to direct action of testosterone on these cells .

The approach taken was to compare neighbonng SNB cells in tfm mosaic animals both in the presence of chronic systernic testosterone and in its absence .

lt was hypothesized that only SNB cells which are androgen sensitive (as determined by androgen receptor immunocytochemistry) would show a decrease in CGRP expression, (as measured by CGRP immunocytochemistry) in the presence of systemic testosterone . To test this hypothesis, the proportion of androgen sensitive SNB cells immunoreactive for CGRP was compared to the proportion of tfm SNB cells immunoreactive for CG RP in both testosterone treated and untreated mosaic animals .

Anirnals

Al1 animals were obtained from our breeding colony . Sixteen mosaic animals were generated by the breeding of known carriers of the ffm mutation . Carriers were identified by the presence of interna1 testis in pups of previous litters . Identified carriers were placed with a sexually vigorous male until copulatory plugs

were seen and the appearance of plugs was taken as the day of conception,

embryonic day O (EmO) . Pregnant carriers received daily injections of 2mg

testosterone proprionate (TP) S.C. in oil from day Em16 until day Em21 . The

prenatal injections of TP serve to maximize SNB motoneuron survival (Ward et al.

1996) as well as distinguish between wild type females, whose nipple lines are

completely masculinized by prenatal testosterone (Goldman et al. 1976) and

mosaic fernales, who form a partial nipple line, due to the presence of androgen

insensitive nipple tissue in tfm mosaic animals . As exogenous perinatal testosterone administration inhibits vaginal delivery, liners not delivered by Em22 were removed by cesarean section from treated carriers under ether anesthesia .

Pups delivered by cesarean section were cross foçtered to a lactating wild type female having delivered within four days prior . Carriers undergoing cesarean section were irnmediately euthanized via a lethal dose of sodium pentobarbitol .

Pups were injected with 1 mg T4P S.C. dissolved in 0.1 ml corn oil on postnatal days 1 and 3 (Pol and Po3) in order to maximally rnasculinize SNB motoneurons

(Ward et al. 1996) .

Animals were identified at day Po30 according to the following system of phenotypic markers (see Table 1) . All identification was verified by dissection of reproductive organs at the end of the experiment .

Androaen maniwlation in aduithood Nine rnosaic animals received 2-20 mm silastic implants (1 -57 mm inner diameter,

3.1 8 mm outer diameter) packed with crystalline testosterone (Steraloids) at 60 days of age and eleven mosaic animals received 2 - 20 mm empty silastic implants placed subcutaneously between the scapulae, under ether anesthesia .

Testosterone implants of this size yield high physiological doses of plasma T4 in castrated male rats (Smith et al. 1977) .

After 4-6 weeks, the implants were removed under ether anesthesia. After a further 24 -36 hours, animals received either 0.2 (n=ll) or 2 (n=9) mg hydroxy flutamide (OH-F) in propylene glycol subcutaneously, which binds to the androgen receptor and induces nuclear translocation of the androgen receptor without altering gene transcription (Wong et al. 1994), facilitating irnmunocytochemical analysis . Al1 animals were perfused 4-6 hours following OH-F injection . Table 1.

Method of identifvinq pup kawotype by phenofv~icmarkers.

Presurned pup ka~otv~e Testis Nip~les tfm affected male X~ Y undescended present

Mosaic fe male xm xW absent present

Wild Type female X~ xW absent absent

Wild Type Male X? external absent f erfusion and tissue ~reparation

Anirnals received an overdose of sodium pentobarbital (approximately 46 mg

intraperitoneally) . On achievement of deep anesthesia, as measured by the

disappearance of deep reflexes, animals were perfused transcardially with 200 ml

ice cold phosphate buffered (pH = 7.4) physiological saline (PBS) over 20 minutes

followed by 200 ml ice cold 4% paraformaldehyde - PBS (pH = 7.4) over 20

minutes . Spinal cords were removed by dissection of the spinal column rostral to the pelvis and caudal to the last 2 ribs followed by extrusion into 4%

paraformaldehyde for 2 hours of post fixation at 4OC . Tissue was then transferred to 20% sucrose - PBS at 4°C until th9 spinal cords sank . Tissue was then sectioned coronally at 50 Fm on a freezing microtome . Spinal sections corresponding to L5-6 and SI were sequentially collected into three eppendorf tubes containing de Olmos solution (propylene glycol based cryopreservent previously demonstrated to preserve antigens over long periods of time - Watson et al. 1986) and stored at -20°C until ICC was performed . Only every third SNB sections (Le. one eppendorf tube) collected from each animal was used for immunocytochemistry and data collection .

lmmunocytochemistrv

Tissue in de Olmos solution was reconstituted in tissue wells with five 10 minute washes in PBS solution containing 0.1 % gelatin and 0.3% Triton-XI00 (Sigma) and sections were sequentially double labeled for androgen receptor and CGRP

as follows. Free floating sections were incubated with PBS-GT containing 10%

normal goat serum (NGS) for 1 hour at 24OC to prevent non-specific secondary

antibody binding . Androgen receptor immunoreactivity was then assessed using the rabbit polyclonal prirnary antibody PG21 directed against the 21 amino acid

COOH epitope of the rat androgen receptor (gift of Gai1 Prins, University of

Chicago) . PG21 has been previously characterized in the rat SNB , and SNB AR-

IR has been shown to discriminate wild type and tfm affected neurons (Freeman et

al. 1995) . Tissue was incubated at 4°C for 36-48 hours under constant agitation on a rnixing platform in PG21 primary antiserum diluted in PBS-GT containing 1%

NGS at a dilution of 1 : 3000 . After rinsing, tissue was incubated with biotinilated goat anti-rabbit antiserum (Vector Laboratories) at 1: 250 dilution in PBS - GT for one hour at 24OC . Following rinsing, tissue was incubated with avidin-biotin peroxidase cornplex (Vectastain elite kit, Vector Laboratories) for 1 hour at 24°C .

Peroxidase (and hence androg en receptor immunoreactivity) was detected by activating the chromagen 3' 5' diaminobenzidine (DAB) with hydrogen peroxide in the presence of nickel chloride in Tris buffer (pH = 7.2),resulting in a blue-black label . Tissue was thoroughly rinsed of DAB solution and CGRP immunocytochemistry begun .

CGRP immunoreactivity was assessed with rabbit polyclonal antisera generated against synthetic rat CGRP (Penninsula). Sections were again incubated in 10%

NGS for 1 hour at 24°C following which sections were incubated for 48 hours in CGRP antisera at 1: 16000 dilution at 4OC under constant agitation on a mixing platform . Sections were rinsed and incubated with biotinilated goat anti-rabbit secondary antisera in PBS-GT (1 : 250 dilution, Vector Laboritories) for 1 hour at

24°C. Following rinsing, sections were incubated in avidin-biotin complex

(Vectastain elite kit, Vector Laboratories) for 1 hour at 24OC, rinsed and visualized with DAB without nickel enhancement, yielding a red-brown label . Sections were thoroughly rinsed of DAB solution, mounted on gelatin coated slides, and coverslipped with Permount (Anachemia Science) following dehydration through graded alcohols and clearing in xylene.

