Mystery Rays from Outer Space

Meddling with things mankind is not meant to understand. Also, pictures of my kids

October 30th, 2007

Rube Goldberg and hypersensitivity: Frame-shifting, part II

Rube Goldberg machineAntigen processing is not only interesting and important in itself,1 but it’s been used extensively to tease apart fundamental cell biology — things like protein folding, intracellular proteolysis, protein trafficking, and ER-associated degradation have been identified or studied via antigen processing. There are a bunch of reasons why MHC has been such a Swiss army knife of cell biology. One of the reasons is that MHC can amplify a tiny, tiny signal into a blatant, unmistakable readout.

That’s because cytotoxic T lymphocytes recognize MHC/peptide combinations, recognize it incredibly well, and respond with easily-observed events. CTL can recognize as few as 10 (maybe fewer) specific peptides per cell, even though for every one of those peptide/MHC complexes there are ten thousand other complexes with other peptides, smothering it. And CTL respond by destroying the cell, which gives you a simple, black-and-white, binary outcome.

It’s obviously useful to have a highly sensitive2 readout. But it’s a curse as well as a blessing. In particular, because the outcome is binary (alive or dead) it’s really hard to get quantitative information out of the system. Once you’re over the very low threshold, everything is positive.

What this means is that detecting something with a CTL readout doesn’t tell you if that something is common, unusual, rare, or sui generis. CTL readouts over the years have demonstrated the existence of events that (I believe) are really very unusual — they aren’t representative of “normal” cell biologic processes, but rather represent the far end of the curve, things that, yeah, can happen, but have to be pushed. For example, there’s proteasome splicing : biochemically a really cool phenomenon, that got picked up by CTL readouts — but it’s really not likely that it happens very often, or is a real player in the normal function of the cell.3

On the other hand, of course, some things that have turned out to be common and important processes were identified up in the same ultrasensitive way. For example, exactly this sort of thing turned out to be a very early demonstration of ER-associated degradation,4 which is now known to be a major and critical pathway.

HIV-1 frameshift inducing element
HIV-1 frameshift inducing element

So — following on from my post earlier this week — the immediate question that comes to mind when Nilabh Shastri’s group publishes about frame-shifted epitopes5 is whether this is a major, common phenomenon, or it is the end-product of a Rube Goldbergesqe sequence of events that isn’t going to happen very often?

Shastri has long been a fan of the idea that frame-shifting — reading proteins from abnormal start sites, or by hiccups during translation — could be a common source of antigenic peptides (epitopes). In his latest paper, he demonstrates a frame-shift epitope from HIV; he and some other groups have demonstrated frame-shift epitopes before, but those were mostly fairly minor, and were easy to ignore. This example seems to be a relatively potent epitope, and is harder to ignore. Are frame-shifts common in the cell? Are they common sources of CTL epitopes?

ResearchBlogging.orgInterestingly, they identified the epitope by bioinformatic analysis of a known frame-shift product. (In other cases, the identification went the other way around, from identifying the epitope sequence to the frame-shifted precursor.) This raises one point: If frame-shifted proteins really are common sources of CTL epitopes, then for one thing the task of the bioinformatician is six times harder, because they will have to survey all six reading frames, not just the known proteins, of a viral genome, to look for epitopes. But (for all the criticism I’ve leveled at epitope prediction software) that doesn’t seem to be a major factor; predictions do find epitopes (however inefficiently) and they find them in true proteins.

In the small handful of cases where a full CTL response to a virus has been analyzed fairly completely — that is, where almost all the epitopes recognized the CTL have been identified — they almost all have been identifiably from authentic viral proteins.6 That said, there are some that haven’t been identified yet; for example, in mouse cytomegalovirus a number of epitopes remain unmapped,7 and might be from frame-shifted precursors.

Proponents of the unconventional precursors argue that many MHC-associated peptides (identified by mass spectrometry, for example) don’t have an authentic protein precursor in the various databases. I think that’s true, but far more are identifiable (I don’t know the ratio of identifiable to unidentifiable, though), and most of the anonymous ones probably represent, say, un-sequenced alleles or something like that.

Overall, I think the bulk of the findings from epitope identification really argue that things like frame-shifted epitopes, or proteasome-spliced epitopes, or non-ATG-initiated epitopes — things that we think should be rare, based on what we know about cell biology — really are rare. The fact that they do appear and can be captured by this exquisitely sensitive8 system, probably goes to show that there is more slop in the system than is often believed — more aberrant, defective products sneak through into RNA and protein than is really appreciated, and in all likelihood error correction is just as important as error prevention in normal cell function.


  1. And people who research antigen processing are invariably suave, attractive, and charming. Well-known fact![]
  2. I’m desperately trying to avoid saying “exquisitely sensitive” here, because every paper and review on the subject calls it “exquisitely sensitive”[]
  3. I reserve the right to deny I ever said this, if proteasome splicing ever turns out to be important.[]
  4. Skipper, J. C., Hendrickson, R. C., Gulden, P. H., Brichard, V., Van Pel, A., Chen, Y., Shabanowitz, J., Wolfel, T., Slingluff, C. L., Jr., Boon, T., Hunt, D. F., and Engelhard, V. H. (1996). An HLA-A2-restricted tyrosinase antigen on melanoma cells results from posttranslational modification and suggests a novel pathway for processing of membrane proteins. J. Exp. Med. 183, 527-534.[]
  5. Maness, N. J., Valentine, L. E., May, G. E., Reed, J., Piaskowski, S. M., Soma, T., Furlott, J., Rakasz, E. G., Friedrich, T. C., Price, D. A., Gostick, E., Hughes, A. L., Sidney, J., Sette, A., Wilson, N. A., and Watkins, D. I. (2007). AIDS virus specific CD8+ T lymphocytes against an immunodominant cryptic epitope select for viral escape. J Exp Med 204:2505-2512 []
  6. Kotturi, M. F., Peters, B., Buendia-Laysa, F. J., Sidney, J., Oseroff, C., Botten, J., Grey, H., Buchmeier, M. J., and Sette, A. (2007). The CD8+ T-cell response to lymphocytic choriomeningitis virus involves the L antigen: uncovering new tricks for an old virus. J Virol 81, 4928-4940. []
  7. Munks, M. W., Gold, M. C., Zajac, A. L., Doom, C. M., Morello, C. S., Spector, D. H., and Hill, A. B. (2006). Genome-wide analysis reveals a highly diverse CD8 T cell response to murine cytomegalovirus. J Immunol 176, 3760-3766.[]
  8. Couldn’t keep it up[]
October 29th, 2007

RNA, protein, and information

ENCODE logo Not long ago there was some keruffle over the ENCODE data,1 and the unrelated but almost simultaneous Cell paper,2 that demonstrated widespread transcription even from apparently-inactive genes — for example, this discussion and this one at Ars Technica’s Nobel Intent , and this one at Sandwalk. The observations were considered surprising because RNA was (and is) usually considered to be fairly tightly regulated. The ENCODE data, in particular, were used as arguments pro and con “junk” RNA — non-functional transcription.

