Introduction

Recently, there has been a renewed interest in phage-bacteria interactions because phages have not only profound influence on the biology of bacterial pathogens1,2. Despite comprehensive studies on phage genomes15 and the role of S. aureus phages in horizontal transfer of resistance and virulence genes among clones and species12, the molecular interactions mediating phage adsorption to the staphylococcal cell surface remain poorly understood.

The mechanism underlying S. aureus phage adsorption has often been assumed to be similar to that of phages infecting gram-negative bacteria. However, as gram-positive bacteria have a very different cell wall structure compared to that of gram-negative bacteria, phages infecting gram-positive bacteria may employ adsorption mechanisms different from those infecting gram-negative bacteria. Accounting for over 50% of the cell wall mass, WTAs are the most abundant surface molecules in the cell wall of bacteria belonging to the order Bacillales, which includes genera such as Bacillus, Listeria and Staphylococcus7,35. Hence, it is most likely that phages infecting bacteria of these genera need to interact with WTAs for successful adsorption.

In this study, we demonstrated that GlcNAc residues on WTAs are essential for ϕ11 adsorption regardless of their anomeric configurations. We also found that 6-O-acetylation of muramic acid residues in peptidoglycan is involved in ϕ11 adsorption. We showed that Gp45 and Gp54 are two baseplate proteins critical for ϕ11 infection, as both antisera can neutralize ϕ11 infection dose-dependently. Recombinant Gp45 inhibits ϕ11 adsorption in a dose-dependent manner and binds to glycosylated WTAs, demonstrating that Gp45 is the RBP of ϕ11. Unfortunately, recombinant Gp54 purified from E. coli was not stable and hence unsuitable for cell wall binding studies and its functions could not be tested.

Staphylococcal pathogenicity islands (SaPIs) have an intimate relationship with temperate staphylococcal phages. Phages can induce the SaPI cycle, which allows the SaPIs to be efficiently encapsidated into special small phage heads commensurate with their size10. Previous mutational analyses of the genes present in the morphogenesis cluster of ϕ11 demonstrated that the Gp45 was essential for both the phage infectivity and transduction of its cognate SaPI13. Of note, Δgp54 seemed to lose its baseplates and failed to plate on S. aureus. Surprisingly, Δgp54 was still able to transduce SaPIs, although with a 100-fold reduction in transduction efficiency when compared with wild-type ϕ11. These results highlight that Gp45 is essential for the recognition process, while the presence of the Gp54 significantly increases the binding affinity between the phage and its receptor. As the N-terminus of Gp54 was predicted to be similar to that of BppU, which maintains the attachment of RBP to the baseplate core in TP901-1, it is tempting to speculate that Gp54 plays an important role in anchoring RBP in the baseplate.

Previously, a tail protein ORF636 from a serogroup A phage phiSLT was characterized as an adhesion protein that binds to poly-glycerolphosphate (GroP) chain of lipoteichoic acids (LTAs)25. Notably, ORF636 shares high homology with Gp45 (62% similarity) and the ORF636 sequence exists in all known serogroup A phages infecting S. aureus. However, it was shown that all tested serogroup A phages can still form plaques on a S. aureus mutant deficient in LTAs, but not on a mutant deficient in WTAs11, suggesting that WTAs but not LTAs are required for S. aureus phage infection. The tight binding of glycerol and glycerolphosphate for the RBPs suggested that LTAs could act as receptors for lactococcal phages36, however the structure of LTAs is well conserved and thought to be too simple to explain the different host specificities of various lactococcal phages. Recently, by mutational analysis, it was demonstrated that cell wall polysaccharide (CWPS) is the host cell surface receptor of tested lactococcus phages of different groups and that differences between the CWPS structures play a crucial role in determining phage host range37.

It is noteworthy that many phages need a protein receptor for adsorption, for example, Fuh A, OmpA, OmpC, LamB for E. coli phages, GamR, YueB for Bacillus phages and Pip for Lactococcus phage38. Interestingly, all these protein receptors are non-essential and many of them were identified by transposon mutagenesis. However, by screening a mutant library of S. aureus we were unable to isolate ϕ11-resistant mutants, which carry transposon insertions in genes encoding membrane proteins39. It is now generally acknowledged that carbohydrate recognizing phages possess a broad baseplate structure with multiple receptor binding sites. Conversely, phages with stubby ends or tail fibres, including the lactococcal c2 phages and the Bacillus phage SPP1, may recognize protein receptors on the cell surface40. The crystal structure of Gp45 was solved and it was found that Gp45 forms six trimers in the baseplates of ϕ11 and that each monomer of Gp45 contains a five-bladed propeller domain with a cavity that could accommodate a GlcNAc moiety (Koc et al., unpublished data). Hence, the presence of 18 receptor binding sites in the baseplate of ϕ11 suggests that its receptors are saccharides but not proteins.