Identification of CGRP-IR and AR-IR cells

Sections were analyzed under a light microscope (Nikon Optiphot -2) at 200X magnification by an experimenter blind to experimental condition . SNB cells were rated as being lightly = 1 or darkly = 2 immunoreactive for CGRP and/or AR (see table II). Cells rated as immunoreactive for either or both proteins were drawn and labeled using a camera lucida attachment . A second experimenter analyzed tissue from six randomly selected anirnals as a reliability check.

Following identification of immunoreactive SNB cells, coverslips were soaked off in xylene and Neutral Red counterstain was applied in order to identify SNB cells unlabeiled by either antiserum and to visualize cells labelled for AR alone . Cells without either label were identified as ffm CGRP-IR negative . All cells identified by Neutra1 Red were drawn using a camera lucida attachment .

RESULTS

All analyses were conducted on SPSS standard version (release 7.5.1) using the general linear mode1 function for omnibus ANOVA and T-tests for planned cornparison .

Only mosaic animals having ten or more cells of each type (AR-IR positive and

AR-IR negative) were included in statistical analyses . Four animals were excluded under this criterion . Further, only cells rated as darkly immunoreactive for CGRP andlor AR were classified as positive, for analysis purposes . Cells rated as lightly or not immunoreactive for CGRP and/or AR were classified as being negative, for analysis purposes .

lnterrater reliabilitv

lnterrater reliability coefficients were established to be 0.98 for AR negative,

CGRP immunopositive cells and 0.997 for AR positive, CGRP immunopositive cells (pc0.05) . Table 2.

Method of identification of immunolabeling of CGRP and AR and desianation of cell tvoe.

Irnmunolabel Cell Tv~e Microscopie amearance

AR/CGRP-IR wf blue-black nucleus, red-brown soma

CGRP-IR ffm unlabeled nucleus, red-brown soma

AR4 R wt blue-black nucleus, pink soma after counterstain no label tfm unlabelled nucleus, pink soma after counterstain Figure 3.

A. Cells rated as immunolabeled for CGRP and/or AR. a.) darkly immunolabeled for androgen receptor but not CGRP (AR-IR) b.) darkly immunolabeled for CGRP but not androgen receptor (CGRP-IR) c.) darkly immunolabeled for both CGRP and androgen receptor (CGRP/AR-I R) . brightfield photomicrograph 200X magnification (scale bar = 50 microns)

Effects of hydroxy flutamide dose on androaen receDtor identification

Animals receiving 0.2 mg (n=9) and animals receiving 2mg (n=7) did not differ in the mean number of rated androgen receptor positive SN8 nuclei (independent samples T-test, df= 16, t= 2.03, pz0.05) .

Dearee of observed mosaicism

The rnean proportion of SN5 cells immunoreactive for androgen receptor was found to be 49.8%, ranging from 30% to 70% . No difference in mean proportion of SNB cells immunoreactive for androgen receptor was observed between treatment groups (independent samples T-test df = 14, t= -0.47, p >0,05) .

Number of SNB cells observed

The mean number of SNB cells observed in one third of spinal segments SI, 15 and L6 was found to be 56.25 (range = 81 - 32, standard deviation = 15.46) .

This agrees very well with previous reports of SNB cell number in androgenized mosaic females (Freeman et al. 1996) . Effects of celI tvpe and testosterone exposure on CGRP imrnunoreactivitv

Animais were assigned four scores corresponding to four possible outcornes for

SNB rnotoneurons : number of CGRP-IR cells, number of AR-IR cells, number of

CGRPIAR -IR cells and nurnber of cells unlabelled by either antisera . Analyses

were carried out by cell type (as determined by androgen receptor

imrnunoreactivity) and by experimental group to which the animal was assigned (T

or B implanted) . The proportion of AR-IR positive cells also immunoreactive for

CGRP were compared to the proportion of AR-IR negative cells immunoreactive

for CGRP within each animal .

2 by 2 ANOVA (Treatrnent by Cell Type) withidbetween subjects cornparison,

revealed a significant effect for cell type (df = 1,14 , F=10.179 , pc0.05) . Planned

cornparisons of group means (paired samples T-test) revealed that a lesser

proportion of wild type (AR-IR) ceils were darkly immunoreactive for CGRP only in the testosterone implanted group (df = 5 , k4.633 , pc0.05) but not the blank

implanted group (df = 9 , t = 1.230 , p>0.05) (see figure 4.). Figure 4.

Effects of testosterone treatment on CGRP-IR in SNE3 motoneurons mosaic for androqen insensitivitv.

wild type androgen [sensitive ] CEUTYPE DISCUSSION

Locus of andro~enicrequlation of SNB CGRP

It seems clear from the present study that the androgenic regulation of CGRP

expression within SNB motoneurons is mediated locally in the adult rat .

Functional androgen receptors within SNB motoneurons are necessary for the

reduction in SNB motoneuronal CGRP expression associated with circulating

androgen . This constitutes the first in vivo demonstration of androgen receptor

dependence of gene expression localized to a specific neural population . The

involvement of androgen receptor within SNB motoneurons in regulation of CGRP

expression runç contrary to the prevailing line of inquiry and as such is novel .

Additionally, the present study provides a dernonstration of the utility of the mosaic

paradigm in elucidating androgen dependent intraneuronal regulation of gene

expression .

As continued androgen exposure is necessary to mairitain low levels of CGRP

mRNA (Popper & Micevych 1990), the most parsimonious interpretation of the present result is a genomic effect of the androgen receptor in down-regulating

CGRP gene expression . This hypothesis is strengthened by the finding that in the trigeminal nucleus, P4 increases CGRP expression (Moussaoul et al. 1996) . As previously noted, the progesterone receptor appears to share a HRE with the androgen receptor . Nevertheless, the possibility that CGRP is significantly downstream from the genomic effects of the androgen receptor in SNB motoneurons remains untested .

The results of this study challenge the sufficiency of androgenic action on the

BCLA in the regulation of CGRP levels in the SNB . While previous work, reporting effects for extract injection resulting in elevated CGRP m RNA levels in the SNB (Popper & Micevych 1989, Popper, Uiibarri & Micevych 1992, Popper,

Abelson & Micevych 1992), suggests that the BCLA response may be sufficient to increase SNE3 CGRP gene expression, it is not clear that this type of regulation is physiological . Further, the inconsistency of results employing the extract injection technique prevents simple interpretation of the reported findings . Muscle may be involved in the regulation of CGRP levels, although the present study indicates that functional androgen receptors in SNB motoneurons are a necessary requirement for this to occur .

The finding that only SNB cells containing functional androgen receptors show a reduction in CGRP expression in the presence of testosterone suggests that in the rnosaic model, muscle effects are trivial . It is important to note, however, that the methodology ernployed in the current study does not distinguish between SNB motoneurons innervating the BC/LA and SNB motoneurons innervating the external anal sphincter . It is conceivable that regulation of SNB CGRP expression is distinct in this muscle and that skews in connectivity could obscure an effect of the BC/LA muscles on SNB CGRP expression .

An interpretation consistent with both findings is that the androgen receptor in

SNB motoneurons has a permissive role in a process of androgenic regulation of

CGRP involving androgenic action on muscles . As the developmental action of androgen on the SNB is mediated through the BClLA muscles, and paralysis of muscles leads to increased CGRP expression in SNB rnotoneurons (Popper &

Micevych 1989), such a mechanism is plausible . As no resolution of rnuscular events was made in the present study, and as al1 animals employed in the extract injection studies (Popper & Micevych 1989, Popper, Abelson & Micevych 1992,

Popper, Ulibarri & Micevych 1992) were intact males, and therefore the vast majority of SNB motoneurons had functional androgen receptors which were activated by circulating androgen, this interpretation has not been tested .

