 Good morning. I hope everyone had a nice weekend. I have to apologize for two minor mishaps last Friday. First, I forgot to press the start button again on the recorder after the quiz, so everything after the quiz is, of course, on the PowerPoint, but not on the video. Second, the TAs, especially Jaren, were quite zealous, commendably so, about grading the quiz, but did not realize that anything within an order of magnitude would be fine for most of the responses. And then Jaren efficiently handed back the quizzes, told me what he had done, and we agreed that everybody got one bonus point to make up for the two-strict grading. So that's how you got the bonus point. We are going to talk about post-synaptic events, and we're going to make the point that synaptic transmission is a delta function in time. We're going to talk about the underlying ion channel physiology, and we are also going to talk about channel blockers. And the readings in Candel are at the bottom, and I believe they are specified. Yeah, they are in Candel. Okay, so last time we talked about the timing of synaptic events. We talked about a very favorable preparation, the giant synapse of the squid, the presynaptic, and then the post-synaptic, which is the squid giant axon. We talked about the timing elements, the synaptic delay, depending on the temperature, is about half a millisecond, and most of that delay is actually caused by opening calcium channels during the presynaptic action potential, presynaptic. And the size and timing of the EPSPs then can be modulated by prolonging the action potential, and this often happens in neuroscience. Now we go to look at post-synaptic events in more detail. And the favorite preparation for teaching about post-synaptic events is the nerve muscle synapse, also called the neuromuscular junction, also called the end plate. And you can see why it's called the end plate here, because there seems to be a bit of a plaque. Now, you remember the trick we did last time, which was to reduce the size of the post-synaptic potential, so that it did not fire a post-synaptic spike, did not reach threshold. Therefore, we can study details of the post-synaptic potential without confusing ourselves with sodium channels. And so the experiment is to place a microelectrode into a muscle cell at various points away from the end plate or the nerve muscle synapse, a few millimeters. And you see that the size of the end plate potential dies off, just as though it were a leaky cable. Leaky cables were a really big deal in electrical engineering at the end of the 18th century, sorry, 19th century, when cables were being laid across the Atlantic and people needed to know how many repeaters to put in. Now, if you look a little more and the reason that the synaptic potential declines in amplitude is that, of course, all of those sodium currents are flowing right in where the sodium channels are at the end plate or at the nerve muscle synapse. And then Kirchhoff's law of conservation of charge, we need to complete the circuit. They flow out more gradually, mostly through potassium and chloride channels further along the fiber. Now, a point that may be missed is that in addition to the amplitude becoming smaller, the peak becomes later as one goes away from the end plate. That's a standard characteristic of the cable properties. So, up until now, then, we have discussed ion channels that are gated by electric fields or more precisely by changes in the electric field. We've talked about voltage gated sodium channels and voltage gated potassium channels and voltage gated calcium channels. And one of you asked quite presciently in the last lecture whether voltage gated calcium channels could also support action potentials. The name is yes under special circumstances. Now, the other great class of ion channels are those gated by chemicals rather than by electricity. In the best example, though, as we'll see, not the only example, are channels gated by neurotransmitters. They go from closed to open as well. Today, we are going to do a specific case of transmitter operated canals, the acetylcholine gated excitatory channel. So, ACH is acetylcholine. And we also, well, I've told you that many basic principles were discovered at the nerve muscle synapse and that acetylcholine is the transmitter there. And I've showed you pictures about the fine structure of the nerve muscle synapse or the end plate and emphasized to you the wondrously complex structure of the synapse. We have the Schwann cell and sheathing it. We have the presynaptic terminal. We have mitochondria in the presynaptic terminal, synaptic vesicles. We have folds along the post-synaptic membrane, presumably, and the hero of the story today is very much the acetylcholine receptors. The function of the selective advantage of the folds is presumably to fit as many acetylcholine receptors as possible so that each acetylcholine molecule when