Brucella abortus virB12 is expressed during infection but is not an essential component of the type IV secretion system

Abstract

The Brucella abortus virB operon, consisting of 11 genes, virB1 to virB11, and two putative genes, orf12 (virB12) and orf13, encodes a type IV secretion system (T4SS) that is required for intracellular replication and persistent infection in the mouse model. This study was undertaken to determine whether orf12 (virB12) encodes an essential part of the T4SS apparatus. The virB12 gene was found to encode a 17-kDa protein, which was detected in vitro in B. abortus grown to stationary phase. Mice infected with B. abortus 2308 produced an antibody response to the protein encoded by virB12, showing that this gene is expressed during infection. Expression of virB12 was not required for survival in J774 macrophages. VirB12 was also dispensable for the persistence of B. abortus, B. melitensis, and B. suis in mice up to 4 weeks after infection, since deletion mutants lacking virB12 were recovered from splenic tissue at wild-type levels. These results show that VirB12 is not essential for the persistence of the human-pathogenic Brucella spp. in the mouse and macrophage models of infection.

FIG. 1.

Construction of virB12 deletion mutants in Brucella. Top: Diagram showing overlap of a putative orf13 with the 5′ end of virB12. Bent arrows indicate the primers used for construction of virB12 deletion mutants. Middle: Map of pUKD/ORF12 showing KIXX flanked by upstream and downstream fragments cloned into the 2.7-kb pCR2.1 backbone (dotted line). The dotted arrow indicates the unique PstI site. Bottom: Replacement of an internal fragment of virB12 with KIXX does not disrupt the predicted orf13. The numbering of the virB12 deletion corresponds to that of the B. abortus virB operon in GenBank accession number AF226278 (22).

FIG. 2.

Characterization of virB12 mutants. (A) Confirmation of virB12 deletion by PCR. Chromosomal DNAs of B. abortus 2308, B. melitensis 16 M, and B. suis 1330 and their respective virB12 mutants were amplified with primer pair ORF12C-F and ORF12C-R (Fig. 1 and Table 2). Wild-type strains yield an amplicon of 520 bp of virB12, while mutants yield one of 1.3 kb, since they contain the Km resistance determinant KIXX. (B) Confirmation of virB12 deletion by Southern blotting. EcoRI-digested genomic DNAs of wild-type Brucella strains and virB12 mutants were hybridized with either a virB2 probe (corresponding to the sequence from 1609 to 1926 of B. abortus virB) or an internal virB12 probe (corresponding to the sequence from 11181 to 11554 of B. abortus virB). The virB2 probe was generated by PCR using the primer pair virB2C-F and virB2C-R. The virB12 probe was generated using primers ORF12I-F and ORF12C-R. Primer sequences are shown in Table 2. (C) Detection of VirB12 protein in stationary-phase cultures of B. abortus 2308, B. melitensis 16 M, and B. suis 1330. Proteins from 1 × 10⁸ CFU of stationary-phase culture were separated on a 12% SDS-PAGE gel, transferred to a PVDF membrane, and probed with anti-VirB12 antiserum.

FIG. 3.

VirB12-specific total IgG in serum samples of mice taken after infection of BALB/c mice with B. abortus 2308 (n = 10; filled bars) or its isogenic virB12 mutant, AK/ORF12 (n = 5; open bars). Samples taken at the indicated time points postinfection were diluted and measured by ELISA using His-VirB12 bound to HisSorb plates (QIAGEN). Bars indicate the averages ± standard deviations of duplicate measurements. The dotted line represents the background reactivity in mice before infection with B. abortus.

FIG. 4.

Recovery of B. abortus, B. suis, and B. melitensis virB12 mutants after coinfection of J774 macrophages with their respective wild-type strains (WTs). A mixed inoculum of the virB12 mutant and the wild type was used to infect J774 cells, and intracellular bacteria were recovered after 48 h as described in Materials and Methods. Data shown are the means of three independent assays done in triplicate ± standard deviations and plotted logarithmically. Top: Log-transformed CFU of the wild type and virB12 mutants recovered from J774 cells are given. Bottom: Competitive index (C.I.), calculated as log(CFU mutant/CFU wild type), is shown on the y axis. A significant difference (*) was seen only in the control experiment using B. abortus mutant BA41 (B. abortus virB1-virB2::mTn5) in comparison with the wild type (P = 0.04).

FIG. 5.

Recovery of B. abortus, B. melitensis, and B. suis virB12 deletion mutants from mouse splenic tissue after mixed infection with their respective wild-type strains (WTs). Bacteria were recovered 4 weeks after i.p. infections with 2 × 10⁵ CFU of 1:1 mixtures of the wild type and the virB12 mutant. Data shown are the averages of groups of four or five mice ± standard deviations and are plotted logarithmically. Top panel: Log-transformed CFU recovered from mouse spleens are given for wild-type and virB12 mutant strains. The dashed line indicates the limit of detection in the spleen. Bottom: Competitive index (C.I.), calculated as log(CFU mutant/CFU wild type), is shown on the y axis. A significant difference (*) was seen in the experiment using B. abortus mutant BA41 (virB1-virB2::mTn5) in comparison with the wild type (P = 0.04).

FIG. 6.

Effect of mutations in virB genes on VirB12 protein levels in B. abortus 2308. Top: Western blots. Proteins from 1 × 10⁸ CFU of stationary-phase culture were separated on a 12% SDS-PAGE gel and transferred to a PVDF membrane, which was followed by Western blotting with polyclonal anti-VirB5 (1:500), anti-VirB12 (1:50,000), or anti-Bcsp31 (1:5,000) serum. Bottom: To determine protein expression levels, immunoblots were quantified by measuring the relative optical densities and areas of the corresponding bands with a computerized image analysis system as described in Materials and Methods. Data were expressed as integrated density values and calculated as ratios of VirB5 to Bcsp31 or VirB12 to Bcsp31. Relative protein expression levels were plotted and compared with that of the wild type (WT), which was set to 1. The strain names corresponding to the genotypes shown are listed in Table 1. Data shown is from a single experiment that was representative of three independent experiments.