ABSTRACT
Staphylococcus aureus and Staphylococcus epidermidis are the leading causes of nosocomial infections in the United States and often are associated with biofilms attached to indwelling medical devices. Despite the importance of biofilms, there is very little consensus about the metabolic requirements of S. aureus during biofilm growth. To assess the metabolic requirements of S. aureus growing in a biofilm, we grew USA200 and USA300 clonal types in biofilm flow cells and measured the extraction and accumulation of metabolites. In spite of the genetic differences, both clonal types extracted glucose and accumulated lactate, acetate, formate, and acetoin, suggesting that glucose was catabolized to pyruvate that was then catabolized via the lactate dehydrogenase, pyruvate formate-lyase, and butanediol pathways. Additionally, both clonal types selectively extracted the same six amino acids (serine, proline, arginine, glutamine, glycine, and threonine) from the culture medium. These data and recent speculation about the importance of arginine in biofilm growth and the function of arginine deiminase in USA300 clones led us to genetically inactivate the sole copy of the arginine deiminase operon by deleting the arginine/ornithine antiporter gene (arcD) in the USA200 clonal type and to assess the effect on biofilm development and pathogenesis. Although inactivation of arcD did completely inhibit arginine transport and did reduce polysaccharide intercellular adhesin accumulation, arcD mutants formed biofilms and achieved cell densities in catheter infection studies that were equivalent to those for isogenic wild-type strains.
Staphylococcus aureus is a leading cause of nosocomial and community-acquired infections. Although the types and severity of diseases produced by this opportunistic pathogen vary, it is a frequent cause of infections associated with indwelling medical devices (e.g., catheters and artificial heart valves). Indwelling device-associated infections commonly involve the formation of a bacterial biofilm on an uncoated plastic surface or on a plastic surface coated with host proteins (58). Due to the importance of S. aureus biofilms in medical device-associated infections, a considerable amount of research has been directed at understanding the mechanisms of biofilm formation. Much of this research has focused on the bacterial mediators of biofilm formation (41, 43, 56, 59), the environmental effectors of biofilm formation (26, 28, 50), and, more recently, the global changes that occur during biofilm development (2, 44, 45, 60). The consensus from transcriptional profiling studies of S. aureus biofilms is that bacteria are growing microaerobically or anaerobically relative to planktonic cultures (2, 45). This is exemplified by increased expression of genes of the arginine deiminase and mixed acid fermentation pathways and pyruvate formate lyase. Support for the idea that staphylococci growing in a biofilm are growing microaerobically can be found in the observations that anaerobiosis increases biofilm formation and polysaccharide intercellular adhesin (PIA) synthesis (11, 53).
Although the number of known requirements for S. aureus biofilm formation is low, a considerable amount of research has provided important information regarding potential mediators of biofilm formation. As examples, S. aureus regulates the formation of biofilms in response to nutrient availability, oxygen tension, and a variety of stresses (11, 21, 25, 33, 43). Importantly, these observations highlight a recurrent regulatory theme in pathogenesis: environmental factors alter the metabolic status of the bacteria, resulting in an alteration of virulence (38) or, in this case, biofilm-forming capacity (6). Despite this being a common regulatory theme, the relationship between environmental factors and pathogenesis is poorly defined. Addressing this relationship is particularly important in the era of “omics,” when genomics, proteomics, and high-throughput mutagenesis screens consistently identify the genes of bacterial physiology and metabolism as being important, or essential, for pathogenesis (2, 3, 17, 30, 32, 39).
To enhance our understanding of the metabolic requirements of S. aureus biofilm development, two S. aureus strains (UAMS-1 and UAMS-1182) were grown in flow cells and the culture medium effluent was collected and analyzed for nutrient extraction, secondary metabolite accumulation, and oxygen concentration. These two strains were chosen because they represent both methicillin-susceptible and methicillin-resistant S. aureus phenotypes and are from distinct genetic backgrounds (USA200 and USA300 clonal types).
