ApoE and the Coronavirus

 BY DEREK LOWE

Apolipoprotein E has generated many surprises over the years. I was working in the Alzheimer's field when it was discovered that a genetic variant of this one, APOE4, is a significant risk factor for Alzheimer's. There was a lot of speculation at the time for why this might be so, because it's not an obvious connection. ApoE carries cholesterol in the blood, and no one had really linked lipid handling to Alzheimer's at the time. Thirty years of work have gone into tracking the details down, and only recently it look like we might know the details: if this hypothesis is correct, the lipoprotein variant leads to unusual lipid profiles in glial cells in the brain, which impairs their function. (More recently, several disruptions in lipid pathways have been linked in one way or another to dementia - see the references in that linked paper).

And Alzheimer's isn't the only disease that is affected by the ApoE4 variant. Lipoprotein behavior is important in the immune system as well, which is one of the reasons that it's so hard to untangle mechanisms in this area. Atherosclerosis, for example, most certainly has a connection with cholesterol trafficking, but it also has a strong connection with inflammation and the immune response to the arterial plaques. These things can have large clinical and public health implications, because while most of the human population has the canonical form of the protein (ApoE3), about 40% of us have one copy of either the APOE2 or APOE4 gene to go along with it. And about 3% of the world population is homozygous for one of those two variant forms, which adds up to a lot of people.

It now looks like these have a connection to the coronavirus pandemic as well. This paper examines mouse models of coronavirus infection, using mice that have had the various human forms of the protein knocked in and exposing them to a mouse-adapted form of SARS-CoV-2. And it turns out that the results are quite different -  the ones with ApoE2 and ApoE4 proteins had notably worse courses of disease and lower survival rates. The outcomes were particularly severe in the male mice, where 100% of the ApoE4 homozygotes died (as opposed to 30% death in the corresponding ApoE3 ones). The disease progressed more quickly in the variant mice, with higher viral loads, and a close look at their immune cell profiles strongly suggests that an impaired immune response to the virus is one of the factors. This (as with human patients with bad outcomes) seems to be part of the problem, with too much immune response (perhaps because other pathways are diminished) doing some of the damage all by itself. But just as mentioned above, working out the exact mechanisms underlying this problem will take a lot of effort.

All this is very suggestive of what goes on in human infections, but although animal models of infectious disease are in many ways about as good as we can get in this work, they can still only take you so far. So the authors here make the connection to human clinical data in the UK. As before, they find that older men are at the most risk for bad outcomes, but when they brought in data from the UK Biobank, it showed that homozygous APOE4 patients had a twofold higher risk of death on top of that. (APOE2 homozygotes had a trend toward increased mortality, but it did not reach statistical significance). The APOE4 result stood up to a number of statistical tests (looking at patients of different genetic backgrounds, adjusting for age and ApoE effects on longevity, etc.) So it appears that ApoE4 homozygosity is the very opposite of a gift that keeps on giving - it's associated not only with increasing risk in chronic diseases such as Alzheimer's, but with increasing risk in response to acute ones like coronavirus infection. Knowing the ApoE status of coronavirus patients, particularly older males, could provide an early warning.

https://www.science.org/content/blog-post/apoe-and-coronavirus