I suppose I thought for a while that if I was talking about dopamine and serotonin and GABA and things enough, people would just kind of "get" neurotransmission. And most people do. But it's still a good thing to cover, partially because it's kind of mind boggling to think about (well, Sci finds it mind-boggling), and partially because it helps you understand why changes in receptors, changes in transporters, or changes in release will have different effects. This comes in very handy when talking about various psychiatric and addictive drugs of which I am very fond. And so, your general post today: Neurotransmission.
And also, I get to DRAW!!! w00t.
The synapse. Do not be fooled by its commonplace appearance. Like so many things, it is not what is on the outside, but what is on the inside that counts. 
So what are we looking at here? That blue bulbous portion that looks like a nose is the presynaptic neuron. The smiley below it in pink is the postsynaptic neuron. And neurotransmission is what gets a signal from one side to the other.
Now the presynaptic neuron has a signal. This stimulus is transmitted as an action potential eletrically down the neuron until it gets to the bulge in the picture, the synaptic buton.
But the electrical signal cannot just bounce on to the next neuron. There's too much space in between the two neurons in the synaptic cleft. Instead, the stimulus of the action potential causes a rush of calcium ions into the synaptic buton, rapidly changing the potential inside.
This change in potential is going to affect little vesicles, little blobs of membrane inside the presynaptic neuron. These vesicles contain neurotransmitters, chemicals synthesized in the presynaptic cell, and stored in the vesicles until stimulated.
When the vesicles are stimulated by this influx of calcium caused by the approaching action potential, the vesicles begin to migrate to the cell membrane. Then, they can either merge with the membrane and release all their neurotransmitter into the synapse, or they can perform a "kiss and run" opening briefly at the membrane and only releasing a little of the neurotransmitter. It's thought right now that the kiss and run is more common than dumping all the neurotransmitter in there.
So now the neurotransitting chemicals are in the synapse. They float across the tiny space in a random way, and in the process, bump into receptors on the other side.
Keep in mind, though. The neurotransmitters are not taken up by the receptors. Instead, they bind, and the receptor, which runs through the membrane in the postsynaptic cell, changes conformation on the inside of the cell, causing activation of pathways.
The receptors here are important. This is because there tend to be many different types of receptor for one type of neurotransmitter. For example, serotonin has 17 known receptors, and there might be more. The type of receptor on the postsynaptic neuron determines how the cell will react to the signal. This is a lot more refined than depending on neurotransmitter release. You can only change the AMOUNT of neurotransmitter released, not whether or not that neurotransmitter will be excitatory or inhibitory. That is left to the receptors. So depending on what the neurotransmitter hits, the result could be excitation or inhibition of the postsynaptic neuron's action potential, or something even more complicated, like activation of specific gene pathways to produce specific proteins.
Not only that, receptor sensitivity to stimulation can change, either by changing the number of receptors at the postsynaptic membrane, or changing the sensitivity of the receptors that are there. There are lots of way to control how much and what kind of signals are getting across, and previous stimulations received will influence how the postsynaptic cell is capable of reacting later. These changes can be short-term or long-term, and can be responsible for starting processes like memory formation, learning, and addiction, as well as tons of other things.
So what happens then? You don't want to leave the neurotransmitter sitting around in the synapse. Because it's floating around at random, sitting in the synapse means it will continue to bump into receptors and pass signals on to the post-synaptic neuron. So the signal must be terminated. Depending on the neurotransmitter you're dealing with (dopamine, serotonin, GABA, glutamate, acetylcholine, the list goes on), there are carious things that can happen. An enzyme can break down the neurotransmitter chemical into its component parts, or the presynaptic neuron can have transporters, which suck the neurotransmitter up back into the synaptic buton, either to be shoved back into vesicles, or to be degraded.
And the synapse clears out, vesicles fill up, calcium goes back out of the presynaptic neuron, and it's all ready to begin again.
