The CO2-dependence of Brucella ovis and Brucella abortus biovars is caused by defective carbonic anhydrases

Abstract

Brucella bacteria cause brucellosis, a major zoonosis whose control requires efficient diagnosis and vaccines. Identification of classical Brucella spp. has traditionally relied on phenotypic characterization, including surface antigens and 5-10% CO₂ necessity for growth (CO₂-dependence), a trait of Brucella ovis and most Brucella abortus biovars 1-4 strains. Although molecular tests are replacing phenotypic methods, CO₂-dependence remains of interest as it conditions isolation and propagation and reflects Brucella metabolism, an area of active research. Here, we investigated the connection of CO₂-dependence and carbonic anhydrases (CA), the enzymes catalyzing the hydration of CO₂ to the bicarbonate used by anaplerotic and biosynthetic carboxylases. Based on the previous demonstration that B. suis carries two functional CAs (CAI and CAII), we analyzed the CA sequences of CO₂-dependent and -independent brucellae and spontaneous mutants. The comparisons strongly suggested that CAII is not functional in CO₂-dependent B. abortus and B. ovis, and that a modified CAII sequence explains the CO₂-independent phenotype of spontaneous mutants. Then, by mutagenesis and heterologous plasmid complementation and chromosomal insertion we proved that CAI alone is enough to support CO₂-independent growth of B. suis in rich media but not of B. abortus in rich media or B. suis in minimal media. Finally, we also found that insertion of a heterologous active CAII into B. ovis reverted the CO₂-dependence but did not alter its virulence in the mouse model. These results allow a better understanding of central aspects of Brucella metabolism and, in the case of B. ovis, provide tools for large-scale production of diagnostic antigens and vaccines.

Figure 1

CAI is dispensable for CO₂-independent growth of B. abortus 2308W. Growth of B. abortus 2308W and the corresponding insertion mutant in CAI were tested under normal atmospheric conditions in Peptone-Glucose. Each point represents the mean ± standard error (error bars are within the size of the symbols) of technical triplicates. The experiment was repeated three times with similar results.

Figure 2

CAI and CAII are functional in B. suis 1330 and 513. Growth of B. suis 1330 and B. suis 513 and the corresponding insertion mutants in CAI (B. suis 1330::pJQKm-CAI; B. suis 513::pJQKm-CAI) and CAII (B. suis 1330::pJQKm-CAII; B. suis 513::pJQKm-CAII) were tested under normal atmospheric conditions in Peptone-Glucose. Each point represents the mean ± standard error (error bars are within the size of the symbols) of technical triplicates. The experiment was repeated three times with similar results.

Figure 3

B. abortus 292 and 544 carrying a functional CAII become CO₂-independent. Growth of B. abortus strains 292 and 544 and the derivative strains carrying plasmid pRH001_Ba2308WCAII or a stable Tn7_Ba2308WCAII insertion in the genome were tested under normal atmospheric conditions in Peptone-Glucose. Each point represents the mean ± standard error (error bars are within the size of the symbols) of technical triplicates. The experiment was repeated three times with similar results.

Figure 4

The CO₂-independence mediated by CAI is conditioned by nutrient availability. Growth of B. suis strain 513 and the CA insertion mutants (B. suis 513::pJQKm-CAI; B. suis 513::pJQKm-CAII) were tested under normal atmospheric conditions in Glutamate–Lactate–Glycerol and Glucose. Each point represents the mean ± standard error (error bars are within the size of the symbols) of technical triplicates. The experiment was repeated three times with similar results.

Figure 5

Insertion of the gene of a functional CAI in B. abortus 2308W does not increase growth rates. Growth of B. abortus 2308W and its derivative strain carrying B. suis 513 CAI in the genome (B. abortus 2308W Tn7_Bs513CAI) were tested under normal atmospheric conditions in Peptone-Glucose. Each point represents the mean ± standard error (error bars are within the size of the symbols) of technical triplicates. The experiment was repeated three times with similar results.

Figure 6

An active CAII reverts the CO₂-dependence of B. ovis PA and does not alter its multiplication in the mouse model. A Growth of B. ovis PA and its derivative strain carrying B. abortus 2308W CAII in the genome (B. ovis PA Tn7_Ba2308WCAII) under normal atmospheric conditions in Peptone-Glucose-Yeast Extract-Serum. Each point represents the mean ± standard error (error bars are within the size of the symbols) of technical triplicates. The experiment was repeated three times with similar results. B Bacterial loads of B. ovis PA and B. ovis PA Tn7_Ba2308WCAII in the spleens of BALB/c mice at 3 and 8 weeks post-infection. No statistical differences were found (Student’s t-test).