Daily digest — 2026-10-02

Linker optimization is still mostly empirical, so orthogonal readouts of how architecture shapes the complex interface are welcome. Gas-phase data are not solution cooperativity, and I'd not rank degraders by them alone, but pairing them with cellular degradation could help rationalize why two warheads behave so differently. A preprint, so treat the conclusions as provisional until peer review.

Nav1.7 has burned a lot of programs on selectivity and translation, so a clean off-target panel plus knockout-neuron controls is what I'd look at first. Potency is still modest and the dose needed is high, so this is a starting scaffold rather than a candidate. The practical takeaway is MicroED: assigning chirality from nanocrystals of an early lead saves weeks of enantiomer chasing.

p300/CBP is a good test of when degradation earns its molecular weight: scaffolding functions that inhibitors leave intact are exactly what a degrader removes. The HAT-domain and bromodomain optimization stories are useful reading for anyone with a stubborn chromatin target. I'd read the degrader sections skeptically on paralog selectivity, since p300 and CBP are nearly identical.

Going after the substrate pocket rather than the ATP site is the sensible way to buy HSP70 selectivity, and D2B makes iterating on it cheap. The cell-line-selective killing is the part I'd treat cautiously: it could reflect UPR dependence or something off-target. Orthogonal engagement data like CETSA are encouraging; I'd want a close inactive analog next.

Most of us accept PROTAC bulk as a tax; fusing the E3 ligand into the warhead's cap group attacks that directly and is a transferable idea for other zinc-enzyme targets. The question I'd press is whether the shared motif compromises either binding event or cereblon neosubstrate selectivity. If in vivo degradation holds with better properties, the 'linkerless' direction deserves more attention.

Two gating layers, shuttle for entry and ROS cage for activation, is clever systems design, but it moves the problem from medicinal chemistry to conjugate CMC and PK. I'd look hard at the unmasked-PROTAC fraction in periphery and at the shuttle's dependence on aged-endothelium biology, which could narrow the patient window. A real step for CNS degraders if the in vivo data hold up.

This paper is the full med-chem story behind a compound that already has human data, which is rarer and more useful than most 'discovery of' papers. EP4 is crowded, so the interesting part is how the new series differentiates on PK and safety against earlier antagonists. I'd read it for the scaffold-hop logic and watch whether phase I translates into combination activity with checkpoint blockade.