That's a really, really basic picture of what's going on at a synapse. But what, you may ask, is so mind-boggling about that? What boggles Sci's mind is the tiny scale on which this is happening (the order of microns, a micron is 0.000001m), and the SPEED. This happens FAST. Every movement of your fingers requires THOUSANDS of these signals. Every new fact you learn requires thousands more. Heck, every word your are looking at, just the ACT of LOOKING and visual signals coming into your brain. Millions of signals, all over the brain, per second. And out of each tiny signal, tiny things change, and those tiny changes determine what patterns are encoded and what are not. Those patterns can determine something like what things you see are remembered or not. And so, those millions of tiny signals will determine how you do on your calculus test, whether you swerve your car away in time to miss the stop sign, and whether you eat that piece of cake.
If that's not mind-boggling, what IS?!
That is, indeed, mind-boggling.
Can I steal that post?
Complex stuff; excellently, put simply.
Hi Sci, some minor points:
So what are we looking at here? That blue bulbous portion that looks like a nose is the presynaptic neuron. The smiley below it in pink is the postsynaptic neuron. And neurotransmission is what gets a signal from one side to the other.
The drawing may confuse people ("why is one neuron shaped like a doorknob, the other like a crescent?"). Really, what you drew was a bouton facing the postsynaptic membrane of maybe a dendritic spine. So it's parts of the two neurons, not the whole neurons.
But the electrical signal cannot just bounce on to the next neuron. There's too much space in between the two neurons in the synaptic cleft.
It's probably not the size across the cleft but fact that the membrane resists the flow of charge.
Instead, the stimulus of the action potential causes a rush of calcium ions into the synaptic buton, rapidly changing the potential inside...This change in potential is going to affect little vesicles, little blobs of membrane inside the presynaptic neuron.
As far as I know, it is not the change in potential that affects the vesicles (besides, the action potential alone is a huge change in potential, so there'd be no need for calcium-carried potential) but the presence of increased calcium levels itself. This is because the calcium ions bind to proteins that guide the vesicles to the membrane, etc.
The neurotransmitters are not taken up by the receptors. Instead, they bind, and the receptor, which runs through the membrane in the postsynaptic cell, changes conformation on the inside of the cell, causing activation of pathways.
Well, sometimes it causes activation of pathways. Sometimes it just opens an ion channel.
I'm going to skip over your excellent explanation to say that the illustrations are wonderful, especially the Pac-Man enzymes at the end.
cm: excellent points. I left most of them out in an effort to get this across as simply as possible, but they're really good for those who want more detail, but yeah, I left out the ion channel. Oops. I also left out things like VMAT and how vesicles bind to the membrane, which of course only makes the bind boggle MORE when you learn about all the complexity.
Mr Ian: cite your source, and steal away.
Neurons are totally awesome, and excitability rocks the house too. Especially the action potential part. Synaptic tranmission is all well and good, but let's remember it starts with an action potential. Not that I'm biased or anything.
just to comment on CM's post. The calcium doesn't guide the vesicles to the membrane, rather it causes conformational changes in a particular synaptic vesicle protein (the calcium sensor which is thought to be synaptotagmin) which allows for vesicle fusion and release of contents. But this I know is far too detailed for the general audience.
great post..
reading this reminded me of a professor in grad school that was very very particular about what you call a neurotransmitter.
Her opinion, was that if it didn't gate a channel directly, then it wasn't a neurotransmitter.
Nat: I've got an awe-inspiring post on the action potential, if I do say so myself.
neuropostdoc: yeah, bit too detailed. But awesome. I could spend ages on all the little conformational changes...
Pinus: when you say "gate a channel directly", do you mean open an ion channel? That seems like a bit of a limiting definition. Most neurotransmitters DO have activity at an ion channel or two, but often most of the receptors hit are G-protein coupled (consider, for example, serotonin, where the only ion channel receptor is the 5-HT3). And what about dopamine, where the all receptors are G-protein coupled, but that activation could OPEN ion channels? Did she have another term for chemicals released across the synapse to hit receptors on the other side, if those receptors are not ion channels? And was she basing her distinction on findings in the lit on chemical distinction? This is the first time I've heard this idea, please let me know!
yes, I mean ion channel.
so basically...GABA, Glycine, Glutamate, Acetylcholine and 5HT (at 5HT3).
everything else would be a neuromodulator.
