Bear Chapter 2 PDF: Neurons and Glia

CHAPTER TWO Neurons and Glia INTRODUCTION THE NEURON DOCTRINE The Golgi Stain Cajal’s Contribution BOX 2.1 OF SPECIAL INTEREST: Advances in Microscopy THE PROTOTYPICAL NEURON The Som...

CHAPTER TWO Neurons and Glia INTRODUCTION THE NEURON DOCTRINE The Golgi Stain Cajal’s Contribution BOX 2.1 OF SPECIAL INTEREST: Advances in Microscopy THE PROTOTYPICAL NEURON The Soma The Nucleus Neuronal Genes, Genetic Variation, and Genetic Engineering BOX 2.2 BRAIN FOOD: Expressing One’s Mind in the Post-Genomic Era BOX 2.3 PATH OF DISCOVERY: Gene Targeting in Mice, by Mario Capecchi Rough Endoplasmic Reticulum Smooth Endoplasmic Reticulum and the Golgi Apparatus The Mitochondrion The Neuronal Membrane The Cytoskeleton Microtubules BOX 2.4 OF SPECIAL INTEREST: Alzheimer’s Disease and the Neuronal Cytoskeleton Microfilaments Neurofilaments The Axon The Axon Terminal The Synapse Axoplasmic Transport BOX 2.5 OF SPECIAL INTEREST: Hitching a Ride with Retrograde Transport Dendrites BOX 2.6 OF SPECIAL INTEREST: Intellectual Disability and Dendritic Spines CLASSIFYING NEURONS Classification Based on Neuronal Structure Number of Neurites Dendrites Connections Axon Length Classification Based on Gene Expression BOX 2.7 BRAIN FOOD: Understanding Neuronal Structure and Function with Incredible Cre GLIA Astrocytes Myelinating Glia Other Non-Neuronal Cells CONCLUDING REMARKS 23 023–054_Bear_02_revised_final.indd 23 12/20/14 2:58 AM  24 PART ONE FOUNDATIONS INTRODUCTION INTRODUC CTION All tissues and organs in the body consist of cells. The specialized func- tions of cells and how they interact determine the functions of organs. The brain is an organ—to be sure, the most sophisticated and complex organ that nature has devised. But the basic strategy for unraveling its functions is no different from that used to investigate the pancreas or the lung. We must begin by learning how brain cells work individually and then see how they are assembled to work together. In neuroscience, there is no need to separate mind from brain; once we fully understand the individual and concerted actions of brain cells, we will understand our mental abilities. The organization of this book reflects this “neurophiloso- phy.” We start with the cells of the nervous system—their structure, func- tion, and means of communication. In later chapters, we will explore how these cells are assembled into circuits that mediate sensation, perception, movement, speech, and emotion. This chapter focuses on the structure of the different types of cells in the nervous system: neurons and glia. These are broad categories, within which are many types of cells that differ in structure, chemistry, and function. Nonetheless, the distinction between neurons and glia is impor- tant. Although there are approximately equal numbers of neurons and glia in the adult human brain (roughly 85 billion of each type), neurons are responsible for most of the unique functions of the brain. It is the neurons that sense changes in the environment, communicate these changes to other neurons, and command the body’s responses to these sensations. Glia, or glial cells, contribute to brain function mainly by insulating, supporting, and nourishing neighboring neurons. If the brain were a chocolate chip cookie and the neurons were chocolate chips, the glia would be the cookie dough that fills all the other space and suspends the chips in their appropriate locations. Indeed, the term glia is derived from the Greek word for “glue,” giving the impression that the main function of these cells is to keep the brain from running out of our ears! Although this simple view belies the importance of glial function, as we shall see later in this chapter, we are confident that neurons perform most information processing in the brain, so neurons receive most of our attention. Neuroscience, like other fields, has a language all its own. To use this language, you must learn the vocabulary. After you have read this chap- ter, take a few minutes to review the key terms list and make sure you understand the meaning of each term. Your neuroscience vocabulary will grow as you work your way through the book. THE NEURON NEURO ON DOCTRINE To study the structure of brain cells, scientists have had to overcome several obstacles. The first was the small size. Most cells are in the range of 0.01–0.05 mm in diameter. The tip of an unsharpened pencil lead is about 2 mm across; neurons are 40–200 times smaller. (For a review of the metric system, see Table 2.1.) Because neurons cannot be seen by the naked eye, cellular neuroscience could not progress before the development of the compound microscope in the late seventeenth century. Even then, obstacles remained. To observe brain tissue using a microscope, it was necessary to make very thin slices, ideally not much thicker than the diameter of the cells. However, brain tissue has a con- sistency like a bowl of Jell-O: not firm enough to make thin slices. Thus, 023–054_Bear_02_revised_final.indd 24 12/20/14 2:58 AM  CHAPTER 2 NEURONS AND GLIA 25 TABLE 2.1 Units of Size in the Metric System Unit Unit Abbr Ab Abbreviation brev evia iati tion on Mete Me Meter terr Eq Equi Equivalent uiva vale lent nt Real Re Real-World al-Wor World ld E Equ Equivalent quiv ival alen entt Kilometer km 103 m About two-thirds of a mile Meter m 1m About 3 feet Centimeter cm 10⫺2 m Thickness of your little finger Millimeter mm 10⫺3 m Thickness of your toenail Micrometer ␮m 10⫺6 m Near the limit of resolution for the light microscope Nanometer nm 10⫺9 m Near the limit of resolution for the electron microscope the anatomical study of brain cells had to await a method to harden the tissue without disturbing its structure and an instrument that could produce very thin slices. Early in the nineteenth century, scientists dis- covered how to harden, or “fix,” tissues by immersing them in formalde- hyde, and they developed a special device called a microtome to make very thin slices. These technical advances spawned the field of histology, the micro- scopic study of the structure of tissues. But scientists studying brain structure faced yet another obstacle. Freshly prepared brain tissue has a uniform, cream-colored appearance under the microscope, with no differences in pigmentation to enable histologists to resolve individual cells. The final breakthrough in neurohistology was the introduction of stains that selectively color some, but not all, parts of the cells in brain tissue. One stain still used today was introduced by the German neurologist Franz Nissl in the late nineteenth century. Nissl showed that a class of basic dyes would stain the nuclei of all cells as well as clumps of material surrounding the nuclei of neurons (Figure 2.1). These clumps are called Nissl bodies, and the stain is known as the Nissl stain. The Nissl stain is extremely useful for two reasons: It distinguishes between neurons and glia, and it enables histologists to study the arrangement, or cytoarchi- tecture, of neurons in different parts of the brain. (The prefix cyto- is from the Greek word for “cell.”) The study of cytoarchitecture led to the realization that the brain consists of many specialized regions. We now know that each region performs a different function. The Golgi Stain The Nissl stain, however, could not tell the whole story. A Nissl-stained neuron looks like little more than a lump of protoplasm containing a ▲ FIGURE 2.1 Nissl-stained neurons. A thin slice of brain tissue has been stained with cresyl violet, a Nissl stain. The clumps of deeply stained ma- terial around the cell nuclei are Nissl bodies. (Source: Hammersen, 1980, Fig. 493.) 