Bear Chapter 5 Synaptic Transmission PDF

CHAPTER FIVE Synaptic Transmission INTRODUCTION BOX 5.1 OF SPECIAL INTEREST: Otto Loewi’s Dream TYPES OF SYNAPSES Electrical Synapses Chemical Synapses CNS Chemical Synapses...

CHAPTER FIVE Synaptic Transmission INTRODUCTION BOX 5.1 OF SPECIAL INTEREST: Otto Loewi’s Dream TYPES OF SYNAPSES Electrical Synapses Chemical Synapses CNS Chemical Synapses BOX 5.2 PATH OF DISCOVERY: For the Love of Dendritic Spines, by Kristen M. Harris The Neuromuscular Junction PRINCIPLES OF CHEMICAL SYNAPTIC TRANSMISSION Neurotransmitters Neurotransmitter Synthesis and Storage Neurotransmitter Release BOX 5.3 BRAIN FOOD: How to SNARE a Vesicle Neurotransmitter Receptors and Effectors Transmitter-Gated Ion Channels BOX 5.4 BRAIN FOOD: Reversal Potentials G-Protein-Coupled Receptors Autoreceptors Neurotransmitter Recovery and Degradation Neuropharmacology BOX 5.5 OF SPECIAL INTEREST: Bacteria, Spiders, Snakes, and People PRINCIPLES OF SYNAPTIC INTEGRATION The Integration of EPSPS Quantal Analysis of EPSPs EPSP Summation The Contribution of Dendritic Properties to Synaptic Integration Dendritic Cable Properties Excitable Dendrites Inhibition BOX 5.6 OF SPECIAL INTEREST: Startling Mutations and Poisons IPSPs and Shunting Inhibition The Geometry of Excitatory and Inhibitory Synapses Modulation CONCLUDING REMARKS 109 109–142_Bear_05_revised_final.indd 109 12/20/14 3:43 AM  110 PART ONE FOUNDATIONS INTRODUCTION INTRODUC CTION In Chapters 3 and 4, we discussed how mechanical energy, such as a thumbtack entering your foot, can be converted into a neural signal. First, specialized ion channels of the sensory nerve endings allow positive charge to enter the axon. If this depolarization reaches threshold, then action potentials are generated. Because the axonal membrane is excitable and has voltage-gated sodium channels, action potentials can propagate with- out decrement up the long sensory nerves. For this information to be pro- cessed by the rest of the nervous system, however, these neural signals must be passed on to other neurons—for example, the motor neurons that control muscle contraction, as well as neurons in the brain and spinal cord that lead to a coordinated reflex response. By the end of the nineteenth cen- tury, it was recognized that this transfer of information from one neuron to another occurs at specialized sites of contact. In 1897, English physiologist Charles Sherrington gave these sites their name: synapses. The process of information transfer at a synapse is called synaptic transmission. The physical nature of synaptic transmission was debated for almost a century. One attractive hypothesis, which nicely explained the speed of synaptic transmission, was that it was simply electrical current flowing from one neuron to the next. The existence of such electrical synapses was finally proven in the late 1950s by Edwin Furshpan and David Potter, American physiologists who were studying the nervous system of crayfish at University College London, and Akira Watanabe, who was studying the neurons of lobster at the Tokyo Medical and Dental University. We now know that electrical synapses are common in the brains of invertebrates and vertebrates, including mammals. An alternative hypothesis about the nature of synaptic transmission, also dating back to the 1800s, was that chemical neurotransmitters trans- fer information from one neuron to another at the synapse. Solid support for the concept of chemical synapses was provided in 1921 by Otto Loewi, then the head of the Pharmacology Department at the University of Graz in Austria. Loewi showed that electrical stimulation of axons innervating the frog’s heart caused the release of a chemical that could mimic the effects of neuron stimulation on the heartbeat (Box 5.1). Later, Bernard Katz and his colleagues at University College London conclusively demonstrated that fast transmission at the synapse between a motor neuron axon and skeletal muscle was chemically mediated. By 1951, John Eccles of the Australian National University was studying the physiology of synaptic transmission within the mammalian central nervous system (CNS) using a new tool, the glass microelectrode. These experiments indicated that many CNS syn- apses also use a chemical transmitter; in fact, chemical synapses comprise the majority of synapses in the brain. During the last decade, new methods of studying the molecules involved in synaptic transmission have revealed that synapses are far more complex than most neuroscientists anticipated. Synaptic transmission is a large and fascinating topic. The actions of psy- choactive drugs, the causes of mental disorders, the neural bases of learning and memory—indeed, all the operations of the nervous system—cannot be understood without knowledge of synaptic transmission. Therefore, we’ve devoted several chapters to this topic, mainly focusing on chemical synapses. In this chapter, we begin by exploring the basic mechanisms of synaptic transmission. What do different types of synapse look like? How are neu- rotransmitters synthesized and stored, and how are they released in response to an action potential in the axon terminal? How do neurotransmitters act on the postsynaptic membrane? How do single neurons integrate the inputs provided by the thousands of synapses that impinge upon them? 109–142_Bear_05_revised_final.indd 110 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 111 BOX 5.1 OF SPECIAL INTEREST Otto Loewi’s Dream O ne of the more colorful stories in the history of neuro- science comes from Otto Loewi, who, working in Austria in I awoke again, at three o’clock, and I remembered what it was. This time I did not take any risk; I got up imme- the 1920s, showed definitively that synaptic transmission diately, went to the laboratory, made the experiment on between nerves and the heart is chemically mediated. The the frog’s heart, described above, and at five o’clock the heart has two types of innervation; one type speeds the chemical transmission of the nervous impulse was conclu- beating of the heart, and the other slows it. The latter type sively proved... Careful consideration in daytime would of innervation is supplied by the vagus nerve. Loewi isolated undoubtedly have rejected the kind of experiment I per- a frog heart with the vagal innervation still intact, stimulated formed, because it would have seemed most unlikely that the nerve electrically, and observed the expected