Mystery Rays from Outer Space

Hepatitis

(Spending a few days in Toronto with my kids, so I can go to the Darwin exhibit at the Royal Ontario Museum.¹ Depending on when I get back, my next blog post may be a little delayed.)

Even though cytotoxic T lymphocytes are called “cytotoxic”, it was no surprise when new techniques in the mid-90s suggested that CTL might have other strings to their bows. (I talked about this the other day.) Some experiments were also pointing the same way. For example, Frank Chisari’s group offered evidence that CTL might be able to control hepatitis B virus without actually being cytotoxic.

Hepatitis B virus was a particularly difficult case for study in general, because it’s pretty much a human-specific virus. (HBV is a pain in tissue culture. There are a couple of animal models, but they’re hardly convenient. I mean, ducks? Woodchucks?) From studies of the natural disease, it seemed pretty clear that natural HBV infection was often handled by CTL; people who control HBV show a strong CTL response, those in whom HBV progresses and becomes chronic usually do not. But testing the function of CTL in HBV infection wasn’t easy without an animal model.

So Chisari made an animal model. He made transgenic mice that express the HBV genome (that is, the mouse genome contained the HBV genome) under a liver-specific promoter.² There you go, hepatitis B virus in the mouse liver. It can’t spread and infect new cells as would normally happen in humans, but on the other hand you’re expressing the virus in essentially all the liver cells anyway, so you don’t need to infect new cells.

Hepatitis B virus-transgenic mice

Of course, the “virus” is “self” under those conditions, meaning that the immune system doesn’t react to it. So Chisari’s group immunized other mice to raise HBV-specific CTL, and transferred the CTL into the transgenic mice.³ These HBV-specific CTL did two things: They apparently controlled the virus “infection”, and they damaged the liver.⁴ The presumption was that the two findings were the same effect, and the CTL were trying to eliminate virus by killing liver cells: “Our data show that antigen-specific immune effector mechanisms can destroy HBV-positive hepatocytes in vitro and in vivo … “.

Hepatitis viruses

Six years later, though, they were suggesting the opposite, demonstrating that in fact CTL were controlling the “virus” through a non-cytolytic mechanism.⁵ (Actually, they showed very similar data a couple of years earlier than that,⁶ but the 1996 paper went further with the mechanisms and was overall more solid.) Essentially, they demonstrated that CTL were shutting down virus expression in the liver by releasing interferon and other cytokines. There is some liver damage, but it’s not necessarily because of direct cytotoxity; it’s a side-effect of the cytokines, not directly related to the viral clearance. (Recently, it’s been proposed that the CTL make a decision which route to go — cytotoxicity vs. cytokine-mediated control — based on the amount of virus antigen. ⁷ This might offer a way to manipulate this and drive the response to the most effective system in other infections, though it’s far from trivial to adjust amount of viral antigen.)

Non-cytotoxic control of HBV

So what? What difference does it make how the CTL are clearing virus, so long as they do clear it? Here are some reasons to care:

• This way a few CTL can affect many target cells. There are some 10¹¹ hepatocytes, and hepatitis viruses can infect a lot of them, very fast. There are different estimates for how long it takes for a CTL to deliver a lethal hit when it’s being cytotoxic, but let’s say something like 30 - 60 minutes per cell. That’s a long, long time for CTL to kill off all the infected cells. If all they have to do is release interferon, they can probably pick off many cells at once and move on. This is a faster way to control viruses, and it doesn’t take as many CTL. Guidotti et al transferred 5 x 10⁶ HBV-specific CTL into the transgenic mice; within 24 hours, all of the virus had been shut down (remember. these are transgenic mice expressing “virus” in every liver cell).
• Again: There are some 10¹¹ hepatocytes, and hepatitis viruses can infect a lot of them, very fast. If CTL have to kill all the virus-infected cells, what’s left of the liver to do it’s usual liverly duties? If the infection can be shut down without killing the cells, you’ve saved your liver. Those 5 x 10⁶ CTL shut down the virus without killing more than 10% of the liver cells.
• Potentially, any inflammation in the liver can lead to protection. If you have HBV, and you’re infected by lymphocytic choriomeningitis, (well, first you’d likely be a mouse), the inflammation induced by the LCMV might shut down the HBV, as a handy side effect. This cross-protection from other viruses has actually been seen both in mice⁸ and — maybe — in humans as well.⁹

So the finding that CTL are capable of shutting down virus replication without having to kill the infected cells, fit very nicely with the new technology showing that CTL commonly have these mechanisms available.

