The strains used in the study include S. aureus JE2 (plasmid-cured derivative of USA300 LAC) and all derivative strains within the Nebraska Transposon Mutant Library (NTML), consisting of 1920 unique transposon mutants with inactivation of non-essential genes (Fey et al., 2013). The bursa aurealis transposon used to create the collection contains the resistance cassette ermB conferring resistance to erythromycin (Fey et al., 2013). All bacterial strains were cultured at 37°C in tryptic soy broth (TSB) or on tryptic soy agar (TSA), with antimicrobial agents added as indicated. Antimicrobial agents used in the study were erythromycin (Sigma), ciprofloxacin (Sigma), oxacillin (Sigma), linezolid (Sigma), fosfomycin (Sigma), daptomycin (Santa Cruz Biotechnology), mupirocin (Sigma), vancomycin (Sigma) and gentamicin (Sigma). Transduction of atpA::ΦNΣ into wild type (WT) JE2 was performed with bacteriophage φ11, by selecting for transductants on erythromycin plates (5 μg/ml) (Fey et al., 2013).

To screen for increased susceptibility among the NTML strains, we first determined the minimum inhibitory concentration (MIC) of WT strain JE2 to ciprofloxacin, oxacillin, linezolid, fosfomycin, daptomycin, mupirocin, vancomycin, and gentamicin. MIC was determined using a twofold microbroth dilution assay according to Clinical and Laboratory Standards Institute (CLSI) guidelines (CLSI, 2009), except that cation-adjusted Mueller Hinton broth was substituted with TSB.

The NTML is stored in glycerol at -80°C in 20 96-well microtiter plates. Material from the frozen stock was transferred directly with a Deutz 96 cryoreplicator (Duetz et al., 2000) from the 96-wells microtiter plates onto TSA plates supplemented with 5 μg/ml erythromycin (to prevent the growth of contaminants during the screen, as all the strains in the NTML are resistant to erythromycin; Fey et al., 2013) and 0.5x MIC of the respective antimicrobial agent. The plates were incubated at 37°C for 24 h and visually inspected for lack of growth of individual mutants.

The MIC was determined for all strains displaying growth deficiency in the hypersusceptibility screen to the respective antimicrobial agent and WT strain JE2 using E-tests (bioMérieux) performed on TSA plates. No erythromycin was supplemented to the plates during MIC determination with E-tests. The MIC was determined upon incubation at 37°C for 22 h and interpreted according to the guidelines of the manufacturer.

To investigate if increased antimicrobial susceptibility could be detected in vivo, we employed the Galleria mellonella infection model (Desbois and Coote, 2011; Ramarao et al., 2012). Healthy 5th-instar wax moth larvae weighting approximately 250 mg were randomly picked from a batch purchased at a local pet store and divided into six groups with 20 larvae in each. Virulence of S. aureus WT and the atpA mutant were compared by injecting 20 larvae with 10 μl (containing 1 × 10⁷ CFU) with a Hamilton syringe into the hemocoel through the lowest left proleg. Survival of G. mellonella was monitored for 120 h. For treatment efficacy of gentamicin against S. aureus, 20 larvae for each group were injected with 1 × 10⁷ CFU of WT or the atpA mutant. One hour post infection, the larvae were injected with 10 μl gentamicin (1 mg/kg bodyweight) in the lowest right proleg. Gentamicin therapy was repeated every 24 h for a total of 72 h and survival of G. mellonella was monitored for 120 h. A control group for toxicity of gentamicin (1 mg/kg bodyweight) as well as for non-treated (inoculated with phosphate-buffered saline) were included. The data was analyzed in GraphPad Prism 4 (GraphPad Software Inc.) using the Kaplan–Meier method and statistical difference determined using log rank test.