In general, the mosaic animals employed in this studied displayed a degree of

SNB androgen receptor mosaicism w hich agreed well with previously observed results (Freeman et al. 1996) . Haif (rnean = 49.8 %) of SNB cells were immunoreactive for androgen receptor . While slightly higher than the roughly

35% AR positive SNB neurons observed by Freeman et al. 1996, it rernains significantly less than the roughly 75% AR positive SNB neurons reported in wild type littermates of rnosaic animals (Freeman et al. 1996) . The rnethod of identifying androgen receptor positive cells likewise proved consistent across dose (hydroxy flutamide doses = 0.2 mg or 2 mg) . These observations support the

accuracy of the method of androgen receptor identification and hence rnosaic cell

type classification employed in this study .

The overall proportion of SNB rnotoneurons rated as immunoreactive for CGRP

departed significantly from the values reported elsewhere (see figure 3) . Although

testosterone treatment reduced the number of SN8 neurons expressing CGRP by

approximately one half in both Popper and Micevych's (1989) study and in the

androgen sensitive population in our study, we observed fewer labeled cells

overall . This difference can be attributed to a number of factors : 1) differences in the method of identifying imrnunolabeled neurons, 2) intrinsic differences in androgenic responses in mosaic tissues and 3) intrinsic differences between

males and androgenized female models .

While no clear criteria for rating CGRP immunolabeled cells are reported by

Popper and Micevych (1 989), it is conceivable that differences in immunostaining methodology (for example, in the anti-CGRP antibody ernployed) or in the method of rating cells as immunoreactive might exist . As Popper and Micevych used

Long-Evans rats as subjects whereas the animals in the present study were King-

Holtzmann / Wistar crosses, a strain difference is a possibility . In a developmental study of CGRP immunoreactivity in the SNB conducted by other investigators for exarnple, roughly 65% of male SNB motoneurons were reported to be CGRP immunoreactive at 60 days (Jordan et al. 1993), versus the roughly

40% reported by Popper and Micevych (1989) .

P hvsioloqical consequences of direct reaulation of CG R P exoression within SN 8 motoneurons

Questions pertaining to the ultimate causation of androgen receptor regulation of

SNB CG RP expression are difficult to address . Many functional roles for CG RP in neuromodulation of cholinergic transmission (Liou & Fu 1995, Hodges-Savola

& Fernandez 1995) and maintenance of synaptic integrity (New & Mudge, 1986,

Tsujimoto & Kuno 1988, Changeux 1991, Lu et al. 1993) have been put forward , and discerning a unitary role for CGRP in the neuromuscular junction is elusive .

CGR P appears to modulate functional parameters of neuromuscular transmission.

A neuromodulator such as this could be important in potentiating the restoration of functional transmission at the neuromuscular junction in infrequently used muscle groups . It appears that CGRP expression in muscles is inversely related to the frequency of use, or activational history (Blanco et al. 1997) . Studies of CGRP expression in motoneurons innervating constituently active muscles, such as the diaphragm, indicate low CGRP expression, whereas muscles frequently active

(such as the soleus) are innervated by motoneurons expressing intermediate levels of CGRP . The highest levels of CGRP expression are found in the muscles which are infrequently active, such as the BC/LA which is devoted to sexual behaviour

Artificially inducing inactivity, by paralyzing the BC muscle increases motoneuronal CGRP mRNA synthesis in the SNB (Popper, Ulibarri & Micevych

1992) . That CGRP appears to regulate nicotinic receptor expression in muscle by a signal transduction pathway distinct from that with which cholinergie transmission does so (Fontaine et al. 1987) strengthens the argument that CGRP rnay potentiate synaptic integrity in infrequently active muscles .

If such a role for CGRP in neuromuscular transmission proves accurate, then androgen serves, paradoxically, to decrease the synaptic integrity at neuromuscular junctions specific for reproduction . This paradox rnay be resolved by considering how variation in androgen levels and activity in the SNB might occur ecologically in the rat.

Male rats which are not actively breeding, such as rats undergoing seasonal attenuation of reproduction, would presumably have lower activity in the SNB-

BCILA system and also lower circulating testosterone values (Karnel et al. 1975) .

Male rats frequently breeding would have frequently active SNB-BC/LA systems and high testosterone levels . In this way, androgen decreasing CGRP expression in the SNB would be economical . In periods of high sexual activity, transmission at the SNB-BC/LA neuromuscular junction would be relatively frequent and the trophic action of cholinergic transmission presurnably sufficient to mediate synaptic integrity . In periods of low sexual activity, transmission at the SNB-

BC/LA neuromuscular junction would be infrequent, but increased CGRP expression in SNB motoneurons due to lowered circulating androgen would potentiate both cholinergic transmission (Lu et al. 1993) and the trophic actions of cholinergic transmission (New & Mudge 1986, Tsujimoto & Kuno 1988) .

Presynaptic CGRP, then, may play a role in the maintenance of neuromuscular synaptic integrity . In this case, variation in androgen levels resulting from reproductive behaviour would then enact an adaptive response in CGRP expression in SNB motoneurons . Low levels of androgen resulting from periods of relative sexual inactivity would increase CGRP expression presynaptically and thus contribute to synaptic integrity . Essentially, androgenic regulation of CGRP within SNB motoneurons would serve to keep the BC/LA system ready for use .

Androaenic reaulation of CGRP in the nervous svstem

The finding that CGRP is under direct control of the androgen receptor in SNB motoneurons raises the possibility that androgen regulates CGRP levels in other motoneuron pools, and in other neural populations . Androgen mediated cellular events may not be as drarnatic in other neural populations as the SNB is remarkable in androgen sensitivity and absence of estrogen sensitivity . Other motoneuron pools show varying degrees of androgen sensitivity, but the proportion of androgen sensitivity in these motoneuron pools is less than that of the SNB (see above). The sexual dimorphism present in the SNB is remarkable for motoneuron pools, as such, it seems unlikely that androgen would be a major regulator of CGRP expression in motoneurons innervating muscles whose activity is not contingent on androgen .

If it is assumed that androgen regulates CGRP in the same rnanner in al1 motoneuron pools, CGRP expression does not correspond in the expected way to degree of androgen sensitivity . If the SN6 is taken as the most androgen sensitive motoneuron pool, it also shows the highest levels of CGRP expression

(Blanco et al. 1997), which does not agree with the hypothesized attenuation of

CGRP expression in motoneurons by androgen . It could be, however, that androgenic regulation of CGRP is similar in other motoneuron pools, but that other factors regulating CGRP expression (Durham et al. 1997) would obscure androgen's effects .

Steroid regulation of CGRP has been described in the hypothalamus and thalamus of mammals . Studies in the MPOA and periventricular nucleus have indicated that, in female rats, estrogen increases CGRP expression (Yuri &

Kawata 1992) . Sirnilarly, testosterone appears to increase hypothalarnic CGRP expression in male mice, although it is not clear that this is not a consequence of aromatization, nor is this effect consistent in trigeminal or cervical spinal cord

areas examined (Moussaoul et al. 1996). Doubtless CGRP is regulated by

steroids, but the nature of this regulation is not obviously conserved across neural

populations .

