P504L
Category 3/4 — Most DruggablePathogenicTransmembrane · predictedEditorialProline → Leucine at position 504 inside wolframin's sixth transmembrane helix (TM6). ClinVar Pathogenic with the broadest clinical spectrum in the gene — Wolfram syndrome 1, Wolfram-like syndrome, DFNA6 hearing loss, Cataract 41, type 2 diabetes. AlphaMissense 0.797, DynaMut2 ΔΔG -0.56 kcal/mol (destabilising). One of two adjacent variants (with C505Y) characterizing a vulnerable TM6 region.
Interactive 3D Structure
Bond changes · DynaMut2 interaction analysis
| Interaction type | Wild-type partner | Mutant partner | Status |
|---|---|---|---|
| Hydrogen bond | L484 | — | Lost |
| Hydrogen bond | S502 | S502 | Preserved |
| Hydrogen bond | L507 | L507 | Preserved |
| Hydrogen bond | Y508 | Y508 | Preserved |
| Polar contact | S502 | S502 | Preserved |
| Polar contact | L507 | L507 | Preserved |
| Polar contact | Y508 | Y508 | Preserved |
| Van der Waals | S502 | — | Lost |
| Van der Waals | P885 | — | Lost |
| Hydrophobic | L468 | L468 | Preserved |
| Hydrophobic | P885 | P885 | Preserved |
Lost / gained / preserved interatomic contacts at the variant residue, from the DynaMut2 (Arpeggio) interaction analysis of the wild-type and energy-minimized mutant structures.
Computational Predictions
Clinical Evidence
Observed at very low frequency in gnomAD.
Structural Context
Position 504 sits inside TM6, immediately adjacent to C505 (Atlas card adjacent). The AlphaFold model places P504 within 5 Å of VAL503 (2.5 Å), CYS505 (2.5 Å), SER502 (4.1 Å), PRO885 (4.1 Å, from TM11 — same cross-helix contact as C505/P885), LEU507 (4.3 Å), and LEU506 (4.5 Å). The proline at position 504 occupies the same structural micro-environment as the C505 next to it, and both contact PRO885 in TM11 at ~4.1 Å.
The wild-type proline at 504 plays a deliberate kinking role in TM6. Just as P885 kinks TM11 (see P885L Atlas card), P504 introduces a controlled helix bend in TM6. This is structurally remarkable: TM6 contains a proline-induced kink at position 504, TM11 contains a proline-induced kink at position 885, and the two kinks face each other across the membrane at the C505/P885 interface. The geometry suggests an intentional structural register that the protein's evolutionary design depends on.
Replacing the TM6 proline with leucine removes the helix kink — TM6 becomes more linear than the wild-type. The PRO885 in TM11 remains in place, but its docking partner across the helix interface has changed shape. The TM6-TM11 register slips. DynaMut2 reports a |ΔΔG| of 0.56 kcal/mol — the fold absorbs it, but the geometric relationship across the helix interface is materially perturbed.
The clinical breadth of P504L (five documented phenotypes spanning both AD and AR Wolfram presentations) is striking given the AlphaMissense score of 0.797 is the lowest of any variant in this batch. The disconnect suggests AlphaMissense's training is conservative about proline substitutions; the clinical evidence is overwhelming.
Druggability Assessment
The mechanism is removal of a deliberate TM6 helix kink, slipping the TM6-TM11 register at the C505/P504/P885 interface. The therapeutic strategy is geometric: a small molecule that stabilizes the TM6-TM11 packing in the wild-type kink geometry, ideally engaging both sides of the interface (TM6 and TM11) simultaneously.
Compare with P885L (Atlas card adjacent) for the reciprocal mechanism, and with C505Y for an alternative TM6 substitution at the same interface. Three Atlas variants (P504L, C505Y, P885L) converge on a single therapeutic target.
Why this matters
Feed this card to Wolfram Intelligence
Download the P504L PDF below and upload it to Wolfram Intelligence to generate therapeutic-strategy proposals — guanidinium mimetics, sigma-1 agonist docking, NAC thiol-capping. NAC is already on the bench-testing list.