Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2015 Oct 8.
Published in final edited form as: Cell Host Microbe. 2014 Oct 8;16(4):531–537. doi: 10.1016/j.chom.2014.09.002

Inter- and intra-species metabolite exchange promotes virulence of antibiotic-resistant Staphylococcus aureus

Neal D Hammer 1,†, James E Cassat 1,2,†, Michael J Noto 3, Lisa J Lojek 1, Ashley D Chadha 4, Jonathan E Schmitz 1, C Buddy Creech 2, Eric P Skaar 1,*
PMCID: PMC4197139  NIHMSID: NIHMS627821  PMID: 25299336

SUMMARY

Adaptations that enable antimicrobial resistance often pose a fitness cost to the microorganism. Resistant pathogens must therefore overcome such fitness decreases to persist within their hosts. Here we demonstrate that the reduced fitness associated with one resistance-conferring mutation can be offset by community interactions with microorganisms harboring alternative mutations or via interactions with the human microbiota. Mutations that confer antibiotic resistance in the human pathogen Staphylococcus aureus led to decreased fitness whereas co-culture or co-infection of two distinct mutants resulted in collective recovery of fitness comparable to wildtype. Such fitness enhancements result from the exchange of metabolites between distinct mutants, leading to enhanced growth, virulence factor production, and pathogenicity. Inter-species fitness enhancements were also identified as members of the human microbiota can promote growth of antibiotic-resistant S. aureus. Thus, inter- and intra-species community interactions offset fitness costs and enable S. aureus to develop antibiotic resistance without loss of virulence.

INTRODUCTION

The Gram-positive pathogen Staphylococcus aureus is one of the most significant threats to public health and accounts for nearly half of all deaths from antibiotic-resistant pathogens (Prevention, 2013). Multi-drug resistant S. aureus strains are prevalent in healthcare and community settings, limiting treatment options for life-threatening staphylococcal diseases. S. aureus resists antimicrobial treatment through several mechanisms, including expression of antibiotic-modifying enzymes and efflux pumps, and accumulation of mutations that alter drug targets or cell wall composition (Lowy, 2003). Additionally, upon exposure to select antimicrobials, notably aminoglycosides, S. aureus and other pathogens can adopt a respiration-deficient state known as a small colony variant (SCV) (Proctor et al., 2006). SCVs are isolated from chronic infections such as osteomyelitis and from cystic fibrosis (CF) patients, where they are associated with advanced pulmonary disease (Wolter et al., 2013). Human SCV isolates commonly have mutations in the biosynthetic pathways for heme and menaquinone, two essential cofactors for respiration. Because they are often deficient in respiration and dependent on fermentation for energy, SCVs grow slowly and are frequently resistant to many antibiotics. Yet SCVs have an inherent fitness defect relative to wildtype due to their reduced growth and limited virulence factor production in vitro (Proctor et al., 2006). Therefore, the mechanism by which SCVs and other fitness-impaired, antibiotic-resistant bacteria persist and cause disease within hosts remains elusive.

S. aureus and other pathogens are capable of growing in multicellular communities such as biofilms and microcolonies within tissue abscesses (Cheng et al., 2011). These environments facilitate inter-bacterial communication, which may influence pathogen behavior. We hypothesized that the fitness cost associated with a resistance-conferring mutation could be offset by interactions between genetically distinct microorganisms in a bacterial community. In this manuscript, we utilize co-culture and co-infection strategies to demonstrate that distinct antimicrobial-resistant S. aureus SCV strains utilize one another’s metabolites, resulting in enhanced growth yields and virulence. Additionally, our data indicate that metabolite exchange occurs at the interspecies level, as members of the human microbiota can promote growth of antibiotic-resistant S. aureus. These results have important implications for the diagnosis and treatment of antimicrobial-resistant infections, and add to a growing body of evidence that pathogen behavior is influenced by microbial community interactions.

RESULTS

Co-culture of genetically distinct SCVs results in enhanced growth, virulence factor production, and cytotoxicity toward mammalian cells

To test the hypothesis that interactions between distinct S. aureus strains can offset the fitness costs associated with antimicrobial resistance, two aminoglycoside-resistant SCVs were constructed in S. aureus by inactivating enzymes in the biosynthetic pathways for heme (ΔhemB) or menaquinone (ΔmenE). Both ΔhemB and ΔmenE are attenuated for growth (Figures 1A and 1B) and exhibit increased minimum inhibitory concentrations (MICs) to gentamicin relative to wildtype (WT MIC = 2, ΔhemB MIC = 16, ΔmenE MIC = 12). However, growth of ΔhemB and ΔmenE is significantly enhanced when the strains are mixed by cross-streaking on solid media or co-cultured in broth (Figures 1A and 1B). The enhanced growth observed upon cross-streaking of distinct SCVs is not dependent on strain background or which heme or menaquinone biosynthetic enzyme is inactivated (Figures S1A–S1D). The increased growth observed with co-culture of menaquinone and heme biosynthesis mutants is, however, dependent on the ability of the menaquinone biosynthesis mutant to synthesize heme, as a ΔmenB/ΔhemB mutant does not enhance growth of ΔhemA or ΔhemB (Figures S1A and S1B). Interestingly, co-culture of ΔhemB and ΔmenE failed to completely restore gentamicin susceptibility to wildtype levels (MIC = 4 versus WT MIC = 2), suggesting that physiologic adaptions which support the SCV phenotype might engender bacteria with enhanced intrinsic resistance to antimicrobials.

