Mystery Rays from Outer Space

As everyone knows, the incidence of allergies and asthma has exploded over the past 50-odd years. As lots of people also know, while the reasons for this explosion isn’t known (and are probably complex) one of the popular concepts explaining this is the “hygiene hypothesis”. This was originally proposed way back in 1989:¹

These observations . . . could be explained if allergic diseases were prevented by infection in early childhood, transmitted by unhygienic contact with older siblings, or acquired prenatally . . . Over the past century declining family size, improved household amenities and higher standards of personal cleanliness have reduced opportunities for cross-infection in young families. This may have resulted in more widespread clinical expression of atopic disease.

In the nearly 20 years since, this hypothesis hasn’t been proved or disproved. There are quite a few interesting correlations, and the underlying biology seems to make a lot of sense, but as least as far as I know there’s been no smoking-gun study that makes an undisputable link. A Nature Medicine paper² from a couple of weeks ago adds a little more support to the hypothesis, and this one also holds out a distant hope of some kind of intervention as well. It’s long been known that parasitic worms — now rare in the West, but until recently a normal part of the human condition — induce an immune response that is broadly similar to a lot of allergic responses.

Melendez et al show that one class of parasitic worms make a protein that inhibits the anti-parasitic immune response. The protein, ES-62, does this by binding the pathogen pattern receptor molecule TLR4, thereby blocking a signalling pathway that ultimately leads to mast cell activation. This is presumably a parasite immune evasion molecule, analogous in concept to the many viral proteins that block TLR pathways. (The reason I say this is “presumably” a immune evasion molecule is that the other possibility is that the response is driven by the host — that this is more analogous to the way rodents develop a regulatory T cell response to persistent viruses, reducing harmful inflammatory diseases but allowing long-term infection with the virus.)

As the authors say, this is an exciting observation for two reasons. First, while not a smoking gun (that’s a rhetorical question in the title, OK?), it offers a mechanistic explanation for at least part of the hygience hypothesis. Second, the protein offers a handle for therapy of allergic diseases:

Suppression of mast-cell function by ES-62 offers a new explanation for the reason why people harboring worms of at least the filarial nematode type show reduced incidence of allergy, in spite of their elevated serum IgE. … By inhibiting mast-cell effector function, ES-62 offers a new potential therapeutic approach for diseases such as asthma, a medical problem of enormous importance in the developed world. Although ES-62 per se is unlikely to be used for treatment, enough is known about its structure and function to allow one to envisage the development of small, presumably phosphorylcholine-based, derivatives as drugs.

1. Strachan, D. P. (1989). Hay fever, hygiene, and household size. BMJ 299, 1259-1260.[↩]
2. Melendez, A. J., Harnett, M. M., Pushparaj, P. N., Wong, W. S. F., Tay, H. K., McSharry, C. P., and Harnett, W. (2007). Inhibition of Fc[epsi]RI-mediated mast cell responses by ES-62, a product of parasitic filarial nematodes. Nat Med 13, 1375-1381.[↩]

Malaria parasites in mosquito midgut

Why is it so hard to come up with a disproof (or a proof) of overdominant selection in MHC?

I’m starting kind of in mid-sentence here, because this is a continuation of a series of posts on MHC diversity. Briefly: The major histocompatibility complex region in vertebrates is extraordinarily diverse — a hundred times more variable (more alleles) than the average genomic chunk. Even populations that are otherwise inbred and lack diversity throughout their genome, rapidly evolve, or maintain, MHC diversity. There is clearly powerful evolutionary selection for this diversity, and there are several different explanations as to what this driver might be. The two most plausible explanations are frequency-dependent selection (in which rare alleles are selected simply because they are rare, and pathogens haven’t adapted to them) and overdominance, or heterozygote advantage (where individuals with diverse MHC regions, containing many alleles, are selected because they are more resistant to pathogens).

Overdominance was (as far as I know) the first mechanism put forward to explain MHC diversity. ¹ The concept is a simple one: If one MHC allele protects against disease by binding a certain set of peptides, then two alleles should protect against more diseases by cumulatively binding a larger set of peptides. A heterozygous individual should be more resistant to pathogens than individuals that are homozygous for either single allele.

This simple concept, though, turns out to be very difficult to test rigorously. Most importantly, several of the different predictions between overdominance and frequency-dependent selection depend on how the population evolves, over time; but when trying to test predictions, we are usually looking, more or less, at a static snapshot of evolution. In the static state, it’s much harder to differentiate between the two possibilities: Is the allele diversity we see a stable diversity (consistent with overdominance) or is it a dynamic diversity, with different alleles gaining and losing advantage as they become more or less frequent (consistent with frequency-dependent selection)?

