Back

Ligandability Assessment of the LAG-3 D1 Domain Enables Discovery of a Small-Molecule Inhibitor

Garcia-Vazquez, N.; Gabr, M.

2025-12-10 bioinformatics
10.64898/2025.12.07.692800 bioRxiv
Show abstract

LAG-3 is an emerging immune checkpoint whose extracellular D1 domain engages MHC class II through a broad protein-protein interface traditionally considered difficult to modulate with small molecules. To evaluate the ligandability of this region, we combined 100-ns molecular dynamics (MD) simulations, structure-based virtual screening, and biophysical and biochemical assays. MD sampling of the isolated D1 domain revealed a recurrent, transient cavity adjacent to the MHCII-binding surface. A representative pocket-open conformation was used to screen a [~]10,240-compound diversity library, yielding a single validated hit, N05. N05 bound the D1 domain with micromolar affinity measured by microscale thermophoresis (Kd = 59.2 {micro}M, TRIC/MST channel) and by spectral-shift detection (Kd = 56.1 {micro}M), and it partially inhibited the LAG-3/MHCII interaction (EC50 = 42.9 {micro}M; maximal inhibition [~]76%). A 30-ns MD simulation of the LAG-3-N05 complex showed stable ligand engagement within the MD-identified cavity and consistent stabilizing interactions with residues forming the pocket. These results demonstrate that the LAG-3 D1 domain possesses an accessible, dynamically formed binding site capable of accommodating small molecules, providing a structural and biophysical foundation for future exploration of LAG-3 ligandability. Graphical Abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=151 SRC="FIGDIR/small/692800v1_ufig1.gif" ALT="Figure 1"> View larger version (32K): org.highwire.dtl.DTLVardef@177b446org.highwire.dtl.DTLVardef@1dc9efdorg.highwire.dtl.DTLVardef@14a2d6aorg.highwire.dtl.DTLVardef@ebca69_HPS_FORMAT_FIGEXP M_FIG C_FIG Overview of the computational and experimental workflow used to identify and validate a small-molecule binder of the LAG-3 D1 domain.

Published in Journal of Molecular Graphics and Modelling · training set

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.