NorA and Tet38 efflux pumps enable Staphylococcus aureus survival in the cystic fibrosis airway environment, resistance to antibiotics, and coinfection with Pseudomonas aeruginosa

Abstract

Efflux pumps play multiple roles in bacterial physiology, environmental adaptation, and antibiotic resistance. Early cystic fibrosis (CF) airway infections start with Staphylococcus aureus (SA), often later followed by Pseudomonas aeruginosa (PA) infections. In this study, we have evaluated the role of SA pumps NorA and Tet38 in survival and interaction with PA in CF patients. Data showed a ≥4-log₁₀CFU/mL growth deficit of SA mutants ΔnorA and Δtet38 in an artificial sputum medium (ASM), suggesting NorA and Tet38 contributed to SA growth in CF sputum. In ASM, mucin activated norA but inhibited tet38, while extracellular DNA activated tet38 but inhibited norA, demonstrating complementary roles of mucin and DNA in affecting NorA and Tet38 expression. Furthermore, exposure of SA wild type to PA-excreted molecules affected pump expression; 3,4-dihydroxy-2-heptylquinoline PQS caused an increase in tet38 but a decrease in norA, 4-hydroxy-2-heptylquinoline HHQ caused a decrease in norA and tet38, and pyocyanin PYO caused a modest increase in norA, demonstrating differing additional roles of PA-secreted molecules influencing NorA and Tet38 expression. Evaluation of 48 randomly selected unique CF-associated SA showed that 18.8% were NorA-overexpressors and 10.4% were Tet38-overexpressors. NorA-overexpressors showed a fourfold increase in pyocyanin MIC and ≥16-fold in ciprofloxacin MIC. Furthermore, 89% of NorA-overexpressors carried an insertion of CAAT/ACAA/CTAT at the (-10) motif of the norA promoter, and 62.5% of these were co-isolated with PA.These data showed that SA survives PA killing in CF sputum conditions using sputum key components and PA-specific signal molecules to regulate NorA and Tet38 expression.

Keywords: NorA; P. aeruginosa; S. aureus; Tet38; cystic fibrosis; efflux pump.

Fig 1

S. aureus WT and pump mutant growth curves in various ASM media. S. aureus growth curves in ASM medium. WT, efflux pump mutants (A), and mutant strains complemented with the corresponding pump genes (B) were cultured in ASM medium for 20 hours, and then plating and colony counting were performed. The growth curve assays were repeated three times with three different biological samples. The error bars represent the means of (log₁₀CFU/mL ± SD) for each assay. The differences in the log₁₀CFU/mL of mutants ΔnorA and Δtet38 compared to that of WT and mutant ΔnorC were statistically significant as determined by a one-way ANOVA (P < 0.05). WT, S. aureus RN6390; WT(pLI50), RN6390 transformed with the empty plasmid. NorA_compl, mutant ΔnorA transformed with plasmid construct (pLI50-norA). Tet38_compl, mutant Δtet38 transformed with plasmid construct (pLI50-tet38). S. aureus growth curves in ASM, ASM without mucin, and ASM without DNA. WT, efflux pump mutant ΔnorA (C), or Δtet38 (D) were cultured in ASM medium, ASM without 0.5% of mucin or 0.4% of DNA for 20 hours, and then colony counting was performed. The growth curve assays were repeated three times with three different biological samples. The error bars represent the means of (log₁₀CFU/mL ± SD) for each assay. The differences in the log₁₀CFU/mL of mutant ΔnorA or Δtet38 compared to that of WT in three ASM media were statistically significant as determined by a one-way ANOVA (P < 0.05). The differences in the log₁₀CFU/mL between Δtet38 in ASM and ASM without DNA vs. ASM without mucin was statistically significant as determined by a one-way ANOVA (P < 0.05). WT, S. aureus RN6390; ΔnorA, mutant lacks NorA but expresses Tet38; Δtet38, mutant lacks Tet38 but expresses NorA. ASM, medium prepared with mucin and DNA (14). No DNA, ASM medium prepared without 0.4% of fish sperm DNA; No mucin, ASM medium prepared without 0.5% of mucin from porcine stomach. WT vs. ΔnorA, norA mutant compared to WT; WT vs. Δtet38, tet38 mutant compared to WT.

