- Research article
- Open Access
In vivo characterization of an Hfq protein encoded by the Bacillus anthracis virulence plasmid pXO1
© The Author(s). 2017
Received: 21 November 2016
Accepted: 7 March 2017
Published: 14 March 2017
Bacterial Hfq proteins post-transcriptionally regulate gene expression, primarily by mediating the interaction between sRNAs (small RNAs) and their target mRNAs. The role of Hfq-based regulation has been well defined in Gram-negative bacteria, but comparatively less is known about the impact of Hfq proteins in Gram-positive species. The Gram-positive pathogen Bacillus anthracis (causative agent of anthrax) is distinct in that it expresses three homologs of Hfq: Hfq1 and Hfq2 from the chromosome, and Hfq3 from the pXO1 virulence plasmid.
In this study, we utilized overexpression as a strategy to examine the impact of Hfq3 on B. anthracis physiology. The increase in Hfq3 protein levels led to anomalous cell shape and chain formation, which manifested as a severe growth defect. This phenotype was specific to B. anthracis, as Hfq3 expression in B. subtilis at similar levels was not toxic. Toxicity was dependent on residues on the distal face of Hfq3 that are involved in mRNA binding in other bacterial species.
Thus, we hypothesize that Hfq3 interacts with RNA(s) involved in essential functions in the B. anthracis cell, leading to increased binding upon overexpression that either sequesters or accelerates degradation of RNAs important for growth. These results not only aid in elucidating the role of Hfq proteins in B. anthracis, but also contribute to our current understanding of Hfq in Gram-positive bacteria.
Hfq proteins primarily function as chaperones in the bacterial cell, facilitating the interaction between small RNAs (sRNAs) and their target mRNAs to regulate gene expression [1, 2]. Hfq has also been implicated in the direct modulation of the expression of certain mRNAs . Hfq proteins are members of the Sm and Sm-like RNA-binding family and generally form hexameric rings that bind RNAs on the distal and proximal faces, as well as on a charged rim patch on the outer edge of the ring [4–8].
The role and impact of Hfq proteins has been well characterized in Gram-negative bacteria, with Hfq deletion strains exhibiting severe growth defects, increased susceptibility to a range of environmental stresses, and decreased virulence . In E. coli, Hfq overexpression negatively impacts cell physiology by decreasing FtsZ expression, inhibiting proper cell division .
Conversely, the role of Hfq and its associated sRNA regulation has not been fully or clearly elucidated in Gram-positive bacteria, despite the expression of numerous sRNAs in these species. For example, deletion of hfq from Staphylococcus aureus has no major effects on physiology, and sRNA regulation is not impacted . Similarly, while Bacillus subtilis Hfq associates with numerous sRNAs in vivo , Δhfq B. subtilis strains exhibit only subtle effects on their physiology and transcriptome [13, 14]. In contrast, in Listeria monocytogenes, Hfq plays a direct role in modulating stress tolerance and virulence  and interacts effectively with specific sRNAs . A recent study in Clostridium difficile identified multiple phenotypes associated with the knockdown of hfq expression, including the first evidence of an hfq-dependent growth phenotype in a Gram-positive species .
Our recent work  and that of Panda et al.  expanded the understanding of Hfq structure and function within the Bacillus genus by characterization of the structure and function of recombinant B. anthracis Hfq proteins. Hfq1 (BA_1656; Hfq2 in ) does not form the typical hexameric structure expected for the protein family; instead, it behaves as a monomer that appears to disrupt E. coli Hfq’s chaperone activity with a dominant negative phenotype . In contrast, Hfq2 (BA_3842; Hfq1 in ) forms the standard hexamer . Purified Hfq3 (GBAA_RS29265) behaves as a mixture of hexamer and monomer, and His-Hfq3 partially complements the E. coli Δhfq phenotype . Panda et al.  reported binding of RNAs to both Hfq2 and Hfq3 (our nomenclature) in vitro; however, analysis of Hfq3 binding was limited to a mixed pool of labeled cellular RNAs, and no specific RNA-Hfq3 binding pairs were identified.
In this study, we examine the effects of overexpressing Hfq3 in Bacillus anthracis, as a means of identifying potential Hfq3-regulated pathways. We selected Hfq3 for further analysis due to its sequence divergence from other Bacillus Hfq proteins, as well as its ability to form hexamers and its presence on one of B. anthracis’ virulence plasmids. We observed a severe growth defect associated with the overexpression of Hfq3 in B. anthracis, but not in B. subtilis. Based on mutational analysis, we determined that this phenotype is dependent on residues on the distal face of Hfq3 that are associated with mRNA binding in other species. Overall, we report a novel phenotype associated with Hfq overexpression in a Gram-positive species, contributing to our current understanding of structure-function relationships of Hfq proteins in Gram-positive bacteria.
