OUP user menu

The wbiA locus is required for the 2-O-acetylation of lipopolysaccharides expressed by Burkholderia pseudomallei and Burkholderia thailandensis

Paul J. Brett, Mary N. Burtnick, Donald E. Woods
DOI: http://dx.doi.org/10.1111/j.1574-6968.2003.tb11536.x 323-328 First published online: 1 January 2003


Burkholderia pseudomallei and Burkholderia thailandensis express similar O-antigens (O-PS II) in which their 6-deoxy-α-l-talopyranosyl (l-6dTalp) residues are variably substituted with O-acetyl groups at the O-2 or O-4 positions. In previous studies we demonstrated that the protective monoclonal antibody, Pp-PS-W, reacted with O-PS II expressed by wild-type B. pseudomallei strains but not by a B. pseudomallei wbiA null mutant. In the present study we demonstrate that WbiA activity is required for the acetylation of the l-6dTalp residues at the O-2 position and that structural modification of O-PS II molecules at this site is critical for recognition by Pp-PS-W.

  • Burkholderia species
  • O-antigen
  • Virulence determinant
  • trans-Acylase

1 Introduction

Burkholderia pseudomallei, the etiologic agent of melioidosis, is a Gram-negative bacterial pathogen responsible for disease in both humans and animals [1,2]. Previous studies have demonstrated that the lipopolysaccharide (LPS) expressed by B. pseudomallei is both a virulence determinant and a protective antigen [36]. Consequently, the O-antigen (O-PS II) has become a significant component of the various sub-unit vaccine candidates that we are currently developing for immunization against melioidosis [7].

The O-PS II moiety produced by B. pseudomallei is an unbranched heteropolymer consisting of disaccharide repeats having the structure 3)-β-d-glucopyranose-(1→3)-6-deoxy-α-l-talopyranose-(1→ in which ∼33% of the 6-deoxy-α-l-talopyranose (l-6dTalp) residues possess 2-O-methyl and 4-O-acetyl substitutions while the remainder of the l-6dTalp residues bear only 2-O-acetyl modifications [8,9]. Studies have also demonstrated that the non-pathogenic species Burkholderia thailandensis synthesizes an O-antigen with the same repeating unit [10]. Recently, we demonstrated that the O-antigen (O-PS) expressed by Burkholderia mallei, the causative agent of glanders, is virtually identical to O-PS II except that it lacks acetyl modifications at the O-4 position of the l-6dTalp residues [11]. Curiously, however, pairwise comparisons between the B. mallei and B. pseudomallei O-polysaccharide biosynthetic clusters failed to reveal any sequence differences that could account for the structural dissimilarities observed between O-PS and O-PS II [5,11].

In the current study, we used a combination of molecular and physical approaches to further characterize the role of the wbiA locus which is thought to be involved in the acetylation of O-PS II antigens [5].

2 Materials and methods

2.1 Strains, plasmids and growth conditions

The bacterial strains and plasmids used in this study are described in Table 1. Escherichia coli, B. pseudomallei and B. thailandensis strains were grown at 37°C in Luria–Bertani (LB) broth or on LB agar. B. mallei strains were grown at 37°C in LB broth containing 4% glycerol or on LB agar containing 4% glycerol. For E. coli, antibiotics were used at the following concentrations: ampicillin (Ap) 100 µg ml−1, gentamicin (Gm) 15 µg ml−1 and kanamycin (Km) 25 µg ml−1. For B. pseudomallei and B. thailandensis, antibiotics were used at the following concentrations: Gm 25 µg ml−1, streptomycin (Sm) 100 µg ml−1 and trimethoprim (Tp) 100 µg ml−1. Bacterial strains were maintained at −70°C in 20% glycerol suspensions.