Figure 1. Co-culture of genetically distinct SCVs results in enhanced growth, virulence factor production, and cytotoxicity toward mammalian cells.

Figure 1

Open in a new tab

(a) Cross-streaking of ΔhemB and ΔmenE on TSA. Arrows indicate the streak direction, and the strain applied first is denoted with an open arrow. (b) Growth in TSB was monitored using optical density (O.D.600) over time. Mix indicates ΔhemB-ΔmenE co-culture. ΔhemB* and ΔmenE* indicate growth with supplementation of 2 μM heme or 12.5 μM menadione, respectively. *** denotes p<0.05 for ΔhemB and ΔmenE relative to mix. (c) Quorum sensing in co-cultured SCVs was monitored using agr-dependent YFP expression. YFP fluorescence (relative fluorescent units [RFUs], yellow bars) was recorded at 15 hours and normalized to O.D.600. *** denotes p<0.05. The average from three independent experiments is shown in b and c. Error bars represent one SD from the mean. (d) SDS-PAGE of supernatants from WT, ΔhemB, ΔmenE, and mix. (e) Immunoblot for alpha hemolysin in supernatants from the indicated strains, mix, and ΔhemB or ΔmenE supplemented with heme or menadione, respectively (*). (f) Osteoblast viability after exposure to concentrated supernatant from the indicated strains. Supernatants from ΔhemB and ΔmenE were concentrated an additional 3-fold relative to WT and the 1:1 ΔhemB-ΔmenE mix to account for differences in growth yield. Results are expressed as percent control (equivalent volume of media). N=10 per group and data represent three independent experiments. *** and # denote p<0.001 relative to WT and mix, respectively. See also Figure S1, S2, and S3.

To further delineate the involvement of metabolite exchange in enhanced growth of co-cultured SCVs, we created a menaquinone-deficient strain that only synthesizes heme if provided the precursor aminolevulinic acid (ALA) (ΔhemA/ΔmenE). Strain ΔhemA/ΔmenE only enhances growth of the heme-deficient SCV ΔhemB when ALA is supplemented in the media (Figures S2A–S2I). This result illustrates the requirement for de novo heme synthesis by the menaquinone-deficient strain and transfer to the heme-deficient strain to enhance growth during co-culture, strongly suggesting that metabolite exchange is occurring between these two strains. Moreover, addition of the heme-binding protein IsdI to broth culture significantly reduces the growth of the ΔhemB-ΔmenE mixture (Figure S2J), implying that exchange of the metabolite heme is necessary for enhanced growth during co-culture. Collectively, these results indicate that distinct SCVs can complement each other to enhance growth through metabolite exchange.

Increased bacterial density in ΔhemB-ΔmenE co-cultures was accompanied by enhanced activation of quorum sensing mediated by the accessory gene regulator (agr) system. In contrast, culture of ΔhemB or ΔmenE alone led to decreased activation of quorum sensing, even after adjusting for the reduced bacterial density of these cultures (Figure 1C; Figures S3A and S3B). Consistent with a known role of the agr system in activation of S. aureus virulence factor production (Novick, 2003), activation of quorum sensing in ΔhemB-ΔmenE co-culture lead to increased protein secretion, including extracellular proteases and the secreted toxin alpha-hemolysin (Figures 1D–1E; Figures S3C and S3D). Activation of quorum sensing and extracellular protease production have previously been demonstrated to inversely correlate with biofilm formation (Lauderdale et al., 2009; Mrak et al., 2012). Moreover, S. aureus SCVs are capable of increased biofilm formation relative to wildtype (Singh et al., 2010). We therefore assessed the functional impact of metabolite exchange on biofilm formation during ΔhemB-ΔmenE co-culture. Relative to either SCV alone, co-culture of ΔhemB and ΔmenE resulted in increased biofilm/biomass (O.D.580/O.D.600) ratios at early time points, likely reflecting initial adherence. At later time points, however, ΔhemB-ΔmenE co-culture yielded lower average biofilm/biomass ratios (Figures S3E and S3F). Finally, to investigate whether increased agr activation and secreted virulence factor production during SCV co-culture leads to increased host cell death, secreted proteins isolated from ΔhemB-ΔmenE co-culture were incubated with murine osteoblasts. As compared to supernatants from wildtype, ΔhemB, and ΔmenE, supernatants obtained from ΔhemB-ΔmenE co-culture incite significantly more osteoblast cell death (Figure 1F). Collectively, these results demonstrate that interactions between distinct SCVs enhance quorum sensing, virulence factor production, and cytotoxicity towards mammalian cells.

