Linking Virulence and Iron Limitation Response in Staphylococcus aureus: The sRNA IsrR Is Involved in SaeRS Activation

Abstract

The Gram-positive opportunistic pathogen Staphylococcus aureus colonizes ∼30% of the human population but also causes various diseases. Precise regulation of genes involved in virulence and metabolic functions is required to adapt to changing host conditions, such as severe restriction of iron availability. In addition to the global regulator Fur (ferric uptake regulator), the iron limitation response of S. aureus is shaped by the recently identified sRNA IsrR (iron sparing response regulator). IsrR mediates an iron sparing response by inhibiting the synthesis of nonessential iron-containing proteins, which are in particular involved in the central metabolism. In addition, we demonstrate that isrR expression is positively associated with α-hemolysin levels and the hemolysis activity of S. aureus HG001. To investigate the influence of IsrR on virulence factor production, we performed a mass spectrometry-based secretome analysis of isrR-expressing and nonexpressing strains under iron-limited and iron-sufficient conditions. The SaeR regulon was positively influenced by the presence of IsrR, and IsrR is likely involved in the activation of the Sae system. Additionally, IsrR also positively affected the protein levels of the isdABCDEFGH-encoded heme uptake system (e.g., IsdB). Taken together, IsrR establishes a link between the iron limitation response and the virulence in S. aureus.

Keywords: SaeRS TCS; Staphylococcus aureus; exoproteome; heme uptake; hemolysin; iron limitation; regulatory RNA; sRNA; secretome; virulence factor.

1

Hla protein levels, hla transcript levels, and hemolysis activity are influenced by IsrR. (A) Protein levels of Hla according to Ganske et al. (2024). The bar chart depicts the amount (mean maxLFQ protein level) of Hla between isrR-expressing (HG001, pJLisrR) and nonexpressing strains (ΔisrR, pJLctrl). Error bars represent the standard deviation of the four biological replicates. Statistics: Welch-t test on protein levels (p < 0.001: ***, p < 0.01: **, and p < 0.05: *). (B) The effect of IsrR on hla mRNA abundance was examined by Northern blot analysis. For each sample, 4 μg of total RNA was loaded per lane. (C) The effect of IsrR on hemolysis activity. Strains were cultivated in TSB at 37 °C and at an OD_540nm of 1, 10 μL of culture was spotted on 5% sheep blood Columbia agar, and plates were incubated for 24 h. Hemolysis activity was determined based on the area of hemolysis around the spot of growing cells. Boxplots represent the median of three biological and two technical replicates each. Statistics: Kruskal–Wallis test on hemolysis activity and Wilcoxon test with Benjamini–Hochberg p-value adjustment as post hoc test. Results of relevant post hoc pairwise comparisons are depicted. (D) Hemolysis activity after 48 h. Strains were cultivated in TSB at 37 °C and at an OD_540nm of 1, 10 μL of culture was spotted on 5% sheep blood Columbia agar, and plates were incubated for 48 h. Hemolysis activity was determined based on the area of hemolysis around the spot of growing cells.

2

The effect of IsrR on Hla is SaeRS-dependent. (A) The effect of SaeRS and IsrR on the hemolysis activity. Strains were cultivated in TSB and at an OD_540nm of 1, and 10 μL of culture was spotted on 5% sheep blood Columbia agar. Plates were incubated for 24 h. Hemolysis activity was determined based on the area of hemolysis around the spot of growing cells. Boxplots represent the median of three biological and two technical replicates each. Statistics: Kruskal–Wallis test on hemolysis activity and Wilcoxon test with Benjamini–Hochberg p-value adjustment as post hoc test. Results of relevant post hoc pairwise comparisons are depicted. (B) The effect of IsrR on hla, chp, coa, and saePQRS mRNA abundance representing SaeR targets was examined by Northern blot analysis. For each sample, 4 μg of total RNA was loaded per lane.