released does find a receptor. Now, in addition to this example from the peripheral nervous system, I like to point out examples from the central nervous system. And this is an example of acetylcholine receptors playing an important role in the central nervous system. This is a coronal section of a mouse. And it is stained with a special stain that reveals one of the enzymes that make dopamine. It's called tyrosine hydroxylase. And indeed, these are dopamine-ergic neurons. These are neurons that release the neurotransmitter dopamine. And you can see, in fact, that they are in the shape of a handlebar moustache. The handlebars are the so-called substantia nigra. And those cells are packed very close together. So that part of the substantia nigra is called the pars compacta. The other part of the substantia nigra is a little more spread out. It's called the dentral tegmental area. Both of those regions in the handlebar moustache, the handlebars and the upper lip, make dopamine. And both regions have lots of acetylcholine receptors on them. They have a special type of acetylcholine receptors called nicotinic acetylcholine receptors. The mammalian genome actually contains lots of subunits for nicotinic receptors. The major brain nicotinic acetylcholine receptors have alpha-4 and beta-2 subunits in them. Their stoichiometry is uncertain. Other brain nicotinic receptors are HOMO pentamers of the so-called alpha-7 subunit. And so this is one region where nicotine acts very importantly, particularly in the ventral tegmental area, because the ventral tegmental area is the region that produces a sense of well-being or reward when nicotine enters the central nervous system. The substantia nigropars verticulata is also another region called GABA, that is inhibitory. Now for today's tough question, who introduced nicotine to European culture? Would anyone like a hint? Here is the hint. Yes, it was Columbus. His crew first sampled tobacco in 1492. All right, here is a nicotinic receptor. I have to remove that for next year. Here is a nicotinic receptor in most of its glory as a protein. There are actually two views of the nicotinic receptor. One of them is alpha helices, the red helices, plus beta strands, beta sheets, which are green. The other view is colored by subunit. There are five subunits. There is the region on the outside of the membrane, the extracellular solution, where ligands, where acetylcholine or nicotine bind. They bind to the same place. I'll show you that in a moment. There is the transmembrane region. We'll discuss that in a moment. There is the cytoplasmic region, which is responsible for interactions with other proteins for putting the receptor in the right place and other very important cell biological functions. About 2,200 amino acids in five subunits, a pretty big protein. If we look right here and expand this image a great deal, we find that acetylcholine or nicotine bind at the interface between two subunits. This is a picture of nicotine binding at the interface between two subunits, and people at Caltech have studied this interaction a great deal. This is an x-ray crystallographic picture from another lab. We know within a fraction of an angstrom how nicotine binds and how it then activates the channel. To give you an overview of how this might work, we now move into the transmembrane part of the channel. Here is the acetylcholine or nicotine binding. Here is the transmembrane part of the channel. Because there are five pseudo-symmetric subunits, there are five subunits facing the channel. We think that the M2 helix of each subunit faces the channel. Only two of the subunits are shown here, not all five. You could think of gating of a receptor channel as convincing five Caltech professors all to stand around in a circle with their knees all facing inward, and their binding sites up here. When acetylcholine or nicotine binds to those channels up here, all Caltech professors do something that Caltech professors don't all do. They all act at the same time in agreement, and they swing their knees. That is important because we believe that swinging the knees, I'll show you on the next slide, that there are great controversies about what the biophysicists all do, the Caltech professors. In some accounts, they simply swivel their knees. In other accounts, they straighten up. But in any case, they change the conducting pathway. So the question is whether they twist or they straighten up between the closed and the open positions. But in any case, what appears to happen is that in the closed form of the channel, these professors all have oily patches on their knees, which prevent ions from flowing through. But the sides of their pants or legs all have the side chains with hydrophilic amino acids, lots of hydroxyl groups. And so these hydrophilic amino acids trick the permeant ions into thinking they are still in water and are hydrogen bonding with water, and so