023–054_Bear_02_revised_final.indd 25 12/20/14 2:58 AM  26 PART ONE FOUNDATIONS nucleus. Neurons are much more than that, but how much more was not recognized before Italian histologist Camillo Golgi devised a new method (Figure 2.2). In 1873, Golgi discovered that soaking brain tis- sue in a silver chromate solution, now called the Golgi stain, makes a small percentage of neurons become darkly colored in their entirety (Figure 2.3). This revealed that the neuronal cell body, the region of the neuron around the nucleus that is shown with the Nissl stain, is actu- ally only a small fraction of the total structure of the neuron. Notice in Figures 2.1 and 2.3 how different histological stains can provide strik- ingly different views of the same tissue. Today, neurohistology remains an active field in neuroscience, along with its credo: “The gain in brain is mainly in the stain.” The Golgi stain shows that neurons have at least two distinguishable parts: a central region that contains the cell nucleus and numerous thin tubes that radiate away from the central region. The swollen region con- taining the cell nucleus has several names that are used interchangeably: cell body, soma (plural: somata), and perikaryon (plural: perikarya). The thin tubes that radiate from the soma are called neurites and are of ▲ FIGURE 2.2 Camillo Golgi (1843–1926). two types: axons and dendrites (Figure 2.4). (Source: Finger, 1994, Fig. 3.22.) The cell body usually gives rise to a single axon. The axon is of uni- form diameter throughout its length, and any branches from it generally extend at right angles. Because axons can extend over great distances in the body (a meter or more), histologists of the day immediately recog- nized that axons must act like “wires” that carry the output of the neu- rons. Dendrites, on the other hand, are rarely longer than 2 mm. Many dendrites extend from the cell body and generally taper to a fine point. Soma Dendrites Neurites Axon ▲ FIGURE 2.3 Golgi-stained neurons. (Source: Hubel, 1988, p. 126.) ▲ FIGURE 2.4 The basic parts of a neuron. 023–054_Bear_02_revised_final.indd 26 12/20/14 2:58 AM  CHAPTER 2 NEURONS AND GLIA 27 Early histologists recognized that because dendrites come in contact with many axons, they must act as the antennae of the neuron to receive in- coming signals, or input. Cajal’s Contribution Golgi invented the stain, but a Spanish contemporary used it to great- est effect. Santiago Ramón y Cajal was a skilled histologist and artist who learned about Golgi’s method in 1888 (Figure 2.5). In a remark- able series of publications over the next 25 years, Cajal used the Golgi stain to work out the circuitry of many regions of the brain (Figure 2.6). Curiously, Golgi and Cajal drew completely opposite conclusions about neurons. Golgi championed the view that the neurites of different cells are fused together to form a continuous reticulum, or network, similar to the arteries and veins of the circulatory system. According to this ▲ FIGURE 2.5 reticular theory, the brain is an exception to the cell theory, which Santiago Ramón y Cajal (1852–1934). states that the individual cell is the elementary functional unit of (Source: Finger, 1994, Fig. 3.26.) all animal tissues. Cajal, on the other hand, argued forcefully that the neurites of different neurons are not continuous with each other and communicate by contact, not continuity. This idea that cell theory also applies to neurons came to be known as the neuron doctrine. Although Golgi and Cajal shared the Nobel Prize in 1906, they re- mained rivals to the end. The scientific evidence over the next 50 years strongly supported the neuron doctrine, but final proof had to wait for the electron microscope in the 1950s (Box 2.1). With the increased resolving power of the electron microscope, it was finally possible to show that the neurites of different neurons are not continuous with one another (Figure 2.7). Thus, our start- ing point in the exploration of the brain must be the individual neuron. ▲ FIGURE 2.6 One of Cajal’s many drawings of brain circuitry. The letters label the different ele- ments Cajal identified in an area of the human cerebral cortex that controls voluntary move- ment. We will learn more about this part of the brain in Chapter 14. (Source: DeFelipe and Jones, 1998, Fig. 90.) 023–054_Bear_02_revised_final.indd 27 12/20/14 2:58 AM  28 PART ONE FOUNDATIONS BOX 2.1 OF SPECIAL INTEREST Advances in Microscopy T he human eye can distinguish two points only if they are separated by more than about one-tenth of a millimeter sensitive detectors, and the computer takes these data and reconstructs the image of the neuron. Unlike the traditional (100 ␮m). Thus, we can say that 100 ␮m is near the limit methods of light and electron microscopy, which require tis- of resolution for the unaided eye. Neurons have a diameter sue fixation, these new techniques give neuroscientists the of about 20 ␮m, and neurites can be as small as a frac- ability to peer into brain tissue that is still alive. Furthermore, tion of a micrometer. The light microscope, therefore, was a they have allowed “super-resolution” imaging, breaking the necessary development before neuronal structure could be limits imposed by traditional light microscopy to reveal struc- studied. But this type of microscopy has a theoretical limit tures as small as 20 nm across. imposed by the properties of microscope lenses and visible light. With the standard light microscope, the limit of resolu- tion is about 0.1 ␮m. Because the space between neurons is only 0.02 ␮m (20 nm), it’s no wonder that two esteemed scientists, Golgi and Cajal, disagreed about whether neurites were continuous from one cell to the next. This question could not be answered until about 70 years ago when the electron microscope was developed and applied to biological specimens. The electron microscope uses an electron beam instead of light to form images, dramatically increasing the resolving power. The limit of resolution for an electron microscope is about 0.1 nm—a million times better than the unaided eye and a thousand times better than a light microscope. Our insights into the fine structure of the inside of neurons—the ultrastructure—have all come from electron microscopic ex- amination of the brain. Today, microscopes on the leading edge of