MATERIALS AND METHODS
Bacterial strains and growth conditions.By pulsed-field gel electrophoresis (PFGE), strain UAMS-1 is a methicillin-susceptible S. aureus strain of the USA200 clonal group, while UAMS-1182 is a methicillin-resistant S. aureus strain indistinguishable from the USA300 clonal group (data not shown) (37). PFGE control strains (i.e., known USA clonal groups) were provided by the Network on Antimicrobial Resistance in Staphylococcus aureus (www.narsa.net ). In addition, UAMS-1182 was identified as belonging to the USA300 clonal group based on the presence of the arginine catabolic mobile element (19) as determined by PCR (data not shown). Strains and plasmids used in this study are listed in Table 1. Escherichia coli strains were grown in 2× YT broth (47) or on 2× YT agar, and S. aureus strains were grown in tryptic soy broth containing 0.25% glucose (TSB) (BD Biosciences) or on TSB containing 1.5% agar. Unless otherwise stated, all bacterial cultures were inoculated 1:200 from an overnight culture (normalized for growth) into TSB, incubated at 37°C, and aerated at 225 rpm with a flask-to-medium ratio of 10:1. Antibiotics were purchased from Fisher Scientific or Sigma Chemical and, when used, were used at the following concentrations: ampicillin at 100 μg/ml (E. coli) and erythromycin at 8 μg/ml and chloramphenicol at 8 μg/ml (S. aureus).
Strains and plasmids used in this study
Construction of S. aureus arc mutants.A 1.8-kb internal fragment of the arcD gene was amplified by PCR and cloned into the SmaI site of pBluescript II KS(+) (Stratagene) to construct the plasmid pYF-1. The ermB cassette of pEC4 (7) was inserted into the NdeI site within the arcD fragment of pYF-1 to generate the plasmid pYF-2. Next, the arcD::ermB fragment was cloned into the KpnI-EcoRI site of the temperature-sensitive shuttle plasmid pTS1 (22) to generate the plasmid pYF-3. The temperature-sensitive plasmid pYF-3 was electroporated into S. aureus strain RN4220 (48) and was then introduced into strain UAMS-1 by φ85 phage transduction (40). Strain UAMS-1 containing pYF-3 was used to construct an arcD mutant by the method described by Foster (18). Putative mutants were confirmed by PCR, Southern blotting (52), and enzymatic assays.
To facilitate monitoring of the course of the catheter-associated infections, we generated an arcD mutation in S. aureus strain Xen40 (Caliper Life Sciences, Hopkinton, MA). Xen40 is strain UAMS-1 with a chromosomal insertion of the luxABCDE operon modified for expression in gram-positive bacteria (27). The mutant was made using the pKOR1 mutagenesis vector as previously described (1), using upstream primers ArcD1-attB2F and ArcD1-SacIIR and downstream primers ArcD2-SacIIF and ArcD2-attB1R (Table 2). The PCR products were digested with SacII, ligated together, amplified using the two outside primers (ArcD1-attB2F and ArcD2-attB1R), and ligated into pKOR1. The recombinant pKOR1 arcD construct was transformed into strain Xen40, and the arcD mutant was constructed as previously described (1). Inactivation of arcD was verified by PCR using the ArcD1-attB2F and ArcD2-attB1R primers, by confirming bioluminescence using the IVIS-200 imaging system (Caliper Life Sciences), and by enzymatic assays. A Xen40 arcD mutant strain (UAMS-1272) with growth characteristics indistinguishable from those of UAMS-1 arcD::ermB was chosen for the animal experiments.
Primers used in this study
PIA immunoblot assay.PIA accumulation was determined essentially as described previously (57). Briefly, bacterial cultures were grown for 16 h at 37°C with a flask-to-medium ratio of 5:1 and aerated at 160 rpm. Equivalent cell densities of S. aureus were harvested by centrifugation, and PIA was extracted with 0.5 M EDTA (pH 8.0) by boiling for 5 min. After boiling, the samples were centrifuged and the supernatants were harvested and incubated with proteinase K (20 mg/ml; Sigma-Aldrich) for 30 min. Aliquots (100 μl) of PIA samples were applied to Nytran nylon membranes (Waterman Inc.) in triplicate, air dried, and blocked overnight with 5% skim milk. Membranes were incubated for 2 h with anti-PIA antiserum and for 2 h with an anti-rabbit immunoglobulin G-alkaline phosphatase conjugate. The presence of PIA was detected by the addition of Sigma Fast 5-bromo-4-chloro-3-indolylphosphate/nitroblue tetrazolium substrate (Sigma-Aldrich). The integrated density values of spots on autoradiographs were determined with the TotalLab software (Nonlinear Dynamics Ltd.).
