Phage P68 Virion-Associated Protein 17 Displays Activity against Clinical Isolates of Staphylococcus aureus

Phage-encoded murein hydrolases are either part of the lysiscassette or found as structural components of the phage virion.Here, we show that Staphylococcus aureus bacteriophage P68 containsa virion-associated muralytic enzyme. Protein 17 has a compositestructure. The N-terminal part comprises the muralytic activity,whereas the C-terminal part is required for binding to the cellsurface. A high multiplicity of infection with phage P68 causedrapid lysis, and purified protein 17 triggered premature lysiswhen added to S. aureus cells prior to infection with P68, suggestingthat it functions to weaken the murein at the site of phageDNA entry. Protein 17 displayed activity against clinical S.aureus isolates, which are resistant to infection by phage P68,demonstrating that the protein targets surface structures distinctfrom the phage receptor. This broad activity spectrum of protein17 could qualify virion-associated muralytic enzymes as attractiveantimicrobials.

INTRODUCTION

Top

Introduction

In 1940, Max Delbrück described two different types ofphage-induced lysis (4) termed lysis from within (LI) and lysisfrom without (LO). LI occurs at a genetically determined timeafter infection. LI permits the release of phage progeny andgenerally depends on the activity of phage-encoded lysis genes(46). Double-stranded DNA phage, like phage ${lambda}$ , typically employa holin-endolysin system to release phage progeny (46, 47).Endolysins form a group of unrelated enzymes with differentmuralytic activities that generally accumulate fully foldedin the cytoplasm. Holins are small proteins, which form lesionsin the cytoplasmic membrane through which phage-encoded endolysinsgain access to the peptidoglycan at the end of the replicationcycle. In contrast, small RNA or DNA phage encode neither anendolysin nor a holin. Phage progeny release seems to be achievedby phage-encoded proteins, which perturb the activity of hostenzymes involved in peptidoglycan biosynthesis (3, 48).

LO was first described upon superinfection with phage T6 (4).Likewise, T4 virions caused rapid lysis of the host cell uponinfection with high multiplicities of infection (MOI) (4). PhageT4 encodes two different lysozyme activities. The T4 gene eproduct is made during late protein synthesis and is responsiblefor degradation of the peptidoglycan layer of the host cellenvelope during LI (35). Protein 5, with N-acetylmuramidaseactivity (13, 36), is anchored in the base plate structure ofthe virion tail (8, 14) and is responsible for LO. Similarly,protein P5, a lytic endopeptidase of the double-stranded RNAbacteriophage ${phi}$ 6 is associated with the nucleocapsid (30). Morerecently, the presence of peptidoglycan hydrolase activity hasbeen demonstrated in the virions of Staphylococcus aureus phage ${phi}$ 11 and ${phi}$ 85 (32) as well as in the Lactococcus lactis phage Tuc2009(16). In phage T7, the putative transglycosylase protein 16was shown to increase the efficiency of phage genome translocationinto the host cell (31, 33). In addition, homology searcheshave identified putative lytic transglycosylases among the structuralproteins of phage PRD1 and P1 (18). PRD1 virions have been shownto contain two activities, the lysins P15 and P7. Neither proteinis essential for infectivity, but P7 plays an accessory rolein genome penetration into infected cells (38, 39).

In this study, we have identified protein 17 of the S. aureuslytic phage P68 (45) as a structural component of the virionwith muralytic activity. The catalytic site was localized inthe N-terminal part of the protein, whereas the C-terminal partrepresents the substrate recognition and/or binding domain.Protein 17 is the first virion-associated component characterizedfrom a representative of the Podoviridae infecting S. aureuswith broad activity against clinical S. aureus strains.

MATERIALS AND METHODS

Top

Materials and Methods

Bacterial strains, phage, plasmids, and growth conditions. Bacterial strains, phage, and plasmids used in this study arelisted in Table 1. Staphylococcus aureus strains were grownin BHI broth (Merck) or on agar at 37°C. Escherichia colistrains were cultivated in Luria-Bertani (LB) broth at 37°C,and when appropriate, ampicillin (100 µg/ml) was added.Bacteriophage P68 was propagated on its S. aureus host 68 usingdouble-layer agar plates, and progeny viruses were purifiedusing CsCl density gradient centrifugation (40). P68 DNA wasextracted and purified as previously described (45). Plaqueformation of P68 on different S. aureus strains was assayedby following standard procedures (40).

