Iron availability and oxygen tension regulate the Yersinia Ysc type III secretion system to enable disseminated infection

Abstract

The enteropathogen Yersinia pseudotuberculosis and the related plague agent Y. pestis require the Ysc type III secretion system (T3SS) to subvert phagocyte defense mechanisms and cause disease. Yet type III secretion (T3S) in Yersinia induces growth arrest and innate immune recognition, necessitating tight regulation of the T3SS. Here we show that Y. pseudotuberculosis T3SS expression is kept low under anaerobic, iron-rich conditions, such as those found in the intestinal lumen where the Yersinia T3SS is not required for growth. In contrast, the Yersinia T3SS is expressed under aerobic or anaerobic, iron-poor conditions, such as those encountered by Yersinia once they cross the epithelial barrier and encounter phagocytic cells. We further show that the [2Fe-2S] containing transcription factor, IscR, mediates this oxygen and iron regulation of the T3SS by controlling transcription of the T3SS master regulator LcrF. IscR binds directly to the lcrF promoter and, importantly, a mutation that prevents this binding leads to decreased disseminated infection of Y. pseudotuberculosis but does not perturb intestinal colonization. Similar to E. coli, Y. pseudotuberculosis uses the Fe-S cluster occupancy of IscR as a readout of oxygen and iron conditions that impact cellular Fe-S cluster homeostasis. We propose that Y. pseudotuberculosis has coopted this system to sense entry into deeper tissues and induce T3S where it is required for virulence. The IscR binding site in the lcrF promoter is completely conserved between Y. pseudotuberculosis and Y. pestis. Deletion of iscR in Y. pestis leads to drastic disruption of T3S, suggesting that IscR control of the T3SS evolved before Y. pestis split from Y. pseudotuberculosis.

Fig 1. Mutation of the identified IscR binding site in the lcrF promoter disrupts apo-IscR binding and type III secretion.

(A) Nucleotide sequences of the IscR binding site in the WT strain (top) and the lcrF^pNull mutant (bottom). Bases previously shown to be important for IscR binding are in bold [25]. Mutated residues in the lcrF^pNull mutant are underlined. (B) Electrophoretic mobility shift assays (EMSAs) were performed on DNA fragments (217 bp) containing the WT and lcrF^pNull mutant promoter sequences upstream of the yscW-lcrF operon. Concentrations of E. coli apo-IscR, afforded by the IscR-C92A mutant protein, are denoted above the gel lanes. One representative experiment out of three is shown. (C) Y. pseudotuberculosis was grown in low calcium 2xYT media containing iron and the T3SS was induced at 37°C for 2 hours. T3SS cargo proteins secreted into culture supernatant and precipitated with TCA were visualized with Coomassie blue. One representative experiment out of three is shown.

Fig 2. The Yersinia Ysc type III secretion system is induced by oxygen.

(A) Y. pseudotuberculosis was grown under anaerobic, iron-replete conditions at 26°C and optical density at 600 nm (OD_600nm) measured every hour. Y. pseudotuberculosis pYV^- lacks the pYV virulence plasmid and serves as a T3SS-deficient control strain. Data shown represent an average of three biological replicates. (B-C) Y. pseudotuberculosis WT was grown in M9/high glucose (iron-replete) in the presence or absence of oxygen and type III secretion induced by shifting to 37°C for 4 hrs. (B) RNA was extracted from the bacterial cultures and iscR and lcrF gene expression analyzed using qPCR. Relative mRNA levels for each gene of interest were normalized to 16S rRNA levels. Data shown represent the average of three independent experiments. **** p ≤ 0.0001, as determined by unpaired t-test. (C) Equal amounts of cell lysates were probed with antibodies for RpoA, IscR, YopE, and LcrF. T3SS cargo proteins secreted into culture supernatant and precipitated with TCA were probed with YopE antibody. All samples shown are from the same experiment, with each lysate split into two gels, one subsequently used for IscR and YopE probing and the other for LcrF probing. RpoA was blotted on each membrane as a loading control. One representative experiment out of three is shown.

Fig 3. Iron limitation does not affect IscR, LcrF, YopH, or YopE expression in the presence of oxygen.

