Low Concentrations of Nitric Oxide Modulate Streptococcus pneumoniae Biofilm Metabolism and Antibiotic Tolerance

Abstract

Streptococcus pneumoniaeis one of the key pathogens responsible for otitis media (OM), the most common infection in children and the largest cause of childhood antibiotic prescription. Novel therapeutic strategies that reduce the overall antibiotic consumption due to OM are required because, although widespread pneumococcal conjugate immunization has controlled invasive pneumococcal disease, overall OM incidence has not decreased. Biofilm formation represents an important phenotype contributing to the antibiotic tolerance and persistence ofS. pneumoniaein chronic or recurrent OM. We investigated the treatment of pneumococcal biofilms with nitric oxide (NO), an endogenous signaling molecule and therapeutic agent that has been demonstrated to trigger biofilm dispersal in other bacterial species. We hypothesized that addition of low concentrations of NO to pneumococcal biofilms would improve antibiotic efficacy and that higher concentrations exert direct antibacterial effects. Unlike in many other bacterial species, low concentrations of NO did not result inS. pneumoniaebiofilm dispersal. Instead, treatment of bothin vitrobiofilms andex vivoadenoid tissue samples (a reservoir forS. pneumoniaebiofilms) with low concentrations of NO enhanced pneumococcal killing when combined with amoxicillin-clavulanic acid, an antibiotic commonly used to treat chronic OM. Quantitative proteomic analysis using iTRAQ (isobaric tag for relative and absolute quantitation) identified 13 proteins that were differentially expressed following low-concentration NO treatment, 85% of which function in metabolism or translation. Treatment with low-concentration NO, therefore, appears to modulate pneumococcal metabolism and may represent a novel therapeutic approach to reduce antibiotic tolerance in pneumococcal biofilms.

FIG 1

SNP treatment of in vitro pneumococcal biofilms reduced biofilm viability and biomass. Forty-eight-hour S. pneumoniae serotype 14 in vitro biofilms were treated with SNP for 2 h, the biomass was assessed by absorbance (OD₆₀₀), and the viability was measured. Significant reductions in total biomass and in the number of viable cells remaining within the biofilm were observed following 1 mM SNP treatment. ***, P ≤ 0.001. The error bars indicate standard deviations.

FIG 2

SNP treatment of in vitro S. pneumoniae biofilms reduced the viable-cell population in the surrounding supernatant. The viability of 48-h S. pneumoniae serotype 14 in vitro biofilm and supernatant populations was measured by CFU enumeration following treatment with 1 mM SNP. SNP treatment significantly reduced both the biofilm and supernatant populations. *, P ≤ 0.05. The error bars indicate standard deviations.

FIG 3

In vitro S. pneumoniae biofilms treated with SNP demonstrated reduced viability and no evidence of dispersal. Forty-eight-hour S. pneumoniae serotype 14 in vitro biofilms were treated with 1 mM SNP for 2 h and then imaged using confocal microscopy and LIVE/DEAD staining. (a to d) SNP-treated biofilms (1 mM) (b) demonstrated no obvious change in biomass compared with untreated biofilms (a); however, reduction in the number of Syto9-stained live bacteria in the 1 mM SNP-treated biofilms (d) was less than in untreated biofilms (c), commensurate with CFU enumeration data. (e and f) Scanning electron microscopy with alcian blue staining further demonstrated no obvious changes in biofilm ultrastructure between untreated (e) and 1 mM SNP-treated (f) biofilms (magnification, ×4,000; scale bar, 10 μm).

FIG 4

SNP treatment reduced the in vitro S. pneumoniae planktonic growth rate. S. pneumoniae serotype 14 in vitro planktonic cultures were treated with SNP during exponential growth phase, and the growth rate was measured by the change in absorbance (OD₅₉₅) over 2 h and compared with the growth rate of untreated cultures. A significant reduction in the growth rate was observed using 500 μM SNP, and complete cessation of growth was observed with concentrations greater than 5 mM. *, P ≤ 0.05; ****, P ≤ 0.0001. The error bars indicate standard deviations.

FIG 5

The response of S. pneumoniae to treatment with SNP is NO mediated. (a) S. pneumoniae serotype 14 exponential planktonic cultures were treated with the NO donors SNP, DEA/NO, nitrate, and nitrite and the CN⁻ anion control KCN over 2 h. Significant decreases in the growth rate were observed upon treatment with two independent NO donors, SNP and DEA/NO, indicating that the response was NO mediated. KCN treatment had no effect on the growth rate, confirming that the response to SNP was not CN⁻ mediated (P = 0.528). Sodium nitrate (P = 0.321) and sodium nitrite (P = 0.078) treatments also had no effect on the growth rate, suggesting that nitrate and nitrite were not utilized as sources of NO. (b and c) The addition of the NO scavenger carboxy-PTIO (b) and the peroxynitrite scavenger l-methionine (c) reduced the response to SNP treatment, suggesting the response may be mediated by either NO or peroxynitrite. **, P ≤ 0.01; ***, P ≤ 0.001. The error bars indicate standard deviations.

FIG 6

Adjunctive treatment of S. pneumoniae in vitro biofilms with SNP enhanced antibiotic efficacy. Forty-eight-hour S. pneumoniae serotype 14 (ST124), 19F, 23F, and D39 in vitro biofilms were treated for 2 h, and the remaining viable cells were measured by CFU enumeration. When used separately, both SNP and AMC treatments reduced the viable biofilm cell population; however, combined SNP and AMC treatment resulted in a further significant reduction in viability. *, P ≤ 0.05. The error bars indicate standard deviations, with the central horizontal lines indicating means.

FIG 7

Adjunctive treatment of S. pneumoniae biofilms on ex vivo adenoid tissue with SNP enhanced antibiotic efficacy. Adenoid tissue samples (n = 11) were dissected into four equal sections (each with a similar proportion of luminal surface) and treated for 2 h, and the viability of S. pneumoniae was measured by CFU enumeration. SNP treatment alone had no significant effect on viable pneumococci (P = 0.722), whereas AMC treatment alone resulted in a significant reduction (P = 0.005). Combined SNP and AMC treatment, however, resulted in enhanced antibiotic efficacy (P = 0.041). *, P ≤ 0.05 (Wilcoxon signed-rank test). The error bars indicate standard deviations.

FIG 8

Treatment of S. pneumoniae in vitro biofilms with SNP resulted in the differential expression of a small subset of quantitatively identified proteins. Comparative iTRAQ analyses of SNP-treated (100 μM SNP for 2 h) and untreated S. pneumoniae serotype 14 7-day-old in vitro biofilms quantitatively identified 112 proteins, 13 of which were differentially expressed following treatment.

FIG 9

Treatment of S. pneumoniae in vitro biofilms with SNP resulted in a change in metabolic and translation protein expression levels. Comparative iTRAQ analyses of SNP-treated (100 μM SNP for 2 h) and untreated S. pneumoniae serotype 14 in vitro biofilms quantitatively identified 13 differentially expressed proteins (a) and qualitatively identified 12 differentially expressed proteins (b) following treatment. The quantitative inclusion criteria were as follows: ≥3 peptide matches, a protein score of ≥50, and ≥5% sequence coverage (P < 0.05). The qualitative inclusion criteria were as follows: 2 peptide matches, a protein score of ≥50, and ≥5% sequence coverage (P < 0.05). Comparative protein data with ratios of >1.3 and <0.77 were identified as showing differential protein expression.