View this table:
[in this window]
[in a new window]
TABLE 1. Bacterial strains, plasmids, phage, and oligonucleotides used in this study

DNA manipulations and plasmid construction. Plasmid DNA was prepared with the QIAGEN midi plasmid kit. PCRamplifications were carried out using Pwo polymerase (Peqlab)and the primers listed in Table 1. The PCR fragments were purifiedusing QIAquick spin PCR purification columns (QIAGEN). To assignthe enzymatically active domain of protein 17, two 3'-truncatedfragments of P68 gene 17 were generated by PCR.

The full-length N-terminally His₆-tagged protein 17 (653 aminoacids [aa]) was obtained as follows. The oligonucleotides I27and J27 (Table 1) were annealed to template P68 DNA. The resultingPCR product was cleaved with NdeI/XhoI and cloned into plasmidpET-16b restricted with NdeI/XhoI (Table 1). This created plasmidpMUH17, which harbors an in-frame fusion between His₆ codonsand the entire gene 17.

Amplification of gene 17 with oligonucleotides I27 and K27 (Table1) generated a 1,440-bp PCR fragment containing a 3'-terminallydeleted gene 17 (codons 1 to 480) flanked by the restrictionsites NdeI and XhoI. The NdeI/XhoI-restricted PCR fragment wascloned into pET-16b, resulting in plasmid pMUH17S. Likewise,the oligonucleotide pair I27-L27 (Table 1) was utilized to createa gene 17 fragment comprising codons 1 to 594. The 1,782-bpfragment was cloned into pET-16b, resulting in plasmid pMUH17L.All DNA cloning steps were initially performed with E. coliTOP 10.

Testing of S. aureus clinical isolates for methicillin resistance. S. aureus clinical isolates were tested for methicillin resistanceby the disk diffusion method using oxacillin as the class drug.Disk diffusion testing was performed according to the CLSI (formerlyNCCLS) guidelines (37) using 1-µg-oxacillin-containingdisks (Sigma-Aldrich) on Mueller-Hinton agar (Oxoid, England).The plates were incubated in ambient air at 35°C and readafter 24 h. Interpretive zone diameters for oxacillin were asfollows: $≤$ 10 mm was resistant and $≥$ 13 mm was susceptible. Intermediatezone sizes were between 10 and 13 mm. Any growth, includinglight growth within the 10-mm-diameter zone around the diskwas taken to indicate resistance.

MALDI mass spectrometry. Matrix-assisted laser desorption ionization (MALDI) mass spectrometrywas carried out on a Bruker Reflex III time of flight mass spectrometerequipped with a 26-sample SCOUT source and video system, a nitrogenUV laser ( ${lambda}$ _max = 336 nm), and dual-channel plate detector (BrukerDaltonik, Bremen, Germany) using standard procedures.

Purification of protein 17 and variants thereof. High-level synthesis of the His-tagged proteins 17 (653 aa),17S (486 aa; aa 1 to 480 of protein 17), and 17L (600 aa; aa1 to 594 of protein 17) was achieved in E. coli BL21(DE3) harboringplasmids pMUH17, pMUH17S, and pMUH17L, respectively. After overnightgrowth in Luria-Bertani broth at 37°C, the respective cultureswere diluted 20-fold with LB medium and incubated at 37°Cfor 4 h when expression of the cloned genes was induced by theaddition of isopropyl-ß-D-thiogalactopyranoside (IPTG)to a final concentration of 1 mM. After further incubation for3 h at 37°C, the cells were harvested by centrifugationin a Sorvall RC 5C centrifuge at 6,000 rpm for 20 min. The cellpellets were resuspended in 10 ml lysis buffer (300 mM NaCl,10 mM imidazole, 50 mM NaH₂PO₄, pH 8.0), and the cells werelysed by sonication (Branson Sonifier 250). The cell lysatewas centrifuged at 28,000 x g for 30 min, followed by loadingof the supernatant onto a 5-ml Ni-nitrilotriacetate-agarose(QIAGEN) column at 4°C. The N-terminal His-tagged proteinswere eluted from the column with 250 mM imidazole followed bydialysis in 25 mM Tris (pH 7.5).