Bacterial strains, media, and growth conditions
All bacterial strains, plasmids, and primers used in this study are listed in Additional file 1: Tables S2–S4. TOP10, DH5-α, and SCS110 E. coli strains were used for cloning. B. anthracis strain Ames 35 (pXO1+, pXO2−; derived from ) and B. subtilis strain PY79  were grown in LB broth/plate or NBY broth/plate (containing 0.8% nutrient broth, 0.3% yeast extract, and 0.9% sodium bicarbonate). LB and NBY/0.9% sodium bicarbonate cultures were incubated at 37 °C in air or a CO2 environment, respectively. E. coli Hfq in vivo complementation strains (YN585 derivatives) were grown on MacConkey agar plates (BD Difco; Franklin Lakes, NJ) incubated at 37 °C in air. When appropriate, antibiotics were added to both liquid and solid media: ampicillin (100 μg/mL); kanamycin (15 μg/mL).
Cloning and transformation
Each hfq gene was cloned into the pSW4 shuttle expression vector  using the NdeI and BamHI restriction sites. After passage through SCS110 to prepare unmethylated DNA, pSW4 vectors were transformed into B. anthracis via electroporation. pSW4 expression vectors were transformed into B. subtilis strain PY79 via natural competency using Modified Competency Media (3 mM sodium citrate, 2% glucose, 20 μg/mL ferric ammonium citrate, 0.1% casein hydrolysate, 0.2% potassium glutamate, 100 mM potassium phosphate, 10 mM MgSO4) .
Total lysate and western blot analysis
Bacteria were grown in LB or NBY/0.9% sodium bicarbonate cultures in either air (LB) or 15% CO2 (NBY). Cells were harvested via centrifugation, and cell pellets were lysed in 1X PBS via beadbeating. Protein concentrations of lysate supernates were estimated via an A280 measurement, and gels/blots were normalized by loading equal amounts of total protein per well, which was validated by Coomassie Blue staining. Proteins were separated on 4–20% Tris-glycine gels (Life Technologies; western blot) or 16% Tricine gels (Life Technologies; total lysate). Samples were run as semi-native (not heated) vs. boiled (5 min, 99 °C), and were prepared with either 2X Tris-glycine SDS loading buffer (Life Technologies; final [SDS] 1%) or 2X Tricine SDS loading buffer (Life Technologies; final [SDS] 4%). Both gel systems utilized running buffer containing 0.1% SDS. Following SDS-PAGE, 16% Tricine gels were Coomassie Blue stained, and 4–20% Tris-glycine gels were electroblotted onto nitrocellulose. Membranes were blocked with Odyssey Blocking Buffer (LI-COR Biosciences, Lincoln, NE) and probed with anti-Hfq3 antibodies (1:2000 dilution) overnight at room temperature. An anti-mouse IR800-conjugated secondary antibody (LI-COR Biosciences) was utilized for detection, and blots were visualized on the Odyssey LI-COR infrared scanner.
Point mutations were introduced into Hfq3 in the pSW4 and pQE80 vectors by the Q5 site-directed mutagenesis kit (New England Biolabs; Ipswich, MA) using the primers listed in Additional file 1: Table S4, as well as the primers in  for the Q4A mutation in pQE80. Each mutant gene was fully sequenced in the coding region.
Samples for transmission electron microscopy (TEM) and differential interference contrast microscopy (DIC) were prepared from LB plate scrapings. For TEM, cells were washed with PBS and resuspended in a 2% paraformaldehyde and 0.1 M sodium cacodylate fixative. Images were collected using a Hitachi H-7500 transmission electron microscope. For DIC imaging, a Leica TCS SP5 confocal microscope DMI 6000 (Leica Microsystems GmbH, Manheim, Germany) was used to collect the images with a 63x oil immersion objective NA 1.4. Images were post-processed using Imaris image processing software (Bitplane, Germany) and Adobe Photoshop 7.0 (Adobe Systems).
Complementation in E. coli
pQE80 His-Hfq3 expression plasmids (Additional file 1: Table S2) were transformed into a ΔlacZΔhfq strain of E. coli carrying an rpoS-lacZ translational fusion (strain YN585). Resulting strains were streaked to MacConkey (+100 μg/mL ampicillin) agar plates (BD Difco) and incubated for ≈ 16 h at 37 °C to observe color development. Lactose in the MacConkey plates induces constitutive expression of His-Hfqs from the plasmids. β-galactosidase assays were conducted on overnight MacConkey plate scrapings essentially as described in reference . To observe total Hfq protein expression levels, cells were scraped from MacConkey plates and lysed with an 8 M urea buffer (pH 8). Lysates were visualized on 16% Tricine gels by Coomassie Blue staining.
We observed that deletion of the His-tag from the wild-type Hfq3 construct reduced the pink color on MacConkey plates; this may be due to reduction in protein accumulation in the absence of the tag, impacts on other properties of heterologous expression, and/or impacts of the tag on the biochemistry of the protein itself. Therefore, the original His-Hfq3 system for heterologous production in E. coli was retained. (We note that all Bacillus phenotypes described in this paper are in the absence of any additional tag residues, with the native protein sequence).