Co-infection with genetically distinct SCVs results in increased intraosseous bacterial burdens and bone destruction in a murine model of osteomyelitis

SCVs are frequently isolated from patients suffering from osteomyelitis and contribute to the treatment recalcitrance of this and other chronic infections (Proctor et al., 2006). Because SCVs grow slowly and are deficient in virulence factor production in vitro, it is unclear how these variants proliferate and cause tissue destruction in vivo. To evaluate if interactions between distinct SCVs enhance growth within host tissues, wildtype, ΔhemB, ΔmenE, or a 1:1 mixture of ΔhemB and ΔmenE were used to inoculate mice in an experimental model of osteomyelitis (Cassat et al., 2013). Relative to wildtype, infection with either ΔhemB or ΔmenE resulted in significantly lower bacterial burdens in infected femurs (Figure 2A). However, mice infected with the ΔhemB-ΔmenE mixture sustained bacterial burdens that were significantly increased by approximately two logs relative to each SCV alone. Mice infected with the ΔhemB-ΔmenE mixture also experienced substantially increased weight loss relative to mice infected with either SCV alone (Figure 2B). To ascertain the relative proportions of ΔhemB and ΔmenE SCVs present after in vivo co-infection, we inoculated groups of mice with a 1:1 mixture of ΔhemB and ΔmenE after in vitro growth with varying levels of heme and menadione supplementation. Regardless of the supplementation level prior to infection, we found that all infected mice harbored both ΔhemB and ΔmenE SCV populations, with ΔhemB isolated at a higher frequency (Table S2). Even with minimal or no supplementation during in vitro pre-growth, mice infected with the ΔhemB-ΔmenE mixture sustained average bacterial loads higher than that of either supplemented SCV alone (Figure 2C). Bone destruction during S. aureus osteomyelitis is dependent on production of extracellular virulence factors (Cassat et al., 2013). To test if co-infection with distinct SCVs results in increased bone destruction, groups of mice were infected with WT, ΔmenE, ΔhemB, or ΔhemB-ΔmenE mixture and bone destruction was assessed 14 days later by micro-computed tomography. Femurs from ΔmenE- and ΔhemB-infected mice sustained drastically reduced bone destruction relative to wildtype (Figures 2D–2H). Conversely, mice infected with the ΔhemB-ΔmenE mixture sustained bone destruction that was not significantly different from WT, but was profoundly increased relative to ΔmenE or ΔhemB infection. Enhanced production of virulence factors during co-culture of ΔhemB and ΔmenE therefore correlates with increased bone destruction in vivo. Thus, interactions between genetically distinct staphylococci can overcome fitness costs associated with antimicrobial resistance to facilitate growth and virulence within mammalian hosts.

Figure 2. Co-infection with genetically distinct SCVs results in increased intraosseous bacterial burdens and bone destruction in a murine model of osteomyelitis.

Figure 2

Open in a new tab

(a) Groups of mice were subject to osteomyelitis by inoculation of the indicated strains following in vitro growth in 2 μM heme (ΔhemB) or 12.5 μM menadione (ΔmenE). Bacterial burdens were determined after 7 days. Horizontal bar represents the mean. ** denotes p<0.01 and *** denotes p<0.001. N=5 mice per group. (b) Percent initial weight was calculated daily for mice infected in (a). (c) The same as in (a) except that strains were grown in vitro in minimal (0.25 μM heme and 5 μM menadione) or no supplementation prior to mixing and inoculation. (d) Quantification of cortical bone destruction. N=5 mice per group. Error bars represent standard error of the mean. * denotes p<0.05 and *** denotes p<0.001 relative to both WT and mix. N.S. = not significant. (e–h) Representative micro-computed tomography images of femurs infected with WT (e), ΔmenE (f), ΔhemB (g), and ΔhemB-ΔmenE mix (h). Three femurs with median bone loss are shown for each group. See also Table S2.

Interspecies interactions promote the emergence and growth of antibiotic-resistant S. aureus