View this table:
Table 1

Bacterial strains and plasmids used in this study

Strains and plasmidsRelevant characteristic(s)Reference or source
E. coli
SM10Mobilizing strain: expresses RP4 tra genes; Kmr Sms[21]
TOP10High efficiency transformationInvitrogen
B. pseudomallei
1026bClinical isolate: Gmr Kmr Smr Pmr Tps[5]
DD5031026b derivative: Δ(amrR-oprA) rpsL; Smr Pmr Gms Kms Tps[5]
PB604DD503 derivative: wbiA::dhfrIIb-p15A oriV; Tpr[5]
PB605PB604 (pUCP31T); Gmr TprThis study
PB606PB604 (p31wbiA); Gmr TprThis study
B. thailandensis
ATCC 700338Type strain (soil isolate): Gmr Kmr Smr Pmr Tps[10]
DW503ATCC 700338 derivative: rpsL; Smr Pmr Gms Kms Tps[22]
BT604DW503 derivative: wbiA::dhfrIIb-p15A oriV; TprThis study
BT605BT604 (pUCP31T); Gmr TprThis study
BT606BT604 (p31wbiA); Gmr TprThis study
B. mallei
ATCC 23344Type strain (human isolate)USAMRIID
pCR2.1-TOPOTA cloning vector: ColE1 ori; Apr KmrInvitrogen
pUCP31TBroad host range vector: OriT pRO1600 ori; Gmr[23]
p31wbiA1.37-kb B. pseudomallei wbiA PCR product cloned into the Xba I/Kpn I sites of pUCP31T; GmrThis study
  • United States Army Medical Research Institute of Infectious Diseases, Fort Detrick, MD, USA.

2.2 DNA manipulations and transformations

Molecular and cloning techniques were performed essentially as described by Sambrook et al. [12]. Plasmids were purified using QIAprep spin plasmid minipreps (Qiagen). Genomic DNA was isolated using the Wizard™ Genomic DNA Isolation kit (Promega). Competent E. coli were transformed using standard methods.

2.3 PCR amplification and sequence analysis of wbiA genes

The wbiA genes from B. thailandensis ATCC 700388 and B. pseudomallei 1026b were PCR amplified from purified chromosomal DNA samples using the wbiA-5′ (5′-GCTCTAGACATGAGATCGTGCTTGAGCG-3′) and wbiA-3′ (5′-GGGGTACCGATAAAGCCAGCCCCACCGG-3′) primer pair; the Xba I and Kpn I sites in the linker regions are underlined. The primers were designed at the 3′-end of wzt and the 5′-end of wbiB using the previously described B. pseudomallei O-PS II biosynthetic gene cluster (GenBank database accession number AF064070). Reactions were performed using Taq polymerase (Invitrogen) as per manufacturer's instructions except that the denaturing temperature was increased to 97°C to compensate for the high G/C content of the chromosomal DNAs. The resulting PCR products were then cloned into pCR2.1-TOPO and sequenced on both strands. Sequence analyses were conducted with the aid of DNASIS version 2.5 (Hitachi) as well as the BLASTX and BLASTP programs [13]. The Shigella flexneri bacteriophage SF6 oac GenBank accession number is X56800. The B. thailandensis nucleotide sequence reported in this study was entered into the GenBank database under accession number AY028370.

2.4 Construction and complementation of wbiA mutants

B. pseudomallei PB604, a strain harboring an insertionally inactivated wbiA gene, was previously constructed by DeShazer et al. [5]. The wbiA gene of B. thailandensis was insertionally inactivated using the allelic exchange vector pPB604Tp resulting in strain BT604. Allelic exchange was performed as previously described [5,14]. Mutants were complemented in trans using the broad host range vector pUCP31T harboring a wild-type copy of the B. pseudomallei wbiA locus. Plasmids were conjugated to B. pseudomallei and B. thailandensis as previously described [15].

2.5 Western blot and silver stain analysis

Whole cell lysates were prepared as previously described [16] and used in both Western immunoblot and silver stain analyses. Overnight bacterial cultures were pelleted, resuspended in lysis buffer and boiled prior to SDS–PAGE analysis on 12% gels. Immunoblots were performed as previously described [17] using rabbit polyclonal antisera specific for B. pseudomallei O-PS II. Silver stain analyses were performed as previously described [18].

