D-galactan II is an immunodominant antigen in O1 lipopolysaccharide and affects virulence in Klebsiella pneumoniae: implication in vaccine design

Abstract

In the O1 strain of Klebsiella, the lipopolysaccharide (LPS) O-antigen is composed of D-galactan I and D-galactan II. Although the composition of the O1 antigen of Klebsiella was resolved more than two decades, the genetic locus involved in the biosynthesis of D-galactan II and the role of D-galactan II in bacterial pathogenesis remain unclear. Here, we report the identification of the D-galactan II-synthesizing genes by screening a transposon mutant library of an acapsulated Klebsiella pneumoniae O1 strain with bacteriophage. K. pneumoniae strain deleted for wbbY exhibited abrogated D-galactan II production; altered serum resistance and attenuation of virulence. Serologic analysis of K. pneumoniae clinical isolates demonstrated that D-galactan II was more prevalent in community-acquired pyogenic liver abscess (PLA)-causing strains than in non-tissue-invasive strains. WbbY homologs, WbbZ homologs, and lipopolysaccharide structures based on D-galactan II also were present in several Gram-negative bacteria. Immunization of mice with the magA-mutant (K(-) 1 O1) (that is, with a LPS D-galactan II-producing strain) provided protection against infection with an O1:K2 PLA strain. Our findings indicate that both WbbY and WbbZ homologs are sufficient for the synthesis of D-galactan II. D-galactan II represents an immunodominant antigen; is conserved among multiple species of Gram-negative bacteria and could be a useful vaccine candidate.

Keywords: D-galactan II; Klebsiella pneumoniae; immunodominant antigen; lipopolysaccharide; vaccine.

Figure 1

The K. pneumoniae O1 mutants. (A) The K. pneumoniae NTUH-K2044 wild-type, magA deletion, wbbO deletion, magA wbbO double-deletion mutants and magA deletion transposon mutant strains, 16-1C and 17-9F, were grown to mid-log phase at 37°C and spread on LB plates. Serial dilutions of SM buffer containing phage O1-1 were dotted onto the bacterial lawn and grown at 37°C for 16–18 h. PFU, plaque forming unit. (B) Transposon insertion sites within the wb cluster in the NTUH-K2044 magA mutants. Arrows denote the orientations of the ORFs. (C) The composition of K. pneumoniae O1 LPS and the gene responsible for each step. (D) Genetic organization of wbbY, wbbZ (or their homologs), and their flanking regions in K. pneumoniae NTUH-K2044 (upper) and E. coli SMS-3-5 (lower). Arrows denote the orientations of the ORFs.

Figure 2

Genetic construction, growth, and CPS production of K. pneumoniae NTUH-K2044 and its isogenic mutants. (A) Mapping of transcription start sites of the wbbY and wbbZ genes. The promoter regions and the proposed −10 and −35 regions are indicated by gray shading. The transcription start sites are shown by +1 with arrowheads and in larger font. Deduced amino acid sequences (in single-letter code) are indicated beginning with an ATG start codon. (B) Schematic diagram representing the construction of deletion mutations in the NTUH-K2044 background. The locations and orientations of the ORFs described in this study are indicated by arrows. The dashed lines refer to joining of the regions flanking the deleted target. (C) Mutation of wbbY or wbbZ does not impair bacterial growth. Overnight cultures of the K. pneumoniae NTUH-K2044 wild-type strain, wbbY, wbbZ, and wbbZ-NP (non-polar) mutant strains were individually inoculated into fresh LB medium and grown at 37°C. The growth of bacteria was monitored periodically every hour by plating of serial dilutions on LB agar and counting CFU. The data represent the means of three independent trials; the error bars represent the standard deviations. (D) The amount of K1 CPS production of the K. pneumoniae NTUH-K2044 wild-type, and of the wbbY, wbbZ, wbbZ-NP, and magA single-mutant strains, was determined by assaying uronic acid content. The data represent the means of three independent trials; the error bars represent the standard deviations. ^**P < 0.01 by Student's t-test (compared to the wild-type strain); other comparisons were not statistically significant (P ≥ 0.05).

Figure 3

The LPS profiles of K. pneumoniae NTUH-K2044 and its isogenic mutants. (A) EPS specimens were prepared from K. pneumoniae NTUH-K2044, from wbbY and wbbZ single mutants, and from the same mutants harboring complementation plasmids. Extracts from normalized bacterial suspensions (10⁸ CFU) were separated by SDS-PAGE and visualized by silver staining (upper) or immunoblotting with anti-D-Gal II antiserum (lower). HMW-LPS, high-molecular-weight LPS. LMW-LPS, low-molecular-weight LPS. (B) EPS specimens were prepared from K. pneumoniae NTUH-K2044, from wbbZ and wbbZ-NP single mutants, and from mutants harboring complementation plasmids. Extracts from normalized bacterial suspensions (10⁸ CFU) were separated by SDS-PAGE and visualized by silver staining (upper) or immunoblotting with anti-D-Gal II antiserum (lower). (C,D) EPS extracted from Klebsiella K72:O2, E. coli DH5α, and K. pneumoniae wbbY mutant strains harboring the indicated plasmids were separated by SDS-PAGE and the LPS patterns were visualized by silver staining (upper) or immunoblotting with anti-D-Gal II antiserum (lower). (E) The ¹H-NMR spectra of the LPS of the wbbY mutant. LPS specimens were prepared from the Klebsiella K72:O2 capsule-deficient mutant (K72Δwza wzb), the K. pneumoniae NTUH-K2044 magA deletion strain, the magA wbbY double-deletion mutants, and the magA wbbY double-deletion strain containing the wbbY-expressing plasmid. The four D-galactosyl residues are indicated (a to d) according to decreasing anomeric ¹H shifts.

