Structural bioinformatics enhances the interpretation of somatic mutations in KDM6A found in human cancers

Young-In Chi; Timothy J Stodola; Thiago M De Assuncao; Elise N Leverence; Brian C Smith; Brian F Volkman; Angela J Mathison; Gwen Lomberk; Michael T Zimmermann; Raul Urrutia

doi:10.1016/j.csbj.2022.04.028

Structural bioinformatics enhances the interpretation of somatic mutations in KDM6A found in human cancers

Comput Struct Biotechnol J. 2022 Apr 28:20:2200-2211. doi: 10.1016/j.csbj.2022.04.028. eCollection 2022.

Authors

Young-In Chi^{1

2}, Timothy J Stodola¹, Thiago M De Assuncao^{1

2}, Elise N Leverence¹, Brian C Smith³, Brian F Volkman³, Angela J Mathison^{1

2}, Gwen Lomberk^{1

2

4}, Michael T Zimmermann^{1

3

5}, Raul Urrutia^{1

2

3

5}

Affiliations

¹ Genomic Sciences and Precision Medicine Center (GSPMC), Medical College of Wisconsin, Milwaukee, WI, United States.
² Division of Research, Department of Surgery, Medical College of Wisconsin, Milwaukee, WI, United States.
³ Department of Biochemistry, Medical College of Wisconsin, Milwaukee, WI, United States.
⁴ Department of Pharmacology and Toxicology, Medical College of Wisconsin, Milwaukee, WI, United States.
⁵ Clinical and Translational Sciences Institute, Medical College of Wisconsin, Milwaukee, WI, United States.

Abstract

The histone demethylase KDM6A has recently elicited significant attention because its mutations are associated with a rare congenital disorder (Kabuki syndrome) and various types of human cancers. However, distinguishing KDM6A mutations that are deleterious to the enzyme and their underlying mechanisms of dysfunction remain to be fully understood. Here, we report the results from a multi-tiered approach evaluating the impact of 197 KDM6A somatic mutations using information derived from combining conventional genomics data with computational biophysics. This comprehensive approach incorporates multiple scores derived from alterations in protein sequence, structure, and molecular dynamics. Using this method, we classify the KDM6A mutations into 136 damaging variants (69.0%), 32 tolerated variants (16.2%), and 29 variants of uncertain significance (VUS, 14.7%), which is a significant improvement from the previous classification based on the conventional tools (over 40% VUS). We further classify the damaging variants into 15 structural variants (SV), 88 dynamic variants (DV), and 33 structural and dynamic variants (SDV). Comparison with variant scoring methods used in current clinical diagnosis guidelines demonstrates that our approach provides a more comprehensive evaluation of damaging potential and reveals mechanisms of dysfunction. Thus, these results should be taken into consideration for clinical assessment of the damaging potential of each mutation, as they provide hypotheses for experimental validation and critical information for the development of mutant-specific drugs to fight diseases caused by KDM6A dysfunctions.

Keywords: 2OG, 2-oxoglutarate; COSMIC, Catalog of somatic mutations in cancer; Cancer; DV, Dynamics variants; Epigenetic regulator; Genomic variation; HAT, Hydrogen atom transfer; HMT, Histone methyltransferase; Histone demethylase; JmjC, Jumonji C domain; KDM6A; KDM6A, Histone lysine(K)-specific demethylase 6A; Kabuki syndrome; MD, Molecular dynamics; Molecular dynamics; Mutational impact analysis; PDB, Protein data bank; Protein structure; RMSD, Root mean square deviation; RMSF, Root mean square fluctuation; Rg, Radius of gyration; SASA, Solvent-accessible surface area; SDV, Structural & dynamics variants; SNP, Single nucleotide polymorphism; SV, Structural variants; TCGA, The Cancer Genome Atlas; TPR, Tetratricopeptide repeat; VUS, Variant of uncertain (unknown) significance; dbSNP, Single nucleotide polymorphism database; gnomAD, genome aggregation database.

Abstract

Grants and funding