2.6 Purification of LPS and O-PS

LPS was purified using a previously described hot aqueous phenol extraction protocol [7,9]. Delipidation of the LPS molecules was achieved via mild acid hydrolysis (2% acetic acid) followed by size exclusion chromatography (Sephadex G-50) as previously described by Perry et al. [9]. Carbohydrate positive fractions were detected using a phenol–sulfuric acid assay [19]. The purity of the carbohydrate preparations was determined to be >90% in all instances. Protein contamination was determined using bicinchoninic acid assays (Pierce) while nucleic acid contamination was estimated from OD260/280 measurements.

2.7 Nuclear magnetic resonance (NMR) spectroscopy analysis

13C-NMR spectra were recorded at 100.5 MHz and the chemical shifts were recorded in ppm relative to an internal acetone standard (31.07 ppm [13C]; Spectral Data Services, Champaign, IL, USA).

3 Results and discussion

3.1 Comparison of wbiA alleles from B. thailandensis and B. pseudomallei

The wbiA allele from B. thailandensis ATCC 700388 was cloned and sequenced as described in Section 2. Analysis of the 1239-bp open reading frame contained within the cloned PCR product demonstrated sequence identities of 93.6% at the nucleotide and 95.0% at the amino acid levels in comparison to the previously characterized B. pseudomallei 1026b wbiA allele (Fig. 1). Based upon these preliminary results we predicted that the function of WbiA would be similar in both B. pseudomallei and B. thailandensis.

Figure 1

Amino acid alignment of B. pseudomallei 1026b (Bp) wbiA, B. thailandensis ATCC 700388 (Bt) wbiA and S. flexneri phage SF6 (Sf) oac gene products. Shaded residues represent identity amongst the aligned sequences. Dots indicate dissimilarities between the Bp and Bt proteins. Asterisks indicate residues conserved amongst members of the family of integral membrane proteins involved in the acylation of exported carbohydrates.

Further analysis of the wbiA gene products expressed by the two Burkholderia species demonstrated the presence of conserved amino acid motifs that defines a family of inner membrane trans-acylases. The structural and functional significance of these motifs, however, has yet to be determined. The family includes Salmonella typhimurium OafA, Shigella flexneri bacteriophage SF6 Oac, Rhizobium meliloti ExoZ and Legionella pneumophila Lag1 [20]. Interestingly, all are involved in the acetylation of bacterial polysaccharides [20]. A gapped sequence alignment of the WbiA homologues with the Oac trans-acylase revealed overall sequence identities of approximately 30% (Fig. 1), a result that is consistent with the family in general.

3.2 Phenotypic characterization of wbiA null mutants

To determine the effect of the wbiA null mutations on the synthesis of O-PS II, B. pseudomallei PB604 and B. thailandensis BT604 were phenotypically characterized using a variety of genetic and immunological approaches. Silver staining of SDS–PAGE fractionated whole cell lysates demonstrated that BT604 was capable of expressing full-length LPS molecules based upon the presence of a characteristic LPS banding pattern (data not shown). The LPS was also shown to be immunologically similar to that expressed by the type strain and DW503 due to the reactivity of the antigen with the O-PS II polyclonal antiserum (data not shown). Interestingly, however, neither the BT604 whole cell lysates nor the purified LPS molecules reacted with the O-PS II specific monoclonal antibody (mAb) Pp-PS-W suggesting that the wbiA locus was required for the expression of a native O-PS II moieties (Fig. 2). By complementing BT604 with the broad host range vector, p31wbiA, we were able to restore the reactivity of the whole cell lysates and purified LPS with the Pp-PS-W mAb (Fig. 2). Similar results were observed for the B. pseudomallei strains (data not shown).

Figure 2

Western immunoblot analysis of purified B. thailandensis LPS antigens. The primary antibody used was the O-PS II specific Pp-PS-W mAb. Lane 1, DW503 LPS; lane 2, BT605 LPS; lane 3, BT606 LPS.

