A Klebsiella pneumoniae Regulatory Mutant Has Reduced Capsule Expression but Retains Hypermucoviscosity

Abstract

The polysaccharide capsule is an essential virulence factor for Klebsiella pneumoniae in both community-acquired hypervirulent strains as well as health care-associated classical strains that are posing significant challenges due to multidrug resistance. Capsule production is known to be transcriptionally regulated by a number of proteins, but very little is known about how these proteins collectively control capsule production. RmpA and RcsB are two known regulators of capsule gene expression, and RmpA is required for the hypermucoviscous (HMV) phenotype in hypervirulent K. pneumoniae strains. In this report, we confirmed that these regulators performed their anticipated functions in the ATCC 43816 derivative, KPPR1S: rcsB and rmpA mutants are HMV negative and have reduced capsule gene expression. We also identified a novel transcriptional regulator, RmpC, encoded by a gene near rmpA The ΔrmpC strain has reduced capsule gene expression but retains the HMV phenotype. We further showed that a regulatory cascade exists in which KvrA and KvrB, the recently characterized MarR-like regulators, and RcsB contribute to capsule regulation through regulation of the rmpA promoter and through additional mechanisms. In a murine pneumonia model, the regulator mutants have a range of colonization defects, suggesting that they regulate virulence factors in addition to capsule. Further testing of the rmpC and rmpA mutants revealed that they have distinct and overlapping functions and provide evidence that HMV is not dependent on overproduction of capsule. This distinction will facilitate a better understanding of HMV and how it contributes to enhanced virulence of hypervirulent strains.IMPORTANCEKlebsiella pneumoniae continues to be a substantial public health threat due to its ability to cause health care-associated and community-acquired infections combined with its ability to acquire antibiotic resistance. Novel therapeutics are needed to combat this pathogen, and a greater understanding of its virulence factors is required for the development of new drugs. A key virulence factor for K. pneumoniae is the capsule, and community-acquired hypervirulent strains produce a capsule that causes hypermucoidy. We report here a novel capsule regulator, RmpC, and provide evidence that capsule production and the hypermucoviscosity phenotype are distinct processes. Infection studies showing that this and other capsule regulator mutants have a range of phenotypes indicate that additional virulence factors are in their regulons. These results shed new light on the mechanisms controlling capsule production and introduce targets that may prove useful for the development of novel therapeutics for the treatment of this increasingly problematic pathogen.

Keywords: HMV; RmpA; RmpC; capsular polysaccharide; hypervirulent.

FIG 1

Capsule production and mucoviscosity are reduced in the DBD mutants. Saturated overnight cultures were subcultured into fresh M9-CAA and grown for 6 h at 37°C. Uronic acid (A) and mucoviscosity (B) were assessed as described in Materials and Methods. The data presented here are from a representative assay. The strains used are KPPR1S (wild type [WT]), VK506 (ΔmanC), VK248 (ΔrcsB), VK352 (ΔrmpA), VK487 (ΔrmpC), VK277 (ΔkvrA), and VK410 (ΔkvrB). The one-way ANOVA test was performed to determine statistically significant differences between each mutant and WT. ****, P ≤ 0.0001; ns, not significant.

FIG 2

DBD mutants have altered host cell associations. BMDMs were inoculated with the indicated strains at an MOI of 50 and allowed to interact for 1 h. Adherent (A) and intracellular (B) bacteria were determined as described in Materials and Methods. The data presented here are from a representative assay. The strains used are the same strains used in Fig. 1. The one-way ANOVA test was performed to determine statistically significant differences between each mutant and WT. *, P ≤ 0.05; **, P ≤ 0.01; ****, P ≤ 0.0001.

