Mycobacterium tuberculosis (Mtb) and Mycobacterium smegmatis (Msmeg) contain two ε (dnaq) exonuclease homologs.

Transcription

1 Supplementary Figure 1 Mycobacterium tuberculosis (Mtb) and Mycobacterium smegmatis (Msmeg) contain two ε (dnaq) exonuclease homologs. (a) Sequence alignment of the ε-exonuclease homologs from four different species. Conserved catalytic residues of E. coli ε are indicated by blue triangles below the sequences. The clamp-binding motif of E. coli ε is boxed in green. (b) Active site of the E. coli ε exonuclease. (c) Computational model of the active site of Mtb Rv3711c. (d) Computational model of the active site of Mtb Rv2191.

2 Supplementary Figure 2 Mycobacterium tuberculosis Rv3711c (Rv3711c MTB ) is a 3 5 DNA exonuclease but does not form a stable complex with DnaE1 MTB. (a) Coomassie-stained gel showing purified E. coli ε (ε EC ), DnaE1 MTB and Rv3711c MTB. (b) Gel showing a 3 5 exonuclease activity assay with ε EC, DnaE1 MTB and Rv3711c MTB. (c) Analytical size exclusion chromatography shows that E. coli PolIIIα (PolIIIα EC ) and ε EC form a stable complex at concentrations as low as 1.5 M. (d) In contrast, DnaE1 MTB and Rv3711c MTB do not show any interaction, even at 10 M protein concentration (all equimolar amounts).

3 Supplementary Figure 3 PHP active sites in bacterial replicative DNA polymerases. (a) Alignment of the PHP domain sequences from replicative DNA polymerases. Conserved metal-binding residues of the PHP domain are indicated by blue triangles below the sequences. Cyan squares indicate residues in E. coli that deviate from the consensus metalbinding motif. (b) Computational model of the Mtb DnaE1 PHP domain based on the crystal structure of T. aquaticus PolIII (shown in c). Black circles indicate residues mutated for the experiments performed in this study. (c) The PHP domain active site of T. aquaticus

4 PolIII. (d) The PHP domain active site of E. coli PolIII. Underlines indicate residues in E. coli that deviate from the consensus metalbinding motif.

5 Supplementary Figure 4 Mycobacterium tuberculosis DnaE1 wild-type (DnaE1 MTB WT) and PHP mutants are properly folded. (a) SDS-PAGE analysis of purified proteins. Each lane contains 3.5 pmol (~0.5 μg) protein. The gel was stained with Coomassie Brilliant Blue. (b) Purified proteins do not show any aggregation, as judged by size exclusion chromatography. The arrow indicates the void volume (at 0.8 ml). For clarity, graphs are shifted vertically by 150 mau. (c) Circular dichroism spectra show that DnaE1 MTB and PHP mutants are properly folded. (d) Thermal denaturation curves show that WT and mutant DnaE1 have similar melting temperatures of C. (e) Time course of exonuclease activity on single-stranded DNA. DnaE1 MTB WT shows robust 3 5 exonuclease activity but not 5 3 exonuclease activity.

6 Supplementary Figure 5 Primer extension from mismatched substrates requires exonuclease activity. (a) Primer extension from mismatched substrates by Mycobacterium tuberculosis DnaE1 wild-type (DnaE1 MTB WT) and E. coli PolIII (PolIII EC ) + ε EC is blocked by a phosphorothioate linkage (denoted by -S-) that is resistant to exonuclease activity. In contrast, a matched primer with a terminal phosphorothioate linkage can be extended normally. (b) Addition of ε EC exonuclease in trans allows DnaE1 MTB PHP mutants to extend from mismatched DNA substrates.