3.3 Spectroscopic analysis of the O-polysaccharide antigens

The O-polysaccharides from B. thailandensis DW503, BT604 and BT606, B. pseudomallei DD503, PB604 and PB606 and B. mallei ATCC 23344 were isolated and purified as described in Section 2. The 13C-NMR spectrum of the DW503 antigen demonstrated four anomeric carbon signals between 98.5 and 102.6 ppm, two O-acetyl signals at 174.1 and 174.6 ppm (CH3C O) as well as 21.2 and 21.4 ppm (C H3CO), two 6-deoxyhexose CH3 signals at 16.0 and 16.2 ppm and an O-methyl signal at 58.8 ppm (Fig. 3A), all of which are consistent with previously published values [9]. Similar spectra were also obtained for the BT606, DD503 and PB606 samples (data not shown). In contrast, the 13C-NMR spectrum of the BT604 sample demonstrated four anomeric carbon signals between 98.5 and 102.2 ppm, one O-acetyl signal at 174.6 ppm (CH3C O) and 21.2 (C H3CO), two 6-deoxyhexose CH3 signals at 16.0 and 16.3 ppm and an O-methyl signal at 58.8 ppm (Fig. 3B). A similar spectrum was recorded for the PB604 sample (data not shown). Based upon these results it was apparent that the O-polysaccharides expressed by BT604 and PB604 were lacking one of the two O-acetyl moieties associated with native O-PS II molecules.

Figure 3

13C-NMR spectra of native and mutant O-polysaccharides expressed by B. thailandensis strains (A) DW503 and (B) BT604.

To determine which of the O-acetyl groups was missing a comparison of the DW503 and BT604 13C-NMR spectra with the 13C-NMR spectrum obtained for B. mallei ATCC 23344 O-PS was conducted. Based upon an analysis of the spectral data we were able to establish that BT604 lacks O-acetyl modifications at the O-2 position of the l-6dTalp residues since O-polysaccharides lacking O-acetyl substitutions only at the O-4 position would have produced spectra consistent with that obtained for B. mallei O-PS (Fig. 4). Similar conclusions can also be drawn for B. pseudomallei PB604. Based upon these observations, it is highly probable that a second unlinked locus is responsible for the O-acetylation of l-6dTalp residues at the O-4 position since the wbiA locus is the only predicted trans-acylase in the O-PS II biosynthetic operon. Studies are currently under way to examine this hypothesis.

Figure 4

13C-NMR spectra of B. thailandensis and B. mallei O-polysaccharides expanded between the region of 15–25 ppm. (A) DW503, (B) BT604 and (C) ATCC 23344. The peaks around 16 ppm represent 6-deoxyhexose CH3 signals while those around 21 ppm represent O-acetyl (C H3CO) signals.

3.4 Characterization of the epitope recognized by the Pp-PS-W mAb

We have recently demonstrated that the O-PS antigen expressed by B. mallei does not react with Pp-PS-W [11]. A comparison of the O-antigens expressed by B. pseudomallei and B. thailandensis with those expressed by B. mallei strains suggested that this phenomenon was likely due to differences in the O-acetylation patterns exhibited by the O-PS and O-PS II molecules (Fig. 5). Based upon the results of the current study, it is now apparent that the mAb reacts only with 3)-β-d-glucopyranose-(1→3)-6-deoxy-α-l-talopyranose-(1→ polymers in which the l-6dTalp residues are coordinately acetylated at the O-2 and O-4 positions. Whether or not the 2-O-acetyl modification imposes conformational constraints upon the O-polysaccharides or serves more directly as a structural epitope remains yet to be determined. Needless to say, however, these observations have proven to be a valuable reminder of the importance of maintaining the structural integrity of O-PS II during the synthesis of the glycoconjugate vaccine candidates.

Figure 5

Structures of (A) B. pseudomallei O-PS II and (B) B. mallei O-PS. In B. pseudomallei R′=O-methyl or O-acetyl and R″=O-acetyl or OH. In B. mallei R′=O-methyl or O-acetyl.


This work was funded by the Department of Defense Contract No. DAMD 17-98-C-8003 and CIHR MOP 31343. M.N.B. was the recipient of an Alberta Heritage Foundation for Medical Research Studentship Award.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
View Abstract