the ions flow through the channel. So that is more or less the way a typical chemical operated or neurotransmitter operated or ligand operated channel works. There are subunits which prevent the ions from flowing. The subunits change confirmation so that ions can more easily flow through the channel. Any questions? So we've talked about subunits, we've talked about Columbus, we've talked about Caltech professors. Let's talk about ions now. This is a recording from a muscle just sort of held there with no, held at minus 100 millivolts by putting a steady current. So it's not a voltage clamp experiment, people just put current into the muscle, or they hold it at plus 80 or various voltages in between millivolts. And then pulse acetylcholine onto that muscle so that the acetylcholine receptors open up. When held at minus 100, the acetylcholine tends to depolarize the muscle, make it less negative, more positive, goes towards zero. When held at plus 80, the acetylcholine tends to make the membrane less positive. And the result is that if you hold it at around zero or minus five millivolts, there are still channels opening, but there's no change in the current. And so most excitatory ligand-gated channel. Now, if you think about what we've told you so far, the Nernst potential for sodium is very positive, the Nernst potential for potassium is very negative. Clearly, this so-called equilibrium potential is neither. In fact, it is a blend of sodium and potassium, sometimes some calcium. And instead of calling it a Nernst potential or an equilibrium potential, because it's not, we call it a reversal potential because the sign of the response reverses around the reversal potential. And so at the reversal potential, the agonist has little effect on the membrane potential, negative to the reversal potential, which is where most cells like to sit right here around minus 80. The acetylcholine pulse gives a positive push to the membrane potential. That positive push, if it's pushy enough, exceeds threshold, the muscle or the nerve fires an action potential, and the synapse has done its work. So here we have the, we're going to introduce a new terminology for the conductance. We have resting potentials in the, resting channels in the muscle or the nerve, which holds the muscle or the nerve roughly at the Nernst potential for potassium around minus 90, minus 80 millivolts. And now we define a conductance, a bunch of little gammas for the excitatory postsynaptic potential, also called the in plate potential for a muscle. The conductance is in series with the battery, that battery is not a Nernst potential, it's a combination of Nernst potentials, mostly equal for sodium and potassium. So it's mostly in average around minus five millivolts. So we have, everybody explain what's being diagrammed here. So as usual, at the resting potential, potassium channels are always open and they dominate and bring the voltage right around here minus 80. But with an excitatory postsynaptic response or an in plate potential, the mixed sodium potassium channels open too. We have the, we go then to the membrane potential, which is dominated by the E EPSP, which is around minus, Detroit minus five. And as long as we get more positive than about minus 50, we fire a spike and everybody is happy. The synapse has done its work. Another aspect of ligand gated channels, acetylcholine gated channels, also recalls voltage gated channels. Here is a voltage, here is an experiment done in my lab at Caltech, in which we put acetylcholine receptors into a frog egg expressing alpha four beta two receptors, but you could do it in many different ways. We add nicotine for about 20 seconds, channels open, and then they don't remain open, they start to close. Now this may recall the voltage clamp experiments that I showed you with sodium channels. When you depolarize the sodium channels, they open and then they inactivate, or if you look at the current, the current reaches a negative peak and then it inactivates. So we call that inactivation at a voltage gated channel. We call that desensitization at a ligand gated channel, at a neurotransmitter gated channel. Actually naming something gives us no mechanistic insight at all into the cause, except that we can say in general that it appears as though the open conformation of the channel, right here activated, seems to be metastable in a physical chemical sense. That is if we wait long enough, the usual synaptic potential takes a millisecond and goes back and forth here, but if we keep the agonist on for a long time, it finds it gets over an energy hump, goes to a more stable state. If we keep the agonist open, if we keep the ligand available for longer, it finds another desensitized state until like any good physical chemical system, as you learned in chem one, it reaches a state of lowest free energy, which is in fact, like Titus binding. It is doubtful whether these longer