During infection, pathogens may be exposed to multiple selective pressures, including host defenses, other microorganisms, and antimicrobial compounds. In CF patients whose lungs are infected by S. aureus, staphylococci are exposed not only to antibiotics such as aminoglycosides, but also frequently encounter other bacterial pathogens such as Pseudomonas aeruginosa (Hubert et al., 2013; Schneider et al., 2008). During co-infection, P. aeruginosa secretes the exotoxin pyocyanin which hinders S. aureus respiration (Hoffman et al., 2006; Voggu et al., 2006). In response to this pressure, S. aureus can grow as an SCV, rendering it resistant to pyocyanin (Biswas et al., 2009). We hypothesized that pyocyanin-induced SCVs might enhance growth of aminoglycoside-induced SCVs during co-culture. SCVs were generated by exposure of S. aureus to either gentamicin (SCVgent) or pyocyanin (SCVpyo). Both SCVs exhibited elevated MICs to gentamicin (SCVgent = 24; SCVpyo = 48). SCVpyo exhibited an enhanced growth rate on menadione-supplemented media, suggesting that it is a menadione auxotroph. Conversely, SCVgent did not display enhanced growth after supplementation with menadione, heme, or thymidine, suggesting one or more alternative auxotrophies (data not shown). Co-culture of SCVgent with SCVpyo results in enhanced growth (Figures 3A and 3B). Moreover, co-culture of SCVgent with SCVpyo increases growth over each single SCV even in the presence of gentamicin (Figure 3C). Accordingly, the gentamicin MIC of mixed SCVgent - SCVpyo culture was 16. We found that SCVpyo could complement the growth of 18 out of 20 tested SCVs generated by exposure to gentamicin. Collectively, these data reveal that staphylococcal SCVs induced by distinct selective pressures can interact to overcome the fitness defect associated with respiration arrest.

Figure 3. Pyocyanin-resistant SCVs enhance the growth of antimicrobial-resistant SCVs.

Figure 3

Open in a new tab

(a) SCVgent and SCVpyo cross-streaking on TSA. (b) Growth of WT, SCVgent, SCVpyo, and SCVgent-SCVpyo mix in TSB was monitored using O.D.600 over time. (c) Growth of the indicated strains in TSB containing various concentrations of gentamicin was monitored using O.D.600 after 8 hours. The average from three independent experiments is shown. Error bars represent one SD from the mean. See also Figure S4.

To test if inter-species interactions that are not antagonistic can also promote the growth of antibiotic-resistant pathogens, the ability of Enterococcus faecalis to enhance the growth of S. aureus SCVs was determined. Enterococci are frequently isolated with staphylococci in diabetic soft tissue infections (Citron et al., 2007). Cross-streaking of E. faecalis with S. aureus SCVs inactivated for menaquinone biosynthesis results in enhanced growth. This process is dependent on heme synthesis by S. aureus as E. faecalis cannot synthesize heme and therefore does not enhance growth of ΔhemB. In total these findings imply that the growth enhancement requires either heme exchange from S. aureus, menaquinone exchange from E. faecalis, or both (Figure S4) (Whittenbury, 1964).

Co-culture of distinct, antimicrobial-resistant SCVs isolated from CF patients enhances growth

Previous studies utilizing in vitro assays and animal models have demonstrated that both intra- and interspecies bacterial interactions can elicit changes in bacterial physiology and virulence. However, there is a paucity of data to implicate such interactions in the virulence of antibiotic-resistant bacterial pathogens in vivo in humans. To investigate if distinct, slowly growing microorganisms from humans can enhance one another’s growth, we isolated and archived bacteria from the respiratory tract of CF patients. Within these samples, we identified members of the human upper respiratory microbiota including S. epidermidis, Streptococcal species, and Diphtheroid-like organisms that are capable of enhancing the growth of S. aureus ΔhemB, ΔmenE, or both (Figure 4A). Furthermore, we isolated distinct, gentamicin-resistant S. aureus SCVs from the same patient and discovered that a subset of these SCVs could enhance growth of laboratory-derived SCVs (Figures 4B–4C) as well as one another (Figure 4D) during co-culture. These data suggest that metabolite exchange may also occur within human microbial communities to enhance growth of antibiotic-resistant bacteria.

Figure 4. Co-culture of distinct, antimicrobial-resistant SCVs isolated from CF patients enhances growth.

Figure 4

Open in a new tab

(a) Slowly growing bacterial colonies were isolated and archived from the sputa of adult and pediatric patients suffering from CF at Vanderbilt University Medical Center. Polymicrobial samples were labeled successively as VUCF001, VUCF002, etc. Individual, slowly growing bacterial colonies were isolated from each sample and archived as VUCF001.1, VUCF001.2, etc. Individual small colonies were then co-cultured with ΔhemB or ΔmenE on solid media to assess for enhanced growth. A representative subset of isolated SCVs is shown. (b,c) Isolated, gentamicin-resistant S. aureus SCVs from a single patient were co-cultured with ΔhemB (b) or ΔmenE (c) on solid media to assess for enhancement of growth. (d) Select gentamicin-resistant S. aureus were co-cultured with each other on solid media to demonstrate growth enhancement. A representative subset of growth-enhancing pairs is shown. Strain identities are as follows: VUCF001.1 = Staphylococcus epidermidis; VUCF001.2 and VUCF001.3 = Diphtheroid-like Gram-positive rods; VUCF002.1 = Streptococcus sobrinus; VUCF003.1 through VUCF003.5 = S. aureus.