True overdominance is explicitly not dependent on allele frequency. ² There are conditions (that are not true overdominance) in which heterozygotes will have a selective advantage over homozygotes, where the advantage is strictly dependent on allele frequency. Therefore … ³

overrepresentation of HLA heterozygotes among individuals with favorable disease outcomes (which we term population heterozygote advantage) need not indicate allele-specific overdominance. On the contrary, partly due to a form of confounding by allele frequencies, population heterozygote advantage can occur under a very wide range of assumptions about the relationship between homozygote risk and heterozygote risk. In certain extreme cases, population heterozygote advantage can occur even when every heterozygote is at greater risk of being a case than either corresponding homozygote.

There are a fair number of studies on more or less wild populations that have claimed to show evidence for overdominance, but few (if any) deal with the frequency problem. For that reason, most of the claims in the literature are at best consistent with overdominance, but are not proof of it.

A second complication is that individuals heterozygous at the MHC are quite likely to be heterozygous generally. How can specific effects of MHC heterozygosity be distinguished from a general heterozygote advantage? Again, this makes studies on wild populations hard to interpret cleanly.

There’s a third complication: Overdominance is most likely to be a factor in infections with more than one pathogen. ⁴

MHC-mediated resistance to a single pathogen is inherited as a dominant trait. This means that there will be no differences in susceptibility between a homozygote MHC allele or haplotype and a heterozygote carrying the focal allele plus a different one. Therefore, heterozygote advantage is difficult to detect in single pathogen challenges.

An exception to this might be when there’s technically a single pathogen, but it’s highly antigenically diverse — the obvious example being HIV, which mutates rapidly and regularly throws out antigenic variants during the course of infection. It’s interesting, then, that HIV infection is one of the situations where heterozygote advantage has been observed, ⁵ though I don’t think population allele frequency was taken into account in these studies. On the other hand, malaria is antigenically diverse as well, but experiments have not shown overdominance in that case.⁶

So we’re left with the difficult situation where you need to have fairly large numbers and multiple generations, in order to detect selection; yet you probably can’t use most natural populations to strictly the test the theory. Setting up a large and relatively diverse, yet well-controlled, lab animal population, and then infecting with multiple pathogens; or following a reasonably well-controlled field population; is a daunting task.

In the next post in this series I’ll mention a few cases where this has been done.

1. Doherty, P. C., and Zinkernagel, R. M. (1975). A biological role for the major histocompatibility antigens. Lancet 1, 1406-1409.[↩]
2. At least, that’s how I understand it; and I repeat that I’m not an expert. Anyone knowledgeable about this, feel free to jump in.[↩]
3. Lipsitch, M., C. T. Bergstrom, and R. Antia. 2003. Effect of human leukocyte antigen heterozygosity on infectious disease outcome: the need for allele-specific measures. BMC Med Genet 4: 2. [↩]
4. Wegner, K. M., M. Kalbe, H. Schaschl, and T. B. Reusch. 2004. Parasites and individual major histocompatibility complex diversity–an optimal choice? Microbes Infect 6: 1110-1116. [↩]
5. For example, Carrington, M., G. W. Nelson, M. P. Martin, T. Kissner, D. Vlahov, J. J. Goedert, R. Kaslow, S. Buchbinder, K. Hoots, and S. J. O’Brien. 1999. HLA and HIV-1: heterozygote advantage and B*35-Cw*04 disadvantage. Science 283: 1748-1752. [↩]
6. Wedekind, C., Walker, M., and Little, T. J. (2005). The course of malaria in mice: major histocompatibility complex (MHC) effects, but no general MHC heterozygote advantage in single-strain infections. Genetics 170, 1427-1430.[↩]

One of the hottest topics in immunology today is the regulatory T cell — thousands of publications, hundreds of reviews. But few if any of the reviews go into any detail on the history of the TReg, instead coughing nervously, shuffling their feet, and hastily pointing out some shiny distracting fact over there. This is probably partly because the authors think everyone already knows the whole sordid story, partly because they don’t want to waste review space on spilled milk, and a lot because the story is embarrassing for those involved.

TRegs are a type of T cell that, as you might guess, regulates the immune response — dampens reactions so that you’re not overwhelmed by the inflammatory response, and reduces the risk of autoimmunity. People and animals lacking TRegs have horrible, often fatal autoimmune diseases. TRegs were defined back in the mid 1990s or so, and the 2000s have seen an explosive growth in the field.

