by Donald Pfaff
Ribeiro, A. C., S. Musatov, A. Shteyler, S. Simanduyev, I. Arrieta-Cruz, S. Ogawa, and D. W. Pfaff. 2012. “siRNA Silencing of Estrogen Receptor-α Expression Specifically in Medial Preoptic Area Neurons Abolishes Maternal Care in Female Mice.” Proceedings of the National Academy of Sciences of the United States of America 109 (40): 16324–16329.
Schumacher, M., H. Coirini, D. W. Pfaff, and B. S. McEwen. 1990. “Behavioral Effects of Progesterone Associated with Rapid Modulation of Oxytocin Receptors.” Science 250: 691–694.
7
BRAIN–BODY RELATIONS
Problem: The need to put together brain mechanisms for an entire chain of behaviors that are harmonized with organs throughout the rest of the body. This chapter follows the story through body-space and body-time. Body-wide adaptations must be understood in order to satisfy one of my main points: body and brain are coordinated in the regulation of vertebrate behaviors, including social behaviors.
Through Space: Coordination Body-Wide
This book focuses on brain and behavior, but I like to contemplate the virtual symphony of biological and physiological processes, an orchestration that coordinates the preparations for as well as the results of reproduction. It all starts with the egg. Here I rely on David Albertini writing in the Physiology of Reproduction (2015). The egg is a unique cell in that is, to use a term from stem cell biology, totipotent. That is, with appropriate microsignaling from its environment and various different programs of transcription factors, the egg can be transformed after fertilization into any cell type in the body.
In the ovary an egg grows from a very small cell, the product of meiosis, to a much larger “activated” cell located in an ovarian follicle. The early stages of growth do not depend on pituitary gonadotrophic hormones, but the latest of the stages do. During the preovulatory stage of the cycle of any mammal, in addition to signaling by follicle-stimulating hormone (FSH), estrogens work through both estrogen receptors, α and β (ER-α and ER-β), to regulate follicular development. Estrogens work not just by making follicles grow faster (a confusing situation because ER knockouts lack negative feedback on the pituitary) but by regulating normal follicular and luteal structures in the ovary and by providing an adequate number of oocytes available for ovulation.
Estrogenic effects in the uterus are equally dramatic. They determine the uterine estrous cycle, which means that they cause massive uterine growth before the time of ovulation and that they are required for fertility. Females lacking progesterone receptors have normal uterine development but also are infertile (Binder et al. 2015). Some of the reasons for infertility include defects in implantation of the fertilized egg.
The uterus will be prepared for implantation, necessary for further embryonic development, by the previously described hormone effects but also by the secretions of other chemicals from endometrial glands, compounds such as prostaglandins and interferons. Implantation includes the attachment of the blastocyst to the uterine wall, followed by cell proliferation and vascularization.
Organs concerned with energy balance are also involved. Neural and hormonal mechanisms required for ovulation are highly likely to depend on nutrition. For young women, the classic findings by Rose Frisch and others have stated that a certain amount of body fat is required for complete pubertal development. This makes sense, biologically—how could a mother adequately nurse her baby if she herself is starving? The endocrine and molecular signals of a female’s current metabolic state to the reproductive system are currently being investigated.
Spanish cell biologist Manuel Tena-Sempere (2015) gives much credit to the metabolic cue leptin. Leptin, a hormone produced in fat cells, was cloned by Jeffrey Friedman and his team at Rockefeller University, and has been implicated in body weight control and, importantly, in diabetes. If Rose Frisch’s hypothesis is correct, Friedman’s discovery may prove as important for the physiology of reproduction as it does for weight reduction. Taking into account that patients with abnormally low amounts of leptin suffer from a delay or absence of puberty, I think their suppositions may well be correct. Tena-Sempere is interested in investigating how such metabolic signals get to the kisspeptin neurons, which allow gonadotropin-releasing hormone (GnRH) neurons to do their job. For example, the messenger RNA (mRNA) encoding the leptin receptor is expressed by those neurons in the arcuate nucleus of the hypothalamus; low leptin levels cause reduced kisspeptin mRNA expression. Leptin injections in low-leptin animal models cause increased kisspeptin expression. Leptin is a major player, but Tena-Sempere cautions that many other routes of metabolic signaling also are in play. That is, not just body fat but several intestinal organs are likely involved.