1. Also I promise it’s a complete coincidence that my Red Sox were playing the Blue Jays here last few days. The Jays gave them a whuppin’, but William and I had a fine time at the ballpark anyway.[↩]
2. Chisari, F. V., Pinkert, C. A., Milich, D. R., Filippi, P., McLachlan, A., Palmiter, R. D., and Brinster, R. L. (1985). A transgenic mouse model of the chronic hepatitis B surface antigen carrier state. Science 230, 1157-1160.[↩]
3. As you can see, the system is pretty elaborate, and I’ve never really felt all that comfortable with it — just too Rube Goldberg-ish for my liking — even though there’s nothing specific I can point to; and Chisari’s group is conscientious about their controls.[↩]
4. Moriyama, T., Guilhot, S., Klopchin, K., Moss, B., Pinkert, C. A., Palmiter, R. D., Brinster, R. L., Kanagawa, O., and Chisari, F. V. (1990). Immunobiology and pathogenesis of hepatocellular injury in hepatitis B virus transgenic mice. Science 248, 361-364.[↩]
5. Guidotti, L. G., Ishikawa, T., Hobbs, M. V., Matxke, B., Schreiber, R., and Chisari, F. V. (1996). Intracellular inactivation of the hepatitis B virus by cytotoxic T lymphocytes. Immunity 4, 25-36.[↩]
6. Guidotti, L. G., Ando, K., Hobbs, M. V., Ishikawa, T., Runkel, L., Schreiber, R. D., and Chisari, F. V. (1994). Cytotoxic T lymphocytes inhibit hepatitis B virus gene expression by a noncytolytic mechanism in transgenic mice. Proc Natl Acad Sci U S A 91, 3764-3768.
   also
   Tsui, L. V., Guidotti, L. G., Ishikawa, T., and Chisari, F. V. (1995). Posttranscriptional clearance of hepatitis B virus RNA by cytotoxic T lymphocyte-activated hepatocytes. Proc Natl Acad Sci U S A 92, 12398-12402.[↩]
7. Gehring, A. J., Sun, D., Kennedy, P. T., Nolte-’t Hoen, E., Lim, S. G., Wasser, S., Selden, C., Maini, M. K., Davis, D. M., Nassal, M., and Bertoletti, A. (2007). The level of viral antigen presented by hepatocytes influences CD8 T-cell function. J Virol 81, 2940-2949.[↩]
8. Guidotti, L. G., Borrow, P., Hobbs, M. V., Matzke, B., Gresser, I., Oldstone, M. B., and Chisari, F. V. (1996). Viral cross talk: intracellular inactivation of the hepatitis B virus during an unrelated viral infection of the liver. Proc Natl Acad Sci U S A 93, 4589-4594.[↩]
9. Thio, C. L., Netski, D. M., Myung, J., Seaberg, E. C., and Thomas, D. L. (2004). Changes in hepatitis B virus DNA levels with acute HIV infection. Clin Infect Dis 38, 1024-1029.[↩]

I’ve previously posted on regulatory T cells (TRegs) and their potential role in transplants. Briefly, TRegs are capable of specifically shutting off immune responses to particular antigens; they’re normal components of an immune system. TRegs can be damaging in some contexts — for example, in cancer, where it seems that TRegs often shut off immune responses to tumors, so that the tumor can escape immune clearance; and they can be beneficial in other context — for example, in some persistent virus infections, where a chronic immune response would be damaging, TRegs apparently modulate the immune response so that the virus persists but doesn’t cause severe damage.

