A Unique B-Family DNA Polymerase Facilitating Error-Prone DNA Damage Tolerance in Crenarchaeota

Sulfolobus islandicus codes for four DNA polymerases: three are of the B-family (Dpo1, Dpo2, and Dpo3), and one is of the Y-family (Dpo4). Western analysis revealed that among the four polymerases, only Dpo2 exhibited DNA damage-inducible expression. To investigate how these DNA polymerases could contribute to DNA damage tolerance in S. islandicus, we conducted genetic analysis of their encoding genes in this archaeon. Plasmid-borne gene expression revealed that Dpo2 increases cell survival upon DNA damage at the expense of mutagenesis. Gene deletion studies showed although dpo1 is essential, the remaining three genes are dispensable. Furthermore, although Dpo4 functions in housekeeping translesion DNA synthesis (TLS), Dpo2, a B-family DNA polymerase once predicted to be inactive, functions as a damage-inducible TLS enzyme solely responsible for targeted mutagenesis, facilitating GC to AT/TA conversions in the process. Together, our data indicate that Dpo2 is the main DNA polymerase responsible for DNA damage tolerance and is the primary source of targeted mutagenesis. Given that crenarchaea encoding a Dpo2 also have a low-GC composition genome, the Dpo2-dependent DNA repair pathway may be conserved in this archaeal lineage.


Δdpo2 ΔpyrEFΔlacSΔdpo2
This work

Δdpo3 ΔpyrEFΔlacSΔdpo3
This work

Δdpo4 ΔpyrEFΔlacSΔdpo4
This work

Plasmids Features
pSeSD A Sulfolobus-E.coli shuttle vector carrying an expression cassette controlled under a synthetic strong promoter ParaS-SD Peng et al., 2012(Peng et al., 2012 pSeSD_dpo2 pSeSD carrying Dpo2 encoding sequence This work pSe-Rp The plasmid contains a DNA fragment of two tandem copies of CRISPR repeat sequences for the construction of the artificial mini-CRISPR array Peng et al., 2015(Peng et al., 2015 pAC-dpo1, -2, -3, -4 pSe-Rp carrying a spacer matching to the protospacer in the coding region of dpo1, -2, -3, -4 gene correspondingly in genome This work pGE-dpo1, -2, -3, -4 The genome-editing plasmid derived from pAC-dpo1, -2, -3, -4 respectively, with the corresponding This work  Figure S1. Effect of NQO on cell growth and expression of DNA polymerases in S. islandicus (A) Growth curve of the wild-type strain of S. islandicus E233S in the presence of NQO. NQO was added to exponetial growth cultures (A600nm=0.2) at the concentration of 0, 1, 2 and 3 μM, and incubated for 24 h during which cell samples were taken for monitoring their A600 values.
(B) Expression profiles of the four DNA polymerases revealed by western analysis. 10 μg of total cell extracts of NQO-treated samples (1, 2, 3 µM) and the untreated reference (CK) were used for the immunoblotting analysis using antibodies against each DNA polymerase. PCNA1, which has a constant expression upon DNA damage, serves as a loading control. (C) Quantification of relative amounts of Dpo2 in samples taken from cultures incubated with different concentrations of NQO.

Supplementary Figure S2. Cell growth and western blotting analysis of dpo2-overexpression strain and its reference
(A) Exponentially growing cultures (A600nm=0.2) were incubated with 0, 1, 2 and 3 μM NQO for 24 hours. The A600nm value of each culture at 0h and 24 hour after NQO addition was plotted. Three independent experiments were performed with the standard deviation shown in the error bar. Unpaired t test was performed for each group of data, with p values indicated in the graph. inthe (B) Cell samples were taken at 6 hours after NQO addition and cell extracts were obtained by sonication and centrifugation. 10 μg of cell extracts of NQO-treated and control (CK) samples were used for immunoblotting analysis using Dpo2 and Penta-His Tag antibodies. To estimate the relative amounts of overexpressed Dpo2, samples of overexpression strain were diluted by 20 times individually and used for the western blot analysis. PCNA1 serves as an internal control. 1N, 2N and 3N refers to the sample incubated with 1, 2 and 3 μM NQO respectively.

Supplementary Figure S3. Expression of each DNA polymerase in different strains upon NQO treatment
Exponentially growing cultures (A600nm=0.2) were incubated with 2 μM NQO for 6 hours and samples were taken for the preparation of cell extracts. Euqal amount of cell extracts (10 μg) for each sample were used in western blotting assay using antibodies against each DNA polymerase. PCNA1 serves as an internal control.

Supplementary Figure S4. Expression of each DNA polymerase in different strains post UV irradiation
Exponentially growing cultures (A600nm=0.2) were exposed to 50 J/m 2 UV-C light, then, the treaed cultures were allowed to recover for 6 h under the dark condition with shaking. Cell extracts were prepared and euqal amounts of cell extracts (10 μg) for each sample were used for the western blotting assay using antibodies against each DNA polymerase. PCNA1 serves as an internal control.

Supplementary Figure S5. Dpo2 homologues carry a substitution at the PolC motif
Structure-based sequence alignment of the PolC motif of Dpo2 homologues. The mutated aspartate in the PolC motif (YGDTDS) was indicated by the asterisk symbol. The structure of Thermococcus gorgonarius PolB (1tgo) was used as the template for the structure-based sequence alignment, which was performed using PROMALS3D webserver (Pei et al., 2008) and depicted using Espript 3 (Robert and Gouet, 2014). SisDpo1 harboring the canonical PolC motif was shown as the control.

Supplementary Figure S6. Sequence alignment of conserved regions of Dpo2 homologues
The structure of Thermococcus gorgonarius PolB (1tgo) was used as the template for the structurebased sequence alignment, which was performed using PROMALS3D webserver (Pei et al., 2008) and depicted by Espript 3 (Robert and Gouet, 2014). SisDpo1 belonging to PolB1 family was shown as a control. Framed sequences indicates a 12 aa sequence insertion in Thermogladius calderae PolB2 (Tcal).

Supplementary Figure S8. Phylogenetic tree of crenarchaeal species
Phylogenetic trees of representative crenarchaeal species were constructed using their 16S rDNA sequences retrieved from the NCBI databases. The 16S rDNA sequences were first aligned using MUSCLE program, then, poorly aligned regions were removed by Gblocks program (v0.91b) using the default setting. The phylogenetic tree was constructed using the trimmed sequences with the PhyML program (v3.0) and visualized by the TreeDyn program (v198.3). The 16s rDNA sequence of Haloferax volcanii DS2 was used as the outgroup.

Supplementary Figure S9. Sequence alignment of YxGG/A and PolA motif of archaeal PolB2 homologs
The structure of Thermococcus gorgonarius PolB (1tgo) was used as the template to conduct structure-based sequence allignments using PROMALS3D webserver (Pei et al., 2008), and the resulting data were depicted by Espript 3 (Robert and Gouet, 2014). Conserved residues are highlighted by yellow background and identical ones are in the red. Variations in the YxGG/A and PolA motifs of PolB2s from Aeropyrum pernix K1 (Ape) and Aeropyrum camini SY1 are framed and indicated by red arrows.

Supplementary Figure S10. Spontaneous mutation spectra of apt3 locus in WT and Δdpo4
Mutated bases are shown in red on the top of original ones. Single base deletions are indicated with blue "-" signs above the deleted bases, and single base insertions are indicated by green "+" signs beneath the bases immediately before the insertion positions with the inserted shown in green. Large insertions (>2bp) are shown with black triangle signs. Numbers in the bracket indicate the sample size (total number of analyzed mutants).