Our discussion of gene regulatory pathways up to this point has focused on bacteria. Historically, however, much of the early research on gene regulation was done on viruses—specifically, bacteriophages (phages), viruses that infect bacterial cells. Phage genomes are small, yet the genes they contain must be carefully regulated to enable efficient infection and viral reproduction. In addition, phage genes are frequently transferred to or from host cell chromosomes, thereby providing a critical means of transferring genetic information between organisms (i.e., from one host to another). Such horizontal gene transfer plays a significant role in driving the evolution of new bacterial traits, including resistance to drugs and toxins.

We focus here on the bacteriophage λ (λ phage) system, one of the best-studied systems of gene regulation in all of biology. Gene regulation in λ phage is intimately coupled to the state of the host cell, and we discuss some examples illustrating the principles involved. Gene regulation in λ phage involves a series of integrated pathways that exemplify the coordinated gene expression occurring in cells: some of the mechanisms used by λ phage have been found to govern gene expression in other systems. For example, animal cells take advantage of differential binding affinities and cooperative interactions of regulators to turn genes and gene networks on and off during development. Thus, insights from the λ phage system continue to guide our understanding of more complex organisms.

716

Most organisms are susceptible to infection by viruses, which usurp the host cell machinery to produce more viral particles. One well-studied class of bacterial viruses, the lysogenic bacteriophages, uses two kinds of replication mechanism to ensure viral propagation and transmission. As we noted in Chapter 14, after introduction of its DNA into a host cell, a phage has the potential to enter one of two pathways for propagation (Figure 20-20). Most of the time, viruses use the lytic pathway, in which phage DNA is immediately replicated and viral proteins are synthesized to construct the viral coating around the DNA. Once many viral particles have been made, they burst from the cell, killing it, and enter the extracellular environment; they can now infect other cells. Occasionally, viruses enter the lysogenic pathway, whereby phage DNA integrates into the host cell chromosome and is replicated along with the chromosome as the cell divides. In this state, the bacteriophage is called a prophage, and the cell carrying it is a lysogen. Under normal conditions the prophage is stable within the lysogen, but if the host cell is stressed by DNA-damaging agents, nutritional limitations, or other conditions that threaten survival, the prophage can rapidly excise from the chromosome and enter the lytic pathway. The switch from lysogenic to lytic growth is referred to as prophage induction.

717

Many phages are capable of this kind of genetic switch, but the underlying mechanisms are best understood for λ phage, a virus of E. coli that consists of a double-stranded linear DNA encapsulated in a head structure and attached tail region made of virally encoded proteins. This phage has been used extensively as a model system for studying integrated gene regulatory networks, and it has also provided many tools useful in molecular biology research. For example, engineered versions of λ phage have been co-opted to transfer and express genes in E. coli and to inactivate host genes.

Two regulatory proteins, the λ repressor (also called cI) and Cro, govern the growth pathway of the virus. When the λ repressor protein is predominant, λ phage enters the lysogenic pathway; its DNA integrates into the chromosome, and only the λ repressor itself is expressed. When Cro predominates, however, λ phage enters the lytic pathway; most of the λ genes are expressed, and viral replication and packaging ensue. Eventually, the host cell is broken open by cell lysis, a process that releases the progeny phages.

Whether the phage enters lytic or lysogenic growth is determined early in a λ phage infection and depends largely on two other λ proteins: cII and cIII. The cII protein is a transcription activator that enhances transcription of the λ repressor gene and hence stimulates production of the repressor protein (cI). When cII is abundant and active, infection proceeds through the lysogenic pathway. But because cII is susceptible to degradation by bacterial proteases, it is often destroyed before it can trigger substantial production of the λ repressor. This tends to occur under nutrient-rich conditions, when proteases are present in high concentrations. It makes sense for the phage to enter the lytic pathway, triggered by low cII levels, when plenty of materials are available for viral reproduction and particle assembly.

The cIII protein stabilizes cII, probably by acting as a decoy, or alternative substrate, for protease molecules that would otherwise degrade cII. In this way, elevated cIII levels can trigger the switch to lysogen formation by enhancing cII concentrations, which increases production of the λ repressor.