Accounting for over 50% of the cell wall mass, WTAs are considered to be the most abundant surface molecules in S. aureus and have been implicated in various critical processes and interactions such as staphylococcal cell division, biofilm formation, β-lactam resistance and staphylococcal pathogenesis41,42. Due to the in-homogeneity of WTA, its analysis has proven to be very challenging. Unlike research carried out on DNA, RNA or protein, methods available for studying WTA function are very limited. Despite technical limitations, a few WTA-interacting proteins such as FmtA43, WTA antibody, MBL44 and SREC-145 have recently been identified. Here, we report Gp45 as a new WTA-interacting protein. Our results may eventually provide new tools for labelling and detecting the subdomain structures in the cell wall of S. aureus. Additionally, this study establishes a solid basis for the development of sensitive affinity-based infection diagnostics46 and therapeutics for MRSA infection.

Materials and Methods

Bacterial strains and growth conditions

S. aureus strains used in these studies are listed in Table 1. Bacteria were grown at 37 °C in BM broth (1% tryptone, 0.5% yeast extract, 0.5% NaCl, 0.1% K2HPO4, 0.1% glucose) under agitation.

Table 1 Bacterial strains used in this study.
Full size table

Construction of S. aureus mutants

The deletion mutants ∆pip1, ∆pip2, and ∆pip1/2 were constructed by allelic exchange. For knockout plasmid construction, the primers listed in Table S1 in the Supplementary Information were used. For deletion of pip1, flanking regions were amplified with primer pairs pip1-F1-up/pip1-F1-dn and pip1-F2-up/pip1-F2-dn. Purified PCR products were digested with SalI/NheI and NheI/EcoRI respectively and subsequently ligated into the SalI/EcoRI digested knockout vector pBASE647. The resulting plasmid was used for allelic exchange48. For the construction of the pip2 deletion mutant, a similar approach was pursued. The flanking regions of pip2 were amplified with primer pairs pip2-F1-up/pip2-F1-dn and pip2-F2-up/pip2-F2-dn, digested with XbaI and ligated. Afterwards this marker-less knockout cassette was subcloned into pKOR-1 and the resulting plasmid was used for mutant construction via allelic exchange48.

Overexpression and purification of the recombinant Gp45 and Gp54

Both gp45 and gp54 were amplified by PCR from S. aureus strain SA113, which is a ϕ11 lysogen. The primers used for the PCR reaction are listed in Table S1 in the Supplementary Information. The amplified gp45 or gp54 genes were subcloned into the expression vector pET28a between the NheI and XhoI sites. The resulting plasmids were transformed into E. coli BL21 for overexpression of Gp45 or Gp54. Both proteins were fused to a hexa-histidine-tag at the N-terminus to facilitate purification. After IPTG induction of the host cells, recombinant Gp45 was extracted and purified according to the procedure described previously49. Briefly, cells were lysed via ultrasonication (Digital Sonifier, Branson). After centrifugation at 38.000 × g for 55 min, cell debris was removed and the supernatant containing recombinant Gp45 protein was loaded on a 5 mL Ni-NTA-column (GE Healthcare). Fractions containing Gp45 were pooled and concentrated to 1 mg/mL using Vivaspin 20 centrifugal concentrators with a molecular size cut-off of 50,000 (Sartorius, Göttingen, Germany). The concentrated sample was then loaded on a size-exclusion chromatography column SD200 pre-equilibrated with SEC-buffer containing 25 mM HEPES, 150 mM NaCl, 1 mM DTT. Fractions containing Gp45 were pooled and concentrated as pure Gp45 preparations. The purity and folding of the recombinant Gp45 were assessed with SDS-PAGE, Circular dichroism (CD) spectroscopy and dynamic light scattering (DLS). Gp54 was purified by the same procedure as for Gp45.

Preparation of cell wall from S. aureus strains

The cell wall was extracted according to the procedure described previously50. Briefly, S. aureus overnight cultures were harvested by centrifugation at 5000 × g for 10 minutes. The cells were washed with 20 mM NH4Ac buffer (pH 4.8) and re-suspended in the same buffer. After disruption in a cell disrupter (Euler, Frankfurt am Main, Germany), the cell lysates were centrifuged at 5000 × g to remove the intact cells. The supernatant was collected as a crude extract of cell wall and mixed well with 5 mM MgSO4, 40 U/mL DNase and 80 U/mL RNase at final concentrations before overnight incubation at 37 °C. Next, to remove any cell membrane contamination, SDS was added to a final concentration of 2%, followed by ultra-sonication for 15 min. After heating at 65 °C for one hour, the cell wall preparations were washed six times with 20 mM NH4Ac buffer by centrifugation at 12,000 × g. Finally, the cell wall preparations were re-suspended in distilled water and quantified by measuring the amount of inorganic phosphate using the QuantiChromTM Phosphate Assay Kit (BioAssay Systems, USA) as described previously39.

