Staphylococcus epidermidis biofilms undergo metabolic and matrix remodeling under nitrosative stress

Abstract

Staphylococcus epidermidis is a commensal skin bacterium that forms host- and antibiotic-resistant biofilms that are a major cause of implant-associated infections. Most research has focused on studying the responses to host-imposed stresses on planktonic bacteria. In this work, we addressed the open question of how S. epidermidis thrives on toxic concentrations of nitric oxide (NO) produced by host innate immune cells during biofilm assembly. We analyzed alterations of gene expression, metabolism, and matrix structure of biofilms of two clinical isolates of S. epidermidis, namely, 1457 and RP62A, formed under NO stress conditions. In both strains, NO lowers the amount of biofilm mass and causes increased production of lactate and decreased acetate excretion from biofilm glucose metabolism. Transcriptional analysis revealed that NO induces icaA, which is directly involved in polysaccharide intercellular adhesion (PIA) production, and genes encoding proteins of the amino sugar pathway (glmM and glmU) that link glycolysis to PIA synthesis. However, the strains seem to have distinct regulatory mechanisms to boost lactate production, as NO causes a substantial upregulation of ldh gene in strain RP62A but not in strain 1457. The analysis of the matrix components of the staphylococcal biofilms, assessed by confocal laser scanning microscopy (CLSM), showed that NO stimulates PIA and protein production and interferes with biofilm structure in a strain-dependent manner, but independently of the Ldh level. Thus, NO resistance is attained by remodeling the staphylococcal matrix architecture and adaptation of main metabolic processes, likely providing in vivo fitness of S. epidermidis biofilms contacting NO-proficient macrophages.

Keywords: Staphylococcus epidermidis; biofilm metabolism; confocal laser scanning microscopy (CLSM); nitrosative stress; nuclear magnetic resonance (NMR).

Figure 1

Staphylococcus epidermidis metabolic pathways relevant to this study. Schematic representation of the Embden–Meyerhof–Parnas (EMP; glycolytic) pathway, amino sugar pathway (GlmS, GlmM, and GlmU), pyruvate metabolism, and nitrate and nitrite respiration. AckA, acetate kinase; ETC, electron transport chain; Fru-6-P, fructose-6-phosphate; Glc-6-P, glucose-6-phosphate; GlcN-6-P, glucosamine-6-phosphate; GlcN-1-P, glucosamine-1-phosphate; GlmS, glutamine fructose-6-phosphate transaminase; GlmM, phosphoglucosamine mutase; GlmU, glucosamine 1-phosphate N-acetyltransferase; IcaA, poly-beta-1,6-N-acetyl-d-glucosamine synthase; Ldh, lactate dehydrogenase; MQ, menaquinone; Ndhs, NADH dehydrogenases; NarGHJI, nitrate reductase; NirBD, nitrite reductase; Nos, nitrous oxide reductase; NO₃ ⁻, nitrate; NO₂ ⁻, nitrite; NO, nitric oxide; N₂O, nitrous oxide; N₂, dinitrogen; Pyr, pyruvate; PIA, polysaccharide intercellular adhesin; PNAG, poly-N-acetylglucosamine; qNor, quinol-dependent nitric oxide reductase; UDP-GlcNAc, UDP-N-acetylglucosamine.

Figure 2

Mass and viability of Staphylococcus epidermidis biofilms exposed to NO and oxamate. (A) Amount of biofilm formed by S. epidermidis 1457, RP62A, and M12 strains grown for 24 and 48 h in high-glucose DMEM/FBS, in the absence (Ctr, red dots) and presence of 1 mM of NO (blue dots). Biofilm amounts of strains 1457 and RP62A were also assessed in the presence of 5 mM of oxamate (yellow dots) and oxamate+NO (green dots). Biofilm mass was determined via the crystal violet assay by measuring absorbance at 590 nm. Representative staining of the 24-h biofilms with crystal violet in well plates is shown in the graph. (B) Viable biofilm-encased cells determined by CFU counting. Scattered symbols represent individual measurements, and horizontal lines indicate median values and interquartile range; n ≥ 14 for Ctr, n ≥ 12 for NO and oxamate conditions, and n ≥ 9 for oxamate+NO. Comparisons were performed using Welch’s t-tests. Asterisks represent statistically significant differences (****, p ≤ 0.0001; ***, p ≤ 0.001; **, p ≤ 0.01; *, p ≤ 0.05). DMEM, Dulbecco’s modified Eagle medium; FBS, fetal bovine serum; CFU, colony-forming unit.