Figure 4

Analysis of WbbY and WbbZ homologs in K. pneumoniae and other Gram-negative bacteria. (A) PCR detection of wbbY (830 bp) and wbbZ (583 bp)—like genes. lane M, 1 kb DNA ladder; NC, no-template control. (B) EPS specimens were prepared from K. pneumoniae NTUH-K2044; isogenic mutants of K2044; strain A5054; the Klebsiella K72 wild type; K72 wza wzb double mutant; and E. coli O19 ab K⁻:H7 F8188-41. Extracts from normalized bacterial suspensions (10⁸ CFU) were separated by SDS-PAGE and visualized by immunoblotting with anti-D-Gal II antiserum. (C) EPS extracted from the wbbY and wbbZ single-mutant and Klebsiella K72:O2 strains harboring the indicated plasmid was separated by SDS-PAGE and the LPS patterns were analyzed by immunoblotting with anti-D-Gal II antiserum. (D) Phylogenetic analysis of WbbY and WbbZ homologs from different Gram-negative bacteria using the SMART (Simple Modular Architecture Research Tool) database.

Figure 5

Serum sensitivity and virulence of NTUH-K2044 and its isogenic mutants. (A–C) Serum sensitivity assays of resistance to killing by non-immune healthy human serum of the NTUH-K2044 wild-type; wbbY, wbbZ, wbbZ-NP, magA, and wbbO single-deletion mutants; and the respective complemented strains. The data represent the means of three independent trials; the error bars represent the standard deviations. A mean survival ratio ≥1 corresponds to serum resistance. ^**P < 0.01 or ^*P < 0.05 by Student's t test (compared to the wild-type strain). (D) Eight mice per group were infected with the NTUH-K2044 wild-type and the wbbY mutant strains at an intraperitoneal (IP) dose of 1 × 10² CFU/animal. Survival of mice was monitored for 4 weeks. ■, NTUH-K2044; △, wbbY mutant (wbbY vs. parent, P = 0.024; log-rank test). (E) Survival of the isogenic K. pneumoniae wbbY (open triangles) mutant in the in vivo bacterial competition model, using competition against the fully virulent placZ mutant in groups of eight mice. The ratio of lacZ⁺ or lacZ⁻ K. pneumoniae (close squares) colonies in the contents of the spleen and liver was determined from a single mouse at sacrifice (24 h post-infection). Each symbol represents the competitive index (CI) for each inoculum; and the medians and SDs of the values are shown (for the wild-type strain group vs. placZ mutant group, the CI of spleen and liver were 1.016 ± 0.206 and 1.009 ± 0.286, respectively; wbbY mutant group vs. placZ mutant group, the CI of spleen and liver were 0.553 ± 0.089 and 0.564 ± 0.123, respectively; P = 0.001, Wilcoxon's signed rank test). (F) Survival of mice immunized with magA mutant and then challenged with NTUH-A4528 (O1:K2). Mice (10 per group) were inoculated three times by once-weekly IP injections with 1 × 10⁶ CFU of magA mutant or magA wbbY double-mutant strain. Age-matched, unimmunized control mice were inoculated with saline. On the fourth week, immunized or unimmunized groups were challenged with NTUH-A4528 (1 × 10³ CFU per animal, IP). Survival was assessed for 28 days following infection. ▲, non-immunized, NTUH-A4528 challenged; △, magA mutant immunized, NTUH-A4528 challenged; ◊, magA wbbY double mutant immunized, NTUH-A4528 challenged. By log-rank test: P < 0.0001, magA/NTUH-A4528 vs. non-immune/NTUH-A4528 or magA wbbY/NTUH-A4528. (G) Immune response of anti-D-Gal II IgG in magA mutant-immunized mice. Immunoblots developed with magA mutant immune mouse serum (1:500) or magA wbbY double-mutant immune mouse serum (1:500). (H) Anti-D-Gal II specific antibodies reduce growth of NTUH-A4528 (O1:K2) K. pneumoniae and E. coli F8188-41 (O19) in mice. Groups of 4 BALB/c mice were treated with magA mutant immune mouse serum (IMS–positive control, IMS-PC) or with magA wbbY double-mutant immune mouse serum (IMS–negative control, IMS-NC). One hour after injection, mice were infected with 1 × 10³ CFU of the NTUH-A4528 (KP-IMS-PC and KP-IMS-NC) or 1 × 10⁶ CFU of E. coli F8188-41 (EC-IMS-PC and EC-IMS-NC) by IP injection. Three hours after infection, the numbers of bacteria in the liver and spleen were determined. Log₁₀ CFU was standardized per 0.1 gram wet organ weight. The black and white bars represent the means for each group for liver and spleen, respectively. Data are presented as means ± SD. ^**P < 0.01 or ^*P < 0.05 by Student's t test (KP-IMS-PC vs. KP-IMS-NC and EC-IMS-PC vs. EC-IMS-NC).