Abstract
CFTR, whose failure causes cystic fibrosis, is a chloride channel, which belongs to the ATP binding cassette (ABC) transporter family. Like other ABC proteins, CFTR consists of two halves, each containing a cytosolic nucleotide-binding domain (NBD1, NBD2) and a transmembrane spanning domain (TMD1, TMD2). ATP binding and hydrolysis at the NBDs control the CFTR gate, presumed to be in the TMDs. It remains unclear precisely how the NBD/TMD coupling is mediated.We used correlation analysis to identify possible pairs of energetically coupled residues which might mediate direct interactions between the NBDs and the TMDs. First, since CFTR belongs to a subgroup of ABCs, which contain two very divergent NBDs, we constructed an alignment containing only similar, asymmetric transporters. We implemented 5 different correlation algorithms. The major difficulty with correlation analysis is the prevalence of false positives due to non-independence of sequences resulting from evolutionary constraints. Only one algorithm corrects for non-independence by referring to an inferred phylogenetic tree, resulting in a smaller output. The other 4 methods assume independence and assign a score to every possible pair of alignment positions. Only pairs scoring in the top 1% were included.To select candidate pairs at the NBD/TMD interface we looked for pairs found with more than one algorithm and separated by less than ∼15A on a Sav1866 based homology model of CFTR. We also checked the distributions of amino acids at the two positions on the inferred phylogenetic tree, to eliminate correlations clearly due to evolutionary branching. We identified several pairs which are close to sites which can be cross-linked after cysteine substitution. Functional characterization of the effects of mutations on single-channel kinetics is underway.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have