The striking homology of the Mycobacterium tuberculosis mammalian cell entry operons (mce1, mce2, mce3 and mce4) with other mycobacterial species and the proposed role of the mammalian cell entry protein 1A (Mce1A) of M. tuberculosis to facilitate invasion of host cells have led us to look into the finer details of these proteins in order to better understand their structure–function relationship.

We performed sequential alignments and secondary structure predictions of Mce1A, Mce2A, Mce3A and Mce4A, and compared these results with results from homology modeling by fold prediction and threading.

Sequential alignments showed that Mce1A and Mce2A are highly homologous, close to 70%, while the other combinations gave only about 30% similarities. The major parts of the proteins aligned without gaps and there were striking similarities by secondary structure predictions indicating that the proteins would have similar folds and to be ⍺/β proteins like the previously reported Mce1A model based on Colicin N. Fold prediction showed that the best templates for Mce2A were substrate-binding domain of DnaK and slow processing precursor penicillin acylase from Escherichia coli while the ⍺-domains of Mce3A and Mce4A could both be modeled using the cytoplasmic domain of serine chemotaxis receptor as template.

Although different templates had to be used to model the MceA proteins, functional information may be derived that is relevant for their overall function in M. tuberculosis. The β-domain is probably involved in binding with the receptors on target cells while the ⍺-domain is more likely to be involved in pore formation. As predicted from the folds, Mce3A and Mce4A model structures indicate a lipid bound conformation and therefore may be required in signaling events of the mammalian cell entry process.