effect: the if a nervous impulse released a transmitting agent, it would slowing of the heartbeat. The critical demonstration that this do so not just in sufficient quantity to influence the effec- effect was chemically mediated came when he applied the tor organ, in my case the heart, but indeed in such an ex- solution that had bathed this heart to a second isolated frog cess that it could partly escape into the fluid which filled heart and found that the beating of this one also slowed. the heart, and could therefore be detected. Yet the whole The idea for this experiment had actually come to Loewi in nocturnal concept of the experiment was based on this a dream. Below is his own account: eventuality, and the result proved to be positive, contrary In the night of Easter Sunday, 1921, I awoke, turned to expectation. (Loewi, 1953, pp. 33–34.) on the light, and jotted down a few notes on a tiny slip of The active compound, which Loewi called vagusstoff paper. Then, I fell asleep again. It occurred to me at six (literally “vagus substance” in German), turned out to be o’clock in the morning that during the night I had written acetylcholine. As we see in this chapter, acetylcholine is down something most important, but I was unable to deci- also a transmitter at the synapse between nerve and skeletal pher the scrawl. That Sunday was the most desperate day muscle. Unlike at the heart, acetylcholine applied to skeletal in my whole scientific life. During the next night, however, muscle causes excitation and contraction. TYPE ES OF SYNAPSES TYPES We introduced the synapse in Chapter 2. A synapse is the specialized junction where one part of a neuron contacts and communicates with an- other neuron or cell type (such as a muscle or glandular cell). Information generally flows in one direction, from a neuron to its target cell. The first neuron is said to be presynaptic, and the target cell is said to be postsynaptic. Let’s take a closer look at the different types of synapse. Electrical Synapses Electrical synapses are relatively simple in structure and function, and they allow the direct transfer of ionic current from one cell to the next. Electrical synapses occur at specialized sites called gap junctions. Gap junctions occur between cells in nearly every part of the body and interconnect many non-neural cells, including epithelial cells, smooth and cardiac muscle cells, liver cells, some glandular cells, and glia. When gap junctions interconnect neurons, they can function as electri- cal synapses. At a gap junction, the membranes of two cells are separated by only about 3 nm, and this narrow gap is spanned by clusters of special proteins called connexins. There are about 20 different subtypes of con- nexins, about half of which occur in the brain. Six connexin subunits com- bine to form a channel called a connexon, and two connexons (one from each cell) meet and combine to form a gap junction channel (Figure 5.1). The channel allows ions to pass directly from the cytoplasm of one cell 109–142_Bear_05_revised_final.indd 111 12/20/14 3:43 AM  112 PART ONE FOUNDATIONS Cell 1 ▼ FIGURE 5.1 Gap junction A gap junction. (a) Neurites of two cells connected by a gap junction. (b) The enlargement shows gap junction channels, which bridge the cytoplasm of the two cells. Ions and small molecules can pass in both directions through these channels. (c) Six connexin subunits comprise one connexon, two connexons comprise one gap junction channel, and many gap junction channels comprise one gap junction. Cell 2 (a) Cell 1 Gap junction channels cytoplasm Gap junction 3.5 nm 20 nm Connexon Connexin Cell 2 cytoplasm Ions and Channel formed by pores (b) small molecules in each membrane (c) to the cytoplasm of the other. The pore of most gap junction channels is relatively large. Its diameter is about 1–2 nm, big enough for all the major cellular ions and many small organic molecules to pass through. Most gap junctions allow ionic current to pass equally well in both directions; therefore, unlike the vast majority of chemical synapses, elec- trical synapses are bidirectional. Because electrical current (in the form of ions) can pass through these channels, cells connected by gap junctions are said to be electrically coupled. Transmission at electrical synapses is very fast and, if the synapse is large, nearly fail-safe. Thus, an action potential in the presynaptic neuron can produce, with very little delay, an action potential in the postsynaptic neuron. In invertebrate species, such as the crayfish, electrical synapses are sometimes found between sensory and motor neurons in neural pathways mediating escape reflexes. This mechanism enables an animal to beat a hasty retreat when faced with a dangerous situation. Studies in recent years have revealed that electrical synapses are common in every part of the mammalian CNS (Figure 5.2a). When two neurons are electrically coupled, an action potential in the presynaptic neuron causes a small amount of ionic current to flow across the gap junc- tion channels into the other neuron. This current causes an electrically mediated postsynaptic potential (PSP) in the second neuron (Figure 5.2b). Note that, because most electrical synapses are bidirectional, when that second neuron generates an action potential, it will in turn induce a PSP in the first neuron. The PSP generated by a single electrical syn- apse in the mammalian brain is usually small—about 1 mV or less at its peak—and may not, by itself, be large enough to trigger an action 109–142_Bear_05_revised_final.indd 112 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 113 Cell 1 Action Vm of cell 1 0 potential Dendrite Record Vm of cell 1 – 65 Record Vm 0 1 2 3 of cell 2 Time (msec) Gap – 63 junction Vm of cell 2 – 64 Electrical PSP Dendrite – 65 0 1 2 3 Time (msec) Cell 2 (a) (b) ▲ FIGURE 5.2 Electrical synapses. (a) A gap junction interconnecting the dendrites of two neu- rons constitutes an electrical synapse. (b) An action potential generated in one neuron causes a small amount of ionic current to flow through gap junction chan- nels into a second neuron, inducing an electrical PSP. (Source: Part a from Sloper and Powell, 1978.) potential in the postsynaptic cell. One neuron usually makes electrical synapses with many other neurons, however, so several PSPs occurring simultaneously may strongly excite a neuron. This is an example of syn- aptic integration, which is discussed later in the