Protein structural features predict responsiveness to pharmacological chaperone treatment for three lysosomal storage disorders

Abstract

Three-dimensional structures of proteins can provide important clues into the efficacy of personalized treatment. We perform a structural analysis of variants within three inherited lysosomal storage disorders, comparing variants responsive to pharmacological chaperone treatment to those unresponsive to such treatment. We find that predicted ΔΔG of mutation is higher on average for variants unresponsive to treatment, in the case of datasets for both Fabry disease and Pompe disease, in line with previous findings. Using both a single decision tree and an advanced machine learning approach based on the larger Fabry dataset, we correctly predict responsiveness of three Gaucher disease variants, and we provide predictions for untested variants. Many variants are predicted to be responsive to treatment, suggesting that drug-based treatments may be effective for a number of variants in Gaucher disease. In our analysis, we observe dependence on a topological feature reporting on contact arrangements which is likely connected to the order of folding of protein residues, and we provide a potential justification for this observation based on steady-state cellular kinetics.Author summaryPharmacological chaperones are small molecule drugs that bind to proteins to help stabilize the folded state. One set of diseases for which this treatment has been effective is the lysosomal storage disorders, which are caused by defective lysosomal enzymes. However, not all genotypes are equally responsive to treatment. For instance, missense mutants that are particularly destabilized relative to WT are less likely to respond. The availability of datasets containing responsiveness data for large numbers of mutants, along with crystal structures of the protein involved in each disease, make machine learning methods incorporating sequence-based and structural data feasible. We hypothesize that data from two diseases, Fabry and Pompe disease, may be useful for predicting responsiveness of variants in the related Gaucher disease. Results suggest that many rare variants in Gaucher disease could be amenable to existing drugs. Results suggest that drug responsiveness depends on protein topology is such a way that mutations in early-to-fold residues are more likely to be non-responsive to pharmacological chaperone treatment, which is consistent with a simple kinetic model of stability rescue. This study provides an example of how machine learning can be used to inform further studies towards personalized treatment in medicine.

Circuit topology predicts pathogenicity of missense mutations

“…In this study, a large number of inverse parallel relations was predictive of nonresponsiveness to pharmacological chaperone treatment. For a triosephosphate isomerase (TIM) barrel protein, the structure of which is contained in three lysosomal storage disorder proteins studied, residues with many inverse parallel relations were in fact early to fold, according to hydrogen‐deuterium exchange experiments, consistent with a model where late to fold residues are more likely to be rescued 49,51 . It was found that the dependence on the inverse parallel relation according to a simple decision tree is consistent with a simple kinetic model of folding, binding, and export.…”

“…This work relates to a recent study evaluating the importance of structural features in predicting and explaining drug responsiveness in lysosomal storage disorders. 49 In this study, a large number of inverse parallel relations was predictive of nonresponsiveness to pharmacological chaperone treatment. For a triosephosphate isomerase (TIM) barrel protein, the structure of which is contained in three lysosomal storage disorder proteins studied, residues with many inverse parallel relations were in fact early to fold, according to hydrogen-deuterium exchange experiments, consistent with a model where late to fold residues are more likely to be rescued.…”

“…This work relates to a recent study evaluating the importance of structural features in predicting and explaining drug responsiveness in lysosomal storage disorders 49 . In this study, a large number of inverse parallel relations was predictive of nonresponsiveness to pharmacological chaperone treatment.…”

“…The inverse parallel relation also shows substantial contribution to pathogenicity. According to models of folding, 30,49 residues with many inverse parallel relations are likely to be early to fold, indicating that they would be expected to disrupt rate of folding. The hypothesis that the relative importance of the inverse parallel and cross relations indicate an importance of folding kinetics and non‐native interactions in determining pathogenicity can be validated in the future using newly developed all‐atom models of (un)folding and misfolding 50 …”

“…The inverse parallel relation also shows substantial contribution to pathogenicity. According to models of folding, 30,49 residues with many inverse parallel relations are likely to be early to fold, indicating that they would be expected to disrupt rate of folding.…”