The roles of the lambda c3 gene and the Escherichia coli catabolite gene activation system in the establishment of lysogeny by bacteriophage lambda

ATP-dependent proteinases in bacteria

Cytoplasmic proteolysis is an indispensable process for proper function of a cell. Degradation of many intracellular proteins is initiated by ATP-dependent proteinases, which are involved in the regulation of the level of proteins with short half-lives. In addition, they remove many damaged and abnormal proteins and thus play also an important role during stress. ATP-dependent proteinases are large multi-subunit assemblies composed of proteolytic core domains and ATPase-containing regulatory domains on a single polypeptide chain or on distinct subunits, which can act as molecular chaperones. This review briefly summarizes the data about four main groups of these proteinases in bacteria (i.e. Lon, Clp family, HslUV and FtsH) and characterizes their structure, mechanism of action and properties.

Regulation of bacteriophage lambda development by guanosine 5'-diphosphate-3'-diphosphate

On infection of its host, Escherichia coli, bacteriophage lambda can follow one of two alternative developmental pathways: lytic or lysogenic. Here we demonstrate that the "lysis-versus-lysogenization" decision is influenced by guanosine tetraphosphate (ppGpp), a nucleotide that is synthesized in E. coli cells in response to amino acid or carbon source starvation. We found that the efficiency of lysogenization is the highest at ppGpp concentrations somewhat higher than the basal level; too low and too high levels of ppGpp result in less efficient lysogenization. Maintenance of the already integrated lambda prophage and phage lytic development were not significantly influenced in the host lacking ppGpp. We found that the level of HflB/FtsH protease, responsible for degradation of the CII protein, an activator of "lysogenic" promoters, depends on ppGpp concentration. The highest levels of HflB/FtsH was found in bacteria lacking ppGpp and in cells bearing increased concentrations of this nucleotide. Using lacZ fusions, we investigated the influence of ppGpp on activities of lambda promoters important at the stage of the lysis-versus-lysogenization decision. We found that each promoter is regulated differentially in response to the abundance of ppGpp. Moreover, our results suggest that the cAMP level may influence ppGpp concentration in cells. The mechanism of the ppGpp-mediated control of lambda development at the stage of the lysis-versus-lysogenization decision may be explained on the basis of differential influence of guanosine tetraphosphate on activities of p(L), p(R), p(E), p(I), and p(aQ) promoters and by dependence of HflB/FtsH protease level on ppGpp concentration.

Stochastic kinetic analysis of developmental pathway bifurcation in phage lambda-infected Escherichia coli cells

Fluctuations in rates of gene expression can produce highly erratic time patterns of protein production in individual cells and wide diversity in instantaneous protein concentrations across cell populations. When two independently produced regulatory proteins acting at low cellular concentrations competitively control a switch point in a pathway, stochastic variations in their concentrations can produce probabilistic pathway selection, so that an initially homogeneous cell population partitions into distinct phenotypic subpopulations. Many pathogenic organisms, for example, use this mechanism to randomly switch surface features to evade host responses. This coupling between molecular-level fluctuations and macroscopic phenotype selection is analyzed using the phage lambda lysis-lysogeny decision circuit as a model system. The fraction of infected cells selecting the lysogenic pathway at different phage:cell ratios, predicted using a molecular-level stochastic kinetic model of the genetic regulatory circuit, is consistent with experimental observations. The kinetic model of the decision circuit uses the stochastic formulation of chemical kinetics, stochastic mechanisms of gene expression, and a statistical-thermodynamic model of promoter regulation. Conventional deterministic kinetics cannot be used to predict statistics of regulatory systems that produce probabilistic outcomes. Rather, a stochastic kinetic analysis must be used to predict statistics of regulatory outcomes for such stochastically regulated systems.