technology use laser beams to illuminate tissue and computers to cre- Figure A ate digital images (Figure A). Neuroscientists now routinely A laser microscope and computer display of a fluorescent neuron introduce into neurons molecules that fluoresce when il- and dendrites. (Source: Dr. Miquel Bosch, Massachusetts Institute luminated by laser light. The fluorescence is recorded by of Technology.) ▲ FIGURE 2.7 Neurites in contact, not continuity. These neurites were reconstructed from a series of images made using an electron microscope. The axon (colored yellow) is in contact with a dendrite (colored blue). (Source: Courtesy of Dr. Sebastian Seung, Princeton University, and Kris Krug, Pop Tech.) 023–054_Bear_02_revised_final.indd 28 12/20/14 2:58 AM  CHAPTER 2 NEURONS AND GLIA 29 THE P PROTOTYPICAL ROTO OTYPICAL NEURON N As we have seen, the neuron (also called a nerve cell) consists of several parts: the soma, the dendrites, and the axon. The inside of the neuron is separated from the outside by the neuronal membrane, which lies like a circus tent on an intricate internal scaffolding, giving each part of the cell its special three-dimensional appearance. Let’s explore the inside of the neuron and learn about the functions of the different parts (Figure 2.8). The Soma We begin our tour at the soma, the roughly spherical central part of the neuron. The cell body of the typical neuron is about 20 ␮m in diameter. The watery fluid inside the cell, called the cytosol, is a salty, potassium- rich solution that is separated from the outside by the neuronal mem- brane. Within the soma are a number of membrane-enclosed structures called organelles. The cell body of the neuron contains the same organelles found in all animal cells. The most important ones are the nucleus, the rough endo- plasmic reticulum, the smooth endoplasmic reticulum, the Golgi appara- tus, and the mitochondria. Everything contained within the confines of the cell membrane, including the organelles but excluding the nucleus, is referred to collectively as the cytoplasm. The Nucleus. Its name derived from the Latin word for “nut,” the nucleus of the cell is spherical, centrally located, and about 5–10 ␮m across. It is contained within a double membrane called the nuclear envelope. The nuclear envelope is perforated by pores about 0.1 ␮m across. Within the nucleus are chromosomes which contain the genetic ma- terial DNA (deoxyribonucleic acid). Your DNA was passed on to you from your parents and it contains the blueprint for your entire body. The DNA in each of your neurons is the same, and it is the same as the DNA in the cells of your liver and kidney and other organs. What distinguishes a neuron from a liver cell are the specific parts of the DNA that are used to assemble the cell. These segments of DNA are called genes. Each chromosome contains an uninterrupted double-strand braid of DNA, 2 nm wide. If the DNA from the 46 human chromosomes were laid out straight, end to end, it would measure more than 2 m in length. If we were to compare this total length of DNA to the total string of letters that make up this book, the genes would be analogous to the individual words. Genes are from 0.1 to several micrometers in length. The “reading” of the DNA is known as gene expression. The final product of gene expression is the synthesis of molecules called proteins, which exist in a wide variety of shapes and sizes, perform many different functions, and bestow upon neurons virtually all of their unique charac- teristics. Protein synthesis, the assembly of protein molecules, occurs in the cytoplasm. Because the DNA never leaves the nucleus, an interme- diary must carry the genetic message to the sites of protein synthesis in the cytoplasm. This function is performed by another long molecule called messenger ribonucleic acid, or mRNA. mRNA consists of four differ- ent nucleic acids strung together in various sequences to form a chain. The detailed sequence of the nucleic acids in the chain represents the information in the gene, just as the sequence of letters gives meaning to a written word. The process of assembling a piece of mRNA that contains the informa- tion of a gene is called transcription, and the resulting mRNA is called 023–054_Bear_02_revised_final.indd 29 12/20/14 2:58 AM  30 PART ONE FOUNDATIONS Mitochondrion Neuronal membrane Nucleus Rough ER Ribosomes Polyribosomes Golgi apparatus Smooth ER Axon hillock Microtubules Axon ▲ FIGURE 2.8 The internal structure of a typical neuron. 023–054_Bear_02_revised_final.indd 30 12/20/14 2:58 AM  CHAPTER 2 NEURONS AND GLIA 31 Gene Gene Promoter Terminator DNA DNA Exon 1 Exon 2 Exon 3 1 Transcription Intron 1 Intron 2 DNA Transcription RNA polymerase RNA RNA 2 RNA processing Splicing mRNA transcript (b) mRNA 3 Export from nucleus Cytoplasm (a) ▲ FIGURE 2.9 Gene transcription. (a) RNA molecules are synthesized by RNA polymerase and then processed into mRNA to carry the genetic instructions for protein assembly from the nucleus to the cytoplasm. (b) Transcription is initiated at the promoter region of the gene and stopped at the terminator region. The initial RNA must be spliced to remove the introns that do not code for protein. the transcript (Figure 2.9a). Interspersed between protein-coding genes are long stretches of DNA whose functions remain poorly understood. Some of these regions, however, are known to be important for regulat- ing transcription. At one end of the gene is the promoter, the region where the RNA-synthesizing enzyme, RNA polymerase, binds to initiate transcription. The binding of the polymerase to the promoter is tightly regulated by other proteins called transcription factors. At the other end is a sequence of DNA called the terminator, or stop sequence, that the RNA polymerase recognizes as the end point for transcription. In addition to the non-coding regions of DNA that flank the genes, there are often additional stretches of DNA within the gene itself that cannot be used to code for protein. These interspersed regions are called introns, and the coding sequences are called exons. Initial transcripts con- tain both introns and exons, but then, by a process called RNA splicing, 023–054_Bear_02_revised_final.indd 31 12/20/14 2:58 AM  32 PART ONE FOUNDATIONS the introns are removed and the remaining exons are fused together (Figure 2.9b). In some cases, specific exons are also removed with the introns, leaving an “alternatively spliced” mRNA that actually encodes a different protein. Thus, transcription of a single gene can ultimately give rise to several different mRNAs and protein products. mRNA transcripts emerge from the nucleus via pores in the nuclear envelope and travel to the sites of protein synthesis elsewhere in the neu- ron. At these sites, a protein molecule is assembled much as the mRNA molecule was: by linking together many small molecules into a chain. In the case of protein, the building blocks are amino acids, of which there are 20 different kinds. This assembling