The discovery that the hypothalamus is responsible for controlling both reproductive hormones and behavior suggested various mechanisms by which hormonal and behavioral cycles are inexorably linked, and even co-regulated. While our knowledge about this process has grown dramatically, our understanding of the essential control circuits that operate during normal reproduction, or fail in abnormal functioning, is still limited. Various neurotransmitter candidates have been proposed as essential elements of the systems that regulate reproduction. However, few are involved in so many aspects of reproduction as are the opioid peptides, which play a critical or supporting role in (a) controlling hormonal cycling in females (Akabori and Barraclough 1986; Kalra 1985; Wiesner et al. 1984), (b) regulating reproductive behavior in males (Hughes et al. 1988; Matuszewich and Dornan 1992; Myers and Baum 1979) and females (Pfaus and Pfaff 1992; Sirinathsinghji 1986; Wiesner and Moss 1986a), and even (c) modulating mesolimbic dopamine release mediated by reinforcing sexually relevant olfactory stimuli (Mitchell and Gratton 1991).

Gonadal steroid regulation of hypothalamic opioids represents an important feedback system by which to control reproduction. Hypothalamic (opioid) circuits regulate hormonal releasing hormones that control pituitary secretion. This regulation in turn affects gonadal steroid hormones, which act centrally to alter opioid function and facilitate reproductive behavior. Since such feedback is vitally important for the regulation of hormonal and behavioral events during the estrous cycle, most of this discussion will be limited to opioid action in females. Many of the experiments that we will describe utilized models of hormone manipulation to investigate natural regulation of hypothalamic opioid systems in animals, primarily rodents.

The expression of sexually differentiated patterns of behavior is a characteristic of many vertebrate species and often correlates with sex differences in the relative abundance of neurons in brain regions thought to control such behaviors. In general, two fundamental processes determine the number of neurons that survive into adulthood. First, changes in the number of neuroblasts formed in the ventricular zone can occur in response to mechanisms that are intrinsic to a particular population of cells or that are controlled by extrinsic factors such as cell–cell interactions, neuronal growth factors, and circulating hormones. Second, similar extrinsic cellular and hormonal factors can determine the number of cells of a particular lineage that reach their permanent destination, establish appropriate connections, and achieve a regionally specific functional phenotype. Sex steroid hormones can affect both processes but appear to exert their most pronounced influences on neuronal development during a restricted perinatal critical period (Arnold and Gorski 1984; Breedlove 1986; Goy and McEwen 1980; Harris and Levine 1965; Rhees et al. 1990). Thus, treatment of female neonates with sex steroids during the first few postnatal days alters the number of neurons residing in certain nuclei, as well as the morphology, synaptology, and neurotransmitter expression of individual neurons (Arai et al. 1986; Arnold and Jordan 1988; De Vries 1990; Gorski 1985; Raisman and Field 1973; Simerly 1989, 1991).

Introduction

Gonadal steroid hormone receptors are ligand-activated transcription factors that are part of a complex superfamily of such factors (Evans 1988). These receptors alter the transcription of genes containing specific promoter or enhancer sequences. In this way, estrogen, acting through the estrogen receptor (ER), alters the transcription of genes in the cells where the receptors are found.

Neurons that contain gonadal steroid hormone receptors are a key to the mechanisms through which these hormones govern the behavioral and neuroendocrine processes underlying reproduction (Morrell et al. 1975; Morrell and Pfaff 1983). A complex and only partly understood cascade of events occurs subsequent to the genomic regulation initiated by these ligand-activated transcription factors (Yamamoto 1985). The presence of gonadal steroid hormone receptors in neurons has been documented by means of steroid hormone autoradiography, biochemical assays for binding, and immunocytochemistry (Blaustein and Olster 1989; DonCarlos et al. 1991; Giordano et al. 1991; Morrell et al. 1992). Now the tools of molecular biology provide a means of investigating the mRNA from which gonadal steroid hormone receptors are translated.

The differential sensitivity of brain regions to steroid hormones is based on the combination of the regional location of neurons containing these receptors, the number of neurons containing the receptors per brain region, and the number of receptors per neuron. The degree of sensitivity to steroid hormones is not a static property of the brain as there is increasing evidence that the endocrine and behavioral status of the adult mammal can govern the sensitivity of brain regions to steroid hormones by regulating either the number of neurons expressing the receptors or the amount of receptor per neuron (Hnatczuk et al. 1994; Koch and Ehret 1989; Pearson et al. 1993; Simerly and Young 1991).