Fig 2

Gene transcript levels in ASM media and ASM media without mucin or DNA. S. aureus WT (A), mutant ΔnorA (B), and mutant Δtet38 (C) were cultured in TSB, ASM, ASM without mucin (ASM – Mucin), and ASM without DNA (ASM – DNA) (initial CFU/mL ~ 10⁴) for 1 hour, then quantitative real-time RT-PCR assays (qPCR) were performed to assess the level of norA and tet38 of WT (A), tet38 of mutant ΔnorA (B), and norA of mutant Δtet38 (C). The relative transcript levels of norA and tet38 were expressed as the fold change (FC) in pump gene transcripts of bacteria grown in ASM, ASM without mucin, or ASM without DNA versus bacteria grown in TSB. The assays were repeated three times with three different biological samples. The error bars represent the means of FC ± SEM for each assay. The (*) represents the differences in the FC of norA or tet38 of WT or mutants grown in various ASM versus TSB; and (**) represents the differences in the FC of norA or tet38 of WT or mutants grown in ASM without mucin or DNA versus ASM. The differences were statistically significant as determined by an ANOVA (P < 0.05).

Fig 3

Variation of SA norA and tet38 transcripts in different conditions. norA and tet38 transcripts of SA WT exposed to supernatants of PA. The PA supernatants were prepared from overnight cultures in TSB. SA WT grew in TSB or ASM until OD₆₀₀ ~0.5, then 30% PA supernatant was added for 1 hour. Quantitative real-time RT-PCR assays (qPCR) were performed to assess the level of norA and tet38 of WT exposed to PA in TSB (A) or WT exposed to PA in ASM (B). The relative transcript levels of norA and tet38 were expressed as the fold change (FC) in pump gene transcripts of SA WT grown in TSB or ASM and exposed to PA supernatants versus non-exposed SA WT. The assays were repeated three times with three different biological samples. The error bars represent the means of FC ± SEM for each assay. The (*) represents the differences in the FC of norA or tet38 of SA WT in TSB or ASM and exposed to PA supernatants versus SA WT non-exposed in TSB or ASM. The differences were statistically significant as determined by an ANOVA (P < 0.05). SA WT, S. aureus RN6390; PA WT, P. aeruginosa PA14. norA and tet38 transcripts of SA WT exposed to PA QS molecules. SA WT (10⁵ CFU/mL) was exposed to PA WT supernatant, HHQ, PQS, and PYO at 0.5 × MIC concentrations (C), or to PYO concentrations (0–3 µg/mL) (D) for 1 hour, and then norA and tet38 transcripts were determined by qPCR assays. The relative transcript levels of norA and tet38 were expressed as the fold change (FC) in pump gene transcripts of SA WT grown in TSB and exposed to PA WT supernatant or QS molecules versus non-exposed SA WT. The assays were repeated three times with three different biological samples. The error bars represent the means of FC ± SEM for each assay. The (*) in (C) represents the differences in the FC of tet38 of SA WT exposed to PA WT supernatant or QS molecules versus SA WT non-exposed. The (**) in (C) represents the differences in the FC of norA or tet38 of SA WT exposed to QS molecules versus SA WT exposed to PA WT supernatant. The (*) in (D) represents the differences in the FC of norA of SA WT exposed to PYO at 0.25 µg/mL versus SA WT non-exposed (0 µg/mL). The differences were statistically significant as determined by an ANOVA (P < 0.05). SA WT, S. aureus RN6390; PA WT, P. aeruginosa PA14.

