One of the big (and so far unanswered) questions about the coronavirus epidemic is what kind of immunity people have after becoming infected. This is important for the idea of “re-infection” (is it even possible?) and of course for vaccine development. We’re getting more and more information in this area, though, and this new paper is a good example. A team from the La Jolla Institute for Immunology, UNC, UCSD, and Mt. Sinai (NY) reports details about the T cells of people who have recovered from the virus. To get into this, a quick explainer seems appropriate, so the next bit will be on the background of T cells and adaptive immunity – then we’ll get into these latest results.
So everyone’s heard of the broad category of white blood cells. One group of those are the lymphocytes (literally “lymph cells”, where they’re most easily found), and the lymphocytes include T cells, B cells, and NK cells. You’re looking at three big branches of the immune system right there. The NK (“natural killer”) cells are part of the innate immunity, the nonspecific kind, and they’re in the cell-mediated cytotoxic wing of that. The other side of the immune system is adaptive immunity. The B cells feature in my antibody background posts, because as part of the adaptive system they’re the ones that produce more of some specific antibody once one of the zillions of them present in the body turns out to fit onto a new antigen. The T cells are in the adaptive side as well, but they’re in the cell-mediated part of that army.
T cells come from the thymus (thus the “T”), so if you’ve been wondering what your thymus has done for you lately, that’s one good answer. They all have a particular surface protein, the T cell receptor. Similar to the way that the immune system generates a huge number of antibodies by shuffling and mixing protein expression, there are a huge number of different T cell receptors waiting to recognize what antigens may come along. The precursors of T cells come from the bone marrow and migrate to the thymus, where they branch out into different lines (and that branching out continues even once they leave the thymus and begin circulating in the lymph and in the blood).
The most direct of those are the cytotoxic T cells, also known as CD8+ T cells and by several other names. CD8 is another particular cell-surface protein that distinguishes this type. These cells aren’t going after viral particles; they’re going after the body’s own virus-infected cells and killing them off before they can break open and spread more viral particles. They’ll kill off bacterial cells in the same way. These are also the ones that the CAR-T therapies are trying to mobilize so that they’ll recognize cancer cells and do the same thing to them. How do they accomplish the deed? They’re thorough; there are several deadly mechanisms that kick in. One general one is to secrete cytokines, especially TNF-alpha and interferon-gamma, that alert other cellular systems to the fact that they’ve detected targets to attack. (The monoclonal antibody drugs for arthritis are actually aimed to shut down that TNF-alpha pathway, because in RA the T cells are – very inappropriately – attacking the body’s own joint tissue). A second CD8+ action is to release “cytotoxic granules”. These are payloads of destruction aimed at the target cell once the T cell is closely connected to it (the “immune synapse”). You need that proximity because cytotoxic granules are bad news – they contain proteins that open up pores in the target cell, and blunderbuss serine protease enzymes that slide in through them, whereupon they start vigorously cleaving intracellular proteins and causing general chaos (and eventually cell death). And the third killing mode is via another cell-surface protein the CD8+ cells have called FasL – it binds to a common protein on the target cells called Fas, and that sets off a signaling cascade inside the target cells that also leads to cell death. (Interestingly, the CD8+ cells use this system after an infection has subsided to kill each other off and get their levels back down to normal!)
And then there’s another crowd, the CD4+ T cells, also known as T-helper cells and by other names. They work with another class of immune cells, the antigen-presenting cells, which go around taking in all sorts of foreign proteins and presenting them on their cell surfaces. A CD4+ cell, when it encounters one of those, goes through a two-stage activation process kicks in (the second stage is sort of a verification check to make sure that it’s really a foreign antigen and not something already present in the body). If that’s successful, they start to proliferate. And you’re going to hate me for saying this, but that’s where things get complicated. Immunology! The helper T cells have a list of immune functions as long as your leg, interacting with many other cell types. Among other things, they help set off proliferation of the CD8+ cells just detailed, they activate B cells to start producing specific antibodies, and they’re involved with secretion of more cytokine signaling molecules than I can even stand to list here. These are in fact the cells targeted by HIV, and it’s the loss of such crucial players in the immune response that makes that disease so devastating.
