In August 2022, over 2 months ago, the august journal Science published an important paper on the genetics of PSP. I had difficulty wrapping my head around the complicated, cutting-edge technical aspects of the work, so I procrastinated in relaying it to you.
But last week, at CurePSP’s annual International Scientific Symposium in New York City, the paper’s senior author, Daniel Geschwind of UCLA, presented the work clearly enough for a non-lab person like myself and I now feel comfortable telling you that this paper is a real game-changer for PSP. The first author is a very junior member of Dr. Geschwind’s lab, Yonatan Cooper, a recent PhD who’s studying for his MD. The name of the paper is “Functional regulatory variants implicate distinct transcriptional networks in dementia.”
Until now, pretty much all we’ve known about the molecular genetics of PSP is that there are two places in MAPT (the gene encoding the tau protein), where one version of the gene is a little more common in people with PSP than in healthy people, and that there’s similar incrimination of a handful of other genes on other chromosomes. These variants are all in “markers,” rather than in the genes themselves. That is, a spot near the gene is the thing whose variant is statistically over-represented in those with the disease relative to healthy people. That doesn’t tell us for sure which of the dozens of genes in the vicinity of the marker is the actual disease-associated gene and it definitely doesn’t tell us the nature of the gene’s defect, or how it contributes to brain cell loss.
But now, the researchers at UCLA have analyzed the actual function of the genes in the chromosomal neighborhood of each of 9 markers associated with PSP. This is a huge undertaking, so they use a new technique called a massively parallel reporter assay (MRPA), which reveals gene expression. That is, it shows the types and amounts of proteins encoded by each of the 9 PSP-related genes and the other nearby genes incriminated by that marker.
The result was that the PSP-associated genes didn’t encode proteins themselves, but rather, served a regulatory function. The two genes most heavily associated with PSP were PLEKHM1 and KANSL1. Both are on chromosome 17, very near the MAPT gene. The disordered DNA sequences for PSP were transcription factor binding sites, the places in the gene where regulatory proteins can attach in order to do their job of adjusting the amount and composition of the protein encoded by that particular stretch of DNA.
So, what does this mean? To quote the paper, “These analyses support a mechanism underlying noncoding genetic risk, whereby common genetic variants drive disease risk in aggregate through polygenic cell type-specific regulatory effects on specific gene networks.” The English-language version is that they showed that the genetic contribution to PSP consists of variants in members of groups of genes that work together to regulate a specific cellular function. An individual with PSP simply has the bad luck to harbor enough such genes to get the disease process going.
The research paper shows that the gene variants themselves don’t directly encode a toxic version of a normal protein, as occurs in Huntington’s disease or other highly heritable brain degenerations. The toxic levels of tau in PSP must therefore be the indirect result of the disordered gene regulation, and as Dr. Geschwind emphasized, this and many other possible indirect effects of genetic variation contributing to the cause(s) of PSP remain to be discovered.
The fact that multiple genes must conspire together to produce the disease could explain why PSP is almost never familial: it’s very rare that more than one member of a family would have enough of the gene variants to accomplish any nefarious purpose. Someone with PSP would have had to inherit some variants from Mom and some from Dad, neither of whom had enough variants to cause the disease in themselves. Then, of course, one or more environmental exposures or experiences are probably also necessary but insufficient co-conspirators. But that wasn’t part of this project.
Enough for now. In a future post I’ll speculate with abandon on the implications for anti-PSP drug development.