A docking model of dapsone bound to HLA-B*13:01 explains the risk of dapsone hypersensitivity syndrome

Dapsone (4,4'-diaminodiphenylsulfone) has been widely used for the treatment of infections such as leprosy. Dapsone hypersensitivity syndrome (DHS) is a major side effect, developing in 0.5-3.6% of patients treated with dapsone, and its mortality rate is ∼10%. Recently, human leukocyte antigen (HLA)-B*13:01 was identified as a marker of susceptibility to DHS.

To investigate why HLA-B*13:01 is responsible for DHS from a structural point of view.

First, we used homology modeling to derive the three-dimensional structures of HLA-B*13:01 (associated with DHS) and HLA-B*13:02 (not so associated despite strong sequence identity [99%] with HLA-B*13:01). Next, we used molecular docking, molecular dynamic simulations, and the molecular mechanics Poisson-Boltzman surface area method, to investigate the interactions of dapsone with HLA-B*13:01 and 13:02.

We found a crucial structural difference between HLA-B*13:01 and 13:02 in the F-pocket of the antigen-binding site. As Trp95 in the α-domain of HLA-B*13:02 is replaced with the less bulky Ile95 in HLA-B*13:01, we found an additional well-defined sub-pocket within the antigen-binding site of HLA-B*13:01. All three representative docking poses of dapsone against the antigen-binding site of HLA-B*13:01 used this unique sub-pocket, indicating its suitability for binding dapsone. However, HLA-B*13:02 does not seem to possess a binding pocket suitable for binding dapsone. Finally, a binding free energy calculation combined with a molecular dynamics simulation and the molecular mechanics Poisson-Boltzman surface area method indicated that the binding affinity of dapsone for HLA-B*13:01 would be much greater than that for HLA-B*13:02.

Our computational results suggest that dapsone would fit within the structure of the antigen-recognition site of HLA-B*13:01. This may change the self-peptides that bind to HLA-B*13:01, explaining why HLA-B*13:01 is a marker of DHS susceptibility.