Portal de Periódicos da CAPES

Lambda's Switch: Lessons from a Module Swap

2006; Elsevier BV; Volume: 16; Issue: 12 Linguagem: Inglês

10.1016/j.cub.2006.05.037

1879-0445

Mark Ptashne,

Bacterial Genetics and Biotechnology

A recent experiment has replaced Cro, a crucial component of lambda's genetic switch, with the lac repressor (plus two lac operators). The resulting hybrid phage is viable, but a subtle phenotypic defect explains a puzzle concerning the workings of the switch. A recent experiment has replaced Cro, a crucial component of lambda's genetic switch, with the lac repressor (plus two lac operators). The resulting hybrid phage is viable, but a subtle phenotypic defect explains a puzzle concerning the workings of the switch. The evolution of complexity, modularity, and systems biology — all of these matters come to mind when reading Atsumi and Little's [1Atsumi S. Little J.W. Role of the lytic repressor in prophage induction of phage lambda as analyzed by a module-replacement approach.Proc. Natl. Acad. Sci. 2006; 103: 4558-4563Crossref PubMed Scopus (35) Google Scholar] recent analysis of phage lambda's 'genetic switch'. The switch, which I will describe in a bit more detail below, comprises an integrated set of simple protein–protein and protein–DNA interactions, all of which have been extensively characterized and quantified. A circuit diagram describing these interactions, a part of which is illustrated in Figure 1, has been at hand for some time (reviewed in [2Ptashne M. A Genetic Switch: Phage Lamba Revisited.Third edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York2004Google Scholar]). Lambda's switch might strike one as 'irreducibly complex' — perhaps taking away any part of it would be disasterous. But, as shown by Atsumi and Little [1Atsumi S. Little J.W. Role of the lytic repressor in prophage induction of phage lambda as analyzed by a module-replacement approach.Proc. Natl. Acad. Sci. 2006; 103: 4558-4563Crossref PubMed Scopus (35) Google Scholar], and by previous work primarily from the same lab [3Little J.W. Shepley D.P. Wert D.W. Robustness of a gene regulatory circuit.EMBO J. 1999; 18: 4299-4307Crossref PubMed Scopus (207) Google Scholar], various parts of the switch can be removed without totally destroying its function. It seems that evolution can start with a crude version of a switch and, by adding parts seriatum, make the switch work incrementally better, just as Darwin would have liked. The modularity of the switch — which might well have been obscured by later evolutionary modifications — has been so well conserved that, as shown here, one control element can be removed and replaced with a heterologous one without impairing switch function. And — the main point of the new paper under discussion [1Atsumi S. Little J.W. Role of the lytic repressor in prophage induction of phage lambda as analyzed by a module-replacement approach.Proc. Natl. Acad. Sci. 2006; 103: 4558-4563Crossref PubMed Scopus (35) Google Scholar] — this hybrid switch can be used to address a thorny question concerning the physiological significance of one of the reactions of the switch. It is sobering to realize that, despite all our knowledge about the system, we evidently cannot calculate its behavior very precisely, and outstanding questions have to be addressed by sophisticated experimentation such as that reviewed here. To understand these matters we need a brief overview of the switch in action. Consider, to begin with, the lambda phage repressor, the DNA-binding protein that, in a lysogen, turns off transcription of lytic phage genes. The repressor also controls expression of its own gene (cI), both positively and negatively. The combination of these effects maintains the concentration of repressor at the right level so as to poise the lysogen to respond to an inducing signal such as UV irradiation. That signal induces lytic phage growth by destroying the repressor, and as lytic genes are turned on the repressor gene is turned off. The switch is amazingly efficient: absent an inducing agent, a lysogen is stable for countless generations, and yet, upon receiving such a signal, in virtually every cell of the population lytic growth of the previously dormant phage ensues [2Ptashne M. A Genetic Switch: Phage Lamba Revisited.Third edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York2004Google Scholar, 4Dodd I.B. Shearwin K.E. Egan J.B. Revisited gene regulation in bacteriophage lambda.Curr. Opin. Genet. Dev. 2005; 15: 145-152Crossref PubMed Scopus (98) Google Scholar]. These regulatory events are effected by protein–protein and protein–DNA binding reactions shown in part in Figure 1A. In a lysogen, repressor bound to sites OR1 and OR2 in the right operator (OR) activates transcription of its own gene by contacting RNA polymerase and thereby recruiting it to the adjacent promoter (PRM) of the repressor gene (cI). These DNA-bound repressors simultaneously exclude polymerase from the lytic promoter PR. When bound to the weaker site OR3 repressor turns off transcription of cI by excluding RNA polymerase from its promoter. In mutants lacking this auto-negative control, the level of cI expression is elevated some two-three-fold. This small increase is sufficient to significantly impair induction, a fact that becomes important later in this story [4Dodd I.B. Shearwin K.E. Egan J.B. Revisited gene regulation in bacteriophage lambda.Curr. Opin. Genet. Dev. 2005; 15: 145-152Crossref PubMed Scopus (98) Google Scholar, 5Dodd I.B. Perkins A.J. Tsemitsidis D. Egan J.B. Octamerization of lambda CI repressor is needed for effective repression of PRM and efficient switching from lysogeny.Genes Dev. 2001; 15: 3013-3022Crossref PubMed Scopus (143) Google Scholar]. Omitted from Figure 1 is a second operator–promoter sequence (OL, PL) positioned some 2400 base pairs away. Interactions between repressors bound simultaneously to sites in OL and OR (with concomitant DNA looping) aid the reactions shown here, and are particularly important for the binding of repressor to OR3[6Dodd I.B. Shearwin K.E. Perkins A.J. Burr T. Hochschild A. Egan J.B. Cooperativity in long-range gene regulation by the lambda CI repressor.Genes Dev. 2004; 18: 344-354Crossref PubMed Scopus (147) Google Scholar]. Cooperative binding of repressor subunits to DNA is mediated by discrete contacts between repressor molecules [7Bell C.E. Frescura P. Hochschild A. Lewis M. Crystal structure of the lambda repressor C-terminal domain provides a model for cooperative operator binding.Cell. 2000; 101: 801-811Abstract Full Text Full Text PDF PubMed Scopus (108) Google Scholar, 8Bell C.E. Lewis M. Crystal structure of the lambda repressor C-terminal domain octamer.J. Mol. Biol. 2001; 314: 1127-1136Crossref PubMed Scopus (41) Google Scholar], just as the activation of cI transcription by repressor is mediated by a specific contact between repressor and polymerase [9Jain D. Nickels B.E. Sun L. Hochschild A. Darst S.A. Structure of a ternary transcription activation complex.Mol. Cell. 2004; 13: 45-53Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar]. One might have guessed that this rather elaborate machinery — built as we have indicated from simple parts — would suffice to make a good switch. But we have long known that there is another key player: the protein called Cro, the gene for which lies adjacent to, and is transcribed in the opposite direction from, cI (Figure 1B). The cro gene, silent in a lysogen, is one of the first to be transcribed upon induction, and its activity is required for efficient lytic growth. One essential function of Cro is to turn down expression of lytic genes (including itself) that are expressed at a high level immediately following induction. It does this by binding to OR1 and OR2, the same sites recognized by repressor [2Ptashne M. A Genetic Switch: Phage Lamba Revisited.Third edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York2004Google Scholar]. One of the early genes turned off (or down) by Cro encodes a protein, CII, that initiates transcription of cI, which then (as mentioned above) becomes self-sustaining, a classic epigenetic event. Thus, Cro indirectly discourages the establishment of repressor synthesis and thereby encourages the phage to enter the lytic cycle after UV irradiation of a lambda lysogen. This picture is extensively supported by genetic and physiological experiments (see, for example, [10Kobiler O. Rokney A. Friedman N. Court D.L. Stavans J. Oppenheim A.B. Quantitative kinetic analysis of the bacteriophage lambda genetic network.Proc. Natl. Acad. Sci. USA. 2005; 102: 4470-4475Crossref PubMed Scopus (91) Google Scholar]). But perhaps — and as suggested early on [11Johnson A.D. Poteete A.R. Lauer G. Sauer R.T. Ackers G.K. Ptashne M. Lambda repressor and cro–components of an efficient molecular switch.Nature. 