chapter. The precise roles of electrical synapses vary from one brain region to another. They are often found where normal function requires that the activity of neighboring neurons be highly synchronized. For example, neu- rons in a brain stem nucleus called the inferior olive can generate both small oscillations of membrane voltage and, more occasionally, action po- tentials. These cells send axons to the cerebellum and are important in motor control. They also make gap junctions with one another. Current that flows through gap junctions during membrane oscillations and ac- tion potentials serves to coordinate and synchronize the activity of inferior olivary neurons (Figure 5.3a), and this in turn may help to control the fine timing of motor control. Michael Long and Barry Connors, working at Brown University, found that genetic deletion of a critical gap junction protein called connexin36 (Cx36) did not alter the neurons’ ability to gen- erate oscillations and action potentials but did abolish the synchrony of these events because of the loss of functional gap junctions (Figure 5.3b). Gap junctions between neurons and other cells are particularly com- mon early in development. Evidence suggests that during prenatal and postnatal brain development, gap junctions allow neighboring cells to share both electrical and chemical signals that may help coordinate their growth and maturation. Chemical Synapses Most synaptic transmission in the mature human nervous system is chemical, so the remainder of this chapter and the next will now focus exclusively on chemical synapses. Before we discuss the different types of 109–142_Bear_05_revised_final.indd 113 12/20/14 3:43 AM  114 PART ONE FOUNDATIONS (a) With gap junctions: Action potential Vm of cell 1 –0 Record Vm of cell 1 Oscillations 1 – 65 Gap junction Vm of cell 2 –0 2 – 65 Record Vm of cell 2 (b) Without gap junctions: Vm of cell 3 Record Vm –0 of cell 3 3 No gap junction – 65 4 Vm of cell 4 –0 – 65 Record Vm 0 1 2 3 4 5 of cell 4 Time (sec) ▲ FIGURE 5.3 Electrical synapses can help neurons to synchronize their activity. Certain brain stem neurons generate small, regular oscillations of Vm and occasional action potentials. (a) When two neurons are connected by gap junctions (cells 1 and 2), their oscillations and action potentials are well synchronized. (b) Similar neurons with no gap junctions (cells 3 and 4) generate oscillations and action potentials that are entirely uncoordinated. (Source: Adapted from Long et al., 2002, p. 10903.) chemical synapses, let’s take a look at some of their universal character- istics (Figure 5.4). The presynaptic and postsynaptic membranes at chemical synapses are separated by a synaptic cleft that is 20–50 nm wide, 10 times the width of the separation at gap junctions. The cleft is filled with a matrix of fibrous extracellular protein. One function of this matrix is to serve as a “glue” that binds the pre- and postsynaptic membranes together. The presynaptic side of the synapse, also called the presynaptic element, is usually an axon terminal. The terminal typically contains dozens of small membrane-enclosed spheres, each about 50 nm in diameter, called synaptic vesicles (Figure 5.5a). These vesicles store neurotransmitter, the chemical used to communicate with the postsynaptic neuron. Many axon terminals also contain larger vesicles, each about 100 nm in diameter, called secretory granules. Secretory granules contain soluble protein that appears dark in the electron microscope, so they are sometimes called large, dense-core vesicles (Figure 5.5b). 109–142_Bear_05_revised_final.indd 114 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 115 Axon terminal (presynaptic element) Secretory granules Mitochondria Synaptic Active zone Membrane cleft Postsynaptic differentiations Synaptic density vesicles Receptors Postsynaptic dendrite ▲ FIGURE 5.4 The components of a chemical synapse. Dense accumulations of protein adjacent to and within the mem- branes on either side of the synaptic cleft are collectively called mem- brane differentiations. On the presynaptic side, proteins jutting into the cytoplasm of the terminal along the intracellular face of the mem- brane sometimes look like a field of tiny pyramids. The pyramids, and the membrane associated with them, are the actual sites of neurotrans- mitter release, called active zones. Synaptic vesicles are clustered in the cytoplasm adjacent to the active zones (see Figure 5.4). The protein thickly accumulated in and just under the postsynaptic membrane is called the postsynaptic density. The postsynaptic density contains the neurotransmitter receptors, which convert the intercellular chemical signal (i.e., neurotransmitter) into an intracellular signal (i.e., a change in membrane potential or a chemical change) in the postsynaptic cell. As we shall see, the nature of this postsynaptic response can be quite varied, depending on the type of protein receptor that is activated by the neurotransmitter. CNS Chemical Synapses. In the CNS, different types of synapse may be distinguished by which part of the neuron is postsynaptic to the axon terminal. If the postsynaptic membrane is on a dendrite, the synapse is said to be axodendritic. If the postsynaptic membrane is on the cell body, the synapse is said to be axosomatic. In some cases, the postsynaptic membrane is on another axon, and these synapses are called axoaxonic (Figure 5.6). When a presynaptic axon contacts a postsynaptic dendritic spine, it is called axospinous (Figure 5.7a). In certain specialized neu- rons, dendrites actually form synapses with one another; these are called 109–142_Bear_05_revised_final.indd 115 12/20/14 3:43 AM  116 PART ONE FOUNDATIONS Mitochondria Presynaptic terminal Postsynaptic cell Active zone (a) Synaptic vesicles Dense-core vesicles (b) ▲ FIGURE 5.5 Chemical synapses, as seen with the electron microscope. (a) A fast excitatory synapse in the CNS. (b) A synapse in the PNS, with numerous dense-core vesicles. (Source: Part a adapted from Heuser and Reese, 1977, p. 262; part b adapted from Heuser and Reese, 1977, p. 278.) Soma Synapse (a) (b) (c) Dendrite Axon ▲ FIGURE 5.6 Synaptic arrangements in the CNS. (a) An axodendritic synapse. (b) An axoso- matic synapse. (c) An axoaxonic synapse. 