7 Supplementary Figure 6 The per base pair mutation rate of Mycobacterium smegmatis estimated from fluctuation analysis. (a) Fluctuation analysis was used to determine the rate at which wild-type M. smegmatis acquired resistance to rifampicin. Circles represent the mutant frequency (number of rifampicin-resistant mutants per cell plated in a single culture). The red bar represents the estimated mutation rate (mutations conferring rifampicin resistance per generation), with error bars representing the 95% confidence interval (CI). (b) The number of mutations in rpob (Ms1367) that confer rifampicin resistance in our fluctuation analysis was determined by sequencing 150 independent rifampicin-resistant isolates. This analysis identified ten unique mutations. The per base pair mutation rate, in vitro, was determined by dividing rifampicin by the target size.

8 Supplementary Figure 7 Loss-of-function mutations in the dnae1 PHP domain are rarely found in clinical Mycobacterium tuberculosis isolates. (a) dnae1 (Rv1547) PHP domain SNPs observed in clinical Mtb isolates. SNP prevalence refers to the number of clinical strains containing the indicated SNP as compared to the total number of clinical strains analyzed. See Supplemental Table 1 for additional information. (b) Fluctuation analysis was used to determine the rates at which the indicated M. smegmatis strains acquired resistance to rifampicin. With the exception of wild-type M. smegmatis, these strains harbor a deletion of the endogenous dnae1 (Ms3178) gene and have been complemented with the indicated M. tuberculosis dnae1 (Rv1547) gene. Circles represent the mutant frequency (number of rifampicin-resistant mutants per cell plated in a single culture). The red bar represents the estimated mutation rate (mutations conferring rifampicin resistance per generation), with error bars representing the 95% confidence interval (CI). *P < 0.05 in comparison of mutant frequencies by Wilcoxon rank-sum test.

9 Supplementary Table 5. Bacterial strains used in this study. For the sake of simplicity, Mtb DnaE1 numbering was used throughout the paper. Mtb DnaE1 D23 = Msmeg DnaE1 D25 Mtb DnaE1 D226 = Msmeg DnaE1 D228 M. smegmatis All strains are derivatives of mc with the exception of the protein production strain, which is a derivative of mc Strain # Genotype 1 wild-type mc dnae1::hyg dnae1::l5(zeo) [Ms3178] 22 dnae1::hyg dnae1::l5(kan) [Ms3178] 30 dnae1::hyg dnae1::l5(kan) [Rv1547] 48 dnae1-myc::l5(kan) [Ms3178] 58 dnae1-d228n-myc::l5(kan) [Ms3178] 62 P UV15-Tet -dnae1-myc::l5(kan) [Ms3178] 90 P UV15-Tet -dnae1-d228n-myc::l5(kan) [Ms3178] 121 dnae1-d25n-myc::l5(kan) [Ms3178] 128 groelδc P acet -T7 RNA polymerase::l5(kan) 215 P UV15-Tet -dnae1-myc::l5(kan) [Rv1547] 217 P UV15-Tet -dnae1-d23n-myc::l5(kan) [Rv1547] 221 P UV15-Tet -dnae1-d226n-myc::l5(kan) [Rv1547] 223 P UV15-Tet -dnae1-d25n-myc::l5(kan) [Ms3178] 248 dnae1::hyg dnae1-k95n::l5(kan) [Rv1547] 251 dnae1::hyg dnae1-t249i::l5(kan) [Rv1547] 285 dnae1::hyg dnae1-a149v::l5(kan) [Rv1547] CF1 ΔdnaQ (Ms6275) CF2 ΔMs4259 CF3 ΔdnaQ (Ms6275) ΔMs4259 MA3 dnae1::hyg dnae1-d228n+silent::l5(kan) [Ms3178] MA4 dnae1::hyg dnae1-d25n+silent::l5(kan) [Ms3178] M. tuberculosis (all strains are derivatives of H37Rv) CF4 CF5 wild-type H37Rv ΔdnaQ (Rv3711c) E. coli BL21- Gold(DE3) E. coli B F - ompt hsds(r 8 - m 8 - ) dcm + Tet r gal λ(de3) enda Hte