Antibiotic sensitivity and cell autolysis assays
Minimum inhibitory concentration (MIC) testing was performed based on the protocol described in reference , with modifications indicated below. LB plate scrapings of individual colonies of Hfq3 or GFP overexpression strains were diluted in LB-Kanamycin (15 μg/ml). Vancomycin and ampicillin were serially diluted in LB-Kanamycin across a 96-well plate with final concentrations ranging from 32–0.06 μg/ml (for vancomycin) and from 32–0.001875 μg/ml (for ampicillin). Hfq3 and GFP overexpression strains were inoculated to a final CFU count of 1.4–6.8 × 104/well. Plates were incubated at 37 °C with shaking at 200 rpm for 19 h, and then visually observed for the presence of turbidity to determine MICs. MICs in Additional file 1: Table S5 represent results for 12 GFP overexpression strain colonies and 21 Hfq3 overexpression strain colonies aggregated across at least three independent experiments.
To monitor cell autolysis, LB plate scrapings of Hfq3 or GFP overexpression strains were resuspended to an OD600 of 0.7–1.0 in LB-Kanamycin (15 μg/ml). Due to the small colony size of the Hfq3 strain, replicates for both strains were created from multiple colonies. Cultures were incubated at 37 °C with shaking at 225 rpm for 30 min to resume active growth, followed by addition of Triton X-100 to a final concentration of 0.05% (v/v) and continued incubation at 37 °C.
Female Balb/cJ mice (6–8 weeks old) were purchased from Jackson Laboratories (Bar Harbor, ME). All animal studies were performed in accordance with National Institutes of Health and Animal Welfare Act guidelines on protocol LPD8E, approved by the Animal Care and Use Committee of the National Institute of Allergy and Infectious Diseases, National Institutes of Health.
To generate mouse polyclonal antibodies against Hfq3, Balb/cJ mice were immunized -12, -9, -6 and -4 weeks prior to terminal serum bleeds. Immunizations were performed subcutaneously with 10–30 μg of purified His-Hfq3. Aluminum hydroxide gel (Alum; Sigma, St. Louis, MO) was used as adjuvant and premixed with antigen at 1:1 (v/v Hfq3: Alum). Antibody reactivity and specificity were assessed by western blotting (Additional file 2: Figure S1A, B) and by dilution ELISA. The anti-Hfq3 antibody was specific for purified Hfq3 (as compared to Hfq1 or Hfq2; Additional file 2: Figure S1B). The antibody was able to detect overexpressed Hfq3 in B. anthracis cell lysates, but not endogenous levels of Hfq3 in B. anthracis lysates (Additional file 2: Figure S2A). We previously demonstrated that Hfq3 is expressed at the RNA level in log-phase B. anthracis . Based on western blotting profiles, it is likely that this antibody recognizes a conformational epitope(s), resulting in better recognition of Hfq3 hexamer vs. Hfq3 monomer. The in vivo structure of overexpressed Hfq3 as observed by western blot (mix of hexamer and monomer) is consistent with our previous in vitro work with purified His-Hfq3 .
An Hfq3 overexpression system as an investigational tool
Our previous work compared the biochemical properties of the three Hfq proteins of B. anthracis. While the Sm2 RNA-binding motif is well conserved in Hfq3, the unstructured C-terminus present in E. coli Hfq is significantly truncated (Fig. 1). In searching for physiological roles of these proteins, we did not observe defects in the growth rates of hfq deletion strains, including a Δhfq1hfq2hfq3 strain , consistent with previous findings in B. subtilis . This led us to investigate phenotypes linked to overexpression of Hfq proteins in B. anthracis as an alternative means of identifying roles of this protein family in Bacilli. Intentional overexpression is a useful genetic tool for the identification of cellular pathways linked to a gene of interest in wild-type cells, and can identify phenotypes that are not evident from deletion analysis in some cases .
Characterization of the Hfq3 toxic growth phenotype in B. anthracis
To further characterize the impact of Hfq3 overexpression on B. anthracis, we directly compared the growth of Ames 35 (parent strain), Ames 35 overexpressing Hfq2, Ames 35 overexpressing Hfq3, and Ames 35 expressing an unrelated protein (GFP variant). These plates recapitulated the negative impact of Hfq3 overexpression both at 15% CO2 (Fig. 2c, iii) and at 5% CO2 (Additional file 2: Figure S2B). Overexpressed Hfq3 has a toxic phenotype in air (Fig. 2c, iii, left) that is variable (Fig. 2c, iii, left; Additional file 2: Figure S2E). This can be ascribed to leaky expression from the pSW4 vector, since Hfq3 expression was observed under both non-inducing and inducing conditions, with an increase in Hfq3 production upon induction (compare LB to NBY/CO2; Fig. 2d). Under inducing conditions, the overexpressed Hfq3 accumulates to lower levels than overexpressed Hfq2, relative to total cellular protein, as is visible in Fig. 2e. The GFP variant protein is expressed more abundantly than either of the Hfqs (Fig. 2e), indicating that overexpression of proteins in general does not impair growth in B. anthracis.