SDS-polyacrylamide gel electrophoresis and zymogram assays. Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresiswas performed as described by Laemmli (19), and the zymogramassays were carried out as outlined by Lepeuple et al. (20).Briefly, 0.2% Staphylococcus aureus 68 autoclaved cells wereincluded in 10% polyacrylamide gels for detection of bacteriolyticactivity. P68 phage particles were solubilized in protein samplebuffer (62.5 mM Tris-HCl [pH 6.8], 2.3% SDS, 1% mercaptoethanol,10% glycerol, 0.01% bromophenol blue), heated for 10 min at100°C, and then separated on SDS-phosphonoacetic acid (PAA)gels containing the autoclaved cells. After electrophoresis,the zymograms were washed for 30 min with distilled water atroom temperature and then transferred into buffer containing25 mM Tris (pH 7.5) and 0.1% Triton X-100, followed by furtherincubation for 16 h at 37°C. The zymograms were rinsed withdistilled water, stained with 0.1% methylene blue in 0.001%KOH for 2 h at room temperature, and then destained with distilledwater. Peptidoglycan hydrolase activity was detected as a clearzone on a dark blue background. Proteins of the clear zone ofthe zymogram were eluted using the ElutaTube extraction anddialysis kit (MBI).

Antimicrobial activity of protein 17. The 50% inhibitory concentration of protein 17 was determinedon S. aureus strain 68, which was grown in brain heart infusion(BHI) broth at 37°C. Then 5 pg/ml, 50 pg/ml, 500 pg/ml,5 ng/ml, 50 ng/ml, or 50 µg/ml of protein 17 was addedto 1 x 10⁷ CFU. Twenty minutes after the addition of the proteins,serial dilutions were plated on BHI agar plates, and the numberof CFU was determined after bacterial growth for 16 h at 37°C.The experiment was performed in triplicate. The antimicrobialspectrum of protein 17 on 35 clinical S. aureus isolates wasdetermined similarly by growing the strains in BHI broth at37°C until they reached an optical density at 600 nm of0.5 ( $~$ 1 x 10⁷ CFU). Then protein 17 (50 ng/ml) was added, and20 min after the addition of the protein, serial dilutions wereplated on BHI agar plates. The number of CFU was determinedin triplicate for each strain after bacterial growth for 16h at 37°C.

Binding of protein 17 to the bacterial surface. A cell wall binding assay was used to determine the bindingof proteins 17, 17L, and 17S, respectively, to the bacterialsurface. Bacteria grown to late log phase were harvested bycentrifugation and concentrated 10-fold in phosphate-bufferedsaline (PBS) buffer (100 mM NaCl, 3 mM KCl, 10 mM NaH₂PO₄, 2mM KH₂PO₄, pH 7.5), and then stored on ice. One hundred microlitersof cells and 100 µl (50 ng) purified protein were mixedand incubated at room temperature for 10 min. Cells were thenremoved from the supernatant by centrifugation (16,000 x g,60 s). The cells were washed twice with 500 µl of PBS,and the pellet was resuspended in 50 µl PBS buffer. Thesame volume of protein sample buffer (62.5 mM Tris-HCl [pH 6.8],2.3% SDS, 1% mercaptoethanol, 10% glycerol, 0.01% bromophenolblue) was added, and samples were heated for 10 min at 100°Cand loaded onto an SDS-PAA gel. The presence of proteins 17,17L, and 17S in the cell pellet (bound protein) and in the supernatant(unbound protein) was determined by a standard immunodetectionprocedure using antibodies raised against whole-phage particles.

