The Role and Targets of the RNA-Binding Protein ProQ in the Gram-Negative Bacterial Pathogen Pasteurella multocida

Abstract

The Gram-negative pathogen Pasteurella multocida is the causative agent of many important animal diseases. While a number of P. multocida virulence factors have been identified, very little is known about how gene expression and protein production is regulated in this organism. One mechanism by which bacteria regulate transcript abundance and protein production is riboregulation, which involves the interaction of a small RNA (sRNA) with a target mRNA to alter transcript stability and/or translational efficiency. This interaction often requires stabilization by an RNA-binding protein such as ProQ or Hfq. In Escherichia coli and a small number of other species, ProQ has been shown to play a critical role in stabilizing sRNA-mRNA interactions and preferentially binds to the 3' stem-loop regions of the mRNA transcripts, characteristic of intrinsic transcriptional terminators. The aim of this study was to determine the role of ProQ in regulating P. multocida transcript abundance and identify the RNA targets to which it binds. We assessed differentially expressed transcripts in a proQ mutant and identified sites of direct ProQ-RNA interaction using in vivo UV-cross-linking and analysis of cDNA (CRAC). These analyses demonstrated that ProQ binds to, and stabilizes, ProQ-dependent sRNAs and transfer RNAs in P. multocida via adenosine-enriched, highly structured sequences. The binding of ProQ to two RNA molecules was characterized, and these analyses showed that ProQ bound within the coding sequence of the transcript PmVP161_1121, encoding an uncharacterized protein, and within the 3' region of the putative sRNA Prrc13. IMPORTANCE Regulation in P. multocida involving the RNA-binding protein Hfq is required for hyaluronic acid capsule production and virulence. This study further expands our understanding of riboregulation by examining the role of a second RNA-binding protein, ProQ, in transcript regulation and abundance in P. multocida.

Keywords: Pasteurella multocida; ProQ; RNA-binding proteins; sRNA.

FIG 1

Genomic location of proQ and growth of the wild-type P. multocida VP161, proQ mutant, and complemented strains in different growth media. (A) Schematic representation of the location of proQ in the genome and the positions of the TargeTron intron (orange box), prc transcriptional start site, and stem-loop sequences. (B and C) Growth curves of WT[EV] (circles), proQ[EV] (squares), and proQ[proQ] (triangles) over 24 h with shaking at 37°C when incubated in heart infusion (HI) broth (B) or HI with 300 mM NaCl (C). (D) Growth curves of WT[EV] (squares), proQ[EV] (circles), and proQ[proQ] (triangles) over 24 h, with shaking at 37°C, when grown in peptone water containing 1% glucose, either untreated (0 μM) (black) or iron-depleted using pretreatment with dipyridyl (150 μM) (gray). Data shown are means ± standard deviations (SD) (n = 3).

FIG 2

Transcriptomic analysis following perturbation of proQ expression in P. multocida strain VP161. (A) Volcano plots showing the log₂ fold change in gene expression and P value for each transcript. Each plot shows those transcripts that were significantly differentially expressed in a single comparison (black dots), all three comparisons (pink dot), proQ[EV] versus WT[EV] and proQ[proQ] versus WT[EV] (blue dots), proQ[EV] versus WT[EV] and proQ[proQ] versus proQ[EV] (green dots), and proQ[proQ] versus WT[EV] and proQ[proQ] versus proQ[EV] (orange dots). (B) Venn diagram showing the number of differentially expressed genes in each of the strains and those common to one or more strains. Genes with increased expression are represented by the up arrow and those with decreased expression by the down arrow. Blue-shaded circle labeled proQ[EV] versus WT[EV] shows differential expression of genes in the proQ mutant containing empty vector (AL3358) compared to expression in wild-type VP161 containing empty vector (AL3356). Red-shaded circle labeled proQ[proQ] versus WT[EV] represents the number of differentially expressed genes in the complemented proQ mutant (AL3357) compared to expression in the wild-type strain containing empty vector (AL3356). Green-shaded circle represents the number of differentially expressed genes in proQ[proQ] strain AL3357 compared to expression in the proQ[EV] strain AL3358. *, fourteen genes showed decreased expression in the proQ mutant but increased expression when proQ was expressed on a plasmid; ^, one gene displayed decreased expression in the proQ mutant and increased expression in the complemented proQ mutant. (C) Differentially expressed genes identified by RNA-seq transcriptomic analysis (see Table S1 in the supplemental material), grouped according to cellular function. The number of differentially expressed genes belonging to each function is shown relative to the following analysis groups: proQ[EV] versus WT[EV] (AL3358 versus AL3356, blue bars), proQ[proQ] versus WT[EV] (AL3357 versus AL3356, red bars), and proQ[proQ] versus proQ[EV] (AL3357 versus AL3358, green bars).