While Hfq2 overexpression did not reproducibly inhibit growth on the plate patches (Fig. 2c; Additional file 2: Figure S2B), about 50% of the Ames 35 pSW4-Hfq2 transformants across 52 individual colonies and 9 experiments exhibited toxicity under inducing conditions (NBY/CO2). In parallel experiments, >95% of independent pSW4-Hfq3 transformants exhibited growth defects. This defect is evident after 24 h of growth at 30 °C for the Hfq2 overexpression strain (Additional file 2: Figure S2C). However, we did not observe any evidence on the plates of pSW4-Hfq2 toxicity under non-inducing conditions (low level of Hfq expression, as described in Fig. 2a). Therefore, we propose that overexpression of Hfq2 in B. anthracis has a toxic effect once cellular protein levels reach higher concentrations than for Hfq3, which appears to be toxic to the cells even at low, leaky levels of expression. Additionally, we hypothesize that variability in Hfq2 toxicity is due to variations in protein expression levels across independent transformants.
Distal face residues are required for Hfq3 toxicity
One potential etiology of the Hfq3 overexpression phenotype could be the interaction of overexpressed Hfq3 with endogenous Hfq proteins. However, an Ames 35 Δhfq1hfq2hfq3-Hfq3 overexpression strain retains the toxic phenotype (Fig. 3d; Additional file 2: Figure S2E).
B. anthracis Hfq3 mutant nomenclature with analogous Hfq residues in E. coli and S. aureus
B. anthracis Hfq3 Mutant
Analogous E. coli Hfq Residue
Analogous S. aureus Hfq Residue
E5A, D6A (“ED Patch”)
I12R, E13R (“Rim Rev”)
The impact of each mutation on the growth of Hfq3-overexpressing Ames 35 cells was examined under inducing conditions (NBY/CO2). Mutation of residues on the distal face (F21D, V26D, R27A) resulted in rescue of normal growth, as observed by the plate method (Fig. 4c, right plate; Additional file 2: Figure S3A). In contrast, mutation of key residues on the proximal face of Hfq3 (Q4A, K53A, ED Patch) did not restore normal growth (Fig. 4c; Additional file 2: Figure S3A). Reversion of the rim residues to the E. coli sequence would have been expected to increase toxicity, if this site increased binding of critical RNAs; instead, toxicity was somewhat decreased (Fig. 4c; Additional file 2: Figure S3A), which may be due to the disruption in hexamer formation (Fig. 4d). The relative importance of the distal face mutants is suggestive of mRNA binding by Hfq3, as opposed to an mRNA-sRNA pair leading to toxicity.
Western blot analysis confirmed that each of the three Hfq3 distal face mutants is abundantly expressed in B. anthracis, indicating that their normal growth is not due to loss of protein expression (Fig. 4d). The K53A (proximal) and rim revertant mutants are expressed, but do not exhibit stable hexamer formation under these gel conditions (Fig. 4d). Similarly, the corresponding K56A E. coli Hfq mutant is able to bind RNA, despite apparent lack of hexamer formation as assessed on a semi-native gel [26, 31]. We hypothesize that these mutants form transient hexamers that are capable of RNA binding, but dissociate too quickly under semi-native gel conditions to be observed here. Additionally, the toxicity of the K53A overexpression strain provides evidence that the toxicity of Hfq3 is not due to Hfq aggregation.
Distal face Hfq3 mutants also disrupt an E. coli complementation phenotype
Mutants in His-Hfq3 exhibited defects in complementation of the E. coli Δhfq phenotype (Fig. 5a, b), with the largest decrease in complementation observed for the distal face mutants F21D and V26D (Additional file 1: Table S1). The proximal face mutants, as well as the distal face mutant R27A, exhibited a partial loss of the complementation phenotype (Fig. 5b; Additional file 1: Table S1). (However, we note that the results in Fig. 4d suggest that the K53A and rim revertant mutants exhibit unstable hexamer formation, so we have not focused on those mutants here). Denaturing lysate gels reveal that Hfq3 protein accumulates in each of the mutants (Additional file 2: Figure S3B), suggesting that color differences cannot simply be ascribed to lack of His-Hfq3 expression. Therefore, residues involved in the toxic Hfq3 overexpression phenotype were also involved in the complementation phenotype.
Hfq3 expression in B. subtilis does not negatively impact growth
To further probe the physiological mechanism of Hfq3 toxicity, we transformed pSW4-Hfq3 into B. subtilis strain PY79. Expression from pSW4 is not induced in B. subtilis by CO2, but occurs in both air and CO2 conditions (observable from the color of GFP expression in Additional file 2: Figure S4A, iv). In contrast to B. anthracis, the B. subtilis Hfq3 expression strain grew normally (Fig. 5c; Additional file 2: Figure S4A; Additional file 2: Figure S5A). Hfq3 protein was expressed in B. subtilis (Additional file 2: Figure S5B, C), indicating that the lack of phenotype is not due to lack of protein expression. The Hfq3-R27A mutant had a change in appearance (opaque to more translucent on streaks); however, the Hfq3 overexpression phenotypes in B. subtilis were clearly distinct from the phenotypes in B. anthracis. Similar to Hfq2 overexpression in B. anthracis, Hfq2 expression in B. subtilis yielded an intermediate toxic effect, at a frequency of approximately 28% in 36 transformants (Fig. 5c; Additional file 2: Figure S4A-C). Therefore, the Hfq3 overexpression phenotype exhibits species specificity.