I have used the term neurotransmitter loosely in manuscripts before and had reviewers tell me that I should replace neurotransmitter with neuromodulator.
her distinction was based on the fact that only chemicals that directly gate ion channels...everything else modifies excitability or responsiveness...thus they were modulators.
pinus: *headdesk* now that you explain it, I HAVE heard that before. I guess it's more specific, but it seems to me that it's almost a question of semantics. Those that directly gate ion channels would just be directly modulating the excitability (by causing excitation), rather than indirect modulation through making the area less or more responsive. Is there concern that people are formulating ideas improperly because of lack of clarity? I could see that happen with grad students who haven't had the receptor ideas drilled into their heads yet.
Nice post!
Depending on the neurotransmitter you're dealing with ... there are carious things that can happen... the presynaptic neuron can have transporters
Hey, let's not forget our glial friends! They mop up both GABA and glutamate (the MAJOR CNS neurotransmitters -- heresy I know) from the synapse.
Where does all this 'excitation-secretion coupling' leave neurotransmitters like adenosine, neuropeptides, endocannabinoids, etc.?
The neurotransmitter/neuromodulator distinction seems like bunk to me.
My definition is that a neurotransmitter is something released from a neurons that transmits information to a separate cell. I might further limit the information to electrical excitability, rather than other intracellular signalling events.
Now some of those transmitters will have rapid effects on excitability (via ionotropic receptors) and some will have slower effects (via metabotropic receptors).
I can kinda get the idea of neurotransmitter as that thing mediating a fast effect, whereas a neuromodulator mediates a slower effect. What about in the middle though, where metabotropic responses can be rapid? Is it still a modulator is the response is complete within 1 second? In 250 ms?
It's not clear to me what purpose the distinction serves.
Oh yeah, I liked your action potential post too (IIRC my first comment on Neurotopia was there).
I am not saying I buy in to it, just wanted to point out that there are people who think this way, and they review papers and grants.
What is mind boggling is that a pre synaptic neuron looks like a penis, and a post synaptic neuron looks like a vagina.
Or maybe they look like that because it the best template for that kind of job.
Nat: kisses up against the presynaptic neuron, and the whole thing is surrounded by glia, which can ALSO take stuff up and break stuff down. At least, it might look like that...
pinus: a fair point.
Tex: Excellent point about the glia, especially with GABA and glutamate. I tend to speak from the neurotransmitters I know the most about, and forget stuff like that.
And speaking of stuff I don't know a lot about: adenosine, neuropeptides, and cannabinoids would be among them. I shall have to do some more reading, I think.
This is very nice. Reminiscent of a lecture I had once regarding ion exchanges in blood. This, as everyone else has pointed out, is very, very well done.
Nice post, Sci. cm's clarifications are very on point, as is the very important distinction between ligand-gated ion channel neurotransmitter receptors--such as those for glutamate, GABA, glycine, acetylcholine, and ATP--and the metabotropic receptors, which are all seven-transmembrane-domain G protein-coupled receptors (as are just about all neuropeptide receptors).
neuropostdoc said:
just to comment on CM's post. The calcium doesn't guide the vesicles to the membrane, rather it causes conformational changes in a particular synaptic vesicle protein (the calcium sensor which is thought to be synaptotagmin) which allows for vesicle fusion and release of contents. But this I know is far too detailed for the general audience.
But I didn't write that "calcium guides the vesicles in the membrane." That would be wrong twice: calcium can guide nothing, and what would it mean for vesicles to be guided *in* the membrane?
What I wrote was, "calcium ions bind to proteins that guide the vesicles to the membrane, etc." Perhaps "guide" and "etc." was too sloppy of a shorthand form of "controls the elaborate process of docking, priming, fusion pore opening, clathrin (un)coating, etc.", but I didn't want to go into it that much. But sure, I gave credit to the synaptic proteins as doing the "vesicle management" work.
Great post, Scicurious. I wish I could create posts like that.
Couple comments:
I've never heard of neuromodulators; IIRC there are G protein-coupled receptors for GABA, glycine, and glutamate, (so says Wiki re GABA and glutamate, IIRC the G protein-coupled GABA receptor also responds to glycine although I couldn't find ref's in a quick search). This sort of breaks down the distinction between neurotransmitters and neuromodulators.
