Holins: the protein clocks of bacteriophage infections

Breaking free: "protein antibiotics" and phage lysis

Bacteriophages must destroy the bacterial cell wall to lyse their host and release their progeny into the environment. There are at least two distinct mechanisms by which phages destroy the cell wall. Bacteriophages with large genomes use a holin-endolysin system, while bacteriophages with small genomes encode a single lysis protein. Three unrelated single protein lysis systems are known and these proteins will be the focus of the review. Recent results indicate that at least two of these proteins inhibit cell wall synthesis and are thus the phage analogs of antibiotics like penicillin.

Antibiotic tolerance in pneumococci

When bacteria such as Staphylococcus aureus and Streptococcus pneumoniae are exposed to lytic antibiotics such as penicillin and vancomycin, a self-induced killing process is initiated in the organism. This killing occurs via both non-lytic and lytic processes. Recent data suggest that the non-lytic killing system, which might affect the cytoplasmic membrane, secondarily activates murein hydrolases that eventually lyse the cell. Disturbances in this suicide pathway can lead to antibiotic tolerance, a process whereby the antibiotic still exerts its bacteriostatic effects but the self-induced killing system is impaired. In mutants obtained in vitro, signaling pathways have been affected that show either increased or decreased antibiotic-induced killing. Among clinical isolates of S. pneumoniae that are tolerant to penicillin and/or vancomycin, we do not yet know whether these signaling pathways are affected. We could, however, demonstrate that the activity of murein hydrolases is negatively controlled by the production of capsular polysaccharides in one vancomycin-tolerant isolate. Hence, type and level of capsular expression might constitute one factor that determines the degree of lysis, once the killing signal has been elicited by the antibiotic.

A bacteriolytic agent that detects and kills Bacillus anthracis

The dormant and durable spore form of Bacillus anthracis is an ideal biological weapon of mass destruction. Once inhaled, spores are transported by alveolar macrophages to lymph nodes surrounding the lungs, where they germinate; subsequent vegetative expansion causes an overwhelming flood of bacteria and toxins into the blood, killing up to 99% of untreated victims. Natural and genetically engineered antibiotic-resistant bacilli amplify the threat of spores being used as weapons, and heighten the need for improved treatments and spore-detection methods after an intentional release. We exploited the inherent binding specificity and lytic action of bacteriophage enzymes called lysins for the rapid detection and killing of B. anthracis. Here we show that the PlyG lysin, isolated from the gamma phage of B. anthracis, specifically kills B. anthracis isolates and other members of the B. anthracis 'cluster' of bacilli in vitro and in vivo. Both vegetative cells and germinating spores are susceptible. The lytic specificity of PlyG was also exploited as part of a rapid method for the identification of B. anthracis. We conclude that PlyG is a tool for the treatment and detection of B. anthracis.

The Escherichia coli FKBP-type PPIase SlyD is required for the stabilization of the E lysis protein of bacteriophage phi X174

Most bacteriophages abruptly terminate their vegetative cycle by causing lysis of the host cell. The ssDNA phage phi X174 uses a single lysis gene, E, encoding a 91-amino-acid membrane protein that causes lysis of Escherichia coli by inhibiting MraY, a conserved enzyme of murein biosynthesis. Recessive mutations in the host gene slyD (sensitivity to lysis) absolutely block E-mediated lysis and phi X174 plaque formation. The slyD gene encodes a FKBP-type peptidyl-prolyl cis-trans isomerase (PPIase). To investigate the molecular basis of this unique FKBP-dependence, spontaneous plaque-forming mutants of phi X174 were isolated on a slyD lawn. All of these Epos ('plates on slyD') suppressors encode proteins with either a R3H or L19F change. The double mutant was also isolated and generated the largest plaques on the slyD lawn. A c-myc epitope tag sequence was incorporated into the parental E and Epos genes without effect on lytic function. Western blots and pulse-chase labelling experiments showed that both Epos and E are highly unstable in a slyD background; however, Epos is synthesized at a higher rate, allowing a lysis-sufficient level of Epos to accumulate. Our results indicate that SlyD is required for stabilizing the E protein and allowing it to accumulate to the levels required to exert its lytic effect. These data are discussed in terms of a model for the specific role of the SlyD PPIase in E folding, and of the use of the very strict SlyD- dependence phenotype for identifying elements of PPIase selectivity.