Flow cell biofilm formation. S. aureus strains were grown in flow cell apparatuses (Stovall Life Sciences, Greensboro, NC) coated with 20% human serum. To initiate the biofilm, 500 μl of dilute overnight cultures (optical density at 600 nm, 0.15) was injected into each chamber and the flow cell was incubated upside down at 37°C for 1 h. Following the incubation period, a continuous flow of TSB containing 0.5% glucose and 3% NaCl was delivered at a flow rate of 0.5 ml per minute per chamber. At 12 h postinoculation, effluent samples were collected at 4-h intervals and analyzed for pH and dissolved oxygen concentration or were stored at −20°C until required.
Measurement of dissolved oxygen.The dissolved oxygen concentration was monitored using an Accumet AR60 m instrument equipped with a YSI 5905 oxygen sensor (Fisher Scientific).
Measurement of glucose, lactate, acetate, formate, ammonia, and acetoin concentrations.Glucose, lactate, acetate, formate, and ammonia concentrations were determined with kits purchased from R-Biopharm, Inc., and used according to the manufacturer's directions.
Acetoin concentrations were determined essentially as described previously (23). Briefly, 1 ml of effluent was mixed sequentially with 0.2 ml 0.5% creatine, 0.2 ml 5% α-naphthol, and 0.2 ml 40% potassium hydroxide. The mixture was vortexed for 30 seconds and incubated at room temperature for 1 h. The absorbance at 540 nm was measured, and the unknown acetoin concentrations were determined with a standard curve generated using known acetoin concentrations.
Enzymatic activity assays.Bacteria were grown without agitation overnight in TSB supplemented with 20 mM arginine. Cell-free lysates of S. aureus were prepared as follows. Aliquots (3 ml) were harvested by centrifugation, washed with cold 0.01 M potassium phosphate buffer (pH 7.0), and suspended in 3.0 ml of lysis buffer containing 0.1 M potassium phosphate buffer and 25 μg/ml lysostaphin (AMBI). The samples were incubated at 37°C for 15 min and passed through a French press (twice at 15,000 lb/in2). The lysate was centrifuged for 5 min at 20,800 × g at 4°C. Arginine deiminase activity assays were performed essentially as described previously (34). In brief, cell-free lysates (100 μl) were added to Eppendorf tubes containing prewarmed (37°C) reaction buffer (400 μl) (5.8 mM l-arginine and 131 mM potassium phosphate [pH 5.8]) and incubated at 37°C for 15 min. After 15 min, the reaction was stopped by the addition of 50 μl of ice-cold 70% perchloric acid and the samples were clarified by centrifugation. The supernatant (400 μl) was analyzed for citrulline production as described by Sugawara et al. (54). One unit of arginine deiminase activity was defined as the amount of enzyme that catalyzed the formation of 1 μmol of citrulline per min. Protein concentrations were determined by the method of Lowry et al. (35).
Assay of amino acids.The concentrations of free amino acids in TSB were determined with a Beckman model 6300 amino acid analyzer (Scientific Research Consortium, Inc., St. Paul, MN).
