The wxacO gene of Xanthomonas citri ssp. citri encodes a protein with a role in lipopolysaccharide biosynthesis, biofilm formation, stress tolerance and virulence

Abstract

Xanthomonas citri ssp. citri (Xcc) causes citrus canker, one of the most economically damaging diseases affecting citrus worldwide. Biofilm formation is important for the pathogen to survive epiphytically in planta prior to the induction of canker symptoms. In this study, two EZ-Tn5 transposon mutants of Xcc strain 306, affected in biofilm formation, were isolated; subsequent analyses led to the identification of a novel gene locus XAC3596 (designated as wxacO), encoding a putative transmembrane protein, and the rfbC gene, encoding a truncated O-antigen biosynthesis protein. Sodium dodecylsulphate-polyacrylamide gel electrophoresis revealed that lipopolysaccharide (LPS) biosynthesis was affected in both wxacO and rfbC mutants. The wxacO mutant was impaired in the formation of a structured biofilm on glass or host plant leaves, as shown in confocal laser scanning microscopy analysis of strains containing a plasmid expressing the green fluorescent protein. Both wxacO and rfbC mutants were more sensitive than the wild-type strain to different environmental stresses, and more susceptible to the antimicrobial peptide polymyxin B. The two mutants were attenuated in swimming motility, but not in flagellar formation. The mutants also showed reduced virulence and decreased growth on host leaves when spray inoculated. The affected phenotypes of the wxacO and rfbC mutants were complemented to wild-type levels by the intact wxacO and rfbC genes, respectively. This report identifies a new gene influencing LPS production by Xcc. In addition, our results suggest that a structurally intact LPS is critical for survival in the phyllosphere and for the virulence of Xcc.

Figure 1

Ability of Xanthomonas citri ssp. citri strains to form biofilms in borosilicate glass tubes as determined using crystal violet staining. The experiment was repeated three times with five replicates each time. Averages and standard errors from one experiment of three with similar results are shown. Wt, X. citri ssp. citri wild‐type strain 306; 247C8(wxacO), wxacO mutant; 247C8V, 247C8 (wxacO) complemented with empty vector pUFR053 without the wxacO gene; C247C8, strain 247C8 (wxacO) complemented with pJU3596; 236E9(rfbC), rfbC mutant; 236E9V, 236E9 (rfbC) complemented with empty vector pUFR053 without the rfbC gene; C236E9, strain 236E9 (rfbC) complemented with pJU3598; CK^‐, nutrient broth (NB) medium without inoculation of bacteria. A _590 nm, absorbance at 590 nm.

Figure 2

Sequence analysis of EZ‐Tn5 insertions in wxacO (XAC3596) and rfbC mutants. (A) Genetic organization of the putative lipopolysaccharide (LPS) gene cluster of Xanthomonas citri ssp. citri strain 306 and transposon insertion sites in the wxacO and rfbC mutants. The length of each arrow represents the relative open reading frame (ORF) size and indicates the direction of transcription. The triangles indicate the Tn5 insertion sites. Gene colour represents operon membership. The annotation information and sizes of the genes were obtained from the genome sequence database of Xcc strain 306 [National Center for Biotechnology Information (NCBI) Accession No: AE008923]. (B) Domain structure analyses of the putative WxacO and RfbC proteins. The domain structure prediction was performed using the smart program via the ExPASy Proteomics Server http://www.ca.expasy.org. Domain symbols: Glycos_transf_2, glycosyl transferase family 2 domain; T, transmembrane domain.

Figure 3

Sodium dodecylsulphate‐polyacrylamide gel electrophoresis (SDS‐PAGE) analysis of lipopolysaccharide (LPS) from proteinase K‐digested whole‐cell lysates of Xanthomonas citri ssp. citri strain 306 and its mutants carrying EZ‐Tn5 insertions in the wxacO and rfbC genes. Both wxacO and rfbC mutants were affected in the biosynthesis of O‐antigen. The wild‐type (Wt) LPS profile was restored in the complemented strains. S, LPS standard from Salmonella typhimurium (12.5 µg; Sigma). Lanes: 1, wild‐type strain 306; 2, C247C8 (wxacO mutant complemented with pJU3596); 3, 247C8 (wxacO) (wxacO mutant); 4: 247C8V (wxacO mutant complemented with empty vector pUFR053 without the wxacO gene); 5, 236E9 (rfbC) (rfbC mutant); 6, 236E9V (rfbC mutant complemented with empty vector pUFR053 without the rfbC gene); 7, C236E9 (rfbC mutant complemented with pJU3598). The gel was run in the SDS–glycine buffer system and silver stained using a silver stain kit (Bio‐Rad Laboratories, Inc.) following the manufacturer's instructions. The lost bands in the mutants are indicated by arrows.

Figure 4

Confocal laser scanning microscopy (CLSM) of biofilms formed by wild‐type and wxacO mutant of Xanthomonas citri ssp. citri strain 306. (A) Green fluorescent protein (GFP)‐labelled cells grown on the glass bottoms of culture dishes were visualized at different stages of biofilm formation, 4 days post‐inoculation. The panels (left to right) show cell attachment to the glass surface. Scale bars, 75 µm. (B) Biofilm formation on grapefruit leaves. Bacterial suspensions were inoculated onto the abaxial leaf surface by spray. Scale bars, 34.78 µm. WT, wild‐type strain 306; 247C8 (wxacO), wxacO mutant.

Figure 5

Virulence assays and growth of Xanthomonas citri ssp. citri strain 306 and derivative strains in planta. Bacterial strains were inoculated onto the abaxial leaf surfaces by spray. (A) Disease symptoms on grapefruit leaves inoculated with X. citri ssp. citri wild‐type strain 306 (Wt), wxacO mutant [247C8(wxacO)], 247C8 (wxacO) complemented with empty vector pUFR053 without the wxacO gene (247C8V), 247C8 (wxacO) complemented with pJU3596 (C247C8), rfbC mutant [236E9(rfbC)], 236E9 (rfbC) complemented with empty vector pUFR053 without the rfbC gene (236E9V), 236E9 (rfbC) complemented with pJU3598 (C236E9) and water control (WC). The inoculated leaves were photographed at 14 days after spraying. (B) Growth of X. citri ssp. citri wild‐type strain 306 (Wt), wxacO mutant [247C8 (wxacO)] and 247C8(wxacO) complemented with pJU3596 (C247C8) on grapefruit leaves. (C) Growth of X. citri ssp. citri wild‐type strain 306 (Wt), rfbC mutant [236E9 (rfbC)] and 236E9 (rfbC) complemented with pJU3598 (C236E9) on grapefruit leaves. Bacterial populations were determined by homogenizing leaf discs (0.8 cm in diameter) in 0.85% (w/v) NaCl, followed by dilution plating. The growth assays were repeated three times with similar results with three replicates each time. Data from one representative experiment are shown. Vertical bars represent the standard errors of the means.