DISCUSSION

In this report, we demonstrate that distinct S. aureus strains can interact in vitro and in vivo with one another and with the human microbiota to overcome the fitness cost associated with antimicrobial resistance mutations. Enhancement of growth and virulence is dependent on the mutual production and procurement of specific metabolites by distinct mutants and allows groups of fitness-impaired, antibiotic-resistant S. aureus to achieve a collective fitness comparable to wildtype. Our results strongly suggest that the mechanism of growth enhancement in SCV co-culture is exchange of metabolites. However, the precise mechanism by which these metabolites exit the cell, are incorporated into the recipient cell, and are ultimately utilized for cellular processes remains to be determined. After in vivo co-infection with heme and menadione-deficient SCVs, we found a predominance of heme-deficient mutants. This may reflect the relative efficiency of heme exchange as compared to menadione. Alternatively, these results may reflect the presence of dedicated heme acquisition systems in S. aureus, or the increased bioavailability of host heme as compared to menadione. Nevertheless, these experiments were performed in the absence of antibiotic selection, which we predict would lead to an iterative process whereby multiple antibiotic-resistant mutants are generated during the course of infection, a subset of which could then interact to enhance growth and virulence. Our data have significant implications for the diagnosis and treatment of chronic infections caused by SCVs. Current diagnostic practices may be inadequate, as a single antibiotic resistance profile is often interpreted as evidence of clonality in clinical specimens, yet our data suggest that such samples may contain genetically distinct bacterial populations whose interactions can alter antimicrobial resistance and virulence properties. Concerning treatment, these data suggest that fitness costs associated with antimicrobial resistance, and specifically the S. aureus SCV phenotype, can be overcome in vivo to promote virulence. Diversity within a bacterial community, both at the strain and species level, therefore affords S. aureus the ability to tolerate fitness-reducing mutations that promote persistence in their host while maintaining virulence potential.

EXPERIMENTAL PROCEDURES

Ethics Statement

All experiments involving animals were reviewed and approved by the Institutional Animal Care and Use Committee of Vanderbilt University and performed according to NIH guidelines, the Animal Welfare Act, and US Federal law. Human clinical samples were obtained under a protocol approved by the Vanderbilt Institutional Review Board.

Bacterial strains and growth conditions

Unless otherwise noted, all experiments utilized derivatives of the S. aureus strain JE2 (WT) (Fey et al., 2013). Bacteria were grown on tryptic soy broth (TSB) solidified with 1.5% agar (TSA) at 37°C or in TSB with shaking at 180 rpm, unless otherwise indicated. TSB was supplemented with 10 μg/ml erythromycin where indicated. Aminolevulinic acid (ALA) was supplemented to solid growth media at a final concentration of 75 μg/ml and onto sterile Whatman paper discs at a concentration of 50 mg/ml. IsdI was purified from Escherichia coli BL21 (DE3) as previously described (Skaar et al., 2004). Bacterial strains and plasmids used in this study are listed in Table S1. Phage Φ85-mediated transduction of the hemB and menE erythromycin cassette-disrupted allele was used to transfer these insertion mutations from S. aureus Newman to JE2. Gentamicin MICs were determined using Etest® strips (Biomerieux) according to manufacturer’s instructions. Quorum sensing was monitored during growth of S. aureus using the fluorescent reporter plasmid pDB59 (agr-P3-YFP) (Malone et al., 2009; Yarwood et al., 2004). Bacterial cultures were diluted 1:100 from overnight culture and optical density was measured as absorbance at 600 nm. YFP fluorescence was monitored using a BioTek plate reader with excitation and emission filters of 485 nm and 525 nm, respectively. Differences in growth yields and YFP fluorescence between strains were analyzed by Student’s t-test. Strain SCVgent was created by inoculating WT S. aureus into TSB containing 10 μg/ml gentamicin and growing at 37°C for 8 hours. The resulting culture was plated at limiting dilution onto TSA and grown overnight at 37°C. Isolated SCVs were passaged 10 times on TSA and then plated onto TSA containing 10 μg/ml gentamicin to ensure stability of the SCV phenotype and gentamicin resistance. Strain SCVpyo was created by plating S. aureus Newman on plates containing 20 μg/ml of pyocyanin and selecting a resultant SCV. The auxotrophy of SCVgent and SCVpyo was tested by plating either strain on TSA supplemented with 2 μM heme, 5 μM menadione, or 25 μM thymidine.

Isolation of slowly growing microorganisms from cystic fibrosis patients

Bronchoalveolar lavage (BAL) fluid and sputum samples were obtained from adult and pediatric patients with cystic fibrosis at Vanderbilt University Medical Center. Samples were de-identified of all personally identifying information by the Vanderbilt Clinical Microbiology Laboratory staff prior to further microbiologic analysis. BAL fluid or sputum samples were serially diluted in sterile phosphate-buffered saline (PBS) and plated onto TSA. Freezer stocks of the entire bacterial population from each patient sample were created by removing bacterial growth from the TSA plates and resuspending in a solution of 5% bovine serum albumin and 5% L-glutamic acid sodium salt. Slowly growing bacterial colonies were subsequently identified from patient samples after 48–72 hours of growth on TSA at 37°C. Individual small colonies were passaged on TSA to ensure maintenance of the small colony phenotype, and then co-cultured with other small colonies in the patient sample or with ΔhemB and ΔmenE by cross-streaking. To identify gentamicin-resistant small colonies, archived bacterial populations from each patient sample were plated onto TSA supplemented with 6 μg/ml gentamicin. Gentamicin-resistant small colonies were then passaged on TSA as above before co-culture with other small colonies in the patient sample or with ΔhemB and ΔmenE. Freezer stocks of individual small colonies were also created as above. Species identification of slowly growing microorganisms was accomplished by a combination of Gram stain, catalase testing, Staphaurex® agglutination assay, as well as automated identification on a Becton Dickinson Phoenix™ Automated Microbiology System.

