MutSbeta exceeds MutSalpha in dinucleotide loop repair

Abstract

Background: The target substrates of DNA mismatch recognising factors MutSalpha (MSH2+MSH6) and MutSbeta (MSH2+MSH3) have already been widely researched. However, the extent of their functional redundancy and clinical substance remains unclear. Mismatch repair (MMR)-deficient tumours are strongly associated with microsatellite instability (MSI) and the degree and type of MSI seem to be dependent on the MMR gene affected, and is linked to its substrate specificities. Deficiency in MSH2 and MSH6 is associated with both mononucleotide and dinucleotide repeat instability. Although no pathogenic MSH3 mutations have been reported, its deficiency is also suggested to cause low dinucleotide repeat instability.

Methods: To assess the substrate specificities and functionality of MutSalpha and MutSbeta we performed an in vitro MMR assay using three substrate constructs, GT mismatch, 1 and 2 nucleotide insertion/deletion loops (IDLs) in three different cell lines.

Results: Our results show that though MutSalpha alone seems to be responsible for GT and IDL1 repair, MutSalpha and MutSbeta indeed have functional redundancy in IDL2 repair and in contrast with earlier studies, MutSbeta seems to exceed MutSalpha.

Conclusion: The finding is clinically relevant because the strong role of MutSbeta in IDL2 repair indicates MSH3 deficiency in tumours with low dinucleotide and no mononucleotide repeat instability.

Figure 1

Western blot analysis of the MMR protein contents in the NEs used in the functional assay. HeLa, a positive control, includes all five MMR proteins, MLH1, PMS2, MSH2, MSH3 and MSH6. HCT116 lacks MLH1, PMS2 and MSH3. Both GP5d and LoVo lack MSH2, MSH3 and MSH6. As an assay control, Sf9 TE are included with and without the overexpressed WT MMR proteins. The loading control, α-tubulin is not shown.

Figure 2

MMR efficiency of HCT116, LoVo and GP5d NEs complemented with MutSα, MutSβ and MutLα complexes for 5′GT, 5′IDL1 and 5′IDL2 substrates. (A) Mock represents heteroduplex only, with no added NE or recombinant protein. MMR-proficient HeLa NE including all five MMR proteins is used as a positive control. MMR-deficient HCT116, LoVo and GP5d NEs and NEs complemented with Sf9 insect cell TE are used as negative controls. The top fragment (3193 bp) represents the unrepaired linearised heteroduplexes and the two lower fragments (1833 and 1360 bp) show the migration of the repaired and double-digested DNA molecules. The repair percentages (R%) represent fractions of repaired DNA calculated as a ratio of double-digested DNA relative to total DNA added to the reaction. Values are a mean of three independent experiments. (B) The comparison of substrate-specific repair efficiencies of the MMR protein complexes (repair efficiency R% and s.d.±%). MutSα is able to repair all three substrates (5′GT/5′IDL1/5′IDL2), whereas MutSβ does not repair 5′GT or 5′IDL1 in any extracts. However, complementation of HCT116 NE (lacking MLH1, PMS2 and MSH3) with MutLα alone yields a considerably lower IDL2 repair percentage (12%, s.d.±5%) than after co-complementation with MutLα and MutSβ (38%, s.d.±5%). Moreover, complementation of LoVo and GP5d NEs (lacking MSH2, MSH3 and MSH6) with MutSα yields lower IDL2 repair percentages, 17% (s.d.±17%) and 13% (s.d.±4%), than when complemented with MutSβ, 31% (s.d.±21%) and 18% (s.d.±9%), respectively.

Figure 3

The functional analysis of MSH3-R796W. Both MutSβ-WT (20%, s.d.±5%) and MutSβ-R796W (18%, s.d.±8%) restore the repair capability of GP5d+MutLα. The values indicated are averages obtained from three separate experiments.