If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

Self vs. non-self immunity

The immune system's main job is to fight off harmful invaders like bacteria and viruses. This is achieved through a two-step process. First, B cells and T cells that react to our own proteins are eliminated during their development. Second, even if some of these cells escape, they need a matching T cell to activate them. This system isn't perfect, and when it fails, it can lead to autoimmune diseases. Created by Patrick van Nieuwenhuizen.

Want to join the conversation?

  • leafers tree style avatar for user Roger Gerard
    During immuno-suppressive illnesses like HIV how do the innate and adaptive cells of the immune systems lose their functions?
    (18 votes)
    Default Khan Academy avatar avatar for user
    • blobby green style avatar for user Scott Becker
      From Wikipedia: "HIV infects vital cells in the human immune system such as helper T cells (specifically CD4+ T cells), macrophages, and dendritic cells.[4] HIV infection leads to low levels of CD4+ T cells through a number of mechanisms, including apoptosis of uninfected bystander cells,[5] direct viral killing of infected cells, and killing of infected CD4+ T cells by CD8 cytotoxic lymphocytes that recognize infected cells.[6] When CD4+ T cell numbers decline below a critical level, cell-mediated immunity is lost, and the body becomes progressively more susceptible to opportunistic infections."
      (13 votes)
  • hopper cool style avatar for user ☣Ƹ̵̡Ӝ̵̨̄Ʒ☢ Ŧeaçheя  Simρsoɳ ☢Ƹ̵̡Ӝ̵̨̄Ʒ☣
    @ 8::07 He says they begin to differentiate, but doesn't explain the process. I'm fuzzy on cellular differentiation, how is it that the cells know when to differentiate, and what to differentiate into? Is the process of cellular differentiation fully understood?
    (8 votes)
    Default Khan Academy avatar avatar for user
    • blobby green style avatar for user jake.a.ruddy
      I can't answer your question fully, but I have studied some papers on stem cell differentiation. People have found that stem cells differentiate based on the concentrations of different chemicals and ligands they are exposed to, and based on the properties of the surface the cells are growing on (stem cells grown on stiffer surfaces were found to become more bone-like, while stem cells grown on more compliant surfaces were found to be more like fat cells).
      (18 votes)
  • mr pink red style avatar for user Derek Jones
    Around you mention that if there is some bacteria in the bone marrow it is killed or goes away? How does it get killed if the immature b-cells binding to it are killed? Therefore, can mature b-cells (or others t-cell, dendritic etc) go into the bone marrow to fight the infection? If not, can't the infection/virus/cancer take a hold?

    Also at you say that the pathogen is broken up and presented on an MHC2 molecule, how long do these MHC2 molecules + antigen last if they do not bump into it?

    Thank you :)
    (14 votes)
    Default Khan Academy avatar avatar for user
    • aqualine ultimate style avatar for user Abdul Ekiyoyo
      The MHC2 molecules should last a long time, I haven't seen anything to suggest otherwise but if T cells don't see a MHC molecule, they will undergo apoptosis (just like how if B cells don't encounter antigens). And for your first question, he assumed that the bacteria in the bone marrow will only exist for a short period of time, I am not sure why that is. But he also said that hopefully you have additional B cells outside of the bone marrows that can find the bacteria. This will be the case if you have a vaccine for a disease and you have made multiple memory B cells in advance.
      (5 votes)
  • blobby green style avatar for user Abbie Biggers
    How are antibodies not formed against the various flora in the gut? These are not a part of our "self", but the immune system does not attack the symbiotic bacteria.
    (11 votes)
    Default Khan Academy avatar avatar for user
    • leaf green style avatar for user AERO9977
      They do attack, but the immune system can't wipe out the good bacteria fast enough before the bacteria replicate. So this equilibrium is created where the immune systems attack but the good bacteria still lives. The reason the immune system can't fully wipe out the good bacteria is because of multiple factors.
      (4 votes)
  • blobby green style avatar for user Aram Durgerian
    Is the process of vetting B and T cells in the bone marrow and thymus referred to as negative selection?
    (5 votes)
    Default Khan Academy avatar avatar for user
  • orange juice squid orange style avatar for user Todd Mayer
    @ ish, he mentions that B-cells create antibodies which will go bind and kill "self" proteins, isn't this detrimental, and won't this result in more loss of "self" proteins? If memory cells are created as a result of the B/T-Cell "kiss", this will result in creation of memory B-Cells, which will then be trained to kill self-proteins in the future, can anyone expound on how this is of benefit? Am I misunderstanding? Would this lead to auto-immune disease?
    (5 votes)
    Default Khan Academy avatar avatar for user
    • leaf blue style avatar for user Thomas Lambert
      I believe that the point being illustrated is that if a b cell escaped with the 'self' presenting mhcII protein, it still needs the unlikely instance of a same coded cd4+ and cd8+ t cells to become an autoimmune disease process. This makes the 2 step antigen verification process vital to avoid such mistakes.
      (4 votes)
  • blobby green style avatar for user Sazy Plew
    Is it my naivety &/or lack of complete/complex understanding or it's too simple for someone with several years of training to consider...let's take celiacs disease, life threatening allergy to gluten (really any allergen)...couldn't you over time eliminate the allergic reaction by stimulating the body to respond to eliminate the unnecessary reactive molecule at the source; get the body to do what it should've done in the first place? I'm thinking maybe these molecules, for whatever reason, didn't make it to the bone marrow in the first place or a rouge responder accidentally got out and once responsive proliferated. I imagine doing this by directly injecting the antigen into the bone marrow and providing bosters until the pre-circulating responders have died off...effectively eliminating the allergic response. Or would an allergic response happen in the bone marrow...I guess I'm seeing the bone marrow as the training grounds. This would work for any allergy, say a nut allergy, and better, as it'd eventually be a permanent change, than the current method of small exposures over time to change one's IgG & IgE ratio which is temporary at best.