Mechanism of androaen rece~torreaulation of CGRP expression

The present study indicates that the androgen receptor in SNB motoneurons is

necessary for androgenic regulation of CGRP but it does not discriminate at what

levei this regulation is occurring within the ceIl . CGRP could be significantly

downstrearn from the primary genomic effects of the androgen receptor . No

definitive ARE has been identified, and possible HREs in the calcitonin gene

promoter regions have not been described, rnaking interpretation of in vivo

assessrnents of CGRP regulation difficult .

The present results indicate that SNB CGRP is regulated directly through the

androgen receptors within SNB motoneurons . Indirect methods of determining the primary site of androgenic action in the regulation of SNB CGRP have lead to

misleading results, which may represent artifact of the method of testing . No support was found for the mode1 of BClLA muscles regulating SNB CGRP in response to systemic testosterone . The mosaic paradigm provides a powerful tool for the analysis of steroid receptor action in vivo . The present study provides a demonstration of its utility in studies of gene expression . Localizing steroid action has traditionally been difficult and addressed very indirectly . The rnosaic paradigrn allows for a direct test of these concerns . 60

LIST OF ABBREVIATIONS USED

% = percent

OC= degrees centigrade

3a-di01 = 3a-androstenediol

3P-diol = 3B-androstenediol pm = microns

[3H] = tritiated

A4 = androstenedione

ANOVA = analysis of varience

AR = androgen receptor

ARE = androgen response element

B = blank silastic implant

SC = medial and lateral bulbocavernosus muscles

BCILA = media1 and lateral bulbocavernosus muscles and levator ani muscles

CGRP = calcitonin gene-related peptide cm = centirneters

CNS = central nervous system

COOH = carboxyl

DAB = 3' 5' diaminobenzidine

DH EA = 5-androstene-3P-ol-17-one, dehydroepiandrosterone

DHT = dihydrotestosterone

DNA = deoxyribonucleic acid E2 = 17p-estradiol

Em = embryonic day

ER = a- estrogen receptor

GABA = y - arninobutyric acid

HRE = hormone response element

ICC = immunocytochemistry

IR = immunoreactive

IR+ = displaying immunoreactivity

IR- = displaying no immunoreactivity

L = lumbar segment of spinal cord

LA = Ievator ani muscle m-RNA = messenger ribonucleic acid ml = milliliters mm = millimeters

MMTV-CAT = mammilary mouse tumor virus chloramphenicol acetyl transferase

MPOA = medial preoptic area of the hypothalamus

NGS = normal goat serurn

NH3 = amino

OH-F = hydroxy flutamide

P4 = progesterone

PBS = phosphate buffered saline

PBS-GT = 0.3 percent triton X-100, 0.1 percent gelatin phosphate buffered saline Po = postnatal day

PRE = progesterone response element

RDLN = retrodorsolateral nucleus

S = sacral segment of spinal cord

SHBG = steroid hormone binding globulin

SNB = spinal nucleus of the bulbocavernosus

sxr = sex reve rsal rnutation

T = testosterone filled silastic implant

T4 = testosterone

T4P = testosterone proprionate

tfm = testicular feminization mutation

TIC = transcription initiation complex

X = X chromosome xH"= wild type X chromosome xmn= X chromosome with the tfm mutation

Y = Y chromosome REFERENCES

Abdelgadir, S.E., Resko, J.A., Ojeda, S.E., Lephart, E.D., McPhaul, M-J., Roselli, C.E. (1 994) . Androgens regulate aromatse cytochrome P450 messenger ribonucleic acid in rat brain. Endocrinoloav . 135(1) , 395 - 401.

Al-Shamma, H.A., Arnold, A.P. (1 995) . Importance of target innervation in recovery from axotomy-induced loss of androgen receptor in rat perineal motoneurons. Journal of Neurobiolo~v.28(3) , 341 - 353.

Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., Watson, J.D. (Eds). (1 992) . The molecular biolo~yof the cell. (3rd ed.). New York : Garland .

Amara, S.G.,Jonas, V., Rosenfeld, M.G., Ong, ES., Evans, R.M. (1 982) . Alternative RNA processing in calcitonin gene expression generates mRNAs encoding different polypeptide products. Nature (London). 298 , 240 - 244.

Argente, J., Chowen- Breed, J.A., Steiner, RA., Clifton, D.K. (1990) . messenger RNA in hypothalamic neurons is lncreased by testosterone through activation of androgen receptors and not by aromatization to estradiol. Neuroendocrinoloav. 52 , 342 - 349.

Arnold, A.P. (1 996) . Genetically triggered sexual differentiation of bain and behavior. Hormones and Behavior. a,495 - 505.

Arnold, A.P., Breedlove, S.M. (1 985). Organization and activational effects of sex steroids on brain and behavior : A reanalysis. Hormones and Behavior. , 469 - 498.

Bakker, J., Slob, A.K., van Leeuwen, F.W., Sonnemans, M., Pool, C.W., (1 997). Quantitative estimation of estrogen and androgen receptor- immunoreactive cells in the forebrain of neonatally estrogen-deprived male rats. Neuroscience .77(3)$91 1-91 7

Beato, M. (1 989) . Gene regulation by steroid hormones. Cell. 56 , 335 - 344.

Beato, M., Sanchez-Pacheco, A. (1 996) . Interaction of steroid hormone receptors with the transcription initiation complex. Endocrine Reviews. u(6), 587 - 609. Berger, F.G. (1 989) . Androgen-regulated gene expression. Annual Review of Physiolociv. a,51 - 65.

Blanco, C.E.,Popper, P., Micevych, P. (1997) . Alpha-CGRP mRNA levels in motoneurons innervating specific rat muscles. Brain Research Molecular Brain Research. g , 253-61. Breedlove, S. M., (1984). Steroid influences on the development and function of a neuromuscular system. Proaress in Brain Research. g,147-1 70.

Breedlove, S.M. (1985) . Hormonal control of the anatomical specificity of motoneuron-to-muscle innervation in rats. Science . 227 , 1357 - 1359.

Breedlove, S.M. (1 986) . Cellular analyses of hormone influence on motoneuronal development and function. Hormones and Behavior . B(4) , 469 - 98.

Breedlove, S.M. (1 992) . Sexual dimorphism in the vertebrate nervous system. The Journal of Neuroscience . -(Il) ,4133 - 4142. Breedlove, S.M., Arnold, A.P. (1980) . Hormone accumulation in a sexually dimorphic motor nucleus in the rat spinal cord. Science .210 , 564 - 566.

Breedlove, S.M., Arnold, A.P. (1 981) . Sexually dimorphic motor nucleus in the rat lumbar spinal cord : Response to adult hormone manipulation, absence in androgen-insensitive rats. Brain Research . 225, 297-307. Breedlove, S.M.,Arnold, A.P. (1983a) . Hormonal control of a developing neuromuscular system : 1. Cornplete demasculinization of the male rat spinal nucleus of the bulbocavernosus using the anti-androgen flutamide. The Journal of Neuroscience . 3,417 - 423.

Breedlove, S.M., Arnold, A.P. (1 983b) . Sex differences in the pattern of steroid accumulation by motoneurons of the rat lumbar spinal cord. The Journal of Com~arativeNeuroloov . 21 5 ,211 - 216. Burke, K.A., Widows, M.R., Senglaub, D.R. (1997) . Synergistic effects of testosterone metabolites on the developrnent of motoneuron morphology in a sexually dimorphic rat spinal nucleus. Journal of Neurobiology. 33(1) , 1 - 10.

