That last post on vaccine reactions mentioned autotimmune disease several times, and that brings up a question that’s been outstanding for many decades now. If you look at the distribution of most autoimmune syndromes in the population, you find many obvious differences between male and female incidence rates. Guillain-Barré (discussed yesterday) is less common in women (about 35/65 overall) particularly in older patients, but the younger patients (less common overall) tend to be even or even slightly more female. Ankylosing spondylitis is slightly less common in women than men (45/55) and type I diabetes is about 50/50. But multiple sclerosis is about 75/25 towards women, rheumatoid arthritis is around 80/20, lupus and Hashimoto’s are about 90%, and Sjögren’s is even higher than that. I’m unaware of any autoimmune diseases that affect men over women to those degrees, and when you add up the total number of diseases and the total number of patients, it’s clear that women are far more affected.
Why is that? A lot of work has gone into that question over the years, and as this overview at Nature shows, two of the big obvious possibilities have a lot of evidence behind them: differences in hormone levels and the presence of two X chromosomes (and the genes on them). There are definite fluctuations in some of these diseases as hormone levels change (puberty, pregnancy, menopause), although the details are still being argued out. Which hormones are involved, and are the problems due to an excess of some or to a deficit of others? But there are also clear effects from the X chromosome. Indeed, if you engineer mice to have otherwise identical hormone levels but just bring in an extra X chromosome, they show up more frequently with conditions like lupus. The same thing happens in Klinefelter syndrome in humans, where a male ends up with an extra X chromosome.
So it’s not either/or; both of these factors contribute. And an underlying adaptation might be behind both – the immune system adaptations needed during pregnancy. A pregnant woman is devoting her metabolism to carrying a ever-larger and more complex fetus that is immunologically different from her own tissues, so the immune system clearly has a lot of delicate work to do. Evolutionarily, immune attacks on one’s developing progeny is surely a bad choice, but lowering the immune defenses too much for the mother during the same period is not going to work out well, either. You’re linking directly into the driveshaft of selection pressure either way, and as the Nature article makes clear, we are far from understanding the details of the solutions that have been worked out over the eons. Developmental biology and immunology are both notorious jungles of thickly overgrown pathways and interconnected systems, so their intersection is going to be extremely hard to comprehend.
But one theory is that since women in general spend less of their lives in pregnancy in modern times than they did during the vast stretches of evolutionary time, that the various fine-tunings of the immune response that go on during pregnancy (often mediated, it is believed, by placental tissue) are infrequent enough to throw the balance over towards autoimmunity. That makes for a neat story, and it appears that autoimmune diseases have increased in areas with lower fertility, but there’s a lot more to be worked out. The article describes studies that are looking at large repositories of medical data, trying to correlate autoimmune diagnoses with number of pregnancies, how young a person was at first childbirth, comparisons with women who did not have children, and so on. Of course, you can also ask why some autoimmune syndromes don’t respond in this direction as well: what is it about older men that makes them at greater risk for Guillain-Barré, or what is is about older women that protects them?
It would not surprise me if as we learn more about these things that we find that autoantibody status is a really significant variable in human health and disease. Obviously that’s the case with outright autoimmune disease, but I would have to think that there are a great many people who seem phenotypically average until the right (or wrong!) stimulus comes along. They might be more likely to develop rheumatoid arthritis as they age, or are more likely to have side effects like Sjögren’s syndrome in response to particular drugs, or are at greater risk for something like Guillain-Barré in response to an otherwise unremarkable respiratory viral infection. Perhaps (as that link above suggests) they are at greater risk for severe disease symptoms when infected by an ordinary transmissible virus as well. Many people have wondered about these situations and others, but the challenge has been to get enough understanding to do something about any of them in advance. I hope that we’re getting there.