There are a couple of obvious scenarios where it would be nice to be able to control TRegs. There’s a lot of interest in reducing TReg activity in cancer, such as with CTLA4 antagonists. There’s also a lot of interest in increasing TReg activity in organ transplants, and there have actually been a couple of cases where it’s seemed to have worked.

A recent paper in PNAS¹ offers steps toward a more general procedure, that could in theory lead to controlled, planned generation of TRegs for any antigen.

A key aspect of TRegs is that they are antigen-specific. They don’t randomly suppress immune responses; they identify particular antigens that should be tolerated, and shut off immunity to those antigens. That allows fine control over the response, but it also makes it harder to catch a TReg; T cells (not just TRegs) that recognize any particular antigen are very rare events, hiding in a blizzard of other specificities. What if you could force T cells for an antigen you choose to enter the TReg pathway?

This has already been done, in fact, but in a very artificial system — in mice with transgenic T cell receptors. These mice overwhelmingly express a single TcR in all of their T cells — there’s no snowflake in a blizzard problem, because the entire blizzard is made of identical flakes. Harold von Boehmer’s group has shown that you can drive these transgenic T cells into the TReg pathway by offering very, very low levels of antigen, under defined conditions, over a long period. ² The recent paper¹ shows that you can do the same thing in normal, non-transgenic, mice; and by doing this you can force graft tolerance. (They used female mice and drove tolerance to the male antigen H-Y antigen. The tolerized female mice then became tolerant of male grafts, while the control female mice rejected the male grafts.)

The key, at least for this particular protocol, seems to be to use very low dose antigen and “suboptimal” conditions (where “optimal” refers to conditions that drive conventional immune responses. The vocabulary of immune responses is really kind of misleading, because it’s focused on easily-measured responses like protection against viruses or graft rejection. Regulatory T cell responses are just as active, and probably are just about as common and important, but it’s hard to talk about them without giving the impression that they’re somehow passive, or abnormal, or defective).

One problem with moving this into the clinic is that you would need to know what the target antigen is, which in an outbred population like humans you do not know a priori. However, as bioinformatic and experimental techniques for identifying antigen peptides improve, it may become more practical to run this for patients before their transplants. The potential payoff would be very high, because you might be able to remove immunosuppression altogether:

If a procedure as simple as peptide infusion, which permits de novo induction of Tregs from mature T cells, prevents transplant rejection or GVHD, it could offer a realistic opportunity to induce tolerance to a variety of antigens such as allergens, transplantation antigens, and antigens causing autoimmunity while minimizing undesirable side effects often associated with general immunosuppression.

1. Verginis, P., McLaughlin, K.A., Wucherpfennig, K.W., von Boehmer, H., Apostolou, I. (2008). Induction of antigen-specific regulatory T cells in wild-type mice: Visualization and targets of suppression. Proceedings of the National Academy of Sciences, 105(9), 3479-3484. DOI: 10.1073/pnas.0800149105[↩][↩]
2. Kretschmer, K., Apostolou, I., Hawiger, D., Khazaie, K., Nussenzweig, M. C., and von Boehmer, H. (2005). Inducing and expanding regulatory T cell populations by foreign antigen. Nat Immunol 6, 1219-1227.
   Apostolou, I., and von Boehmer, H. (2004). In vivo instruction of suppressor commitment in naive T cells. J Exp Med 199, 1401-1408.[↩]

I have a very sporadic and idiosyncratic series in which I talk about “classic papers”, and in my idiosyncratic series Vic Engelhard’s paper on tyrosinase processing counts as a classic paper. It was one of the early indications that proteins in the ER must be degraded in the cytosol, and as such it’s one of a number of ways that antigen presentation has helped fundamental understanding of cell biology; but I think it hasn’t received as much recognition as it could have.

But perhaps I should begin at the beginning.