Overexpression of Hfq3 leads to aberrant B. anthracis cellular morphology
The rounded appearance of the Hfq3-overexpressing cells and the visible defects in cell wall appearance suggest that the increased levels of Hfq3 may impact the activity of MreB and/or the activity of a murein hydrolase in B. anthracis. An mreB mutant in B. subtilis exhibits issues with the maintenance of cell width as well as cell viability; these growth defects were reversible by propagation in media containing magnesium and sucrose . However, supplementation of the NBY/bicarbonate medium by sucrose and magnesium (SMM) did not restore normal growth of the Hfq3 overexpression strain (Fig. 6c).
Finally, to characterize the impact of Hfq3 overexpression on the cell envelope, we compared the physiology of the B. anthracis Hfq3 overexpression strain to the B. anthracis GFP overexpression strain. Both strains exhibited similar sensitivities (MICs) to both vancomycin and ampicillin (Additional file 1: Table S5). Additionally, both strains exhibited similar sensitivities to lysis by 0.05% Triton X-100 (Fig. 6d). The Hfq3 overexpression strain was capable of spore formation, and the Hfq3 overexpression plasmid was still toxic when transformed into a Spo− B. anthracis strain.
Discussion & conclusions
Previously, our work elucidated the molecular properties of the unusual set of three distinct Hfq proteins produced in the Bacillus anthracis cell, including Hfq3, expressed from a gene located on the pXO1 virulence plasmid of the pathogen. Here, we exploited the pSW4 plasmid overexpression system as a tool to examine functions of Hfq3 in vivo in B. anthracis. A similar overexpression strategy was successfully utilized to identify new binding activities of Hfq in E. coli [10, 36], although we note that the E. coli system is distinct in its clear and diverse phenotypes observed in the Δhfq E. coli mutant. In B. anthracis, we observed a striking toxic effect of Hfq3 overexpression, even at the lower levels observed under LB growth conditions. The toxic effect is dependent on Hfq distal face residues implicated in mRNA binding in E. coli, but independent of proximal face residues implicated in small RNA binding. Strikingly, toxicity was species specific and not simply a general effect; overexpression of Hfq3 in B. subtilis did not cause the same growth defect.
We propose that increased intracellular concentrations of Hfq3 lead to increased levels of binding of Hfq3-interacting mRNAs in the B. anthracis cell. This could interfere with cellular physiology either by sequestering and blocking translation of these mRNAs due to the high Hfq3 concentrations, or, alternatively, by increasing the rate of degradation and cycling of these mRNAs. In E. coli, (ARN)X sequences allow for mRNA binding to the distal face of E. coli Hfq; binding can then target mRNAs for degradation in some cases . The results for Hfq3 overexpression in B. subtilis seem to suggest that the toxic phenotype is not due to general sequestration of all mRNAs. The distal face mutagenesis does not directly demonstrate RNA binding, and it is alternatively possible that the Hfq3 overexpression phenotype is due to protein-protein interactions. However, the importance of residues previously demonstrated to disrupt mRNA binding to E. coli Hfq is suggestive of an mRNA binding mechanism in the Hfq3-overexpressing cells.
A recent study in Clostridium difficile—a species in which Hfq knockdown leads to significant growth defects (unique for Gram-positive bacteria)—identified changes in transcript levels of genes involved in cell wall structure and transport in the absence of Hfq . Importantly, Hfq depletion in C. difficile led to cell elongation and cell wall deformation, leading Boudry et al.  to propose that Hfq regulation influences C. difficile cell wall composition. While the Δhfq strain of B. anthracis used here exhibits normal growth in rich media, our findings suggest that Hfqs may be playing more subtle roles in the physiology of B. anthracis. The collapse of normal cell shape and chain formation in the Hfq3 overexpressing cells could be consistent with an impact on cell wall formation and cell division, correlating with some of the functions in C. difficile. Alternatively, the toxic phenotype could be due to other cellular impacts, such as loss of membrane integrity or sequestration of RNAs involved in other essential cell functions. The results of the antibiotic sensitivity and autolysis assays suggest that the phenotype does not involve differences in peptidoglycan crosslinking/teichoic acid structure [38–41].
While overexpression of Hfq2 appears to lead to toxicity as well, the Hfq3 phenotype is distinct in its presence at lower relative levels of protein expression, as observed from the difference in cellular appearance under air (non-inducing) conditions. The Hfq2 phenotype may be the result of weaker binding of the same RNA(s) to the Hfq2 hexamer when the protein reaches higher levels in the cell. We note that we cannot rule out the possibility that growth of the Hfq2 overexpression strain reflects rapid accumulation and growth of suppressor mutants.
