How the Vertebrate Brain Regulates Behavior
Page 1
HOW THE VERTEBRATE BRAIN REGULATES BEHAVIOR
DIRECT FROM THE LAB
DONALD PFAFF
Cambridge, Massachusetts & London, England
2017
Copyright © 2017 by the President and Fellows of Harvard College
All rights reserved
Cover Design: Graciela Galup
Cover photo: Ryan McVay / Thinkstock
978-0-674-66031-1 (alk. paper)
978-0-674-97877-5 (EPUB)
978-0-674-97876-8 (MOBI)
978-0-674-97875-1 (PDF)
The Library of Congress has cataloged the printed edition as follows:
Names: Pfaff, Donald W., 1939– author.
Title: How the vertebrate brain regulates behavior : direct from the lab / Donald Pfaff.
Description: Cambridge, Massachusetts : Harvard University Press, 2017. | Includes bibliographical references and index.
Identifiers: LCCN 2016042682
Subjects: LCSH: Brain. | Mammals—Behavior. | Neurophysiology. | Neuroendocrinology. | Molecular neurobiology.
Classification: LCC QP376 .P447 2017 | DDC 573.8/619—dc23
LC record available at https://lccn.loc.gov/2016042682
CONTENTS
Introduction
1. Hormone Receptors
2. Discovering the Neural Circuit for a Vertebrate Behavior Essential to Reproduction
3. Hormonal Regulation of Gene Expression in the Brain
4. Genes Regulating Behavior
5. Neuropeptide: Gonadotropin-Releasing Hormone
6. Neuropeptide: Oxytocin
7. Brain–Body Relations
8. Central Nervous System Arousal Fueling Instinctive Behaviors
9. Sex Difference
10. Summary
Acknowledgments
Index
INTRODUCTION
When I began in neurobiology, there was a longstanding problem: most scientists felt that it was necessary to work with extremely simple animals to understand how nervous systems govern behavior. Vertebrates and especially mammals were considered to be too hard, too complex. A new paradigm was needed to attack the scientific problems of how the vertebrate brain regulates behavior. To fill that need, I analyzed and then exploited the effects of hormones on behavior. In doing so I reaped the analytic advantages of chemical specificity (steroid hormones) and neuronal specificity (circuitry for a simple mammalian behavior) so as to understand how a biologically important mammalian behavior is regulated.
To put it another way, a neuroscientist might win by studying complex functions in a simple organism or by studying a simple behavior in a complex organism. I chose the latter. Having done so, I succeeded in linking molecular chemistry to physiology and ultimately to the causation of behavior. In recounting my own laboratory’s work I am going through all the steps that any laboratory must to solve problems as our science of the vertebrate brain develops.
In the first several chapters of this scientific account, a review of some of my laboratory’s accomplishments, I will show that we produced the first demonstrations of specific molecular changes in particular neurons that drive a chain of social behaviors of extraordinary biologic importance. That is, I proved that the female’s behavioral response to the male’s mount, which is essential for fertilization (lordosis behavior), is driven by specific estrogen-dependent molecular events in ventromedial hypothalamic neurons, which regulate the neuronal circuit that we subsequently worked out.
As mentioned, the first creative step was to find a problem worth solving that, in fact, would turn out to be solvable: how the chain of behaviors essential for reproduction is organized and regulated. My findings, starting 50 years ago, are related here, together with the work of others who have enlarged the field of neuroendocrinology to the point where it presents the best opportunities for relating molecular genetics to neuroscience and behavior.
Some people think that mechanistic explanations for behavior should be set up against evolutionary explanations. No! I note that my type of work—discovering brain mechanisms for behavior—does not contradict those who emphasize how behavioral regulation has evolved, an aspect of biology and medicine led by E. O. Wilson, at Harvard (Wilson 1975). My field of neuroscience is complementary to Wilson’s field, including his seminal work on sociobiology. Wilson thinks about how biologically important behaviors have evolved through time, while I have discovered mechanisms that are the result of that evolutionary process and drive behaviors right now. In fact, it is those very mechanisms that we neuroscientists study, which actually evolve!