Proteolysis is a normal part of cell function. Proteins that are damaged, misfolded, or mis-translated, as well as proteins that have simply reached the end of their useful life, are degraded and converted to amino acids that can be recycled into new proteins. In the early- to mid-1990s, there were three general systems that were known to degrade proteins, depending on which subcellular compartment they were in:

• In the cytosol and nucleus, proteins are predominately degraded by proteasomes.
• Proteins taken up from the exterior of the cell can be degraded in acidic lysosomes
• Mitochondrial proteins are degraded by a number of proteases within the mitochondrion¹

That leaves an obvious gap. What happens to proteins that are in the endoplasmic reticulum (ER)? This is particularly relevant because the ER is a ferociously active site of protein synthesis, folding, and assembly; when any of those steps goes awry, the protein is supposed to be degraded, a process known as “quality control”. It was clear in the 1980s that proteins that failed quality control in the ER were degraded; in human cells, a well-known example was the cystic fibrosis transmembrane conductance regulator (CFTR), which folds inefficiently and is rapidly degraded². But it was not clear where the degradation happened (in the ER? The cytosol? Somewhere else?), and what proteases were responsible.

At first it was believed that “what happens in the ER stays in the ER” — ER proteins were degraded in the ER, by ill-defined proteases in that compartment. But I don’t think there was much enthusiasm for that belief, and basically, the field was a mess, as you can see from this introductory paragraph from the time:³

Other membrane proteins are also known to be degraded at the ER, but the process is poorly understood, and the responsible enzymes have not been identified. For example, some of these proteolytic events are ATP dependent, but some are not; some occur within the lumen while others take place on the cytoplasmic side; some exhibit inhibitor sensitivities characteristic of serine proteases, whereas others do not.

This was relevant to me as an immunologist, because viral glycoproteins (which, of course, are synthesized in the ER) are popular targets for cytotoxic T lymphocyte (CTL) recognition (a quick review of MHC class I antigen presentation is here). We knew in the early 1990s that most if not all CTL targets — even those derived from ER proteins — were generated in the cytosol. As I wrote in a 1996 review:⁴

Proteins targeted into the ER by signal sequences can also be presented on MHC I molecules. Since these molecules are cotranslationally transported into the endoplasmic reticulum, they might be expected to bypass hydrolysis in the cytosol. However, where analyzed, the presentation of most of these antigens is dependent on the TAP-transporter and on proteasome activity, and therefore the presented peptides are probably being generated in the cytosol.

The three obvious possible explanations were that either the putative glycoprotein never made it into the ER and was degraded as a mistargeted protein (Jon Yewdell would call that a “DRiP”; I can’t remember exactly when I heard him propose that first, but it was around that time); that the glycoprotein went in to the ER, got degraded there, and the peptides were first transfered to the cytosol; or that the glycoprotein was transfered from the ER to the cytosol and degraded there.

The simplest explanation, at the time, seemed to be the first one –  proteins never made it in to the ER, and were degraded in the cytosol. However, it wasn’t an explanation that we liked very much, for various reasons. Vic Engelhard’s paper (Remember Engelhard’s paper? This here’s a post about Engelhard’s paper) cleared that question up, at least for one epitope.

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.

His finding is simple enough to describe, although it relied on a technically very difficult mass spec analysis:

1. A peptide presented on MHC class I was derived from an ER protein (which they knew from its sequence);
2. the protein had actually gone into the ER, because it had been N-glycosylated, which only happens in the ER;
3. yet the peptide itself was probably generated in the cytosol, because enzymes that modified it are mainly found in the cytosol.

The most surprising and exciting part was the second point: Clear evidence that the protein had actually gone into the ER before the peptide was generated. ⁵ This wasn’t by any means definitive proof that ER proteins are degraded in the cytosol (a follow-up paper in 1998⁶ took it a bit further) but it certainly was suggestive.