Burkholderia thailandensis E125 harbors a temperate bacteriophage specific for Burkholderia mallei

Burkholderia thailandensis is a nonpathogenic gram-negative bacillus that is closely related to Burkholderia mallei and Burkholderia pseudomallei. We found that B. thailandensis E125 spontaneously produced a bacteriophage, termed phiE125, which formed turbid plaques in top agar containing B. mallei ATCC 23344. We examined the host range of phiE125 and found that it formed plaques on B. mallei but not on any other bacterial species tested, including B. thailandensis and B. pseudomallei. Examination of the bacteriophage by transmission electron microscopy revealed an isometric head and a long noncontractile tail. B. mallei NCTC 120 and B. mallei DB110795 were resistant to infection with phiE125 and did not produce lipopolysaccharide (LPS) O antigen due to IS407A insertions in wbiE and wbiG, respectively. wbiE was provided in trans on a broad-host-range plasmid to B. mallei NCTC 120, and it restored LPS O-antigen production and susceptibility to phiE125. The 53,373-bp phiE125 genome contained 70 genes, an IS3 family insertion sequence (ISBt3), and an attachment site (attP) encompassing the 3' end of a proline tRNA (UGG) gene. While the overall genetic organization of the phiE125 genome was similar to lambda-like bacteriophages and prophages, it also possessed a novel cluster of putative replication and lysogeny genes. The phiE125 genome encoded an adenine and a cytosine methyltransferase, and purified bacteriophage DNA contained both N6-methyladenine and N4-methylcytosine. The results presented here demonstrate that phiE125 is a new member of the lambda supergroup of Siphoviridae that may be useful as a diagnostic tool for B. mallei.

Complete nucleotide sequence and likely recombinatorial origin of bacteriophage T3

We report the complete genome sequence (38,208 bp) of bacteriophage T3 and provide a bioinformatic comparative analysis with other completely sequenced members of the T7 group of phages. This comparison suggests that T3 has evolved from a recombinant between a T7-like coliphage and a yersiniophage. To assess this, recombination between T7 and the Yersinia enterocolitica serotype O:3 phage phiYeO3-12 was accomplished in vivo; coliphage progeny from this cross were selected that had many biological properties of T3. This represents the first experimentally observed recombination between lytic phages whose normal hosts are different bacterial genera.

C-terminal domains of Listeria monocytogenes bacteriophage murein hydrolases determine specific recognition and high-affinity binding to bacterial cell wall carbohydrates

Listeria monocytogenes phage endolysins Ply118 and Ply500 share a unique enzymatic activity and specifically hydrolyse Listeria cells at the completion of virus multiplication in order to release progeny phage. With the aim of determining the molecular basis for the lytic specificity of these enzymes, we have elucidated their domain structure and examined the function of their unrelated and unique C-terminal cell wall binding domains (CBDs). Analysis of deletion mutants showed that both domains are needed for lytic activity. Fusions of CBDs with green fluorescent protein (GFP) demonstrated that the C-terminal 140 amino acids of Ply500 and the C-terminal 182 residues of Ply118 are necessary and sufficient to direct the murein hydrolases to the bacterial cell wall. CBD500 was able to target GFP to the surface of Listeria cells belonging to serovar groups 4, 5 and 6, resulting in an even staining of the entire cell surface. In contrast, the CBD118 hybrid bound to a ligand predominantly present at septal regions and cell poles, but only on cells of serovars 1/2, 3 and 7. Non-covalent binding to surface carbohydrate ligands occurred in a rapid, saturation-dependent manner. We measured 4 x 104 and 8 x 104 binding sites for CBD118 and CBD500 respectively. Surface plasmon resonance analysis revealed unexpected high molecular affinity constants for the CBD-ligand interactions, corresponding to nanomolar affinities. In conclusion, we show that the CBDs are responsible for targeting the phage endolysins to their substrates and function to confer recognition specificity on the proteins. As the CBD sequences contain no repeats and lack all known sequence motifs for anchoring of proteins to the bacterial cell, we conclude that they use unique structural motifs for specific association with the surface of Gram-positive bacteria.