    Thank you Sal for starting up khan and more importantly making it free! Your site has been especially helpful the last couple of years for me on a personal level. Currently, academically, khan has been a help with my microbiology class at College of Marin as my instructor J.D. is constantly mixing up concepts; to name a few examples-- antigen w antibody w mhc receptors, difference btwn to deliver vs to contain pipette, and the AB model w toxins. Khan has been a tremendous aid given her misteachings makes it difficult, compounded with the fact my reading is still getting back to normal so I have to reread much of the text, my class notes are lacking, and along w working with other hardships. You clear up A LOT and make it simpleIs... THANK YOU! I wish I could do more than give a verbal thanks. I'm sure you help millions upon millions. You are quite possibly the best teacher ever. You should consider offering some sort of formal degree, certificate, or something to your users. YOUR teachings are better than the online classes I've taken through my local junior college. ( Could you ramp up your physics section though?). Words can't express enough thanks, but it's all I have at the moment-- Thank You Sal.
    (5 votes)
    Default Khan Academy avatar avatar for user
  • duskpin ultimate style avatar for user Trinety Davison
    Is the "kiss" a scientific word for that or are you just messing with me?
    (3 votes)
    Default Khan Academy avatar avatar for user
    • leaf green style avatar for user Joanne
      At and he uses the word kiss. He is just trying to say the cells have to touch. Some call it bind together, some say 'handshake'. He is using an analogy. This binding will then allow the activation of the B cell so it will start making antibodies.
      (3 votes)
  • blobby green style avatar for user M E
    The video mentions that stored within the bone marrow are proteins that the body uses to weed out immune cells that attack "self". What about B-Cells and T-Cells that react, or "connect", to other body cells, not just proteins? Surely we can't have one of every type of cell found in the body within the bone marrow.

    (e.g. In the case of Type I Diabetes where the immune system attacks insulin producing cells on the pancreas?)
    (4 votes)
    Default Khan Academy avatar avatar for user
  • purple pi purple style avatar for user brewbooks
    At it's said: "You only want that T-cell receptors to react to foreign things, to non-self things and not to self-things." When T-cells are inactivated, cancer's can occur more frequently. How do T-cells react to cancerous cells, which are self things?
    (1 vote)
    Default Khan Academy avatar avatar for user
    • leaf blue style avatar for user dysmnemonic
      As cancerous cells pick up more and more mutations, there are changes to more of their proteins. Because parts of the proteins can change, they can then present non-self peptides, so CD8+ T cells can attack those cells. The immune system isn't very good at attacking cancers, though, because they do mostly have self antigens on the cell surfaces.
      (6 votes)