In vivo biofilm formation.A catheter-based murine model was used to assess biofilm formation in vivo (9, 46). Briefly, anesthesia was induced by intraperitoneal injection of 0.75 mg of a 2,2,2 tribromoethanol in tert-amyl alcohol (Sigma-Aldrich) per gram of body weight. Sterile 1-cm Teflon intravenous catheter segments (B. Braun, Bethlehem, PA) were then implanted subcutaneously in each flank of 6- to 8-week-old NIH-Swiss mice (Harlan Industries, Inc., Indianapolis, IN). After wound closure with Vetbond surgical adhesive (3 M, St. Paul, MN), each catheter lumen was inoculated with 100 μl of phosphate-buffered saline (PBS) containing 5 × 105 CFU. Groups of 10 mice (20 catheters total) were infected with each strain. Mice were imaged immediately after inoculation for 5 min using the IVIS-200 imaging system. All mice from each group were humanely sacrificed at 2 days postinfection and imaged a second time immediately following sacrifice. To quantify bioluminescence, a defined region of interest corresponding to the size of the implanted catheter was created and used to measure total flux (photons/second) from each catheter. This was done both immediately after inoculation and at the time of sacrifice. Statistical significance was assessed using one-way analysis of variance with all pairwise multiple comparison (Tukey test) as formatted in SigmaStat software (SPSS Inc., Chicago, IL). To quantify biofilm formation by each strain, infected catheters were removed immediately after sacrifice, gently washed three times in sterile PBS to remove nonadherent bacteria, and subsequently placed in 5 ml of sterile PBS. Adherent bacteria were removed from the catheters by sonication. The number of recovered bacteria was then quantified by serial dilution and plate counting on the appropriate selective media. Statistical significance of quantitative plate counts was assessed using the Mann-Whitney rank sum test as formatted in SigmaStat software. All mouse infection experiments were reviewed and approved by the University of Arkansas for Medical Sciences Animal Care and Use Committee and comply with animal welfare legislation and NIH guidelines and policies.
RESULTS
Staphylococcus aureus carbohydrate metabolism during biofilm growth.Transcriptional profiling experiments suggest that S. aureus growing in a biofilm is growing microaerobically or anaerobically (2, 45). To biochemically test this possibility, we grew strains UAMS-1 and UAMS-1182 in a flow cell apparatus with a continuous flow of TSB culture medium and analyzed the effluent for the extraction of glucose and accumulation of common fermentation products (Fig. 1 and 2). Consistent with the transcriptional profiling data, S. aureus growing in a biofilm extracted glucose from the culture medium and accumulated the fermentation products lactic acid, acetic acid, and formic acid (Fig. 2). The accumulation of organic acids in the culture medium occurred coincident with a decreased pH and induction of the butanediol pathway, as suggested by the accumulation of acetoin (Fig. 2). Bacteria induce the butanediol pathway in response to a decrease in the environmental pH (5) and as a means to prevent a redox imbalance by oxidizing NADH to NAD+. The accumulation of organic acids and alcohols in the culture medium effluent is strongly indicative of oxygen-limited growth. To determine if the biofilm was oxygen limited, we measured the oxygen concentration of the culture medium effluent for strains UAMS-1 and UAMS-1182 (Fig. 3). As expected, as the biofilm matured, it depleted oxygen from the culture medium (Fig. 3). Taken together, these data suggest that reduced oxygen tension induced a fermentative metabolism resulting in increased pH stress.
Growth of S. aureus strains UAMS-1, UAMS-1182, and UAMS-1 arcD::ermB in a three-chamber flow cell. Bacterial strains were grown at 37°C with a continuous flow (0.5 ml per minute per chamber) of TSB containing 0.5% glucose and 3% NaCl. The results are representative of multiple (>3) independent experiments. In the figure, strain UAMS-1 arcD::ermB is abbreviated as U1-arcD::ermB.
Metabolite extraction and accumulation by S. aureus strains UAMS-1 and UAMS-1182 during biofilm growth in flow cells. (A) The decreased pH of the culture medium corresponds to decreasing glucose concentrations. (B) Ammonia accumulation in the culture medium effluent. (C and D) Accumulation of organic acids and acetoin in the culture medium effluent of strains UAMS-1 (C) and UAMS-1182 (D). (E and F) Free amino acid extraction from the culture medium by strains UAMS-1 (E) and UAMS-1182 (F). The results presented are representative of at least two independent experiments.
Dissolved oxygen concentration in the culture medium effluent of strains UAMS-1 and UAMS-1182 grown in flow cells. The results presented are representative of at least three independent experiments.