FIG 3

DBD mutants are attenuated in a mouse pneumonia model. Mice were inoculated with 2 × 10⁴ CFU of the indicated strains and euthanized at 24 or 72 hpi. Lungs and spleens were removed, macerated, and plated for determination of bacterial burden as CFU per gram of tissue. The data in panels A (lungs) and B (spleens) were compiled from three experiments. Representative experiments with complemented strains are presented in panels C and D. The strains used are KPPR1S (WT), VK248 (ΔrcsB), VK352 (ΔrmpA), VK487 (ΔrmpC), VK532 (rcsB_comp), VK379 (rmpA_comp), and VK487 with pKW185 (pRmpC). The Mann-Whitney test was performed to determine statistically significant differences between mutants and WT. *, P ≤ 0.05; ****, P ≤ 0.0001.

FIG 4

Capsule gene expression is affected by loss of DBD genes. (A) Schematic of the capsule locus containing genes for sugar precursor biosynthesis, polymerization, and export. (B to E) Saturated cultures of WT and mutant strains carrying plasmids with transcriptional gfp fusions were subcultured and grown as described in the legend to Fig. 1. Relative fluorescence units (RFU) were measured, normalized first to the culture OD₆₀₀, and then to WT (set at 100). The three characterized promoters tested were galF-gfp (pPROBE_galF) (B), wzi-gfp (pPROBE_wzi) (C), manC-gfp (pPROBE_manC) (D). (E) Complementation assays were performed using strains transformed with manC-gfp and individual complementation plasmids (pRcsB [pKW173], pRmpA [pKW184], pRmpC [pKW185], pKvrA [pTM006], and pKvrB [pTM007]). The one-way ANOVA test was performed to determine statistically significant differences between each mutant and WT. ***, P ≤ 0.001; ****, P ≤ 0.0001.

FIG 5

RcsB, KvrA, and KvrB control rmpA expression. (A) rmpC is in an operon with rmpA. Standard PCR was performed with primers CB472 and CB498 (black arrows) using wild-type genomic DNA (gDNA), samples from cDNA synthesis reactions with reverse transcriptase (RT+) and without reverse transcriptase (RT−), or with no template (nt). The RNA used to generate the cDNA was isolated from strain KPPR1S grown in LB from a previously published data set (23). Promoter-gfp fusions for rmpA (rmpA-gfp, pKW174) (B), kvrB (kvrB-gfp, pPROBE_kvrB) (C), kvrA (kvrA-gfp, pPROBE_kvrA) (D), and rcsDB (rcsDB-gfp, pKW170) (E) were transformed into the indicated strains and grown as described in the legend to Fig. 1. (F and G) Strains containing manC-gfp were transformed with the indicated complementing plasmids; plasmid names are as given in the legend to Fig. 4. The one-way ANOVA test was performed to determine statistically significant differences between each mutant and WT. ***, P ≤ 0.001; ****, P ≤ 0.0001.

FIG 6

RmpA and RmpC have overlapping and independent functions. The WT, ΔrmpA, and ΔrmpC strains were transformed with manC-gfp (pPROBE_manC) and pRmpA (pKW184), pRmpC (pKW185), or pRmpAC (pKW186), grown as described in the legend to Fig. 1 and assayed for manC expression (A) or mucoviscosity (B). The one-way ANOVA test was performed to determine statistically significant differences between each mutant and WT. **, P ≤ 0.01; ***, P ≤ 0.001; ****, P ≤ 0.0001.

FIG 7

Alignment of RmpA amino acid sequences from hv strains. DNA sequences were obtained from the nucleotide accession numbers, translated, and aligned using Geneious v5.3.6. The strains, accession numbers, and open reading frames were as follows: KPPR1S, GenBank accession no. CP009208.1, ORF VK055_5097 (36); NTUH-c, accession no. AP006725, ORF KP1_3619 (37); NTUH-p, accession no. AP006726, ORF KP1_p020 (37); CG43, accession no. AY378100.1 (pLVPK) (20), ORF LV255; Kp52.145, accession no. (plasmid II) NZ_FO834905, ORF BN49_RS00655 (22).