Coevolution and Disease‐Causing Mutations in Glucose‐6‐Phosphate Dehydrogenase (G6PD)

Glucose‐6‐phosphate dehydrogenase (G6PD) is essential for protection against oxidative stress. Over 160 different missense mutations in G6PD reduce the activity and/or stability of the enzyme, causing G6PD deficiency, which results in anemia and neonatal jaundice, and contributes to many other diseases. Severity of G6PD deficiency is categorized into four clinical classes based on G6PD activity in patient blood samples; however, the structural location and class of a mutation do not fully predict its biochemical effects. The goal of this work is to further understand how mutations affect different parameters of G6PD function. Towards this goal, sequence coevolution analysis (SCA) was used to identify coevolving groups (sectors) of G6PD residues. Previous applications of SCA to protein domains have identified coevolving sectors that control distinct parameters of protein function. Application of a novel SCA algorithm to G6PD identified several sectors that were enriched in disease‐causing mutations. Overall, the sectors select for known G6PD‐deficiency‐causing mutations and exclude predicted non‐phenotypic mutations. G6PD mutants from several sectors, as well as mutants not associated with any sectors, were biochemically characterized. Principal component analysis of the biochemical data showed that the mutants do not cluster by sector, but do cluster by clinical classification. We conclude that coevolution may be a good predictor of whether a G6PD mutation causes G6PD deficiency, but coevolving sectors do not correlate to specific biochemical parameters of G6PD. Support or Funding Information NSF Graduate Research Fellowship Program Stanford Interdisciplinary Graduate Fellowship (SIGF) Stanford Child Health Research Institute (CHRI) NIH