An ancient player unmasked: T4 rI encodes a t-specific antiholin

Phage T4 effects lysis by its holin T and its endolysin E. Lysis is inhibited (LIN) if the infected cell is subjected to secondary infections by T4 phage particles. The T4 rI gene is required for LIN in all hosts tested. Here, we show that a cloned rI gene can impose a T-specific LIN on T-mediated lysis in the context of the phage lambda infective cycle, in the absence of other T4 genes and without secondary infection by T4. Moreover, it is shown that the T holin accumulates in the membrane during LIN, forming SDS-resistant oligomers. We show by cross-linking experiments that a T-RI heterodimer is formed during LIN, demonstrating that RI belongs to the functional class of antiholins, such as the S107 protein of lambda, which heterodimerizes with its cognate holin, S105. Finally, we show that the addition of Ni(2+) ions to the medium can block lysis by a T protein hexahistidine-tagged at its C-terminus, suggesting that liganding of the periplasmic domain is sufficient to impose lysis inhibition. The results are discussed in terms of a model in which the LIN-inducing signal of the secondary infecting phage influences a conformational equilibrium assumed by RI in the periplasm.

Holins kill without warning

Holins comprise the most diverse functional group of proteins known. They are small bacteriophage-encoded proteins that accumulate during the period of late-protein synthesis after infection and cause lysis of the host cell at a precise genetically programmed time. It is unknown how holins achieve temporal precision, but a conserved feature of their function is that energy poisons subvert the normal scheduling mechanism and instantly trigger membrane disruption. On this basis, timing has been proposed to involve a progressive decrease in the energized state of the membrane until a critical triggering level is reached. Here, we report that membrane integrity is not compromised after the induction of holin synthesis until seconds before lysis. The proton motive force was monitored by the rotation of individual cells tethered by a single flagellum. The results suggest an alternative explanation for the lysis "clock," in which holin concentrations build to a critical level that leads to formation of an oligomeric complex that disrupts the membrane.

A protein antibiotic in the phage Qbeta virion: diversity in lysis targets

A(2), a capsid protein of RNA phage Qbeta, is also responsible for host lysis. A(2) blocked synthesis of murein precursors in vivo by inhibiting MurA, the catalyst of the committed step of murein biosynthesis. An A(2)-resistance mutation mapped to an exposed surface near the substrate-binding cleft of MurA. Moreover, purified Qbeta virions inhibited wild-type MurA, but not the mutant MurA, in vitro. Thus, the two small phages characterized for their lysis strategy, Qbeta and the small DNA phage phiX174, effect host lysis by targeting different enzymes in the multistep, universally conserved pathway of cell wall biosynthesis.

Phi29 family of phages

Continuous research spanning more than three decades has made the Bacillus bacteriophage phi29 a paradigm for several molecular mechanisms of general biological processes, such as DNA replication, regulation of transcription, phage morphogenesis, and phage DNA packaging. The genome of bacteriophage phi29 consists of a linear double-stranded DNA (dsDNA), which has a terminal protein (TP) covalently linked to its 5' ends. Initiation of DNA replication, carried out by a protein-primed mechanism, has been studied in detail and is considered to be a model system for the protein-primed DNA replication that is also used by most other linear genomes with a TP linked to their DNA ends, such as other phages, linear plasmids, and adenoviruses. In addition to a continuing progress in unraveling the initiation of DNA replication mechanism and the role of various proteins involved in this process, major advances have been made during the last few years, especially in our understanding of transcription regulation, the head-tail connector protein, and DNA packaging. Recent progress in all these topics is reviewed. In addition to phi29, the genomes of several other Bacillus phages consist of a linear dsDNA with a TP molecule attached to their 5' ends. These phi29-like phages can be divided into three groups. The first group includes, in addition to phi29, phages PZA, phi15, and BS32. The second group comprises B103, Nf, and M2Y, and the third group contains GA-1 as its sole member. Whereas the DNA sequences of the complete genomes of phi29 (group I) and B103 (group II) are known, only parts of the genome of GA-1 (group III) were sequenced. We have determined the complete DNA sequence of the GA-1 genome, which allowed analysis of differences and homologies between the three groups of phi29-like phages, which is included in this review.

Bacillus subtilis locus encoding a killer protein and its antidote

We have isolated mutations that block sporulation after formation of the polar septum in Bacillus subtilis. These mutations were mapped to the two genes of a new locus, spoIIS. Inactivation of the second gene, spoIISB, decreases sporulation efficiency by 4 orders of magnitude. Inactivation of the first gene, spoIISA, has no effect on sporulation but it fully restores sporulation of a spoIISB null mutant, indicating that SpoIISB is required only to counteract the negative effect of SpoIISA on sporulation. An internal promoter ensures the synthesis of an excess of SpoIISB over SpoIISA during exponential growth and sporulation. In the absence of SpoIISB, the sporulating cells show lethal damage of their envelope shortly after asymmetric septation, a defect that can be corrected by synthesizing SpoIISB only in the mother cell. However, forced synthesis of SpoIISA in exponentially growing cells or in the forespore leads to the same type of morphological damage and to cell death. In both cases protection against the killing effect of SpoIISA can be provided by simultaneous synthesis of SpoIISB. The spoIIS locus is unique to B. subtilis, and since it is completely dispensable for sporulation its physiological role remains elusive.

