
Full text loading...
Category: Bacterial Pathogenesis; Microbial Genetics and Molecular Biology
Phage Lysis, Page 1 of 2
< Previous page | Next page > /docserver/preview/fulltext/10.1128/9781555816506/9781555813079_Chap06-1.gif /docserver/preview/fulltext/10.1128/9781555816506/9781555813079_Chap06-2.gifAbstract:
This chapter is aimed at achieving a useful overview of phage lysis, with perhaps more emphasis on rationale and perspective than on molecular detail. Host lysis is the only macroscopic real-time phenotype in phage biology. In addition, some phages of gram-positive bacteria have lysozymes with endopeptidase activities directed against one of several different peptide linkages in the oligopeptide cross-links. Recently, a dual-start holin-antiholin system was described for the Listeria phage A118. Detailed molecular and genetic studies on the lysis of dsDNA phages have been pursued only for λ, T4, and more recently, P1 and P2. A section of the chapter describes the current state of our understanding of these four systems and then covers other phage systems, for which there is rarely more than fragmentary information. It is the thesis of this chapter that phage lysis is a highly evolved, sophisticated process which, along with its biological bookend, the adsorption-penetration process, has never received sufficient attention commensurate with the richness of its physiological, molecular, and phylogenetic character. Thus, the future is bright, and the coming years should reveal even more fascinating new elements of the phage lysis phenomenon. It is time that humans pay proper attention to a biological event that happens perhaps 1028 times per day in our biosphere.
Full text loading...
Gram-negative bacterial peptidoglycan structure and sites of cleavage by various endolysins. Abbreviations: GlcNac, N-acetylglucosamine; MurNac, N-acetylmuramic acid; L- and D-Ala, L- and D-alanine; D-Glu, D-glutamic acid; m-DAP, meso-diaminopimelic acid; Lys, lysine. Adapted from reference 130 with permission.
Gram-negative bacterial peptidoglycan structure and sites of cleavage by various endolysins. Abbreviations: GlcNac, N-acetylglucosamine; MurNac, N-acetylmuramic acid; L- and D-Ala, L- and D-alanine; D-Glu, D-glutamic acid; m-DAP, meso-diaminopimelic acid; Lys, lysine. Adapted from reference 130 with permission.
(A) Holin and antiholin topologies. The membrane topologies of class I (λ S105, P1 LydA, and P2 Y), class II (S21 68), and class III (T4 gpt) holins are shown, along with those of their cognate antiholins. The topologies shown for the S gene products (J. Deaton and R. Young, unpublished data) ( 41 , 42 , 45 ), the S 21 gene product, LydB (Xu et al., unpublished data), and the T4 gpt proteins (Tran and Young, unpublished data) are supported by experimental data, whereas the other topologies are predictions based on analyses of the primary structures. (B) Topology and landmarks of the S105 holin of λ. Numbering of the residues begins with position 3 and ends with the C-terminal residue 107. Each TMD is depicted as a helical projection according to the best estimate of the number of embedded residues, based on hydropathy and TMD prediction algorithms at the ExPASy website (http://www.expasy.org) ( 39 ) and on chemical accessibility experiments ( 44 ). The numbers at the beginning and end of each TMD represent the probable first residues in the bilayer. TMD1 is shown tilted to the plane of the membrane because of its predicted length of 24 residues. Charged residues are indicated along the polypeptide chain, and the formylated state of the N terminus of S105 is shown. Residues at which lysis-defective mutations have a dominant-negative character (i.e., delayed lysis in the presence of the wt allele) are shown in gray. Residues at which lysis-defective mutations have an early dominant character are shown as squares; early dominant alleles are nonlytic alleles that accelerate lysis in the presence of the wt (see text). Gly22 is shown as a gray box because both dominant- negative and early dominant alleles exist at this position ( 93 ).The three Ala residues along one face of TMD2 (Ala48,Ala52, and Ala55) and Cys51 are indicated by single-letter codes.