Meanwhile, the anterior pituitary gland is regulated by hormones and by the central nervous system (CNS), both in a tonic (steady over time) fashion and in a dynamic fashion that permits ovulation. Elaborate experiments by Jon Levine and his students, then at Northwestern University, proved that the negative feedback effects of ovarian steroid hormones are exerted both at the level of the pituitary itself and in the hypothalamus. Neuroendocrine regulation of ovulation, as summarized by Plant and Zeleznik, is something completely different. At the end of the follicular phase of estrus or the menstrual cycle after prolonged estrogen exposure, the negative feedback actions are flipped into so-called positive feedback actions. That is, more and prolonged ovarian sex steroids cause gonadotropin release, which fosters still more sex steroid release. Difficult experiments with monkeys suggest strongly that estrogens acting just at the pituitary can be sufficient to cause an ovulatory luteinizing hormone surge. In the rodent, however, it is clear that GnRH neurons in the preoptic area and anterior hypothalamus provide a signal that is superimposed on pituitary mechanisms and may be accompanied by increased sensitivity to GnRH at the level of the pituitary.
In light of this short sketch of nonbehavioral components in reproduction, consider the adaptations of breast tissue for feeding the baby once born. Obviously, mammary tissue is essential to the definition of being a female mammal. Steven Anderson, at the University of Colorado Medical School has described the cell proliferation that begins in breast tissue during the earliest days of pregnancy, and the beginning of lactogenesis during midpregnancy (Anderson et al. 2015). The hormonal determinants of successful lactation are complex. Not only estrogens and progestins but also prolactin, oxytocin, and growth hormone are required to achieve successful breast feeding. Insulin signaling and insulin-like growth factors also play important roles.
Thus, even before considering the neuraxis, it is clear that tissues throughout the body must be harmonized in their preparation for successful reproduction.
Behavioral Components
Just as the nonbehavioral components of well-organized reproduction stretch from one end of the body to the other, so do the behavioral components. The adequate sensory stimulus for lordosis behavior, as detailed in Chapter 2, is light cutaneous pressure on the skin of the flanks and also the rump in the region of the genitalia. We had pursued the question systematically and electrophysiologically in experimental animals (Chapter 2), then later on we had the benefit of working with cell biologist Nieves Martín-Alguacil and surgeon Justine Schober, who obtained appropriate access to human tissue (Schober et al. 2015). Little information was available regarding the sensory nerve endings within the glabrous skin of the external female genitalia in human patients. The diversity of possible sensations suggests a variety of receptor types. Comprehensive knowledge of the sensory stimuli, including stimulus positions, changes in temperature, and pressure and pain, is critical for addressing pain and sexual function disorders clinically. So the aim of our neurohistological study was to document the presence and characteristics of cutaneous sensory receptors in human female genital tissue.
The protein PGP 9.5, localized immunocytochemically, was the most sensitive neural marker for identifying cutaneous sensory receptors. Free nerve endings in the papillary dermis appeared as thin fibers that were varicose, branched, or s
ingle processed, straight or bent. In the labia minora, free nerve endings were identified in the strata basale, spinosum, and granulosum of the epidermis. Noncapsulated (Meissner-like) corpuscles in the dermal papillae interdigitated with epidermal ridges of the skin. Capsulated corpuscles protruded from the deep dermis into the epidermis. Encapsulated corpuscles and cells located in the inner and outer cores were strongly positive for PGP 9.5. We concluded that Meissner’s corpuscles and Pacinian corpuscles are present in the female labia minora and exhibit characteristic staining patterns. We could infer that the cutaneous sensory receptors important for sexual behaviors were quite similar between humans and laboratory animals.