To remove proteins or lipids from the cell surface, S. aureusstrain 68 was subjected to various treatments. Cells from 50-mlaliquots of log-phase cultures resuspended in 250 µl PBSbuffer served as starting material for the following treatments:heating (121°C, 15 min), extraction with 4% SDS or 1% TritonX-100 (50°C, 30 min), extraction with 50% phenol-50% chloroform,or digestion with 20 µg proteinase K (50°C, 30 min).After these treatments, the cells were extensively washed withPBS buffer to remove any residual detergents, solvents, or enzymes.

RESULTS

Top

Results

Virion protein 17 of phage P68 has muralytic activity. To assess whether P68 virions contain a structural componentwith muralytic activity, zymogram assays were performed withautoclaved S. aureus 68 cells. Upon electrophoretic separationof P68 phage proteins, the zymogram revealed a single $~$ 70-kDaband with muralytic activity (Fig. 1, lane 1). The correspondingphage protein was electroeluted from the clear zone and againseparated on an SDS-PAA gel in parallel with purified phageP68 particles. As shown in Fig. 1, lane 4, only one proteinband appeared after elution from the zymogram, which appearedto correspond in size to the minor structural protein 17 ofP68 (M_r = 74,806) (45). The identity of protein 17 was confirmedby mass spectrometry of the protein eluted from the zymogramand of the corresponding phage virion protein eluted from theSDS-PAA gel. The determination of N- and C-terminal peptidesby mass spectrometry further demonstrated that the protein hadthe predicted molecular weight, i.e., that it was not processedat either end (not shown).

View larger version (119K):
[in this window]
[in a new window]
FIG. 1. Protein 17 of phage 68 has muralytic activity. 10¹¹ phage particles (lane 1) and 10 µg of lysostaphin (lane 2) were resolved in the zymogram as described in Materials and Methods. Proteins from the clear area of the zymogram (lane 1) were eluted and resolved on a 10% SDS-PAA gel (lane 4) in parallel with purified P68 particles (lane 3). The positions of molecular mass markers (in kilodaltons) are indicated to the right.

Computer-based similarity searches revealed that protein 17showed 27% similarity within a 493-amino-acid overlap to theminor structural protein 89 of Lactobacillus delbrueckii phageLL-H (29) and 25% similarity to the tail tip protein Tal₂₀₀₉of L. lactis phage Tuc2009, belonging to the Siphoviridae. TheTal₂₀₀₉ has recently been shown to possess cell wall-degradingactivity (16).

High MOI and addition of purified protein 17 trigger premature lysis of phage P68-infected cells. S. aureus cells infected with an MOI of 5 of phage P68 enteredthe lysis phase at approximately 45 to 50 min postinfection,and the culture cleared within the next 10 min (Fig. 2). Incontrast, infection with a high MOI ( $~$ 100) of P68 virions caused,similarly to E. coli phage T4 (4), nearly complete lysis ofS. aureus cells within 15 min of infection (Fig. 2), indicatingthat protein 17 may induce LO. When 50 ng/ml of protein 17 wasadded to P68-infected S. aureus cells 10 min prior to infection,premature lysis was observed. Premature lysis was also triggeredby addition of protein 17 at 20 min postinfection, whereas theaddition of protein 17 immediately before the start of vegetativelysis had no effect. These experiments suggested that P17 weakensthe peptidoglycan from without.

View larger version (17K):
[in this window]
[in a new window]
FIG. 2. Infection of S. aureus strain 68 with an MOI of 100 (•) and an MOI of 5 ( ${circ}$ ) at time zero culminated in lysis 15 and 40 min thereafter, respectively. Purified protein 17 (50 ng/ml) was added 10 min prior to infection ( ${square}$ ) or 20 min postinfection ( ${triangleup}$ ) with phage P68 or immediately before the onset of lysis (x). Results from a representative experiment of three independent experiments are shown. OD₆₀₀, optical density at 600 nm.