of proteins from amino acids under the direction of the mRNA is called translation. The scientific study of this process, which begins with the DNA of the nucleus and ends with the synthesis of protein molecules in the cell, is known as molecular biology. The “central dogma” of molecular biology is summarized as follows: Transcription Translation DNA mRNA Protein Neuronal Genes, Genetic Variation, and Genetic Engineering. Neurons differ from other cells in the body because of the specific genes they express as proteins. A new understanding of these genes is now possible because the human genome—the entire length of DNA that comprises the genetic information in our chromosomes—has been sequenced. We now know the 25,000 “words” that comprise our genome, and we know where these genes can be found on each chromosome. Furthermore, we are learning which genes are expressed uniquely in neurons (Box 2.2). This knowledge has paved the way to understanding the genetic basis of many diseases of the nervous system. In some diseases, long stretches of DNA that contain sev- eral genes are missing; in others, genes are duplicated, leading to overex- pression of specific proteins. These sorts of mishaps, called gene copy num- ber variations, often occur at the moment of conception when paternal and maternal DNA mix to create the genome of the offspring. Some instances of serious psychiatric disorders, including autism and schizophrenia, were recently shown to be caused by gene copy number variations in the af- fected children. (Psychiatric disorders are discussed in Chapter 22.) Other nervous system disorders are caused by mutations—“typographical errors”—in a gene or in the flanking regions of DNA that regulate the gene’s expression. In some cases, a single protein may be grossly abnormal or missing entirely, disrupting neuronal function. An example is fragile X syn- drome, a disorder that manifests as intellectual disability and autism and is caused by disruption of a single gene (discussed further in Chapter 23). Many of our genes carry small mutations, called single nucleotide polymor- phisms, which are analogous to a minor misspelling caused by a change in a single letter. These are usually benign, like the difference between “color” and “colour”—different spelling, same meaning. However, sometimes the mutations can affect protein function (consider the difference between “bear” and “bare”—same letters, different meaning). Such single nucleotide polymorphisms, alone or together with others, can affect neuronal function. Genes make the brain, and understanding how they contribute to neu- ronal function in both healthy and diseased organisms is a major goal of neuroscience. An important breakthrough was the development of tools for genetic engineering—ways to change organisms by design with gene mutations or insertions. This technology has been used most in mice because they are rapidly reproducing mammals with a central nervous 023–054_Bear_02_revised_final.indd 32 12/20/14 2:58 AM  CHAPTER 2 NEURONS AND GLIA 33 BOX 2.2 BRAIN FOOD Expressing One’s Mind in the Post-Genomic Era S equencing the human genome was a truly monumen- tal achievement, completed in 2003. The Human Genome n1 Brain Brain 2 Project identified all of the approximately 25,000 genes in human DNA. We now live in what has been called the “post- genomic era,” in which information about the genes ex- pressed in our tissues can be used to diagnose and treat diseases. Neuroscientists are using this information to tackle long-standing questions about the biological basis of neuro- logical and psychiatric disorders as well as to probe deeper into the origins of individuality. The logic goes as follows. The brain is a product of the genes expressed in it. Differences in gene expression between a normal brain and a diseased brain, or a brain of unusual ability, can be used to identify the molecular basis of the observed symptoms or traits. Vial of mRNA Vial of mRNA The level of gene expression is usually defined as the from brain 1, from brain 2, number of mRNA transcripts synthesized by different cells labeled red labeled green and tissues to direct the synthesis of specific proteins. Thus, the analysis of gene expression requires comparing the rela- tive abundance of various mRNAs in the brains of two groups Mix Mix Mi i applie applied ied ed of humans or animals. One way to perform such a compari- to DNA DNA microarray micr m crroar cro ro ray Spot of synthetic son is to use DNA microarrays, which are created by robotic DNA with gene- machines that arrange thousands of small spots of synthetic specific sequence DNA on a microscope slide. Each spot contains a unique DNA sequence that will recognize and stick to a different spe- cific mRNA sequence. To compare the gene expression in two brains, one begins by collecting a sample of mRNAs from each brain. The mRNA of one brain is labeled with a chemical tag that fluoresces green, and the mRNA of the other brain Microscopic Gene with Gen Genene with Gene w ith with slide is labeled with a tag that fluoresces red. These samples are reduced equivalent reduced d then applied to the microarray. Highly expressed genes will expression expression express sion expression produce brightly fluorescent spots, and differences in the rel- in brain 2 in both in brain 1 ative gene expression between the brains will be revealed by brains differences in the color of the fluorescence (Figure A). Figure A Profiling differences in gene expression. system similar to our own. Today, it is common in neuroscience to hear about knockout mice, in which one gene has been deleted (or “knocked out”). Such mice can be used to study the progression of a disease, like frag- ile X, with the goal of correcting it. Another approach has been to generate transgenic mice, in which genes have been introduced and overexpressed; these new genes are called transgenes. Knock-in mice have also been cre- ated in which the native gene is replaced with a modified transgene. We will see many examples in this book of how genetically engineered animals have been used in neuroscience. The discoveries that allowed genetic modification of mice have revolutionized biology. The researchers who did this work were recognized with the 2007 Nobel Prize in Physiology or Medicine: Martin Evans of Cardiff University, Oliver Smithies of the University of North Carolina at Chapel Hill, and Mario Capecchi of the University of Utah (Box 2.3). 