109–142_Bear_05_revised_final.indd 116 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 117 (a) Axons (b) Presynaptic terminals Postsynaptic dendritic spine Axon Axon Postsynaptic (c) elements (d) Presynaptic terminals (a) Active zones Axon ▲ FIGURE 5.7 Various shapes and sizes of CNS synapses. (a) Axospinous synapse: A small presynaptic axon terminal contacts a postsynaptic dendritic spine. Notice that presynaptic terminals can be recognized by their many vesicles, and postsynaptic elements have postsynaptic densities. (b) An axon branches to form two presyn- aptic terminals, one larger than the other, and both contact a postsynaptic soma. (c) An unusually large axon terminal contacts and surrounds a postsynaptic soma. (d) An unusually large presynaptic axon terminal contacts five postsynaptic dendritic spines. Notice that larger synapses have more active zones. dendrodendritic synapses. The sizes and shapes of CNS synapses also vary widely (Figure 5.7a-d). The finest details of synaptic structure can be studied only under the powerful magnification of the electron microscope (Box 5.2). CNS synapses may be further classified into two general categories based on the appearance of their presynaptic and postsynaptic membrane differentiations. Synapses in which the membrane differentiation on the postsynaptic side is thicker than that on the presynaptic side are called asymmetrical synapses, or Gray’s type I synapses; those in which the mem- brane differentiations are of similar thickness are called symmetrical 109–142_Bear_05_revised_final.indd 117 12/20/14 3:43 AM  118 PART ONE FOUNDATIONS BOX 5.2 PAT H O F D I S C O V E RY For the Love of Dendritic Spines by Kristen M. Harris T he first time I looked through the microscope and saw a dendritic spine, it was love at first sight, and the affair has Adult Rat Glia vesicles simply never ended. I was a graduate student in the new neu- Spine PSD Dendrite robiology and behavior program at the University of Illinois, us Apparatus Axon and it was indeed an exciting time in neuroscience. The 1979 Glia Society for Neuroscience meeting had only about 5,000 at- Spine tendees (attendance is now about 25,000), and the member Mitochondrion number I obtained during my first year of graduate school 1 micron was and remains 2500. I had hoped to discover what a “learned” dendritic spine Figure B looks like by training animals and then using the Golgi stain- ing method to quantify changes in spine number and shape. microscopy (3DEM). I was truly hooked. With 3DEM, one Eagerly, I developed a high-throughput project, preparing the could reconstruct dendrites, axons, and glia, and not only brains from many rats at once, sectioning through the whole measure and count dendritic spines but also see where syn- brains, checking that the silver staining had worked, and then apses formed, what was inside them, and how glia asso- storing the tissue sections in butanol while engaging under- ciated with synapses (Figure B). The 3DEM platform offers graduates to help mount them on microscope slides. To our enormous possibilities for discovery. My life continues to be dismay, we found several months later that all the silver had devoted to uncovering the processes of synapse formation been dissolved out of the cells. There were no cells to see, and plasticity during learning and memory in the brain. and the project died an untimely death. Early in my career, while the revolution of molecular biology I was fortunate, however, to meet Professor Timothy Teyler was sweeping the field, only a rare student or fellow scien- at a Gordon Research Conference. He had recently brought tist shared my enthusiasm for 3DEM. That bias has shifted the hippocampal slice preparation to the United States from dramatically as neuroscientists have come to recognize the Norway and was moving his lab from Harvard to a new medi- importance of understanding how molecules work in consort cal school in Rootstown, Ohio. I was completely enamored by with intracellular organelles in small spaces like dendrites and the experimental control that brain slices might offer, so I de- spines. Furthermore, all maps of neural circuitry must include veloped a Golgi-slice procedure and moved to complete my synapses. These endeavors have drawn scientists from nearly PhD with Teyler. This time, I prepared one slice at a time, and every field, making 3DEM even more exciting as many of the as can be seen in Figure A, the spines were exquisitely visible. imaging and reconstruction processes previously done man- Unfortunately, accurate counts and shape measurements of the ually are being automated. Figure C shows a recent 3DEM tiny spines were just beyond the resolution of light microscopy. rendering, with color-coding of organelles and synaptic com- While I was a graduate student, I talked my way into the ponents. It is indeed thrilling to be part of this growth. New esteemed summer course in neurobiology at the Marine findings abound regarding the plasticity of synapse structure Biological Laboratories in Woods Hole, Massachusetts. during normal changes in brain function and as altered by There I first learned serial-section three-dimensional electron diseases that tragically affect who we are as human beings. Axonal 3D-EM of Dendrite with Synapses (red) and Organelles Bouton Dendritic Dendrites Spine Neuron Cell Body Axons Dendrites Rapid Golgi preparation (Harris, 1980) Figure A Figure C 109–142_Bear_05_revised_final.indd 118 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 119 synapses, or Gray’s type II synapses (Figure 5.8). As we shall see later in the chapter, these structural differences reveal functional differences. Gray’s type I synapses are usually excitatory, while Gray’s type II syn- apses are usually inhibitory. The Neuromuscular Junction. Synaptic junctions also exist outside the CNS. For example, axons of the autonomic nervous system innervate glands, smooth muscle, and the heart. Chemical synapses also occur Asymmetrical Symmetrical membrane membrane between the axons of motor neurons of the spinal cord and skeletal (a) differentiations (b) differentiations muscle. Such a synapse is called a neuromuscular junction, and it has many of the structural features of chemical synapses in the CNS ▲ FIGURE 5.8 Two categories of CNS synaptic (Figure 5.9). membrane differentiations. (a) A Gray’s Neuromuscular synaptic transmission is fast and reliable. An action type I synapse is asymmetrical and potential in the motor axon always causes an action potential in the mus- usually excitatory. (b) A Gray’s type II cle cell it innervates. This reliability is accounted