Antimicrobial activity of protein 17. In further experiments, we tested the antimicrobial activityof purified protein 17 by exposing 1 x 10⁷ CFU of exponentiallygrowing S. aureus 68 cells to different concentrations of theprotein for 20 min. The addition of 5 pg/ml of protein 17 toS. aureus 68 cells had no measurable effect (Fig. 3). The additionof 50 pg/ml, 500 pg/ml, 5 ng/ml, 50 ng/ml, and 50 µg/mlof protein 17 to the S. aureus culture resulted in a reductionof the plating efficiency of 25% ± 5%, 50% ± 2%,90% ± 3%, 99% ± 1%, and 99.5% ± 0.2%, respectively(Fig. 3). From these results, we calculated the 50% inhibitoryconcentration for protein 17 as $~$ 500 pg/ml.

View larger version (22K):
[in this window]
[in a new window]
FIG. 3. Antimicrobial activity of protein 17. The antimicrobial activity of protein 17 was tested by exposing 1 x 10⁷ CFU of exponentially growing S. aureus strain 68 for 20 min to 0 pg/ml, 5 pg/ml, 50 pg/ml, 500 pg/ml, 5 ng/ml, 50 ng/ml, and 50 µg/ml of protein 17. The experiment was performed in triplicate. The error bars represent standard deviations.

The antimicrobial spectrum of protein 17 was further testedon 35 clinical isolates of S. aureus, $~$ 50% of which were oxacillinresistant (Table 2). Purified protein 17 (50 ng/ml) was addedto exponentially growing cells (1 x 10⁷ CFU), and serial dilutionsof the cells were plated after 20 min of exposure. This treatmentresulted in a reduction in the plating efficiency of $~$ 99% ofthe majority of the tested clinical S. aureus isolates (Table2).

View this table:
[in this window]
[in a new window]
TABLE 2. Host range of S. aureus phage P68 and antimicrobial spectrum of purified protein 17

Protein 17 showed a broader antimicrobial spectrum comparedto the host range of phage P68. As shown in Table 2, the phageplated only on 11 of the clinical strains, whereas protein 17displayed activity against 26 strains. This suggested that thebacterial receptor required for phage attachment differs fromthe ligand required for protein 17 binding. Interestingly, thephage plated on two S. aureus isolates (P5 and P27) (Table 2)which were insensitive to protein P17. This finding could indicatethat P17 is not necessarily essential for infection by P68 ofevery S. aureus strain.

The C terminus of protein 17 is required for binding to the cell surface. Many muralytic enzymes of gram-positive bacteria and their phagehave a composite structure with the catalytic active site locatedin the N-terminal region, while the C-terminal part containsthe target-specific binding domain (1, 12, 22, 42, 44). To testwhether protein 17 has a cell surface binding domain, two C-terminallytruncated variants of protein 17, lacking the C-terminal 53aa (protein 17L) and 167 aa (protein 17S), respectively, weregenerated by means of recombinant DNA technology and purified.As shown in Fig. 4, both truncated versions of protein 17 displayedmurein hydrolase activity in zymograms containing murein ofS. aureus strain 68, demonstrating that the catalytic site ofthe protein 17 is located within the first 480 aa residues.

View larger version (60K):
[in this window]
[in a new window]
FIG. 4. Enzymatic activity of protein 17 resides in the N-terminal part. The muralytic activities of proteins 17 (lane 1), 17S (lane 2), and 17L (lane 3), respectively, were tested in zymograms containing murein of S. aureus strain 68.

The addition of 50 ng/ml of truncated proteins 17S and 17L,respectively, to exponentially growing S. aureus 68 cells andto the clinical strains susceptible to protein 17 did not affecttheir viability (not shown), which led us to test whether celltargeting is abolished in 17S and 17L, respectively. Fifty nanogramsof either protein 17, 17S, or 17L was added to S. aureus strain68, and binding was assessed. As shown in Fig. 5A, both C-terminallytruncated variants of protein 17 did not bind to the cells,demonstrating that the C terminus is indeed required for celltargeting of protein 17. Moreover, in agreement with the antimicrobialspectrum (Table 2), protein 17 did not bind to the clinicalstrains P12 and P27, whereas binding was observed to the protein17-sensitive strains P6 and P21 (Fig. 5B).

