Protein Deimmunization with a Generative Diffusion Model
Protein immunogenicity is a major cause of biologic drug failure, leading to the production of anti-drug antibodies (ADA) that accelerate drug clearance, rejection of repeated doses, abolished efficacy, and sometimes life threatening anaphylaxis in patients. These responses constrain the clinical use of synthetic protein therapeutics from non-human origin. To efficiently deimmunize protein…
Protein immunogenicity often leads to biologic drug failure, causing anti-drug antibodies that decrease drug efficacy and can even cause life-threatening anaphylaxis. Synthetic protein therapeutics from non-human sources face significant challenges due to this immunogenicity. Researchers have created a generative discrete diffusion model called MESSI to tackle this issue.
MESSI takes protein structures as input and produces sequences that retain structure while eliminating peptides that bind to major histocompatibility complex class II molecules. This model designs sequences with high structural recovery and can target specific MHC molecules, achieving lower epitope prediction rates compared to traditional post-hoc filtering methods.
In experiments, MESSI was used to deimmunize luciferase and a mini-binder designed from scratch against PD-1. The resulting variants showed no predicted epitopes and improved activity compared to the original proteins. Additionally, MESSI was coupled with evolution guidance to deimmunize uricase, a clinically approved enzyme with severe immunogenicity. The deimmunized variants exhibited higher catalytic activity than their natural counterparts.
By enabling the generative design of functional proteins with minimized immunogenicity, MESSI has the potential to make protein therapeutics safer and more effective.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
