Fig 1.
Generation and characterization of uracil excision repair mutants.
(A) Schematic depicting the generation of gene replacement mutant of ung. The insertion of hygromycin resistance cassette disrupted the native allele. (B) PCR using F2-R2 (gene-specific primers) resulted in the amplification of 684 bp in Rv and ~2 kb in RvΔung. PCR using F1-R1 (primers beyond the 5’ and 3’ flank) resulted in the amplification of ~3 kb in Rv and ~4.5 kb in RvΔung. PknB gene amplification was used as the positive control. (C) Schematic depicting the generation of gene replacement mutant of udgB. The insertion of chloramphenicol resistance cassette disrupted the native allele. (D) PCR using F4-R3 resulted in amplification in RvΔudgB but not in Rv. PCR using F3-R3 resulted in amplification of ~2.5 kb and ~3 kb in Rv and RvΔudgB, respectively. (E) In the background of RvΔung, the udgB native allele was disrupted by the insertion of the chloramphenicol resistance cassette. Indicated primers were used for screening the RvΔdKO. (F) Immunoblot analysis for the confirmation of gene replacement mutants. 50 and 100 μg of Rv, RvΔung, RvΔudgB, and RvΔdKO. WCLs were resolved on 16% Tris-Tricine-Urea gel, transferred to a nitrocellulose membrane, and probed with α-Ung (1:2000) and α -UdgB (1:2000) antibodies, respectively. WCL (10 μg) was resolved on 10% SDS-PAGE, transferred to a membrane, and probed with α-PknB (1:10000) antibody. (G) Schematic representation of UDG activity assay. SSU9 and GU9 were incubated with various lysates or Ugi. (H) Cell extracts (10 μg) of Rv, RvΔung, RvΔudgB, and RvΔdKO and the 5’-32P end-labeled SSU9 and GU9 (25000 c.p.m) were used for performing UDG assay. Ugi (25ng) was preincubated with Rv, RvΔung, RvΔudgB, and RvΔdKO for performing UdgB activity assays. Product and Substrate are labeled as ‘P’ and ‘S’.
Fig 2.
Characterization of complementation strains of uracil DNA glycosylases.
(A and B) Growth profiles of Rv, RvΔung, RvΔudgB, and RvΔdKO in 7H9 or Sauton’s minimal medium were determined by CFU enumeration on 7H11-OADC plates at indicated time points. Data represent two biologically independent experiments. Each experiment was performed in quadruplets. Data represent mean and standard deviation. (C) Schematic representation of complementation constructs pST-Ki-ung, pST-Ki-udgB, pST-Ki-ung-udgB. (D and E) UDG activity assays were performed in the absence or presence of Ugi using Rv, RvΔung, RvΔudgB, and RvΔdKO, RvΔung::ung, RvΔudgB::udgB and RvΔdKO::ung-udgB as described in Methods.
Fig 3.
Uracil DNA glycosylase gene mutants exhibit hypermutability.
(A) Mid-log-phase cultures of Rv, RvΔung, RvΔudgB, and RvΔdKO were used for genomic DNA preparation. Schematic showing the procedure employed for WGS. (B) WGS of RvΔung, RvΔudgB, and RvΔdKO were compared with WGS of Rv grown in vitro. The heat map shows the percent SNPs present in RvΔung (n = 4), RvΔudgB (n = 4), and RvΔdKO (n = 3). Blosum score provides information about the synonymous and non-synonymous changes due to SNP. (C) Schematic representation of spontaneous mutation rate analysis. The mutation rate was calculated using a fluctuation test. (D) Rifampicin and isoniazid resistance rates for Rv, RvΔung, RvΔung::ung, RvΔudgB, RvΔudgB::udgB, RvΔdKO and RvΔdKO::ung-udgB. Data is representative of biological triplicates. Graphpad prism was used for the statistical analysis (one-way ANOVA). Data represent mean and SD. (E)The table represents a fold increase of the mutation rate of Rv, RvΔung, RvΔung::ung, RvΔudgB, RvΔudgB::udgB, RvΔdKO and RvΔdKO::ung-udgB with respect to Rv. (F) Mutation frequency was calculated after subjecting the Rv, RvΔung, RvΔung::ung, RvΔudgB, RvΔudgB::udgB, RvΔdKO and RvΔdKO::ung-udgB to nitrosative or oxidative stress, respectively. Cells were plated on no antibiotic and rifampicin (10 μg/ml)-containing plates. Data represent one of the two biological experiments and each experiment was performed in triplicates. Data represent mean and SD. *p<0.01, **p<0.001 and *** p<0.0001.
Fig 4.
Uracil DNA glycosylase mutants exhibit hypervirulent phenotype.
