Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Aug 18;85(9):e00068-17.
doi: 10.1128/IAI.00068-17. Print 2017 Sep.

Identification and Characterization of Two Klebsiella Pneumoniae lpxL Lipid A Late Acyltransferases and Their Role in Virulence

Affiliations
Free PMC article

Identification and Characterization of Two Klebsiella Pneumoniae lpxL Lipid A Late Acyltransferases and Their Role in Virulence

Grant Mills et al. Infect Immun. .
Free PMC article

Abstract

Klebsiella pneumoniae causes a wide range of infections, from urinary tract infections to pneumonia. The lipopolysaccharide is a virulence factor of this pathogen, although there are gaps in our understanding of its biosynthesis. Here we report on the characterization of K. pneumoniaelpxL, which encodes one of the enzymes responsible for the late secondary acylation of immature lipid A molecules. Analysis of the available K. pneumoniae genomes revealed that this pathogen's genome encodes two orthologues of Escherichia coli LpxL. Using genetic methods and mass spectrometry, we demonstrate that LpxL1 catalyzes the addition of laureate and LpxL2 catalyzes the addition of myristate. Both enzymes acylated E. coli lipid A, whereas only LpxL2 mediated K. pneumoniae lipid A acylation. We show that LpxL1 is negatively regulated by the two-component system PhoPQ. The lipid A produced by the lpxL2 mutant lacked the 2-hydroxymyristate, palmitate, and 4-aminoarabinose decorations found in the lipid A synthesized by the wild type. The lack of 2-hydroxymyristate was expected since LpxO modifies the myristate transferred by LpxL2 to the lipid A. The absence of the other two decorations is most likely caused by the downregulation of phoPQ and pmrAB expression. LpxL2-dependent lipid A acylation protects Klebsiella from polymyxins, mediates resistance to phagocytosis, limits the activation of inflammatory responses by macrophages, and is required for pathogen survival in the wax moth (Galleria mellonella). Our findings indicate that the LpxL2 contribution to virulence is dependent on LpxO-mediated hydroxylation of the LpxL2-transferred myristate. Our studies suggest that LpxL2 might be a candidate target in the development of anti-K. pneumoniae drugs.

Keywords: Klebsiella pneumoniae; LpxL; lipid A; pathogenesis; virulence factors.