S. aureus growing in a biofilm selectively extracts amino acids.Bacteria regulate the selective transport of amino acids into the cytosol; that is, they take what they need when they need it (2, 17, 24, 49, 51). This is an advantageous adaptation allowing bacteria to acquire necessary amino acids from their environment and redirect energy from amino acid biosynthesis into growth. It has been postulated that amino acid catabolism is important for staphylococcal biofilm formation and pH homeostasis (2, 44, 45). To determine which amino acids S. aureus extracted from the culture medium during biofilm growth, we measured the free amino acid concentrations of culture medium effluent and expressed this as a function of the free amino acid concentration in sterile culture medium (Fig. 2E and 2F). Although the extent to which amino acids were extracted varied, strains UAMS-1 and UAMS-1182 immediately began selectively extracting the free amino acids glutamine, serine, proline, glycine, threonine, and arginine, while other amino acids (e.g., valine and tryptophan) remained in the medium. The extraction of these six amino acids differs somewhat from that seen in planktonic growth, where during exponential growth serine, glycine, glutamine, glutamate, alanine, and threonine are selectively extracted from the culture medium and proline and arginine are not extracted until carbon becomes limiting (51; G. A. Somerville, unpublished data). The selective extraction of amino acids from the culture medium during biofilm growth suggested that the amino acids were being used for carbon and/or energy production and not just for protein synthesis. Amino acid catabolism usually involves the release of ammonia through a deamination step; thus, it is possible to determine the fate of amino acids by measuring the concentration of ammonia in the culture medium effluent (Fig. 2B). Concomitant with the extraction of amino acids from the culture medium was the accumulation of ammonia in the effluent, indicating amino acid catabolism. These data led us to speculate that interfering with amino acid transport and/or catabolism might limit biofilm formation, PIA synthesis, or virulence. To test these possibilities, we chose to genetically inactivated the arcD gene (arcD encodes the arginine/ornithine antiporter) of strain UAMS-1 and assessed its ability to synthesize PIA and form a biofilm. The choice to inactivate the arginine transporter, as opposed to a transporter of one of the other selectively extracted amino acids, was based on previous transcriptional profiling data (2, 45, 60), proteomic data (44), and speculation about the potential importance of the arginine deiminase pathway in staphylococcal pathogenesis (15).
Growth of the UAMS-1 arcD mutant strain.In S. aureus, the genes coding for the anaerobic catabolism of arginine are contained within the arginine deiminase (arc) operon. The arc operon is comprised of genes for arginine deiminase (arcA), ornithine transcarbamylase (arcB), the arginine/ornithine antiporter (arcD), and carbamate kinase (arcC). The arginine/ornithine antiporter is a cationic exchange system that catalyzes a one-to-one stoichiometric exchange of arginine and ornithine (12, 16, 42); hence, as the bacteria extract arginine from the culture medium, ornithine accumulates in the medium. As expected, inactivation of the arcD gene prevented the extraction of arginine from the culture medium and the accumulation of ornithine in the medium during biofilm growth (Fig. 4A). Inactivation of arcD also reduced the accumulation of ammonia in the culture medium (after 48 h of growth, the concentration of ammonia in the culture medium effluent was 3.31 mM for UAMS-1 and 1.38 mM for UAMS-1 arcD::ermB.) Additionally, under microaerobic growth conditions (5:1 flask-to-medium ratio and 100 rpm aeration), the arcD mutant strain (UAMS-1 arcD::ermB) accumulated a significantly lower concentration of ammonia in the culture medium than isogenic wild-type strain (Fig. 4D), suggesting that arginine catabolism was inhibited. Taken together, these data confirm that the correct gene was inactivated and that it was the sole arginine transporter.
Growth characteristics of strains UAMS-1 and UAMS-1 arcD::ermB grown under biofilm and planktonic conditions. (A) Arginine extraction and ornithine accumulation in culture medium of strains UAMS-1 and UAMS-1 arcD::ermB grown in biofilm flow cells. (B) pH of the biofilm culture medium effluent from panel A. (C and D) Planktonic growth of strains UAMS-1 and UAMS-1 arcD::ermB under microaerobic conditions (5:1 flask-to-medium ratio, TSB, and 100 rpm aeration.) (C) Growth and pH of the culture medium of strains UAMS-1 and UAMS-1 arcD::ermB. (D) Ammonia accumulation in the culture medium as a function of growth (A600) for the cultures in panel C. The results presented are representative of at least two independent experiments.