The presence of non-functional RNA, though, didn’t strike me as very surprising at all, and part of the reason for that was that I had been primed to think about efficiency in cellular processes by Jon Yewdell’s “DRiPs” hypothesis.3

Briefly (I want to talk about DRiPs in detail some other time) Yewdell suggests that peptides that are presented on MHC class I to cytotoxic T lymphocytes are usually derived, not from full-length, functional proteins, but from “defective ribosomal products” — proteins that began translation and got screwed up partway through, or that completed translation and failed to fold properly. Proteins, in other words, that were defective from the get-go, that never had a chance to contribute to the whole happy economy of the cell. This contrasts to the traditional view, that antigenic peptides are derived from proteins during their normal turnover, often over a period of many hours.

Pollock Untitled (Green Silver)
Pollock’s drips: “Untitled (Green Silver)”

Yewdell argued that in fact a large percentage of translation (he’s offered various percentages, but let’s say 30% of translation) ends up in this defective pool, and because it’s defective it’s destroyed very rapidly by the proteasome — again, he’s offered various numbers, but let’s say for the sake of argument that it’s destroyed within a handful of minutes.

I don’t mind saying that I was extremely skeptical when I read his initial paper, and I am still quite skeptical about the overall contribution; but over the years I have (reluctantly) come a long way to accepting the general principle. But — unlike many of the people who disagreed with the DRiP hypothesis — I didn’t find the principle of DRiPs per se implausible.

In fact, it was one of those things that I had never thought of, but that made immediate sense to me as soon as I read the idea. I think of it as an information theory thing: Preventing errors in translation must take a certain amount of energy; at some point the incremental energy needed to reduce the error rate from N to N-1 would be greater than the amount of energy needed to degrade a defective product. And as soon as you consider it as that equation, it becomes a slider, and the set-point could be almost anywhere. It’s quite plausible (to me, anyway) that the amount of energy used in error prevention is relatively high, whereas the energy loss in protein degradation is relatively low — and so it’s cheaper, energetically, to simply make error-riddled protein, and let the proteasome sort it out after the fact.

(I’m simplifying all the arguments here, pro and con. I’ll probably take them up in bits and pieces later on.)

Anyway, exactly the same reasoning applies to transcription. The amount of energy that it would take to clamp down and make absolutely perfect identification of proper transcriptional start sites, must at some point be greater than the amount of energy involved in destruction of aberrant RNA. So this is why I thought it was quite predictable that there would be widespread, low-level, transcription of non-functional RNA that would then run into the next level of information processing.

Blogging on Peer-Reviewed ResearchThe reason I thought about this today, months after the fuss has pretty much died down, is a paper from Nilabh Shastri’s group4 that demonstrates another instance of what, I suspect, is another example of aberrance that’s tolerated by cells. (There’s also a commentary on the paper,5 by Yewdell.) Today’s post was in fact supposed to be entirely about that paper, but what with all the time I’ve spent blathering about the background, I’ll finish up with a new post later this week. In any case it’s time for me to go read “Green Eggs and Ham” to my kids.

(I’m trying out including the BPR3 icon here. I’m not entirely convinced by the BPR3 rationale, but I’m willing to see what happens for a while, anyway.)


  1. Identification and analysis of functional elements in 1% of the human genome by the ENCODE pilot project. Nature 2007 447, 799-816. []
  2. Guenther, M. G., Levine, S. S., Boyer, L. A., Jaenisch, R., and Young, R. A. (2007). A chromatin landmark and transcription initiation at most promoters in human cells. Cell 130, 77-88.[]
  3. Yewdell, J. W., Aton, L. C., and Benink, J. R. (1996). Defective ribosomal products (DRiPs): A major source of antigenic peptides for MHC class I molecules? J. Immunol. 157, 1823-1826.[]
  4. Maness, N. J., Valentine, L. E., May, G. E., Reed, J., Piaskowski, S. M., Soma, T., Furlott, J., Rakasz, E. G., Friedrich, T. C., Price, D. A., Gostick, E., Hughes, A. L., Sidney, J., Sette, A., Wilson, N. A., and Watkins, D. I. (2007). AIDS virus specific CD8+ T lymphocytes against an immunodominant cryptic epitope select for viral escape. J Exp Med 204:2505-2512 []
  5. Yewdell, J. W., and Hickman, H. D. (2007). New lane in the information highway: alternative reading frame peptides elicit T cells with potent antiretrovirus activity. J Exp Med 204:2501-2504 []
October 25th, 2007

Invertebrate memory, or wishful thinking?

Copepod
Copepod

I’ve been fascinated by recent findings on evolution of immune systems. On the one hand, the roots of the classic vertebrate adaptive immune system have been pushed back further back in time (e.g. the identification of RAG in sea urchins); and on the other, invertebrate immune systems are looking much more complex than previously thought, through a completely separate path.

Traditionally, immunological memory has been considered to be a hallmark of the vertebrate immune system. “Memory” here means the ability to respond to a second pathogen exposure in a more rapid and more aggressive manner than on the first exposure. The example we’re all familiar with is vaccination; if your first exposure to, say, smallpox or measles antigen is via a relatively harmless form, then when you are exposed to the harmful, fully-virulent pathogen you develop a very rapid, if not instantaneous, immune response and you are protected from disease. Invertebrates are not, traditionally, believed to have this memory response.

Still, a half-dozen papers over the last five years have claimed to show immunologic memory in invertebrates. I think this started with a 2003 paper on viral infection in copepods,1 but there are also several examples in insects.2 I honestly haven’t examined these papers all that closely, leaving them on the long “Cool papers I will look at carefully when I have time” list. A typical conclusion is:

This work contradicts the paradigm that insect immune responses cannot adapt and will promote the search for similar responses overlooked in organisms with an adaptive immune response.3

DrosophilaA paper that just came out in BioEssays4 raises a cautionary note on these conclusions — though I’m not very comfortable with at least part of the message. They argue that all the observations that have been interpreted as evidence for immunological memory could just as plausibly have non-immunological explanations: Because the papers only measure end-points (like survival or reproductive success), they risk missing other explanations:

The failure to make assessment of immunological parameters is a consistent weakness in most papers purporting to demonstrate priming, memory or adaptivity in the invertebrate innate immune system.

They also observe that these recent claims fly5 in the face of a great deal of previous research:

The idea that invertebrates have a primitive form of adaptive immune is not new and was explored extensively from the 1960s onwards. … Despite the prevailing zeitgeist of that time, phenomenological observations and experimental analyses failed to produce consistent and convincing evidence for immune memory or specificity.

