MtlABFD operon did not impair survival from AFAs, in contrast to inactivation of mtlD alone. Proportionately reduced AFA survival was observed with an mtlD but not an mtlABFD inactivation in S. aureus Newman (Liv1027 and Liv1028, respectively; Table 1) (data not shown). Reduced survival of the mtlD mutant was fully complemented with the entire mtlABFD operon present on pMJH71 using strain Liv1098 (SH1000 mtlD::tet pMJH71) (Figure 2). The reduced survival of Liv1023 (SH1000 mtlD::tet) on linoleic acid agar was supported with a significantly reduced linoleic acid MIC (0.4560.02 mM) (p,0.004) in BHI medium, compared to SH1000 (0.960.04 mM), Liv1024 (0.6960.02 mM) and Liv1098 (0.8560.03 mM). Strain Liv1023 (SH1000 mtlD::tet) exhibited a profound growth defect when cultured in broth containing Mtl as the carbohydrate source (peptone 10 g l21, Mtl 10 g l21, beef extract 1 g l21, NaCl 10 g l21) (Figure 6A). Substituting the sugar alcohol Mtl for theS. aureus Mannitol Utilisation and SurvivalFigure 3. Schematic representation of the mtlABFD locus. Position of the transposon insertion and allelic replacements created during this study. doi:10.1371/journal.pone.0067698.gsugars fructose or glucose restored normal growth, demonstrating the Mtl-specific defect (data not shown). S. aureus accumulates intracellular Mtl following incubation in the presence of glucose. To test if this accumulation affected survival from AFAs, the relative survival of exponential cells (OD600 = 1) of SH1000 incubated in PBS containing 1 (w/v) glucose was determined after growth on 1 mM linoleic acid agar. No clear difference in survival of the strains was observed. All strains grew equally well at 37uC in BHI broth (data not shown), however a pronounced reduction in growth rate was observed for strain Liv1023 (SH1000 mtlD::tet) when cultured in BHI broth at 25uC (Figure 6B). This defect was specific to inactivation of mtlD but not for deletion of the complete operon. Starvation survival with limiting glucose was not impaired in mtl mutant strains [35,37]. Growth of S. aureus SH1000 was tested in the absence or presence of mannitol (0.1 M, 0.5 M), with or without 1 mM linoleic acid to test for synergy. Mannitol was shown to have similar properties as ethanol [13], by acting synergistically with linoleic acid as evident by the reduced viable count with increasing mannitol concentration (Figure 7).Analysis of Cellular MetabolitesA comparative metabolomics analysis was undertaken to identify the intracellular metabolites of exponentially growing cells of strains SH1000, Liv1023 (SH1000 mtlD::tet) and Liv1024 (SH1000 mtlABFD::tet). This revealed that inactivation of mtlD Title Loaded From File resulted in an accumulation of Mtl and Mtl-P, the latter being undetectable in both SH1000 and Liv1024 (Table 2 and supplementary table 1). The total relative levels of Mtl species were over 20-fold Title Loaded From File greater in Liv1023 (SH1000 mtlD::tet) than SH1000. The near absence of Mtl in strain Liv1024 supports data that the MtlAB PTS transporter is the main portal for Mtl uptake [38]. Inactivation of mtlD and mtlABFD resulted in the absence of cellular Sorbitol-6-P (Table 2). Further clear differences in metabolite levels were evident in strain Liv1023 (mtlD::tet) relative to SH1000 and Liv1024 (Table S1).Figure 4. Mtl fermentation capability of S. aureus strains. Mtl fermentation is revealed by acid formation and colour change of the pH indicator to yellow. Liv1023 (SH1000 mtlD::tet) and Liv1024 (SH1000 mtlABFD::tet).MtlABFD operon did not impair survival from AFAs, in contrast to inactivation of mtlD alone. Proportionately reduced AFA survival was observed with an mtlD but not an mtlABFD inactivation in S. aureus Newman (Liv1027 and Liv1028, respectively; Table 1) (data not shown). Reduced survival of the mtlD mutant was fully complemented with the entire mtlABFD operon present on pMJH71 using strain Liv1098 (SH1000 mtlD::tet pMJH71) (Figure 2). The reduced survival of Liv1023 (SH1000 mtlD::tet) on linoleic acid agar was supported with a significantly reduced linoleic acid MIC (0.4560.02 mM) (p,0.004) in BHI medium, compared to SH1000 (0.960.04 mM), Liv1024 (0.6960.02 mM) and Liv1098 (0.8560.03 mM). Strain Liv1023 (SH1000 mtlD::tet) exhibited a profound growth defect when cultured in broth containing Mtl as the carbohydrate source (peptone 10 g l21, Mtl 10 g l21, beef extract 1 g l21, NaCl 10 g l21) (Figure 6A). Substituting the sugar alcohol Mtl for theS. aureus Mannitol Utilisation and SurvivalFigure 3. Schematic representation of the mtlABFD locus. Position of the transposon insertion and allelic replacements created during this study. doi:10.1371/journal.pone.0067698.gsugars fructose or glucose restored normal growth, demonstrating the Mtl-specific defect (data not shown). S. aureus accumulates intracellular Mtl following incubation in the presence of glucose. To test if this accumulation affected survival from AFAs, the relative survival of exponential cells (OD600 = 1) of SH1000 incubated in PBS containing 1 (w/v) glucose was determined after growth on 1 mM linoleic acid agar. No clear difference in survival of the strains was observed. All strains grew equally well at 37uC in BHI broth (data not shown), however a pronounced reduction in growth rate was observed for strain Liv1023 (SH1000 mtlD::tet) when cultured in BHI broth at 25uC (Figure 6B). This defect was specific to inactivation of mtlD but not for deletion of the complete operon. Starvation survival with limiting glucose was not impaired in mtl mutant strains [35,37]. Growth of S. aureus SH1000 was tested in the absence or presence of mannitol (0.1 M, 0.5 M), with or without 1 mM linoleic acid to test for synergy. Mannitol was shown to have similar properties as ethanol [13], by acting synergistically with linoleic acid as evident by the reduced viable count with increasing mannitol concentration (Figure 7).Analysis of Cellular MetabolitesA comparative metabolomics analysis was undertaken to identify the intracellular metabolites of exponentially growing cells of strains SH1000, Liv1023 (SH1000 mtlD::tet) and Liv1024 (SH1000 mtlABFD::tet). This revealed that inactivation of mtlD resulted in an accumulation of Mtl and Mtl-P, the latter being undetectable in both SH1000 and Liv1024 (Table 2 and supplementary table 1). The total relative levels of Mtl species were over 20-fold greater in Liv1023 (SH1000 mtlD::tet) than SH1000. The near absence of Mtl in strain Liv1024 supports data that the MtlAB PTS transporter is the main portal for Mtl uptake [38]. Inactivation of mtlD and mtlABFD resulted in the absence of cellular Sorbitol-6-P (Table 2). Further clear differences in metabolite levels were evident in strain Liv1023 (mtlD::tet) relative to SH1000 and Liv1024 (Table S1).Figure 4. Mtl fermentation capability of S. aureus strains. Mtl fermentation is revealed by acid formation and colour change of the pH indicator to yellow. Liv1023 (SH1000 mtlD::tet) and Liv1024 (SH1000 mtlABFD::tet).