M9/high glucose iron-limited aerobic cultures of Y. pseudotuberculosis were subsequently grown in iron-replete (+Fe) or iron-limited (-Fe) aerobic conditions and type III secretion induced by shifting to 37°C for 4 hrs. (A) RNA was isolated and qPCR used to determine relative expression of iscR, lcrF, and yopE by normalizing to 16S rRNA levels. Data shown represent the average of four independent experiments. *p ≤ 0.05, ***p ≤ 0.001, ****p≤0.0001, as determined by one-way ANOVA with Tukey post-test. (B) IscR and RpoA protein levels were determined by Western blot analysis. One representative experiment out of four is shown. (C) T3SS cargo proteins secreted into culture supernatant and precipitated with TCA were probed with antibodies for YopE and YopD. Cell lysates were probed with antibodies for RpoA, YopE, YopD, and LcrF. The ΔyscNU negative control strain does not express a functional T3SS. All samples shown are from the same experiment, with each lysate split into two gels, one subsequently used for LcrF probing and the other for YopE probing. RpoA was used as a loading control. One representative experiment out of four is shown. (D) Y. pseudotuberculosis PyopH-mCherry was used to assess yopH expression under M9/high glucose aerobic iron-replete (+Fe) or iron-limited (-Fe) conditions at 37°C, in the genetic backgrounds indicated. mCherry fluorescence intensity normalized to OD₆₀₀ is shown, with T = 0 being the start of incubation at 37°C. Data shown represent the average of three independent experiments. Densitometric quantification of the bands was performed using Image Lab software (Bio-Rad), setting the WT (+Fe) band to 1.00.

Fig 4. Iron limitation induces iscR, lcrF, and yopE expression and type III secretion under fermentative conditions.

Iron-limited Y. pseudotuberculosis was shifted to iron-replete (+Fe) or iron-limited (-Fe) M9/high glucose media in the absence of oxygen and type III secretion induced at 37°C for 4 hrs. (A) RNA was isolated and qPCR used to determine relative expression of iscR, lcrF, and yopE by normalizing to 16S rRNA levels. Data shown represent the average of four independent experiments. *p ≤ 0.05, **p ≤ 0.005, ****p≤0.0001****p≤0.0001, as determined by one-Way ANOVA with Tukey post-test. (B) T3SS cargo proteins secreted into culture supernatant and precipitated with TCA were probed with antibodies for YopE and YopD. Cell lysates were probed with antibodies for RpoA, YopE, and LcrF. All samples shown are from the same experiment, with each lysate split into two gels, one subsequently used for LcrF probing and the other for YopE probing. RpoA was used as a loading control. One representative experiment out of four is shown. Densitometric quantification of the bands was performed using Image Lab software (Bio-Rad), setting the WT (+Fe) band to 1.00.

Fig 5. Iron limitation induces iscR, lcrF, and yopE expression under anaerobic respiration conditions.

Iron-limited Y. pseudotuberculosis was shifted to iron-replete (+Fe) or iron-limited (-Fe) M9 media supplemented with sodium nitrate and mannitol in the absence of oxygen. Type III secretion was induced at 37°C for 4 hrs. T3SS cargo proteins secreted into culture supernatant and precipitated with TCA were probed with antibodies for YopE. Cell lysates were probed with antibodies for RpoA, YopE, and LcrF. All samples shown are from the same experiment, with each lysate split into two gels, one subsequently used for LcrF and YopE probing and the other for IscR probing. RpoA was used as a loading control. One representative experiment out of three is shown. Dotted lines delimit the area of each well. Densitometric quantification of the bands was performed using Image Lab software (Bio-Rad), setting the WT (+Fe) band to 1.00.

Fig 6. Ectopic expression of IscR is sufficient to rescue T3SS expression under anaerobic, iron-replete conditions.

Y. pseudotuberculosis was grown in iron-replete M9/high glucose and in the absence of oxygen and type III secretion induced at 37°C for 4 hrs, as shown in Fig 2. T3SS cargo proteins secreted into culture supernatant and precipitated with TCA were probed with an antibody for YopE. Cell lysates were probed with antibodies for RpoA, IscR, YopE, and LcrF. All samples shown are from the same experiment, with each lysate split into two gels, one subsequently used for LcrF and YopE probing and the other for IscR probing. RpoA was used as a loading control. One representative experiment out of three is shown. Dotted lines delimit the area of each well. Densitometric quantification of the bands was performed using Image Lab software (Bio-Rad), setting the WT band to 1.00.

Fig 7. Proper regulation of lcrF by IscR is necessary for disseminated infection.

Mice were infected with ~2x10⁸ Y. pseudotuberculosis using a bread feeding model, and organs and intestinal contents harvested 5 days post-inoculation. CFUs were normalized to the weight of the organ in grams. Each symbol represents one animal. Unfilled symbols are set at the limit of detection for each individual organ based on weight, and represent CFU that were below this limit. Graphs show data combined from five independent experiments. **p<0.01, *** p<0.001, ****p <0.0001, as determined by an unpaired Mann-Whitney rank sum test. Dashes represent geometric means.

Fig 8. IscR is critical for type III secretion in Y. pestis.

WT and ΔiscR Y. pseudotuberculosis (pstb) and Y. pestis pgm- were grown in BHI low calcium or high calcium media at 37°C. Proteins in the bacterial culture supernatant were precipitated with TCA and type III secretion probed using an anti-YopE antibody. Results are representative of three independent experiments. * YopE specific band. Doublet bands for YopE have been previously reported [58].