Chromosomal reconstruction of the atpA mutant was achieved by use of the temperature-sensitive shuttle vector pBASE6 (Geiger et al., 2012). A chromosomal region encompassing atpA was PCR amplified from WT S. aureus JE2 chromosomal DNA using primer pair: 5′-ATATGAGCTCGAAGAGTTAGATAAGATTGTCAAACTAG-3′/5′-GATACAAGATCTGATGGTTTGTATTGCTACTTGC-3′ and cloned into pBASE6 via SacI/BglII. This plasmid was purified from E. coli IM08B (Monk et al., 2015) and transformed directly into JE2 atpA::ΦNΣ (NE592) at 30°C followed by chromosomal integration by plating on TSA (10 μg/ml chloramphenicol) at 44°C overnight. Plasmid cross-out was performed by passage at 30°C followed by plating on TSA (500 ng/ml anhydrotetracycline) and successful allelic exchange of the transposon insertion with the intact atpA gene was selected for by replica plating of colonies and screening for sensitivity toward erythromycin and chloramphenicol. Reconstruction of the atpA locus was verified by PCR amplification using primers 5′-CAAGTATGCTAAAGCATTATTTGACGTGTC-3′/5′-CGTAATTTCTGCTTGTCTCGCTCTG-3′ positioned outside the chromosomal region used for homologous recombination.

The complete NTML of 1920 single-gene inactivations (Fey et al., 2013) was screened for mutants displaying increased susceptibility to eight antimicrobial agents with different mechanism of action, by inspecting for lack of growth on agar plates supplemented with 0.5x MIC of the respective antimicrobials. The antimicrobials tested were ciprofloxacin, oxacillin, linezolid, fosfomycin, daptomycin, mupirocin, vancomycin, and gentamicin (Table 1). For the mutants identified in the initial screen as being unable to grow at 0.5x MIC, the MIC was determined using E-tests. A total of 68 mutants were confirmed to display twofold or greater increase in susceptibility to one or more antimicrobial agents compared to the WT strain S. aureus JE2 (Figure 1). Eight of these strains displayed increased sensitivity to multiple agents.

The S. aureus JE2 strain carries the acquired resistance genes mecA and fosB on the chromosome (Fey et al., 2013), which confer resistance to β-lactams (Zapun et al., 2008) and fosfomycin (Thompson et al., 2014), respectively. As a verification of the screen, we indeed identified the mecA and fosB mutants as more susceptible to oxacillin and fosfomycin, respectively (Figure 1). In our experimental setup the MIC values of oxacillin and fosfomycin for WT were within the susceptibility range according to CLSI (CLSI, 2009), even though they possess mecA and fosB. Choice of medium and NaCl concentration may affect the absolute MIC value for oxacillin (Huang et al., 1993) and discrepancies in absolute fosfomycin MIC between E-test and liquid assays has been reported (Díez-Aguilar et al., 2013; Mihailescu et al., 2014). Additionally, S. aureus JE2 is resistant to ciprofloxacin due to mutations in the essential topoisomerase genes gyrA and grlA (Sreedharan et al., 1990; Ferrero et al., 1995; Diep et al., 2006; Highlander et al., 2007; Kennedy et al., 2008). The lowering in MIC for these three antimicrobials is potentially related to the effect of the resistance genes or resistance mutations and may not necessarily classify as intrinsic resistance determinants. S. aureus JE2 is susceptible toward the remaining antimicrobials and the inactivations of genes conferring increased susceptibility are considered intrinsic resistance determinants.

Gentamicin belongs to the bactericidal antibiotic class of aminoglycosides, which interferes with protein synthesis through binding to the 16S rRNA of the 30S small ribosomal unit causing mistranslation of proteins (Shakil et al., 2008). Aminoglycosides are polycationic compounds, which is an important feature for their cell interactions and uptake (Shakil et al., 2008). Twenty six mutants displayed increased sensitivity to gentamicin (Figure 1). Inactivation of the gene rpsT encoding 30S ribosomal protein S20 was found to render the cell more susceptible to gentamicin suggesting that alteration of the ribosome assembly can influence gentamicin sensitivity through some currently unknown mechanism.

