Coming early to the late genes

When I was an undergraduate at Caltech, I worked in Bob Edgar’s laboratory on an obscure aspect of phage T4 morphogenesis, which had previously been studied by a little-known South African geneticist, Sydney Brenner [1]. I remember a journal club presented by one of the graduate students in the lab, about the lac repressor binding to DNA. I don’t remember the precise experiments but I know I was impressed by the logic and the stepwise manner in which they proceeded. It was molecular biology — exciting, elegant and revealing. The Edgar lab worked on bacteriophage morphogenesis, and so I was well aware of the components (heads, tails, tailfibers, baseplates, whiskers, collars, and so on) that make up the phage particle, and of their corresponding genes, which were called ‘late’ genes. I was standing outside the Edgar lab one day, next to the enormous map of the phage T4 rII locus that adorned the wall and something struck me that I hadn’t appreciated before. It made sense that the genes encoding the phage head and tail proteins were expressed late, but what was the mechanism that ensured that they were expressed late? Even more striking, why were these genes not expressed if DNA replication, an early event, was blocked? I went on to graduate school at the Massachusetts Institute of Technology with a continuing interest in the control of gene expression, thinking that I would work in some area of developmental biology. I didn’t intend to study phage but I found it impossible to give up working with an organism where I could do lots of experiments and ask lots of questions, and so I joined Ethan Signer’s lab to work on bacteriophage λ. During my first year, at a lab journal club, I heard about a paper by Dambly, Couturier and Thomas [2] concerning a phenomenon they called ‘functional rescue’. They were studying mutants of λ that were defective in one of their late genes, the R gene. Mutants lacking R cannot lyse the host cell and thus do not form plaques on bacterial lawns. Phage λ can exist in a silent form integrated into the host bacterial chromosome, as a prophage. Although a λ prophage will normally repress an infecting λ phage, these experiments were done using an infecting λ phage that is insensitive to repression by the λ prophage. The Dambly et al. paper [2] made the striking observation that such an insensitive phage with a defective R gene could activate the normally silent R gene of a prophage and cause the cell to lyse. How was it possible for the infecting λ to turn on the prophage R gene? Answering this question involves understanding why the R gene of the prophage is silent under ordinary circumstances. The reason is that transcription of R requires expression of the Q gene, which is not expressed by the λ prophage. Dambly et al. showed that an infecting phage could turn on the prophage R gene by synthesizing Q protein (which it was able to do because the infecting phage was insensitive to the prophage’s repressor). The Q protein could then activate transcription of the prophage R gene and thereby provide the missing function for the infecting phage to grow lytically (see Figure 1). Although λ genetics has a reputation for being arcane and complex (which perhaps developed after I entered the field), this paper made a lot of sense to me. The experiments were straightforward, I loved the logic and I could see how this type of experiment illuminated the mechanisms underlying the control of gene expression. There may also have been some personal resonance with these experiments because they involved late genes. As I sat through this journal club, I was compulsively doodling ‘Functional Rescue’ in highly floral, psychedelic style (it was, after all, 1968) and thinking how wonderful it would be to do this type of work. Towards the end of my first year, Ethan Signer suggested I should check out some strains in the lab collection that contained various deletions of the λ prophage, in particular, one strain with a deletion that ended in the S gene, which is adjacent to R. We knew the R gene must be present but it couldn’t be Magazine R827