To put it another way, I figured out how the nervous system works by solving the problem posed by a specific behavioral function. Rather than asking “How does the brain work?” I worked out how a specific function is accomplished. I had the advantage of working with a hormone-controlled behavior, so I could “triangulate” brain mechanisms, viewing them from one angle as hormone targets and from another as behavioral response producers. By Chapter 5 in this book, you will see that we achieved a realization of these brain mechanisms in physical terms.
I also note that although traditionally neuroscientists considered the central nervous system to be an isolated entity and studied it as such, in the course of my analyses of lordosis behavior mechanisms I was studying a behavior that is necessarily and essentially social.
At MIT
Again, the first and most important issue a scientist faces is to decide on the topic to investigate. What questions are deeply interesting yet solvable? As the great immunologist Sir Peter Medawar said, “Science is the art of the soluble.”
That is, when I started laboratory research the core problem was to figure out how to solve the mysteries of how higher, vertebrate brains regulate entire normal behaviors. For a long time people could study simple nervous systems. So the choice was between examining complex behavior in simple animals and trying to unravel mechanisms for simple behaviors in complex animals. Eric Kandel (1965, 1976), a Nobel Prize–winning professor at Columbia, did the former. I figured out how to do the latter.
A seven-word sentence, uttered more than 50 years ago, combined with study in organic chemistry did the trick. As a Harvard undergraduate I had volunteered to serve on the back ward of a state mental hospital. I wanted to be a psychiatrist or neurologist. But the medical care in that ward was terrible. My surgeon father would not have approved. Then when I went to the Massachusetts Institute of Technology (MIT) for graduate school I was helping my thesis advisor, Joseph Altman (who had discovered postnatal neurogenesis), trace how newly divided neurons migrate to their final positions in the rat brain. Thinking back to my medical intentions, I said, “Joe, I’d like to do something more closely related to brain function”—that is, to behavior. He answered in seven words: “I’d like to do something with hormones.”
Well, I had been a good student in organic chemistry and knew quite well that steroids are not such complex molecules! I could not resist once I realized that some important hormones are steroids, and I found that some steroid hormones control simple, biologically crucial animal behaviors. The core problem to be solved was how two major communication systems in the body—in all vertebrate bodies, including humans—signal to each other to regulate biologically crucial behaviors in biologically adaptive ways. This book gives the essence of the solution.
This Book
The findings recounted in this book redefined what neuroscientists studying the mammalian brain can accomplish. Using different levels of investigation—single genes, ion channels, single nerve cell physiology, neural circuits, and an entire social behavior—we built a behavioral mechanism. Taken togeth
er, these sets of discoveries proved for the first time exactly how specific chemicals (hormones) acting on specific nerve cell groups can determine a complete, natural mammalian behavioral response.
Chapter 1 summarizes my discovery about 50 years ago of hormone receptors in the brain. The limbic-hypothalamic system of neurons with nuclear receptors for sex hormones was discovered in the rat brain but turned out to be universal among vertebrate brains—“from fish to philosopher.”
Chapter 2 explains how the hormone-sensitive hypothalamic neurons summarized in Chapter 1 proved to be one of three entry points for working out the first neural circuit for a vertebrate behavior. The other two were the sensory inputs (followed neuroanatomically and neurophysiologically) and the motor outputs (from muscles back to motoneurons and so forth). Applying the logic of Sherringtonian physiology (Sherrington, 1947) but using modern methods and working with a large laboratory group over several years, we were able to demonstrate a spinal-midbrain-spinal circuit for lordosis behavior, a circuit that is regulated by an estrogen-dependent output from the hypothalamus: the first circuit for a vertebrate behavior and a social behavior at that. Principle: the hierarchically organized circuit features modules, each of which contributes its unique physiological regulation.