Both of those points make sense to me.6 The argument I’m not comfortable with is this one:

We argue that the case for adaptive immunity in invertebrates based only on such phenomena is weak and flawed, as it can only be upheld if supported by descriptions of the underlying mechanisms. (My emphasis. IY)

It’s an argument I’m somewhat sympathetic with. Science should be built on understanding, not description — that’s why “phenomenology” is an insult. On the other hand, one needs observations to build hypotheses; otherwise, where do you start? All we know today of vertebrate immunity is built on observations that were, at one point, simply phenomena without an understood mechanism. (To be fair, that line is from the abstract, and in the body of the paper they basically say the same thing as me: “We fully acknowledge that phenomenological investigations are a logical starting point.“)

As I say, I haven’t looked closely at the papers they’re criticizing, and I’m not very familiar with the older literature they mention on invertebrate immunity; but from what I have seen, Hauton and Smith are making reasonable points. My bias is that I want invertebrates to have some form of adaptive immunity just for the cool factor — but I am certainly much more skeptical than I was before reading this.


  1. Kurtz, J., and Franz, K. (2003). Innate defence: evidence for memory in invertebrate immunity. Nature 425, 37-38. []
  2. For example, Pham, L. N., Dionne, M. S., Shirasu-Hiza, M., and Schneider, D. S. (2007). A specific primed immune response in Drosophila is dependent on phagocytes. PLoS Pathog 3, e26. []
  3. Pham, L. N., Dionne, M. S., Shirasu-Hiza, M., and Schneider, D. S. (2007). A specific primed immune response in Drosophila is dependent on phagocytes. PLoS Pathog 3, e26. []
  4. Hauton, C., and Smith, V. J. (2007). Adaptive immunity in invertebrates: A straw house without a mechanistic foundation. Bioessays 29, 1138-1146. []
  5. Har![]
  6. At least, in principle — I haven’t read through all the literature closely enough to be confident they’re completely right[]
October 21st, 2007

Persistent viruses and regulatory T cells

Deer mouse Some viruses can persistently infect their host indefinitely, without doing any particular damage. On the other hand, the same virus can infect a different host species and cause devastating harm. Why?

The obvious answer is that the virus has become adapted to the first host, but not the second. Leaving aside the question of whether that actually means anything helpful, I’ve previously pointed out that it’s actually not all that common for a virus to evolve toward reduced virulence, and probably needs a very specific set of host/pathogen interactions to work.

The next most obvious answer is that the host has adapted to the virus. What might that mean, mechanistically? And is it something we can take advantage of clinically?

Sin Nombre virus
Sin Nombre virus

The hantavirus family include viruses that infect rodents, such as such as Sin Nombre virus,1 which normally infects deer mice, and Seoul virus, which normally infects Norway rats. These viruses usually cause a persistent infection, and their rodent hosts show little if any disease; but occasionally the viruses can jump into humans, and in humans hantavirus diseases are often pretty severe. Presumably the rodents and their viruses have evolved some kind of arrangement that allows persistent infection without a lot of disease, and humans lack that adaptation.

Two recent papers2 show how that works, though I think there are some interesting questions still open. Both papers found that regulatory T cells, affectionately known as TRegs, are responsible for the viral persistence.

TRegs are one of the immunological flavours of the month, or even of the past few years. There’s a kind of amusing history to TRegs that I won’t go into here that led to an embarrassed silence on the subject for a few years in the 1990s, but recently they’ve been implicated in more and more interesting stuff. In general, you need TRegs to balance the pro-inflammatory side of the immune system: people and mice lacking TRegs have massive, lethal, autoimmune disease. It turns out that mice infected with Sin Nombre virus, and rats infected with Seoul virus, have elevated TRegs. In the rats at least3 removing TRegs reduced the viral load. So apparently, one reason the hantaviruses are able to persist, is that their hosts turn on a TReg response when they’re infected; that reduces the inflammatory response to the virus (that’s what TRegs do), so the immune system doesn’t eliminate it.

Hantaviruses themselves are non-cytopathic, meaning they don’t directly kill cells — disease associated with hantavirus infection is probably mainly because of the immune response killing infected cells. For example, when humans are infected with hantaviruses, they can develop cardiopulmonary syndrome and hemorrhagic fever with renal syndrome. These are diseases of excessive inflammation. By increasing their TReg responses, rodents suppress immune-mediated killing of the infected cells, and prevent this sort of immune-mediated disease. 4 The virus can persist because the immune response is quiet, but it doesn’t cause disease, because the immune response is quiet.

TReg
TReg

Is this something we might consider pursuing with human infections? Not just with hantaviruses: What about other persistent viruses like hepatitis C virus (HCV), or even HIV? Remember that simian immunodeficiency virus in its natural host (say, sooty mangabeys) is a persistent virus that doesn’t cause disease, but in non-natural hosts (like rhesus macaques) it causes a progressive disease. Could this be because TRegs prevent immune-mediated disease and let SIV persist harmlessly in their natural host? Could upregulating TRegs cure AIDS?

No. Yes. Maybe.

This is one of those things where there’s no simple answer. It turns out that HCV,5 HIV,6 and SIV7 in macaques are all already good at upregulating TRegs. In all likelihood that’s one of the reasons they can establish a persistent infection in the first place.

On the other hand, TRegs can do good as well as the opposite.8 Even though having too many TRegs lets HCV establish infection, once the infection is under way the TRegs seem to reduce the damage associated with the infection; like hantavirus infections in rodents.

The whole thing is very complicated, 9 and probably is very dependent on the fine details, location, and timing of the host/virus interaction. It’s definitely not a panacea. Still, it’s worth looking at more, to try to understand what the important details are, in the hope that this powerful system can eventually become a tool in the clinics.