023–054_Bear_02_revised_final.indd 33 12/20/14 2:58 AM  34 PART ONE FOUNDATIONS BOX 2.3 PAT H O F D I S C O V E RY Gene Targeting in Mice by Mario Capecchi H ow did I first get the idea to pursue gene targeting in mice? From a simple observation. Mike Wigler, now at Cold of transgenic mice through the injection and random inte- gration of exogenous DNA into chromosomes of fertilized Spring Harbor Laboratory, and Richard Axel, at Columbia mouse eggs, or zygotes. To achieve the high efficiency of University, had published a paper in 1979 showing that ex- expression of the exogenous DNA in the recipient cell, I had posing mammalian cells to a mixture of DNA and calcium to attach small fragments of viral DNA, which we now un- phosphate would cause some cells to take up the DNA in derstand to contain enhancers that are critical in eukaryotic functional form and express the encoded genes. This was gene expression. exciting because they had clearly demonstrated that exog- But what fascinated me most was our observation that enous, functional DNA could be introduced into mammalian when many copies of a gene were injected into a cell nucleus, cells. But I wondered why their efficiency was so low. Was it all of these molecules ended up in an ordered head-to-tail a problem of delivery, insertion of exogenous DNA into the arrangement, called a concatemer (Figure B). This was as- chromosome, or expression of the genes once inserted into tonishing and could not have occurred as a random event. the host chromosome? What would happen if purified DNA We went on to unequivocally prove that homologous recom- was directly injected into the nucleus of mammalian cells in bination, the process by which chromosomes share genetic culture? information during cell division, was responsible for the in- To find out, I converted a colleague’s electrophysiology corporation of the foreign DNA (Folger et al., 1982). These station into a miniature hypodermic needle to directly inject experiments demonstrated that all mammalian somatic cells DNA into the nucleus of a living cell using mechanical micro- contain a very efficient machinery for swapping segments of manipulators and light microscopy (Figure A). The procedure DNA that have similar sequences of nucleotides. Injection of worked with amazing efficiency (Capecchi, 1980). With this a thousand copies of a gene sequence into the nucleus of a method, the frequency of successful integration was now cell resulted in chromosomal insertion of a concatemer con- one in three cells rather than one in a million cells as for- taining a thousand copies of that sequence, all oriented in merly. This high efficiency directly led to the development the same direction. This simple observation directly led me to Holding pipette Fertilized mouse egg Micropipette Figure A with DNA Fertilized mouse egg receiving an injection of foreign DNA. (Image solution courtesy of Dr. Peimin Qi, Division of Comparative Medicine, Massachusetts Institute of Technology.) 023–054_Bear_02_revised_final.indd 34 12/20/14 2:58 AM  CHAPTER 2 NEURONS AND GLIA 35 (EC). Then I heard a rumor that Martin Evans in Cambridge, England was isolating more promising cells, which he called EK cells, that resembled EC cells but were derived from a nor- mal mouse embryo rather than from tumors. I called him and asked if the rumor was correct, and he said it was. My next question was whether I could come to his laboratory to learn how to work with those cells, and his an- swer again was yes. Christmas time, 1985, was beautiful in Cambridge. My wife, who worked with me, and I had a wonderful couple of weeks learning how to maintain these marvelous cells and use them to generate mice capa- ble of germ line transmission. Figure B Investigators often have a pre- conceived idea about the particular envision mutating any chosen gene, in any chosen manner, role of their gene of interest in mouse biology, and they are in living mice by gene targeting. usually very surprised by results when the gene is knocked Excited by this possibility, in 1980, I submitted a grant out. Gene targeting has taken us in many new directions, to the U.S. National Institutes of Health (NIH) proposing to including most recently pursuing the role of microglia, cells directly alter gene DNA sequences in mammalian cultured that migrate into the brain after being generated in the bone cells by homologous recombination. They rejected the pro- marrow along with immune and blood cells. Mutating these posal, and their arguments were not unreasonable. They cells in mice results in a pathology remarkably similar to the argued that the probability of the exogenously added DNA human condition called trichotillomania, a type of obsessive- sequence ever finding the DNA sequence similar enough compulsive disorder characterized by strong urges to pull to enable homologous recombination in living mammalian out one’s hair. Amazingly, transplanting normal bone mar- cells (containing 3 ⫻ 109 nucleotide base pairs) was van- row into mutant mice permanently cures them of this path- ishingly small. Fortunately, my grant application contained ological behavior (Chen et al., 2010). Now, we are deeply two other proposals that the NIH reviewers liked, and they immersed in trying to understand the mechanism of how funded those projects. I used those funds to support the microglia control neural circuit output and, more impor- gene targeting project. Four years later, we had results sup- tantly, exploring the intimate relationship between the im- porting our ability to do gene targeting in cultured mam- mune system (in this case microglia) and neuropsychiatric malian cells. I then resubmitted a new NIH grant application disorders such as depression, autism, schizophrenia, and to the same review panel, now proposing to extend gene Alzheimer’s disease. targeting to generating mutant mice. Their evaluation sheet in response began this way: “We are glad you didn’t follow Ref erences Capecchi MR. 1980. High efficiency transformation by direct micro- our advice.” injection of DNA into cultured mammalian cells. Cell 22:479–488. It took 10 years to develop gene targeting in mice (Thomas Chen SC, Tvrdik P, Peden E, Cho S, Wu S, Spangrude G, Capecchi & Capecchi, 1987). Prior to this success, we had to under- MR. 2010. Hematopoietic origin of pathological grooming in stand the homologous recombination machinery in eukary- Hoxb8 mutant mice. Cell 141(5):775–785. otic cells. Also, because the frequency of gene targeting was Folger KR, Wong EA, Wahl G, Capecchi MR. 1982. Patterns of in- low, if we were to be successful in transferring our technology tegration of DNA microinjected into cultured mammalian cells: to mice, we needed mouse embryonic stem cells capable of evidence for homologous recombination between injected plasmid DNA molecules. Molecular and Cellular Biology 2:1372–1387. contributing to the formation of the germ line—the sperm and Thomas KR, Capecchi MR. 1987. Site-directed mutagenesis by eggs—in mature animals. I was getting depressed from our gene targeting in mouse embryo-derived stem cells. Cell 51: lack of success using cells derived from embryonal carcinoma 503–512. 