Many books are subject to the fundamental questions “Why this topic?” and “Why now?” Scientific texts are perhaps most susceptible because they often present similar topics. As a partial answer to these questions, we paraphrase P. B. Medawar in his Advise to a Young Scientist: We have tried to prepare the kind of book that we ourselves would like to read and have as a reference.

In recent years, the field of reproductive neuroendocrinology has experienced a renaissance brought about by the application of cellular and molecular biological techniques. We have made significant progress in understanding the mechanisms that underlie central nervous system control of reproductive behavior. This progress has been well documented at various meetings and in individual papers. We felt it was necessary, therefore, to offer a collection of essays by some of those who have contributed to this renaissance. We hasten to add that the chapters in this volume do not necessarily reflect all of the vital issues of behavioral neuroendocrinology. Rather, they represent brief reviews by and current data from a number of productive scientists in this field.

Because of a limitation of space, several important topics are not discussed or are only briefly presented in this volume. These include the spinal nucleus of the bulbocavernosus system, cell membrane steroid receptors, interactions of steroids with γ-aminobutyric acid receptors, the songbird neural circuitry, as well as the insect and amphibian models of reproduction and metamorphosis. Each of these models has proved to be extremely useful for studying the effects of sex steroid hormones on the nervous system.

Introduction

For neuroscientists, the study of sex differences in the brain promises at least two benefits. Investigations of their development can elucidate the processes that form brain structure during ontogeny that generates specific functions and behaviors, while investigations of the functional significance of these sex differences can reveal how brain morphology and function are related. Except for the fact that sex-related differences in the number of spinal motoneurons have been linked to sex-related differences in the number of specific muscle cells (Kelley 1988; Breedlove 1992), these benefits have been difficult to achieve, however. The complexity of the neuroanatomical connections to and from the brain regions where these differences are found and technical difficulties in manipulating specific sexually dimorphic elements in these areas have delayed the desired result.

This complexity, however, can be exploited. Given that all brain areas contain heterogeneous populations of cells and inputs, focusing on the neurotransmitter content of cells and inputs could reveal whether sexual differentiation selectively affects particular cell populations. This, in turn, could facilitate our understanding of the cellular processes underlying differentiation. Focusing on the neurotransmitter content may also help to reveal the anatomical connections of sexually dimorphic areas, and therefore to assess the impact of a particular dimorphism on other brain areas. Finally, knowing the neurotransmitter systems involved would allow specific manipulation of sexually dimorphic elements by applying specific agonists and antagonists (De Vries 1990).

Steroid hormones act on the brain to influence its organization during development and its activity in adulthood, thereby regulating behavior and physiology. Achieving these effects on the brain is often the culmination of a complex set of events within the endocrine system. During development, the sequence begins with the differentiation of steroid-secreting organs and their expression of steroidsynthetic enzymes. In adulthood, it continues with the regulation of the activities of one or more of these enzymes by pituitary trophic factors. After secretion, but before the steroid reaches targets within specific brain cells, the hormone is subject to a variety of regulatory influences. These can include steps to-inactivate the molecule by peripheral catabolism and excretion. The presence of carrier proteins in blood can limit the availability of free steroid to enter tissues. Having reached a target organ, the steroid may encounter additional enzymes that catalyze changes in its structure, rendering the molecule inactive. Alternatively, the steroid may be converted to a molecule with increased biological activity or one that functions along an alternative steroid-activating pathway. When these transformations are complete, the steroid is available to influence cellular function by interacting with intracellular protein receptors. Once bound to ligand, the steroid receptor can bind to specific DNA hormone response elements to influence the transcription of specific genes. The active steroid may also influence cell function without changing gene expression by interacting directly with cell membranes or with other cellular processes.