Figure 3

Substrate consumption and major end-products formed by NO- and oxamate-exposed biofilms of strains 1457 and RP62A. Lactate accumulation (A, B), glucose consumption (C, D), and acetate accumulation (E, F) by biofilms of 1457 (A, C, E) and RP62A (B, D, F) strains, grown in high-glucose DMEM/FBS, unexposed (Ctr, red) and exposed to 1 mM of NO (blue) and 5 mM of oxamate (yellow) and oxamate+NO (green). Biofilm supernatant samples for substrate and end-product concentration analysis by ¹H-NMR were harvested at the time of inoculation (0 h) and after 24 and 48 h of biofilm formation. To allow better observation of the differences between the conditions, the points at each hour in each graph were represented with a small shift, which does not indicate measurements at different times. Error bars represent concentrations’ means ± SD (n ≥ 6 for 48-h data and n ≥ 3 for 24-h data). Comparisons were performed using Welch’s t-tests. Asterisks represent statistically significant data relative to control (****, p ≤ 0.0001; ***, p ≤ 0.001; **, p ≤ 0.01; *, p ≤ 0.05).

Figure 4

Mass and viability of biofilm-free cells covering Staphylococcus epidermidis biofilms exposed to NO and extracellular lactate accumulated by planktonically grown S. epidermidis exposed to NO. Mass (A) and viability (B), measured by OD₆₀₀ and CFUs, respectively, of free cells recovered from the media covering S. epidermidis biofilms of strains 1457 and RP62A grown for 24 and 48 h in high-glucose DMEM/FBS, exposed (blue dots) or not (Ctr, red dots) to 1 mM of NO. Lactate and acetate concentrations (C) were determined by ¹H-NMR in culture supernatants of S. epidermidis strains 1457 and RP62A grown in high-glucose DMEM/FBS in planktonic conditions for 24 h in the absence (Ctr, red bar) and presence of 1 mM of NO (blue bar). Scattered symbols represent individual measurements (n ≥ 14 and n ≥ 18 for free cells mass and viability, respectively), and horizontal lines indicate median values and interquartile range. Error bars represent mean ± SD (n = 3). Welch’s t-tests were performed. Asterisks represent statistically significant data relative to control (****, p ≤ 0.0001; **, p ≤ 0.01). CFUs, colony-forming units; DMEM, Dulbecco’s modified Eagle medium; FBS, fetal bovine serum.

Figure 5

NO-induced ldh expression changes in Staphylococcus epidermidis 1457 and RP62A biofilms. Log-2-fold changes in gene expression of ldh from 24-h biofilms of (A) strain RP62A as compared to strain 1457, in the absence and presence of NO, and (B) 1457 and RP62A biofilms with NO as compared to control without NO addition. Boxes represent the interval between the 25th and 75th percentiles, and intermediate lines mark medians. Error bars represent minimum and maximum values (n ≥ 12). The expression ratios of the genes were normalized relative to the 16S constitutive gene of S. epidermidis.

Figure 6

Amount of matrix polymers and NO-induced expression changes in biofilms of Staphylococcus epidermidis exposed to NO. Quantification by confocal microscopy of the amount of exopolysaccharides (A) and proteins (B) in the matrix of 24-h biofilms of S. epidermidis strains 1457 and RP62A in the presence (blue dots) and absence (Ctr, red dots) of 1 mM of NO and 5 mM of oxamate (yellow dots) and oxamate+NO (green dots). Log-2-fold changes in gene expression of (C) glmM and icaA from 24-h biofilms of strain RP62A as compared to strain 1457, in the absence and presence of NO, and (D) glmM, glmU, and icaA of 1457 and RP62A biofilms with NO as compared to control without NO addition. (E) Confocal images, depicted as Z and orthogonal projections, representative of the S. epidermidis 1457 and RP62A biofilms exposed or unexposed (control) to 1 mM of NO, with matrix proteins and polysaccharides in red and green, respectively. Each image shows a 75 × 75 µm section of the biofilm with varying heights. Each component was detected by staining with appropriate fluorophores. Scattered symbols represent individual measurements (n ≥ 15 for protein and exopolysaccharides), and horizontal lines indicate median values and interquartile range. All comparisons were performed using t-tests with Welch’s correction, when appropriate. Asterisks represent statistically significant data relative to control (****, p ≤ 0.0001; **, p ≤ 0.01; *, p ≤ 0.05).

Figure 7

Structural properties and phenotype of the biofilms of Staphylococcus epidermidis 1457 and RP62A strains exposed to NO. 1457 and RP62A biofilms stained with crystal violet (A) and resuspended in PBS (1×) inside Eppendorf tubes (B) with or without (Ctr) the addition of 1 mM of NO. Both strains were grown in contact with silicon catheter-mimicking tubes, in high-glucose DMEM/FBS medium for 24 h, and in the absence (Ctr, red dots) or presence of 1 mM of NO (blue dots) and 5 mM of oxamate (yellow dots) and oxamate+NO (green dots). The amount of biofilm attached to the tube was measured by crystal violet staining (C). At 24 h of growth in the same conditions, visible alterations in the biofilm were present in the absence (Ctr) versus presence of NO (D). Each figure shows a representative experiment of an n ≥ 7. Single image means thickness (µM) (E) and roughness (F) of the biofilm matrices of individual confocal images of 1457 and RP62A biofilms unexposed (Ctr, red dots) and exposed to NO (blue dots). Extracellular proteins and polysaccharides were visualized by staining with appropriate fluorophores. Scattered symbols represent individual measurements (n ≥ 15), and horizontal lines indicate median values and interquartile range. All comparisons were performed using Welch’s t-tests. Asterisks represent statistically significant data relative to control (****, p ≤ 0.0001; ***, p ≤ 0.001; *, p ≤ 0.05). PBS, phosphate-buffered saline; DMEM, Dulbecco’s modified Eagle medium; FBS, fetal bovine serum.