If Vic’s paper had come out a year or two earlier, it would probably have made much more of a splash than it did, but in February of 1996 it was only a nose ahead of several more focused papers. The field had started to clear up in the mid-1990s, with some observations in yeast in 1993 and 1994 ⁷ and around 1995 moving into mammalian cells with (among others) the Jensen et al. paper I quoted above. ⁸  And later in 1996, the iceberg tipped over altogether, with a whole bunch of almost simultaneous papers that showed quite clearly that ER degradation wasn’t done by ER proteases at all, but was in fact performed by proteasomes. ⁹ All the papers demonstrated that there’s an export step before degradation: ER proteins that fail quality control are shunted out into the cytosol, where the proteasomes can grab onto them and chop them up. (The export step is still not all that well understood in molecular detail, though in 2007 it started to open up some, I think.)

Though Engelhard’s 1996 paper is reasonably widely cited (256 citations as I write this) it clearly didn’t have the impact on cell biology in general that it did on me, probably because it came out around the same time as a bunch of more specific papers. This blogpost is an attempt to give a bit more retroactive credit to a very nice example of logical reasoning from indirect evidence.

1. I believe that some, though not all, of the mitochondrial proteases were identified in the late 1980s/early 1990s, and that the broad outline of mitochondrial proteolysis was understood in the early 1980s (Desautels, M. and Goldberg, A. L. (1982) Liver mitochondria contain an ATP-dependent, vanadate-sensitive pathway for the degradation of proteins. Proc Natl Acad Sci USA 79 , pp. 1869-1873.). I don’t know all that much about mitochondrial proteolysis, though, so if someone wants to correct me, please do so.[↩]
2. For example, Ward C, Kopito R (October 14, 1994) Intracellular turnover of cystic fibrosis transmembrane conductance regulator. Inefficient processing and rapid degradation of wild-type and mutant proteins. J. Biol. Chem. 269.:25710-25718[↩]
3. Taken from Jensen TJ, Loo MA, Pind S, Williams DB, Goldberg AL, et al. (October 6, 1995) Multiple proteolytic systems, including the proteasome, contribute to CFTR processing. Cell 83:129-35. with references removed[↩]
4. York, I. A., and Rock, K. L. (1996). Antigen processing and presentation by the class I major histocompatibility complex. Annual review of immunology Annu Rev Immunol 14, 369-396.[↩]
5. Technical explanation: The mass spec analysis showed that the asparagine that is encoded in the DNA was actually an aspartic acid in the presented peptide; deglycosylating enzymes that were believed to only be present in the cytosol remove carbohydrates from Asn to generate Asp.[↩]
6. Mosse, C. A., Meadows, L., Luckey, C. J., Kittlesen, D. J., Huczko, E. L., Slingluff, C. L., Shabanowitz, J., Hunt, D. F., and Engelhard, V. H. (1998). The class I antigen-processing pathway for the membrane protein tyrosinase involves translation in the endoplasmic reticulum and processing in the cytosol. J Exp Med 187, 37-48.[↩]
7. (Sommer T, Jentsch S (September 9, 1993) A protein translocation defect linked to ubiquitin conjugation at the endoplasmic reticulum. Nature 365.:176-9.
   and
   Kölling R, Hollenberg CP (July 15, 1994) The ABC-transporter Ste6 accumulates in the plasma membrane in a ubiquitinated form in endocytosis mutants. EMBO J 13.:3261-71.[↩]
8. Jensen TJ, Loo MA, Pind S, Williams DB, Goldberg AL, et al. (October 6, 1995) Multiple proteolytic systems, including the proteasome, contribute to CFTR processing. Cell 83:129-35. [↩]
9. I may be missing some:
   Hampton RY, Gardner RG, Rine J (December 1996) Role of 26S proteasome and HRD genes in the degradation of 3-hydroxy-3-methylglutaryl-CoA reductase, an integral endoplasmic reticulum membrane protein. Mol Biol Cell 7.:2029-44.
   Werner ED, Brodsky JL, McCracken AA (November 26, 1996) Proteasome-dependent endoplasmic reticulum-associated protein degradation: an unconventional route to a familiar fate. Proc Natl Acad Sci U S A 93.:13797-801.
   Hiller MM, Finger A, Schweiger M, Wolf DH (September 20, 1996) ER degradation of a misfolded luminal protein by the cytosolic ubiquitin-proteasome pathway. Science 273.:1725-8.
   Qu D, Teckman JH, Omura S, Perlmutter DH (September 13, 1996) Degradation of a mutant secretory protein, alpha1-antitrypsin Z, in the endoplasmic reticulum requires proteasome activity. J Biol Chem 271.:22791-5.
   Wiertz EJ, Jones TR, Sun L, Bogyo M, Geuze HJ, et al. (March 8, 1996) The human cytomegalovirus US11 gene product dislocates MHC class I heavy chains from the endoplasmic reticulum to the cytosol. Cell 84.:769-79.[↩]

