As many will know, there is a whole set of what are termed “polyglutamine repeat disorders“, which themselves are a large part of a bigger set of trinucleotide repeat disorders. Huntington’s disease is a well-known one: the gene (HTT) for the huntingtin protein ends up with a series of CAG nucleotide repeats, which after translation to protein gives it a long set of glutamine residues. It’s been known for a long time that the severity of the disease seems to track with the number of glutamine residues: if there are fewer than 35 in a row, a person will be totally asymptomatic, with fewer than 28 being considered completely normal. 36 to 40 repeats takes a patient into a variable zone where the disease will show effects, but the age of onset and their severity will vary. And greater than 40 is full-blown Huntington’s.
The way these form seems to be slippage during DNA copying or repair, through several mechanisms. There are many interesting and suggestive features to all these diseases: for one thing, you don’t see “dinucleotide repeat” syndromes, probably because that messes up the reading frame. Trinucleotide repeats form all sorts of loops and diversions in DNA, but it’s at least still (marginally) functional. Meanwhile, CAG is apparently the most common of the repeats, and the polyglutamines that result are thought to be likely to aggregate (leading to disease pathology). It seems that most of the repeat disorders, though, CAG or not, have approximately that 35-repeat threshold for trouble. But there’s a separate category of much larger repeats, this time showing up outside protein-coding regions, that are also associated with things like Friedrich’s ataxia. But in most all of these, the phenotypic trouble tends to show up in very specific cell types and lineages for each disease (even though these are somatic mutations), for reasons that are not well understood. Many of the trinucleotide repeat diseases have been documented as getting slightly worse in each generation as well (“anticipation”) as the repeats grow larger.
A lot of the work in this field has focused on the proteins and their polyglutamine residues, which is natural enough – that’s generally the “business end” of a lot of genetic diseases, of course (dysfunctional proteins). But there have been clues over the years that at least some of the trouble starts back at the properties of the DNA. That 35-repeat threshold, for example, could be argued either way, but it’s for sure that these repeats can lead to DNA hairpins, triple-helix snarls, unwound regions (sometimes with RNA joining in on one strand, a so-called “R loop”, and more. Now this new paper puts all the evidence together in one place that it’s the CAG problem and not so much the glutamine problem.
They find that the onset of disease for a specific repeat length turns out to be correlated with somatic variations in a number of DNA-repair enzymes, for one thing. And there’s the recent discovery of the details of a spinocerebellar ataxia disease that has CAG repeats but whose protein product doesn’t express the glutamines at all. The etiology is still quite messy, but it’s mostly back at the DNA level:
Ultimately, the disease presentation in any of the trinucleotide repeat disorders may depend on a combination of (1) the rate at which CAG repeats expand in the critical target cell type(s), (2) the degree to which modifiers act to influence expansion rate, (3) the threshold somatic expansion size for cellular damage to occur, (4) the mechanism, polyglutamine toxicity or other, by which that somatically expanded repeat causes the cellular damage, and (5) the degree to which the damage mechanism is influenced by modifiers.
That leaves a place for polyglutamine to cause trouble, but its place as the primary driver of these diseases apparently needs to be given up. Splicing, translation, mRNA effects – there are plenty of places for the wheels to come off back before you make the aberrant protein. People have mostly studied DNA repair and maintenance enzyme systems in the context of oncology, but this paper argues that they should be targets for the trinucleotide repeat diseases as well. But you’d want to start early – it’s already known that the duration of the overt-disease phase seems to be independent of the CAG repeat length in general, which suggests that the mechanisms for disease onset are different from the ones that are triggered once it’s underway. But a new approach to therapy for these conditions is definitely worth paying attention to. As it is, almost all of them have no treatments available at all, so if there’s progress to be made by building on the work done in oncology, then so much the better.