The Escherichia coli hflA locus encodes a putative GTP-binding protein and two membrane proteins, one of which contains a protease-like domain

The hflA (high frequency of lysogenization) locus of Escherichia coli governs the lysis-lysogeny decision of bacteriophage lambda by controlling stability of the phage cII protein. hflA contains three genes, hflX, hflK, and hflC, encoding polypeptides of 50, 46, and 37 kDa, respectively. We have determined the nucleotide sequence of 3843 base pairs containing hflA and have found three large open reading frames corresponding to hflX, hflK, and hflC. HflX contains the three sequence motifs typical of GTP-binding proteins and appears to be a member of a distinct family of putative GTPases. HflC and HflK appear to be integral membrane proteins which show some similarity to each other and to a human membrane protein. The C-terminal region of HflC contains a domain resembling the catalytic domain of ClpP, a bacterial ATP-dependent protease. We hypothesize that HflK and HflC constitute a distinct membrane-bound protease whose activity may be modulated by HflX GTPase.

Membrane localization of the HflA regulatory protease of Escherichia coli by immunoelectron microscopy

The hflA locus of Escherichia coli specifies a multisubunit protease that selectively degrades the cII transcriptional activator of phage lambda. The regulated turnover of cII is critical for the choice between the lytic and lysogenic pathways of viral development. Previous cell fractionation work has indicated that HflA is associated with the inner membrane fraction. We have sought to demonstrate that the HflA protease is localized in the cell membrane of intact cells. To achieve this goal, we have combined electron microscopy of thin-sectioned E. coli cells with antibody tagging by a colloidal gold label. Using antibody to purified HflA protein, we have found preferential membrane labeling for hflA+ cells but not for hflA mutant cells. We conclude that HflA protease is localized in the cell membrane. The membrane location for HflA protein may serve as a component of a targeting mechanism to limit the action of the regulatory protease to selected cytoplasmic proteins.

The activity of the CIII regulator of lambdoid bacteriophages resides within a 24-amino acid protein domain

The CIII protein of lambdoid bacteriophages promotes lysogeny by stabilizing the phage-encoded CII protein, a transcriptional activator of the repressor and integrase genes. We have isolated a set of missense mutations in the cIII gene of phage lambda and of phage HK022 that yield inactive CIII proteins. All the mutations are located in the relatively conserved central region of the protein. A comparative analysis of the CIII protein sequence in lambda, HK022, and the lambdoid bacteriophage P22 leads us to suggest that this central region assumes an amphipathic alpha-helical structure. This part of the lambda cIII gene was cloned within a fragment of the lacZ gene (the alpha-complementing fragment). The resulting fusion protein displays CIII activity. Mutations that yield a nonfunctional fusion protein map within its CIII moiety. These results indicate that the central portion of the CIII protein is both necessary and sufficient for CIII activity.

An additional function for bacteriophage lambda rex: the rexB product prevents degradation of the lambda O protein

The rex operon of bacteriophage lambda excludes the development of several unrelated bacteriophages. Here we present an additional lambda rexB function: it prevents degradation of the short-lived protein lambda O known to be involved in lambda DNA replication. We have shown that it is the product of rexB that is responsible for the stabilization of lambda O: when a nonsense mutation is present in rexB, lambda O protein is labile; suppression of the mutation by the corresponding nonsense suppressor causes partial restabilization of lambda O. lambda rexB also stabilizes lambda O in trans. We discuss our results in relation to the function of rexB in lambda DNA replication and its role in the protein degradation pathways of bacteriophage lambda.

Levels of Epstein-Barr virus DNA in lymphoblastoid cell lines are correlated with frequencies of spontaneous lytic growth but not with levels of expression of EBNA-1, EBNA-2, or latent membrane protein

The process of Epstein-Barr virus (EBV)-induced transformation of human B lymphocytes results in a cell line that is a mixture of latently and lytically infected cells, with the lytic cells composing roughly 5% to less than 0.0001% of the overall population. A set of nine normal lymphoblastoid cell lines that span a 100- to 200-fold range in average EBV DNA content were studied, and the frequency with which these cells entered a lytic phase of viral growth correlated with their EBV DNA copy number (as a population average). However, neither factor correlated with the levels of expression of transcript for the viral genes EBNA-1, EBNA-2, and latent membrane protein, nor did they correlate with the levels of EBNA-2 protein and latent membrane protein. The rate at which a cell line enters into lytic growth spontaneously is therefore not dependent on the overall steady-state levels of expression of these latent-phase genes.

