Predicting functionally informative mutations in Escherichia coli BamA using evolutionary covariance analysis.

Area of Science:

• Molecular biology
• Bioinformatics
• Structural biology

Background:

• The outer membrane protein BamA is crucial for assembling beta-barrel proteins in Escherichia coli.
• BamA undergoes conformational changes during assembly, suggesting complex residue interactions.
• Genetic analysis of BamA mutants is challenging due to a lack of efficient selection methods.

Purpose of the Study:

• To employ a bioinformatic approach, specifically direct coupling analysis (DCA), to identify functionally important residues in BamA.
• To validate the effectiveness of DCA in predicting residue proximity in large beta-barrel proteins.
• To investigate the roles of specific BamA residues (R661 and D740) in protein function and stability.

Main Methods:

• Direct Coupling Analysis (DCA) was applied to the BamA paralog FhaC to assess its predictive power.
• A novel structured prior was incorporated into the DCA empirical correlation matrix to improve accuracy.
• DCA was then used to analyze BamA, identifying candidate residues for mutagenesis.

Main Results:

• DCA accurately predicted residue proximity in FhaC, with an improved true positive rate of 0.88 after incorporating a structured prior.
• DCA implicated residues R661 and D740 in BamA's function.
• Mutations R661G and D740G resulted in outer membrane permeability defects and destabilized the BamA beta-barrel.
• Synthetic phenotypes and cross-suppressors indicated a functional link and potential direct interaction between R661 and D740.

Conclusions:

• Direct coupling analysis is a powerful tool for identifying functionally interacting residues in complex protein systems like BamA.
• The identified residues R661 and D740 play critical roles in BamA function and beta-barrel stability.
• This bioinformatic approach offers a valuable strategy for mutagenesis in systems lacking genetic selections and for studying dynamic proteins.