Computational protocol: RGS7 is recurrently mutated in melanoma and promotes migration and invasion of human cancer cells

Similar protocols

Protocol publication

[…] The atomic coordinates of RGS7 RGS domain (PDB id 2D9J) were recovered from the Protein Data Bank ( E383K mutant was manually created and then was minimized using the Polak-Ribiere Conjugate Gradient (PRCG) minimization algorithm, the OPLS2005 force field and implicit solvent available in MacroModel v. 10.8 (Schrödinger, LLC, New York, NY, 2015).RGS7 N-terminal domain was modeled using the full-length X-ray structure of the homologous RGS9 (in complex with Gβ5, PDB id 2PBI), which has 40% identity and 56% similarity to RGS7). Chain A was selected as template for modeling automatically by I-TASSER v. 4.4 and the Phyre2 web servers, and manually by Modeller v. 9.15 and Prime v. 4.0 (Schrödinger, LLC, New York, NY, 2015),. Due to the high sequence similarity between the RGS7 and RGS9 N-terminal domains, the same sequence alignment was obtained by different methods. Using default settings, five models were generated by I-TASSER and Modeller, and one model by Prime and Phyre2, respectively. All models were used as input for a side chain refinement with SCWRL4. After side chain refinement, we observed that orientations of the R44 residue in the different models cluster to similar positions (Supplementary Fig. ). R44 switches between alternative conformations where it establishes H-bonds with S58, with D61, or with both S58 and D61. The representative structure is shown in Fig. . Residues around R44C were analyzed as potential positions for cysteine substitutions to form a disulfide bridge (Supplementary Fig. ). H-bonds are automatically visualized according to default distance criteria defined in the software Maestro (Schrodinger): maximum distance of 2.8 Å, donor minimum angle of 120, acceptor minimum angle of 90. The distances are calculated between the hydrogen and acceptor atom, and can be manually measured by the “Measure” tool available in Maestro.The substitution of an arginine with the shorter cysteine in the mutant RGS7 makes positions 58 and 61 too far from 44 to allow for disulfide bridge formation without affecting the loop conformation. In contrast, V56C resulted in optimal distance for a disulfide bond formation with R44C (Cα56-Cα44 distance of 5.3 Å, Cβ56-Cβ44 distance of 4.0 Å). The V56C and R44C mutations and the disulfide bridge were manually created in the representative model of the wild-type RGS7 model. The resulting R44C/V56C model was minimized using the Polak-Ribiere Conjugate Gradient (PRCG) minimization algorithm, the OPLS2005 force field and implicit solvent available in MacroModel v. 10.8 (Schrödinger, LLC, New York, NY, 2015). Native cysteines in this domain are located too far to be able to form alternative disulfide bridges (Fig. ). [...] WT RGS7, R44C RGS7 or R44C/V56C mutant RGS7 were overexpressed in A375  cells, immunoprecipitated with a Flag antibody, eluted from the beads and the cysteine were modified with N-Ethylmaleimide (NEM) without prior reduction in order to block formation of new disulfide bonds. The samples were then separated on SDS-PAGE and the slices at the expected size were cut. The samples were digested by trypsin and analyzed by LC-MS/MS on Q-Exactive-plus mass spectrometer fitted with a capillary HPLC (Thermo-Fisher Scientific). The peptides were identified by Discoverer software version 1.4 vs human uniprot and decoy databases (in order to determine the false discovery rate (FDR), and versus the specific sequences, using the Sequest search engine. Although the MS/MS spectrum was not clear enough to provide a statistically significant identification, the expected mass of the peptides was accurately observed. […]

Pipeline specifications

Software tools I-TASSER, Phyre, MODELLER, SCWRL, Comet
Applications MS-based untargeted proteomics, Protein structure analysis
Organisms Homo sapiens
Diseases Melanoma, Neoplasms
Chemicals Cysteine, Hydrogen