Sequence homology between bacteriorhodopsin and G-protein coupled receptors: exon shuffling or evolution by duplication?

Bacteriorhodopsin (BR) is a membrane protein of known structure, widely used for the homology modeling of G-protein-coupled receptors (GPCR). The observation of apparently transposed sequence similarities between some of the helical domains of BR and GPCR has led to the suggestion that exon shuffling may have occurred in the later evolution of GPCR, which would necessitate a different folding pattern for the seven transmembrane helices of GPCR. An alternate hypothesis is that duplication occurred in the evolution of an ancestral gene, such that helices 5-7 originated as duplicates of helices 1-3, leading to intragenic as well as intergenic similarities between helices 1-3 and 5-7 of BR and various GPCR. Analyses of GPCR and BR sequences suggest that such a duplication may have occurred; symmetry within the BR structure is also consistent with homology between these two regions. The hypothesis of evolution by duplication is consistent with the conventional, unshuffled homology model, which is also supported by the obvious conservation of the retinal binding Lys moiety on helix 7 in both BR and the mammalian opsins.