Cattanach, B.M., Pollard, C.E., Hawkes, S.G. (1971). Sex reversed mice : XX and XO males . Cvtoaenetics. , 31 9-37. Changeux, J.P. (1 991) . Compartrnentalized transcription of acetylcholine receptor genes during motor endplate epigenisis. New Bioloriist .3(5),41 3 - 429. Cohen, J., Dore, C., Robaire, B., Ruf, K.B. (1984) . Plasma concentrations of free 5 -androstane-3 , 17 -di01 and related gonadal steroids during spontaneous and induced sexual maturation in the fernale rat. Bioloqv of Reproduction. 30 , 105-111.

Conley, A.J., Bird, I.M. (1997) . The rofe of ctyochrome P450 17 - hydroxylase and 3 -hydroxysteroid dehydrogenase in the integration of gonadal and adrenal steroidogenisis via the 5 and 4 pathways of steroidogenesis in mamrnals. Biolow of Reproduction. s,789 - 799. Cote, G. J., Nyuyen, LN., Berget, S.M., Gagel, R.F. (1 990) . Calcitonin exon sequences influence alternative RNA processing. Molecular Endocrinoloav. A(Il), 1744 - 1749.

CsiHik, B., Tatji, L., Kovacs, T., Kukla, E., Rakic, P., Knyihar-Csillik, E. (1 993) . Distribution of calcitonin gene-related peptide in vertebrate neuromuscular junctions : relationship to the aceytlcholine receptor. The Journal of Histochemistw and Cytochemistrv .4J(10) , 1547 - 1555. Desjardins, C. (1 981) . Endocrine signaling and male reproduction. Bioloav of Re~roduction.a , 1-21.

Deslypere, J., P., McPhaul, M., J., Wilson,J., O., Young, M. (1992). Testosterone and 5 alpha-dihydrotestosterone interact differently with the androgen receptor to en hance transcription of the MMTV-CAT reporter gene. Molecular and Cellular Endocrinoloqv . 88(1-3),15 - 22.

Drengler, S.M., Jones, K.J., Handa, R.J., (1996a). Sex differences in androgen receptor mRNA levels and regulation in hamster facial motoneurons. Brain Research Molecular Brain Research .35(1-2),: 131 -1 38.

Drengler, S.M., Handa, R.J., Jones, K.J. (1 996b) . Regulation of androgen receptor mRNA expression in hamster facial motoneurons : Differential effects of non-aromatizable and aromatizable androgens. Molecular Brain Research . 0 , 8 - 15. Drews, U., Schleicher, G., Stumpf, W.E., Theidemann, K.U. (1 988) . Androgen receptor-deficient cells in the mosaic epididimis of sex-reversed mice heterozygous for ffm: An autoradiographic study with [3H]-estrodiol. Anatomical Record. 220(4) ,393 - 396.

Dube, J.Y., Lesage, R., Trernblay, R.R. (1 975) . Androgen and estrogen binding in rat skeletal and perineal muscles. Canadian Journal of Biochernistry. 50- 55.

Durham, P.L., Russo, A.F., Sharrna, R.V., (1 997) . Repression of the calcitonin gene-related peptide prornoter by 5-HT1 receptor activation. Journal of Neuroscience. l7(24), 9545-53.

Emery, D.E., Sachs, B.D. (1975) . Ejaculatory pattern in female rats without androgen treatment. Science. 190 ,484 - 486. Erkular, S.D., Wetzel, D.M. (1 989) -5Alpha -dihydrotestosterone has nonspecific effects on membrane channels and possible genomic effects on ACh- activated channels. Journal of Neuro~hysiology.a(5) , 1036 - 1052. Erskine, M.S. (1983) . Effects of an anti-androgen and 5 alpha -reductase inhibitors on estrus duration in the cycling fernale rat. Phvsiolo~vand Behavior. 30 , 51 9 - 524. Evans, R.M. (1 988) . The steroid and thyroid hormone receptor superfamiiy. Science. 240 ,889 - 895. Fishrnan, RB., Breedlove, S.M. (1 985) . The androgenic induction of spinal sexual dimorphism is independent of supraspinal afferents. Develo~mentaiBrain Research. a ,255 - 258. Fishman, R.B., Breedlove, S.M. (1988). Neonatal androgen maintains sexually dimorphic muscles in the absence of innervation. Muscle Nerve. u,553 - 560. Fishman RB., Chism, L., Firestone, G.L., Breedlove, S.M. (1 990) . Evidence for androgen receptors in sexually dimorphic penneal muscles of neonatai male rats : Absence of androgen accumulation by the perineal rnotoneurons. Journal of Neurobioloav. 21, 694 - 704 . Fishman, R.B., Breedlove, S.M. (1992). Local perineal implants of anti- androgen block masculinization of the SNB. Developmental Brain Research. El 283 - 286.

Fontaine, B., Klarsfeld, A., Changeux, J-P. (1 987) . Calcitonin gene-related peptide and muscle activity regulate acetylcholine alpha-subunit mRNA levels by distinct intracellular pathways. The Journal of Cell Bioloav. 105 , 1337 - 1342.

Forger, N.G., Hodges, L.L., Roberts, S.L., Breedlove, S.M. (1 992) . Regulation of motoneuron death in the spinal nucleus of the bulbocavernosus. Journal of Neurobioloqy -23,1 192 - 1203.

Forger, N.G., Hodges, L.L., Breedlove, S.M. (1993) . Ontogeny of calcitonin gene-related peptide immunoreactivity in rat lumbar motoneurons : Delayed appearance and sexual dimorphism in the spinal nucleus of the bulbocavernosus. The Journal of Com~arativeNeuroloav. 330 , 51 4 - 520.

Freeman, L.M., Padgett, B.A., Prins, G.S., Breedlove, S.M. (1 995) . Distribution of androgen receptor immunoreactivity in the spinal cord of wild-type, androgen-insensitive and gonadectomized male rats. Journal of Neurobioloa~. (l), 51 - 59. Freeman, L.M., Watson, N.V., Breedlove, S.M. (1 996) . Androgen spares androgen - insensitive motoneurons from apoptosis in the spinal nucleus of the bulbocavernosus in rats. Hormones and Behavior . a,424 - 433 . Frye, C.A., Duncan, J.E., Basham, M., Erskine, M.S. (1996) . Behavioral effects of 3 alpha- androstanediol II : Hypothalamic and preoptic area actions via a GABAergic mechanism . Behavioral Brain Research . , 119-1 30. Gibson, S.J., Polak, J. M., Bloom, S.R., Sabate, LM., Mulderry, P.M., Ghatei, M.A., McGregor, G.P., Morrison, J.F.B., Kelly, J.S., Evans, R.M., Rosenfeld, M.G. (1 984) . Calcitonin gene-related peptide in the spinal cord of man and eight other species. The Journal of Neuroscience. A(I2) ,3101 - 31 1 1. Gladue, B.A., Clemens, L.G. (1 982) . Development of ferninine sexual behavior in the rat : Androgenic and temporal influences. Phvsioloqv and Behavior. a,263 - 267. Glass, C.K. (1994) . Differential recognition of target genes by nuclear receptor rnonorners, dimers and heterodimers. Endocrine Reviews . l5(3) , 391 - 407.