Preparation of S. aureus culture supernatant

To prepare concentrated culture supernatants, 24 hour (ΔhemB and ΔmenE) or 12 hour (WT) TSB cultures were back-diluted 1:100 into 50 ml RPMI supplemented with 1% casamino acids in a 250 ml flask. A 1:1 ΔhemB-ΔmenE mix RPMI culture was inoculated using a 1:200 dilution of ΔhemB TSB culture and a 1:200 dilution of ΔmenE TSB culture. RPMI cultures were grown for 15 hours at 37°C and 180 rpm shaking, after which time supernatants were collected by centrifugation. Supernatants were then filtered through a 0.22 μm filter and concentrated with Amicon Ultra 3kDa nominal molecular weight limit centrifugal filter units (Millipore, Billerica, MA). Supernatants from ΔhemB and ΔmenE were concentrated an additional 3-fold relative to WT and the 1:1 ΔhemB-ΔmenE mix to account for differences in growth yield. Secreted proteins in culture supernatants were resolved by SDS-PAGE after addition of an equivalent volume of 2X SDS buffer and heating to 95°C. Gels were subsequently stained with Protein Assay Dye Reagent (Bio Rad). Alternatively, the resolved proteins were transferred to a nitrocellulose membrane and subjected to immunoblot using a mouse monoclonal anti-Hla antibody (Sigma). An Alexa Fluor-conjugated secondary antibody was used for visualization (Life Sciences).

Osteoblast cytotoxicity assays

MC3T3-E1 subclone 4 cells were obtained from the American Type Culture Collection (ATCC) and propagated according to ATCC recommendations. Cells were grown at 37°C and 5% CO2 with replacement of media every 2 or 3 days. Osteoblast cytotoxicity assays were then performed as previously reported (Cassat et al., 2013). Differences in cell viability after exposure to various supernatants were analyzed by Student’s t-test.

Biofilm assay

Cultures of WT, ΔhemB, and ΔmenE were grown overnight in TSB supplemented with 3% NaCl and 0.5% glucose. Biofilms were grown in plasma-coated 96-well plates with duplicate rows by inoculating TSB with 3% NaCl and 0.5% glucose with a 1:50 ratio of WT, ΔhemB, ΔmenE, and a mix of equal volumes ΔhemB and ΔmenE. Biofilms were incubated statically at 37°C and processed every 2 hours for 12 hours, then every 12 hours for 3 days. To account for differences in growth rate among strains, the biofilm data are expressed as a ratio of crystal violet staining (O.D.580) to biomass (O.D.600), and statistical significance was determined by Student’s t-test. Biomass of each strain was determined in duplicate wells by mechanically disrupting the biofilm with a pipette tip and vigorous pipetting, followed by resuspension of the disrupted biofilm with the planktonic growth and reading the optical density at 600 nm (O.D.600). Crystal violet staining for biofilm formation was performed as follows: biofilms were washed 3 times with phosphate buffered saline (PBS), fixed with 100% ethanol, and then stained with crystal violet before washing an additional 3 times with PBS. Biofilms were destained with 33% acetic acid and the crystal violet stain was read at an optical density at 580 nm (O.D.580).

Murine model of osteomyelitis and micro-computed tomography

Osteomyelitis was induced in 6–8 week female C57BL/6J mice as previously reported (Cassat et al., 2013). For experiments to enumerate bacterial burdens from infected femurs, strains ΔhemB and ΔmenE were grown in TSB supplemented with 2 μM heme or 12.5 μM menadione, respectively to ensure that all strains were respiration-proficient at inoculation and that in vivo proliferation would not be influenced by in vitro growth rates. For experiments to calculate bone destruction in infected femurs, strains ΔhemB and ΔmenE were grown in TSB supplemented with 0.25 μM heme or 5 μM menadione, to provide the minimum amount of supplementation to support in vitro growth rates approximately equivalent to WT. For experiments to enumerate the proportions of ΔhemB and ΔmenE SCVs present in infected femurs after in vivo co-infection, groups of 5 mice were infected with a 1:1 mixture ΔhemB-ΔmenE after in vitro growth in full supplementation (2 μM heme, 12.5 μM menadione), minimal supplementation (0.25 μM heme, 5 μM menadione) or no supplementation. Prior to inoculation, bacteria were collected by centrifugation, washed with PBS, and resuspended to a concentration of 1x106 colony forming units (CFU) in 2 μl PBS. Importantly, ΔhemB and ΔmenE were not mixed until the moment of inoculation to ensure that there was no in vitro growth enhancement prior to infection. Infection was allowed to proceed for 7 or 14 days, at which time mice were euthanized and the left femur was removed and either processed for CFU enumeration or imaged by microCT. For CFU enumeration, femurs were homogenized and plated on TSA, TSA with 2 μM heme, or TSA with 5 μM menadione to determine the total number of bacteria present and the number of ΔhemB and ΔmenE mutants present, respectively. A minimum of 50 colonies were counted in duplicate when determining the proportion of ΔhemB and ΔmenE mutants in infected femurs. Differences in mean bacterial burdens from groups of mice were analyzed using Student’s t-test.

