Fifty years fused to lac.

I recount the history of how I became interested in the use of gene fusions for studying biological problems. Initially, selections for mutations that would restore function to an inactivated lac operon unexpectedly yielded fusions in which lac was expressed from the controlling elements of upstream genes. Subsequently, by chance, I generated strains in which the lac operon was transposed from its normal position on the chromosome to a position close to the trp operon, thus facilitating sets of useful fusions of the two operons. The development of a more generalized technique for obtaining fusions by my student Malcolm Casadaban opened up a much broader set of biological problems that could be approached with fusions. Work on these problems included the study of protein translocation across membranes, the analysis of membrane protein topology, and the discovery of the pathway of electron transfer that leads to disulfide bond formation in proteins.