023–054_Bear_02_revised_final.indd 35 12/20/14 2:58 AM  36 PART ONE FOUNDATIONS Rough Endoplasmic Reticulum. Neurons make use of the information in genes by synthesizing proteins. Protein synthesis occurs at dense globular structures in the cytoplasm called ribosomes. mRNA transcripts bind to the ribosomes, and the ribosomes translate the instructions contained in the mRNA to assemble a protein molecule. In other words, ribosomes use the blueprint provided by the mRNA to manufacture proteins from raw material in the form of amino acids. In neurons, many ribosomes are attached to stacks of membrane called rough endoplasmic reticulum, or rough ER (Figure 2.10). Rough ER abounds in neurons, far more than in glia or most other non-neuronal cells. In fact, we have already been introduced to rough ER by another name: Nissl bodies. This is the organelle stained with the dyes that Nissl introduced over 100 years ago. Rough ER is a major site of protein synthesis in neurons, but not all ribosomes are attached to rough ER. Many are freely floating and are Nucleus called free ribosomes. Several free ribosomes may appear to be attached Nuclear by a thread; these are called polyribosomes. The thread is a single envelope strand of mRNA, and the associated ribosomes are working on it to make Nuclear pore multiple copies of the same protein. What is the difference between proteins synthesized on the rough ER and those synthesized on the free ribosomes? The answer appears to de- pend on the intended fate of the protein molecule. If it is destined to re- side within the cytosol of the neuron, then the protein’s mRNA transcript shuns the ribosomes of the rough ER and gravitates toward the free ri- Rough ER Ribosomes bosomes (Figure 2.11a). However, if the protein is destined to be inserted ▲ FIGURE 2.10 into the membrane of the cell or an organelle, then it is synthesized on Rough endoplasmic reticulum, or the rough ER. As the protein is being assembled, it is threaded back rough ER. and forth through the membrane of the rough ER, where it is trapped (Figure 2.11b). It is not surprising that neurons have so much rough ER because, as we shall see in later chapters, special membrane proteins are what give these cells their remarkable information-processing abilities. Smooth Endoplasmic Reticulum and the Golgi Apparatus. The remain- der of the cytosol of the soma is crowded with stacks of membranous or- ganelles that look a lot like rough ER without the ribosomes, so much so that one type is called smooth endoplasmic reticulum, or smooth ER. Smooth ER is heterogeneous and performs different functions in different locations. Some smooth ER is continuous with rough ER and is believed to be a site where the proteins that jut out from the membrane are care- fully folded, giving them their three-dimensional structure. Other types of smooth ER play no direct role in the processing of protein molecules but instead regulate the internal concentrations of substances such as calcium. (This organelle is particularly prominent in muscle cells, where it is called sarcoplasmic reticulum, as we will see in Chapter 13.) The stack of membrane-enclosed disks in the soma that lies farthest from the nucleus is the Golgi apparatus, first described in 1898 by Camillo Golgi (Figure 2.12). This is a site of extensive “post-translational” chemical processing of proteins. One important function of the Golgi apparatus is believed to be the sorting of certain proteins that are destined for delivery to different parts of the neuron, such as the axon and the dendrites. The Mitochondrion. Another very abundant organelle in the soma is the mitochondrion (plural: mitochondria). In neurons, these sausage- shaped structures are about 1 ␮m long. Within the enclosure of their outer membrane are multiple folds of inner membrane called cristae (singular: crista). Between the cristae is an inner space called matrix (Figure 2.13a). 023–054_Bear_02_revised_final.indd 36 12/20/14 2:58 AM  CHAPTER 2 NEURONS AND GLIA 37 mRNA mRNA mRNA Free Rough ER ribosome mRNA being mRNA being translated translated Newly created protein ▲ FIGURE 2.11 Protein synthesis on a free ribosome and on rough ER. mRNA binds to a ri- bosome, initiating protein synthesis. (a) Proteins synthesized on free ribosomes are destined for the cytosol. (b) Proteins Newly synthesized synthesized on the rough ER are des- membrane-associated tined to be enclosed by or inserted into protein the membrane. Membrane-associated proteins are inserted into the membrane (a) Protein synthesis on a free (b) Protein synthesis on rough ER as they are assembled. ribosome Rough ER Newly synthesized Golgi protein apparatus ▲ FIGURE 2.12 The Golgi apparatus. This complex organ- elle sorts newly synthesized proteins for delivery to appropriate locations in the neuron. 023–054_Bear_02_revised_final.indd 37 12/20/14 2:58 AM  38 PART ONE FOUNDATIONS Mitochondria are the site of cellular respiration (Figure 2.13b). When a mitochondrion “inhales,” it pulls inside pyruvic acid (derived from sugars and digested proteins and fats) and oxygen, both of which are floating in the cytosol. Within the inner compartment of the mitochondrion, pyruvic acid enters into a complex series of biochemical reactions called the Krebs cycle, named after the German-British scientist Hans Krebs, who first proposed it in 1937. The biochemical products of the Krebs cycle provide energy that, in another series of reactions within the cristae (called the electron-transport chain), results in the addition of phosphate to adenos- ine diphosphate (ADP), yielding adenosine triphosphate (ATP), the cell’s energy source. When the mitochondrion “exhales,” 17 ATP molecules are released for every molecule of pyruvic acid that had been taken in. ATP is the energy currency of the cell. The chemical energy stored in ATP fuels most of the biochemical reactions of the neuron. For example, Outer membrane as we shall see in Chapter 3, special proteins in the neuronal membrane use the energy released by the breakdown of ATP into ADP to pump cer- Inner tain substances across the membrane to establish concentration differ- membrane ences between the inside and the outside of the neuron. Cristae The Neuronal Membrane The neuronal membrane serves as a barrier to enclose the cytoplasm inside the neuron and to exclude certain substances that float in the fluid that bathes the neuron. The membrane is about 5 nm thick and is studded with proteins. As mentioned earlier, some of the membrane-associated proteins pump substances from the inside to the outside. Others form Matrix pores that regulate which substances can gain access to the inside of the neuron. An important characteristic of neurons is that the protein com- (a) position of the membrane varies depending on whether it is in the soma, the dendrites, or the axon. The function of neurons cannot be understood without understanding the structure and function of the membrane and its associated proteins. + O2 + CO2 In fact, this topic is so important that we’ll spend much of the next four chapters looking at how the membrane endows neurons with the remark- able ability to transfer electrical signals throughout the brain and body. Pyruvic acid The Cytoskeleton Protein Dietary and stored Earlier, we compared the neuronal membrane to a circus tent draped Sugar energy sources over an