reached states have any selective advantage, and as you'll see during normal synaptic transmission, the ligand is not around long enough to reach those states. So typically the ligand, the transmitter is around for only a millisecond or so. Desensitization presumably is just states of more stable drug receptor or chemical interactions. So let's go back to this idea that Feynman told me about in 1973. He said, to record single channels well enough, what you want to do is to get a pipette near enough to the membrane so that all the current flows through an electronic ammeter and you can measure it. So here is indeed a nicotinic acetylcholine receptor exposed to acetylcholine, and it opens for a few milliseconds, and then we have another one open. We don't know whether the second one that opens is the same one that opened first, because they are little gammas that are identical, but we know that there were two openings. And we know that we can measure just a couple of picohamperes, which is about 10 to the fourth ions per millisecond for about 20 milliseconds, and that this is quite accurate. So this magic seal that I told Feynman could never occur, and does in fact occur, that prevents current from flowing under the membrane is electrically tight prevents current from flowing under the membrane. Here we have a patch pipette in red. We have the rest of the cell with ion channel flowing. We have the ones that interest us here under the pipette. We make the seal with gentle suction by simply sucking up on the membrane through the pipette, and we get a nice seal. And so if we look in detail at what must be going on in this seal, we know it's electrically tight. Well, what does that mean? How tight is this seal? Well, if we use the stuff you learned in Phys 1B, which is resistivity, we can compute the resistance of this pathway by taking the resistivity times the length dividing it by the area. The resistance of one of these patch pipettes is around 10 to the ninth ohms. Resistivity of a typical blood solution is 22 ohms centimeters. We work it through and we discover that the thickness must be around two times 10 to the minus 11th meters, or less than an angstrom, which is ridiculous. So what this really tells us is that there is, in fact, the chemical reaction between the membrane and the glass of the pipette wall that makes a pretty good seal. And we know a little bit about this chemical reaction that's probably noncovalent between the head groups of the lipids and the silicates in the glass, but we're not sure. Now the other interesting aspect is that the giga ohm seal is chemically tight as well. So what chemically tight means is that it doesn't allow stuff inside the pipette to mix with stuff outside the pipette. So that, for instance, in this particular case, if we put acetylcholine in the pipette, it opens channels in the pipette, but not channels outside the pipette. If we put acetylcholine outside the pipette, it would open channels outside the pipette, we would not measure them, but not inside the pipette. And so we have a very nice tool for compartmentalization, which is really cool if you want to do some cell biology about where things are. All right, now there's this electronic ammeter that Kandel and Albert's talk about. And so we put a microelectrode in here. And here is an image from little Albert's, the essential Albert's, which you may take and buy one. And pretty embarrassingly, this oscilloscope shows a trace, but it doesn't even show you the electronic ammeter. It just shows you why are connected to an oscilloscope, which is not real. But luckily there's a guy named Sigworth. And he put the sensitive electronic ammeter in it right here. And so who is Sigworth? Well, I told you a story a couple of sessions ago about meeting Delbrook and Feynman. I also told you briefly about Carver Mead, who is now Professor Emeritus here. And during this session in 1973, Carver Mead told me, well, these are going to be very small signals. And if you want to measure small noisy signals, I have a senior who can help. Now, you may remember that Carver Mead invented a lot of very large scale integrated circuits. He also first enunciated Moore's Law. So when he tells you about electronic circuits, you need to listen. So the next day, this person, Fred Sigworth, 74, showed up in our lab. Now, if I remember correctly, Math 1 uses Apostle. And that's really the great thing about math and the difference between math and biology. In biology, you have to change the entire first chapter of the textbook every 10 years or so. In math, Apostle wrote this in the 60s, and it's still fine. Anyway, Fred Sigworth was a bit of an electronics genius. And Apostle, one of the great advantages of having written a textbook that will withstand the test of time, that is a math book, is that you can get your lectures down just perfectly. So here's Apostle talking to undergraduates. Anybody know which