Video transcript

Voiceover: The purpose of the immune system is to fight things. To fight bacteria and viruses and other things that you don't want in your body. That's what it does. The question we can ask is how does the immune system know not to attack your own body, and that might seem like a strange question or an obvious question but it's actually not obvious the answer to it. In this video we're gonna go into why it's not obvious and then how the body actually does prevent its immune system from attacking itself. Another way of saying is this how does the body distinguish self versus non-self? By that we mean how does it tell the difference between your own body, your own proteins, your own cells and foreign proteins or cells or things that shouldn't be in your body. How does it know to attack these and not to attack this. To find out why the question itself is not obvious let's go back to B cells and talk about B cells. Here we have a B cell and here is its nucleus with some DNA and the most important part of the B cell that we care most about is its B cell receptor which later if it becomes active can be released as an antibody. Now this B cell receptor is what's gonna bind to foreign, pieces of foreign bacteria or viruses and these antibodies are gonna bind to those things and help your body get rid of them. The important thing to remember about B cells is that these antibodies or these B cell receptors are coded in the DNA of the B cell, but that they're different for every B cell. Every B cell has a unique set of antibodies and B cell receptors that it makes. We'll give them slightly different colors to make that obvious. Here's one with slightly different DNA and a slightly different B cell receptor. The really critical point to remember is that these B cell receptors that will become antibodies are generated at random. Your body kind of shuffles the DNA here and creates a unique B cell receptor and antibody for each B cell. It's precisely that fact which is that they're created at random which means that your body is in danger of creating B cell receptors and antibodies that can react to your own body because while this guy might be good, while he might react to let's say a bacteria out here that you want to get rid of, while that's good. This guy might easily create a B cell receptor and later an antibody that can react to something you don't want it to react to. For example, let's say this is an important protein in your body maybe it's insulin. You would not want a B cell to be created which will react to insulin because then maybe it will start creating antibodies that bind all the insulin in your blood and if you don't know what insulin does don't worry about it but by binding to insulin it will prevent insulin from doing its function which is very important. How can you keep your body from making B cells that would react to yourself? In fact there's no way to do it because as I said, this process of creating different B cell receptors and antibodies is totally random. There's no way to keep your body from making B cell receptors or B cells that will react to yourself. What does that mean? That means that you're gonna make them but you need to find a way to figure out which ones are reacting to you and to get rid of them. You need to figure out a way to kill the ones you don't want. By the way we're talking about B cells here. B cells. Everything we're saying is equally applicable to T cells. Let me draw one here. T cells it's equally applicable because T cells also have a T cell receptor that's generated at random and you only want that T cell receptor to react to foreign things. To non-self things and not to self things. The processes we're gonna talk about are equally true, maybe even in some cases more so true for T cells and B cells. Let's go to the bone marrow to figure out how this process works, and we're going to the bone marrow because that's where B cells come from. It's where they get their unique antibody, your B cell receptor. They get that by changing their DNA a tiny little bit by shuffling pieces around. Let's look at a couple of these B cells which are still young. They haven't yet been allowed out of the bone marrow. They haven't been vetted to see if they should be allowed out. Each one has its unique receptor. Let's draw a few of those receptors here. Let's say that one of these guys reacts to self. One of them reacts to some protein in your own body that you don't want it to. Again, that just happened at random because you're creating these receptors really at random. Let's say that the guy that we're gonna want to get rid of this one because he reacts to let's say it's insulin again like up there. Really we should have drawn insulin in yellow to show that it goes with this receptor. Let's say this guy reacts to insulin. How can you figure out that this guy reacts to self? The answer is actually quite simple. The answer is that you just need to keep around the various proteins that your body uses. You need to keep them around in the bone marrow while these B cells are being vetted. For example here you'll have a little insulin. Very small amount but it will be there. You'll have a little bit of some other protein. Let's say maybe hemoglobin. You'll have some other protein here. You'll have yet another protein over here. All these proteins will be around. What your body does at this stage and development is it says whatever B cell binds to something, wherever B cell binds to something with its B cell receptor in the bone marrow will be killed. This B cell right here that recognizes this insulin protein the fact that it recognizes it, it means that it will bind and that will cause a little bit of a chemical change in the B cell or something, and one thing will lead to another and the whole system will be programmed so that as a result this guy will die. Every B cell that recognizes self if it sees that self molecule in the bone marrow it will be killed. This works because your bone marrow will have most of the abundant proteins in your body. They'll be present there so that you can make sure that you weed out all the B cells that react to self. Now what happens after this step is that these guys who have been vetted they can proceed onwards to maybe a lymph node. Somewhere where they can begin to actually be active now that they've sort of gone through basic training here in the bone marrow. You might ask yourself, well, what about here? What about here when one of these B cells that doesn't react to self, what about