OK, there’s some background for this new paper. What it’s looking at in detail are the virus-specific CD8+ and CD4+ cells that have been raised up in response to the infection in recovering patients. As you’ve seen, both of these subtypes are adaptive; they’re recognizing particular antigens and responding to those – so how robust was this response, and what coronavirus antigens set things off? You can see how important these details are – depending on what happens, you could have an infection that doesn’t set off enough of a response to leave behind B and T cells that will remember what happened, leaving people vulnerable to re-infection. Or you could set off too huge a response – all those cytokines in the “cytokine storm” that you hear about? CD4+ cells are right in the middle of that, and I’ve already mentioned the TNF-alpha problems that are a sign of misaligned CD8+ response. The current coronavirus is pretty good at evading the innate immune system, unfortunately, so the adaptive immune system is under more pressure to deliver. And one reason (among many) that the disease is more severe in elderly patients is that the number of those antigen-presenting cells decline with age, so one of the key early steps of that response gets muted. That can lead to a too-late too-heavy T cell response when things finally do get going, which is your cytokine storm, etc. In between the extremes is what you want: a robust response that clears the virus, remembers what happened for later, and doesn’t go on to attack the body’s own tissues in the process.
Comparing infected patients with those who have not been exposed to the coronavirus, this team went through the list of 25 viral proteins that it produces. In the CD4+ cells, the Spike protein, the M protein, and the N protein stood out: 100% of the exposed patients had CD4+ cells that responded to all three of these. There were also significant CD4+ responses to other viral proteins: nsp3, nsp4, ORF3s, ORF7a, nsp12 and ORF8. The conclusion is that a vaccine that uses Spike protein epitopes should be sufficient for a good immune response, but that there are other possibilities as well – specifically, adding in M and N protein epitopes might do an even more thorough job of making a vaccine mimic a real coronavirus infection to train the immune system.
As for the CD8+ cells, the situation looked a bit different. The M protein and the Spike protein were both strong, with the N protein and two others (nsp6 and ORF3a) behind it. Those last three, though, were still about 50% of the response, when put together, so there was no one single dominant protein response. So if you’re looking for a good CD8+ response, adding in epitopes from one or more of those other proteins to the Spike epitope looks like a good plan – otherwise the response might be a bit narrow.
And here’s something to think about: in the unexposed patients, 40 to 60% had CD4+ cells that already respond to the new coronavirus. This doesn’t mean that people have already been exposed to it per se, of course – immune crossreactivity is very much a thing, and it would appear that many people have already raised a response to other antigens that could be partially protective against this new virus. What antigens those are, how protective this response is, and whether it helps to account for the different severity of the disease in various patients (and populations) are important questions that a lot of effort will be spent answering. As the paper notes, such cross-reactivity seems to have been a big factor in making the H1N1 flu epidemic less severe than had been initially feared – the population already had more of an immunological head start than thought.
So overall, this paper makes the prospects for a vaccine look good: there is indeed a robust response by the adaptive immune system, to several coronavirus proteins. And vaccine developers will want to think about adding in some of the other antigens mentioned in this paper, in addition to the Spike antigens that have been the focus thus far. It seems fair to say, though, that the first wave of vaccines will likely be Spike-o-centric, and later vaccines might have these other antigens included in the mix. But it also seems that Spike-protein-targeted vaccines should be pretty effective, so that’s good. The other good news is that this team looked for the signs of an antibody-dependent-enhancement response, which would be bad news, and did not find evidence of it in the recovering patients (I didn’t go into these details, but wanted to mention that finding, which is quite reassuring). And it also looks like the prospects for (reasonably) lasting immunity after infection (or after vaccination) are good. This, from what I can see, is just the sort of response that you’d want to see for that to be the case. Clinical data will be the real decider on that, but there’s no reason so far to think that a person won’t have such immunity if they fit this profile.
Onward from here, then – there will be more studies like this coming, but this is a good, solid look into the human immunology of this outbreak. And so far, so good.