Arginine transport and/or catabolism enhances the accumulation of PIA.Two different S. aureus clonal types selectively extracted six amino acids, including arginine, from the culture medium during growth in a biofilm (Fig. 2E and F). These data and the fact that arcD inactivation does not cause a growth defect during planktonic growth (Fig. 4B) suggest that arginine may be important for one or more aspects of biofilm maturation. To test this possibility, we assessed the ability of the UAMS-1 and UAMS-1 arcD::ermB strains to grow in biofilm flow cells. The gross morphologies of biofilms formed by strain UAMS-1 arcD::ermB and by strains UAMS-1 and UAMS-1182 were very similar (Fig. 1), demonstrating that bacterial attachment and biofilm maturation can occur independent of arginine transport or catabolism. To determine if arcD inactivation affected PIA accumulation, we determined the amount of cell-associated PIA during the exponential and stationary phases of planktonic growth. Interestingly, the accumulation of PIA was lower in strain UAMS-1 arcD::ermB than in the isogenic strain UAMS-1 (46 to 51% reduction; P = 0.0015), suggesting that the biofilm formed by the arcD mutant strain (Fig. 1) might be PIA limited.
Effect of arginine transport on catheter-associated infections.To facilitate monitoring of the course of the catheter-associated infections, we deleted the arcD gene from a strain of UAMS-1 carrying the luxABCDE operon (strain Xen40) and measured in vivo bioluminescence. Consistent with the results for strain UAMS-1 arcD::ermB, the Xen40 arcD deletion mutant (strain UAMS-1272) had reduced PIA accumulation (46 to 51% reduction; P = 0.0015) and formed a biofilm in a flow cell equivalent to those for the isogenic wild-type strain (data not shown). Additionally, genetic inactivation of arcD in both UAMS-1 and Xen40 inhibited arginine deiminase activity (data not shown). The in vivo growth of the arc operon mutant and the wild-type strains was statistically equivalent using the Mann-Whitney rank sum test (P = 0.218). Similarly, bioluminescence on day 2 was equivalent for the wild-type and arcD mutant strains (data not shown). Taken together, these data demonstrate that in vitro and in vivo biofilms can mature in the absence of the arginine deiminase/arc operon.
DISCUSSION
Indwelling medical device-associated infections caused by bacterial biofilms are a significant source of morbidity in the United States, with two of the more common causative agents of these infections being S. aureus and Staphylococcus epidermidis (55). In spite of the prevalence of biofilm-associated staphylococcal infections, there is little consensus about the requirements necessary to form a biofilm. To address the metabolic aspects of this deficiency, we grew two genetically distinct S. aureus strains in biofilm flow cells, measured the extraction and accumulation of metabolites, and assessed the importance of one metabolite, arginine, on biofilm development and pathogenesis.