Fig 4

CF-SA norA and tet38 transcript levels and roles of efflux pumps in S. aureus ability to grow in ASM medium. 16 CF-SA with wild-type norA and tet38 promoters were cultured in TSB and ASM media (initial CFU/mL ~ 10⁴) for 1 hour, then quantitative real-time RT-PCR assays (qPCR) were performed to assess the level of norA (A) and tet38 (B) transcripts. The relative transcript levels of norA and tet38 were expressed as the fold change (FC) in pump gene transcripts of bacteria grown in ASM versus grown in TSB. The assays were repeated three times with three different biological samples. The error bars represent the means of FC ±SEM for each assay. The (*) represents the differences in the FC of norA or tet38 of WT versus norA or tet38 of CF-SA. The (**) represents the differences in the FC of norA or tet38 of S. aureus in ASM versus norA or tet38 of S. aureus in TSB. The differences were statistically significant as determined by an ANOVA (P < 0.05). (A) norA transcript level; (B) tet38 transcript level; CF-SA/PA, co-isolated pairs; WT, S. aureus Newman.

Fig 5

Relative norA transcript levels and MICs of ciprofloxacin and levofloxacin of SA transformants carrying plasmid-borne norA promoter mutations. norA transcript level of SA transformants in ASM versus TSB. SA RN6390 (WT) and norA mutant (∆norA) transformed with plasmid constructs: pLI50-[norA_(WT)p-norA], pLI50-[norA_(CAAT)p-norA], pLI50-[norA_(ACAA)p-norA], and pLI50-[norA_(CTAT)p-norA] were cultured in TSB or ASM (initial CFU/m L~10⁴) supplemented with chloramphenicol (Cm = 10 µg/mL, selection pressure) for 1 hour, then qPCRs were performed to assess the norA transcript levels of transformants. (A) The relative norA transcript levels were expressed as the fold change (FC) in norA transcripts of transformants grown in ASM versus in TSB, and the fold change between transformants with plasmid constructs carrying wild-type norA promoter versus mutated norA promoter. The assays were repeated three times with three different biological samples. The error bars represent the means of FC ± SEM for each assay. The (*) represents the differences in the norA transcript levels of WT or ∆norA mutant with plasmid constructs (WT or mutated norA promoters) versus WT or ∆norA mutant with empty plasmid (pLI50) in TSB. The (**) represents the differences in the norA transcript levels of WT or ∆norA mutant with plasmid constructs (mutated norA promoters) versus WT or ∆norA mutant with plasmid construct pLI50-[norA_(WT)p-norA] in ASM. The differences were statistically significant as determined by an ANOVA (P < 0.05). MICs of ciprofloxacin and levofloxacin of SA transformants. The MICs were carried out in TSB media. (B) The (*) represents the differences between the MICs of CIP or LFX between transformants with a norA wildtype versus a mutated norA promoter. The differences were statistically significant as determined by an ANOVA (P < 0.05). WT or ∆norA (norA_(WT)p), S. aureus with construct pLI50-norA_(WT)p-norA. WT or ∆norA (norA_(CAAT)p), S. aureus with construct pLI50-norA_(CAAT)p-norA. WT or ∆norA (norA_(ACAA)p), S. aureus with construct pLI50-norA_(ACAA)p-norA. WT or ∆norA (norA_(CTAT)p), and S. aureus with construct pLI50-norA_(CTAT)p-norA.

Fig 6

Growth of SA in ASM medium. SA WT (RN6390), CF-SA with CAAT insertion (CF-25, CF-26), CF-SA with wild-type promoter (CF-2, CF-3), and CF-SA with polX insertion (CF-17, CF-18) (initial CFU/mL ~ 10⁴) were cultured in ASM medium for 24 hours, and then colony counting was performed. The growth curve assays were repeated three times with three different biological samples. The error bars represent the means of (log₁₀CFU/mL ± SD) for each assay. The differences in the log₁₀CFU/mL of CF-17 and CF-18 (polX insertion) compared to that of SA WT and other CF-SA with wildtype or CAAT insertion at 24 hours of growth were statistically significant as determined by an ANOVA (P < 0.05).