10 Supplementary Table 6. Plasmids generated for this study. Plasmid # Description 14 dnae1::l5(kan) [Rv1547] 17 dnae1::lox-hyg-lox [Ms3178] 23 dnae1::l5(kan) [Ms3178] 30 dnae1::l5(zeo) [Ms3178] 40 dnae1-d228n::l5(kan) [Ms3178] 42 dnae1-myc::l5(kan) [Ms3178] 50 dnae1-d228n-myc::l5(kan) [Ms3178] 53 P UV15-Tet -dnae1-myc::l5(kan) [Ms3178] 57 dnae1-d25n-myc::l5(kan) [Ms3178] 65 P UV15-Tet -dnae1-d228n-myc::l5(kan) [Ms3178] 84 dnae1-myc::l5(kan) [Rv1547] 87 dnae1-d228n+silent::l5(kan) [Ms3178] 113 dnae1-d25n::l5(kan) [Ms3178] 143 P UV15-Tet -dnae1-myc::l5(kan) [Rv1547] 144 P UV15-Tet -dnae1-d23n-myc::l5(kan) [Rv1547] 146 P UV15-Tet -dnae1-d226n-myc::l5(kan) [Rv1547] 147 P UV15-Tet -dnae1-d25n-myc::l5(kan) [Ms3178] 161 dnae1-d25n+silent::l5(kan) [Ms3178] UFL1 6xHis-dnaE1 [Rv1547; pyub28b-hyg] UFL2 6xHis-dnaE1-D23N [Rv1547; pyub28b-hyg] UFL3 6xHis-dnaE1-D226N [Rv1547; pyub28b-hyg] UFL4 PolIII alpha (E. coli) [pet3c-amp] UFL5 6xHis-epsilon (E. coli) [pet28a-pp-kan] UFL6 6xHis-dnaQ [Rv3711c; pet28a-pp-kan]

11 Supplementary Table 7. DNA substrates used in this study Red letters in bold/underline indicate position of mismatched bases. Blue, lower case s indicates position of the phosphorothioate linkage. Real-time primer extension assay matched mismatched matched, phosphorothioate mismatched, phosphorothioate 5 TAGGACGAAGGACTCCCAACTTTAGGTGCG 3 ATCCTGCTTCCTGAGGGTTGAAATCCACGCCCCCCCCC-[6FAM] 5 TAGGACGAAGGACTCCCAACTTTAGGTGCT 3 ATCCTGCTTCCTGAGGGTTGAAATCCACGCCCCCCCCC-[6FAM] 5 TAGGACGAAGGACTCCCAACTTTAGGTGCsG 3 ATCCTGCTTCCTGAGGGTTGAAATCCACG-CCCCCCCCC-[6FAM] 5 TAGGACGAAGGACTCCCAACTTTAGGTGCsT 3 ATCCTGCTTCCTGAGGGTTGAAATCCACG-CCCCCCCCC-[6FAM] Gel analysis primer extension assay matched mismatched -1 TG mismatched -1 TC mismatched -2 TG mismatched -2 TC ara-a inhibition assay Gel analysis exonuclease assay single-stranded, with 5 label single-stranded, with 3 label 3 CTCAGGAAGCAGGAATTTTGCTGTCTGTGAA 3 CTCAGGAAGCAGGAGTTTTGCTGTCTGTGAA 3 CTCAGGAAGCAGGACTTTTGCTGTCTGTGAA 3 CTCAGGAAGCAGGGATTTTGCTGTCTGTGAA 3 CTCAGGAAGCAGGCATTTTGCTGTCTGTGAA 3 CTCAGGAAGCAGGAAGTCAGTCAGTCAGTCA [6FAM]-5 GTTCACGAGACCTACTGACACTGA 5 GTTCACGAGACCTACTGACACTGA -[6FAM]