3

Overview of secretome profiles. (A) Schema of the experimental set up of the secretome analysis. The HG001 wild-type, the ΔisrR mutant, and the ΔsaePQRS (Δsae) mutant were cultivated under iron-limited conditions. Strains constitutively expressing isrR (pJLisrR) and control strains with the empty vector (pJLctrl) were cultivated under iron-rich conditions. The strains marked by the box were introduced in Ganske et al. (2024). (B) PCA displaying the first and second component. The PCA was calculated based on the 636 proteins of interest for the secretome. Each strain and sampling condition was labeled individually. Replicates are displayed as points. (C) Scree plot of principal components. The percentage of global variance described by the respective component is displayed. Components describing more than 5% of variance are colored in light gray, and components describing more than 1% of variance are displayed. The 5% threshold is depicted as the dotted line. (D) Separation profiles for the principal components. Negative decadal logarithms of q-values (FDR-adjusted p-values) of Kruskal–Wallis tests for separation of each condition subcategory (growth phase, medium, Sae status, and IsrR status) for the most important five principal components are displayed. The q-value threshold for a significance of 0.05 is depicted as the dotted line. (E) GSEA regulon analysis of proteins spanning the principal components. One-sided GSEA analysis was performed on the percentage of weight of each protein into each principal component for transcription factor regulons according to AureoWiki. Negative decadal logarithms of q-values (FDR-adjusted p-values) are displayed for the five most important principal components. Regulons with a q-value less than 0.1 (dotted line) for at least one principal component are shown, and only regulons with more than five identified members were considered.

4

Heat map of protein abundances of known Sae-dependent virulence factors in the secretome. MaxLFQ values were normalized to the mean OD_540nm per condition at the harvest time of the supernatant sample and min–max scaled per protein across iron-sufficient and iron-limited conditions. The SaeR regulon was depicted according to AureoWiki. (A) Global representation of protein levels in the secretome of the known SaeR regulon. (B) Strictly SaeR-dependent regulon members with higher protein levels during the exponential growth phase are depicted for the isrR-expressing and nonexpression Sae-positive strains. (C) Strictly SaeR-dependent regulon members with higher protein levels during the stationary growth phase are depicted for the isrR-expressing and nonexpression Sae-positive strains.

5

Overview of IsrR-driven effects on the secretome of . (A) Identification of IsrR targets based on the changes of protein abundances in the secretome. For each of the comparisons, candidates were considered if they were significantly altered (|fold change| > 1.5 and q-value <0.05) in at least one growth phase and if they were not discordantly altered between the two growth phases. Candidates identified in at least two comparisons were considered as secretome target candidates. (B) Integration of the 86 identified secretome target candidates with in silico IsrR target prediction. Predictions according to IntaRNA2.0 specifically for NCTC 8325 (Table S7) and Ganske et al. (2024) were considered. (C) Protein levels of Lpl9 in the stationary growth phase as an example of a protein pattern representing negative IsrR regulation. The bar chart depicts the amount (mean maxLFQ protein level) of Lpl9 between IsrR-expressing (HG001, pJLisrR, and Δsae pJLisrR) and nonexpressing (ΔisrR, pJLctrl, and Δsae pJLctrl) strains. Error bars represent the standard deviation of the biological replicates. Statistics: Welch-t test on protein levels (p < 0.001: ***, p < 0.01: **, and p < 0.05: *).

6

IsrR positively influences the protein levels of the heme uptake system and impacts the siderophore-mediated iron uptake of . (A) Heat map of Fur regulon proteins identified in the secretome. MaxLFQ values were normalized to the mean OD_540nm per condition at a harvest time point of the supernatant sample and the min–max scaled per protein. (B) CAS-Agar diffusion assay to determine siderophore activity. Strains were cultivated in TSB or TSBDP, and the supernatant was harvested in stationary growth phase. Sterile filtered supernatant was applied to CAS agar plates. Siderophore activity was determined based on the area of orange iron–poor complex formation around the spot. Siderophore activity was subsequently normalized to OD_540nm. Boxplots represent the median of four biological replicates. Statistics: Kruskal–Wallis test on siderophore activity and Wilcoxon test with Benjamini–Hochberg p-value adjustment as post hoc test. Results of relevant post hoc pairwise comparisons are depicted.

7

Working model of IsrR linking SaeR-regulated virulence and the iron limitation response in S. aureus.