Then we got lucky. The hormone receptors I had discovered in the brain turned out to be ligand-activated transcription factors. Thus, I could use molecular techniques to study hormone-facilitated gene transcription and, in turn, link the eventual products of such transcription to reproductive behavior. Chapter 3 covers this work. The product of this work, showing how transcription of specific genes in specific neurons leads to an estrogen-dependent hypothalamic output that regulates a defined neural circuit—thus producing a biologically crucial behavior—has been called a high-water mark in neuroscience.
Chapter 4 deals with gene expression causally linked to the reproductive behaviors mentioned previously. Taken together, these four advances—hormone receptors, neural circuitry for female behavior, hormone effects on gene expression, and gene / behavior causal relations—proved for the first time exactly how specific chemicals acting in specific parts of the brain could determine a complete (vertebrate) behavioral response. The behavior explained is a social behavior essential for reproduction.
Chapters 5 and 6 deal with specific genes involved in lordosis behavior regulation, genes expressing gonadotropin-releasing hormone (GnRH) and oxytocin, respectively. The GnRH discoveries revealed, surprisingly, how a simple decapeptide regulates reproduction body-wide. Then Chapter 7 wraps up the story of the unity of the body, explaining neural and behavioral controls of reproduction and discussing seven principles of hormone / behavior relations. Neural and endocrine systems are united.
I have argued elsewhere that, deep to the sexual arousal that fosters mating behavior, there operates a fundamental neuronal “force,” a concept called generalized arousal. In Chapter 8, I carry that argument forward because I feel that elevated generalized arousal is probably essential for initiating all motivated behaviors.
Sex differences in brain and behavior (Chapter 9) are crucial, obviously, for reproductive physiology; for the neurobiology of the normal brain, remote from reproduction, they are of minimal importance. For certain maladies, though, such as autism, a new science of the importance of sex difference is emerging. Arthur Arnold, at the University of California–Los Angeles, works on these subjects, examining the implications of reproductive biology for sexually differentiated phenomena, all significant for medicine and public health.
The final chapter, Chapter 10, summarizes discoveries that turned out to be universal among vertebrates.
The subtitle of this book, “Direct from the Laboratory,” reflects the vibrancy and immediacy of my field of work. Actually two fields of science developed as recounted here. First, the neuroscience of the vertebrate brain and behavior developed over the last 50 years. Torsten Wiesel and David Hubel opened up the visual cortex. Sten Grillner (2005, 2007) and Anders Lundberg elucidated motor mechanisms. Eric Kandel (2012) and Paul Greengard explained mechanisms of learning and reward. In my case developing the science of neurobiology consisted of applying a Sherringtonian ([1906] 1947) type of logic and work to the long-standing problem of solving a mammalian behavior. Second, within neuroscience, neuroendocrinology used to be thought of as a “boutique” field, amazingly, despite its role to unite two major signaling systems in the body, neural and hormonal. Now we recognize that hormone actions on brain and behavior offer an “intellectual interstate highway” to the linkage of modern molecular biology and genomics to neuroscience by virtue of the types of nuclear hormone receptors (ligand activated transcription factors) that I discovered in brain tissue (Chapter 1).
So, putting it another way, 50 years ago I decided to understand how the nervous system works by solving the problem posed by a specific behavioral function. I had the advantage of working with a hormone-controlled behavior—detailed in Chapter 1—so I could “triangulate” brain mechanisms, viewing them from one angle as hormone targets and (Chapter 2) from another angle as behavioral response producers. Chapters 3 and 4 add the genetics and genomics. By Chapter 5 in this book you will see that we achieved a realization of these brain mechanisms for a reproductive behavior in physical terms. Yes, it is possible to analyze fully the mechanisms for a vertebrate behavior.
Further Reading
Grillner, S., A. Kozlov, P. Dario, C. Stefanini, A. Menciassi, A. Lansner, and J. Hellgren Kotaleski. 2007. “Modeling a Vertebrate Motor System: Pattern Generation, Steering and Control of Body Orientation.” Progress in Brain Research 165: 221–34.