for, in part, by struc- synapse is symmetrical and usually tural specializations of the neuromuscular junction. Its most important inhibitory. specialization is its size—it is one of the largest synapses in the body. The presynaptic terminal also contains a large number of active zones. In addition, the postsynaptic membrane, also called the motor end- plate, contains a series of shallow folds. The presynaptic active zones are precisely aligned with these junctional folds, and the postsynaptic membrane of the folds is packed with neurotransmitter receptors. This structure ensures that many neurotransmitter molecules are focally released onto a large surface of chemically sensitive membrane. Because neuromuscular junctions are more accessible to researchers than CNS synapses, much of what we know about the mechanisms of synaptic transmission was first established here. Neuromuscular junc- tions are also of considerable clinical significance; diseases, drugs, and poisons that interfere with this chemical synapse have direct effects on vital bodily functions. PRIN PRINCIPLES NCIPLE ES OF CHEMICAL S SYNAPTIC YNAPTIC TRAN NSMISSSION TRANSMISSION Consider the basic requirements of chemical synaptic transmission. There must be a mechanism for synthesizing neurotransmitter and packing it into the synaptic vesicles, a mechanism for causing vesicles to spill their contents into the synaptic cleft in response to a presynaptic action poten- tial, a mechanism for producing an electrical or biochemical response to neurotransmitter in the postsynaptic neuron, and a mechanism for remov- ing neurotransmitter from the synaptic cleft. And, to be useful for sensa- tion, perception, and the control of movement, all these things must often occur very rapidly, within milliseconds. No wonder physiologists were ini- tially skeptical about the existence of chemical synapses in the brain! Fortunately, thanks to several decades of research on the topic, we now understand how many of these aspects of synaptic transmission are so efficiently carried out. Here we’ll present a general survey of the basic principles. In Chapter 6, we will take a more detailed look at the indi- vidual neurotransmitters and their modes of postsynaptic action. Neurotransmitters Since the discovery of chemical synaptic transmission, researchers have been identifying neurotransmitters in the brain. Our current understand- ing is that the major neurotransmitters fall into one of three chemical 109–142_Bear_05_revised_final.indd 119 12/20/14 3:43 AM  120 PART ONE FOUNDATIONS ▲ FIGURE 5.9 The neuromuscular junction. The postsynaptic membrane, known as the motor end-plate, contains junctional folds with numerous neurotransmitter receptors. Motor neuron Muscle fibers Myelin Axon Neuromuscular junction Synaptic vesicles Active zone Synaptic cleft Receptors Junctional fold Postsynaptic muscle fiber Presynaptic Motor end-plate terminals (postsynaptic) region categories: (1) amino acids, (2) amines, and (3) peptides (Table 5.1). Some representatives of these categories are shown in Figure 5.10. The amino acid and amine neurotransmitters are all small organic molecules containing at least one nitrogen atom, and they are stored in and released from synaptic vesicles. Peptide neurotransmitters are large molecules—chains of amino acids—stored in and released from secretory granules. As discussed ear- lier, secretory granules and synaptic vesicles are frequently observed in 109–142_Bear_05_revised_final.indd 120 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 121 TABLE 5.1 The Major Neurotransmitters Amino Acids Amines Peptides Gamma-aminobutyric Acetylcholine (ACh) Cholecystokinin (CCK) acid (GABA) Dopamine (DA) Dynorphin Glutamate (Glu) Epinephrine Enkephalins (Enk) Glycine (Gly) Histamine N-acetylaspartylglutamate (NAAG) Norepinephrine (NE) Neuropeptide Y Serotonin (5-HT) Somatostatin Substance P Thyrotropin-releasing hormone Vasoactive intestinal polypeptide (VIP) COOH COOH CH2 CH2 CH2 CH2 NH2 CH COOH NH2 CH NH2 CH2 COOH (a) Glu GABA Gly HO O CH3 OH + CH3 C O CH2 CH2 N CH3 HO CH CH2 NH2 CH3 (b) ACh NE Carbon Oxygen Nitrogen Hydrogen Arg Pro Lys Pro Gln Gln Phe Phe Gly Leu Met Sulfur (c) Substance P ▲ FIGURE 5.10 Representative neurotransmitters. (a) The amino acid neurotransmitters gluta- mate, GABA, and glycine. (b) The amine neurotransmitters acetylcholine and nor- epinephrine. (c) The peptide neurotransmitter substance P. (For the abbreviations and chemical structures of amino acids in substance P, see Figure 3.4b.) 109–142_Bear_05_revised_final.indd 121 12/20/14 3:43 AM  122 PART ONE FOUNDATIONS the same axon terminals. Consistent with this observation, peptides often exist in the same axon terminals that contain amine or amino acid neu- rotransmitters. As we’ll discuss in a moment, these different neurotrans- mitters are released under different conditions. Different neurons in the brain release different neurotransmitters. The speed of synaptic transmission varies widely. Fast forms of synaptic transmission last from about 10–100 msec, and at most CNS synapses are mediated by the amino acids glutamate (Glu), gamma-aminobutyric acid (GABA), or glycine (Gly). The amine acetylcholine (ACh) medi- ates fast synaptic transmission at all neuromuscular junctions. Slower forms of synaptic transmission may last from hundreds of milliseconds to minutes; they can occur in the CNS and in the periphery and are medi- ated by transmitters from all three chemical categories. Neurotransmitter Synthesis and Storage Chemical synaptic transmission requires that neurotransmitters be synthe- sized and ready for release. Different neurotransmitters are synthesized in different ways. For example, glutamate and glycine are among the 20 amino acids that are the building blocks of protein (see Figure 3.4b); consequently, they are abundant in all cells of the body, including neurons. In contrast, GABA and the amines are made primarily by the neurons that release them. These neurons contain specific enzymes that synthesize the neurotransmit- ters from various metabolic precursors. The synthesizing enzymes for both amino acid and amine neurotransmitters are transported to the axon termi- nal, where they locally and rapidly direct transmitter synthesis. Once synthesized in the cytosol of the axon terminal, the amino acid and amine neurotransmitters must be taken up by the synaptic vesicles. Concentrating these neurotransmitters inside the vesicle is the