Six mutants with annotated membrane transport proteins exhibited increased sensitivity to gentamicin, with the most profound effect on sensitivity observed for vraG and trkA, displaying sixfold reductions in MIC. Gentamicin uptake into the cell is known to be dependent on membrane potential, where hyperpolarization of the membrane leads to increased uptake, while depolarization reduces the uptake of gentamicin (Taber et al., 1987). Hyperpolarization and accordingly increased gentamicin uptake is suspected for the trkA mutant, as a trkA deletion mutant in Mycobacterium smegmatis displays hyperpolarization and increased susceptibility to aminoglycosides (Castañeda-García et al., 2011). Lack of the TrkA transporter also potentiates the efficacy of aminoglycosides in Pseudomonas aeruginosa (Lee et al., 2009; Gallagher et al., 2011). Inactivation of vraG has previously been shown to increase the susceptibility toward gentamicin, vancomycin and cationic antimicrobial peptides in S. aureus, potentially by alteration of the cell surface charge (Meehl et al., 2007; Yang et al., 2012). The greatest sensitization for gentamicin, being a 16-fold reduction in MIC, was observed upon inactivation of genes encoding for subunits of the ATP synthase, namely atpA, atpB, atpG and atpH. Transduction of atpA into WT JE2, confirmed that transposon inactivation of atpA increases sensitivity of S. aureus toward gentamicin. Inactivation of atpA also potentiated efficacy of gentamicin in E. coli, which was hypothesized to occur via increased uptake of aminoglycosides due to elevated membrane potentials arising from altered respiration (Lobritz et al., 2015). While we were unable to complement the atpA-inactivated mutant with a functional atpA gene on a plasmid, we successfully performed allelic exchange of the transposon insertion with the intact atpA gene, generating a strain displaying gentamicin sensitivity as WT (1.5 μg/ml).

Besides the numerous established genes modulating gentamicin sensitivity, the NTML screen identified several novel genes of unknown function contributing to the intrinsic sensitivity to gentamicin (Figure 1).

Ciprofloxacin is a fluoroquinolone interfering with bacterial replication via inhibition of the enzymes DNA gyrase and topoisomerase IV (Blondeau, 2004). DNA gyrase, composed of two subunits GyrA and GyrB, is essential for replication initiation as the enzyme introduces negative supercoils ahead of the replication fork (Drlica and Zhao, 1997), while topoisomerase IV, composed of GrlA and GrlB, is responsible for decatenation of the replicated daughter chromosomes (Drlica and Zhao, 1997). Interaction of ciprofloxacin with these enzymes leads to a halt in replication and ultimately to DNA strands breakage (Drlica and Zhao, 1997; Blondeau, 2004). S. aureus JE2 is resistant to ciprofloxacin, due to amino acid changes in the DNA gyrase GyrA (Ser84Leu) and topoisomerase IV GrlA (Ser80Tyr), encoded by gyrA and grlA respectively (Sreedharan et al., 1990; Ferrero et al., 1995).

Elucidation of the intrinsic resistome in E. coli revealed that multiple gene products related to DNA replication and repair aid this bacterium to survive DNA damage generated by fluoroquinolone treatment (Tamae et al., 2008; Liu et al., 2010). Four of the 13 identified mutants in our screen with increased sensitivity to ciprofloxacin also display inactivation of genes related to DNA replication and repair, namely recG, xerC, recX and xseA (Figure 1). The RecG protein is a DNA helicase and strains with deletion of recG display reduced recombination and DNA repair (Al-Deib et al., 1996). Identification of recG corroborates previous work in S. aureus on increased susceptibility to quinolones in a recG mutant (Niga et al., 1997). A mutant of xseA, encoding the large subunit of exonuclease VII, was also observed in E. coli to be hypersensitive to ciprofloxacin (Tamae et al., 2008).