(A) Holin and antiholin topologies. The membrane topologies of class I (λ S105, P1 LydA, and P2 Y), class II (S21 68), and class III (T4 gpt) holins are shown, along with those of their cognate antiholins. The topologies shown for the S gene products (J. Deaton and R. Young, unpublished data) ( 41 , 42 , 45 ), the S 21 gene product, LydB (Xu et al., unpublished data), and the T4 gpt proteins (Tran and Young, unpublished data) are supported by experimental data, whereas the other topologies are predictions based on analyses of the primary structures. (B) Topology and landmarks of the S105 holin of λ. Numbering of the residues begins with position 3 and ends with the C-terminal residue 107. Each TMD is depicted as a helical projection according to the best estimate of the number of embedded residues, based on hydropathy and TMD prediction algorithms at the ExPASy website (http://www.expasy.org) ( 39 ) and on chemical accessibility experiments ( 44 ). The numbers at the beginning and end of each TMD represent the probable first residues in the bilayer. TMD1 is shown tilted to the plane of the membrane because of its predicted length of 24 residues. Charged residues are indicated along the polypeptide chain, and the formylated state of the N terminus of S105 is shown. Residues at which lysis-defective mutations have a dominant-negative character (i.e., delayed lysis in the presence of the wt allele) are shown in gray. Residues at which lysis-defective mutations have an early dominant character are shown as squares; early dominant alleles are nonlytic alleles that accelerate lysis in the presence of the wt (see text). Gly22 is shown as a gray box because both dominant- negative and early dominant alleles exist at this position ( 93 ).The three Ala residues along one face of TMD2 (Ala48,Ala52, and Ala55) and Cys51 are indicated by single-letter codes.
The S holin kills without warning. Cells tethered to coverslips by antiflagellin antibodies and carrying an inducible clone of the λ lysis cassette were induced at time zero. Single frames were chosen from recordings of two representative cells and are depicted here to illustrate the process of cell lysis. Starting from the time points indicated to the left of the panels, single frames were captured every 200 ms. After induction of the lysis genes, the tethered cells rotate at a high and constant speed (first row). About 20 min after induction, the rotation of the cell abruptly slows and stops completely within 1 to 3 s (second row). Cell lysis, due to digestion of the cell wall by the λ R endolysin, occurs within seconds after the sudden stop in rotation (third and fourth rows). Adapted from reference 46 with permission.
The S holin kills without warning. Cells tethered to coverslips by antiflagellin antibodies and carrying an inducible clone of the λ lysis cassette were induced at time zero. Single frames were chosen from recordings of two representative cells and are depicted here to illustrate the process of cell lysis. Starting from the time points indicated to the left of the panels, single frames were captured every 200 ms. After induction of the lysis genes, the tethered cells rotate at a high and constant speed (first row). About 20 min after induction, the rotation of the cell abruptly slows and stops completely within 1 to 3 s (second row). Cell lysis, due to digestion of the cell wall by the λ R endolysin, occurs within seconds after the sudden stop in rotation (third and fourth rows). Adapted from reference 46 with permission.
Death raft model of holin lesions. Holins (gray circles) are depicted accumulating in two-dimensional rafts in the plane of the bilayer in this top-down view. At some point dependent on the overall intramolecular and intermolecular interhelical interactions of the raft of holins, a transitory spontaneous channel opens up, leading to a local de-energization of the membrane, which in turn triggers the neighboring holins, leading rapidly to hole formation, a largely quaternary rearrangement. If an uncoupler is added early in the accumulation phase, the triggering leads to small holes that are capable of releasing endolysin (gray circles with jaws). However, mature, spontaneously triggered holes are large enough to allow the release of endolysin– β-Gal tetrameric hybrids (endolyins with dotted ovals). Adapted from reference 120 with permission.
Death raft model of holin lesions. Holins (gray circles) are depicted accumulating in two-dimensional rafts in the plane of the bilayer in this top-down view. At some point dependent on the overall intramolecular and intermolecular interhelical interactions of the raft of holins, a transitory spontaneous channel opens up, leading to a local de-energization of the membrane, which in turn triggers the neighboring holins, leading rapidly to hole formation, a largely quaternary rearrangement. If an uncoupler is added early in the accumulation phase, the triggering leads to small holes that are capable of releasing endolysin (gray circles with jaws). However, mature, spontaneously triggered holes are large enough to allow the release of endolysin– β-Gal tetrameric hybrids (endolyins with dotted ovals). Adapted from reference 120 with permission.
Lysis cassette of phage λ. The lysis genes S, R, Rz, and Rz1 are shown, drawn to scale, with the λ late promoter, pR′, and an enlargement depicting the dual translational starts of S leading to the synthesis of the holin (S105) and antiholin (S107). Adapted from reference 130 with permission.
Lysis cassette of phage λ. The lysis genes S, R, Rz, and Rz1 are shown, drawn to scale, with the λ late promoter, pR′, and an enlargement depicting the dual translational starts of S leading to the synthesis of the holin (S105) and antiholin (S107). Adapted from reference 130 with permission.