Usually, λ phage infection leads to lytic growth. In addition to nutrient levels, the multiplicity of infection (MOI) also affects this growth pathway. This is because infections typically occur at a low MOI, when there are many more host cells than viral particles. In this situation, it is advantageous for viruses to grow lytically so that more progeny can be made and released to infect the available host cells. However, once there is an abundance of viral particles relative to host cells, multiple viruses begin to infect each host. In this circumstance, the high MOI leads to more frequent lysogen formation. The virus propagates silently within the chromosome and awaits future opportunities to enter the lytic pathway when host cells are again abundant. The mechanisms underlying this fascinating genetic switch have been elucidated over many years, using numerous genetic and biochemical methods.

Most of the ∼50 genes of the λ phage genome are required for viral replication and packaging; a relatively small region of the genome is critical for the gene regulation necessary to induce lytic versus lysogenic growth (Figure 20-21). On initial infection of a host cell by λ DNA, viral transcription begins at two opposing promoters, P_L (leftward promoter) and P_R (rightward promoter), to produce the “immediate early” transcripts. This leads to the production of two proteins, N and Cro, which begin to increase in concentration. Two additional promoters, P_RM (promoter for repressor maintenance) and P_RE (promoter for repressor establishment), drive transcription of the cI gene, which encodes the λ repressor. However, unlike the strongly constitutive P_L and P_R, P_RM and P_RE are weak promoters and require activators to recruit RNA polymerase. Thus, at first, the cI gene is not transcribed and no λ repressor is produced.

718

Overlapping the P_L, P_RM, and P_R promoters are two operator sites, O_L and O_R, each of which contains three binding sites for the λ repressor and Cro. Like the Lac repressor, both the λ repressor and Cro form homodimers in which two DNA-binding domains in the protein recognize DNA through a helix-turn-helix motif. In x-ray crystallographic structures of these proteins bound to DNA, researchers observed that each DNA-binding domain in the dimer binds to half of the 17 bp inverted repeat of the operator sequence (Figure 20-22). Each of the six operator binding sites, three sites in O_R and three in O_L (see Figure 20-21), can bind a λ repressor dimer or a Cro dimer. However, these sites have different affinities for the regulatory proteins, so they are not all occupied at once, or at random. In addition, operator-protein binding at one site influences binding at the other sites, an example of cooperative interaction.

During initial viral infection, once appreciable levels of Cro are expressed, this protein binds the O_R3 site; because O_R3 overlaps P_RM, Cro blocks the access of RNA polymerase to P_RM, and the cI gene is not transcribed. Cro does not bind as well to O_R1 and O_R2, and thus these operator sites are not occupied by Cro. If this situation continues, lytic growth ensues (Figure 20-23a).

Lysogenic growth requires the activity of P_RE (Figure 20-23b). Initial transcription of the cI gene requires activation of P_RE by cII, another gene product expressed early in the infection process. The cII protein can activate the weak P_RE promoter by binding to a site upstream from the transcription start site (the −35 region; see Chapter 15) and recruiting RNA polymerase to P_RE. However, cII is a substrate of the host cell protease FtsH, which cleaves and inactivates the activator. At a low MOI, in which one viral particle at most has infected any one cell, the concentration of cII does not reach a level sufficient to activate P_RE. But under conditions of high MOI, in which multiple viral particles have infected the cell, more cII is produced because of the multiple copies of the cII gene present in the cell. Furthermore, the phage protein cIII (also expressed early in infection) helps stabilize cII by competing as a substrate for FtsH. In this situation, FtsH cannot keep up with cII production, and the increased levels of cII lead to activation of P_RE.

As a result of the activation of P_RE, the λ repressor protein is produced. The repressor activates P_RM by binding to O_R1 and O_R2, leading to stable expression of the repressor (Figure 20-23c). The P_RM promoter is necessary to maintain ongoing production of the λ repressor because, when the repressor binds operator sites O_R1 and O_R2, P_R is silenced and production of cII stops, leading to loss of the activator (cII) for P_RE. Under these conditions, lysogenic growth is favored.

Note that if Cro were to bind O_R3 before cI (the repressor) bound O_R1 and O_R2, the phage would have trouble establishing lysogeny. This interference by Cro probably never occurs, however, because when cII is highly active, cI production is sufficient to shut off the cro gene before enough Cro is made to turn off P_RM.

