Characterization of a ferrous iron-responsive two-component system in nontypeable Haemophilus influenzae

Abstract

Nontypeable Haemophilus influenzae (NTHI), an opportunistic pathogen that is commonly found in the human upper respiratory tract, has only four identified two-component signal transduction systems. One of these, an ortholog to the QseBC (quorum-sensing Escherichia coli) system, was characterized. This system, designated firRS, was found to be transcribed in an operon with a gene encoding a small, predicted periplasmic protein with an unknown function, ygiW. The ygiW-firRS operon exhibited a unique feature with an attenuator present between ygiW and firR that caused the ygiW transcript level to be 6-fold higher than the ygiW-firRS transcript level. FirRS induced expression of ygiW and firR, demonstrating that FirR is an autoactivator. Unlike the QseBC system of E. coli, FirRS does not respond to epinephrine or norepinephrine. FirRS signal transduction was stimulated when NTHI cultures were exposed to ferrous iron or zinc but was unresponsive to ferric iron. Notably, the ferrous iron-responsive activation only occurred when a putative iron-binding site in FirS and the key phosphorylation aspartate in FirR were intact. FirRS was also activated when cultures were exposed to cold shock. Mutants in ygiW, firR, and firS were attenuated during pulmonary infection, but not otitis media. These data demonstrate that the H. influenzae strain 2019 FirRS is a two-component regulatory system that senses ferrous iron and autoregulates its own operon.

Fig 1

(A) Schematic diagram of the ygiW-firRS operon. The two primary transcriptional start sites are upstream of ygiW and indicated by arrows. The attenuator present between ygiW and firR is represented by a loop. The attenuator is located 14 bp after the ygiW stop codon. (B) Northern blot analysis using internal probes for ygiW and firR. The ygiW probe hybridized with transcripts at 0.5 kb and 2.5 kb, with the 0.5-kb band in much greater abundance. The firR probe only hybridized to the 2.5-kb band. (C) The attenuator stem-loop that is present between ygiW and firR. The line shows the residues that are deleted in JWJ166.

Fig 2

(A) Primer extension mapping of the ygiW-firRS transcriptional start sites. Primer extension analysis identified three transcriptional start sites in wild-type 2019, whereas only TS-2 and TS-3 are present in the NTHI firRS mutant. The inset shows a shorter exposure in which the doublet bands at TS-1 are more apparent. (B) Sequence upstream of the start codon of ygiW illustrating the transcriptional start sites (TS-1, TS-2, and TS-3) and potential −10, −35 (boxed), and +1 sites (bold letter with arrow).

Fig 3

Thermoresponsive induction of ygiW (A) and firR (B). Cultures of wild-type NTHI 2019, the ΔfirR mutant (NB004), the ΔfirS mutant (KK009), the complemented firR mutant (firR⁺; KHS2), and the complemented firS mutant (firS⁺; KHS5) were grown in sRPMI at 37°C to early log phase and shifted to 9°C for 30 min prior to RNA extraction. Expression of each gene was measured by qRT-PCR and compared to the expression of each gene when cultures were incubated at 37°C. The data presented are means and standard deviations from two experiments, each performed in triplicate.

Fig 4

Induction of ygiW-firRS promoter activity in response to ferrous iron. Wild-type 2019 containing a GFP reporter fused to the ygiW promoter (pJJ392) was exposed to various conditions, and GFP fluorescence was measured to determine induction relative to background fluorescence (empty vector control). (A) Various cations were added to cultures at a 100 μM concentration, and GFP fluorescence (RFU, relative fluorescence units) was determined. (B) FeCl₂, FeCl₃, or FeCl₃ and 5 mM sodium ascorbate (Asc) were added to cultures, and GFP fluorescence was determined. (C) GFP fluorescence in wild-type 2019 (black bars), ΔfirR (gray bars; NB004), ΔfirS (white bars; KK009), complemented firR (striped gray bars; KHS2), and complemented firS (striped white bars; KHS5) containing the reporter (pJJ392) in the presence or absence of FeCl₂. GFP fluorescence was measured relative to background fluorescence (empty vector control) to determine RFU. The data presented are means and standard deviations for triplicate determinations in a single experiment. The data presented here are representative of multiple (≥3) experiments performed, from which equivalent results and statistical trends were obtained. Statistical significance (P ≤ 0.05), as determined by the Student two-tailed t test for the comparison of cultures with nothing added versus the other cultures with a compound added, is represented by an asterisk.

Fig 5

Characterization of the FirS iron-binding motif. The induction of ygiW (A) and firR (B) in response to Fe²⁺ was measured by qRT-PCR. The ability of wild-type 2019, the ΔfirS mutant KK009, KHS1 (firS Y149G, R150T), KHS3 (firS D148A), KHS4 (firS E151G, D152S), and KHS5 (firS⁺) to respond to Fe²⁺ was tested. Expression of each gene was measured by qRT-PCR and compared to the expression of each gene in cultures grown without the addition of exogenous FeCl₂. The data presented are means and standard deviations from two experiments, each performed in triplicate.

Fig 6

Characterization of firR mutants in Fe²⁺-responsive induction of ygiW. The expression of ygiW in wild-type 2019, NB004 (ΔfirR), JWJ154 (firR D51A), and KHS2 (firR⁺) was measured by qRT-PCR. Expression of each gene was measured by qRT-PCR and compared to the expression of each gene in cultures grown without the addition of exogenous FeCl₂. The data presented are means and standard deviations from two experiments, each performed in triplicate.

Fig 7

Survival of NTHI in a mouse pulmonary infection model. Mice were infected intranasally with wild-type 2019 (circles), JWJ142 (ΔygiW; squares), NB004 (ΔfirR; upward triangles), and KK009 (ΔfirS; downward triangles), and the log₁₀ CFU in the lungs was determined at 24 and 48 h postinfection. The data presented here are representative of one experiment. Asterisks denote strains with a statistically significant difference with P values of 0.0317 or less compared to the wild type using the Mann-Whitney test.