(A) Schematic outline depicting the pipeline used for WGS of colonies obtained from the guinea pig lungs Rv (n = 11), RvΔung (n = 8), RvΔudgB (n = 8), and RvΔdKO (n = 9). (B) Rv sequenced from guinea pig lungs was compared with the sequence of Rv grown in vitro. A heat map represents the unique SNPs identified in the Rv isolated from guinea pig lungs. Blosum score provides information about the synonymous and non-synonymous changes due to SNP. (C) SNPs per million nucleotides in the sequenced Rv genome, isolated from guinea pig lungs. (D) Heat map showing the SNPs accumulated in Rv, RvΔung, RvΔudgB, RvΔdKO isolated from guinea pig lungs compared with the sequence of Rv grown in vitro. Blosum score provides information about the non-synonymous changes due to SNP. Percentage mutation provides information about the percent of strains sequenced where a particular SNP is detected. (E) Nature of mutations identified in the RvΔdKO isolated from guinea pigs. (F) SNPs per million nucleotides in the sequenced Rv, RvΔung, RvΔudgB, and RvΔdKO. Mutations listed are for one strand of DNA. The mutation spectrum analysis was performed by counting the number of independent mutations per gene; therefore, C➔T mutation is different from G➔A mutation E (G-I) Guinea pigs were challenged with Rv, RvΔdKO, and RvΔdKO::ung-udgB strains. At 56-days p.i., lungs and spleen were isolated. Gross histopathology examined by hematoxylin and eosin staining. CFUs were enumerated by plating lung and spleen homogenate for all seven. (G) Representative gross pathology images of the infected guinea pig lungs and spleen. White arrows shows the discrete tubercle. (H) CFUs were enumerated as described in Methods. Data represent CFU (log10) per lung at day 1 p.i. CFU (log10/ml) from the lungs and spleen of infected guinea pigs (n = 7/per strain) at 56-days p.i. Statistical analysis (one-way ANOVA) was performed using Graphpad prism. ***p<0.0001. (I) Gross histopathology images (40x- magnification) of Rv, RvΔdKO, and RvΔdKO::ung-udgB of infected guinea pig lungs. Black arrows indicate granuloma. G-Granuloma, AS-Alveolar Space.
Fig 5.
Uracil DNA glycosylase mutants accelerate the acquisition of antibiotic resistance.
(A) Spontaneous mutation rate analysis was performed using ciprofloxacin. Data are representative of one of the two biological experiments. Each biological experiment was performed in triplicates. Data represent mean and SD. Statistical analysis (one way ANOVA) was performed using Graphpad prism. ***p<0.0001. (B) Table shows ciprofloxacin resistance rate calculated as described in Fig 2C for Rv, RvΔung, RvΔung::ung, RvΔudgB, RvΔudgB::udgB, RvΔdKO and RvΔdKO::ung-udgB. Fold increase of the mutation rate of Rv, RvΔung, RvΔung::ung, RvΔudgB, RvΔudgB::udgB, RvΔdKO, and RvΔdKO::ung-udgB with respect to Rv. (C) Heat map showing the SNPs accumulated in RvΔdKO (n = 3) grown in vitro and ciprofloxacin-resistant RvΔdKO (n = 13) with respect to Rv grown in vitro. Blosum score provides information about the synonymous and non-synonymous changes due to SNP. Percentage mutation provides information about the percent of strains sequenced where a particular SNP is detected. (D) SNPs per million nucleotides in the sequenced RvΔdKO grown in vitro and ciprofloxacin-resistant RvΔdKO. (E) Circos plot representing the mutations in the genome of ciprofloxacin-resistant RvΔdKO (grey circles; n = 13) with respect to Rv reference sequence (red circle). The spikes in the innermost circle show mutation frequencies. Mutations in gyrA, rv2414c, nanT, and rpoB intergenic regions are highlighted with dotted blue lines. (F) A matrix representing mutation in the gyrA and Rv600-rpoB intergenic region in ciprofloxacin-resistant RvΔdKO.
Fig 6.
RvΔdKO strain displays superior fitness ex vivo and in vivo.
(A) Schematic outline of the competition experiment performed in peritoneal macrophages. (B) Single-cell suspensions of Rv, RvΔdKO, and RvΔdKO::ung-udgB were used to infect peritoneal macrophages. Cells were lysed at the indicated time point, and CFUs were enumerated on 7H11-agar plates. Survival of Rv, RvΔdKO, and RvΔdKO::ung udgB independently at indicated time point shown in (B). (C) Percent CFUs of Rv and RvΔdKO and RvΔdKO::ung udgB were determined by replica plating 100 colonies 7H11-hygromycin plates. While RvΔdKO or RvΔdKO::ung-udgB colonies can grow on 7H11-hygromycin plates, colonies belonging to Rv would not grow on these plates. Ex-vivo infection was performed using two independent biological experiments, and each biological experiment was performed in triplicates. Statistical analysis (Unpaired t-test) was performed using n = 3 for each biological experiment. Graphpad Prism software was employed for performing statistical analysis. *** p<0.0005. Data represents one of the two biological experiments. (D) Schematic representation of a competition experiment in mice (n = 8). Rv (kanamycin-resistant) and RvΔdKO (hygromycin resistant) independently or together were challenged through the aerosol route. CFUs were enumerated at day 1- and 56-days p.i. on 7H11 plates as indicated. (E) Schematic representation of a competition experiment in guinea pigs (n = 7). Rv and RvΔdKO or Rv and RvΔdKO::ung-udgB together were challenged through the aerosol route. CFUs were enumerated at day 1- and 56-days p.i. on 7H11 plates as indicated. CFU(log10) per lung was plotted at day 1 p.i. CFU (log10/ml) was plotted at 56-days p.i. The graph shows CFUs representing the survival of Rv and RvΔdKO or Rv and RvΔdKO::ung-udgB. For in vivo infection experiment, statistical analysis was performed at day 1 using n = 3 (per strain) mice or guinea pigs and n = 8 mice or n = 7 guinea pigs (per strain) at 56-days p.i. Statistical analysis (Unpaired t-test) was performed using Graphpad Prism software. Data represents mean and standard deviation. *** p<0.0005.
Fig 7.
RvΔdKO displays a decisive edge over the parent Rv strain.
(A) Graph represents percent CFUs calculated at day 1 and 56-days p.i. in the mice lungs and guinea pigs’ lungs or spleen. Statistical analysis (Unpaired t-test) was performed at day 1 using n = 3 (per strain) mice or guinea pigs and n = 8 mice or n = 7 guinea pigs (per strain) at 56-days p.i. Data represent mean and standard deviation.*** p<0.0005. (B) The model depicts the biological implication of deletion of uracil DNA glycosylases. In vitro, ex vivo, and in vivo experiments suggest that deletion of uracil DNA glycosylases provides a survival advantage in the host.