Three mutants with inactivated membrane transport systems, norA, trkA and SAUSA300_0924, are at least fourfold more sensitive to ciprofloxacin than the WT strain. This might be due to decreased efflux of this agent, as ciprofloxacin has been observed to being subjected to cellular efflux by pump activity of multiple efflux systems, e.g., NorA (Yamada et al., 1997; Muñoz-Bellido et al., 1999). Moreover, it was confirmed that inactivation of the stress response sigma factor σ^B encoded by rpoF and the σ^B activator RsbU potentiated the effect of ciprofloxacin, which is in agreement with previous observations in S. aureus (Riordan et al., 2006). We also identified the clpP gene encoding the proteolytic subunit of the ClpXP two-component protease to be more susceptible to ciprofloxacin, corroborating a study in P. aeruginosa (Fernández et al., 2012). A possible link between clpP and ciprofloxacin susceptibility is that clpP has been shown to interfere with activation of the SOS DNA-damage response (Cohn et al., 2011). Additionally, inactivation of several hypothetical genes (SAUSA300_1789 and SAUSA300_2311) also conferred a fourfold or greater sensitization to ciprofloxacin relative to the WT strain.

Linezolid belongs to the antimicrobial class of oxazolidinones (Moellering, 2003). Linezolid interacts with the A-site of the 50S ribosomal subunit and inhibits formation of the initiation complex, thus inhibiting protein synthesis (Wilson, 2014). A total of 15 mutants were more susceptible to linezolid (Figure 1), with six mutants displaying at least a fourfold increased sensitivity. None of the identified genes have to our knowledge previously been associated with increased linezolid susceptibility. Interestingly, these genes exert widely different functions, such as zapA encoding the divisome protein ZapA, which stabilizes Z-ring formation (Adams and Errington, 2009), mutS2 encoding MutS2 that in Helicobacter pylori function as an inhibitor of recombination (Pinto et al., 2005) and the ribonuclease R.

The target of mupirocin is the isoleucyl-tRNA synthetase (IleRS), thus inhibiting aminoacylation of isoleucine to the cognate tRNA and thereby prevents protein synthesis (Yanagisawa et al., 1994; Pope et al., 1998). Only three mutants exhibited increased sensitivity toward mupirocin (Figure 1). None of the mutants, sucD, gpmI and SAUSA300_2299, have previously been associated with altered susceptibility to mupirocin. The gene SAUSA300_2299 has been annotated as a putative multidrug resistance transporter and could be involved in active efflux of mupirocin from the cytoplasm.

Oxacillin is a β-lactam interfering with cell wall biosynthesis. The bactericidal activity of oxacillin derives from its binding to penicillin-binding proteins (PBPs), thus preventing cross-linking of peptidoglycan units of the cell wall, which eventually causes cell lysis (Kotra and Mobashery, 1998). In our study, we identified 13 genes displaying increased susceptibility to oxacillin (Figure 1). S. aureus JE2 contains the mecA gene encoding the alternative penicillin binding protein 2a (PBP2a) that enables cross-linking of peptidoglycan units in the presence of β-lactams (Zapun et al., 2008). Indeed inactivation of mecA rendered the strain more susceptible to oxacillin. A functional σ^B response is important for oxacillin susceptibility (Singh et al., 2003; Schulthess et al., 2009) and RsbU together with RsbV are activators of this stress response (Palma and Cheung, 2001), both of these determinants were confirmed as targets for oxacillin sensitizers in our screen. In addition to the previously established contributors to oxacillin intrinsic resistance, several genes of unknown function were identified (SAUSA300_1792 and SAUSA300_2297) that upon inactivation increased susceptibility to oxacillin equally well as mecA inactivation. It was also observed that inactivation of the unknown gene SAUSA300_1003 displayed even greater sensitivity to oxacillin than the mecA mutant. Moreover, this SAUSA300_1003 mutant displayed increased sensitivity to ciprofloxacin, linezolid and daptomycin and thereby it could be of interest to investigate how this gene can potentiate the efficacy of multiple antimicrobial agents.