  1. Best virus name EVAR![]
  2. Schountz, T., Prescott, J., Cogswell, A. C., Oko, L., Mirowsky-Garcia, K., Galvez, A. P., et al. (2007). Regulatory T cell-like responses in deer mice persistently infected with Sin Nombre virus. Proc Natl Acad Sci U S A, 104(39), 15496-15501. And: Easterbrook, J. D., Zink, M. C., & Klein, S. L. (2007). Regulatory T cells enhance persistence of the zoonotic pathogen Seoul virus in its reservoir host. Proc Natl Acad Sci U S A, 104(39), 15502-15507.[]
  3. it wasn’t tested in the mice[]
  4. Interestingly, though, rats infected with Seoul virus had low-level damage in their lungs, and that damage was reduced when the TRegs were reduced. So it’s not a simple black/white thing.[]
  5. Rushbrook, S. M., Ward, S. M., Unitt, E., Vowler, S. L., Lucas, M., Klenerman, P., et al. (2005). Regulatory T cells suppress in vitro proliferation of virus-specific CD8+ T cells during persistent hepatitis C virus infection. J Virol, 79(12), 7852-7859.[]
  6. Nilsson, J., Boasso, A., Velilla, P. A., Zhang, R., Vaccari, M., Franchini, G., et al. (2006). HIV-1-driven regulatory T-cell accumulation in lymphoid tissues is associated with disease progression in HIV/AIDS. Blood, 108(12), 3808-3817.[]
  7. Estes, J. D., Li, Q., Reynolds, M. R., Wietgrefe, S., Duan, L., Schacker, T., et al. (2006). Premature induction of an immunosuppressive regulatory T cell response during acute simian immunodeficiency virus infection. J Infect Dis, 193(5), 703-712.[]
  8. Bad[]
  9. A good review is: Robertson, S. J. & Hasenkrug, K. J. (2006). The role of virus-induced regulatory T cells in immunopathology. Springer Semin Immunopathol, 28(1), 51-62.[]
October 18th, 2007

HSV, latency, and T cells

Baines HSV
Electron tomogram of HSV1

Latency is one of the defining characteristics of the herpesviruses, and it’s also one of the least well understood — especially the role of immunity in latency.

One of the open questions about HSV is what latency really means, in vivo. We know that the virus enters neurons, and then seems to pretty much shut down all its proteins. That should make it mostly invisible to the immune system (or at least to the adaptive immune system). Yet HSV reactivates intermittently (those recurrent cold sores), and it’s generally agreed that reactivation is associated with mild immune suppression. 2

So one question is: If there are no viral proteins being expressed, how can the virus evaluate the status of the immune system outside the neuron? How does the virus know there’s immune suppression out there?

There have been, I think, three general answers to that question:

  1. Viruses reactivate randomly from the neurons, and constantly send out probes into the body. According to this idea, it’s not that the virus needs immune suppression to reactivate, but rather it’s reactivating all the time; it needs immune suppression for the reactivations to take root and become detectable.
  2. Viruses do in fact have some proteins expressed even during latency, and can sense the state of the immune systems that way. (A variant on this is the possibility that non-protein components are doing the sensing. We know that some viral RNAs are present during latency.)
  3. It’s the immune system that keeps the virus latent. Take away the immune system, and out pops the virus.

These are not mutually exclusive possibilities, though it’s likely that just one factor is the main cause.

CD8 in ganglia, Khanna et al 2003In early days, it was tentatively believed that (3) was not likely — you’d need to have some signal to draw the immune response to the appropriate neuron (especially because of course T cells are normally kept away from the brain) and because HSV doesn’t express proteins during neuronal infection, where’s your antigen target? On the other hand, it’s been known for a while that in mice, at least, ganglia that are latently infected with HSV show signs of chronic inflammation.3 It’s never been entirely clear, at least to me, if latent infection in the mouse is precisely like latency in humans, but recently similar findings have been made in human ganglia, even in the complete absence of detectable viral proteins.4

Chronic inflammation means there are immune cells hanging around inside the ganglia. What are these cells? Are they specific for the virus, or are they non-specific cells that have been attracted into the area by the general inflammation? A while ago, I mentioned evidence that antigen-specific T cells are specifically allowed in to the brain, and it was shown a few years ago that in mice the infiltrating cells were antigen-specific, for herpes simplex antigens. (The image at right shows infiltrating T cells in an infected ganglion, from that paper.5) Now, it turns out that the same is true in humans:

CD8+ T cells are selectively located in close vicinity to HSV-1 latently infected neuronal cell bodies. … These results support the recent diversion from the old dogma that HSV-1 latency is an antigenic silent infection. … The collective data argue that the neuron-interacting T cells are most likely HSV-1 specific and recognize those latently infected neurons that intermittently express low amounts of viral proteins below levels detectable by using biochemical means or a particular subset of neurons in which HSV-1 has reactivated from latency. 6

The obvious (though not necessarily correct!) conclusion is that these T cells surrounding infected neurons prevent HSV reactivation from latency, and only when the surrounding T cells are reduced in number or effectiveness — during immunosuppression, in other words — can the virus escape long enough to set up a reactivation.

Interestingly enough, this seems to not be true for Varicella-Zoster virus! VZV (another herpesvirus, of course — chicken-pox and shingles) also establishes latency in the same kinds of ganglia, yet there were no VZV-specific infiltrating T cells. “Neurons latently infected with VZV appear to be invisible for T cells, whereas HSV-1 has adopted mechanisms to impede cytotoxic T lymphocyte-mediated eradication from their latency stronghold.”7


  1. Electron tomogram of a HSV nucelocapsid completing envelopment , from Baines, J. D., C. E. Hsieh, E. Wills, C. Mannella, and M. Marko. 2007. Electron tomography of nascent herpes simplex virus virions. J Virol 81: 2726-2735.[]
  2. This is actually not very solidly proven, in my mind, though it’s widely accepted — but let’s take it as a given that reactivation from latency requires immune suppression.[]
  3. For example, Shimeld C, Whiteland JL, Nicholls SM, Grinfeld E, Easty DL, Gao H, Hill T. (1995). Immune cell infiltration and persistence in the mouse trigeminal ganglion after infection of the cornea with herpes simplex virus type 1. J Neuroimmunol 61:7-16 ; and Liu T, Tang Q, Hendricks RL. (1996). Inflammatory infiltration of the trigeminal ganglion after herpes simplex virus type 1 corneal infection. J Virol 1996, 70:264-27[]
  4. Theil, D., T. Derfuss, I. Paripovic, S. Herberger, E. Meinl, O. Schueler, M. Strupp, V. Arbusow, and T. Brandt. 2003. Latent herpesvirus infection in human trigeminal ganglia causes chronic immune response. Am J Pathol 163: 2179-2184.[]
  5. Khanna, K. M., R. H. Bonneau, P. R. Kinchington, and R. L. Hendricks. 2003. Herpes simplex virus-specific memory CD8+ T cells are selectively activated and retained in latently infected sensory ganglia. Immunity 18: 593-603. []
  6. Verjans, G. M., Hintzen, R. Q., van Dun, J. M., Poot, A., Milikan, J. C., Laman, J. D., Langerak, A. W., Kinchington, P. R., and Osterhaus, A. D. (2007). Selective retention of herpes simplex virus-specific T cells in latently infected human trigeminal ganglia. Proc Natl Acad Sci U S A 104, 3496-3501. []
  7. Could this be the elusive function of immune evasion molecules? Since the HSV immune evasion protein ICP47 does not work in mice, that can’t be the only answer, but it may be a factor in humans.[]
October 15th, 2007

MHC molecules: The sitcom

I talk a lot about MHC molecules, especially MHC class I. Something I haven’t mentioned much is that MHC class I is just one member of a huge extended family of structurally-related molecules. MHC class I and class II are the “classical” MHCs (the Archie Bunker and Edith of histocompatibility complexes); but there’s a long list of spin-off “non-classical” MHCs as well. 1 Many of them are solid citizens that conscientiously perform immune-related tasks, but there are also some whacky neighbours that run around doing zany stuff.

