GST-fusion protein assay to study the interaction regions between hMSH2 and hMSH3 and between hMSH2 and hMSH6. Radiolabeled ([³⁵S]methionine) IVTT protein [IVTT-hMSH(Y)] was precipitated with a GST-fusion protein [GST-hMSH(X)]. The first lane contains 10% of the total IVTT mixture used in each experiment. Three controls are shown for each experiment. The second lane contains the IVTT protein which was added to glutathione beads (only). The third lane contains the glutathione beads which were pretreated with a control E. coli lysate which was similar to the lysates from which the GST moiety and the GST-MSH(X) proteins were isolated. The fourth lane contains the glutathione beads which were pretreated with an E. coli lysate which had been induced for expression of GST moiety (alone) from the vector pGEX-SG1. The fifth lane contains the IVTT-MSH(Y) protein incubated with glutathione beads which had been pretreated with lysates that had been made from E. coli which had been induced for expression of a GST-MSH(X) fusion protein. After stringent washing, the samples were resolved on an SDS–8% PAGE gel and imaged with a PhosphorImager. No other bands were visible on the gels other than those shown.

Interaction regions of hMSH3 with hMSH2. ³⁵S-labeled full-length and truncation mutants of IVTT-hMSH3 were added to glutathione beads which had been pretreated with either GST (alone) or GST-hMSH2. (A) Phosphorimage of samples resolved on an SDS–12% PAGE gel. (B) Illustration of the constructs that were used in this experiment and corresponding region locations. The numbers correspond to the pairs shown in panel A. The hMSH2 interaction regions are shaded gray in the full-length hMSH3.

Interaction regions of hMSH2 with hMSH3. ³⁵S-labeled full-length and truncation mutants of IVTT-hMSH2 were added to glutathione beads which had been pretreated with either GST (alone) or GST-hMSH3. (A) Phosphorimage of samples resolved on an SDS–12% PAGE gel. (B) Illustration of the constructs that were used in this experiment and corresponding region locations. The numbers correspond to the pairs shown in panel A. The hMSH2 interaction regions are shaded gray in the full-length hMSH2.

Linear orientations of the hMSH2-hMSH3 interaction regions. ³⁵S-labeled full-length and truncation mutants of IVTT-hMSH2 were added to a GST construct which contained the amino-terminal interaction region of hMSH3 (aa 1 to 297) (A) or a GST construct which contained the carboxy-terminal interaction region of hMSH3 (aa 1025 to 1129) (B). The carboxy-terminal interaction region of hMSH2 with hMSH3 was further resolved by using GST-hMSH3 (aa 1025 to 1129) and carboxy-terminal fragments containing internal deletions of IVTT-hMSH2 (C). Each of these experiments was resolved on an SDS–15% PAGE gel and visualized with a PhosphorImager. (D) Illustration of the constructs and consensus interactions with amino- and carboxy-terminal regions shown in panels A to C.

Interaction regions of hMSH6 with hMSH2. ³⁵S-labeled full-length and truncation mutants of IVTT-hMSH6 were added to glutathione beads which had been pretreated with either GST (alone) or GST-hMSH2. (A) Phosphorimage of samples resolved on an SDS–12% PAGE gel. (B) Illustration of the constructs that were used in this experiment and corresponding region locations. The numbers correspond to the pairs shown in panel A. The hMSH2 interaction regions are shaded gray in the full-length hMSH6.

Interaction regions of hMSH2 with hMSH6. ³⁵S-labeled full-length and truncation mutants of IVTT-hMSH2 were added to glutathione beads which had been pretreated with either GST (alone) or GST-hMSH6. (A) Phosphorimage of samples resolved on an SDS–12% PAGE gel. (B) Illustration of the constructs that were used in this experiment and corresponding region locations. The numbers correspond to the pairs shown in panel A. The hMSH6 interaction regions are shaded gray in the full-length hMSH2.

Linear orientations of the hMSH2-hMSH6 interaction regions. ³⁵S-labeled full-length and truncation mutants of IVTT-hMSH2 were added to a GST construct which contained the amino-terminal interaction region of hMSH6 (aa 303 to 600) (A) or a GST construct which contained the carboxy-terminal interaction region of hMSH6 (aa 1302 to 1360) (B). The carboxy-terminal interaction region of hMSH2 with hMSH6 was further resolved by using GST-hMSH6 (aa 1302 to 1360) and carboxy-terminal fragments containing internal deletions of IVTT-hMSH2 (C). Each of these experiments was resolved on an SDS–15% PAGE gel and visualized with a PhosphorImager. (D) Illustration of the constructs and consensus interactions with amino- and carboxy-terminal regions shown in panels A to C.

Effects of HNPCC mutations on hMSH2 interaction with hMSH3 or hMSH6. Analysis of the interaction of hMSH2 missense mutations, found in well-defined HNPCC kindreds, with hMSH3 or hMSH6. The binding association of GST-MSH3 or GST-MSH6 with each of these mutations was determined by using IVTT-hMSH2 truncated peptides known to contain the consensus interaction region. Peptides were resolved on SDS–15% PAGE gels and examined by PhosphorImager.

Model of the hMSH2 consensus interaction with hMSH3 or hMSH6. The interaction regions of hMSH2 with hMSH3 and of hMSH2 with hMSH6 are shown in gray and they are connected with lines that illustrate the specificity of each region to their pairing partner. The nucleotide binding regions are shown as black boxes. The location of the HNPCC mutations tested in these studies are illustrated as black diamonds.