job of transporters, special proteins embedded in the vesicle membrane. Quite different mechanisms are used to synthesize and store peptides in secretory granules. As we learned in Chapters 2 and 3, peptides are formed when amino acids are strung together by the ribosomes of the cell body. In the case of peptide neurotransmitters, this occurs in the rough ER. Generally, a long peptide synthesized in the rough ER is split in the Golgi apparatus, and one of the smaller peptide fragments is the active neurotransmitter. Secretory granules containing the peptide neurotransmitter bud off from the Golgi apparatus and are carried to the axon terminal by axoplasmic transport. Figure 5.11 compares the synthesis and storage of amine and amino acid neurotransmitters with that of peptide neurotransmitters. Neurotransmitter Release Neurotransmitter release is triggered by the arrival of an action poten- tial in the axon terminal. The depolarization of the terminal membrane causes voltage-gated calcium channels in the active zones to open. These membrane channels are very similar to the sodium channels we discussed in Chapter 4, except they are permeable to Ca2⫹ instead of Na⫹. There is a large inward driving force on Ca2⫹. Remember that the internal calcium ion concentration—[Ca2⫹]i—at rest is very low, only 0.0002 mM; therefore, Ca2⫹ will flood the cytoplasm of the axon terminal as long as the calcium channels are open. The resulting elevation in [Ca2⫹]i is the signal that causes neurotransmitter to be released from synaptic vesicles. The vesicles release their contents by a process called exocytosis. The membrane of the synaptic vesicle fuses to the presynaptic membrane at the active zone, allowing the contents of the vesicle to spill out into the synaptic cleft (Figure 5.12). Studies of a giant synapse in the squid nervous 109–142_Bear_05_revised_final.indd 122 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 123 Synaptic Precursor Active peptide vesicles peptide neurotransmitter Nucleus 3 4 1 2 3 Secretory granules Ribosome Golgi apparatus Rough ER Precursor molecule 1 Synthesizing enzyme (a) Neurotransmitter ▲ FIGURE 5.11 molecule The synthesis and storage of different types of neurotransmitter. (a) Peptides: ➀ A precursor peptide is synthesized in the rough endoplasmic reticulum. ➁ The 2 Transporter protein precursor peptide is split in the Golgi apparatus to yield the active neurotransmitter. ➂ Secretory vesicles containing the peptide bud off from the Golgi apparatus. ➃ The secretory granules are transported down the axon to the terminal where the peptide is stored. (b) Amine and amino acid neurotransmitters: ➀ Enzymes convert Synaptic precursor molecules into neurotransmitter molecules in the cytosol. ➁ Transporter vesicle proteins load the neurotransmitter into synaptic vesicles in the terminal where they (b) are stored. Presynaptic Synaptic 4 1 vesicle 3 Active 2 zone Synaptic cleft Voltage-gated Neurotransmitter calcium channel molecules Postsynaptic ▲ FIGURE 5.12 The release of neurotransmitter by exocytosis. ➀ A synaptic vesicle loaded with neurotransmitter, in response to ➁ an influx of Ca2⫹ through voltage-gated calcium channels, ➂ releases its contents into the synaptic cleft by the fusion of the vesicle membrane with the presynaptic membrane, and ➃ is eventually recycled by the process of endocytosis. 109–142_Bear_05_revised_final.indd 123 12/20/14 3:43 AM  124 PART ONE FOUNDATIONS system showed that exocytosis can occur very rapidly, within 0.2 msec of the Ca2⫹ influx into the terminal. Synapses in mammals, which gener- ally occur at higher temperatures, are even faster. Exocytosis is quick be- cause Ca2⫹ enters at the active zone precisely where synaptic vesicles are ready and waiting to release their contents. In this local “microdomain” around the active zone, calcium can achieve relatively high concentra- tions (greater than about 0.01 mM). The mechanism by which [Ca2⫹]i stimulates exocytosis has been under intensive investigation. The speed of neurotransmitter release suggests that the vesicles involved are those already “docked” at the active zones. Docking is believed to involve interactions between proteins in the synap- tic vesicle membrane and the presynaptic cell membrane under the active zone (Box 5.3). In the presence of high [Ca2⫹]i, these proteins alter their conformation so that the lipid bilayers of the vesicle and presynaptic mem- branes fuse, forming a pore that allows the neurotransmitter to escape into the cleft. The mouth of this exocytotic fusion pore continues to expand until the membrane of the vesicle is fully incorporated into the presynaptic mem- brane (Figure 5.13). The vesicle membrane is later recovered by the process of endocytosis, and the recycled vesicle is refilled with neurotransmitter (see Figure 5.12). During periods of prolonged stimulation, vesicles are mo- bilized from a “reserve pool” that is bound to the cytoskeleton of the axon terminal. The release of these vesicles from the cytoskeleton, and their docking to the active zone, is also triggered by elevations of [Ca2⫹]i. Secretory granules also release peptide neurotransmitters by exo- cytosis, in a calcium-dependent fashion, but typically not at the active zones. Because the sites of granule exocytosis occur at a distance from the sites of Ca2⫹ entry, peptide neurotransmitters are usually not released in response to every action potential invading the terminal. Instead, the release of peptides generally requires high-frequency trains of action potentials, so that the [Ca2⫹]i throughout the terminal can build to the level required to trigger release away from the active zones. Unlike the fast release of amino acid and amine neurotransmitters, the release of peptides is a leisurely process, taking 50 msec or more. Neurotransmitter Receptors and Effectors Neurotransmitters released into the synaptic cleft affect the postsynaptic neuron by binding to specific receptor proteins that are embedded in the postsynaptic density. The binding of neurotransmitter to the receptor is like inserting a key in a lock; this causes conformational changes in the protein such that the protein can then function differently. Although there are well over 100 different neurotransmitter receptors, they can be classified into two types: transmitter-gated ion channels and G-protein-coupled receptors. Transmitter-Gated Ion Channels. Receptors known as transmitter-gated ion channels are