house this is? Glacker. Okay, sometime in the 60s, I guess, maybe early 70s. So this is Fred Sigworth at about the time of this photo. Apostle had his lectures timed just perfectly so that he would get in to the lecture hall. He would talk about calculus, and he would write the equations on the board. Exactly when he was finished, the clock would ring, and it was time for the next class. Well, Sigworth got to the clock in the back of the lecture hall. And Sigworth sped up the clock by one minute so that when Apostle finished his lecture one day, he looked up at the clock and he was one minute late. That was disconcerting to him. So he spoke a little bit faster during the next lecture, but Sigworth had sped up the clock by an additional minute. This is all in Legends of Caltech, right, which happened to be edited by Sigworth's father. This went on for about two weeks, and then Sigworth's true genius showed because he then started slowing down the clock, completely confusing Apostle. And if you grab Apostle, he'll chuckle at this story too. And Sigworth is today the very distinguished professor of patch clamp physiology at Yale. Professor of cellular and molecular physiology and of biomedical engineering at Yale. And he'll tell you this story when prompted as well. Anyway, Sigworth showed up in our lab in 1973 and he for then did not measure single channels because I wasn't listening to find them. But a few years later, he came up with a beautiful circuit for measuring single channels. And here then would be a typical record, well, a stylized record from one of Sigworth's amplifiers. This is a couple of picolampiers. So you keep Acetylcholine in the pipette, you keep the pipette attached to the cell. And at random intervals being Markovian rate constants, channels open. And at random intervals, they stay close. All right, now let's see how we analyze this. There are several ways to analyze it. But one way to analyze it is to take a pair of electronic scissors and digitally separate all the channels. So cut out the record here, back here, line it up with the first record, cut out this guy, line him up with the first record, and so on and so forth. So we are synchronizing all of the channels artificially on the opening event. Everybody understand what's going on here? And this is done pretty easily with various kinds of software. Now, the next thing that we do is to take all of these artificially synchronized channels and we line them up overlapping each other. And since we didn't get it exactly right, there's a little bit of jitter on the rising phase, but there's much more variation on the folding, on the falling phase, which is how long channels remain open. And so in fact, what I've done here is to show you a single channel or many single channels recorded with time, but what the cell does is to integrate the signal simultaneously from many channels that are acting independently. And in fact, this is more or less what happens at a synapse after the synapse is released to acetylcholine. The pulse of transmitter is rather brief. You could say that it's a delta function. And so all of the post-synaptic channels open nearly synchronously, but they close asynchronously. Everybody follow? Okay. So we can analyze this in more detail. We can make a histogram of many, many open channels after synchronizing on them and ask how long they remain open with time. So this is a histogram of open channel duration, counts per bin versus time. Like any good unimolecular event, this has a single exponential time course, and we can get fancy and give it a time constant. And in fact, we're going to get even fancier and call the time constant the inverse of something else, and I'll show you this in a minute. And this also reminds us of Chem 1B. This is when Jim Heath taught the course. So if a reaction is first order in a reactant A, then the rate of disappearance of A with time is equal simply to a rate constant times the concentration of A. And if we integrate between A0 and infinity, we come out with the concentration of our reactant decays exponentially with time with a rate constant or an inverse time constant equal to K. And so the same thing is in fact happening with ion channels. The same thing is happening with a synapse, brief pulse of acetylcholine or other transmitter, the open channels all open, and then they decay randomly. We can talk about two states, the open state and the closed state. So we talk about a rate constant from going from state 2 to state 1, and that's why we call this K from 2 to 1. So the concentration of acetylcholine at the nerve muscle synapse is very nearly a delta function, but the number of open channels way outlasts that concentration. And in fact, this is a way for the nerve cell or the muscle cell to integrate the contributions from presynaptic events that may occur asynchronously in time. But it's only one of the ways that the synapse integrates. So