Bacteriophage experiments

Using the double layer soft agar method, ϕ11 was propagated with the indicator strain, S. aureus strain RN4220, as a host.

Phage plating efficiencies were determined to investigate the effects of Gp45, Gp54 anti-sera and cell wall preparations on the inactivation of ϕ11. In brief, 100 μL of ϕ11 (3 × 106 PFU/mL) was mixed with 100 μL of cell wall preparations or antisera of certain concentrations and incubated at 37 °C for 10 min. Samples pre-incubated without any cell wall preparations or sera served as controls. Next, the mixtures were diluted before plating on the indicator strain (S. aureus strain RN4220) using double agar overlay methods. After overnight incubation at 37 °C, the plaques were enumerated. The efficiency of plating was calculated relative to that of plating of ϕ11 pre-incubated without any sera or cell wall preparations.

Adsorption assays were performed according to the procedure described previously11. Briefly, 200 μL of S. aureus wild-type or mutant cells containing 8 × 107 CFU were mixed with 100 μL of ϕ11 containing 3 × 105 PFU and incubated at 37 °C for 15 min. The bound phages were separated from the free phages by centrifugation at 13,000 × g for 5 min. Adsorption was calculated by determining the number of PFU of the unbound phage in the supernatant and subtracting it from the total number of input PFU. Adsorption efficiency was expressed relative to the adsorption of wild-type strain RN4220. Each adsorption assay was repeated at least three times. To study the inhibition of adsorption by Gp45, cells were pre-incubated with the purified recombinant Gp45 of indicated concentrations for 15 min before adding phages to the host cells.

Purification of ϕ11 and electron microscopy methods

Phage ϕ11 lysate was centrifuged at 73000 × g, 4 °C for two hours (Beckman Optima XL-80K). The resulting pellet was re-suspended in 500 μL of TMN buffer containing 10 mM Tris-HCl, pH 7.5, 10 mM MgSO4, 500 mM NaCl. The sample was then mixed well with 55% CsCl in TMN-buffer to give a final concentration of 42% CsCl and subjected to ultracentrifugation at 245,000 × g, 15 °C for 20 hours (Beckman). The visible phage band on the CsCl gradient was collected and sequentially dialyzed for two hours each in a D-Tube Dialyzer Mini (Novagen®, Merck Millipore, Darmstadt, Germany) against decreasing concentrations of NaCl in TMN buffer (10 mM Tris-HCl, pH 7.5, 10 mM MgCl2, 4 M NaCl) until the NaCl concentration after each round of dialysis was at 4 M, 2 M, 1 M and 10 mM NaCl, respectively.

For immunogold labelling, purified phage samples were adsorbed to glow discharged, pioloform and carbon-coated grids. The grids were then blocked with 0.2% gelatin in phosphate-buffered saline for 10 min followed by incubation with rabbit anti-Gp45 or rabbit anti-Gp54 serum, which were diluted in blocking buffer at 1:20 and 1:100, respectively. Polyclonal rabbit antisera were raised against purified recombinant Gp45 or Gp54 using a custom antibody service, Speedy 28-Day polyclonal program from Eurogentec (Brussels, Belgium). After blocking at room temperature for 60 min, the grids were washed six times with blocking buffer for a total time of 15 min before incubation with goat anti-rabbit IgG coupled with 12 nm gold colloids (Dianova, Hamburg), which was diluted with blocking buffer at 1:30. After incubation at room temperature for 60 min, the grids were washed three times with blocking buffer for 10 min and three times with phosphate-buffered saline for 10 min, followed by washing four times with double-distilled water for 2 min. Finally, the grids were negatively stained with 1% (w/v) aqueous uranyl acetate before examination with a JEM-1400Plus transmission electron microscope (JEOL, Japan)

Flow cytometry analysis

Flow cytometry was carried out to evaluate the binding of recombinant Gp45 to the S. aureus cell surface. Purified recombinant Gp45 was labelled with biotin using the EZ-Link™ NHS-Biotin kit (Thermo Fisher Scientific). Biotin-labelled Gp45 was then incubated with S. aureus wild-type or mutant cells for 30 min with shaking at room temperature. Cells were washed and stained with strep-Alu488 (Invitrogen) for one hour at 4 °C. Finally, cells were fixed for flow cytometry analysis.

Statistical analysis

Results are expressed as the means ± standard deviations from at least three independent experiments. Statistical analysis was performed using GraphPad Prism (GraphPad Software, Inc., La Jolla, USA, Version 5.04). Statistically significant differences were calculated with two-tailed Student’s t-test or one-way ANOVA with Bonferroni’s post-test as indicated.

Additional Information

How to cite this article: Li, X. et al. An essential role for the baseplate protein Gp45 in phage adsorption to Staphylococcus aureus. Sci. Rep. 6, 26455; doi: 10.1038/srep26455 (2016).