Last week I covered the structure of MHC class I (the Bjorkman structure of HLA-A2). For comparison, here’s a superficial glance at the comparative anatomy of a number of classical and non-classical MHCs. The structural similarity is often pretty remarkable, considering the variety of things they do — iron-regulatory molecules, odor detection, and antigen presentation, for a start! In general, the two things I look at structurally are whether they bind to β2-microglobulin, and what their equivalent of a peptide-binding groove looks like. (Because I’m mostly emphasizing the peptide-binding groove in these images, β2-m is often hard to see, hidden as it is underneath the floor of the groove. If you squint, you can see it — it’s colored blue in the ribbon structures — except for ZAG and MHC class II, which don’t have it at all.)

The molecules (along with their Protein Database accessions, and the Pubmed ID of the article describing their structure) I’ve chosen are as follows:

MHC Family Role Groove binds β2-m PDB PMID
HLA-A2 (MHC class I) Classical Antigen presentation Peptide Yes 2GTW 10843695
HLA-DR (MHC class II) Classical Antigen presentation Peptide No 1DLH 8145819
HLA-E Non-classical NK cells recognition Peptide Yes 1MHE 9660937
CD1d Non-classical Non-classical T cell recognition Lipglycans Yes 2GAZ 16982895
FcRn (Neonatal Fc receptor) Non-classical Neonatal Fc receptor Nothing Yes 1EXU 10933786
ZAG (Zinc-α-2-Glycoprotein) Non-classical Lipid catabolism Something (what?) No 1ZAG 10206894
HFE Non-classical Iron regulation Nothing Yes 1A6Z 10638746
Qa2 Non-classical Immune regulation? Peptide Yes 1k8d 11738047
M10.5 Non-classical Odor receptor chaperone? Nothing? Yes 1ZS8 16089503

First of all, let’s look at some of the guys who bind stuff in their groove (click on each image for a larger version). Here we are, looking “down” from the top of the molecule (as the T cell would looks at HLA-A2, for example) — a surface representation of the MHC molecule with its ligand shown in green. In the lower row, I show the ligand all by itself, as it sits in the groove.

HLA-A2 HLA-DR HLA-E Qa-2 CD1d
HLA-A2 peptide HLA-DR peptide HLA-E peptide Qa-2 peptide CD1d Ligand
MHC class I MHC class II HLA-E Qa-2 CD1d

The odd one out is CD1d, which binds non-peptides — here, a Mycobacterium tuberculosis phosphatidylinositol mannoside. The CD1 binding groove is much deeper and more hydrophobic than that of the peptide-binders’: a tunnel, in places, rather than a groove. Another difference is between MHC class II (HLA-DR in this picture) and most of the other peptide binders; MHC class II peptides can flop out over the edges of the groove (whereas the other guys bury the ends of the peptide) — so MHC class II peptides can be quite a bit longer than the 9 or so amino acids that other MHCs can handle.

Next let’s look at the rest of the crew — the ones that don’t bind peptides, or perhaps anything, in their grooves. For ease of comparison I’m including MHC class I (HLA-A2) again, at the left. In the lower row, I’m showing the same views as ribbon diagrams.

HLA-A2 FcRn ZAG HFE M10.5
HLA-A2 FcRn ZAG HFE M10.5
MHC class I FcRn ZAG HFE M10.5

(Notice that ZAG doesn’t have any β2-m associated with it, but nevertheless forms a very nice binding groove.) You can see very easily for FcRn that the groove is completely screwed up: it’s collapsed and filled in, so there’s nowhere for a peptide, or even a much smaller molecule, to fit in. It’s harder to see for HFE, but the groove there is still too narrow to bind peptides (though it’s not completely collapsed). ZAG and M10.5 are much more interesting. Both have grooves that could actually hold something, but we don’t know what (if anything) they bind.

ZAG binding With ZAG, something actually co-crystallized with the molecule! –but they don’t know what it is. ” … Instead of a peptide, the ZAG groove contains an as yet unidentified ligand that cocrystallizes with the protein.”2 (The figure at right clearly shows something undefined in the ZAG binding groove.)

M10.5 was equally interesting (in the “incident of the dog in the night” sense) — because, even though the groove could actually potentially hold a peptide3 nothing at all was in the crystallized groove. This never happens with classical MHC molecules, which always find something to bind there. But the groove seems (biochemically as well as structurally) to be a complete blank.

M10.5 has an open groove more similar to the peptide-binding classical class I MHC molecules than the non-peptide-binding homologs. … M10.5 electron-density maps show no ordered density corresponding to a peptide … We conclude that the M10.5 groove does not contain a single defined peptidic or non-peptidic occupant or a mixture of compounds with a similar conformation.4

Given that M10.5 seems to be involved in odor recognition — Could it be a pheremone binder? Sadly (because that would be a great story), probably not: “The hydrophobic nature of most pheromones is not complementary to the charged character of the groove, which is much larger than a single pheromone molecule.”

Their final guess is that maybe M10.5 chaperones a V2R — a different receptor:

In this hypothesized scenario, newly synthesized M10 and V2R proteins would be stabilized through mutual interactions with a V2R loop in the M10 groove, enabling the M10 to escort the V2R to the cell surface, rationalizing the observation that M10 proteins are required for cell surface expression of V2Rs. 4

Anyway, there’s a field guide to some of the MHC family. There are many, many more, each with its own beautiful plumage and habitat, but these cover most of the general variations in structure you’ll see.


  1. Actually, many of the “non-classicals” are just about as ancient as the “classical” MHCs, so it’s not really clear which is the spin-off. But I liked the metaphor.[]
  2. Sanchez, L. M., A. J. Chirino, and P. Bjorkman. 1999. Crystal structure of human ZAG, a fat-depleting factor related to MHC molecules. Science 283: 1914-1919. []
  3. “We conclude that the M10.5 groove can accommodate a peptide that adopts a class I MHC-binding conformation, but that differences between the A and F pocket regions of M10.5 and classical class I MHC molecules would require a peptide bound in the M10.5 groove to be anchored differently than a class I MHC-binding peptide”[]
  4. Olson, R., K. E. Huey-Tubman, C. Dulac, and P. J. Bjorkman. 2005. Structure of a pheromone receptor-associated MHC molecule with an open and empty groove. PLoS Biol 3: e257.[][]
October 14th, 2007

T cell skewing?