when it interacts with a bacteria that you actually want it to fight? Is the same thing gonna happen? Is it going to die just because it recognizes the molecule that its made to bind to? The answer is obviously not. You don't want this guy to die because you need him. Because you want to fight this bacteria. The reason why he doesn't die is because well, we're in a different environment. There are different rules, there are different other cells around and this B cell has matured and become different. The rules are different and he's not going to die. This weeding out of B cells that react to self proteins is sort of the first of two mechanisms that I'd like to talk about that the body uses to not react to self. Actually the same thing exact thing happens for T cells except it doesn't happen in the bone marrow, it happens in the thymus because that's where T cells mature. In the thymus we have really the identical process where T cells differentiate and each one has a unique receptor and the ones that reacts to self in the thymus too strongly are killed. It's not a foolproof method or else we wouldn't need step two. Every once in a while a B cell will get out there. A B cell will escape which reacts to self. It's just because every process has its mistakes and maybe you don't have every single protein here in the bone marrow and enough abundance to find the B cells that reacts to a protein of your own body. Let's say this is a B cell that escaped the bone marrow even though it reacts to self. What's gonna happen now? It's gonna find that protein that it was sort of made at random to react with. It's gonna find that protein that your body makes and that your body needs and it's going to bind to it. What's it gonna do now? If you remember it's now going to take that protein ingest it, break it up into little pieces and then present it on an MHC II molecule. If you recall it will just present a small piece of that protein on the MHC II molecule. Maybe it will present a different piece of protein on a different MHC II molecule over here, something like that. The reason it does this is because it needs a T cell to come along. Here's a T cell. It needs a T cell to come along that will recognize that same piece that is put there on its surface. It needs that in order to activate it. It's going to sit there and wait for this T cell to come along who has the perfect receptor. Here's that T cell. They're going to interact and they're gonna have some kind of intracellular kiss that's going to finally allow this B cell to activate. Usually without the T cell coming and recognizing the antigen that the B cell reacts to the B cell cannot activate. It needs this T cell to recognize it. This is exactly the second mechanism of defense that I'd like to bring to your attention which is that even if a B cell escapes that reacts to self. Almost always it's also going to need a T cell that reacts to self to come and activate it. You need both the B cell to escape the weeding out in the bone marrow and the T cell to escape the weeding out in the thymus for you to get an active B cell that's now going to start putting out antibodies that react to self. By the way, this cellular kiss here is usually going on in the lymph node. Looking at this whole process you might have a few complaints. I encourage you think about how it might go wrong. One way you might think it could go wrong is what if bacteria got into the bone marrow and certainly that's very, very possible because when you get infections the thing that's infecting you can move around your body. If this bacterium gets into your bone marrow does that mean that now this B cell is going to bind to it and therefore it's going to be killed? Because at this stage whenever the B cells bind to something they're killed. The answer is yes. This is exactly what happens. The reason why it's not too much of a problem is that even if you have this bacterium here in the bone marrow for a week or two or maybe a month, after that once this bacterium goes away or it's killed then it won't be there anymore and you can start producing these B cells that react to that bacterium again. Hopefully you already had a bunch of these B cells that could react to this bacterium that you had made previously and that were already out in the lymph nodes. Those guys will be there to fight the infection while maybe the infection might be in your bone marrow preventing you from making more of those B cells to kill it. You already have some of those B cells out there in the lymph nodes and they can proliferate out there and sort of lead the battle from there. Now even though your body has these mechanisms to keep your immune system from reacting to yourself it still happens sometimes. The process still goes wrong sometimes and the result is autoimmune disease. It's called autoimmune because you're immune to yourself. Your immune system basically starts attacking your own body and some pretty terrible disease can result. To kind of bring it to life where I'd like to tell you about one example of this. This is a muscle fiber and the way your muscle fibers are activated because you don't want to be flexing all of your muscles all the time the way that they're activated is that they have a little receptor which I'll draw here. This receptor is ready to receive little molecules from a neuron, part of a nerve. It receives little molecules from this neuron that activate this receptor and therefore activate the muscle fiber. If you want to tighten this muscle fiber you just need to send the signal down a neuron and they'll release the little molecules which will activate the muscle fiber. In one example of autoimmune disease you get antibodies against this receptor here on the muscle. They bind to it and that either stops it from functioning. It makes it impossible for it to react to the neuron signals or causes that receptor to be destroyed. Those are two mechanisms that have been seen. In this autoimmune disease what do you think will happen? What happens is that your body can no longer activate muscle fibers as easily. The disease is called myasthenia gravis and the etymology of that is "my" it means muscle and "asthenia" means weakness. Muscle weakness and gravis just means it's serious because it gets serious over time. If you can't activate the muscle fibers in your body you slowly become paralyzed. You don't need to remember this exact mechanism. That's not really important but I just wanted to give an example of one kind of autoimmune disease and how it works.