internal scaffolding. This scaffolding is called the cytoskeleton, Fat and it gives the neuron its characteristic shape. The “bones” of the cy- (b) toskeleton are the microtubules, microfilaments, and neurofilaments ▲ FIGURE 2.13 (Figure 2.14). Unlike the tent scaffolding, however, the cytoskeleton is The role of mitochondria. (a) Compo- not static. Elements of the cytoskeleton are dynamically regulated and nents of a mitochondrion. (b) Cellular res- piration. ATP is the energy currency that are in continual motion. Your neurons are probably squirming around in fuels biochemical reactions in neurons. your head even as you read this sentence. Microtubules. Measuring 20 nm in diameter, microtubules are rela- tively large and run longitudinally down neurites. A microtubule appears as a straight, thick-walled hollow pipe. The wall of the pipe is composed of smaller strands that are braided like rope around the hollow core. Each of the smaller strands consists of the protein tubulin. A single tubulin mol- ecule is small and globular; the strand consists of tubulins stuck together like pearls on a string. The process of joining small proteins to form a long strand is called polymerization; the resulting strand is called a polymer. Polymerization and depolymerization of microtubules and, therefore, of neuronal shape can be regulated by various signals within the neuron. 023–054_Bear_02_revised_final.indd 38 12/20/14 2:58 AM  CHAPTER 2 NEURONS AND GLIA 39 One class of proteins that participate in the regulation of microtubule assembly and function are microtubule-associated proteins, or MAPs. Among other functions (many of which are unknown), MAPs anchor the microtubules to one another and to other parts of the neuron. Pathological changes in an axonal MAP, called tau, have been implicated in the de- mentia that accompanies Alzheimer’s disease (Box 2.4). Microfilaments. Measuring only 5 nm in diameter, microfilaments are about the same thickness as the cell membrane. Found throughout the neuron, they are particularly numerous in the neurites. Microfilaments are braids of two thin strands that are polymers of the protein actin. Actin is one of the most abundant proteins in cells of all types, including neurons, and is believed to play a role in changing cell shape. Indeed, as we shall see in Chapter 13, actin filaments are critically involved in the mechanism of muscle contraction. Like microtubules, actin microfilaments are constantly undergoing as- sembly and disassembly, and this process is regulated by signals in the neu- ron. In addition to running longitudinally down the core of the neurites like microtubules, microfilaments are also closely associated with the membrane. They are anchored to the membrane by attachments with a meshwork of fi- brous proteins that line the inside of the membrane like a spider web. Tubulin molecule Actin Neurofilaments. With a diameter of 10 nm, neurofilaments are inter- molecule mediate in size between microtubules and microfilaments. They exist in all cells of the body as intermediate filaments; only in neurons are they called neurofilaments. The difference in name reflects differences in struc- ture among different tissues. For example, a different intermediate fila- 20 nm ment, keratin, composes hair when bundled together. Of the types of fibrous structure we have discussed, neurofilaments most Microtubule 5 nm closely resemble the bones and ligaments of the skeleton. A neurofilament 10 nm consists of multiple subunits (building blocks) that are wound together Neurofilament Microfilament into a ropelike structure. Each strand of the rope consists of individual ▲ FIGURE 2.14 long protein molecules, making neurofilaments mechanically very strong. Components of the cytoskeleton. The arrangement of microtubules, neurofila- ments, and microfilaments gives the The Axon neuron its characteristic shape. So far, we’ve explored the soma, organelles, membrane, and cytoskeleton. These structures are not unique to neurons but are found in all the cells in our body. Now we’ll look at the axon, a structure found only in neurons and highly specialized for the transfer of information over distances in the nervous system. The axon begins with a region called the axon hillock, which ta- pers away from the soma to form the initial segment of the axon proper (Figure 2.15). Two noteworthy features distinguish the axon from the soma: 1. No rough ER extends into the axon, and there are few, if any, free ribo- somes in mature axons. 2. The protein composition of the axon membrane is fundamentally dif- ferent from that of the soma membrane. These structural differences translate into functional distinctions. Because there are no ribosomes, there is no protein synthesis in the axon. This means that all proteins in the axon must originate in the soma. And the different proteins in the axonal membrane enable it to serve as a wire that sends information over great distances. Axons may extend from less than a millimeter to over a meter long. Axons often branch, and these branches, called axon collaterals, can 023–054_Bear_02_revised_final.indd 39 12/20/14 2:58 AM  40 PART ONE FOUNDATIONS BOX 2.4 OF SPECIAL INTEREST Alzheimer’s Disease and the Neuronal Cytoskeleton N eurites are the most remarkable structural feature of a neuron. Their elaborate branching patterns, critical for infor- changes in the “neurofibrils,” elements of the cytoskeleton that can be stained by a silver solution. mation processing, reflect the organization of the underlying cytoskeleton. It is therefore no surprise that a devastating loss The Bielschowsky silver preparation showed very char- of brain function can result when the cytoskeleton of neurons is acteristic changes in the neurofibrils. However, inside an disrupted. An example is Alzheimer’s disease, which is charac- apparently normal-looking cell, one or more single fibers terized by the disruption of the cytoskeleton of neurons in the could be observed that became prominent through their cerebral cortex, a region of the brain crucial for cognitive func- striking thickness and specific impregnability. At a more tion. This disorder and its underlying brain pathology were first advanced stage, many fibrils arranged parallel showed described in 1907 by the German physician A. Alzheimer in a the same changes. Then they accumulated forming dense paper titled “A Characteristic Disease of the Cerebral Cortex.” bundles and gradually advanced to the surface of the cell. Below are excerpts from the English translation. Eventually, the nucleus and cytoplasm disappeared, and only a tangled bundle of fibrils indicated the site where One of the first disease symptoms of a 51-year-old once the neuron had been located. woman was a strong feeling of jealousy toward her hus- As these fibrils can be stained with dyes different from band. Very soon she showed rapidly increasing memory the normal neurofibrils, a chemical transformation of the impairments; she could not find her way about her