Analysis of cortical bone destruction was determined by microCT imaging as previously described (Cassat et al., 2013). Differences in cortical bone destruction and peripheral new bone formation were analyzed using Student’s t-test.

Statistical Analyses

Data were analyzed by two-tailed Student’s t-test using GraphPad Prism software. Significance was defined as p ≤ 0.05.

Supplementary Material

supplement

Acknowledgments

We thank Drs. Scott Hultgren, David Weiss, Hank Seifert, Vanessa Sperandio, and the Skaar laboratory for review of the manuscript. This research was supported by NIH AI073843 and AI069233 to E.P.S., and T32HD060554-03 and AI113107 to J.E.C. N.D.H. is a CF Foundation Ann Weinberg Memorial Research Fellow. E.P.S. is a Burroughs Wellcome Fellow in the Pathogenesis of Infectious Diseases. We thank Drs. Lynn Hancock for E. faecalis OG1RF. We acknowledge the Network on Antimicrobial Resistance in S. aureus and Nebraska Transposon Mutant Library for supplying strain JE2 and mutant derivatives.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Biswas L, Biswas R, Schlag M, Bertram R, Gotz F. Small-colony variant selection as a survival strategy for Staphylococcus aureus in the presence of Pseudomonas aeruginosa. Appl Environ Microbiol. 2009;75:6910–6912. doi: 10.1128/AEM.01211-09. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Cassat JE, Hammer ND, Campbell JP, Benson MA, Perrien DS, Mrak LN, Smeltzer MS, Torres VJ, Skaar EP. A secreted bacterial protease tailors the Staphylococcus aureus virulence repertoire to modulate bone remodeling during osteomyelitis. Cell Host Microbe. 2013;13:759–772. doi: 10.1016/j.chom.2013.05.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Cheng AG, DeDent AC, Schneewind O, Missiakas D. A play in four acts: Staphylococcus aureus abscess formation. Trends Microbiol. 2011;19:225–232. doi: 10.1016/j.tim.2011.01.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Citron DM, Goldstein EJ, Merriam CV, Lipsky BA, Abramson MA. Bacteriology of moderate-to-severe diabetic foot infections and in vitro activity of antimicrobial agents. J Clin Microbiol. 2007;45:2819–2828. doi: 10.1128/JCM.00551-07. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dunny GM, Brown BL, Clewell DB. Induced cell aggregation and mating in Streptococcus faecalis: evidence for a bacterial sex pheromone. Proceedings of the National Academy of Sciences of the United States of America. 1978;75:3479–3483. doi: 10.1073/pnas.75.7.3479. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Duthie ES, Lorenz LL. Staphylococcal coagulase; mode of action and antigenicity. Journal of General Microbiology. 1952;6:95–107. doi: 10.1099/00221287-6-1-2-95. [DOI] [PubMed] [Google Scholar]
  7. Fey PD, Endres JL, Yajjala VK, Widhelm TJ, Boissy RJ, Bose JL, Bayles KW. A genetic resource for rapid and comprehensive phenotype screening of nonessential Staphylococcus aureus genes. mBio. 2013;4:e00537–00512. doi: 10.1128/mBio.00537-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Hammer ND, Reniere ML, Cassat JE, Zhang Y, Hirsch AO, Indriati Hood M, Skaar EP. Two heme-dependent terminal oxidases power Staphylococcus aureus organ-specific colonization of the vertebrate host. mBio. 2013:4. doi: 10.1128/mBio.00241-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Hoffman LR, Deziel E, D’Argenio DA, Lepine F, Emerson J, McNamara S, Gibson RL, Ramsey BW, Miller SI. Selection for Staphylococcus aureus small-colony variants due to growth in the presence of Pseudomonas aeruginosa. Proc Natl Acad Sci U S A. 2006;103:19890–19895. doi: 10.1073/pnas.0606756104. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Hubert D, Reglier-Poupet H, Sermet-Gaudelus I, Ferroni A, Le Bourgeois M, Burgel PR, Serreau R, Dusser D, Poyart C, Coste J. Association between Staphylococcus aureus alone or combined with Pseudomonas aeruginosa and the clinical condition of patients with cystic fibrosis. J Cyst Fibros. 2013;12:497–503. doi: 10.1016/j.jcf.2012.12.003. [DOI] [PubMed] [Google Scholar]