Role of the medial preoptic area in the hormonal control of male sexual behavior

Mating behaviors of males provide excellent examples of the neurobiological effects of sex steroid hormones. In mammals (Larsson 1979; Michael and Bonsall 1979; Sachs and Meisel 1988), birds (Balthazart 1983; Silver et al. 1979), and other vertebrates (Crews 1979; Crews and Silver 1985; Kelley and Pfaff 1978), testosterone (T) promotes the sexual activity of adult males. Gonadally intact males are more likely to mount receptive females, and to copulate to ejaculation, than are castrated males, unless the castrates are given T. Although these stimulatory effects of T result largely from its action on the brain (Kelley and Pfaff 1978; Larsson 1979; Sachs and Meisel 1988), T also acts on the penis and spinal neurons to affect copulatory performance (Breedlove 1984; Hart 1978; Hart and Leedy 1985; Sachs 1983).

Within the brain, one of the most important areas mediating the effects of T on male sex behavior is the medial preoptic area (MPOA) or the MPOA–anterior hypothalamus (AH) continuum. The MPOA–AH contains many cells that accumulate T or its metabolites, estradiol (E) and dihydrotestosterone (DHT) (Kelley and Pfaff 1978; Luine and McEwen 1985; Michael and Bonsall 1990). However, the conclusion that the MPOA–AH mediates many of the effects of T on male sexual behavior is based on the behavioral changes produced by manipulations of this area (Sachs and Meisel, 1988). Implanting T or E into the MPOA–AH restores mounting in castrated males, and lesioning the MPOA–AH eliminates sexual behavior in males that are exposed to circulating T.

Acetylcholine was the first endogenous chemical to be identified as a neurotransmitter. In addition to its vital role in physiology, acetylcholine has been implicated in the regulation of mammalian behaviors that range from reflexive (Potter et al. 1990) to regulatory (Hagan et al. 1987) to cognitive (Levin et al. 1990). The control of these heterogeneous behaviors appears to be possible because of diffuse cholinergic circuits distributed throughout the mammalian brain (Mesulam et al. 1983). The organization of these systems, particularly in the forebrain and midbrain, places cholinergic neurons in regions involved in sensory, motor, and motivational processes.

Our work during the past decade has indicated that cholinergic systems also play an important role in the regulation of certain behaviors exhibited by mammalian females during mating. It is well known that the sexual behaviors of female rodents are controlled closely by steroid hormones, primarily estrogen and progesterone secreted by the ovaries. However, the sequence of neural events initiated by these hormones to cause complex behavioral responses has not been described fully. The capacity of cholinergic mechanisms to affect rodent sexual behavior suggests a key interface between endocrine activity and cholinergic function. Our primary hypothesis states that ovarian steroids regulate brain function, and consequently behavior, by altering the activity of cholinergic systems within neural target structures. According to current theories of hormone action, steroids could access nuclear genomes (O'Malley and Means 1974) and surface membranes (Schumacher 1990) to alter the nature and number of various proteins associated with cholinergic neurotransmission.

Introduction

Appropriate development and maintenance of neurons and their interconnections are fundamental to central nervous system (CNS) function and plasticity in mammals. The molecules that regulate these processes during development as well as in mature neurons are largely unknown and minimally characterized. Two classes of substances implicated in the regulation of development, survival, and neurotransmitter production of neurons in telencephalic regions that subserve learning, memory, and general cognition are estrogen and the family of endogenous growth- and survival-promoting proteins, the neurotrophins, nerve growth factor (NGF), brain-derived neurotrophin factor (BDNF), neurotrophin-3 (NT-3), and neurotrophin-4/5 (NT-4/5).

The organization of neural circuits controlling a broad spectrum of behavioral functions is permanently influenced by exposure of the developing CNS to gonadal steroid hormones: estrogens and androgens. Exposure to these steroids during pre- and postnatal “critical periods” profoundly and permanently influences the organization of the nervous system. The consequences of this hormonal exposure are evident in striking sexual dimorphisms of neuronal structure and function, which underlie the observed sex differences in behaviors, including those related to cognition. The cellular and molecular mechanisms by which gonadal steroids permanently influence the structural and functional organization of the developing brain remain ill-defined, although it seems clear that effects on cellular aspects of neuronal differentiation are likely to be involved (reviewed by Juraska 1991; Toran-Allerand 1991a). Estrogen enhances the growth and differentiation of neurons and their processes (neurites) within the developing hypothalamus, preoptic area, and cerebral cortex in vivo and in vitro (reviewed by Toran-Allerand 1984, 1991b), as well as in steroid target regions of the adult CNS following injury through steroid deprivation, axotomy, or deafferentation (Matsumoto and Arai 1981; Frankfurt et al. 1990).