To address questions of hormone sensitivity, cell biologist Nieves Martín-Alguacil determined the presence of ER and neuronal nitric oxide synthase (nNOS) in the mouse clitoris (Martín-Alguacil et al. 2008a). A series of sections of the pelvic area, including the preputial glands and clitoris, were assessed by immunocytochemical studies specific for ER-α, ER-β, and nNOS. ER-α was detected in the epithelium of the gland of the clitoris and in the glandular tissue and preputial and apocrine gland. ER-α was also detected in the nuclei of stromal cells around the cavernous tissue and near the epithelium of the clitoris. Cytoplasm ER-α was detected in a few cells in an area ventral to the clitoral gland. There was also nuclear staining in the connective tissue cells surrounding the clitoris.
Very light ER-β immunostaining was detected in the clitoris and in the tissue related to it. There were some cells with nuclear staining in the vessels of the cavernous tissue of the clitoris. Immunostaining of nNOS was detected in the clitoris, the preputial gland, and the connective tissue. We concluded that the ER-α and -β isoforms and nNOS are present in the clitoris and preputial glands of female mice in different cellular locations and with differing levels of expression.
What about projections from the skin regions important for sexual behavior into the spinal cord? Nieves Martín-Alguacil considered that, although genital tactile stimulation is regarded as a precursor to sexual arousal (see Chapter 8) and a recognized initiator of CNS arousal, the specific afferent neural pathways that transmit sensory stimuli of arousal needed neuroanatomical definition. That is, limited knowledge existed of the pathway from the cutaneous receptors of nerves originating in the epithelial tissue of the clitoris and continuing to spinal cord afferents. Accordingly, we defined the neural pathways and mechanisms responsible for arousal originating in the epithelium of the clitoris as well as related neural pathways to the spinal cord in mice (Martín-Alguacil et al. 2008b).
Gross anatomy of the mouse clitoris showed that pudendal and hypogastric nerves have a major role in the innervation of the external genitalia. Neuronal tracing revealed that the greatest neural projection density was in L5 / L6 of the spinal cord. The distribution extended from S1 to L2 with no labeling seen in L3 of the spinal cord. Wheat germ agglutinin–horseradish peroxidase labeling was seen caudal in levels S1 through L4 and rostral in L2. Later in this chapter I give several examples of the conservation of mechanisms as we move from laboratory animals to humans. The surgeon on our team, Justine Schober, was of the opinion that these findings in mice would provide a valuable tool for the study of sexual arousal disorders and the further understanding of sexual function related to neural pathologies and trauma.
Furthering the argument of this chapter that body-wide adaptations must be understood to satisfy body-brain-mind coordination in the regulation of vertebrate behaviors—one of the main points of this book—I will briefly recapitulate the relevant claims in Chapter 2. Once the adequate signal for lordosis behavior reaches the dorsal horn of the spinal cord, the electrophysiological signal becomes tailored to serve the purpose of reproductive behavior control. That is, not all aspects of the initial signal are retained, and some point-by-point regional specificity is lost. In this respect our data coincided with the message of Patrick Wall, the sensory neurophysiologist with the Massachusetts Institute of Technology (MIT) and University College London, about transformations in the contents of sensory messages as one moves from the skin through the neuraxis to the cerebral cortex. For lordosis behavior, of course, the most important ascending projections are to the medullary reticular formation and to the midbrain reticular formation and central grey. Cells there, in the midbrain, will have been turned on by hormone-dependent signals from the ventromedial hypothalamus (VMH).
Now let us consider the hypothalamus (reviewed in Fink et al. 2012). In Chapter 1, I recounted our discovery of sex hormone-binding neurons there. A fuller description for the purpose of this body-wide integrative chapter would include a brief description of two roles for GnRH neurons (see Chapter 5). The pulsatile outputs of GnRH neurons 1) foster the ovulatory release of luteinizing hormone from the anterior pituitary gland and 2) prime VMH neurons to foster lordosis behavior. Brain / behavior and hormonal dynamics are synchronized. In the VMH, preovulatory estrogens will have increased transcription rates for several mRNAs (Chapter 3). Experimental intervention in those transcriptional systems have been proven them important for female reproductive behaviors (Chapter 4).
Once midbrain central grey neurons have been activated by VMH hormone-dependent signals, their elevated electrical activity primes lateral vestibulospinal and medial vestibulospinal neurons. They, in turn, provide excitatory synaptic inputs both directly and through interneurons. Thus, from the skin on the rump, up the neuraxis, all the way to the forebrain, and thence back down to motoneurons, the peripheral and central nervous systems are orchestrated to produce a complete, natural behavior of supreme biological importance. In turn, CNS actions are harmonized with preparations in all the other previously mentioned organs. Various physiological systems are well integrated with each other.