we're going to call it 1A, and in a future lecture, I'll show you about 1B and 2 and 3 and all that. Now what causes this rapid concentration pulse of acetylcholine is an enzyme. I did not emphasize it and should have, but in the synaptic cleft in the extracellular medium between the presynaptic nerve and the postsynaptic cell, including in the extracellular solution and in those folds, there's an enzyme called acetylcholine esterase. Okay, so this is a straightforward course. Acetylcholine, that's the acetate ester of choline, esterase. Well, it's a hydrolytic enzyme that breaks down acetylcholine into, you guessed it, acetate and choline. Neither acetate nor choline are good neurotransmitters, so the receptor stops functioning. Any questions? So I think in this very simple explanation, all of the molecules start here at t equals zero, and they then decay, and the units of this simple rate constant are simply in inverse seconds, the way units of any physical chemical rate constant would be. Acetylcholine esterase is quite efficient. It can hydrolyze one acetylcholine molecule every 100 microseconds, every tenth of a millisecond, and there are many thousands of acetylcholine esterase molecules per square micron of the membrane. So after being initially flooded and overwhelmed by the pulse of acetylcholine, the acetylcholine esterase can take care of the molecules without too much problem. We're going to talk about another kind of blocker, another kind of molecule. We're going to talk about blockers. This is lidocaine. Every molecule in neuroscience and in biology, every drug has several interesting groups. We won't go into this in detail in this course, but we will in neuropharmacology by 155 in the winter of 2017, not next term, but a year from next term. And so every molecule has some substituents. Most drugs that interest us in neuroscience have an amine which can be protonated, charged, and there are aromatic groups and amides. We won't go into that in great detail, except that we're going to say that there is a lidocaine analog. Lidocaine is a local anesthetic. We will come back to it later in the talk. It's like novocaine or procaine or benzocaine. Most use these days for sunburn medications, right? So if you get sunburn, you spray solarcane on you. That has one of the canes. I think it's lidocaine. And if you are unlucky enough to have grown up in a region where the water is not fluoridated, you have cavities, you go to the dentist. The dentist has to do some dental surgery. The typical local anesthetic there will also be lidocaine or a derivative. Now, here are some single molecules with a lidocaine analog. Here is a typical trace with acetylcholine only. And here is a typical trace with a analog of lidocaine with this fancy name QX222. You don't have to remember it. It looks quite different. It looks as though rats have been chewing on it. And if we look in detail now at some aspects of how drugs affect ion channels, what have I told you today? I've told you that some types of drugs compete with acetylcholine or replace acetylcholine at the interface between two subunits. That would be acetylcholine itself or nicotine. But the correct way of looking at blocking drugs such as lidocaine is a quite literal picture of a cork in a drain that plugs up the ion channel. So if we want to expand our state diagram a little bit, we imagine the functioning channel without a blocker in it. So this would be lidocaine. But the drug block channel with a plug in it. So let's define another molecular state of the channel, not open two, closed one, but blocked three. And so we have a few more rate constants to go from open to blocked. And this turns out to be fairly interesting. I won't go into all of the details, but here the units of channels closing normally are inverse seconds. The units of drug block, though, since this is a second-order reaction, we have two molecules, the drug and the channel. Those units, therefore, are inverse moles for inverse seconds, exactly as you learned in Chem 1. So what happens with a drug block channel in our simple diagram? We have current versus time. We've synchronized all the channels. Without a drug present, they close normally. But if a drug can come and block them, producing a premature termination, that's these blue lines here, this blue line here, while many of the closings, not all of them, occur sooner, many of the openings are briefer. Every once in a while, the channel closes on its own. But a large fraction of the time, especially for an effective blocker at a high concentration, the drug comes in and blocks the channel. So if we were to think about two kinds of blockers, we have our standard open and closed states here at the top, a rapidly, more rapidly binding blocker would come in, leave the channel, another one would come in, leave the channel, et cetera, and we have a channel chattering. A drug that binds very slowly