Figures and data
Infection by bacteriophage ΦStaph1N drives the loss of β-lactam resistance in MRSA.
A. Schematic of the experimental setup. Drug-resistant (AbxR), phage-sensitive (PhageS) bacterial cultures are infected with phage. The population of infected cells is passaged and allowed to recover. The surviving cell population is resistant to phage infection (PhageR) but has evolved sensitivity to antibiotics (AbxS). B. ΦStaph1N infects MRSA strains MRSA252, MW2, and LAC (left panel). Following infection with ΦStaph1N, evolved cultures of the three MRSA strains are resistant to ΦStaph1N (right panel). C. ΦStaph1N-treated, evolved MRSA strains show significant loss of resistance against oxacillin, compared to the parental strains. Loss of resistance is indicated by the area of bacterial clearance surrounding the antibiotic resistance strip. D. ΦStaph1N treatment causes loss resistance against different β-lactams. Plotted are the fold reductions of minimal inhibitory concentration (MIC) between treated and mock-treated cells. OXI = oxacillin; CEF = cefazolin; AMX = amoxicillin; AMX+CA = amoxicillin & clavulanic acid; VANC = vancomycin.
Evo2 is a variant of ΦStaph1N with higher activity against MRSA.
A. Evo2 shows comparable infectivity towards MRSA252 but improved infectivity towards MW2 and LAC, relative to ΦStaph1N. The same plaquing data is also shown in Figure S3A. B. Similar to ΦStaph1N, Evo2 infection reduces β-lactam resistance in MRSA. C. Evo2 infection reduces the MIC against oxacillin in clinical isolates of USA300 (ADLs).
Phage infection of MRSA strains produces distinct mutational profiles.
A. Coding sequences (CDS) with mutations from the three MRSA strains following phage treatment or mock treatment. For each strain, three isolates were sequenced and their mutations identified. Mutations are color-coded based on the number of occurrences among the three replicates. B. Categories of genes with mutations that arose in each MRSA strain and treatment condition.
Mutated genes in MRSA following infection with phages ΦStaph1N or Evo2
Phage infection changes the transcriptomic profile of MRSA.
Differential expression analysis was performed on the transcriptomes of MW2 (top panel) and LAC (bottom panel). For both strains, Evo2-infected samples were compared to uninfected controls. 3 biological replicates were analyzed for each condition. Horizontal dotted lines represent an adjusted p-value cut-off of 0.002, while vertical dotted lines represent a log2 fold change of −2 or 2 in expression. Transcripts with a log2 fold change between −2 or 2 and a p-value >0.002 are labeled as grey dots (Not significant, NS); transcripts that pass either the fold change or p-value cutoff (but not the other) are represented as blue and green dots, respectively; transcripts that pass both cutoffs are shown as red dots. Genes discussed in the main text are labeled. Data for all the transcripts with significant fold changes is shown in Table S5.
Phage treatment of MRSA results in attenuated virulence phenotypes.
A. MW2 and LAC strains display hemolytic activity on rabbit blood agar plates, while MRSA252 does not. B. Phage-treated MW2 and LAC strains display reduced hemolysis compared to uninfected cells. C. Surviving cultures of MW2 and LAC treated with either ΦStaph1N (blue) or Evo2 (red) show reduced clumping rates compared to mock untreated cells (teal).
Co-treatment of MRSA with bacteriophage and β-lactam.
A. Checkerboard assays of MRSA strains with gradients of oxacillin and Evo2 (top panels) or ΦStaph1N (bottom panels). The oxacillin gradient is a 2-fold serial dilution of drug concentration (µg/mL), while the phage MOI gradient is a 10-fold serial dilution of MOI. The rows and columns of each plate are labeled with letters and numbers, respectively. The black-white gradient in each well reflects the optical density of the culture and is the mean value from three biological replicates. MRSA strains co-treated with oxacillin and (B) Evo2 or (C) ΦStaph1N were tested for their phage resistance and oxacillin resistance. The letter/number combination reflects the well from which the cells were picked for analysis. Wells that could not produce a viable culture are labeled as NG (no growth). For wells that regrew, we calculated the efficiency of plaquing (EOP) of phage and measured the fold reduction in oxacillin MIC. Cultures that showed no detectable viral plaques are labeled as resistant (R).
Phage sensitivity of MRSA strains.
Efficiencies of plaquing of phages on MRSA252, MW2, and LAC. Phages were 10-fold serially diluted and spotted onto top agar overlays of each strain.
Growth curves of MRSA strains under varying levels of phage infection.
MRSA252, MW2, and LAC cultures were infected with either ΦStaph1N or Evo2 at the indicated multiplicity of infection (MOI). The optical density (OD600) of the cultures was monitored on an automated plate reader. Each condition was tested in 3 independent replicates.
Isolation and sequencing analysis of Evo2.
A. Individual Evo2 plaques appeared in larger ΦStaph1N plaques on LAC. Individual plaques were isolated and propagated in liquid culture. Evo2 shows improved plaquing on MW2 and LAC. Plaquing data in the right panels are the same as in Figure 2A. B. Evo2 is a mutant form of ΦStaph1N with a nonsense mutation in ORF141. The A to C mutation (marked by the arrow) in Evo2 converts Serine 77 of ORF141 into a stop codon.
Phage ∏NM1γ6 infection LAC does not drive the loss of β-lactam resistance.
A. LAC treated with ∏NM1γ6 evolves resistance against ΦNM1γ6, evidenced by the reduction of plaquing from the parental to the evolved populations. B. Evolved and parental LAC populations show comparable MICs against different β-lactams and vancomycin.
Phage SATA8505 infection drives loss of oxacillin resistance.
A. MRSA strains MRSA252, MW2, and LAC treated with SATA8585 evolves resistance against SATA8585, evidenced by the reduction of plaquing from the parental to the evolved populations. B. Evolved and parental MRSA show reduced MICs against oxacillin.
Types of polymorphisms in MRSA strains following infection by phage or a mock treatment.
Plotted are the polymorphisms that were found in a gene with an assigned COG category.
Plaquing efficiency of Evo2 and ΦStaph1N on MW2 and LAC strains with knockouts in mgrA, arl, and sarA.
Effect of phage infection on biofilm formation in MRSA strains.
Cultures were infected or mock-infected with either ΦStaph1N or Evo2. RP62a is a strain of S. epidermidis with known biofilm-forming capability, while LM1680 is a derivative of RP62a that has lost biofilm-forming ability.69,70 Biofilm biomass was assessed by staining with Crystal Violet. Solubilized crystal violet was quantified by measuring absorbance at 600 nm. Values represent averages and standard deviations of three replicates. Statistical significance was determined with a two-tailed t-test.
Bacterial strains and bacteriophages used in this study
MICs (µg/mL) against oxacillin of MRSA strains treated with different MOIs of phage
MICs (µg/mL) of mock- or Evo2-treated MRSA strains against different antibiotics
Efficiencies of plaquing (EOPs)* of ΦStaph1N, Evo2, and ∏NM1γ6 on clinical isolates of USA300 (ADL1-30).
Genes with significant fold changes in MRSA MW2 and LAC strains following Evo2 infection.
