Identification of small RNAs associated with RNA chaperone Hfq reveals a new stress response regulator in Actinobacillus pleuropneumoniae

Abstract

The RNA chaperone Hfq promotes the association of small RNAs (sRNAs) with cognate mRNAs, controlling the expression of bacterial phenotype. Actinobacillus pleuropneumoniae hfq mutants strains are attenuated for virulence in pigs, impaired in the ability to form biofilms, and more susceptible to stress, but knowledge of the extent of sRNA involvement is limited. Here, using A. pleuropneumoniae strain MIDG2331 (serovar 8), 14 sRNAs were identified by co-immunoprecipitation with Hfq and the expression of eight, identified as trans-acting sRNAs, were confirmed by Northern blotting. We focused on one of these sRNAs, named Rna01, containing a putative promoter for RpoE (stress regulon) recognition. Knockout mutants of rna01 and a double knockout mutant of rna01 and hfq, both had decreased biofilm formation and hemolytic activity, attenuation for virulence in Galleria mellonella, altered stress susceptibility, and an altered outer membrane protein profile. Rna01 affected extracellular vesicle production, size and toxicity in G. mellonella. qRT-PCR analysis of rna01 and putative cognate mRNA targets indicated that Rna01 is associated with the extracytoplasmic stress response. This work increases our understanding of the multilayered and complex nature of the influence of Hfq-dependent sRNAs on the physiology and virulence of A. pleuropneumoniae.

Keywords: Galleria mellonella; Pasteurellaceae; extracellular vesicles; porcine pleuropneumonia; trans-acting small RNA.

Figure 1

Identification of novel sRNA candidates associated with the chaperone Hfq. (A) Mapping regions of the RNA sequences from the WT, Ap8hfq::3XFLAG, and Ap8Δhfq strains showing increased intergenic reads in aerobiosis (1) and anaerobiosis (2), corresponding to the 14 sRNA candidates selected in this study. Genomic coordinates of the sRNAs are represented above (rna01 to rna07) or below (rna08 to rna14) of the open reading frames. (B) Heatmap showing the coverage, in reads per kilobase of transcript, per million mapped reads (RPKM), of the 14 putative novel sRNAs identified by co-IP.

Figure 2

Validation of the expression of Hfq-associated sRNAs identified in this work. (A) Expression of sRNAs identified by co-IP using Hfq. Expression was assessed in Actinobacillus pleuropneumoniae Ap8WT (1), Ap8hfq::3XFLAG (2), and Ap8∆hfq (3) strains growing under aerobic and anaerobic conditions until late exponential phase. All membranes were reprobed using the rRNA 5S, expressed constitutively. The red arrow represents the band that corresponds to the predicted size for the respective sRNA. The experiment was performed in triplicate. (B) Genic organization and secondary structure of confirmed sRNA candidates, predicted by RNAfold (Lorenz et al., 2011). Free energy (ΔG) of sRNA structures is showed next to each structure.

Figure 3

Characterization of the new sRNA Rna01. (A) Coding sequence of rna01. The putative promoter and Rho-independent terminator are shown. The putative transcriptional start site (TSS) is represented by the box with an arrow, and the −35 and −10 regions for each putative promoter sequences are underlined. The Rna01 sequence is highlighted in bold. (B) Secondary structure of Rna01 predicted using Rnafold software. The putative stem loops and putative Hfq binding site are shown.

Figure 4

Actinobacillus pleuropneumoniae mRNA Targets predicted to be bound by Rna01. (A) Targets of Rna01 manually predicted in the MIDG2331 genome and their interaction with the seed region 1 of Rna01. Predicted translational start sites of Rna01 targets is represented by “+1.” (B) Interaction of manually predicted targets of Rna01. Grey lines represent protein–protein associations, and their thickness indicates the strength of data support. Associations are meant to be specific and meaningful, for example, proteins that jointly contribute to a shared function. Genes were analyzed with a moderate confidence 0.400 and Markov clustering method (MCL) with inflation parameter 1.1. Associated proteins share colors in the network.

Figure 5

rna01 and ompP2B expression and their effect in the outer membrane protein (OMP) contents. qPCR of rna01. (A) ompP2B (B) during exponential and stationary phases in the WT and mutant strains. (C) OMPs profile of the strains in exponential and stationary phases. *Significative difference by the t-test (p < 0.1). The experiment was performed in triplicate.

Figure 6

Phenotypic characterization of Actinobacillus pleuropneumoniae rna01 mutant strains. (A) Bacterial growth in BHI-NAD at 37°C. Maximum growth rates (μ) are shown in the bottom-right corner. (B) Biofilm formation in polystyrene microtiter plates. The reference strain 5b L20 was used as positive control for biofilm formation. (C) Hemolytic activity on sheep blood agar. (D) Stress tolerance in KCl (0.1 M) and H₂O₂ (0.2 mM). Means with different letters are significantly different by the Tuckey’s test. The experiment was performed in triplicate.

Figure 7

Effects of the deletion of the rna01 gene on Actinobacillus pleuropneumoniae virulence. (A) Killing assay of Galleria mellonella; (B) Visual observation of larval melanization through the course of the experiment (only living larvae are shown, since dead larvae were black and dehydrated). (C) Optical density of larval hemolymph post-infection. Means with different letters are significantly different by the Tuckey’s test. The experiment was performed in triplicate.

Figure 8

Rna01 and extracellular vesicle (EV) production by Actinobacillus pleuropneumoniae. (A) Transmission electron microscopy of EVs produced by A. pleuropneumoniae strains. (B) Size measurement by dynamic scattering light (DLS) of the EVs produced by the strains. Minimum, majority and maximum size of the EVs are shown by colored bars below each graph. (C) Relative abundance of EV production among strains. (D) OMP profile of EVs produced by the strains. (E) Killing assay of G. mellonella after administration of EVs. Means with different letters are significantly different by the Tuckey’s test. The experiment was performed in triplicate.

Figure 9

Comparative analysis among Rna01 and other extracytoplasmic stress associated sRNAs. (A) Heatmap based on matrix identity of the sRNAs. (B) Secondary structure of the sRNAs. The blue shades represent the seed region of interaction with OMP mRNA targets.

Figure 10

Mechanism of activity proposed for the novel sRNA Rna01. (A) The Figure summarizes the expression of rna01 in normal and stress conditions (by σ⁷⁰ and σ^E, respectively) (1), followed by regulation of the targets in an Hfq dependent manner (2), inhibiting (3), or inducing (4) translation of the targets. Also, Rna01 regulates translation of the OMP targets in a Hfq-independent manner (5) by inhibiting translation (6). (B) The activity of Rna01, mediated or not by the chaperone Hfq, affects diverse phenotypes of A. pleuropneumoniae.