Cleavage of the cII protein of phage lambda by purified HflA protease: control of the switch between lysis and lysogeny

The activity of the cII protein of phage lambda is probably the critical controlling factor in the choice of the lytic or lysogenic pathway by an infecting virus. Previous work has established that cII activity is regulated through the turnover of cII protein; the products of the hflA and hflB loci of Escherichia coli are needed for a degradative reaction, and lambda cIII functions in stabilizing cII. By using the cloned hflA locus, we have purified a cII-cleaving enzyme that we term HflA. Purified HflA contains three polypeptides; at least two of the subunits are products of the hflA region, and the third is probably a cleavage product of the larger of these two hflA-encoded polypeptides. The HflA protease activity cleaves cII to small fragments. We conclude that the switch between lambda developmental pathways involves regulated cleavage of cII by the specific protease HflA.

Identification of polypeptides encoded by an Escherichia coli locus (hflA) that governs the lysis-lysogeny decision of bacteriophage lambda

We report the cloning of the Escherichia coli hflA locus, which governs stability of phage lambda cII protein and which has been proposed to encode or regulate a cII-specific protease. The hflA locus was cloned on an 18-kilobase DNA fragment by selecting for plasmids that carry the neighboring purA gene. The boundaries of hflA were delimited by analysis of deletions and insertions constructed in vitro and by use of transposon Tn1000. Maxicell analysis of the proteins encoded by the hflA-containing fragment shows that hflA consists of at least two nonoverlapping genes, hflC and hflK, encoding polypeptides of 37,000 (C) and 46,000 (K) daltons. We observe that insertions into one gene eliminate the corresponding polypeptide and greatly reduce synthesis of the other. We suggest that these two polypeptides (K and C) interact to form a multimeric complex and that free subunits are unstable. We have constructed two types of fusions between hflA and lacZ. One is an hflC-lacZ protein fusion constructed in vitro; the other is an hfl-lacZ operon fusion in which a Mu dX(Apr lac) has inserted into the hflK gene. We have used the operon fusion to infer the direction of transcription of the hflK gene--toward hflC and in the same direction as hflC. Last, we describe evidence that hflA contains an additional gene, hflX, encoding a 50,000-dalton polypeptide.

The catabolite gene activation system of E. coli may be directly involved in regulation of bacteriophage lambda development

The primary structure of bacteriophage lambda DNA has been searched for the presence of consensus CAP binding sites. Four putative CAP binding sites have been found on the lambda genome, indicating that the catabolite gene activation system of E. coli may be directly involved in the regulation of lambda development. Molecular mechanisms of putative cAMP-CAP-mediated stimulation of lysogenic and lytic responses are discussed.

Control of bacteriophage lambda CII activity by bacteriophage and host functions

We have studied the regulation of the lambda cII gene in vivo using cloned lambda fragments. Lambda N protein stimulated cII expression. Surprisingly, although very high cII protein levels were detected by gel electrophoresis, little cII protein activity, measured as stimulation of the lambda pI and pE promoters, was observed. The half-life of cII protein depended critically on its initial level. At low concentrations its half-life was as short as 1.5 min, whereas at high cII protein levels, it could be as long as 22 min. The Escherichia coli mutant ER437 directs lambda towards lysogeny; cII protein was more stable in this strain than in the wild type. On the other hand, although cyclic AMP is required for efficient lysogeny, it did not appear to influence the synthesis, stability, or activity of cII protein.