Combined with the findings in B. anthracis, our mutational analysis in E. coli strengthens the argument for the importance of the distal face in Hfq3’s functional capabilities. Additionally, the difference in the Hfq3 rim sequence is suggestive of differences between Hfq3 and E. coli Hfq in their abilities to bring together sRNAs and their target mRNAs for regulation. Previously, we modeled the structure and charges of the Hfq3 hexamer, based on its amino acid sequence . The proximal face is predicted to have an inner ring of positive charges, similar to S. aureus Hfq, but also exhibits significant patches of RNA-repelling negative charge, in contrast to the findings from E. coli Hfq’s crystal structure. The Hfq3 distal face is predicted to have a greater distribution of positive charge that may be favorable for mRNA binding. Another potential variable is the very short length of the Hfq3 C-terminal tail, which still allows for hexamer formation, but may impact the sRNA-binding function of Hfq3.
Overall, the finding of a B. anthracis-specific overexpression phenotype for Hfq3 is novel for a Gram-positive species, and provides an impetus for the further study of Hfq3-associated functions in B. anthracis. The next steps for research on the function(s) of this protein will be to elucidate the impact of reduction in Hfq3 levels on the B. anthracis transcriptome, with the results presented here as a point of comparison for identification of transcripts potentially linked to impacts on cellular structure or metabolism. Additionally, the results presented here provide the foundation for a much more targeted exploration of potential phenotypes of Δhfq3 B. anthracis cells. Further analysis is important in understanding the level of physiological relevance of the findings from the overexpression system. To this end, we conducted a preliminary transcriptome analysis of an hfq3 deletion mutant of Ames 35, as compared to the wild-type strain grown in NBY + bicarbonate (in CO2) to late log phase (Additional file 3: Experiment 1). These initial results suggest that there are gene expression differences in the absence of Hfq3 that may have impacts on cellular physiology. Of potential future interest would also be the role of Hfqs in the context of the pXO2 plasmid, present in fully virulent strains of B. anthracis and responsible for capsule formation.
Finally, we propose that the relative importance of distal face mRNA-dependent mechanisms, as opposed to sRNA-mRNA pairing mechanisms, may be of broader importance in RNA regulation in Gram-positive species. Such mechanisms would be consistent with the small subset of sRNA-independent mechanisms for post-transcriptional regulation of certain mRNAs in E. coli [37, 42].
The authors wish to thank Dr. Gordon Cheung for construction of the Hfq overexpression plasmids; Dr. Sundar Ganesan from the Research Technologies Branch/NIAID/NIH for the DIC analysis and imaging; Neha Dhasmana for preparing the samples for TEM; RML/NIAID/NIH for the TEM images and assistance with strain characterization; Kishore Kanakabandi for isolation of RNAs, and Stacy Ricklefs and Sarah Anzick for RNA-Seq from the RML Genomics Unit; and Dr. Kumaran Ramamurthi for the B. subtilis transformation media.
This work was funded by the Intramural research programs of the National Institute of Allergy and Infectious Diseases and of the National Cancer Institute, National Institutes of Health. The opinions expressed in this manuscript are entirely those of the authors and are not necessarily official positions of the National Institutes of Health or the U.S. Department of Agriculture. Inclusion of specific product brand names does not imply endorsement.
Availability of data and materials
The main manuscript is not associated with any large datasets. The authors will make plasmids, bacterial strains, etc., as well as any information related to Additional file 3: Experiment 1, available to others upon request.
CEV, SG, and SHL conceived of the study and interpreted data; ABK, CEV, and APP designed, performed, and interpreted microbiological and molecular experiments; EKA performed and interpreted mutagenesis experiments; ABK and CEV drafted the manuscript; SFP designed the transcriptomics experiment and interpreted data; CM analyzed and interpreted transcriptomics data; and MM designed and performed animal experiments to generate mouse antibodies. All authors approved the final manuscript.
The authors declare that they have no competing interests.