FIG 3

Analysis of transcripts that bind directly to ProQ, as determined by CRAC analysis. (A) Distribution of ProQ-bound RNA species: sRNA, small RNA; CDS, protein coding sequence; 3UTR, 3′ untranslated region of CDS; 5UTR, 5′ untranslated region of CDS. (B) Cumulative distribution functions [Fn(x)] for differentially expressed transcripts, defined as ≥1 log₂ change in expression with a false discovery rate of <0.05 (log₂ FC), in proQ mutant (AL3358) compared to the WT (AL3356) (B) and the proQ mutant (AL3358) compared to the complemented proQ mutant (AL3357) (C). Blue lines indicate the Fn(x) for transcripts that bound to ProQ as determined by CRAC analysis, and red lines indicate the Fn(x) for transcripts that did not bind to ProQ. (D) ProQ binding site frequency as a function of position on the transcripts relative to the beginning of the predicted terminator stem-loop sequence. (E) ProQ binding sequence (top), pairing (middle), and structure (bottom) motifs (S, stem; E, external bulge; B, bulge loop; H, hairpin loop; and I, internal loop) as determined from transcripts bound at the beginning of the terminator stem-loop as identified using GraphProt.

FIG 4

Analysis of ProQ interaction with mRNA PmVP161_1121. (A) RNA-seq read coverage over the PmVP161_1121 transcript as determined by CRAC analysis of RNA samples isolated from strains expressing either tagged-ProQ (+) or untagged-ProQ (−). CRAC experiments were performed in biological triplicate (R1 to R3). Read counts are shown on a scale of 0 to 3,000 reads for all plots. (B) The secondary structure of the PmVP161_1121 transcript as predicted using RNAfold and indicating the residues interacting with ProQ region (red circles). (C) Northern blot detection (in biological triplicate) measuring the abundance of the PmVP161_1121 transcript (∼230 bp) in RNA isolated from the wild-type P. multocida VP161 containing empty vector (AL3356), proQ mutant containing empty vector (AL3358), and the complemented proQ mutant (AL3357). (D) Densitometry analysis of 230-nt PmVP161_1121 transcript detected in the northern blot using total RNA production and determined by SYBR-stained rRNA for standardization. Data shown are means ± SD; n = 3. **, P < 0.005; ***, P < 0.0005 by Mann-Whitney test.

FIG 5

Analysis of ProQ interaction with the predicted sRNA Prrc13. (A) RNA-seq read coverage over the Prrc13 transcript following CRAC analysis of RNA samples isolated from strains expressing either tagged ProQ (+) or untagged ProQ (−). CRAC experiments were performed in biological triplicate (R1 to R3). Read counts are shown on a scale of 0 to 3,000 reads for all plots. (B) The secondary structure of the 3′ region of Prrc13 as predicted using RNAfold. Residues binding to ProQ are indicated (red circles). (C) Northern blot detection of Prrc13 transcript. RNA was isolated from the wild-type P. multocida VP161 containing empty vector (AL3356), the proQ mutant containing empty vector (AL3358), and the complemented proQ mutant (AL3357). (D) Densitometry performed on transcript identified in Northern blot. Prrc13 expression is normalized relative to the average RNA production. Data shown are mean ± SD; n = 5 to 6. **, P < 0.005 by Mann-Whitney test.