Promoter for the establishment of repressor synthesis in bacteriophage lambda

Transcription of the lambda repressor gene (cI) is positively regulated by the phage-encoded proteins cII and cIII. We have isolated and characterized the 5'-terminal region of this RNA and shown that it originates at a promoter (pE) located between genes cro and cII. The DNA sequence of this promoter shows little homology to other known promoters. Initiation of transcription from PE is abolished by the cis-dominant mutations cY; these mutations alter the "-10" and "-35" regions of the promoter. We propose that the "-35" region is the site of activation of PE, possibly via the direct interaction of protein cII.

DNA sequence of regulatory region for integration gene of bacteriophage lambda

The cII and cIII proteins specified by bacteriophage lambda direct the lysogenic response to infection through the coordinate establishment of repression and integration of the viral DNA. The regulatory activity of cII/cIII involves positive regulation of two promoter sites: the p(E) promoter, turning on expression of the cI protein that maintains lysogeny, and the p(I) promoter, activating synthesis of the Int protein for integrative recombination. Regulation of the p(I) promoter provides for differential expression of the Int protein with respect to the excision-specific Xis protein from the closely linked int and xis genes. We have determined the DNA sequence of the p(I) promoter region for wild-type lambda DNA and for two classes of mutations: intc mutations, which result in a high rate of Int synthesis in the absence of cII, and deletion mutations, some of which eliminate cII-activated expression of the int gene. We find a sequence with considerable homology (11 of 15 bases) to a "typical" (computer-generated) promoter sequence, adjacent to a region with striking homology (11 of 14 bases) to part of the p(E) promoter region. This presumed p(I) sequence overlaps the start of the xis gene and includes the site of two intc point mutations. A cII-insensitive xis(+) deletion partially removes the proposed p(I) sequence; a deletion that leaves the p(I) sequence intact but terminates 21 bases upstream does not interfere with cII activation of the int gene. From our results and the analysis of the p(E) region, we suggest that cII acts in the promoter -35 recognition region to facilitate binding by RNA polymerase at the -10 interaction region. Differential expression of the int and xis genes results because the p(I) transcript lacks the initiation codon for Xis protein synthesis.

The depression of endolysin synthesis in bacteria infected with high multiplicities of phage lambda

The effect of multiplicity of infection was studied in Escherichia coli with lambda phage, using phage endolysin as an example of a late gene product. A very sensitive endolysin assay method was used so that the initiation time of endolysin synthesis could be more accurately determined. It was observed that high multiplicity of infection (1) increases the rate of lysogenization, (2) progressively delays lysis time, and (3) significantly delays and reduces the synthesis of endolysin in lamdacIII+ cII+ -infected cells. The extent of delay and reduction in endolysin synthesis increases with increasing multiplicity. In contrast, lamdacIII67cII68-infected cells show no delay in endolysin synthesis at high multiplicity of infection when compared with the lamdacIII+ cII+ -infected cells. The results suggest that (1) the expression of cIII and cII genes is multiplicity dependent, (2) high multiplicity of infection enhances the expression of the cIII and cII genes, and (3) the expression of the cIII and cII genes interferes with the expression of the late genes. A model to explain how the expression of the cIII and cII genes interferes with the expression of the late genes is proposed.

Repression and autogenous stimulation in vitro by bacteriophage lambda repressor

Purified lambda repressor protein is shown to reduce the lambda DNA-directed synthesis of proteins in vitro as determined both by net amino-acid incorporation and by analysis of specific lambda-coded proteins resolved by sodium dodecyl sulfate/polyacrylamide slab gel electrophoresis. By means of different lambda DNA templates carrying deletion and point mutations in the operators o-L or o-R, it has been possible to demonstrate repression of the synthesis of two classes of lambda proteins. The synthesis of one, class c, appears to be controlled from the operator o-L and is more efficiently repressed at low concentrations of the repressor than that of the other class of repressible lambda proteins, class d, which is controlled from the operator o-R. Several other proteins synthesized in vitro are not repressible. Some of these are coded by the J-att region. In addition, the repressor appears to have another activity, that of stimulating the synthesis of a protein identified as the repressor itself. Lambda repressor appears to stimulate its own synthesis by acting at prm, a site defined by the cis-acting mutation prm 116.