Grillner, S., H. Markram, E. De Schutter, G. Silberberg, and F. E. LeBeau. 2005. “Microcircuits in Action—From CPGs to Neocortex.” Trends in Neurosciences 28 (10): 525–33.
Kandel, E. R. 1976. Cellular Basis of Behavior. San Francisco: Freeman.
Kandel, E. R., J. H. Schwartz, T. M. Jessell, S. A. Siegelbaum, and A. J. Hudspeth, eds. 2012. Principles of Neural Science. 5th edition. New York: McGraw-Hill.
Kandel, E. R., and L. Tauc. 1965. “Heterosynaptic Facilitation in Neurones of the Abdominal Ganglion of Aplysia depilans.” Journal of Physiology 181 (1): 1–27.
Sherrington, C. S. (1906) 1947. The Integrative Action of the Nervous System. Reprint, New Haven, CT: Yale University Press.
Wilson, E. O. 1975. Sociobiology: The New Synthesis. Cambridge, MA: Harvard University Press.
1
HORMONE RECEPTORS
Problem: Two major signaling systems in the body are the endocrine system and nervous system. How do they communicate? Does this question have anything to do with the brain’s regulation of behavior?
We all have heard physicists say something like this: “You guys in biology and medicine don’t have laws or principles. You have piles of phenomena accompanied by lots of Latin names.” In this chapter, you will see, once I had made the discovery of hormone receptors in the brain, to answer the question above, I plowed on with brains representing every major vertebrate class, using every species for which I could find excellent collaborators—people who knew that species well. By doing that, we established a lawfulness, a massive degree of reliability for my initial finding. Here is the historical backdrop.
The British physiologist Geoffrey Harris had, during the late 1930s, established that neurons in the hypothalamus regulate secretions from the anterior pituitary gland, which in turn controlled the release of protein hormones that would circulate to the ovaries, testes, thyroid, and adrenal glands. Thus, communication of the nervous system to the endocrine system. But what about the reverse? Hormone effects on brain and behavior?
First, based on my interests, I had to establish the existence and specificity of sex hormone actions on behavior (Pfaff 1970b). This was an exercise in “endocrine engineering.” The subjects were ovariectomized female rats and castrated male rats. Indeed, estradiol injections turned on the ability of the females for lordosis behavior, the primary and ess
ential female mating behavior, especially if they were supplemented by a progesterone injection about four hours before the lordosis behavioral assay. I noted that testosterone injections supplemented by progesterone also could elevate lordosis behavior performance, probably (as later shown) because aromatase enzymes in the brain can transform testosterone into estrogen. Likewise, in the males either testosterone or estradiol significantly increased all masculine mating behaviors (such as mounts).
Thus, the existence and strength of steroid hormone effects on sex behaviors were established, and an additional fact popped out of the data. The same females that showed the greatest amount of feminine behavior under one hormonal condition also showed the most under other hormonal conditions. The correlations were as high as 0.88 and 0.90. This indicated that in addition to straightforward hormonal determination of sex behavior, there must have been a tissue sensitivity factor that varied from animal to animal. The level and efficiency of estrogen receptor (ER) binding, transport, and affinity for estrogen response elements on the promoters of estrogen-responsive genes would later be seen as providing the reason for these tissue sensitivity phenomena.
Then for the purposes of behavioral control, I asked, are direct actions of hormones on the brain really necessary? Perhaps estrogenic actions on the pituitary would be of primary importance, and pituitary hormones feeding back upon the brain would provide the main controls over behavior. I addressed these questions using animals in which the pituitary gland had been surgically removed (Pfaff 1970a). The answer was clear. Estrogen and progesterone injections increased female sex behavior (lordosis) performance in hypophysectomized ovariectomized female rats as well as in animals with an intact pituitary gland. Likewise, testosterone injections increased male sex behavior vigor in hypophysectomized castrated male rats as well as in males with intact pituitaries. Thus, direct actions of sex hormones on neurons were indicated. How does that happen?