Lysis genes of P1, P2, and T7. Genes are drawn to scale, with sizes (in numbers of encoded amino acids) and gene names indicated. Holin and antiholin genes have diagonal striping, with the antiholin genes additionally shaded. Endolysin genes have solid shading. The Rz- and Rz1-like genes are dotted or have vertical striping, with the Rz1 embedded gene additionally shaded. Arrows indicate known late transcripts and promoters. None of the lysis genes in this figure share any significant homology with each other, except lydA and lydC of P1. In terms of distinct gene families, there are thus three distinct holin types, three distinct endolysin types, and two unrelated Rz-Rz1 gene pairs. Adapted from reference 131 with permission of the publisher.
Lysis genes of P1, P2, and T7. Genes are drawn to scale, with sizes (in numbers of encoded amino acids) and gene names indicated. Holin and antiholin genes have diagonal striping, with the antiholin genes additionally shaded. Endolysin genes have solid shading. The Rz- and Rz1-like genes are dotted or have vertical striping, with the Rz1 embedded gene additionally shaded. Arrows indicate known late transcripts and promoters. None of the lysis genes in this figure share any significant homology with each other, except lydA and lydC of P1. In terms of distinct gene families, there are thus three distinct holin types, three distinct endolysin types, and two unrelated Rz-Rz1 gene pairs. Adapted from reference 131 with permission of the publisher.
SAR domains of phage endolysins. The N-terminal sequences of a representative group of endolysins encoded by phages or prophages are shown. The SAR sequences demonstrated for P1 Lyz and 21 R and putative SAR sequences of other endolysins are underlined and shown in bold. Basic and acidic residues are highlighted by shaded and shaded italic letters, respectively. All of these endolysins are homologs of the bacteriophage T4 and P22 lysozymes, whose sequences are included for comparison. The position of the catalytic Glu residue, experimentally identified in the soluble lysozymes and in P1 Lyz and predicted by the alignments in the other sequences, is indicated by an asterisk. Adapted from reference 126 with permission of the publisher.
SAR domains of phage endolysins. The N-terminal sequences of a representative group of endolysins encoded by phages or prophages are shown. The SAR sequences demonstrated for P1 Lyz and 21 R and putative SAR sequences of other endolysins are underlined and shown in bold. Basic and acidic residues are highlighted by shaded and shaded italic letters, respectively. All of these endolysins are homologs of the bacteriophage T4 and P22 lysozymes, whose sequences are included for comparison. The position of the catalytic Glu residue, experimentally identified in the soluble lysozymes and in P1 Lyz and predicted by the alignments in the other sequences, is indicated by an asterisk. Adapted from reference 126 with permission of the publisher.
Activation of the SAR endolysin, Lyz, of bacteriophage P1. Lyz is first exported by the host sec system, resulting in a membrane-tethered form of the endolysin. The muralytic activity is suppressed because Cys51, required to be in the sulfhydryl state for catalytic activity, is in disulfide linkage with Cys44 and because the disulfide bond effectively cages the active site. When the membrane potential collapses, the SAR TMD is released from the membrane, liberating the buried Cys13 and resulting in a disulfide bond isomerization that frees the catalytic SH group and permits a conformational change that creates the active site (indicated by a star).
Activation of the SAR endolysin, Lyz, of bacteriophage P1. Lyz is first exported by the host sec system, resulting in a membrane-tethered form of the endolysin. The muralytic activity is suppressed because Cys51, required to be in the sulfhydryl state for catalytic activity, is in disulfide linkage with Cys44 and because the disulfide bond effectively cages the active site. When the membrane potential collapses, the SAR TMD is released from the membrane, liberating the buried Cys13 and resulting in a disulfide bond isomerization that frees the catalytic SH group and permits a conformational change that creates the active site (indicated by a star).