Ephemeroptera

One of the questions in antigen processing is what happens to peptides between the time they’re generated, and the time the they bind to MHC class I.

(The reason we care about peptides and MHC is that antiviral lymphocytes react with a complex of peptides and MHC class I, so this is a central point for antiviral immunity. Peptides are formed as a byproduct of normal protein degradation; an outline of the process, should you care, can be found here.)

In general, the peptides we’re interested in are produced by proteasomes. A protein (say, 500 amino acids long) enters the proteasome, the protein is chopped up, and peptides (between 3 and 30 amino acids long) come out. Almost all of those peptides are further chopped up, to produce amino acids - recycling and replenishing the amino acid pool for new protein synthesis. A small fraction of the peptides (perhaps between 0.01% and 1%), though, escape destruction and manage to bind to MHC class I. We would like to know more about that fraction of peptides, because they drive the lymphocyte attack on virus-infected cells. Why are they not destroyed — is it pure chance, or is there something special about the peptides that are not destroyed? How do they reach the MHC — is it chance again, just random diffusion, or is there some kind of specialized shuttle system that ferries the peptides to the proper subcellular location? Is there any active process modifying the peptides, to make them more (or less) suitable for binding MHC? And so on.

The problem is that it’s really hard to look at those peptides. Ideally, we’d like to grab samples of peptides at every point in the process: Exiting the proteasome, in transit, being degraded and processed, and so on. Then we could analyze what they’re like at each step, and develop a time course of modifications, interactions, and so on. But we can’t do that (yet), because it’s really difficult to measure peptides within a living cell.

A couple of years ago Jacques Neefjes (who always turns out cool papers) put some numbers on just how difficult is is.¹

There were a whole bunch of really cool things about this paper, but just focusing on one: Neefjes’ group came up with a way of measuring the rate of peptide destruction in living cells. They added a fluorescent tag to peptides in such a way that it would only fluoresce when the peptide was degraded; injected the tagged peptides into single cells; and measured (again in single cells) the rate at which the fluorescence appeared.

The injected peptides were destroyed with a half-life of 7 seconds. That is, a single cell can destroy hundreds of thousands, or millions, of peptides within a few seconds. (Most of this destruction, by the way, is performed by aminopeptidases, which are very abundant in cytosol.)

That’s not a long time, and it doesn’t give any individual peptide much chance to find its potential MHC binding partner. “A peptide will thus diffuse through the entire cell in 6 s and has to find TAP within this short period for translocation into the ER lumen.”

Why so fast? Why is the cell so worried about letting peptides hang about? Well, we presume this is because peptides are potentially very toxic. These peptides are generated, pretty much randomly, from active proteins. The peptides will therefore include short chunks of active protein domains, separated from any regulatory context; they could conceivably have biological activities by themselves. Also, you’d get hydrophobic chunks that could cluster into degradation-resistant clumps, if you let them accumulate, and it’s believed that such degradation-resistant complexes are themselves toxic. So you need to get rid of peptides fast, before they accumulate to form dangerous side-effects.

As a result, we antigen processing guys have to pretty much guess and use roundabout, indirect methods to measure peptides. Keeps us off the streets, I guess.