and still blocks would bind here, shorten the channel a little bit, and then stay bound so that the channel could not open again for quite a long time, might unbind here, but if the channel is not ready to open, it won't open until the next time. And so, from single channel records, presuming Markov process is presuming first order transitions among states, we can tell a lot about the behavior of individual molecules, and in fact, single channel recordings were really the first example of detailed single molecule recordings, which have added so much to biology and biophysics and bioengineering over the years. And in particular, then, we were talking about drug block, and I mentioned lidocaine, and lidocaine has this interesting characteristic that in most cases it does not block from the outside, but it blocks from the inside of the channel, and in fact, most sodium channels have a large orifice facing the inside of the cell, facing the cytoplasm, which is an invitation for lidocaine to come in and block that channel. And lidocaine is a weak base. It exists in the neutral form and in the protonated form, that amino group that I told you about can gain or lose a proton. And so, an amazing series of events occur in which lidocaine in its neutral form can easily permeate membranes because it's not charged, it goes through lipids, can pick up or lose a proton in just a couple of milliseconds, and in the protonated form can block channels from the inside. Another very interesting aspect, which we will go into by 155 in greater detail, is the idea that at times, the channel can actually close even though the blocker is in it. So that means that the blocker gets trapped in the channel. It becomes a trapped or a use-dependent blocker. The more the channel is open, the more time the blocker has to find it and then gets stuck in it as though it were a trap. And so, use-dependent blockers are extremely important in medicine. Let's talk about a derivative of procaine, the first local anesthetic called procaine amide. Now imagine these are action potentials, and we can stimulate these action potentials with repetitive pulses. So the channel population of sodium channels opens robustly here. A little bit of procaine amide comes in, blocks the channel, gets stuck in the channel. So the next time we have a smaller response, those channels that are open allow a little bit more procaine to come in, block, it gets stuck, and so on. As a result, if we are stimulating repetitively, after a while, too few sodium channels can open to cause a regenerating action potential, and we lose the action potentials. Now, of course, these procaine amide ions eventually fall off the channel, and so if we space the stimuli rapidly enough, we don't get the kind of decrement that we do. Sorry, if we space the stimuli slowly enough, we don't get the kind of decrement that we get when we rapidly stimulate the neuron or the muscle. So that's why we call procaine amide and other drugs like this use-dependent blockers. And if we put the threshold right around here, we can see that after a while, use-dependent blockers stop a cell from firing spikes, but if we stimulate slowly enough, then the cell can fire spikes for a long time. So, again, we have a functioning channel and a trapped or use-dependent blocker. Now, use-dependent block phenomenon is extremely important if we want to stop cells from firing action potentials too frequently. What's a good example of this? Well, in the heart, arrhythmia is certainly a good example. And in fact, procaine amide, the drug that I just discussed with you, is a favorite for atrial fibrillation, which is the atrium firing too frequently. Another favorite example of use-dependent blockers is anti-apoleptics or anti-convulsants. And so, most anti-convulsive drugs are use-dependent blockers. The best known of them is phenytoin. Its trademark is dilantin, but as usual in this course, we are not responsible for trademarks because they vary so much within companies and around the world. So, local anesthetics, which block, even if the channel doesn't open, are useful for dental surgery, for sunburn medications. Use-dependent blockers that actually get stuck in the channel are quite useful for anti-arhythmic drugs or for anti-apoleptics. Any questions? Later on in this course, we're going to talk about superfamilies of other ligand-gated channels that are synaptic receptors. Today, we've talked about acetylcholine. Each subunit has four transmembrane domains that are five subunits total. We're going to talk about glutamate receptors and their topology, which is different. We won't talk about ATP-gated channels at all. So, a reminder that my office hours are today from 1.15 to 2.00 outside the red door, and if you're interested in learning more about how drugs open and block ion channels, that's part of neuropharmacology, and I'll teach it a year from this winter. See you Wednesday.