The λ repressor is capable of complex regulation, in part because it is a two-domain protein that forms a functional dimer. The N-terminal region of the protein contains the helix-turn-helix DNA-binding motif, and the C-terminal domain is responsible for dimerization. As we have seen, the λ repressor can bind to any of six operator binding sites in the O_L and O_R regions (see Figure 20-21). Despite its name, the repressor bound at O_R2 actually activates transcription from P_RM, and this activation is critical to the lysogenic switch. However, the repressor has highest binding affinity for O_R1, and because binding is cooperative, O_R1 recognition increases the affinity of the bound λ repressor for O_R2. Repressor bound cooperatively at O_R1 and O_R2 blocks RNA polymerase binding at P_R, repressing transcription from that promoter. Similarly, repressor bound cooperatively at O_L1 and O_L2 prevents transcription from P_L. Thus, the repressor can simultaneously activate transcription of its own gene from P_RM and repress transcription of the immediate early genes necessary for lytic growth from P_L and P_R.

Once the lytic-to-lysogenic switch has occurred, the phage DNA integrates into the host chromosome by a mechanism described below. The integrated viral DNA can be maintained and replicated stably as part of the host cell chromosome. But, as described earlier, when the host cell is exposed to agents or conditions that damage DNA, the prophage is rapidly excised, and lytic growth begins. This lysogenic-to-lytic switch comes about because the λ repressor protein resembles the bacterial protein LexA, which undergoes self-cleavage when stimulated by a second bacterial protein, RecA, that is a sensor of DNA damage in the bacterial cell (see Figure 20-13). When RecA is activated in response to DNA damage (the SOS response), the λ repressor protein, if present in the cell, also undergoes self-cleavage. This reaction clips off the C-terminal region of the repressor, removing its dimerization domain and thereby destroying its cooperative binding to O_L and O_R, sites 1 and 2. As a result, the cleaved λ repressor dissociates from the operator DNA, allowing transcription from P_R and P_L, and thus lytic growth.

719

Switching between the activation and repression of promoters enables λ phage to alternate between lytic and lysogenic growth, as dictated by environmental conditions. But additional levels of regulation further control the expression of genes required to establish and maintain early steps in the infection cycle. Two λ proteins, the N and Q proteins, are known as antiterminators because they prevent RNA polymerase from prematurely stopping transcription of the genes they regulate. N protein binds to specific regions of the viral transcript called nut (N utilization) sites (Figure 20-24). The resulting RNA-protein complexes assemble with additional proteins produced from the host cell genes nusA, nusB, nusE, and nusG. Although the functions of these proteins are not known in detail, they somehow help the phage-produced N protein bind to RNA polymerase and enhance its ability to bypass terminator structures downstream from the genes for N and Cro. In this way, elongation of the viral transcript is favored once initial infection is established.

720

One target of the N protein antitermination mechanism is the viral gene encoding the Q antiterminator protein. Unlike N, Q binds DNA sequences between the −10 and −35 regions of PR′, the promoter for genes that are active late in the infection process. In the absence of Q, RNA polymerase pauses shortly after initiation of the transcript, then falls off the template when encountering a terminator hairpin structure, 200 bp further on. When Q is present, it first binds in the promoter region, then transfers to the polymerase when the enzyme pauses after initiation. With Q bound, the polymerase can transcribe through the terminator hairpin, thereby producing full-length transcripts; the mechanism of this antitermination process is not yet clear.

To establish a lysogenic infection, the virus must produce an integrase, an enzyme responsible for integrating phage DNA into the host cell chromosome (see Chapter 14). The int gene, which encodes the integrase, is transcribed from two promoters, P_I and P_L. The cII protein activates P_I, in addition to P_RE (as described earlier); thus, when cII is abundant and the lysogenic state is favored, both repressor and integrase are expressed. Although the int gene is also transcribed from P_L, the two different transcripts have inherently different stabilities due to different RNA structures at their 3′ ends. The P_I-derived transcript ends at a terminator hairpin structure downstream from the integrase-coding sequence, whereas the P_L-derived transcript extends past the terminator—because the transcript is made by a polymerase modified by the N protein (see Figure 20-24). This longer transcript forms an alternative hairpin structure that is a substrate for cellular ribonucleases. As a result, only the P_I-derived transcript is maintained, and it can be translated into integrase protein (Figure 20-25).

721

In a twist on this regulatory mechanism, the λ DNA integrates into the host chromosome such that a small portion of the phage DNA is removed—the portion that, when transcribed, becomes the destabilizing RNA sequence at the end of the P_L-derived integrase transcript. Thus, when λ phage is in the lysogenic state and integrated into the host chromosome, integrase can be produced from transcripts derived from either P_I or P_L. Because integrase is also required to excise the prophage from the host genome, the ability to use either type of int gene transcript ensures that this step is not dependent on cII protein concentration.

722

723

724