Seidler et al. overcame a decade-long failure to crystallize tau's VQIINK segment by using micro-electron diffraction (MicroED), yielding a 1.5 Γ structure of KVQIINKKLD (PDB 5V5B) and a second polymorph of VQIINK (PDB 5K7N). The ten-residue structure revealed two steric-zipper interfaces (A and B), with shape complementarity Sc = 0.77 and buried area Ab = 168 Γ Β²βsubstantially more adhesive than VQIVYK's 0.72 and 75 Γ Β². Engineered K18 constructs confirmed functionally that 2xIN fibrillizes twice as fast as WT (75 vs 150 min half-max), while 2xVY is slowest (225 min) .
Structure-based design produced MINK/WINK (Phase 1, IC50 22.6/28.9 Β΅M) and Phase 2 inhibitors capping all three interfaces, improving potency ~20-fold (W-MINK 1.1 Β΅M). Notably, the established VQIVYK inhibitor TLKIVW failed to block tau40 seeding, and did not synergize with VQIINK inhibitors (MINK+TLKIVW IC50 52.2 Β΅M) .
A potential tension: cryo-EM of AD brain-derived tau fibrils shows the core comprises R3/R4, placing VQIINK (R2) in the fuzzy coat . The authors honestly address this, proposing the solvent-exposed fuzzy-coat VQIINK zippersβnot the buried coreβdrive seeding, and concede the alternative that in vitro fibril cores differ from AD cores. This remains unresolved and is the paper's weakest inferential link.
Confidence: High on structural and in vitro/in-cell claims (directly reported, replicated); low on the tau-strain polymorphism hypothesis, which the authors themselves label speculative.
Know what changed, what holds up, and what remains uncertain. Every Friday. No ads.