The NMDA receptor responds to glutamate but can be "modulated" by voltage, Hallucinogens, Zn2+, pH, Glycine, and Polyamines. This means that "neuromodulators" can affect at least one ion channel (and likely more) as well as 2nd messenger receptors.
I would break things down into neurotransmitters, which are acting across the synaptic cleft, neurohormones, which are released from parts of the axonal arbor (especially the presynaptic areas) in response to action potentials and affect receptors within a short (~1mm) diffusion distance, and hormones, which circulate in the bloodstream. Many chemicals can act as two or all three (e.g. noradrenaline). Since all types can act as modulators, and modulation can take place at almost neurotransmitter speed (when it affects an ion channel such as the NMDA receptor), I would leave them out of the classification and define them entirely functionally.
The issue with glia is very important. When it comes to purely chemical communications, glia such as astrocytes have no disadvantage compared to interneurons, and there's no reason to assume they don't participate in the slower calculations.
I've seen mention of adenosine and ATP receptors, and participation in the communication process by astrocytes, such as , "Regulation of cell-to-cell communication mediated by astrocytic ATP in the CNS" (DOI:10.1007/s11302-005-6321-y, open access, I won't give more than one link because my post will get stalled in the moderation queue, so I'm just giving the DOI here). I've been meaning to write a post on the whole subject, but it's going to be a while, and anyway yours make much better introductions for people not already somewhat familiar with the subject.
BTW, I've run across ref's saying that the same neuron can release both GABA and glycine, and many types of neuron are thought to release both a "fast" neurotransmitter and a slower peptide "modulator", presumably with varying ratios depending on recent activation history (and perhaps many other things).
Tex said:
Where does all this 'excitation-secretion coupling' leave neurotransmitters like adenosine, neuropeptides, endocannabinoids, etc.?
It leaves them in the rather large bin of "stuff in neuroscience that doesn't quite fit the standard conception of how neurons do things". Well, endocannabinoids for sure. Neuropeptides can, AFAIK, be released in the traditional way, due to influx of calcium, just with somewhat different requirements for the levels and dynamics of that calcium (they are also released from a different kind of vesicle called a large dense core vesicle).
Endocannabinoids are actually not THAT different, in a sense. They are also thought to be released due to excitation and rising calcium, just in the postsynaptic cell. And they are not thought to be secreted, but more like oozing out through the membrane. And of course, they are going in the "wrong" direction! They are a fun class of neurotransmitters for sure.
I can't recall the state of knowledge on how adenosine is released or just builds up as ATP is cleaved or both--do you happen to know?
Here's a good link:
Astrocyte Control of Synaptic Transmission and Neurovascular Coupling
Please let me know if you don't like me posting this sort of thing, I'm hoping you'll think they add value.
AK: knock yourself out! I'm learnin' stuff! I would say, though, that the neurotransmitter vs neuromodulator would indeed be a distinction based on the result from their binding at a receptor (ion channel vs metabotropic). This makes sense to me and I can see where people would make the distinction. I could also see making the point that a neurotransmitter acts across the synaptic cleft, regardless of what it ends up binding to. But I think the majority is going to end up making the decision for us.
hah! endocannabinoids are BACKASSWARDS, no less!
OK, here's the next: Astrocytes, from brain glue to communication elements: the revolution continues.
I'd agree that "neurotransmitter vs neuromodulator would indeed be a distinction" the problem being that many (perhaps all) "classic" neurotransmitters also act on metabotropic receptors. Thus the distinction is functional rather than dependent on molecular identity. Another problem is that there are many chemicals that "modulate" the action of ionotropic receptors.
Anyway, I usually post these sort of links to a draft post on my blog, where they sit until I can get around to finishing it, often weeks. This way other people will get the advantage of my search without having to wait until I finish my post (if they even read my blog). And I'll be able to come back an get them when I'm ready to finish the post (if ever). So I'll keep posting them here (one by one) until my posts start going to the moderation queue.
