Introduction and characterization of amber mutations in the bacteriorhodopsin gene.

A fusion between the genes for bacteriorhodopsin and beta-galactosidase was constructed on a multicopy plasmid, pXB/Gal 101. The fusion gene, containing the bacteriorhodopsin gene fused upstream from the beta-galactosidase gene, was under the control of tandem lipoprotein and lac gene promoters. When expressed in Escherichia coli the fusion protein retained beta-galactosidase activity. Mutations in the fusion gene were produced by passage of pXB/Gal 101 through the E. coli mutator strain mut D5. Amber mutations were then selected by examining the loss of the lac+ phenotype imparted by the fusion protein to lac- E. coli cells. Amber mutations occurring within the bacteriorhodopsin gene were localized by replacing the beta-galactosidase region of each mutant plasmid with a beta-galactosidase region which was known to be unmutated. Precise localization of the mutations was achieved first by sizing the prematurely terminated peptides produced by the mutant plasmids in in vitro coupled transcription-translation reactions, and secondly by DNA sequence analysis. Six amber mutants in the gene for bacteriorhodopsin were characterized in this way. One of these was a transversion mutation at a lysine codon; the other five were all transition mutations at tryptophan codons, codons 10, 12, 80, 86, and 137 of the bacteriorhodopsin sequence.