Goldman, A.S., Root, A.W., Duckett, G., Shapiro, G.H. (1975) . Pituitary hormone secretion in the genetically male rat pseudohermaphrodite. Journal of Endocrinoloay . &, 249 - 255.

Goldrnan, A.S., Shapiro, B.H., Neuman, F. (1 976) . Role of testosterone and its metabolites in the differentiation of the mammary gland in rats. Endocrinoloay . B,1490 - 1495. Gorzalka, B.B., Rezek, DL., Whalen, R. E. (1 975) . Adrenal mediation of estrogen-induced ejaculatory behavior in the male rat. Phvsioloçiy and Behavior. -14,373-376. Goy, R.W. (1 970) . Experimental control of psychosexuality. Philoso~hical Transactions of the Roval Societv of London: Series B : Biolo~icalSciences . 259 , 149-162.

Grattan, D.R., Rocca, M.S., Sagrillo, C.A.: McCarthy, M.M., Selmanoff, M. (1 996) . Antiandrogen microimplants into the rostral medial preoptic area decrease gamma-aminobutyric acidergic neuronal activity and increase lutenizing hormone secretion in the intact male rat. Endocrinoloay. 137(10) , 41 67- 41 73.

Hammond, G.L. (1997) . Determinants of steroid hormone bioavailability. Biochemical Societv Transactions. a (2), 577-582. Handa, R.J., Kerr, J.E., DonCarlos, L.L., McGivern, R.F., Hejna, G. (1 996) . Hormonal regulation of androgen receptor messenger RNA in the medial preoptic area of the male rat. Molecular Brain Research. , 57 - 67.

Handa, R.J., Resko, J.A. (1989) . Alpha-adrenergic regulation of androgen receptor concentration in the preoptic area of the rat. Brain Research. 483 , 31 2 - 320.

Hart, B.D. (1 983) . Role of testosterone secretion and penile reflexes in sexual behavior and sperm cornpetit ion in male rats : A theoretical contribution. Phvsioloav of Behavior . a,823 - 827. Hodges-Savola, C.A., Fernandez, H.I. (1 995) . A role for calcitonin-gene related peptide in the regulation of rat skeletal muscle G4 acetylcholinesterase. Neuroscience Letters. 190 , 117 - 1 20.

Imamura. M., Prasad, C. (1 998). Modulation of GABA-gated chloride ion influx in the brain by dehydroepiandrosterone and its metabolites. Biochemical Biophvsical Research Communications . 243(3) , 771-5 . Jordan, CL.,Padgett, B., Hershey, J., Prins, G., Arnold, A. (1997) . Ontogeny of androgen receptor immunoreactivity in lumbar motoneurons and in the sexually dimorp hic levator ani muscle of male rats. Journal of Com~arative Neuroloav. 379 ,88-98.

Johnson, M.H. Everitt, B.J. (1995) : Essential Re~roduction(4th ed.). Boston : Blackwell Science.

Kamel, F., Mock, E.J., Wright, W.W., Frankel, A.I. (1 975) : Alterations in plasma concentrations of testosterone, LH and associated with rnating in the male rat. Hormones and Behavior. 6,277 - 288. Kemppainen, J.A., Lane, M.V., Sar, M., Wilson, E.M. (1992) . Androgen receptor phosphorylation, turnover, nuclear transport and transcriptional activation. The Journal of Bioloaical Chemistrv. 267(2) , 968 - 974.

Koopman, P., Gubbay, J., Vivian, N., Goodfellow, P., Lovell-Badge, R. (1 991) . Male development of chromosornally fernale mice transgenic for sry. Nature (London). 351 , 117 - 121.

Kurtz, E. M., Cover, A.R., Sengelaub, D.R. (1992) . Testosterone fails to Save androgen-sensitive rat motoneurons following early target removal. Develo~mentalBrain Research. 7J , 181 - 189. Lephart, E.D. (1996) . A review of brain aromatase cytochrome P450. Brain Research Reviews. E,1 - 26. Levine, S., Mullins, R.F. Jr. (1 964) . Estrogen administered neonatally affects adult sexual behavior in male and fernale rats. Science . 144 , 185 - 187. Li ,X., Rissman, E. F., Schwartz, P. E. (1997) . Distribution of estrogen receptor-beta-like immunoreactivity in rat forebrain. Neuroendocrinolo~y.66(2) , 63-7.

Lovell-Badge, R., Hacker, A. (1 995) . The molecular genetics of sryand its role in mammalian sex determination . Philoso~hicalTransactions of the Rovai Society of London: Series B : Bioloqical Sciences . 350(1333) , 205 - 21 4. Liou , J-C., Fu, W-M. (1 995) . Additive effects of ADP and CGRP in modulation of the acetylcholine receptor channel in Xenopus embryonic myocytes. British Journal of Pharrnacoloqv. 1 15 , 563 - 568. Lu, B., Fu, W-M., Greengard, P., Poo, M-M. (1993) . Calcitonin gene related peptide potentiates synaptic responses at developing neuromuscular junction. Nature (London). 363 ,76 - 79. Lubischer, J.L., Arnold, A.P. (1 995) . Axotomy transiently down-regulates androgen receptors in motoneurons of the spinal nucleus of the bulbocavernosus. Brain Research. 694 , 61 - 68.

Lumbroso, S., Sultan, C., Privat, A., Brinkmann, A.O., Lobaccaro, J.M., Georget, V., Sandillon, F. (1996). Immunohistochemical localization and immunoblotting of androgen receptor in spinal neurons of male and female rats. Euro~eanJournal of Endocrinoloqv . 134(5), 626-632.

Lustig, R.H. (1994) . Sex hormone modulation of neural development in vitro. Hormones and Behavior. a,383 - 395. Lustig, R.H., Hua, P., Smith, L.S., Wang, C., Chang, C. (1 994) . An in vitro mode1 for the effects of androgen on neurons employing androgen transfected PC12 cells . Molecular and Cellular Neurosciences .5,587 - 596.

Lyon, M.F. (1 961) . Gene action in the X-chromosome of the mouse (Mus musculus L.). Nature(1ondon) . 190 , 372 - 373.

Lyon, M.F., Glenister, P.H. (1980) . Reduced reproductive performance in androgen-resistant tfm/tfm female mice. P hiloso~hical Transactions of the Roval Society of London: Series B : Bioloqical Sciences . 208 , 1-12.

Lyon, M.F., Hawkes, S.G. (1 970) . X-Linked gene for testicular feminization in the mouse. Nature(Londonl. 227 , 1217 - 1219 . Martini, L., Melcangi, R.C., Maggi, R. (1993) . Androgen and progesterone metabolism in the central and peripheral nervous systern. Journal of Steroid Biochernistrv and Molecular Bioloav .47(1-6) , 195 - 205. MacLusky, N.J., Naftolin, F. (1981). Sexual differentiation of the central nervous system. Science .211 , 1294 - 1302.

Matsumoto, A., Prins, G.S.,Arai, Y. (1 996). Androgenic regulation of androgen receptor imrnunoreactivity in motoneurons of the spinal nucleus of the bulbocavernosus of male rats. Journal of Neuroendocrinoloav . -8(7), 553-559.

Maxson, S.C. (1 997). Sex determination in genetic rnechanisms for mammalian brain and behavior. Biomedical Reviews. 7, 85 - 90.