Genetic analysis of the T4 holin: timing and topology

The t protein of bacteriophage T4 shares with other holins the ability to cause the formation of a lethal membrane lesion which allows the phage endolysin to attack the peptidoglycan. Moreover, T, like other holins, acts in a saltatory manner at a precisely programmed time in the vegetative cycle. Unlike other holins, however, T has the unique ability to postpone its lethal function in response to a secondary infection by a T-even phage during the vegetative cycle. A signal transduction system that responds to the secondary infection is thought to be encoded by some of the numerous r genes, defined by mutations that abolish this lysis-inhibition (LIN) response. The primary structure of T differs from two main structural patterns found in more than 30 orthologous groups of holins. Genetic approaches were taken to probe the t sequence for features involved in membrane localization, functional timing and LIN regulation. Gene fusion analysis indicates that T has a single TMD near the N-terminus, with the bulk of the protein residing in the periplasm. Mapping and phenotypic analysis of deletion and point mutations in t indicates that the periplasmic domain of T is the major determinant of the timing mechanism and is involved in the LIN response.

Dimerization between the holin and holin inhibitor of phage lambda

Holins are integral membrane proteins that control the access of phage-encoded muralytic enzymes, or endolysins, to the cell wall by the sudden formation of an uncharacterized homo-oligomeric lesion, or hole, in the membrane, at a precisely defined time. The timing of lambda-infected cell lysis depends solely on the 107 codon S gene, which encodes two proteins, S105 and S107, which are the holin and holin inhibitor, respectively. Here we report the results of biochemical and genetic studies on the interaction between the holin and the holin inhibitor. A unique cysteine at position 51, in the middle of the second transmembrane domain, is shown to cause the formation of disulfide-linked dimers during detergent membrane extraction. Forced oxidation of membranes containing S molecules also results in the formation of covalently linked dimers. This technique is used to demonstrate efficient dimeric interactions between S105 and S107. These results, coupled with the previous finding that the timing of lysis depends on the excess of the amount of S105 over S107, suggest a model in which the inhibitor functions by titrating out the effector in a stoichiometric fashion. This provides a basis for understanding two evolutionary advantages provided by the inhibitor system, in which the production of the inhibitor not only causes a delay in the timing of lysis, allowing the assembly of more virions, but also increases effective hole formation after triggering.

Genetic and biochemical analysis of dimer and oligomer interactions of the lambda S holin

Bacteriophage lambda uses a holin-endolysin system for host cell lysis. R, the endolysin, has muralytic activity. S, the holin, is a small membrane protein that permeabilizes the inner membrane at a precisely scheduled time after infection and allows the endolysin access to its substrate, resulting in host cell lysis. lambda S has a single cysteine at position 51 that can be replaced by a serine without loss of the holin function. A collection of 27 single-cysteine products of alleles created from lambda S(C51S) were tested for holin function. Most of the single-cysteine variants retained the ability to support lysis. Mutations with the most defective phenotype clustered in the first two hydrophobic transmembrane domains. Several lines of evidence indicate that S forms an oligomeric structure in the inner membrane. Here we show that oligomerization does not depend on disulfide bridge formation, since the cysteineless S(C51S) (i) is functional as a holin and (ii) shows the same oligomerization pattern as the parental S protein. In contrast, the lysis-defective S(A52V) mutant dimerizes but does not form cross-linkable oligomers. Again, dimerization does not depend on the natural cysteine, since the cysteineless lysis-defective S(A52V/C51S) is found in dimers after treatment of the membrane with a cross-linking agent. Furthermore, under oxidative conditions, dimerization via the natural cysteine is very efficient for S(A52V). Both S(A52V) (dominant negative) and S(A48V) (antidominant) interact with the parental S protein, as judged by oxidative disulfide bridge formation. Thus, productive and unproductive heterodimer formation between the parental protein and the mutants S(A52V) and S(A48V), respectively, may account for the dominant and antidominant lysis phenotypes. Examination of oxidative dimer formation between S variants with single cysteines in the hydrophobic core of the second membrane-spanning domain revealed that positions 48 and 51 are on a dimer interface. These results are discussed in terms of a three-step model leading to S-dependent hole formation in the inner membrane.