Research Article
Yoojin Hong, Dimitra Chalkia,
Abstract
One of the major challenges in the genomic era is a nnotating structure/function to the vast quantities of se- quence information now available. Indeed, most of t he protein sequence database lacks comprehensive an nota- tion, even when experimental evidence exists. Furth er, within structurally resolved and functionally a nnotated protein domains, additional functionalities contain ed in these domains are not apparent. To add furthe r complica- tion, small changes in the amino-acid sequence can lead to profound changes in both structure and func tion, underscoring the need for rapid and reliable method s to analyze these types of data. Phylogenetic prof iles pro- vide a quantitative method that can relate the stru ctural and functional properties of proteins, as we ll as their evolutionary relationships. Using all of the struct urally resolved Src-Homology-2 (SH2) domains, we de mon- strate that knowledge-bases can be used to create s ingle-amino acid phylogenetic profiles which reliab ly anno- tate lipid-binding. Indeed, these measures isolate the known phosphotyrosine and hydrophobic pockets a s inte- gral to lipid-binding function. In addition, we det ermined that the SH2 domain of Tec family kinases b ind to lipids with varying affinity and specificity. Simulating mutations in Bruton’s tyrosine kinase (BTK) that ca use X-Linked Agammaglobulinemia (XLA) predict that these mutatio ns alter lipid-binding, which we confirm experiment ally. In light of these results, we propose that XLA-caus ing mutations in the SH3-SH2 domain of BTK alter li pid- binding, which could play a causative role in the X LA-phenotype. Overall, our study suggests that the number of lipid-binding proteins is drastically underestimate d and, with further development, phylogenetic profi les can provide a method for rapidly increasing the functio nal annotation of protein sequences.