This paper is so full of exciting implications that I'm not going to try to list them: NMDA Receptors Mediate Neuron-to-Glia Signaling in Mouse Cortical Astrocytes:
This one doesn't say anything about ATP or glia, but it says some important things about synaptic specificity: Synapse-Specific Expression of Functional Presynaptic NMDA Receptors in Rat Somatosensory Cortex (It's also (barely) recent enough for Research Blogging.):
I can only get the summary, so far: GUPEA: The astroglial syncytium:
P2X1 and P2X5 Subunits Form the Functional P2X Receptor (open access):
ATP not only helps control calculations, but axon growth: Inhibition of the ATP-gated P2X7 receptor promotes axonal growth and branching in cultured hippocampal neurons (open access):
Glutamate-induced Exocytosis of Glutamate from Astrocytes (open acess):
Posts like this are why I read here. Well that, and Friday Wierd Sex,..um, Science.
Seriously, thank you.
Fast Subplasma Membrane Ca2+ Transients Control Exo-Endocytosis of Synaptic-Like Microvesicles in Astrocytes (open access):
Substantia nigra osmoregulation: taurine and ATP involvement (open access):
Tonic activation of NMDA receptors by ambient glutamate of non-synaptic origin in the rat hippocampus (open access):
The presence of extra-synaptic NMDARs implies the presence of extra-synaptic currents subject to modulation by anything capable of modifying the general extracellular environment. Since NMDA receptors are subject to modulation by a wide variety of neurohormones, as well as internally generated modulators, this implies that the dendritic tubes are non-passive cables of extreme (potential) intelligence. (A subject I've discussed on my blog.)
Evidently, this intelligence gets input from astrocytes as well as other neurons.
Regulation of Synaptic Transmission by Ambient Extracellular (open access via PMC):
Glutamate Transporters Regulate Extrasynaptic NMDA Receptor Modulation of Kv2.1 Potassium Channels (open access):
More grist for the mill of intelligence in the dendritic membrane.
Potassium Channel Phosphorylation in Excitable Cells: Providing Dynamic Functional Variability to a Diverse Family of Ion Channels (open access):
Dendritic Excitability and Synaptic Plasticity (open access):
Spine Neck Plasticity Controls Postsynaptic Calcium Signals through Electrical Compartmentalization (open access):
Timing and Location of Synaptic Inputs Determine Modes of Subthreshold Integration in Striatal Medium Spiny Neurons (open access, I think: I got in):
Differential Excitability and Modulation of Striatal Medium Spiny Neuron Dendrites (open access):
(You can tell I'm running out of steam: too lazy to do the subscripts and superscripts.)
G-Protein-Coupled Receptor Modulation of Striatal CaV1.3 L-Type Ca2+ Channels Is Dependent on a Shank-Binding Domain (open access, I think, I got in):
Modulation of Voltage- and Ca2+-dependent Gating of CaV1.3 L-type Calcium Channels by Alternative Splicing of a C-terminal Regulatory Domain (Open access):
This is particularly important given that neurons ship RNA into the dendrites for translation, and may be able to distinguish which RNA goes where.
That's it for me tonight. Thanks for letting me use your blog as a notepad, Sci.
Wow there's a lot of comments!
Yay for drawing! My mind is thoroughly boggled! I actually just now read this cool article that is a bit to do with neurotransmission that boggled my mind before, so it's now extra boggled. (If I just failed at html coding, google Your Brain Is a Mess, but It Knows How to Make Fixes by Carl Zimmer)
The modulation also was disrupted by application of peptides targeting the Shank interaction with Homer.
The whole transmitter/modulator/etc stuff IS just semantics. Molecules don't "have the property" of being any of these things, the words are just shorthand for either their most frequently or first described effects. Biologists are terrible about operationally defining a molecule's function and then thinking that's what the molecule somehow "is"--a morphogen, a hormone, a neurotransmitter. This creates a false sense of novelty when we find, for example, a "morphogen" acting as an "axon guidance cue." It's fine when it's jsut used to organize information, but very often it limits the way people think.
Don't get me started on the presumed agency of DNA and "the gene for X."
Anyway, lots of peer reviewers' funerals to go before we purge ourselves of the semantics of early molbio.
A very informative and excellent article. Neurons form networks through which nerve impulses travel. Each neuron receives as many as 15,000 connections from other neurons. Neurons do not touch each other; they have contact points called synapses.