(A) Expression of rI specifically blocks lysis mediated by gpt. Lysogens carrying λt::ΔS were thermally induced to begin the phage lytic cycle (empty squares and filled circles) and supplemented with IPTG (isopropyl- β-D-thiogalactopyranoside) to induce rI cloned under the control of tacPo in a pBR322-derived vector (filled circles) (Tran and Young, unpublished data; also see the work of Ramanculov and Young [ 94 ]). (B) Model of the role of gprI in LIN. Normally, gprI is a highly unstable periplasmic protein. When injection of the DNA of a secondarily infecting T4 phage particle is blocked by the primary infection (presumably through the Imm protein [data not shown]), the phage DNA and/or internal capsid proteins are ectopically localized to the periplasm. By an unknown mechanism, this in turn results in the stabilization of gprI, which then binds to and inhibits the holin, gpt. gprI is shown as a secreted periplasmic protein, but it is also possible that it is tethered to the membrane by an N-terminal TMD.The gprI-gpt antiholin-holin complex is blocked from proceeding down the pathway leading to oligomeric hole formation, which, if allowed to proceed, results in saltatory membrane disruption and release of the cytoplasmic gpe endolysin ( 94 ).
(A) Expression of rI specifically blocks lysis mediated by gpt. Lysogens carrying λt::ΔS were thermally induced to begin the phage lytic cycle (empty squares and filled circles) and supplemented with IPTG (isopropyl- β-D-thiogalactopyranoside) to induce rI cloned under the control of tacPo in a pBR322-derived vector (filled circles) (Tran and Young, unpublished data; also see the work of Ramanculov and Young [ 94 ]). (B) Model of the role of gprI in LIN. Normally, gprI is a highly unstable periplasmic protein. When injection of the DNA of a secondarily infecting T4 phage particle is blocked by the primary infection (presumably through the Imm protein [data not shown]), the phage DNA and/or internal capsid proteins are ectopically localized to the periplasm. By an unknown mechanism, this in turn results in the stabilization of gprI, which then binds to and inhibits the holin, gpt. gprI is shown as a secreted periplasmic protein, but it is also possible that it is tethered to the membrane by an N-terminal TMD.The gprI-gpt antiholin-holin complex is blocked from proceeding down the pathway leading to oligomeric hole formation, which, if allowed to proceed, results in saltatory membrane disruption and release of the cytoplasmic gpe endolysin ( 94 ).
Lysis genes of lytic single-stranded phages. For the ssDNA microvirus ϕX174 and ssRNA leviviruses MS2, AP205, and Qβ, the lysis gene is indicated by shading. L and E are both located in alternate reading frames within other essential genes. A′ in the AP205 genome is thought to be the initial portion of the maturation protein gene A and is connected to A by a translational frame shift ( 67 ).
Lysis genes of lytic single-stranded phages. For the ssDNA microvirus ϕX174 and ssRNA leviviruses MS2, AP205, and Qβ, the lysis gene is indicated by shading. L and E are both located in alternate reading frames within other essential genes. A′ in the AP205 genome is thought to be the initial portion of the maturation protein gene A and is connected to A by a translational frame shift ( 67 ).
ϕX174 E and Qβ A2 inhibit separate enzymes, MraY and MurA, respectively, of the murein synthesis pathway. The soluble cytoplasmic pathway begins with UDP-GlcNac (N-acetylglucosamine), catalyzed by MurA, and ends with the production of a UDP-MurNac (N-acetylmuramic acid) pentapeptide (shown as “pp”).The membrane-embedded enzyme MraY transfers the monosaccharide pentapeptide moiety to the C55 carrier, undecaprenol phosphate (indicated with “squiggles”), by a pyrophosphate linkage, creating lipid I, from which the final precursor, lipid II, is formed by the MurG-catalyzed addition of GlcNac. Lipid II is “flipped” to the outer surface of the cytoplasmic membrane, where the disaccharide-pentapeptide unit is attached to the peptidoglycan by the transglycosylase (TG) activity of a penicillin-binding protein (PBP). Subsequently, cross-linking of the glycan chains through the oligopeptides is catalyzed by the transpeptidase (TP) activities of PBPs, along with other unknown maturation events. Question marks indicate that it is not known how lipid II is flipped to the outer surface of the cytoplasmic membrane or how the unloaded C55 carrier, undecaprenol-P, is returned to the cytoplasm. Stars represent the lysis proteins E and A2.