Figure 8

Effect of nitrate and nitrite on mass and end-products profile of biofilms of Staphylococcus epidermidis 1457 and RP62A. (A) Amounts of biofilm in 1457 and RP62A strains after 48-h growth in the absence (Ctr, blue bars) and presence of nitrate (NO₃ ⁻, brown bars) and nitrite (NO₂ ⁻, yellow bars). Biofilm mass was assessed using the crystal violet assay, measured by absorbance at 590 nm. Error bars represent mean ± SD (n = 6 for Ctr and n = 3 for the +NO₃ ⁻ and +NO₂ ⁻ conditions). (B) Major extracellular metabolites accumulated by biofilms of S. epidermidis 1457 and RP62A strains grown for 48 h in high-glucose DMEM/FBS, in the absence (Ctr) and presence of 40 mM nitrate (NO₃ ⁻) and 5 mM nitrite (NO₂ ⁻). Symbols: pointed bars, lactate; hatched bars, acetate. Comparisons were performed using Welch’s t-tests. Asterisks represent statistically significant data (****, p ≤ 0.0001; ***, p ≤ 0.001; **, p ≤ 0.01; *, p ≤ 0.05).

Figure 9

Biofilm mass, extracellular nitrites, and lactate accumulated in co-cultures of Staphylococcus epidermidis biofilms with NO-proficient macrophages. (A) Evaluation of the extracellular nitrite accumulation in the supernatants of l-NMMA-untreated (light blue and dark blue bars), l-NMMA-treated (orange and red bars), M1-activated (NO, dark blue and red bars), and non-activated (light blue and orange bars) macrophages (MØ) before infection with S. epidermidis. (B, C) Lactate accumulation in the supernatants of l-NMMA-untreated (, ) and l-NMMA-treated (, ), M1-activated (, ), and non-activated (, ) macrophages (MØ) co-cultivated with S. epidermidis strains 1457 (B) and RP62A (C) in inserts/transwells for 24 h in high-glucose DMEM/FBS medium. (D) Lactate accumulation in the supernatants of l-NMMA-untreated (, ) and l-NMMA-treated (, ), M1-activated (, ), and non-activated (, ) macrophages (MØ) grown for 24 h. Nitrites were determined by the Griess method, and lactate concentrations were determined by ¹H-NMR. (E, F) Biofilm formed by S. epidermidis 1457 (E) and RP62A (F) grown for 24 h in high-glucose DMEM/FBS medium in the presence of M1-activated (dark blue and red bars) and MØ non-activated (light blue and orange bars) macrophages that were untreated (light and dark blue bars) or l-NMMA-treated (orange and red bars). Biofilm mass was determined using the crystal violet assay measured by absorbance at 590 nm. (A, E, F) Error bars represent mean ± SD (n = 6 for biofilm mass and n = 18 for nitrites). (B–D) Scattered symbols represent individual measurements, and horizontal lines indicate median values and interquartile range (n = 6). Comparison was performed using Welch’s t-tests. Asterisks represent statistically significant data relative to control (****, p ≤ 0.0001; ***, p ≤ 0.001; *, p ≤ 0.05). DMEM, Dulbecco’s modified Eagle medium; FBS, fetal bovine serum.

Figure 10

Model of nitrosative stress effect on Staphylococcus epidermidis biofilms. NO increases the amount of excreted lactate, whose production is coupled with NADH oxidation to NAD⁺, and decreases the secretion of acetate. The available NAD⁺ allows glycolysis to continue and feeds fructose-6-phosphate necessary for PIA production. Effect of NO in molecules is represented by upward (increase) and downward (decrease) pointing arrows. glmS, glutamine fructose-6-phosphate transaminase; GlmM, phosphoglucosamine mutase; glmU, glucosamine 1-phosphate N-acetyltransferase; icaA, poly-beta-1,6-N-acetyl-d-glucosamine synthase; Ldh, lactate dehydrogenase; PEP, phosphoenolpyruvate; PIA, polysaccharide intercellular adhesion; PNAG, poly-β-1,6-N-acetyl-d-glucosamine; UDP-GlcNAc, UDP-N-acetylglucosamine.