Consent for publication
All animal studies were performed in accordance with National Institutes of Health and Animal Welfare Act guidelines on protocol LPD8E, approved by the Animal Care and Use Committee of the National Institute of Allergy and Infectious Diseases, National Institutes of Health.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- De Lay N, Schu DJ, Gottesman S. Bacterial small RNA-based negative regulation: Hfq and its accomplices. J Biol Chem. 2013;288:7996–8003.View ArticlePubMedPubMed CentralGoogle Scholar
- Valentin-Hansen P, Erickson M, Udesen C. The bacterial Sm-like protein Hfq: a key player in RNA transactions. Mol Microbiol. 2004;51:4525–33.View ArticleGoogle Scholar
- Vogel J, Luisi BF. Hfq and its constellation of RNA. Nat Rev Microbiol. 2011;9:578–89.View ArticlePubMedPubMed CentralGoogle Scholar
- Link TM, Valentin-Hansen P, Brennan RG. Structure of Escherichia coli Hfq bound to polyriboadenylate RNA. Proc Natl Acad Sci U S A. 2009;106:19292–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Panja S, Schu DJ, Woodson SA. Conserved arginines on the rim of Hfq catalyze base pair formation and exchange. Nucleic Acids Res. 2013;41:7536–46.View ArticlePubMedPubMed CentralGoogle Scholar
- Sauer E, Schmidt S, Weichenrieder O. Small RNA binding to the lateral surface of Hfq hexamers and structural rearrangements upon mRNA target recognition. Proc Natl Acad Sci U S A. 2012;109:9396–401.View ArticlePubMedPubMed CentralGoogle Scholar
- Schumacher MA, Pearson RF, Moller T, Valentin-Hansen P, Brennan RG. Structures of the pleiotropic translational regulator Hfq and an Hfq-RNA complex: a bacterial Sm-like protein. Embo J. 2002;21:3546–56.View ArticlePubMedPubMed CentralGoogle Scholar
- Wilusz CJ, Wilusz J. Eukaryotic Lsm proteins: lessons from bacteria. Nat Struct Mol Biol. 2005;12:1031–6.View ArticlePubMedGoogle Scholar
- Chao Y, Vogel J. The role of Hfq in bacterial pathogens. Curr Opin Microbiol. 2010;13:24–33.View ArticlePubMedGoogle Scholar
- Takada A, Nagai K, Wachi M. A decreased level of FtsZ is responsible for inviability of RNase E-deficient cells. Genes Cells. 2005;10:733–41.View ArticlePubMedGoogle Scholar
- Bohn C, Rigoulay C, Bouloc P. No detectable effect of RNA-binding protein Hfq absence in Staphylococcus aureus. BMC Microbiol. 2007;7:10.View ArticlePubMedPubMed CentralGoogle Scholar
- Dambach M, Irnov I, Winkler WC. Association of RNAs with Bacillus subtilis Hfq. PLoS One. 2013;8:e55156.View ArticlePubMedPubMed CentralGoogle Scholar
- Hammerle H, Amman F, Vecerek B, Stulke J, Hofacker I, Blasi U. Impact of Hfq on the Bacillus subtilis transcriptome. PLoS One. 2014;9:e98661.View ArticlePubMedPubMed CentralGoogle Scholar
- Rochat T, Delumeau O, Figueroa-Bossi N, Noirot P, Bossi L, Dervyn E, Bouloc P. Tracking the elusive function of Bacillus subtilis Hfq. PLoS One. 2015;10:e0124977.View ArticlePubMedPubMed CentralGoogle Scholar
- Christiansen JK, Larsen MH, Ingmer H, Sogaard-Andersen L, Kallipolitis BH. The RNA-binding protein Hfq of Listeria monocytogenes: role in stress tolerance and virulence. J Bacteriol. 2004;186:3355–62.View ArticlePubMedPubMed CentralGoogle Scholar
- Christiansen JK, Nielsen JS, Ebersbach T, Valentin-Hansen P, Sogaard-Andersen L, Kallipolitis BH. Identification of small Hfq-binding RNAs in Listeria monocytogenes. RNA. 2006;12:1383–96.View ArticlePubMedPubMed CentralGoogle Scholar
- Boudry P, Gracia C, Monot M, Caillet J, Saujet L, Hajnsdorf E, et al. Pleiotropic role of the RNA chaperone protein Hfq in the human pathogen Clostridium difficile. J Bacteriol. 2014;196:3234–48.View ArticlePubMedPubMed CentralGoogle Scholar
- Vrentas C, Ghirlando R, Keefer A, Hu Z, Tomczak A, Gittis AG, et al. Hfqs in Bacillus anthracis: role of protein sequence variation in the structure and function of proteins in the Hfq family. Protein Sci. 2015;24:1808–19.View ArticlePubMedPubMed CentralGoogle Scholar
- Panda G, Tanwer P, Ansari S, Khare D, Bhatnagar R. Regulation and RNA-binding properties of Hfq-like RNA chaperones in Bacillus anthracis. Biochim Biophys Acta. 1850;2015:1661–8.Google Scholar
- Green BD, Battisti L, Koehler TM, Thorne CB, Ivins BE. Demonstration of a capsule plasmid in Bacillus anthracis. Infect Immun. 1985;49:291–7.PubMedPubMed CentralGoogle Scholar
- Zeigler DR, Prágai Z, Rodriguez S, Chevreux B, Muffler A, Albert T, et al. The origins of 168, W23, and other Bacillus subtilis legacy strains. J Bacteriol. 2008;190:6983–95.View ArticlePubMedPubMed CentralGoogle Scholar