1981; 294: 217-223Crossref PubMed Scopus (294) Google Scholar] — Cro has another role too: by binding to OR3, Cro would turn off transcription of cI directly — just as repressor binding to that site turns off cI transcription. As OR3 is the site in OR with the highest affinity for Cro, that effect, plausibly, would be the initial consequence of production of Cro. Is this reaction important in throwing the switch as the phage enters lytic growth? Or, alternatively, might direct inactivation of repressor by the inducing agent, and the consequent loss of cI autoactivation, suffice to keep the level of repressor low enough to allow lytic growth? The straightforward way to distinguish between these scenarios would be to make a lysogen with a phage bearing a mutation in OR3 that does not bind Cro at that site. If Cro binding to OR3 were important for induction, then such a mutant lysogen would induce only poorly. The problem facing Atsumi and Little [1Atsumi S. Little J.W. Role of the lytic repressor in prophage induction of phage lambda as analyzed by a module-replacement approach.Proc. Natl. Acad. Sci. 2006; 103: 4558-4563Crossref PubMed Scopus (35) Google Scholar] was that every tested mutation in OR3 that diminished Cro binding also diminished repressor binding to that site [4Dodd I.B. Shearwin K.E. Egan J.B. Revisited gene regulation in bacteriophage lambda.Curr. Opin. Genet. Dev. 2005; 15: 145-152Crossref PubMed Scopus (98) Google Scholar, 5Dodd I.B. Perkins A.J. Tsemitsidis D. Egan J.B. Octamerization of lambda CI repressor is needed for effective repression of PRM and efficient switching from lysogeny.Genes Dev. 2001; 15: 3013-3022Crossref PubMed Scopus (143) Google Scholar] (but see below for the latest news on this, which came in while this piece was being finalized for production). And, as mentioned above, if repressor cannot bind to OR3, the consequent overproduction of repressor (a two–three-fold effect) is sufficient to impede induction. And so the question remained: does Cro binding to OR3 play any role in the transition from lysogenic to lytic growth upon induction of a lysogen? (For an experiment suggesting it might not, see [12Svenningsen S.L. Costantino N. Court D.L. Adhya S. On the role of Cro in lambda prophage induction.Proc. Natl. Acad. Sci. USA. 2005; 102: 4465-4469Crossref PubMed Scopus (79) Google Scholar].) Atsumi and Little [1Atsumi S. Little J.W. Role of the lytic repressor in prophage induction of phage lambda as analyzed by a module-replacement approach.Proc. Natl. Acad. Sci. 2006; 103: 4558-4563Crossref PubMed Scopus (35) Google Scholar] addressed this problem by first constructing a hybrid phage bearing the lac repressor in place of Cro (Figure 1C). The hybrid phage was also modified so as to bear lac operators just downstream of PL and PR, leaving the lambda operators in their usual place. In this configuration, each lytic promoter can be repressed by either the lambda or the lac repressor. Thus, in a lysogen, lambda repressor should act as usual, repressing lytic genes and auto-regulating its own gene. During lytic growth, lac repressor (in place of Cro) should bind the lac operators and turn down transcription of lytic genes. This conceptually straightforward enterprise was not so easy to realize in practice. Even though the affinities of the two repressors (Cro and lac repressor) for their corresponding wild-type and mutant operator sites have been extensively characterized, there is no way to be confident that any specified level of lac repressor working on any specified lac operator variant will mimic Cro's action in vivo. And so Atsumi and Little [1Atsumi S. Little J.W. Role of the lytic repressor in prophage induction of phage lambda as analyzed by a module-replacement approach.Proc. Natl. Acad. Sci. 2006; 103: 4558-4563Crossref PubMed Scopus (35) Google Scholar] constructed phage libraries bearing an array of lac operator sequences (with varying affinities for lac repressor) at both PL and PR; and an array of Shine-Delgarno sequences (which determine the efficiency of translation of the