Consistent with previous observations (2, 45), our data demonstrate that S. aureus growing in a biofilm is growing microaerobically (Fig. 2 and 3). We found that mature biofilms maintain a dynamic metabolic flux of carbon and amino acid uptake and excretion (Fig. 2), a situation similar to that proposed for static S. aureus biofilms (45). This dynamic metabolic flux contrasts with that observed in S. epidermidis biofilms, where it was determined that growth in a biofilm leads to “low metabolic” activity (60). Specifically, our data strongly suggest that S. aureus catabolizes glucose to pyruvate by glycolysis and then catabolizes pyruvate via lactate dehydrogenase (EC 1.1.1.27) and pyruvate formate-lyase (EC 2.3.1.54) (Fig. 2C and D). Glycolysis generates two molecules of pyruvate for every molecule of glucose consumed; however, in the process it reduces two molecules of NAD+ to NADH. Reduction of NAD+ to NADH, without an equivalent means to oxidize NADH, can create a redox imbalance and inhibit growth. Under microaerobic or anaerobic growth conditions, the majority of pyruvate is reduced to lactic acid (29, 31), concomitant with a 1:1 stoichiometric oxidation of NADH back to NAD+. To compensate for the diversion of pyruvate into pyruvate formate-lyase, S. aureus also shunts pyruvate into the butanediol pathway via acetolactate synthase (EC 2.2.1.6) (Fig. 2C and D). Shunting pyruvate into the butanediol pathway will facilitate oxidation of NADH and decrease the accumulation of organic acids. These data and deductions suggest that a tenuous redox balance exists during biofilm growth, a potentially exploitable weakness for therapeutic intervention. This suggestion is supported by the observation that inactivation of the butanediol pathway at acetolactate decarboxylase (EC 4.1.1.5) inhibits biofilm formation (8). In addition, this suggestion is a reasonable explanation for the absence of dramatic effects on in vivo or in vitro biofilm formation in arginine transport- and/or catabolism-deficient strains; specifically, the arginine deiminase pathway is not involved in the reduction or oxidation of NAD+ or NADH.
In a complex medium containing glucose, S. aureus will preferentially catabolize glucose for carbon and energy through the glycolytic and pentose phosphate pathways (4, 10, 20, 29, 36). Despite a preference for glucose and a high concentration of glucose in the medium (>15 mM) (Fig. 2A), S. aureus strains growing in a biofilm selectively extract and catabolize six amino acids, including arginine, from the culture medium (Fig. 2 and 4). These data lend credence to the speculation that arginine metabolism may be important for staphylococcal survival and/or pathogenesis (2, 15, 44, 45). This speculation generated two very reasonable hypotheses regarding the role of arginine catabolism in staphylococcal host-pathogen interactions. First, the deamination of arginine by the arginine deiminase pathway results in the extracellular accumulation of ammonia, facilitating bacterial pH homeostasis (2, 44, 45). Second, the staphylococcal arginine deiminase may function in a capacity similar to that in Streptococcus pyogenes (15); specifically, arginine deiminase may aid in evasion of the host immune response by inhibiting peripheral blood mononuclear cell proliferation (13, 14). Hence, one function of arginine may be to induce transcription of the arginine deiminase (arc) operon. To test the first hypothesis, we genetically inactivated the arcD gene, eliminating arginine transport (Fig. 4A), and assessed the effect on ammonia accumulation and the extracellular pH (Fig. 4). Inactivation of arcD completely inhibited the transport of arginine and the accumulation of ornithine (Fig. 4A) and significantly decreased the accumulation of ammonia; however, the pH of the biofilm effluent was unchanged (Fig. 4B). Therefore, in contrast to previous speculation (2, 15, 44, 45), our results demonstrate that the ammonia generated by the arginine deiminase pathway is insufficient to offset the pH decrease due to the accumulation of organic acids (Fig. 4B and C). To test the second hypothesis, we assessed the effect of arcD inactivation in a mouse indwelling device infection model. Mutants lacking arcD have dramatically reduced arginine deiminase activity (data not shown), yet they achieved in vivo cell densities equivalent to that of the isogenic wild-type strain, suggesting that arginine deiminase is not required for biofilm formation and survival in vivo, at least as defined by this model. It is possible that arginine deiminase is important in certain types of infections, such as when preferred nutrients are limiting or during a more active and prolonged immune response; however, additional work will be needed to address these possibilities.
ACKNOWLEDGMENTS
This paper is a contribution of the University of Nebraska Agricultural Research Division, supported in part by funds provided through the Hatch Act, the National Institute of General Medical Sciences (GM076585), and the American Heart Association (0760005Z) to G.A.S.; the National Institute of Allergy and Infectious Diseases (AI43356) to M.S.S.; and the Intramural Research Program of the NIAID to M.O.
We thank the Network on Antimicrobial Resistance in Staphylococcus aureus for providing the PFGE strain standards used in this study.
FOOTNOTES
- Received 9 April 2007.
- Returned for modification 27 May 2007.
- Accepted 10 June 2007.
- Copyright © 2007 American Society for Microbiology
