Demonstration of allelic exchange in the slow-growing bacterium Mycobacterium avium subsp. paratuberculosis, and generation of mutants with deletions at the pknG, relA, and lsr2 loci

Abstract

Mycobacterium avium subsp. paratuberculosis is the causative pathogen of Johne's disease, a chronic inflammatory wasting disease in ruminants. This disease has been difficult to control because of the lack of an effective vaccine. To address this need, we adapted a specialized transduction system originally developed for M. tuberculosis and modified it to improve the efficiency of allelic exchange in order to generate site-directed mutations in preselected M. avium subsp. paratuberculosis genes. With our novel optimized method, the allelic exchange frequency was 78 to 100% and the transduction frequency was 1.1 x 10(-7) to 2.9 x 10(-7). Three genes were selected for mutagenesis: pknG and relA, which are genes that are known to be important virulence factors in M. tuberculosis and M. bovis, and lsr2, a gene regulating lipid biosynthesis and antibiotic resistance. Mutants were successfully generated with a virulent strain of M. avium subsp. paratuberculosis (M. avium subsp. paratuberculosis K10) and with a recombinant K10 strain expressing the green fluorescent protein gene, gfp. The improved efficiency of disruption of selected genes in M. avium subsp. paratuberculosis should accelerate development of additional mutants for vaccine testing and functional studies.

FIG. 1.

Schematic diagram of allelic exchange mutagenesis in M. avium subsp. paratuberculosis. The inserted sequence containing the Hyg gene is the same size in all mutants (1,915 bp), but the size of the deleted sequence varies in the mutants developed in this study (ΔpknG and ΔrelAL, 1,737 bp; ΔrelAS, 873 bp; Δlsr2, 311 bp [Table 2]). Arrows indicate the schematic binding sites and the directions of primers used for PCR identification. The F and R primers are the primers designed to bind outside up- and downstream homologous regions in each mutant. PCRs for ΔrelAS and Δlsr2 were performed with a primer set consisting of F and R primers, because the PCR fragments for the mutant and wild-type strains were clearly distinguished (>1-kb difference). PCRs for ΔpknG and ΔrelAL were performed with specific primer sets consisting of both F and hyg-R primers and hyg-F and R primers because the size differences between the mutant and wild-type strains with the F and R primers were not well distinguished in these cases (178-bp difference). The schematic restriction (RE) and probing (open and filled bars) sites for Southern blot analysis are also shown. Hyg, Hyg^r gene; U and D, up- and downstream homologous regions; RE, restriction enzyme site; open and filled bars, probes for the deleted gene and the hyg gene, respectively.

FIG. 2.

PCR identification for specific gene construction in mutants. (A) PCR for ΔpknG; (B) PCR for ΔrelAL; (C) PCR for ΔrelAS; (D) PCR for Δlsr2. Lane 1, DNA size marker; lane 2, wild type (M. avium subsp. paratuberculosis K10); lane 3, mutant in M. avium subsp. paratuberculosis K10; lane 4, mutant in M. avium subsp. paratuberculosis K10-GFP. The primer sites for ΔpknG (A) and ΔrelAL (B) PCRs were located in the Hyg gene (inserted gene) for the forward primer and outside the downstream homologous region of each disrupted gene for the reverse primer. Note that the wild-type gene was not amplified in panels A and B because of the primer design. The primer sites for ΔrelAS (C) and Δlsr2 (D) PCRs were located outside up- and downstream homologous regions of each disrupted gene, which allowed identification of mutants based on the sizes of the amplified fragments.

FIG. 3.

RT-PCR analysis of gene expression in M. avium subsp. paratuberculosis strains. (A) RT-PCR for pknG (380 bp); (B and C) RT-PCR for relA from ΔrelAL (303 bp) (B) and ΔrelAS (303 bp) (C) mutants; (D) RT-PCR for lsr2 (145 bp). Lane 1, DNA marker; lanes 2 and 3, negative (without RT) and positive controls (gapDH) for RT-PCR, respectively; lanes 4 and 5, target gene expression in mutant and wild-type (M. avium subsp. paratuberculosis K10) strains, respectively. Negative and positive controls for RT-PCR of M. avium subsp. paratuberculosis K10 RNA were also analyzed (data not shown).

FIG. 4.

Southern blot analysis of genomic DNA from mutant and wild-type strains. (A) Southern blot characterization for ΔpknG; (B) Southern blot characterization for ΔrelAS and ΔrelAL; (C) Southern blot characterization for Δlsr2. The restriction and probing sites are shown in Fig. 1. The membrane was hybridized with the first probe (hyg probe). After chemiluminescence detection, the first probe was stripped out, and the membrane was reprobed with the second probe (probe for the deleted gene) and examined. All bands are consistent with the expected sizes.