Structural insights into yeast DNA repair proteins MLH1 and PMS1 by mass spectrometry

DNA mismatch repair (MMR) is essential for the maintenance of genetic material, and the major features have been conserved over time. The primary protein components include members of the MutL family, and the two yeast members of the MutL family that are critical for MMR are MLH1 and PMS1. Mutations in MMR proteins have been implicated in carcinogenesis and can result in a complete loss of MMR activity, possibly due to structural changes in these proteins. Little is known about the tertiary structure of these proteins (especially their C-terminal domains), or their homologs in different species. The specific aims of this study are (1) to investigate the structure and determine the contact and binding sites of the MLH1 and PMS1 heterodimer; and (2) to map the DNA binding site of the N-terminal domain (NTD) of yeast DNA repair protein PMS1. The specific aims have been achieved by a combination of chemical cross-linking, surface modification, limited proteolysis, mass spectrometric characterization and molecular modeling based on the constraints determined in these experiments. The surface modification studies revealed lysine residues located on the surface of both PMS1 and MLH1. These studies implicated residues 665, 675, and 704 of MLH1 to be involved in heterodimerization. The cross-linking studies indicated residues in MLH1 that may be located at a maximum of 12 A apart. A homology model was created for the N-terminal domain of MLH1. A working model was created for the C-terminal domain of MLH1 and the potential heterodimer interaction was modeled using secondary structure prediction, sequence alignment, and the data obtained. We proposed the interaction sites for other MutL homolog heterodimers in yeast and humans. The limited proteolysis DNA binding studies with PMS1 NTD showed no proteolytic cleavage at Arg residue 198 and Lys 364 when DNA is bound, indicating direct involvement in binding or complete protection due to a conformational change. Other residues are also seen to be protected when DNA is bound. These studies have provided insight into the structures of these proteins and allow us to correlate known loss of function and carcinogenic mutations to disruptions in these protein structures.