ϕX174 E and Qβ A2 inhibit separate enzymes, MraY and MurA, respectively, of the murein synthesis pathway. The soluble cytoplasmic pathway begins with UDP-GlcNac (N-acetylglucosamine), catalyzed by MurA, and ends with the production of a UDP-MurNac (N-acetylmuramic acid) pentapeptide (shown as “pp”).The membrane-embedded enzyme MraY transfers the monosaccharide pentapeptide moiety to the C55 carrier, undecaprenol phosphate (indicated with “squiggles”), by a pyrophosphate linkage, creating lipid I, from which the final precursor, lipid II, is formed by the MurG-catalyzed addition of GlcNac. Lipid II is “flipped” to the outer surface of the cytoplasmic membrane, where the disaccharide-pentapeptide unit is attached to the peptidoglycan by the transglycosylase (TG) activity of a penicillin-binding protein (PBP). Subsequently, cross-linking of the glycan chains through the oligopeptides is catalyzed by the transpeptidase (TP) activities of PBPs, along with other unknown maturation events. Question marks indicate that it is not known how lipid II is flipped to the outer surface of the cytoplasmic membrane or how the unloaded C55 carrier, undecaprenol-P, is returned to the cytoplasm. Stars represent the lysis proteins E and A2.
(Top) Sequences of the E lysis proteins of ϕX174 and other Microviridae, with conserved residues shaded ( 4 , 40 , 69 , 70 ). The hydrophobic membrane region is indicated by a shaded box, and the C-terminal region which can be replaced by heterologous protein domains is indicated by a striped box. (Bottom) The membrane topologies of MraY and E are shown, as deduced from gene fusion and bioinformatic analyses ( 20 , 82 ). Mutations selected for their resistance to inhibition by E are shown mapped to two TMDs of MraY ( 10 ;T. Bernhardt and R.Young, unpublished data).
(Top) Sequences of the E lysis proteins of ϕX174 and other Microviridae, with conserved residues shaded ( 4 , 40 , 69 , 70 ). The hydrophobic membrane region is indicated by a shaded box, and the C-terminal region which can be replaced by heterologous protein domains is indicated by a striped box. (Bottom) The membrane topologies of MraY and E are shown, as deduced from gene fusion and bioinformatic analyses ( 20 , 82 ). Mutations selected for their resistance to inhibition by E are shown mapped to two TMDs of MraY ( 10 ;T. Bernhardt and R.Young, unpublished data).
In the top section, the predicted L protein sequences of the leviviruses MS2, fr, GA, and KU1, which infect F+ enterobacteria, are aligned. Basic and acidic residues are highlighted in gray and underlined, respectively. A hydrophobic region with a zero net predicted charge that is capable of forming an integral membrane domain is indicated by a double line above the aligned sequences. Conserved residues are shown below the alignment. The “Δ” symbol indicates the end of an N-terminal deletion which retains lytic function; in this deletion, the vector sequence MAEF . . . is fused to the RSST . . . sequence of L ( 9 ). Below, the putative L protein from AP205, a male-specific RNA phage from Acinetobacter, is shown ( 67 ).
In the top section, the predicted L protein sequences of the leviviruses MS2, fr, GA, and KU1, which infect F+ enterobacteria, are aligned. Basic and acidic residues are highlighted in gray and underlined, respectively. A hydrophobic region with a zero net predicted charge that is capable of forming an integral membrane domain is indicated by a double line above the aligned sequences. Conserved residues are shown below the alignment. The “Δ” symbol indicates the end of an N-terminal deletion which retains lytic function; in this deletion, the vector sequence MAEF . . . is fused to the RSST . . . sequence of L ( 9 ). Below, the putative L protein from AP205, a male-specific RNA phage from Acinetobacter, is shown ( 67 ).
Amino acid sequence, charged residues, and three predicted TMDs (gray) of holin P35 of the lipid phage PRD1 ( 103 ).The positions of amber mutations [K8(Am) and W77(Am)] are indicated by stars, and the lysis phenotypes in different suppressor hosts are shown above the sequence. The missense mutants H1 (D81Y) and H3/H4 (W77C) are indicated by black rectangles, with lysis phenotypes shown below. The insertion sites of mini-Mu transposons ( 117 ) causing truncations and, in some cases, additions of short sequences are indicated by arrows. In each case, the net change in the predicted charge in the insertion mutant is indicated.
Amino acid sequence, charged residues, and three predicted TMDs (gray) of holin P35 of the lipid phage PRD1 ( 103 ).The positions of amber mutations [K8(Am) and W77(Am)] are indicated by stars, and the lysis phenotypes in different suppressor hosts are shown above the sequence. The missense mutants H1 (D81Y) and H3/H4 (W77C) are indicated by black rectangles, with lysis phenotypes shown below. The insertion sites of mini-Mu transposons ( 117 ) causing truncations and, in some cases, additions of short sequences are indicated by arrows. In each case, the net change in the predicted charge in the insertion mutant is indicated.