View larger version (44K):
[in this window]
[in a new window]
FIG. 5. C-terminal part of protein 17 is required for binding to the cell surface. Binding of protein 17 to the cell surface was revealed by means of immunodetection of protein 17 in Western blots using antibodies raised against whole-phage particles. Only the relevant parts of the Western blots are shown. P, the respective protein was found in the cell pellet, i.e., the protein associated with the cell surface; S, the respective protein was only found in the supernatant, i.e., did not bind to the cell surface. (A) Binding of purified proteins 17 (lanes 2 to 4), 17L (lanes 5 to 7), and 17S (lanes 8 to 10) to S. aureus 68 cells. Lane 1, protein 17 was omitted. (B) Binding of purified protein 17 to S. aureus P6, P21, P12, and P27 cells (Table 2). (C) Binding of protein 17 (lower panel) and absorption of phage P68 (upper panel) to untreated S. aureus 68 cells (lane 1) and S. aureus 68 cells treated with proteinase K (PK; lane 2), SDS (lane 3), trichloroacetic acid (TCA; lane 4), and phenol (lane 5). The cell pellet was tested for the presence of protein 17.

Ligand required for binding of protein 17 to the cell surface. To assess the nature of the ligand required for binding of protein17 to the bacterial surface, several treatments (heating, extractionwith solvents and/or detergents, and protease digestion) wereapplied to remove proteins, lipids, and membrane-associatedcomponents such as lipoteichoic acids from the cell wall ofS. aureus strain 68. None of these treatments had an obviouseffect on binding of protein 17 (Fig. 5C). On the other hand,phenol extraction of lipoteichoic acids from S. aureus 68 cellsrendered the cells resistant to phage P68 (Fig. 5C). Taken together,these experiments suggested that lipoteichoic acids serve asphage receptors and that protein 17 may target a carbohydrateligand, which is not primarily required for phage absorption.On the other hand, they also indicated that P17 is not ableto bind stably to the cell wall when it is part of the virion.

DISCUSSION

Top

Discussion

The life cycle of a bacteriophage begins with the insertionof its genetic material into the host and generally culminatesin host cell lysis to allow progeny release. The peptidoglycanis a barrier for the transport of macromolecules and preventsdiffusion of globular proteins larger than $~$ 50 kDa (5, 7). T-evenphage particles have long been known to contain a lysozyme activity(protein 5) in the baseplate. It has been suggested that thelysozyme activity of protein 5 locally digests the peptidoglycan,allowing the T4 tail tube to penetrate the periplasmic spaceand therefore allowing the infection process to be completed(15, 34, 36). More recently, Kenny et al. (16) have demonstratedthat the L. lactis phage Tuc2009 tail tip protein Tal₂₀₀₉ hasmuralytic activity. Tal₂₀₀₉ appears to undergo autocatalyticprocessing, which was attributed to its endopeptidase activity.In this study, we have shown that the virion-associated minorstructural protein 17 of S. aureus phage P68 has muralytic activity.Infection of S. aureus with a high MOI of P68 (Fig. 2) resultedin lysis. In contrast, the same experiment conducted with thephage P68-resistant clinical isolates P4 and P29 did not resultin lysis (not shown), although the latter were sensitive tosoluble P17 (Table 2). This suggests that P68-bound P17 canonly exert its activity upon adsorption of the phage to thehost. Similarly, the addition of the protein 10 min prior toinfection or 20 min postinfection triggered premature lysisof P68-infected S. aureus cells. Taken together, these experimentsindicate that protein 17 causes a localized rupture in the cellwall, which at the initial state of infection, facilitates penetrationof the phage DNA into the cytoplasm. In contrast to the Tal₂₀₀₉protein, we found no evidence for a putative protease motifin P17 and did not observe any evidence for processing of theprotein. Purified P17 was stable, and the MALDI-time of flightdata as well as the apparent molecular weight in SDS-polyacrylamidegels agreed with that deduced from the primary protein sequence.