HLA-A
HLA-A
HLA-B
HLA-B

As a scientist, I’m comfortable with being wrong most of the time. I’m used to it. But when I make a throwaway comment in a blog and get smacked down in PNAS the following week … well, it seems like overkill, you know?

It’s not quite that bad. A couple weeks ago I was talking about the possibility that tuberculosis-specific T cells were mainly restricted by HLA-B as proposed by Lewinsohn et al.1 I wasn’t very enthusiastic about the idea, suggesting that the skewing they saw was probably just chance. I did admit that there was some biological plausibility to the possibility, but I was mainly thinking about some kind of allele-specific immune evasion. Now, Harari et al argue in the latest PNAS that there is a fundamental difference between CTL restricted by HLA-B, and those restricted by HLA-A.

HLA-A and HLA-B are two distinct major histocompatibility class I genes. There are three classical class I genes in humans (and many other species), with A and B being the most important for T cell recognition. (Here is a map of the human MHC region — HLA-A is way over to the left, in among a bunch of non-classical MHC class I genes, while HLA-B and C are closer to the middle, to the right of the pinkish-shaded Class I region.) Each of HLA-A, B, and C has many different alleles among the human population (HLA-A, B, and C are not alleles of each other, a mistake a lot of people make — they are independent, if linked, genes.) HLA-A and B are typically around 80% identical at the protein sequence level (different alleles within A or B are around 90-95% identical), they look pretty much identical structurally2, and in general they’re hard to tell apart. The only things that I can think of where HLA-A and B differ is that (1) HLA-B has been evolving faster, and (2) HIV is more likely to be controlled by HLA-B than A alleles — something that I, and I think just about everyone, put down to the allelic differences (i.e. different viral peptides being bound) rather than any general HLA-A vs. B distinction. So far as I can remember, no one has ever suggested that they have functional differences as far as T cells go.

Recently it’s been suggested that CTL fall into two broad categories, “polyfunctional” and “only effector”, and that the former category is more important for controlling viruses3. Harari et al4 are now suggesting that these different categories are more associated with HLA-B or -A, respectively:

Polyfunctional CD8 T cell responses, also including IL-2 production and Ag-specific proliferation, are predominantly driven by virus epitopes restricted by HLA-B alleles. … Conversely, HLA-A-restricted epitopes are mostly associated with “only effector” IFN–secreting, with cytotoxicity, and with the lack of IL-2 production and Ag-specific proliferation. … Thus, the functional profile of the CD8 T cell response is strongly influenced by the extent to which there is stimulation of polyfunctional (predominantly restricted by HLA-B) versus only effector (restricted by HLA-A) T cell responses.

It’s an interesting idea. However, their data are not really striking — to me at least. The differences they show in polyfunctional CTL have HLA-A restricting, for example, 0.1% to HLA-B’s 0.25% of the population — not so exciting. The stats they show have P values that are officially statistically significant but aren’t really overwhelming — 0.03, 0.02 — and I question whether the stats are done correctly in all cases (e.g. in one part of Figure 3 where there are overlapping error bars and an N of 5, but they show a P value of 0.005, which just doesn’t look right, unless I’m missing something). They do call the effect “skewing”, which is a conservative claim, so I’ll tentatively accept that far. But I’d really like to see this replicated by a different group, and extended in a larger dataset, before I am completely convinced.


  1. Immunodominant Tuberculosis CD8 Antigens Preferentially Restricted by HLA-B. Lewinsohn, D. A., Winata, E., Swarbrick, G. M., Tanner, K. E., Cook, M. S., Null, M. D., Cansler, M. E., Sette, A., Sidney, J., and Lewinsohn, D. M. (2007). PLoS Pathog 3, e127. []
  2. The ribbon diagrams here are of HLA-A*1101 and HLA-B*2709 []
  3. Reviews: Harari, A., Dutoit, V., Cellerai, C., Bart, P. A., Du Pasquier, R. A., and Pantaleo, G. (2006). Functional signatures of protective antiviral T-cell immunity in human virus infections. Immunol Rev 211, 236-254. and Harari, A., Dutoit, V., Cellerai, C., Bart, P. A., Du Pasquier, R. A., and Pantaleo, G. (2006). Functional signatures of protective antiviral T-cell immunity in human virus infections. Immunol Rev 211, 236-254. []
  4. Harari, A., Cellerai, C., Bellutti Enders, F., Kostler, J., Codarri, L., Tapia, G., Boyman, O., Castro, E., Gaudieri, S., James, I., John, M., Wagner, R., Mallal, S., and Pantaleo, G. (2007). Skewed association of polyfunctional antigen-specific CD8 T cell populations with HLA-B genotype. Proc Natl Acad Sci U S A 104:16233-1623 []
October 11th, 2007

Creepy cancer post of the month

Tasmanian DevilThere’s been some buzz about the recent paper on the contagious tumor of Tasmanian devils.1 Clearly the thing is a ghastly disease that’s threatening the Devils with extinction — but from a technical viewpoint, the paper last year on a different transmissible tumor was much more interesting.

The other tumor I’m talking about is canine transmissible venereal tumor (CTVT), and the paper is:
Murgia, C., Pritchard, J. K., Kim, S. Y., Fassati, A., and Weiss, R. A. (2006). Clonal origin and evolution of a transmissible cancer. Cell 126, 477-487 .

Most people haven’t heard of CTVT. (I had, but then I’m a veterinarian originally.) It’s just what it sounds like from the name: A sexually-transmitted cancer that spreads between dogs. (And it does seem to be a pure cancer, not a virus that causes cancer.) Until the Tasmanian Devil contagious tumor, that made CTVT essentially unique; no other tumors are known to be transmissible. (Actually, the Tasmanian Devil paper cites a tumor of Syrian hamsters2 that is apparently transmissible — but I’d never heard of that before, and I’m not clear that it still exists or if it has died out in the past 40 years.)

It turns out that the Tasmanian Devil tumor apparently can spread because it is essentially a self graft. Apparently Devils are highly inbred, and show very little polymorphism at the major histocompatibility complex (MHC) region. As I’ve been pointing out (ad nauseum) lately (see here and the links therein) this is really unusual, as in most vertebrates the MHC region is normally by far the most diverse region of the genome. The MHC region is important in graft rejection — duh, that’s what “histocompatibility” means. Essentially, then, it seems that the Devil tumor can “take” on virtually any other Devil, because it’s recognized as “self” MHC. 3 This is interesting, of course, but there’s nothing surprising about it. Endangered species are commonly inbred — inbred animals lack variation at the MHC4 — and matching MHC allows grafts to take. This is all known.