membrane-spanning proteins consisting of four or five subunits that come together to form a pore between them (Figure 5.14). In the absence of neurotransmitter, the pore is usually closed. When neu- rotransmitter binds to specific sites on the extracellular region of the channel, it induces a conformational change—just a slight twist of the subunits—which within microseconds causes the pore to open. The func- tional consequence of this depends on which ions can pass through the pore. Transmitter-gated channels generally do not show the same degree of ion selectivity as do voltage-gated channels. For example, the ACh-gated ion channels at the neuromuscular junction are permeable to both Na⫹ and K⫹. Nonetheless, as a rule, if the open channels are permeable to 109–142_Bear_05_revised_final.indd 124 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 125 BOX 5.3 BRAIN FOOD How to SNARE a Vesicle Y easts are single-cell organisms valued for their abil- ity to make dough rise and grape juice ferment into wine. We have a way to go before we understand all the mol- ecules involved in synaptic transmission. In the meantime, Remarkably, the humble yeasts have some close similarities we can count on yeasts to provide delightful brain food (and to the chemical synapses in our brain. Recent research has drink) for thought. shown that the proteins controlling secretion in both yeast cells and synapses have only minor differences. Apparently, these molecules are so generally useful that they have been Neurotransmitter Vesicle Synaptotagmin conserved across more than a billion years of evolution, and they are found in all eukaryotic cells. The trick to fast synaptic function is to deliver neurotrans- mitter-filled vesicles to just the right place—the presynaptic membrane—and then cause them to fuse at just the right time, Vesicle when an action potential delivers a pulse of high Ca2⫹ concen- t-SNARES membrane v-SNARE tration to the cytosol. This process of exocytosis is a special case of a more general cellular problem, membrane trafficking. Cells have many types of membranes, including those enclos- ing the whole cell, the nucleus, endoplasmic reticulum, Golgi apparatus, and various types of vesicles. To avoid intracellular chaos, each of these membranes needs to be moved and deliv- ered to specific locations within the cell. After delivery, one type of membrane often has to fuse with another type. A common Presynaptic Calcium molecular machinery has evolved for the delivery and fusion of terminal membrane channel all these membranes, and small variations in these molecules determine how and when membrane trafficking takes place. The specific binding and fusion of membranes seem to depend on the SNARE family of proteins, which were first found in yeast cells. SNARE is an acronym too convoluted to define here, but the name perfectly defines the function of these proteins: SNAREs allow one membrane to snare an- other. Each SNARE peptide has a lipid-loving end that em- beds itself within the membrane and a longer tail that projects into the cytosol. Vesicles have “v-SNAREs,” and the outer membrane has “t-SNAREs” (for target membrane). The cy- tosolic ends of these two complementary types of SNAREs can bind very tightly to one another, allowing a vesicle to “dock” very close to a presynaptic membrane and nowhere else (Figure A). Although complexes of v-SNAREs and t-SNAREs form the main connection between vesicle membrane and target membrane, a large and bewildering array of other presyn- aptic proteins stick to this SNARE complex. We still don’t understand the functions of all of them, but synaptotagmin, a vesicle protein, is the critical Ca2⫹ sensor that rapidly trig- gers vesicle fusion and thus transmitter release. On the pre- synaptic membrane side, calcium channels may form part of the docking complex. As the calcium channels are very close to the docked vesicles, inflowing Ca2⫹ can trigger transmit- ter release with astonishing speed—within about 60 ␮sec in a mammalian synapse at body temperature. The brain has several varieties of synaptotagmins, including one that is spe- Figure A cialized for exceptionally fast synaptic transmission. SNAREs and vesicle fusion. 109–142_Bear_05_revised_final.indd 125 12/20/14 3:43 AM  126 PART ONE FOUNDATIONS ▲ FIGURE 5.13 A “receptor’s eye” view of neurotrans- mitter release. (a) This is a view of the extracellular surface of the active zone of a neuromuscular junction in frog. The particles are believed to be calcium channels. (b) In this view, the presynap- Presumed tic terminal had been stimulated to calcium release neurotransmitter. The exocytotic channels fusion pores are where synaptic vesicles have fused with the presynaptic mem- brane and released their contents. (Source: Heuser and Reese, 1973.) (a) Exocytotic fusion pore (b) Na⫹, the net effect will be to depolarize the postsynaptic cell from the resting membrane potential (Box 5.4). Because it tends to bring the mem- brane potential toward threshold for generating action potentials, this effect is said to be excitatory. A transient postsynaptic membrane depo- larization caused by the presynaptic release of neurotransmitter is called Membrane an excitatory postsynaptic potential (EPSP) (Figure 5.15). Synaptic activation of ACh-gated and glutamate-gated ion channels causes EPSPs. If the transmitter-gated channels are permeable to Cl⫺, the usual net Cytoplasm effect will be to hyperpolarize the postsynaptic cell from the resting mem- brane potential (because the chloride equilibrium potential is usually neg- (a) ative; see Chapter 3). Because it tends to bring the membrane potential away from threshold for generating action potentials, this effect is said to be inhibitory. A transient hyperpolarization of the postsynaptic mem- brane potential caused by the presynaptic release of neurotransmitter is called an inhibitory postsynaptic potential (IPSP) (Figure 5.16). Synaptic activation of glycine-gated or GABA-gated ion channels cause an IPSP. We’ll discuss EPSPs and IPSPs in more detail shortly when we explore the principles of synaptic integration. G-Protein-Coupled Receptors. Fast chemical synaptic transmission is mediated by amino acid and amine neurotransmitters acting on trans- (b) mitter-gated ion channels. However, all three types of neurotransmitter, ▲ FIGURE 5.14 acting on G-protein-coupled receptors, can also have slower, longer The structure of a transmitter-gated lasting, and much more diverse postsynaptic actions. This type of trans- ion channel. (a) Side view of an ACh- mitter action involves three steps: gated ion channel. (b) Top view of the channel, showing the pore at the center 1. Neurotransmitter molecules bind to receptor proteins embedded in the of the five subunits. postsynaptic membrane. 