But well before that — starting in the mid-1970s or before — it was known that there were T cells that could regulate (or, as the term was then, “suppress”) immune responses, and in fact much of the basic biology was worked out then. Immunologists identified the suppressor cell and used mouse crosses to identify its mechanism of action. They were able to precisely localize within the genome a molecule that was critical for suppressor T cell activation. This was the “I-J” determinant,¹ and it was localized within the MHC region of the mouse genome. ² Indeed, based on the biology I-J looked like another member of the MHC family. A great deal of work went into characterizing the I-J determinant,³ including the development on monoclonal antibodies against I-J⁴ and preliminary biochemical characterization of the molecule.⁵ In spite of a minority of skeptics, it was a very exciting time.

So, um, it was a little awkward when I-J turned out, well, not to exist. The region that it had been mapped to was sequenced, and, er, there was nothing there. ⁶

We therefore conclude that the I-J gene is not formed by a DNA rearrangement between the I-A and I-E -subregions. … Our data suggest that the genes encoding I-J serologic determinants expressed by suppressor T cells do not map between the I-A and I-E subregions.

‘hem.

Here’s a chart of publications on I-J by year. See what happens in 1984? That is what a balloon popping looking like. For a few years after people still published stuff that had been in their pipelines, but no one was starting anything new, everyone in the field hurriedy left (mumbling to themselves) for greener pastures, and no one entered the field.

What went wrong? I have no idea, to be honest. The people who were working on I-J included a lot of people who are much smarter than I am (including one of my scientific grandfathers, whose Nobel Prize is only the start of his achievements); my schadenfreude includes a great deal of “That could have been me”.

The good news is that after the embarrassed pause, the underlying phenomenon itself turned out to be real and robust. In the mid-1990s people returned to the field (giving the cells a new name to cover over the I-J fiasco), and TRegs have taken off (the chart below shows TReg publications, in green, compared to the suppressor papers from earlier).

One lesson to be drawn from this, by the way, is the difference between science and pseudoscience. A lot of people had invested heavily in I-J. Yet when evidence disproved their hypothesis, the hypothesis was abandoned. What’s more, returning to the data with an open mind allowed the field to generate new hypotheses, test them, and when they turned out to work, to run with them. TRegs today are one of the most promising handles for a bunch of therapies. If this was pseudoscience, people would still be rummaging around the I-J field, attacking the people who disproved it, and rehashing old and useless experiments based on increasingly convoluted explanations for what turned out to be simply wrong.

1. For example, Murphy, D. B., Herzenberg, L. A., Okumura, K., Herzenberg, L. A., and McDevitt, H. O. (1976). A new I subregion (I-J) marked by a locus (Ia-4) controlling surface determinants on suppressor T lymphocytes. J Exp Med 144, 699-712.[↩]
2. For example, Greene, M. I., Pierres, A., Dorf, M. E., and Benacerraf, B. (1977). The I-J subregion codes for determinants on suppressor factor(s) which limit the contact sensitivity response to picryl chloride. J Exp Med 146, 293-296.[↩]
3. E.g. Murphy, D. B., Yamauchi, K., Habu, S., Eardley, D. D., and Gershon, R. K. (1981). T cells in a suppressor circuit and non-T:non-B cells bear different I-J determinants. Immunogenetics 13, 205-213.[↩]
4. Waltenbaugh, C. (1981). Regulation of immune responses by I-J gene products. I. Production and characterization of anti-I-J monoclonal antibodies. J Exp Med 154, 1570-1583.[↩]
5. Asherson, G. L., Watkins, M. C., Zembala, M. A., and Colizzi, V. C. (1984). Two-chain structure of T-suppressor factor: antigen-specific T-suppressor factor occurs as a single molecule and as separate antigen-binding and I-J+ parts, both of which are required for biological activity. Cell Immunol 86, 448-459.[↩]
6. Kronenberg, M., Steinmetz, M., Kobori, J., Kraig, E., Kapp, J. A., Pierce, C. W., Sorensen, C. M., Suzuki, G., Tada, T., and Hood, L. (1983). RNA transcripts for I-J polypeptides are apparently not encoded between the I-A and I-E subregions of the murine major histocompatibility complex. Proc Natl Acad Sci U S A 80, 5704-5708. Also, Davis, M. M., Cohen, D. I., Nielsen, E. A., Steinmetz, M., Paul, W. E., and Hood, L. (1984). Cell-type-specific cDNA probes and the murine I region: the localization and orientation of Ad alpha. Proc Natl Acad Sci U S A 81, 2194-2198.[↩]