Through Body: Time
Of course, sexuality starts during the first trimester of pregnancy when the SRY gene on the Y chromosome—discovered by Robin Lovell-Badge at Francis Crick Institute at Mill Hill—expresses in such a manner as to kick off a cascade of Sox genes, thus fostering growth of the testes (the medulla in an otherwise indifferent gonad). Although we could discuss very small sex differences in some behaviors in the womb or in the early prenatal period, the real action starts with puberty.
During and after puberty, physiological and behavioral changes are initiated that are central to the maintenance of the species. From the perspective of this book, the main point is that we can prove that an entire sequence of behaviors—from initial arousal (see Chapter 8) through courtship, mating, and maternal behavior—not only depend on regulated expression of the ER-α gene but, more generally, have had their mechanisms elucidated in detail.
Puberty. Given that all mammals approach the time of puberty with sex differences in some brain mechanisms, how do we describe the crucial steps for producing the pubertal changes that will be necessary for all adult reproductive function? Between male and female, the management of the female mammal’s reproductive cycle is more complex.
Sergio Ojeda, head of a laboratory at the Oregon National Primate Research Center, has led mechanistic studies to examine the physiology of puberty in the female at the cellular and molecular levels (Lomniczi et al. 2015). Puberty in the female requires the proper positioning of GnRH neurons, regulated GnRH synthesis, and subsequent pulsatile release of GnRH into the portal circulation, thus causing luteinizing hormone and follicle-stimulating hormone to be released from the pituitary. Regulation of these GnRH neurons involves both an array of excitatory inputs and inhibitory inputs. Excitatory inputs classically have been understood to come from glutamatergic synapses, but recently a peptide called kisspeptin has also been recognized as important. In the human brain, kisspeptin neurons are found in the arcuate nucleus of the hypothalamus. We know they are important because the loss of either the kisspeptin gene or its receptor in patients results in the failure of puberty to come about.
Obviously, one source of inhibitory inputs is supplied by GABA (γ-aminobutyric acid) neurons—in fact, thes
e bind radioactive estrogens and may carry the main explanatory burden for the estrogenic control of GnRH transcription (see Chapter 3) and release. Opioid inputs also inhibit GnRH pulsatile release. The newly recognized inhibitory inputs include RFamide-related peptide, now thought to be the “mammalian ortholog” of a gonadotropin inhibiting factor discovered in birds. Thus, GnRH neuronal activities are regulated in a push-pull fashion, always a good engineering design.
Ojeda and his colleagues were the first to call attention to a central role for glia in the initiation of puberty (reviewed in Lomniczi et al. 2015). One causal route for glial actions depends on the actions of several growth factors. The other causal route uses cell adhesion molecules, centered on the sialyted form of neural cell adhesion molecule (NCAM).
It is clear from the foregoing that not only the gene encoding GnRH and its receptor are important for the regulation of changes during puberty but also several other genes, some of whose failure can simply block puberty. At a more sophisticated level of molecular thinking, Ojeda and his colleagues are studying gene networks that are crucial for transcriptional repression. For example, in addition to other transcriptional repressors he has identified the Polycomb group as a nuclear silencing complex that represses kisspeptin transcription. As a consequence, he recognizes three nuclear proteins in the so-called POK group as increasing pubertal development in the female rat. They do so by inhibiting the Polycomb group so that pubertal changes can be initiated by repressing the repressor.
Inevitably, these transcriptional dynamics will come to be shown as the products of the chemistry of epigenetics. A superficial approach would say that DNA methylation can be considered as relatively simple, but recent studies have called into question the single-factor interpretation that DNA methylation simply represses transcription. Histone N-terminus modifications were already seen as complex; Rockefeller University’s David Allis has made the points 1) in these histones every amino acid counts, 2) an individual N-terminus can have both transcriptional facilitatory and repressive marks, and 3) in some cases the temporal order of histone modifications may matter; in other cases, as far as can be told, they happen simultaneously.