The canine tumor story is very different. This tumor does not match the recipient dogs’ MHC. The same tumor can infect unrelated dogs, with virtually any MHC; Murgia et al looked at tumor-bearing dogs from around the world (“None of the host dogs showed close relatedness to any of the others, consistent with the fact that they came from three locations in Europe, Asia, and Africa and were mongrels”), and they all carried the same tumor. So why is CTVT not rejected as an allograft?

CTVT phhylogeny Most, if not all, tumors in humans show evidence of having been edited by the immune system. That is, the tumors have altered their MHC expression in some way that probably allows them to evade the immune system. That’s one reason that tumors themselves are not rejected. (Presumably, there are many more proto-tumors that arise during our lifetime, that fail to alter their MHC and are destroyed by our immune systems before there are more than a half-dozen abnormal cells. We only see the successful ones.) CTVT has done this, as well, and expresses very low (but detectable) levels of MHC. Also, again like many other tumors, CTVT expresses an immune modulator, TGFβ. 5 Murgia et al suggest that these are enough to make the tumor invisible to the immune system and allow it to engraft.

But I’m not at all convinced. These changes — low MHC expression, high TGFβ — are very common, if not universal, tumor adaptations, yet CVTV is unique — extraordinarily, spectacularly, unique — in its ability to spread and persist within a highly outbred species. CVTV has persisted for hundreds, if not thousands or even tens of thousands of years:

The precise date when CTVT first occurred is difficult to determine. From its indistinguishable histopathology and its ability to grow as an allograft, it is likely that Novinski (1876) studied the same clone, and CTVT could have become established centuries before this date. Our analysis of divergence of microsatellites indicates that the tumor arose between 200 and 2500 years ago. Whether this time period represents the time the tumor first arose or whether it represents a later bottleneck in the tumor’s dispersion as a parasite cannot be resolved. While this estimated date indicates a relatively recent evolutionary origin, CTVT represents the oldest known mammalian somatic cell in continuous propagation, having undergone countless mitoses and host-to-host transfers.

In fact, the tumor may even predate dogs. The figure to the right from Murgia et al shows the relationship between the tumor and dogs and wolves (click for a larger version)– the thing is even closer to wolves than it is to dogs.

Frankly, the thing is damn creepy, and I kind of hope I’m right and there’s much more to its ability to persist than the really very common changes that have been pointed at so far, because I wouldn’t want to think that every tumor was capable of this kind of behaviour. Whatever it is, though, is more novel and scientifically interesting than the Tasmanian Devil tumor. Hopefully the Devil tumor is easier to deal with than the canine one.


  1. Siddle, H. V., Kreiss, A., Eldridge, M. D., Noonan, E., Clarke, C. J., Pyecroft, S., Woods, G. M., and Belov, K. (2007). Transmission of a fatal clonal tumor by biting occurs due to depleted MHC diversity in a threatened carnivorous marsupial. Proc Natl Acad Sci U S A 104:16221-16226 []
  2. Copper, H. L., Mackay, C. M., And Banfield, W. G. (1964). Chromosome Studies Of A Contagious Reticulum Cell Sarcoma Of The Syrian Hamster. J Natl Cancer Inst 33, 691-706.[]
  3. This is apparently the same mechanism as with the Syrian hamster transmissible tumor; hamsters are highly inbred as well. Streilein, J. W., and Duncan, W. R. (1983). On the anomalous nature of the major histocompatibility complex in Syrian hamsters, Hm-1. Transplant Proc 15, 1540-1545.[]
  4. Though not inevitably. The San Nicolas Island foxes I used as an example were inbred, yet had significant MHC diversity.[]
  5. The Devil paper states that CTVT also “up-regulates nonclassical class I expression to avoid the natural killer cell response” — but I don’t know where they get this from. It’s not in the reference they cite, and I can’t find evidence for it anywhere (but that doesn’t mean it’s not correct) []
October 9th, 2007

Heavyweight championship: Overdominance vs. frequency-dependent selection

Mouse lemur
Mouse Lemur. “MHC polymorphism in M. murinus is maintained through pathogen-driven selection acting by frequency-dependent selection”1

Is the major histocompatibility complex so diverse because of overdominance, frequency-dependent selection, or both?

This is part of a series that started with the observation of MHC diversity , discounted as causes high mutation frequency and maternal/fetal interactions , and suggested that mate choice (sexual selection) was not enough to account for the diversity by itself. That leaves us with two strong candidates for the cause of MHC diversity: overdominance, or frequency-dependent selection, or both. The question of which is most important hasn’t yet been definitively answered, and the story is too complex to address really properly in a single post.

What’s the difference between the candidates? According to the overdominance hypothesis, individual who carry more MHC alleles (that is, the individual’s MHC region carries as many different alleles as possible) are more fit (“heterozygote advantage”). According to the frequency-dependent hypothesis, individuals who have rare MHC alleles are more fit. Both hypotheses make biological sense, mathematical models suggest that both could lead to diversity,2 both have some support from field observation (see the picture at top for an example), and both are very difficult to test directly.

There is one major difference between the hypotheses: “A crucial difference between the two types of balancing selection is that overdominance predicts a stable polymorphism, whereas a polymorphism maintained by frequency dependence will be dynamic.” 3 In other words, “in the presence of overdominant selection polymorphic alleles (allelic lineages) may persist in the population for an extremely long time,”4 whereas in contrast “a rare old allele should have no such advantage5 and should disappear from the population by selection or genetic drift.”3

This difference — that alleles should appear and then disappear over time — has been used as an argument against frequency-dependent selection, because some MHC alleles are ancient. “Comparison of human and chimpanzee alleles reveals extensive sharing of polymorphisms, confirming that diversification is a slow process, and that much of contemporary polymorphism originated in ancestral primate species before the emergence of Homo sapiens.”6

Parham 1996 But it depends on your scale: “During the timeframe of mammalian evolution, the lifetimes of a functional class I locus are short and those of individual alleles even shorter.” 7

Even on a short scale, alleles appear and disappear at a great rate. My favourite example of this is the map on the right 8 (I liked it so much I scanned it, years ago; I don’t have access to the 1996 issues of Science on line. Click on the map for a larger version). This shows what happened to MHC diversity during the peopling of the Americas. You can see new alleles popping up down the migration route — but the key point I want to make is made by the authors in a different paper: “Although many new HLA-B alleles have been produced in Latin America, their net effect has been to differentiate populations, not to increase allele diversity within a population.” 9

In other words, rare old MHC alleles are not selected, but disappear, while rare new alleles are selected. This is consistent with the predictions of frequency-dependent selection than of overdominance, I think. But there are also lots of strong arguments for overdominant selection, some of which I’ll mention next time around.