P68 protein P17 differs in another aspect from Tal₂₀₀₉ (16)in that it binds to the cell wall (Fig. 3). In gram-positivebacteria, lysozymes or lytic transglycosylases generally containan N-terminal catalytic domain, while the C-terminal part containsthe target-specific binding domain (10, 42, 44). These C-terminaldomains are composed of amino acid repeats or sequence motifsrequired for anchoring of the enzymes to the bacterial cellsurface, such as (i) an LPXTG motif for covalent linkage tothe peptidoglycan cross-bridges by a sortase function (6, 41),(ii) a hydrophobic C-terminal region for anchoring to the membrane(17), (iii) sequence repeats, in particular those starting witha GW motif for binding to lipoteichoic acids (11), (iv) a bacterialS-layer homology domain for anchoring proteins to modified secondarycell wall polymers (25, 28), and (v) LysM domains, which arethought to contain a general peptidoglycan-binding motif (2).Only full-length protein 17 was shown to bind to and to exertactivity against S. aureus, whereas proteins 17S and 17L displayedmuralytic activity only in the zymogram assay, wherein the catalyticdomain can apparently directly access the peptidoglycan substrate.Our studies indicate that protein 17 has a bipartite structurewith an N-terminal catalytic site and a C-terminal surface bindingdomain. Nevertheless, no typical surface binding motifs wererevealed for the C-terminal end. Two imperfect GW motifs werefound at the C terminus of protein 17 (not shown). Since GWmodules are apparently required for binding to lipoteichoicacids (11) and removal of the latter did not affect bindingof protein 17 (Fig. 5C), it is, however, less likely that theyare involved in targeting of protein 17 to cells. Analogousto endolysins encoded by the L. monocytogenes phage Ply118 andPly500 (23), the ligand required for binding of protein 17 tothe cell wall appears to be different from the phage receptor(Table 2 and Fig. 5C). Since the removal of proteins, lipids,and membrane-associated components from the bacterial cell walldid not affect binding of protein 17, we speculate that thecell ligand for protein 17 is a carbohydrate component, whichis perhaps covalently attached to the peptidoglycan.

Staphylococcus aureus is an important human pathogen responsiblefor a wide variety of diseases. S. aureus is a common causeof nosocomial pneumonia and bloodstream infections as well ascommunity-acquired infections and is known for its resistanceto first-line antibiotics such as penicillin (24, 26), whichunderscores the need for the development of alternatives toantibiotics. Phage therapy has recently received renewed attentionas an alternative for prevention and/or treatment of bacterialinfections (27, 43). Alternatively, phage-encoded endolysinsmight provide effective agents for the control of infectiousdiseases caused by gram-positive bacteria (9), the peptidoglycanof which is externally accessible. Like phage therapy, the specificityof phage-encoded endolysins offers the advantage of targetedkilling of pathogens while leaving the beneficial microfloraintact. For instance, the muramidase Cpl-1 derived from a pneumococcalphage was shown to be efficient in clearing pneumococci fromthe nasopharyngeal mucosa when applied topically but also killedS. pneumoniae in the bloodstream and significantly extendedthe survival of the infected mice (21). Protein 17 displayedactivity against clinical S. aureus isolates, some of whichwere methicillin resistant. As apparent from Table 2, theseagents may have an even broader activity spectrum than the phagefrom which they are derived. It seems, therefore, conceivableto develop virion-associated murein hydrolases toward effectiveantimicrobial agents to combat gram-positive pathogens in asimilar way as shown for phage-derived endolysins by Fischetti(9).

ACKNOWLEDGMENTS

We are grateful to M. Loessner, Technical University of Zürich,for critical reading of the manuscript.

FOOTNOTES

* Corresponding author. Mailing address: Max F. Perutz Laboratories, Department of Microbiology and Immunobiology, University Departments at the Vienna Biocenter, Dr. Bohrgasse 9/4, 1030 Vienna, Austria. Phone: 43 1 4277 54609. Fax: 43 1 4277 9546. E-mail: udo.blaesi{at}univie.ac.at.

REFERENCES

Top