  11. Lauderdale KJ, Boles BR, Cheung AL, Horswill AR. Interconnections between Sigma B, agr, and proteolytic activity in Staphylococcus aureus biofilm maturation. Infect Immun. 2009;77:1623–1635. doi: 10.1128/IAI.01036-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Lowy FD. Antimicrobial resistance: the example of Staphylococcus aureus. J Clin Invest. 2003;111:1265–1273. doi: 10.1172/JCI18535. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Malone CL, Boles BR, Lauderdale KJ, Thoendel M, Kavanaugh JS, Horswill AR. Fluorescent reporters for Staphylococcus aureus. J Microbiol Methods. 2009;77:251–260. doi: 10.1016/j.mimet.2009.02.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Mike LA, Dutter BF, Stauff DL, Moore JL, Vitko NP, Aranmolate O, Kehl-Fie TE, Sullivan S, Reid PR, DuBois JL, et al. Activation of heme biosynthesis by a small molecule that is toxic to fermenting Staphylococcus aureus. Proceedings of the National Academy of Sciences of the United States of America. 2013;110:8206–8211. doi: 10.1073/pnas.1303674110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Mrak LN, Zielinska AK, Beenken KE, Mrak IN, Atwood DN, Griffin LM, Lee CY, Smeltzer MS. saeRS and sarA act synergistically to repress protease production and promote biofilm formation in Staphylococcus aureus. PLoS One. 2012;7:e38453. doi: 10.1371/journal.pone.0038453. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Novick RP. Autoinduction and signal transduction in the regulation of staphylococcal virulence. Mol Microbiol. 2003;48:1429–1449. doi: 10.1046/j.1365-2958.2003.03526.x. [DOI] [PubMed] [Google Scholar]
  17. C.D.C. Antibiotic resistance threats in the United States. 2013;2013 [Google Scholar]
  18. Proctor RA, von Eiff C, Kahl BC, Becker K, McNamara P, Herrmann M, Peters G. Small colony variants: a pathogenic form of bacteria that facilitates persistent and recurrent infections. Nat Rev Microbiol. 2006;4:295–305. doi: 10.1038/nrmicro1384. [DOI] [PubMed] [Google Scholar]
  19. Schneider M, Muhlemann K, Droz S, Couzinet S, Casaulta C, Zimmerli S. Clinical characteristics associated with isolation of small-colony variants of Staphylococcus aureus and Pseudomonas aeruginosa from respiratory secretions of patients with cystic fibrosis. Journal of Clinical Microbiology. 2008;46:1832–1834. doi: 10.1128/JCM.00361-08. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Singh R, Ray P, Das A, Sharma M. Enhanced production of exopolysaccharide matrix and biofilm by a menadione-auxotrophic Staphylococcus aureus small-colony variant. J Med Microbiol. 2010;59:521–527. doi: 10.1099/jmm.0.017046-0. [DOI] [PubMed] [Google Scholar]
  21. Skaar EP, Gaspar AH, Schneewind O. IsdG and IsdI, heme-degrading enzymes in the cytoplasm of Staphylococcus aureus. J Biol Chem. 2004;279:436–443. doi: 10.1074/jbc.M307952200. [DOI] [PubMed] [Google Scholar]
  22. Voggu L, Schlag S, Biswas R, Rosenstein R, Rausch C, Gotz F. Microevolution of cytochrome bd oxidase in Staphylococci and its implication in resistance to respiratory toxins released by Pseudomonas. J Bacteriol. 2006;188:8079–8086. doi: 10.1128/JB.00858-06. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Wakeman CA, Hammer ND, Stauff DL, Attia AS, Anzaldi LL, Dikalov SI, Calcutt MW, Skaar EP. Menaquinone biosynthesis potentiates haem toxicity in Staphylococcus aureus. Molecular Microbiology. 2012;86:1376–1392. doi: 10.1111/mmi.12063. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Whittenbury R. Hydrogen Peroxide Formation and Catalase Activity in the Lactic Acid Bacteria. J Gen Microbiol. 1964;35:13–26. doi: 10.1099/00221287-35-1-13. [DOI] [PubMed] [Google Scholar]
  25. Wolter DJ, Emerson JC, McNamara S, Buccat AM, Qin X, Cochrane E, Houston LS, Rogers GB, Marsh P, Prehar K, et al. Staphylococcus aureus small-colony variants are independently associated with worse lung disease in children with cystic fibrosis. Clin Infect Dis. 2013;57:384–391. doi: 10.1093/cid/cit270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Yarwood JM, Bartels DJ, Volper EM, Greenberg EP. Quorum sensing in Staphylococcus aureus biofilms. J Bacteriol. 2004;186:1838–1850. doi: 10.1128/JB.186.6.1838-1850.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

supplement
NIHMS627821-supplement.pdf (2.4MB, pdf)

RESOURCES