  1. Schad, J., Ganzhorn, J. U., and Sommer, S. (2005). Parasite burden and constitution of major histocompatibility complex in the Malagasy mouse lemur, Microcebus murinus. Evolution Int J Org Evolution 59, 439-450.[]
  2. But “mathematical study alone cannot distinguish between this model [frequency-dependent selection - IY] and the overdominance model.” –Takahata, N., and Nei, M. (1990). Allelic genealogy under overdominant and frequency-dependent selection and polymorphism of major histocompatibility complex loci. Genetics 124, 967-978. []
  3. Slade, R. W., and McCallum, H. I. (1992). Overdominant vs. frequency-dependent selection at MHC loci. Genetics 132(3), 861-864.[][]
  4. Takahata, N., and Nei, M. (1990). Allelic genealogy under overdominant and frequency-dependent selection and polymorphism of major histocompatibility complex loci. Genetics 124, 967-978. []
  5. I believe, though, this is only true with one type of frequency-dependent selection — a pure “minority advantage” model, with no pathogen adaptation, doesn’t predict this. On the other hand, without pathogen adaptation the biological argument for this model is much weaker.[]
  6. Parham, P., Lawlor, D. A., Lomen, C. E., and Ennis, P. D. (1989). Diversity and diversification of HLA-A,B,C alleles. J Immunol 142, 3937-3950.[]
  7. Parham, P. (1994). The rise and fall of great class I genes. Semin Immunol 6, 373-382[]
  8. Parham, P., and Ohta, T. (1996). Population biology of antigen presentation by MHC class I molecules. Science 272, 67-74.[]
  9. Parham, P., Arnett, K. L., Adams, E. J., Little, A. M., Tees, K., Barber, L. D., Marsh, S. G., Ohta, T., Markow, T., and Petzl-Erler, M. L. (1997). Episodic evolution and turnover of HLA-B in the indigenous human populations of the Americas. Tissue Antigens 50, 219-232.[]
October 7th, 2007

It was twenty years ago today

Fig 2aExactly twenty years ago, the structure of HLA-A2 was published: 8 October, 1987.1

That was actually the year before I started grad school, so I didn’t actually know a thing about it at the time, but it was the subject of one of the first journal club presentations I went to, maybe a year after it was published — so it was still fresh and surprising. What a jaw-dropping way to start my new career! I was already interested in MHC, but this was the hook.

At the time, some of the function of MHC class I had been worked out (first by by Doherty and Zinkernagel, and then lots of others) but the mechanisms were still pretty obscure. In particular, although it was known that adding peptide fragments to a cell could sensitize it to killing by cytotoxic T lymphocytes, it was by no means clear how that worked — it was still not known that peptides could physically associated with MHC class I. This was known for MHC class II, but it was not at all appreciated how similar in structure class I and II are — nor was it known how or where peptides bound to class II. However, the connection had been made: “By analogy, it is likely that the homologous class I molecules bind antigenic peptides, and that the HLA-peptide complex is recognized by T-cell receptors in CTL.2

Paper 1

At the time, crystallography was (at least in immunology) a fairly obscure, esoteric technique — mildly interesting, but not what you’d call practical, in terms of answering questions. 3 All that changed with this picture:

Fig6b

(Funny, I remember that picture as being, like, the size of a movie poster — in fact it’s the bottom part of one small corner of a single page. I can’t remember if it was on the cover of that issue of Nature, and the on-line archives don’t seem to have cover illustration that far back.)

That figure shows a little bit of undefined, unresolved mist (in pink) in the middle of the reasonably well-defined HLA-A2 molecule. The location of that little bit of mist, and its very mistiness, were the stunning part of the paper.

Of course, the location represents what we now call the peptide binding groove. (They had offered a drawing of the groove in Fig 2a [reproduced at the top of this post], but it didn’t click with me until I saw the later figure with the mist in it — then I went back and looked again!) 4 It offered a beautiful, clear, simple way to understand MHC function, a perfect dock for a peptide. At the time I had no idea that proteins could actually be so, well, physical — I had vague ideas of interactions and van der Waals forces and fancier stuff that held them together, it had never occurred to me that proteins were shaped like little Legoâ„¢ bricks, that could SNAP together with a satisfying CLICK. Here was something I could practically feel, pick up and move around, and it all made macroscopic sense. I think that I was not the only one who had the same reaction to the structure, too — this was the paper that led to a breakthrough in the use of structural information in immunology.

The even more exciting part was that the mist was mist. Even though the rest of the protein was clear,5 the presumptive peptide was a blur. And the reason for that was instantly obvious, even to me: It was not a single peptide, it was a composite of hundreds or thousands of different peptides, all superimposed on each other. That blur was the secret of MHC specificity, and of MHC complexity. Again, a very physically satisfying explanation.

Wiley’s group had a second article, back to back with this first, on the “Foreign antigen binding site and T cell recognition regions” of the protein. Here they raked together a vast, scattered literature — functional residues, polymorphisms, antibody binding sites, transplant rejections — and showed that it all made sense. Everything could be set into place on the structure they’d shown. Among other things, they predicted that the T cell receptor would bind on top of the “foreign antigen binding site” (remember that going in, there was no direct evidence that antigens even bound to the MHC class I, let alone to any particular site; nor was the T cell receptor identified), recognizing both MHC and peptide residues. They identified contact residues for peptide and for TcR, explained how polymorphisms contributed to diversity … They took the jigsaw puzzle from an outline and a few pieces of blue sky and grass, all the way through the final picture (kindly leaving a few small pieces unplaced, for the rest of us to fill in.)

This second paper was less physical, more cerebral, and it took me longer to understand most of their arguments. It didn’t matter. They had me from that little blue-on-black square in the first paper, and I was happy to spend the time reading and thinking that the second paper needed.

References:
Bjorkman, P. J., Saper, M. A., and Wiley, D. C. (1987). Structure of human class I histocompatibility antigen, HLA-A2. Nature 329, 506-512.

Bjorkman, P. J., Saper, M. A., Samraoui, B., Benett, W. S., Strominger, J. L., and Wiley, D. C. (1987). The foreign antigen binding site and T cell recognition regions of class I histocompatibility antigens. Nature 329, 512-518.


  1. My first son was born on Oct 8, exactly 12 years later, but I swear it was a coincidence. (He is getting a new bike for his birthday, if anyone is interested.) []
  2. All quotes are from the papers in question, from Pamela Bjorkman, in Don Wiley’s lab. []
  3. Remember that this this is all my own viewpoint, and at the time I was a very new student — it’s quite possible that some, or even many, people in the field were more aware of the possibilities than I was. But I think it’s a generally fair statement.[]
  4. I may have got this wrong, but I believe that, in those almost-pre-computer days, Figure 2a was drawn, by hand, by Hidde Ploegh, then a post-doc in the Strominger lab.[]
  5. Relatively speaking — at 3.6 Ã… it is by today’s standards a low-resolution structure[]