Environmental pH controls antimicrobial production by human probiotic Streptococcus salivarius

Abstract

Streptococcus salivarius K12 (SAL) is an oral probiotic used to treat or prevent oral infections caused by human pathogens. SAL produces at least three antimicrobials to exert its antimicrobial activity, namely, salivaricin A and salivaricin B, and the newly identified salivabactin. Salivabactin production is catalyzed by a polyketide/non-ribosomal peptide synthase hybrid biosynthetic gene cluster (BGC), termed as sar-BGC. The sar-BGC expression and salivabactin production are transient during SAL growth in vitro and in vivo, which may negatively impact SAL probiotic efficacy. To understand the molecular basis for transient sar-BGC expression, we assessed the impact of environmental pH on sar-BGC expression. We found that environmental acidification is a critical factor in promoting salivabactin antimicrobial activity and production by inducing sar-BGC expression. We further showed that acidic pH directly influences the quorum-sensing system that controls sar-BGC expression. During environmental acidification, SAL cytosol is acidified, which is sensed by a pH-sensitive histidine switch in the cytosolic transcription regulator, NrpR. The protonation of histidine during cytosolic acidification promotes high-affinity interactions between NrpR and its cognate intercellular signaling peptide, NIP, which leads to upregulation of sar-BGC expression. Collectively, our results indicate that SAL uses a sophisticated regulatory mechanism to orchestrate salivabactin production in an environment that is conducive to its antimicrobial activity.

Importance: Probiotic bacteria are important tools in combating bacterial infections. Probiotics exert their antimicrobial activity via several mechanisms, including antimicrobial production. However, discrepancies exist between the in vitro and in vivo efficacies of probiotics in inhibiting pathogen growth. Understanding the host and environmental factors that influence antimicrobial production and activity is critical for improving probiotic efficacy. In this study, we showed that the antimicrobial salivabactin produced by human oral probiotic Streptococcus salivarius K12 is active at acidic pH. We further elucidated the molecular mechanism by which S. salivarius coordinates salivabactin production in concert with environmental acidification, thereby maximizing salivabactin antimicrobial activity.

Keywords: antimicrobial activity; antimicrobial production; gene regulation; pH sensing; probiotic.

Fig 1

Acidic pH is optimal for salivabactin antimicrobial activity. (A) Genetic organization of the sar biosynthetic gene cluster (sar-BGC). The 14-gene operon encoding sarC-P genes is shown as arrows. The sar-BGC expression is controlled by a divergently transcribed leaderless communication peptide, NIP, and its cognate cytosolic receptor NrpR. (B) GAS (~10⁵ CFU/mL) was inoculated in THY broth adjusted to indicated pH. The GAS growth was supplemented with either carrier (DMSO, untreated) or 0.5× MIC of salivabactin, and inhibition of GAS growth was assessed by enumerating GAS colony-forming units (CFU/mL) after 16 h incubation. P values assessed by t-test are shown (n.s, not significant; **, P < 0.005; ****, P < 0.0001). ND indicates the limit of detection that was set at 10.

Fig 2

SAL upregulates sar-BGC expression in concert with environmental acidification. (A) Wild-type (WT) SAL was grown in THY broth, and samples were collected at the indicated time points. Growth medium pH, sar-BGC (sarD) transcript levels, and bacterial growth by absorption at wavelength 600 nm (A₆₀₀) were determined. Right y-axes represent fold change in sar-BGC transcript levels (green) and A₆₀₀ (red), whereas the left y-axis represents pH of the growth medium (black). Fold changes in transcript levels at indicated time points relative to starting culture (time point t = 0 h) are shown. Data are mean ± standard deviation for three biological replicates. (B) WT SAL grown in THY to late-exponential growth phase (LE, A₆₀₀ ~ 2.0) was harvested by centrifugation, suspended in THY broth adjusted to indicated pH, and incubated for 15 minutes. The fold change in sarD transcript levels relative to WT-LE growth is shown. Data are mean ± standard deviation for three biological replicates. P values assessed by t-test analyses relative to WT LE growth are shown (n.s, not significant; *, P < 0.05; ****, P < 0.0001). (C) WT SAL grown in THY to late-exponential growth phase (LE, A₆₀₀ ~ 2.0) was harvested by centrifugation, suspended in fresh THY broth supplemented with indicated final concentrations of sodium lactate, and incubated for 15 minutes. Fold change differences in sarD transcript levels relative to WT-LE growth are shown. Data are mean ± standard deviation for three biological replicates.