Figures

FIG 1
FIG 1
Synthesis of hexa-acylated lipid A in E. coli and K. pneumoniae. (A) In E. coli, LpxL transfers a laurate (C12) group from an acyl carrier protein (ACP) onto the R-2′-hydroxymyristate acyl chain of Kdo2-lipid IVA. Subsequently, LpxM-dependent addition of myristate (C14) onto the R-3′-hydroxymyristate residue results in the synthesis of hexa-acylated Kdo2-lipid A. (B) In K. pneumoniae, LpxL (which we named LpxL2 in this work) and LpxM transfer myristate onto the R-3′-hydroxymyristate residue to complete the synthesis of hexa-acylated Kdo2-lipid A. (C) Proposed lipid A structures follow previously reported structures for K. pneumoniae (23–27). The modifications and genes responsible for each of them are indicated.
FIG 2
FIG 2
K. pneumoniae LpxL2 acylates the 2′ R-3-hydroxymyristoyl group with C14. Negative-ion MALDI-TOF mass spectrometry spectra of lipid A purified from K. pneumoniae strains 52.145 (Kp52145) (A), 52145-ΔlpxL1 (lpxL1) (B), 52145-ΔlpxL2 (lpxL2) (C), 52145-ΔlpxL1ΔlpxL2 (lpxL1-lpxL2) (D), 52145-ΔlpxL2Com2 (lpxL2Com) (E), and 52145-ΔlpxL1ΔlpxL2Com2 (lpxL1-lpxL2Com) (F) are shown. The data represent the mass-to-charge ratio (m/z) of each lipid A species detected and are representative of those from three extractions.
FIG 3
FIG 3
Deletion of lpxL2 reduces the transcription of phoPQ and pmrAB. (A and B) Activity of the pmrH (A) and pagP (B) promoters in K. pneumoniae 52.145 (Kp52145), 52145-ΔlpxL2 (lpxL2), and 52145-ΔlpxL2Com2 (lpxL2Com) carrying lucFF transcriptional fusions. Values (expressed in RLU) are presented as the mean ± SD from three independent experiments measured in quintuplicate. (C and D) The transcription levels of phoP (C) and pmrA (D) in K. pneumoniae 52.145 (Kp52145), 52145-ΔlpxL2 (lpxL2), and 52145-ΔlpxL2Com2 (lpxL2Com) were determined by RT-qPCR and are shown relative to the expression levels in wild-type bacteria (white bars). Results represent means ± SDs. P values indicate the significance of the differences versus Kp52145 determined using one-way ANOVA with Bonferroni contrasts. **, P < 0.01; ***, P < 0.001; n.s., not significant.
FIG 4
FIG 4
K. pneumoniae LpxL2 acylates the 2′ R-3-hydroxymyristoyl group with C12. The negative-ion MALDI-TOF mass spectrometry spectra of lipid A purified from E. coli BN1 ΔlpxL complemented with K. pneumoniae lpxL2 (BN1 ΔlpxL/pGEMTlpxL2Com) (A), 52145-ΔlpxL1ΔlpxL2 complemented with K. pneumoniae lpxL1 cloned into pGEM-T Easy plasmid (ΔlpxL1ΔlpxL2/pGEMTlpxL1Com) (B), 52145-ΔlpxL1ΔlpxL2 complemented with K. pneumoniae lpxL2 cloned into the Tn7 transposon (ΔlpxL1ΔlpxL2::Tn7lpxL1) (C), E. coli BN1 ΔlpxL complemented with K. pneumoniae lpxL1 cloned into the Tn7 transposon (BN1 ΔlpxL::Tn7lpxL1) (D) are shown. Data represent the m/z of each lipid A species detected and are representative of those from three extractions.
FIG 5
FIG 5
PhoPQ negatively regulates K. pneumoniae lpxL1. (A) Activity of the lpxL1 and lpxL2 promoters in K. pneumoniae 52.145 (Kp52145) carrying lucFF transcriptional fusions. Values (expressed in RLU) are presented as the means ± SDs from three independent experiments measured in quintuplicate. (B) The transcription levels of lpxL1 and lpxL2 in K. pneumoniae Kp52145 were determined by RT-qPCR and are shown relative to the expression levels in wild-type bacteria. The results represent means ± SDs. (C) Activity of the lpxL1 promoter in K. pneumoniae Kp52145, 52145-ΔphoQGB (phoQ), and 52145-ΔphoQGBCom (phoQCom) carrying lucFF transcriptional fusions. Values (expressed in RLU) are presented as the means ± SDs from three independent experiments measured in quintuplicate. (D) The transcription levels of lpxL1 in K. pneumoniae Kp52145, 52145-ΔphoQGB (phoQ), and 52145-ΔphoQGBCom (phoQCom) were determined by RT-qPCR and are shown relative to the expression levels in wild-type bacteria. Results represent means ± SDs. (E to G) Negative-ion MALDI-TOF mass spectrometry spectra of lipid A purified from 52145-ΔphoQGB (phoQ) (E), 52145-ΔphoQGBCom (phoQCom) (F), and 52145-ΔlpxL1ΔphoQGB (lpxL1-phoQ) (G). Data represent the m/z of each lipid A species detected and are representative of those from three extractions. P values indicate the significance of the differences versus Kp52145 or between the indicated comparisons determined using one-way ANOVA with Bonferroni contrasts. *, P < 0.05; **, P < 0.01; ***, P < 0.001; n.s., not significant.
FIG 6
FIG 6
Deletion of lpxL2 decreases K. pneumoniae resistance to polymyxins. The percent survival of K. pneumoniae 52.145 (Kp52145), 52145-ΔlpxL2 (lpxL2), 52145-ΔlpxL2Com2 (lpxL2Com), 52145-ΔlpxO (lpxO), and 52145-ΔlpxOCom (lpxOCom) following 1 h of exposure to polymyxin B (A) and colistin (B) is shown. Values are presented as the means ± SD from three independent experiments measured in duplicate. P values indicate the significance of the differences versus Kp52145 determined using one-way ANOVA with Bonferroni contrasts. **, P < 0.01; n.s., not significant.
FIG 7
FIG 7
Deletion of lpxL2 increases human phagocyte-mediated killing of K. pneumoniae. Three hundred microliters of fresh human blood (from three different donors) was mixed with 1 × 107 CFU of K. pneumoniae 52.145 (Kp52145), 52145-ΔlpxL2 (lpxL2), 52145-ΔlpxL2Com2 (lpxL2Com), 52145-ΔlpxO (lpxO), and 52145-ΔlpxOCom (lpxOCom) and incubated at 37°C for 3 h. The bacterial counts recovered were then divided by the initial counts. Experiments were performed with duplicate samples on three independent occasions. P values indicate the significance of the differences versus Kp52145 determined using one-way ANOVA with Bonferroni contrasts. **, P < 0.01; n.s., not significant.
FIG 8
FIG 8
lpxL2 deletion results in the upregulation of inflammatory responses in macrophages upon infection. The levels of TNF-α (A), IL-6 (B), RANTES (C), and MCP-1 (D) secretion by iBMDM macrophages stimulated for 6 h with UV-killed K. pneumoniae 52.145 (Kp52145), 52145-ΔlpxL2 (lpxL2), 52145-ΔlpxL2Com2 (lpxL2Com), 52145-ΔlpxO (lpxO), and 52145-ΔlpxOCom (lpxOCom) are shown. P values indicate the significance of the differences versus Kp52145 determined using one-way ANOVA with Bonferroni contrasts. *, P < 0.05; **, P < 0.01; n.s., not significant. CON, uninfected cells.
FIG 9
FIG 9
The K. pneumoniae lpxL2 mutant displays decreased virulence in the G. mellonella wax worm infection model. The percent survival of G. mellonella over 72 h postinfection with 105 organisms of K. pneumoniae 52.145 (Kp52145), 52145-ΔlpxL1 (lpxL1), 52145-ΔlpxL2 (lpxL2), 52145-ΔlpxL1ΔlpxL2 (lpxL1-lpxL2), 52145-ΔlpxO (lpxO), 52145-ΔlpxL1ΔlpxL2Com2 (lpxL1-lpxL2Com2), 52145-ΔlpxOCom (lpxOCom), and 52145-ΔlpxL2Com2 (lpxL2Com2) is shown. Thirty larvae were infected in each group. The level of significance was determined using the log-rank (Mantel-Cox) test with the Bonferroni correction for multiple comparisons.

Similar articles

See all similar articles

Cited by 9 articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources

Feedback