McEwen, B.S. (1991) . Non-genomic and genomic effects of steroids on neural activity. Trends in Pharmacoloaical Science . , 141-1 47. McGinnis, M.Y., Katz, S.E. (1996) . Sex differences in cytosolic androgen receptors in gonadectomized male and female rats. Journal of Neuroendocrinoloqy .8 , 193 - 197. Meisel, R.L., Ward, I.L. (1981) . Fetal female rats are masculinized by male littermates located caudally in the uterus. Science. 21 3 , 239-242.

Monahan, E.P., Breedlove, S.M. (1991) . Brain sites projecting to the spinal nucleus of the bulbocavernosus . The Journal of Comparative Neuroloqy . 307 , 370-374.

Monk, M., Harper, M., l., (1 979) . Sequential X chromosome inactivation coupled with cellular differentiation in early mouse embryos. Nature (London) . -281,371 - 313. Moussaoul, S., Duval, P., Lenoir, V., Garret, C., Kerdelhue, 8. (1996) . CGRP in the trigeminal nucleus, spinal cord and hypothalamus: Effect of gonadal steroids. Neuro~e~tides. a, 546-50. Naess, O., Haug, E., Attrarnadal, A., Aakvaag, A., Hansson, V., French, F. (1 976). Androgen receptors in the anterior pituitary and central nervous system of the androgen insensitive" (tfm) rat : Correlation between receptor binding and effects of androgens on gonadotropin secretion . Endocrinoloov . 99(5) , 1295 - 1303.

Neumann, F., Elger, W., Kramer, M. (1966) . Development of a vagina in male rats by inhibiting androgen receptors with an anti-androgen during the critical phase of organogenesis. Endocrinoloov. B,628 - 632. New, H.V., Mudge, A.W. (1 986) . Calcitonin gene-related peptide regulates muscle acetylcholine receptor synthesis . Nature. London . 323(30), 809 - 81 1. Nordeen, E.J., Nordeen, K.W., Sengelaub, D.R., Arnold, A.P. (1985) . Androgens prevent normally occurring cell death in a sexually dimorphic spinal nucleus . Science . 229 ,671 - 672.

Ogawa, S., Pfaff, D.W.,Korach, K.S., Lubahn, D.B. (1 997) . Behavioral effects of estrogen receptor gene disruption in male mice . Proceedinas of the National Academv of Science (USA) .94(4) , 1476-1481.

Olsen, K.L., (1 979a) . Induction of male mating behavior in androgen- insensitive (tfm) and normal (King-Holtzmann) male rats: Effects of testosterone proprionate, estradiol benzoate, and dihydrotestosterone. Hormones and Behavior -l3,66 - 84. Olsen, K.L. (1 979b) . Androgen-insensitive rats are defeminized by their testes. Nature, London .279 ,238 - 239.

Olsen, K.L., Whalen, R.E. (1982) . Estrogen binds to hypothalarnic nuclei of androgen-insensitive (tfm) rats. Ex~erientia. 38(l) , 139 - 140 . Onuf, B. (1 899) . On the arrangement and function of the cell groups in the sacral regioi; of the spinal cord . Journal of Nervous and Mental Disease .a: 498 - 504. Phoenix, C., Goy, R.W., Gerall, A.A., Young, W.C. (1 959) , Organizing action of prenatally administered testosterone proprionate on the tissue mediating behavior in the female guinea pig. Endocrinoloov. a,369 - 382.

Popper, P., Micevych, P.E. (1 989a) . Localization of calcitonin gene-related peptide and its receptors in a striated muscle. Brain Research . 496 , 180 - 186. Popper, P., Micevych, P.E. (1989b) . The effect of castration on calcitonin gene - related peptide in spinal motor neurons. Neuroendocrinoloov . 50,338 - 343.

Popper, P., Micevych, P.E. (1 990) : Steroid regulation of calcitonin gene - related peptide mRNA expression in motoneurons of the spinal nucleus of the bulbocavernosus . Molecular Brain Research .B , 159 - 166. Popper, P., Abelson, L., Micevych, P.E. (1992) . Differential regulation of alpha- calcitonin gene-related peptide and preprocholcystokinin messenger RNA expression in alpha - motoneurons . Neuroscience . a,87 - 96. Popper, P., Ulbarri, C., Micevych, P.€. (1 992) . Muscle activity regulates the expression of calcitonin gene-related peptide in motoneuronç of the spinal nucleus of the bulbocavernosus. Molecular Brain Research -13,43 - 51.

Quarmby, V., E., Wilson, E., M., French, F., S., Lubahn, D., B., Yaibrough, W., G. (1 990) . Autologous down-regulation of androgen receptor messenger ribonucleic acid . Molecular Endocrinoloqy . 4m ,22-8 . Rissman, E.F., Wersinger, S.R., Taylor, J.A., Lubahn, D.B. (1997) . Estrogen receptor function as revealed by knockout studies : Neuroendocrine and behavioral aspects. Hormones and Behavior . a,232 - 243.

Roselli, C.E., Fasasi, T.A. (1 992) , Estradiol increases the duration of nuclear androgen receptor occupation in the preoptic area of the male rat treated with dihydrotestosterone. Journal of Steroid Biochemistrv and Molecular Bioloav .

742(2) , 161-1 68.

Rosenfeld, M.G., Mermod, J.J., Amara, S.G., Swanson, L.W., Sawchenko, P.E., Rivier, J., Vale, W. W., Evans, R.M. (1983) . Production of a novel neuropeptide encoded by the CT gene via tissue specific RNA processing . Nature. London . 304 , 129 - 135.

Sachs, B.D. (1 982) . Role of striated penile muscles in penile reflexes, copulation and induction of pregnancy in the rat. Journal of Reproduction and Fertilitv . 66 , 433 - 443.

Sachs, B.D., Pollak., E.I., Krieger, M.S., Barfield, R.J. (1 973) . Sexual behavior: Normal patterning in androgenized female rats. Science . 181, 770- 771.

Sachs, B.D., Thomas, D.A. (1 985) . Differentiai effects of perinatal androgen treatment on sexually dimorphic characteristics in rats . Phvsioloov and Behavior . a,735 - 742. Sagrillo, C.A., Grattan, D.R. McCarthy, M.M. Selmanoff, M. (1996) . Hormonal and neurotransmitter regulation of GnRH gene expression and related reproductive behaviors . Behavior Genetics .26(3), 241 - 277.

Sato, S., Burgess, SB.,Mcllwain, D.L. (1 994) . Transcription and motoneuron size. Journal of Neurochernistrv. , 1609 - 1615.

Sengelaub, D.R., Jordan, C.L., Kurz, E.M., Arnold, A.P. (1 989a) . Hormonal control of neuron number in sexually dimorphic spinal nuclei of the rat : II. Development of the spinal nucleus of the bulbocavernosus in androgen- insensitive (tfm) rats. The Journal of Comparative Neuroloov . 280 , 630 - 636. Sengelaub, D.R., Nordeen, E.J., Nordeen, K.W., Arnold, A.P. (1989b) . Hormonal control of neuron number in sexually dimorphic spinal nuclei of the rat : III. Differential effects of the androgen dihyrotestosterone. Journal of Com~arative Neuroloav. 280, 637 - 644. Shapiro, B. H., Goldman, A.S., Gustafsson, J-A. (1975) . Masculine-like hypothalamic-pituitary axis in the androgen-insensitive genetically male rat pseudohermaphrodite. Endocrinoloav .97(2), 487 - 492.

