Molecular basis for the functions of a bacterial MutS2 in DNA repair and recombination

Abstract

Bacterial MutS2 proteins, consisting of functional domains for ATPase, DNA-binding, and nuclease activities, play roles in DNA recombination and repair. Here we observe a mechanism for generating MutS2 expression diversity in the human pathogen Helicobacter pylori, and identify a unique MutS2 domain responsible for specific DNA-binding. H. pylori strains differ in mutS2 expression due to variations in the DNA upstream sequence containing short sequence repeats. Based on Western blots, mutS2 in some strains appears to be co-translated with the upstream gene, but in other strains (e.g. UA948) such translational coupling does not occur. Accordingly, strain UA948 had phenotypes similar to its ΔmutS2 derivative, whereas expression of MutS2 at a separate locus in UA948 (the genetically complemented strain) displayed a lower mutation rate and lower transformation frequency than did ΔmutS2. A series of truncated HpMutS2 proteins were purified and tested for their specific abilities to bind 8-oxoG-containing DNA (GO:C) and Holiday Junction structures (HJ). The specific DNA binding domain was localized to an area adjacent to the Smr nuclease domain, and it encompasses 30-amino-acid-residues containing a "KPPKNKFKPPK" motif. Gel shift assays and competition assays supported that a truncated version of HpMutS2-C12 (∼12kDa protein containing the specific DNA-binding domain) has much greater capacity to bind to HJ or GO:C DNA than to normal double stranded DNA. By studying the in vivo roles of the separate domains of HpMutS2, we observed that the truncated versions were unable to complement the ΔmutS2 strain, suggesting the requirement for coordinated function of all the domains in vivo.

Keywords: DNA recombination; Helicobacter pylori; MutS2; Oxidative DNA damage; Specific DNA binding; Translational coupling.

Fig. 1. Expression of mutS2 genes in different H. pylori strains

A: Gene organization and the variable sequences of the mutS2 locus in the H. pylori genome. Thick arrows represent the orientation and extent of ORFs, for which the HP numbers (referring to the 26695 genome sequence) and the gene names are given. The nucleotide sequences between HP0622 and mutS2 from different strains are aligned. The TTG labeled with *** at the beginning of the sequences are the translation start codon of mutS2 annotated for the strain 26695. The stop codons of HP0622 are overlined, and the start codons of mutS2 that we proposed are underlined. Hyphens are introduced for sequence alignment (in the middle region) and for highlighting the coding frame of the mutS2 genes (at the right end). B: Western blot analysis for MutS2 in H. pylori strains. Shown is a representative result from three independent experiments. H. pylori cells were grown in an atmosphere containing 4% O₂ to late log phase. Seven mg of each crude cell extract was resolved on a SDS-12.5% polyacrylamide gel followed by transfer onto a nitrocellulose membrane, and the proteins were detected using anti-MutS2 antiserum and anti-UreA antiserum (as an internal loading control). Lane 1: Purified MutS2 protein; Lane 2: 26695 WT cells; Lane 3: 26695ΔmutS2 cells; Lane 4: UA948 WT cells; Lane 5: UA948ΔmutS2 + P_rdxAmutS2 cells; Lane 6: SS1 WT cells.

Fig. 2. Identification of specific DNA binding domain of HpMutS2

A: Schematic representation of HpMutS2 and truncated constructs. Shown on the top are known functional domains and a linker region with unknown function. Different truncated HpMutS2 gene constructs were prepared by PCR and proteins were purified as described in Table 1. The names of proteins and their corresponding amino acid numbers are shown. B and C: Formation of specific MutS2-DNA complex analyzed by gel shift assay. HpMutS2 variant proteins (1 μg/ml) were incubated with 10 nM ³²P-labelled DNA substrates (GO:C or Holiday Junction). The protein-DNA complexes and the free DNA substrates are marked. “NO” represents the control with no protein.

Fig. 3. Specific DNA binding ability of HpMutS2 truncated variants

A: Schematic representation of HpMutS2 domains and truncated constructs. SDB represents the newly identified Specific DNA-Binding domain, and the corresponding amino acid sequence motif responsible for specific DNA-binding is shown at bottom. B: HpMutS2 truncated variant proteins (C8, C9, C10, C12, C23) were overexpressed in E. coli and purified to near homogeneity as analyzed on a 20% SDS-PAGE. C: A mobility shift assay. One μM HJ DNA substrate was incubated with 1 μM of each HpMutS2 truncated variant protein followed by analysis on a 1.5% agarose gel. The bands of protein-DNA complex and the free DNA substrates are marked. “NO” represents the control with no protein. D: Quantification of the specific DNA binding ability of each HpMutS2 truncated variant protein determined by the filter retention assay. ³²P labelled DNA substrate (GO:C or HJ, 10 nM) was incubated with different proteins (100 nM). Protein-DNA complexes retained on the filter were quantified with a PhosphorImager. Error bars indicate the standard deviation of three independent experiments. E: Alignment of MutS2 sequences (at the identified SDB domain in HpMutS2) from various bacterial species (retrieved from NCBI database, with accession numbers in parentheses). The sequence motif “LDLRG” is one of the most conserved region among Smr sequences, wherein the aspartic residue is involved in the endonuclease activity.

Fig. 3. Specific DNA binding ability of HpMutS2 truncated variants

A: Schematic representation of HpMutS2 domains and truncated constructs. SDB represents the newly identified Specific DNA-Binding domain, and the corresponding amino acid sequence motif responsible for specific DNA-binding is shown at bottom. B: HpMutS2 truncated variant proteins (C8, C9, C10, C12, C23) were overexpressed in E. coli and purified to near homogeneity as analyzed on a 20% SDS-PAGE. C: A mobility shift assay. One μM HJ DNA substrate was incubated with 1 μM of each HpMutS2 truncated variant protein followed by analysis on a 1.5% agarose gel. The bands of protein-DNA complex and the free DNA substrates are marked. “NO” represents the control with no protein. D: Quantification of the specific DNA binding ability of each HpMutS2 truncated variant protein determined by the filter retention assay. ³²P labelled DNA substrate (GO:C or HJ, 10 nM) was incubated with different proteins (100 nM). Protein-DNA complexes retained on the filter were quantified with a PhosphorImager. Error bars indicate the standard deviation of three independent experiments. E: Alignment of MutS2 sequences (at the identified SDB domain in HpMutS2) from various bacterial species (retrieved from NCBI database, with accession numbers in parentheses). The sequence motif “LDLRG” is one of the most conserved region among Smr sequences, wherein the aspartic residue is involved in the endonuclease activity.

Fig. 4. Specific vs non-specific DNA binding ability

A & B: Detection of protein-DNA complex formation analyzed by gel shift assay. One μM DNA substrates (normal dsDNA, GO:C-containing DNA, or Holiday Junction) was incubated with HpMutS2-N31 (A) or HpMutS2-C12 protein (B) at indicated concentrations followed by analysis on 1.5% agarose gel. The bands of protein-DNA complex (C) and the free DNA substrates (F2, dsDNA; F4, four stranded DNA) are marked. C: Competition assay. 100 nM HpMutS2-C12 was incubated with 10 nM ³²P-labeled GO:C DNA substrate and varying concentrations (10, 100, 500 nM) of unlabeled G:C, GO:C, or HJ competitor DNAs, and then the amount of protein-DNA complex was determined using PAGE gel shift assays. The data (% GO:C bound) are averages of three experiments with standard deviations.