Interesting!

I'd suggest anyone interested click through to the study "Does Data-Independent Acquisition Data Contain Hidden Gems? A Case Study Related to Alzheimer’s Disease": https://pubs.acs.org/doi/10.1021/acs.jproteome.1c00558 It includes some ways this isomerization might happen.

If you look at the typical shape of a neuron, it's clear they're particularly dependent on both energy and waste being shuttled around in the cell - so motor proteins running up and down microtubules (If this sounds like Greek, search for 'motor protein' on youtube (or just go here: https://www.youtube.com/watch?v=y-uuk4Pr2i8)

- it's beyond cool. Anyway, busted tau means microtubules collapse - and the cell dies.)

BTW/FWIW there's been some interesting work on transneuronal spread: https://academic.oup.com/brain/article/141/2/550/4775021?login=true

I haven't seen anything pointing to isomerization of Tau (