Shrenker, P., Maxson, SC. (1983) . The genetics of hormonal influence on male sexual behavior of mice and rats. Neuroscience and Biobehavioral Reviews . -7 ,349-59. Shroeder, H.D. (1 980) . Organization of the motoneurons innervating the pelvic muscles of the male rat. The Journal of Com~arativeNeuroloav . 192 , 567 - 587.

Simerly, RB., Chang, C., Muramatsu , Swanson, LW. (1990) . Distribution of androgen and estrogen receptor mRNA - containing cells in the rat brain : An in situ hybridization study. The Journal of Comparative Neuroloov . 294 , 76 - 95. Smith, E.R., Damassa, D.A., Davidson, J.M. (1 977) . Hormone administration: Peripheral and intracranial implants. In R.D. Myer (Ed.), Methods in Psvchobioloav (V0i.3) . New York: Academic Press.

Sodersten, P. (1 973) . Estrogen-activated sexual behavior in male rats. Hormones and Behavior . 4,247 - 256.

Stocco, D.M. (1 997) . A StAR search: Implications in controlling steroidogenesis. Bioloov of Re~roduction. 56 , 328 - 336. Stumpf, W.E., Sar, M. (1976) . Steroid hormone target sites in the brain: The differential distribution of estrogen, progestin, androgen and glucocorticosteroid . Journal of Steroid Biochemistrv . z(l1-12) , 1163-70. Tan, J-A., Marschke, &B., Ho, K-C., Perry, S.,T., Wilson, E.,W., French, F.,S. (1 992). Response elements of the androgen- regu!ated CE gene. The Journal of Bioloqical Chemistw. 267(7) , 4456 - 4466.

Truss, M., Beato, M. (1993) . Steroid hormone receptors: Interactions with deoxyribonucleic acid and transcription factors. Endocrine Reviews. 14(4) , 459 - 479.

Tsujimoto, T., Kuno, M. (1988) . Calcitonin gene-related peptide prevents disuse-induced sprouting of rat motor nerve terminais. The Journal of Neuroscience. g(10) , 3951 - 3957. Uno, M., Kobayashi, S., Ishihara, S., Ehara, H., Deguchi, T. (1998). Prostatic cancer 30 years atter bilateral orchidectomy . British Journal of Uroloav . 81 (3) , 506-7.

van Goozen. S., H., van Engeland, H., Thijssen, J., H., Cohen-Kettenis, P., T., Matthys, W. (1998). Adrenal androgens and aggression in conduct disorder prepubertal boys and normal controls. Bioloaical Psvchiatry . 43(21, 156-8 .

Wallace, S.,J.,R., Hart, B.,L. (1983). The role of the striated penile muscles of the male rat in seminal plug dislodgement and deposition. Phvsioloav & Behavior. 31 , 815-21. Ward, I.,L. (1969) . Differential effect of pre- and postnatal androgen on the sexual behavior of intact and spayed female rats. Hormones and Behavior . 1 , 25-36.

Ward, O., B., Ward, I., L., Eckert, M., A., Carlucci, J., R., Wexler, A., M., (1 996) . Critical periods of sensitivity of sexually dimorphic spinal nuclei to prenatal testosterone exposure in fernale rats. Hormones and Behavior . 30(4) , 407-1 5 . Warner, M., Gustafsson, J-A. (1995) . Cytochrome P450 in the brain : Neuroendocrine functions. Frontiers in Neuroendocrinoloqv .x, 224 - 236.

Watson, R.E., Weigand, S.J., Clough, R.W., & Hoffman, G.E. (1 986) . Use of cryoprotectant to maintain long-term peptide immunoreactivity and tissue morphology. Peptides, 1,1 55-1 59.

Watson N. ,V., Freeman, L.,M., Breedlove, S.,M. (1 996) . Neuronal size in the spinal nucleus of the bulbocavernosus : Direct modulation by androgen in animals with mosaic genetic androgen insensitivity. Societv for Neuroscience Abstracts -26, 697.

Weisz, Jey Ward, I.,L. (1 980) . Plasma testosterone and progesterone titers of pregnant rats, their male and female fetuses and neonatal offspring. Endocrinoloqv . 106 ,306 - 316. Wehling, M. (1997) . Specific nongenomic actions of steroid hormones. Annual Review of Phvsioloav . 59,365 - 93.

Whalen, R.,E., Rezek, D.,L. (1974). Inhibition of lordosis in female rats by subcutaneous implants of testosterone, androdtenedione or dihydrotestosterone in infancy. Hormones and Behavior . S , 125-128. Wimalawansa, S.J.(1996) . Calcitonin gene-related peptide and its receptors: Mo lecular genetics, p hysiology, pathop hysiology, and therapeutic potentials. Endocrine Reviews. II , 533-85. Wong, C-l., Zhou, 2-X., Sar, M., Wilson, E.,M. (1 994) . Steroid requirement for androgen receptor dimerization and DNA binding. Journal of Biolooical Chernistry 268 , 19004 - 1901 2.

Wood, RJ., Newman, S.,W. (1 995) . Androgen and estrogen receptors coexist within individual neurons in the brain of the çyrian hamster. Neuroendocrinolopv . 0 ,487 - 497.

Xu, J., O'Malley, B., W., Tsai, M., J., Tsai, S., Y., DeMayo, F., J., Qiu, Y. (1998) . Partial hormone resistance in mice with disruption of the steroid receptor coactivator-1 (SRC-1) gene. Science . 279153581 , 1922 - 5.

Yarbourough, W.,&, Quarmby, V.,E., Simental, J., A., Joseph, D., R., Sart, M., Lubahn, D.,B., Olsen, K.,L., French, F., S., Wilson, E.,M. (1990). A single base mutation in the androgen receptor gene causes androgen insensitivity in the testicular ferninized rat. Journal of Bioloqicai Chemistrv. 265 , 8893 - 8900.

Yu, W-H., A., McGinnis, M.,Y. (1986). Androgen receptor levels in cranial nerve nuclei and tongue muscles in rats. The Journal of Nauroscience. 6(5),1302 - 1307.

Yuri, K., Kawata, M. (1 992). Semi-quantitative analysis of the effects of estrogen on CGRP- and methionine-enkephalin-immunoreactivity in the periventricular preoptic nucleus and the medial preoptic area of female rats. Brain Research . 578, 282-8.

Zaidi, M., Moonga, B.S., Bevis, P.J., Bascal, Z., A., Breimer, L.,H. (1 990). The calcitonin gene peptides : Biology and clinical relevence. Critical Reviews in Clinical Laboraton, Sciences . -(2), IO9 - 173.

Zorrilla, P., Sirnard, J., Rheaume, E., Labrie, F., Pelletier, G. (1990). Multihormonal control of pre-pro-somatostatin mRNA levels in the periventricular nucleus of the male and female rat hypothalamus. Neuroendocrinoloav. 52, 527 - 536. IMAGE NALUATION TEST TARGET (QA-3)

APPLIED 4 INLAGE , lnc -.-.-- 1653 East Main Street --- - Rochester, NY 14609 USA --= --= -- Phone: 71 61482-0300 __I--- , V F= 716/288-5989

O 1993. Applied Image. lm. Ail Rights Rasewed