109–142_Bear_05_revised_final.indd 126 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 127 BOX 5.4 BRAIN FOOD Reversal Potentials I n Chapter 4, we saw that when the membrane voltage- threshold, the neurotransmitter action would be termed gated sodium channels open during an action potential, Na⫹ excitatory. As a rule, neurotransmitters that open a chan- enters the cell, causing the membrane potential to rapidly nel permeable to Na⫹ are excitatory. If a neurotransmitter depolarize until it approaches the sodium equilibrium poten- causes Vm to take on a value that is more negative than the tial, ENa, about 40 mV. Unlike the voltage-gated channels, action potential threshold, the neurotransmitter action would however, many transmitter-gated ion channels are not per- be termed inhibitory. Neurotransmitters that open a channel meable to a single type of ion. For example, the ACh-gated permeable to Cl⫺ tend to be inhibitory, as are neurotransmit- ion channel at the neuromuscular junction is permeable to ters that open a channel permeable only to K⫹. both Na⫹ and K⫹. Let’s explore the functional consequence of activating these channels. At positive membrane In Chapter 3, we learned that the membrane potential, potentials, ACh causes Vm, can be calculated using the Goldman equation, which outward current takes into account the relative permeability of the membrane to different ions (see Box 3.3). If the membrane were equally permeable to Na⫹ and K⫹, as it would be if many ACh- or Out Membrane glutamate-gated channels were open, then Vm would have current a value between ENa and EK, around 0 mV. Therefore, ionic current would flow through the channels in a direction that brings the membrane potential toward 0 mV. If the mem- Membrane brane potential were ⬍0 mV before ACh was applied, as voltage I-V plot during is usually the case, the direction of net current flow through ACh application the ACh-gated ion channels would be inward, causing a depolarization. However, if the membrane potential were ⬎0 mV before ACh was applied, the direction of net current –60 mV 60 mV flow through the ACh-gated ion channels would be outward, Reversal causing the membrane potential to become less positive. potential Ionic current flow at different membrane voltages can be graphed, as shown in Figure A. Such a graph is called an I-V plot (I: current; V: voltage). The critical value of membrane potential at which the direction of current flow reverses is called the reversal potential. In this case, the reversal potential In would be 0 mV. The experimental determination of a rever- sal potential, therefore, helps tell us which types of ions the membrane is permeable to. At negative membrane potentials, ACh causes If, by changing the relative permeability of the membrane inward current to different ions, a neurotransmitter causes Vm to move to- ward a value that is more positive than the action potential Figure A 2. The receptor proteins activate small proteins, called G-proteins, which are free to move along the intracellular face of the postsynaptic membrane. 3. The activated G-proteins activate “effector” proteins. Effector proteins can be G-protein-gated ion channels in the membrane (Figure 5.17a), or they can be enzymes that synthesize molecules called second messengers that diffuse away in the cytosol (Figure 5.17b). Second messengers can activate additional enzymes in the cytosol that can regulate ion channel function and alter cellular metabolism. Because G-protein-coupled receptors can trigger widespread metabolic effects, they are often referred to as metabotropic receptors. 109–142_Bear_05_revised_final.indd 127 12/20/14 3:43 AM  128 PART ONE FOUNDATIONS Action potential Axon Axon terminal (a) Postsynaptic dendrite Neurotransmitter Record Vm molecules Synaptic cleft EPSP Vm Cytosol – 65 mV Transmitter-gated 0 2 4 6 8 (b) ion channels (c) Time from presynaptic action potential (msec) ▲ FIGURE 5.15 The generation of an EPSP. (a) An action potential arriving in the presynaptic terminal causes the release of neurotransmitter. (b) The molecules bind to transmitter-gated ion channels in the postsynaptic membrane. If Na⫹ enters the postsynaptic cell through the open channels, the membrane will become depolarized. (c) The resulting change in membrane potential (Vm), as recorded by a microelectrode in the cell, is the EPSP. We’ll discuss the different neurotransmitters, their receptors, and their effectors in more detail in Chapter 6. However, you should be aware that the same neurotransmitter can have different postsynaptic actions, de- pending on what receptors it binds to. An example is the effect of ACh on the heart and on skeletal muscles. ACh slows the rhythmic contractions of the heart by causing a slow hyperpolarization of the cardiac muscle cells. In contrast, in skeletal muscle, ACh induces contraction by caus- ing a rapid depolarization of the muscle fibers. These different actions are explained by the different receptors involved. In the heart, a metabo- tropic ACh receptor is coupled by a G-protein to a potassium channel. The opening of the potassium channel hyperpolarizes the cardiac muscle fibers and reduces the rate at which it fires action potentials. In skeletal muscle, the receptor is a transmitter-gated ion channel, specifically an ACh-gated ion channel, permeable to Na⫹. The opening of this channel depolarizes the muscle fibers and makes them more excitable. Autoreceptors. Besides being a part of the postsynaptic density, neu- rotransmitter receptors are also commonly found in the membrane of the presynaptic axon terminal. Presynaptic receptors that are sensitive to the 109–142_Bear_05_revised_final.indd 128 12/20/14 3:43 AM  CHAPTER 5 SYNAPTIC TRANSMISSION 129 Action potential Axon Axon terminal (a) Postsynaptic dendrite Neurotransmitter Record Vm molecules Cl Cl Cl Synaptic cleft IPSP Vm Cytosol – 65 mV Transmitter-gated 0 2 4 6 8 (b) ion channels (c) Time from presynaptic action potential (msec) ▲ FIGURE 5.16 The generation of an IPSP. (a) An action potential arriving in the presynaptic terminal causes the release of neurotransmitter. (b) The molecules b