- Pomerantsev AP, Kalnin KV, Osorio M, Leppla SH. Phosphatidylcholine-specific phospholipase C and sphingomyelinase activities in bacteria of the Bacillus cereus group. Infect Immun. 2003;71:6591–606.View ArticlePubMedPubMed CentralGoogle Scholar
- Boylan RJ, Mendelson NH, Brooks D, Young FE. Regulation of the bacterial cell wall: analysis of a mutant of Bacillus subtilis defective in biosynthesis of teichoic acid. J Bacteriol. 1972;110:281–90.PubMedPubMed CentralGoogle Scholar
- Wiegand I, Hilpert K, Hancock RE. Agar and broth dilution methods to determine the minimal inhibitory concentration (MIC) of antimicrobial substances. Nat Protoc. 2008;3:163–75.View ArticlePubMedGoogle Scholar
- Prelich G. Gene overexpression: uses, mechanisms, and interpretation. Genetics. 2012;190:841–54.View ArticlePubMedPubMed CentralGoogle Scholar
- Zhang A, Schu DJ, Tjaden BC, Storz G, Gottesman S. Mutations in interaction surfaces differentially impact E. coli Hfq association with small RNAs and their mRNA targets. J Mol Biol. 2013;425:3678–97.View ArticlePubMedPubMed CentralGoogle Scholar
- Schu DJ, Zhang A, Gottesman S, Storz G. Alternative Hfq-sRNA interaction modes dictate alternative mRNA recognition. EMBO J. 2015;34:2557–73.View ArticlePubMedPubMed CentralGoogle Scholar
- De Lay N, Gottesman S. Role of polynucleotide phosphorylase in sRNA function in Escherichia coli. RNA. 2011;17:1172–89.View ArticlePubMedPubMed CentralGoogle Scholar
- Mikulecky PJ, Kaw MK, Brescia CC, Takach JC, Sledjeski DD, Feig AL. Escherichia coli Hfq has distinct interaction surfaces for DsrA, rpoS and poly(A) RNAs. Nat Struct Mol Biol. 2004;11:1206–14.View ArticlePubMedPubMed CentralGoogle Scholar
- Panja S, Santiago-Frango A, Schu DJ, Gottesman S, Woodson SA. Acidic residues in the Hfq chaperone increase the selectivity of sRNA binding and annealing. J Mol Biol. 2015;427:3491–500.View ArticlePubMedPubMed CentralGoogle Scholar
- Panja S, Woodson SA. Hfq proximity and orientation controls RNA annealing. Nucleic Acids Res. 2012;40:8690–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Brown L, Elliott T. Efficient translation of the RpoS sigma factor in Salmonella typhimurium requires host factor I, an RNA-binding protein encoded by the hfq gene. J Bacteriol. 1996;178:3763–70.View ArticlePubMedPubMed CentralGoogle Scholar
- Sledjeski DD, Whitman C, Zhang A. Hfq is necessary for regulation by the untranslated RNA DsrA. J Bacteriol. 2001;183:1997–2005.View ArticlePubMedPubMed CentralGoogle Scholar
- Muffler A, Fischer D, Hengge-Aronis R. The RNA-binding protein HF-I, known as a host factor for phage Qbeta RNA replication, is essential for rpoS translation in Escherichia coli. Genes Dev. 1996;10:1143–51.View ArticlePubMedGoogle Scholar
- Formstone A, Errington J. A magnesium-dependent mreB null mutant: Implications for the role of mreB in Bacillus subtilis. Mol Microbiol. 2005;55:1646–57.View ArticlePubMedGoogle Scholar
- Zambrano N, Guichard PP, Bi Y, Cayrol B, Marco S, Arluison V. Involvement of HFq protein in the post-transcriptional regulation of E. coli bacterial cytoskeleton and cell division proteins. Cell Cycle. 2009;8:2470–2.View ArticlePubMedGoogle Scholar
- Vytvyska O, Moll I, Kaberdin VR, von Gabain A, Blasi U. Hfq (HF1) stimulates ompA mRNA decay by interfering with ribosome binding. Genes Dev. 2000;14:1109–18.Google Scholar
- Peschel A, Vuong C, Otto M, Gotz F. The D-alanine residues of Staphylococcus aureus teichoic acids alter the susceptibility to vancomycin and the activity of autolytic enzymes. Antimicrob Agents Chemother. 2000;44:2845–7.View ArticlePubMedPubMed CentralGoogle Scholar
- Sieradski K, Tomasz A. Inhibition of cell wall turnover and autolysis by vancomycin in a highly vancomycin-resistant mutant of Staphylococcus aureus. J Bacteriol. 1997;179:2557–66.View ArticleGoogle Scholar
- Steen A, Palumbo E, Deghorain M, Cocconcelli PS, Delcour J, Kuipers O, et al. Autolysis of Lactococcus lactis is increased upon D-alanine depletion of peptidoglycan and lipoteichoic acids. J Bacteriol. 2005;187:114–24.View ArticlePubMedPubMed CentralGoogle Scholar
- Wecke J, Perego M, Fischer W. D-alanine deprivation of Bacillus subtilis is without effect on cell growth and morphology but affects the autolytic activity. Microb Drug Resist. 1996;2:123–9.View ArticlePubMedGoogle Scholar
- Mohanty BK, Maples VF, Kushner SR. The Sm-like protein Hfq regulates polyadenylation dependent mRNA decay in Escherichia coli. Mol Microbiol. 2004;54:905–20.View ArticlePubMedGoogle Scholar
- Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, et al. Fast, scalable generation of high-quality protein multiple sequences alignments using Clustal Omega. Mol Syst Biol. 2011;7:539.View ArticlePubMedPubMed CentralGoogle Scholar