Introduction

Ovarian hormones have many cellular actions in the central nervous system that result in changes of behaviors and reproductive physiology (Blaustein and Olster 1989). One approach to unraveling the cellular processes by which ovarian hormones act on the brain has been the study of hormonal regulation of female sexual behavior. While the induction of such behavior usually requires stimulation by ovarian hormones, the specific hormonal conditions required for the stimulation and inhibition of sexual behavior vary in accord with the antecedent hormonal conditions in each species. These hormonal conditions include patterns as different as the sequential presence of estradiol and progesterone in rats and guinea pigs (Dempsey et al. 1936; Boling and Blandau 1939), the sequential presence of progesterone and estradiol in sheep (Robinson 1954), the presence of estradiol alone in prairie voles (Dluzen and Carter 1979), and the presence of testosterone metabolized neuronally to estradiol in musk shrews (Rissman 1991). We have studied the hormonal regulation of sexual behavior in rats and guinea pigs by the sequential presence of estradiol and progesterone. In ovariectomized guinea pigs injected with estradiol and progesterone, as during the estrous cycle, the period of sexual receptivity lasts for approximately 8 hours. While the specific cellular endpoints may vary in each of these species, the fundamental cellular processes by which hormones act are likely to be similar in all species.

A new generation of neurobiological work beyond the classical demonstrations of hormone–behavior relations by Beach, Young, and their colleagues has provided the launching platform for serious molecular analyses of some of the neuroendocrine mechanisms involved. Following the determination of cellular targets for estradiol in the brain (Pfaff 1968), a circuit for estrogen-dependent lordosis behavior was elucidated (Pfaff 1980). Principles of steroid hormone/central nervous system/behavior mechanisms include the apparent universality among vertebrates of certain nuclear binding features; the conservation of many endocrine and biochemical reactions from simpler non-neural tissues; the modular construction of the neural circuit for reproductive behavior; and the economic use of neural information within the circuit on both the sensory and motor sides (Pfaff et al. 1994). The explosion of molecular techniques and knowledge surrounding nuclear hormone receptors enables us to analyze brain mechanisms further for steroid-influenced behaviors using mRNA hybridization, genomic structure, protein-DNA binding, in vitro transcription, antisense DNA, and neurotrophic viral vector technologies.

Receptor structure

Steroid hormones influence mammalian tissues by promoting cell development and differentiation as well as acute cellular functions. In target tissues, steroids act by entering the cell and binding to specific receptor proteins. These receptors, which belong to a class of ligand-inducible transcription factors, regulate transcription by interacting with short cis-acting DNA elements – termed hormone response elements (HRE) – in the promoter regions of specific genes (Carson-Jurica et al. 1990; Yamamoto 1985).

Introduction

Steroid hormones produced in the gonads are a prerequisite for mating behavior in the male Syrian hamster, as in most male mammals. While a substantial body of early research was directed toward defining the hormonal requirements (both quantity and identity) for this behavior, more recent studies have focused on hormone metabolism, receptor distribution, and mechanisms of steroid action in the central nervous system (CNS). This chapter explores the transduction of hormonal signals by steroid receptor–containing neurons to facilitate sexual behavior in the male hamster.

Hormonal requirements for copulation

Copulation in the male Syrian hamster

The sequence of copulation in the male Syrian hamster has been reviewed previously (Siegel 1985; Sachs and Meisel 1988). However, a brief description of the behaviors expressed during mating would be helpful for understanding the critical role that hormones play in maintaining this activity. Figure 1.1 illustrates sexual behavior over a 10-minute period in a sexually experienced male. Initial contact with a receptive female is characterized by investigation of the female's head and flank, followed by extensive sniffing and licking of the anogenital region. Through this activity, the male receives chemosensory stimulation via the vomeronasal organ and olfactory mucosa. The ability to perceive chemosensory signals and the integrity of the neural pathways that transmit chemosensory stimuli are essential for copulation, because disruption of chemosensory cues immediately and permanently abolishes mating (reviewed by Sachs and Meisel 1988). In a sexually experienced male, anogenital investigation of the female is followed shortly by a series of mounts and intromissions, interspersed with brief (1- to 2-second) grooming of the penis and perineum.