mRNA) in the lac repressor gene. They looked among these constructs for plaque formers, testing their constructs at a range of different concentrations of IPTG, a molecule that inactivates lac repressor. And to eliminate needless complications, they used a mutant form of lac repressor that, unlike the wild type, forms dimers but not tetramers. Unlike the wild type, a single molecule of this mutant (a dimer in this case) cannot simultaneously bind to separated DNA sites. Atsumi and Little [1Atsumi S. Little J.W. Role of the lytic repressor in prophage induction of phage lambda as analyzed by a module-replacement approach.Proc. Natl. Acad. Sci. 2006; 103: 4558-4563Crossref PubMed Scopus (35) Google Scholar] found that at least one of the hybrid phages grows lytically and forms lysogens, and those lysogens can be induced to produce progeny, a gratifying result. As shown by various experiments, including noting the lethal effect of inactivating lac repressor with high concentrations of IPTG, lac repressor is performing Cro's critical function of turning down expression of lytic genes by binding to the introduced lac operators. But, because this phage has no lac operator in place of OR3, the result strongly suggests that binding of a repressor (Cro or lac) to OR3 — the issue raised above — is not absolutely required for lytic growth. Atsumi and Little [1Atsumi S. Little J.W. Role of the lytic repressor in prophage induction of phage lambda as analyzed by a module-replacement approach.Proc. Natl. Acad. Sci. 2006; 103: 4558-4563Crossref PubMed Scopus (35) Google Scholar] took the matter one step further, by showing that higher levels of UV irradiation are required to induce lysogens of the hybrid than are required to induce wild-type lysogens. And so, evidently, binding of Cro to OR3, while not required for that process, does improve the efficiency of induction. The suggested picture is that at a UV dose sufficient to induce say 50% of the lysogens (the 'set point'), the cells are poised to go one way or the other (induce or remain as lysogens) with equal frequency. The higher set point for lysogens of the hybrid phage suggests that Cro binding to OR3 helps push the decision to lysis, but this effect can be dispensed with at higher UV doses. This conclusion has a familiar ring to it: previous work from the Little lab [3Little J.W. Shepley D.P. Wert D.W. Robustness of a gene regulatory circuit.EMBO J. 1999; 18: 4299-4307Crossref PubMed Scopus (207) Google Scholar, 13Michalowski C.B. Little J.W. Positive autoregulation of cI is a dispensable feature of the phage lambda gene regulatory circuitry.J. Bacteriol. 2005; 187: 6430-6442Crossref PubMed Scopus (26) Google Scholar] has shown that various modifications to the switch, while not inactivating it, do impair its efficiency. For example, eliminating, by mutation, the ability of lambda repressor to activate transcription of its own gene produces a phage that grows lytically very well and even lysogenizes — but the lysogens are less stable (probably because insufficient repressor is made) than are wild-type lysogens. Lambda's switch seems to have evolved from some elemental state by a series of add-ons, each of which improves its function, perhaps as suggested in [2Ptashne M. A Genetic Switch: Phage Lamba Revisited.Third edition. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York2004Google Scholar, 14Ptashne M. Gann A. Imposing specificity by localization: mechanism and evolvability.Curr. Biol. 1998; 8: R812-R822Abstract Full Text Full Text PDF PubMed Google Scholar]. As this dispatch was going to press, a mutant OR3 (bearing three base changes) was described that cannot bind Cro, but binds repressor with nearly normal affinity (K. Shearwin, I. Dodd, R. Schubert and B. Egan, personal communication). Phage bearing the mutant OR3 grow lytically and form lysogens, but the efficiency of induction of those lysogens is decreased compared to wild type. The conclusion fits nicely with that of Atsumi and Little [1Atsumi S. Little J.W. Role of the lytic repressor in prophage induction of phage lambda as analyzed by a module-replacement approach.Proc. Natl. Acad. Sci. 2006; 103: 4558-4563Crossref PubMed Scopus (35) Google Scholar]: Cro binding to OR3 is not essential for induction, but it makes that process more efficient.