home, fibril substance must have taken place. This might be she dragged objects to and fro, hid herself, or sometimes the reason why the fibrils survived the destruction of the thought that people were out to kill her, then she would cell. It seems that the transformation of the fibrils goes start to scream loudly. hand in hand with the storage of an as yet not closely During institutionalization her gestures showed a com- examined pathological product of the metabolism in the plete helplessness. She was disoriented as to time and neuron. About one-quarter to one-third of all the neu- place. From time to time she would state that she did not rons of the cerebral cortex showed such alterations. understand anything, that she felt confused and totally Numerous neurons, especially in the upper cell layers, lost. Sometimes she considered the coming of the doctor had totally disappeared. (Bick et al., 1987, pp. 2–3.) as an official visit and apologized for not having finished her work, but other times she would start to yell in the fear that The severity of the dementia in Alzheimer’s disease is well the doctor wanted to operate on her; or there were times correlated with the number and distribution of what are now that she would send him away in complete indignation, commonly known as neurofibrillary tangles, the “tombstones” uttering phrases that indicated her fear that the doctor of dead and dying neurons (Figure A). Indeed, as Alzheimer wanted to damage her woman’s honor. From time to time speculated, tangle formation in the cerebral cortex very likely she was completely delirious, dragging her blankets and causes the symptoms of the disease. Electron microscopy sheets to and fro, calling for her husband and daughter, reveals that the major components of the tangles are paired and seeming to have auditory hallucinations. Often she helical filaments, long fibrous proteins braided together like would scream for hours and hours in a horrible voice. strands of a rope (Figure B). It is now understood that these Mental regression advanced quite steadily. After four filaments consist of the microtubule-associated protein tau. and a half years of illness the patient died. She was com- Tau normally functions as a bridge between the microtu- pletely apathetic in the end, and was confined to bed in a bules in axons, ensuring that they run straight and parallel to fetal position. (Bick et al., 1987, pp. 1–2.) one another. In Alzheimer’s disease, the tau detaches from the microtubules and accumulates in the soma. This disrup- Following her death, Alzheimer examined the woman’s tion of the cytoskeleton causes the axons to wither, thus im- brain under the microscope. He made particular note of peding the normal flow of information in the affected neurons. travel long distances to communicate with different parts of the nervous system. Occasionally, an axon collateral returns to communicate with the same cell that gave rise to the axon or with the dendrites of neighboring cells. These axon branches are called recurrent collaterals. The diameter of an axon is variable, ranging from less than 1 ␮m to about 25 ␮m in humans and to as large as 1 mm in squid. This variation in axon size is important. As will be explained in Chapter 4, the speed 023–054_Bear_02_revised_final.indd 40 12/20/14 2:58 AM  CHAPTER 2 NEURONS AND GLIA 41 (a) (b) (c) Figure A Neurons in a human brain with Alzheimer’s disease. Normal neurons contain neurofilaments but no neurofibrillary tangles. (a) Brain tissue stained by a method that makes neuronal neurofilaments fluoresce green, showing viable neurons. (b) The same region of the brain stained to show the presence of tau within neurofibrillary tangles, revealed by red fluorescence. (c) Superimposition of images in parts a and b. The neuron indicated by the arrowhead contains neurofilaments but no tangles and therefore is healthy. The neuron indicated by the large arrow has neurofilaments but also has started to show accumulation of tau and therefore is diseased. The neuron indicated by the small arrow in parts b and c is dead because it contains no neurofilaments. The remaining tangle is the tombstone of a neuron killed by Alzheimer’s disease. (Source: Courtesy of Dr. John Morrison and modified from Vickers et al., 1994.) What causes such changes in tau? Attention has focused that leads to neurofibrillary tangle formation and dementia. on another protein that accumulates in the brain of Alzheimer’s Currently, hope for therapeutic intervention focuses on strate- patients, called amyloid. Alzheimer’s disease research is mov- gies to reduce the depositions of amyloid in the brain. The ing very fast, but the consensus today is that the abnormal need for effective therapy is urgent: In the United States alone, secretion of amyloid by neurons is the first step in a process more than 5 million people are afflicted with this tragic disease. 100 nm Figure B Paired helical filaments of a tangle. (Source: Goedert, 1996, Fig. 2b.) of the electrical signal that sweeps down the axon—the nerve impulse— depends on the axonal diameter. The thicker the axon, the faster the im- pulse travels. The Axon Terminal. All axons have a beginning (the axon hillock), a mid- dle (the axon proper), and an end. The end is called the axon terminal or terminal bouton (French for “button”), reflecting the fact that it usually 023–054_Bear_02_revised_final.indd 41 12/20/14 2:58 AM  42 PART ONE FOUNDATIONS appears as a swollen disk (Figure 2.16). The terminal is a site where the axon comes in contact with other neurons (or other cells) and passes in- formation on to them. This point of contact is called the synapse, a word derived from the Greek, meaning “to fasten together.” Sometimes axons have many short branches at their ends, and each branch forms a syn- apse on dendrites or cell bodies in the same region. These branches are collectively called the terminal arbor. Sometimes axons form synapses at swollen regions along their length and then continue on to terminate elsewhere (Figure 2.17). Such swellings are called boutons en passant (“buttons in passing”). In either case, when a neuron makes synaptic contact with another cell, it is said to innervate that cell, or to provide innervation. The cytoplasm of the axon terminal differs from that of the axon in several ways: 1. Microtubules do not extend into the terminal. 2. The terminal contains numerous small bubbles of membrane, called synaptic vesicles, that measure about 50 nm in diameter. Axon hillock 3. The inside surface of the membrane that faces the synapse has a par- ticularly dense covering of proteins. 4. The axon terminal cytoplasm has numerous mitochondria, indicating a high energy demand. Axon collaterals ▲ FIGURE 2.15 The axon and axon collaterals. The axon functions like a telegraph wire to send electrical impulses to distant sites in the nervous system. The arrows indi- cate the direction of information flow. Presynaptic axon terminal Mitochondria Synapse Synaptic vesicles