Fig 3

Environmental acidification activates the NIP signaling pathway by influencing NIP-NrpR interactions. (A) The nip* mutant grown in THY to late-exponential growth phase (LE ~ 2.0) was harvested by centrifugation, suspended in fresh THY adjusted to indicated pH, supplemented with 100 nM synthetic NIP, and incubated for 30 minutes. The sarD transcript levels relative to unsupplemented nip* mutant are normalized to 1 and used as a reference to determine the fold changes in sarD levels in synthetic NIP-supplemented samples. P values assessed by Kruskal-Wallis analyses relative to unsupplemented nip* mutant are shown (n.s, not significant; *, P < 0.05; ****, P < 0.0001). (B) Environmental acidification does not affect the reimport of exogenously added synthetic FITC-NIP into the cytosol. The nip* mutant grown to late-exponential phase (A₆₀₀ ~ 2.0) was harvested by centrifugation and resuspended in pH-adjusted THY broth. Cells were supplemented with either unlabeled NIP (NIP) or FITC-labeled NIP. Cells supplemented with the carrier for the synthetic peptide (DMSO) were used as a control. After 30 minutes of incubation, cells were harvested, washed, and resuspended in PBS, and cytosolic fluorescence in the clarified cell lysates was assessed using excitation and emission wavelengths of 480 and 520 nm, respectively. Cells incubated with FITC-NIP but not lysed (TC, total cells) were used to determine the fluorescence associated with cell surface attached FITC-NIP. Fluorescence in unlysed total cells (TC) The changes in relative fluorescence units (RFU) relative to the unsupplemented nip* mutant are shown. (C) The NrpR-NIP binding constants in binding buffer adjusted to indicated pH. (D) The calculated SAL cytosolic pH values in the tested extracellular pH were determined by the equation derived from the calibration curve.

Fig 4

Crystal structure of NrpR. (A) The crystal structure of an NrpR subunit. The DNA-binding domain (A1-A4) at the amino-terminal domain is colored purple and labeled, and the linker α-helix (A5) connecting the DNA-binding domain and C-terminal domain is shown in green. Each tetratricopeptide repeat motif (TPR1–TPR5; A6-A16) in the C-terminal domain is color-coded and labeled. The N- and C-terminus of NrpR subunit is labeled N and C, respectively. (B) The dimeric structure of NrpR. Each subunit is colored blue and orange, and the DNA-binding and putative peptide-binding TPR domains in one subunit are labeled. The dimerization interface between the two subunits is highlighted by dashed lines. The N- and C-terminus of one subunit is marked as N and C, respectively. (C) The oligomerization state of NrpR as assessed by size exclusion chromatography. A table showing the theoretical molecular weight of various oligomeric states of NrpR and the experimentally calculated molecular weight of NrpR based on the elution profile in a Superdex 200 size exclusion chromatography is shown. (D) Surface representation of the crystal structure of NrpR dimer. One subunit of NrpR dimer is shown as ribbons, and the second subunit is depicted in surface representation. The concave surface in the C-terminal TPR domain containing the putative peptide-binding pocket of NrpR is marked in dashed lines.

Fig 5

Histidine 144 (H144) is critical for pH sensing by NrpR. (A) The location of histidines in the NrpR dimer is shown as red spheres and labeled. The DNA-binding domain is marked and labeled, whereas the putative peptide-binding pocket of NrpR is highlighted in dashed lines. The N- and C-terminus of one subunit is marked as N and C, respectively. (B) Analysis of the sarD transcript levels in nrpR mutant strains containing single alanine substitution at histidines. The sarD expression in the ∆nrpR mutant was used as a reference, and fold changes in sarD transcript levels relative to the reference are shown. Data are mean ± standard deviation for three biological replicates. P values (**P < 0.01 and ****P < 0.0001; n.s, not significant) as assessed by Kruskal-Wallis analyses are shown. (C) NrpR–synthetic NIP-binding constants for the indicated recombinant NrpR mutant proteins as assessed by FP assay are shown.