Fosfomycin is a phosphoenolpyruvate analog that inhibits the first enzymatic step in peptidoglycan biosynthesis, by binding to the enzyme MurA and thereby prevent the formation of N-acetylmuramic acid, an essential precursor of the peptidoglycan cell wall (Michalopoulos et al., 2011; Karageorgopoulos et al., 2012). S. aureus JE2 harbors the fosfomycin resistance gene (fosB) that chemically inactivates fosfomycin (Rigsby et al., 2005; Castañeda-García et al., 2013) and in agreement, the fosB mutant showed increased sensitivity to fosfomycin. Moreover, inactivation of the gene SAUSA300_2279 encoding a regulatory protein of the LysR family also increased susceptibility to fosfomycin. This gene is located directly upstream of fosB and could potentially be a positive regulator of fosB or the transposon insertion in SAUSA300_2279 could prevent proper transcription of fosB. Four other genes also affected fosfomycin susceptibility, with a hypothetical (SAUSA300_0553) and a putative peptide transporter (SAUSA300_0073) being most prominent with threefold greater sensitivity to fosfomycin, which equal the potentiating effect of the fosB mutant.

Vancomycin is a glycopeptide antibiotic readily used in the treatment of methicillin-resistant S. aureus (Liu et al., 2011). The bactericidal activity of vancomycin is derived from the inhibition of cell wall biosynthesis. Vancomycin binds to the D-Ala-D-Ala residues of the pentapeptides on the peptidoglycan precursors N-acetylmuramic acid and N-acetylglucosamine and thus prevents peptide cross-linking of the polysaccharide backbone in peptidoglycan biosynthesis (Reynolds, 1989; Boger, 2001). Inactivation of two genes increased the susceptibility to vancomycin, with a maximum two-fold reduction in MIC (Figure 1). We confirmed the previously established vancomycin intrinsic resistance determinants vraS (Kuroda et al., 2003; Gardete et al., 2012) and vraF of the VraFG ABC transporter system (Meehl et al., 2007). Although our screen identified recognized vancomycin intrinsic resistance determinants, we did not observe mutants of the dlt operon as seen previously in a corresponding vancomycin hypersusceptibility screen (Blake and O’Neill, 2013). No strains in the NTML exist with inactivation of any of the four genes in the dlt operon, which is involved in adding positively charged D-alanine to teichoic acids (Peschel et al., 1999).

Daptomycin is a cyclic lipopeptide displaying bactericidal activity against many Gram-positive organisms (Steenbergen et al., 2005). The bactericidal activity of daptomycin is attributed to the insertion of the lipophilic tail into the bacterial cell membrane, leading to rapid membrane depolarization and potassium ion efflux, with downstream arrest of DNA, RNA and protein synthesis (Steenbergen et al., 2005).

Only a single mutant, SAUSA300_1003, displayed increased susceptibility to daptomycin. The gene SAUSA300_1003 encodes a protein of unknown function and this mutant is also more susceptible to several of the other antimicrobial agents tested.

To verify the potential of enhancing the efficacy of antimicrobial agents by targeting a non-essential gene product in S. aureus, we employed the G. mellonella infection model (Ramarao et al., 2012). We tested the efficacy of gentamicin at a clinically relevant dose (1 mg/kg bodyweight) to promote survival of G. mellonella infected with a lethal dose (10⁷ cells/larvae) of S. aureus JE2 (WT) and the atpA mutant (Figure 2). Gentamicin significantly prolonged survival of G. mellonella infected with the atpA mutant compared to the treatment of WT (P < 0.0001). None of G. mellonella larvae in the group only receiving gentamicin treatment died during the experiment (Data not shown).