1. Reits, E., Griekspoor, A., Neijssen, J., Groothuis, T., Jalink, K., van Veelen, P., Janssen, H., Calafat, J., Drijfhout, J. W., and Neefjes, J. (2003). Peptide diffusion, protection, and degradation in nuclear and cytoplasmic compartments before antigen presentation by MHC class I. Immunity 18, 97-108 .[↩]

Antigen 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

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?

Interestingly, 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[↩]

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’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.

The 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 [↩]

I’ve talked several times (for example, here, here, and here) about predicting cytotoxic T lymphocyte (CTL) epitopes, and emphasized how hard it is (or, at least, how poor the tools are). Here’s an example of why it’s difficult.

(Quick review: CTL recognize virus-infected cells by screening small peptides that are bound to the class I major histocompatibility complex [MHC class I]. The peptides are created by destruction of proteins in the target cell. There’s a handy guide to antigen presentation here, if that helps put things into context.)

In my previous post on the subject, I listed a bunch of different factors that need to be incorporated in the predictions. Number 7 was “Binding to the MHC complex in the ER”, and I commented that peptide binding to MHC class I is probably the second-best understood step in the pathway (behind TAP transport, if you’re keeping score at home).

A paper from earlier this year¹ tried to identify CTL epitopes in influenza viruses. Lots of papers do this, but most don’t follow up with actual, complete tests — too expensive and difficult. Wang et al did the follow through.

They started by looking simply at binding to MHC class I alleles. Without going into details (they were looking for conserved epitopes that matched HLA supertypes, if anyone cares) they identified 167 peptides that they predicted should bind to the various MHC class I alleles; and then they tested them to see if they actually did bind. (They used NetMHC 3.0 ² to predict binding.)

Of the 167 predicted binders, 39 failed to bind altogether, and another 39 only bound very weakly. That leaves 89 peptides (just 53% of their tested pool) that were authentic binders.

Influenza viruses infecting cells of the trachea

Then, they tested to see if their peptides actually reacted with CTL from healthy donors. (They assumed that their healthy donors were immune to a influenza A — reasonable, but not a guarantee, so this is a particularly conservative test, I think.) Just 13 of their peptides were positive by this test (7.8% of their total predicted pool). Unexpectedly, two peptides that were non-binders triggered a response. Wang et al speculated that the very low affinity binding was enough for the CTL, but I wonder if this represented a contamination issue — CTL are famously sensitive, and it’s well known that tiny contaminating peptides in a synthetic prep are enough to trigger CTL, even if they’re barely detectable by other means.

The paper I’ve thought of as the record-holder for accuracy (if I’m being generous with their denominator) is Kotturi et al,³ whose prediction was correct for 25 of 160 potential peptides — about twice as good as the influenza predictions here. But Kotturi et al were dealing with just two MHC class I alleles, H-2D^b and H-2K^b, and those are very intensively-studied alleles. Wang et al. are not only looking at multiple alleles, they were using supertype approaches that allow them to cover almost all (>99%) of the population — a much more difficult prediction. To me, then, their predictions are remarkably successful.

But still: Just over 7% of their predictions were correct. And even limiting to prediction to a single step in the complex pathway — just looking at MHC class I binding of the peptides — they’re barely above 50% accuracy.

It’s a hard job. But I have to say that the field is progressing with impressive speed; these predictions are much more accurate than I would have expected five years ago.

1. Wang, M., Lamberth, K., Harndahl, M., Roder, G., Stryhn, A., Larsen, M. V., Nielsen, M., Lundegaard, C., Tang, S. T., Dziegiel, M. H., Rosenkvist, J., Pedersen, A. E., Buus, S., Claesson, M. H., and Lund, O. (2007). CTL epitopes for influenza A including the H5N1 bird flu; genome-, pathogen-, and HLA-wide screening. Vaccine 25, 2823-2831. [↩]
2. NetMHC is based on these three references — which I’m including as a note to myself: (1) Nielsen, M., Lundegaard, C., Worning, P., Hvid, C. S., Lamberth, K., Buus, S., Brunak, S., and Lund, O. (2004). Improved prediction of MHC class I and class II epitopes using a novel Gibbs sampling approach. Bioinformatics 20, 1388-1397 . (2) Nielsen, M., Lundegaard, C., Worning, P., Lauemoller, S. L., Lamberth, K., Buus, S., Brunak, S., and Lund, O. (2003). Reliable prediction of T-cell epitopes using neural networks with novel sequence representations. Protein Sci 12, 1007-1017 . (3) Buus, S., Lauemoller, S. L., Worning, P., Kesmir, C., Frimurer, T., Corbet, S., Fomsgaard, A., Hilden, J., Holm, A., and Brunak, S. (2003). Sensitive quantitative predictions of peptide-MHC binding by a ‘Query by Committee’ artificial neural network approach. Tissue Antigens 62, 378-384. [↩]
3. The CD8 T-Cell Response to Lymphocytic Choriomeningitis Virus Involves the L Antigen: Uncovering New Tricks for an Old Virus. Maya F. Kotturi, Bjoern Peters, Fernando Buendia-Laysa, Jr., John Sidney, Carla Oseroff, Jason Botten, Howard Grey, Michael J. Buchmeier, and Alessandro Sette. Journal of VIrology, May 2007, p. 4928–4940 [↩]

“Crow 104″ by Meggan Gould

The other day I heard a fascinating talk from Ned Walker, on the ecology and evolution of West Nile Virus in birds and mosquitos. Hopefully some of Ned’s cooler stuff should be published relatively soon, and I’ll talk about it then. In the mean time, Ned’s seminar reminded me of a really baffling observation I remembered reading about, in the mid 1990s, and prompted me to see what had happened to that story. As far as I can find, at present the state of the art regarding it seems to be (A) a big shrug, and (B) the suggestion that it’s an irrelevant side effect — the sort of thing that should make Larry Moran happy.

West Nile Virus (WNV) is a member of the flavivirus family, which are small (~11,000 bases) single-stranded RNA viruses that typically infect many species and often have an insect vector. For unknown reasons, in the mid 1990s WNV started to spread out of its traditional geographical regions (which are, you’ll never guess, the western Nile region of Africa) through much of Africa and Europe, and then hopped into North America and now has spread across the continent. It’s dangerous to humans (4269 cases, 177 fatalities in the USA in 2006) and much worse to birds — causing local extinctions of some species,¹ especially crows² (which, by the way, Ned³ thinks have got a bad rap as carriers — he thinks crows may be dead-end species, while many other species may be the actual routes of transmission).

Anyway, all this flavivirus talk reminded me of this previous observation.⁴ The reason I remember it was that it came out when I was particularly obsessed (even more so than I am today, believe it or not) with viral immune evasion; and the paper described exactly the opposite effect of what I was expecting.

Class I major histocompatibility complexes are recognized by cytotoxic T lymphocytes, which are generally agreed to be a major antiviral force; as such, many viruses target MHC class I and thereby block CTL recognition. ⁵ So it’s pretty common for virus-infected cells to show reduced levels of MHC class I.

West Nile Virus

Mullbacher’s paper showed that flaviviruses do the opposite: They specifically up-regulate surface levels of MHC class I. (As it happens, this had been described earlier, though I think only in specific cell types,⁶ but this was the first time I had run into it.) Mullbacher’s group argued, and still argues, that this is because the virus specifically increases peptide transport into the endoplasmic reticulum (in the 1995 paper, they guessed that a general leakiness might be the cause; later, they argue that it’s a specific effect on the TAP peptide transporter.⁷ ) Other groups believe that it’s a transcriptional effect, through several different (interferon-dependent and -independent) pathways.⁸ Still, the phenomenon seems real, significant, and robust.

How come? What’s the benefit to the virus to up-regulate MHC class I, thus making itself a better target to CTL?

Over the years (I found out, once I picked up on this story again last week) a bunch of different explanations have been proposed — resistance to natural killer cells, and so on; but none of them have been very convincing, and more recently, you get the sense that the researchers are just growing ti