AKT1-Inhibitor VIII co-crystals were harvested into a solution of 25 mM Na-acetate, 25 mM Na-citrate, 21% PEG MME 2000, pH 5.0 and were cryoprotected with 70% harvest solution + 30% ethylene glycol. Cryoprotected crystals were flash cooled in a stream of dry nitrogen vapor held at 100 K. X-ray diffraction data were collected on a Rigaku FR-E Superbright rotating anode X-ray generator, fitted with a Cu anode and an RAXIS IV++ image plate detector (Rigaku, TX, USA). The diffraction data were processed using Mosflm [26] and scaled using the program Scala [27] (link).
The crystals belonged to space group P21 with unit cell dimensions of a = 49.31 Å, b = 69.94 Å, c = 61.85 Å, β = 100.6°.
The crystal structure was solved by molecular replacement, with all calculations performed using the program Molrep [27] (link). The molecular replacement calculations were performed in two steps: In the first step, a search model consisting of residues 147–440 of the inactive AKT2 kinase domain (PDB code: 1MRV) was used in a standard rotation function/translation function calculation, resulting in a single solution with an R-factor of 0.505 (similar searches with an active conformation of AKT1 kinase domain failed to find a reliable solution). The quality of this molecular replacement solution was improved slightly by brief crystallographic refinement to 2.8 Å resolution in Refmac5. In the second step, a search model consisting of residues 2–106 of the unliganded AKT1 PH domain (PDB code: 1UNP) was used in a rotation function/phased translation function search, using phase information calculated from the coordinates of the AKT2 kinase domain solution. A single PH domain plus kinase domain solution was found with an R-factor of 0.417. This combined solution was subjected to multiple cycles of refinement in Refmac5 to 2.7 Å resolution [28] (link), followed by model rebuilding in the program O [29] (link). The final round of model rebuilding was guided by the use of simulated annealing composite omit maps followed by crystallographic refinement in CNX 2005 [30] (link) to generate the final molecular model.
The final structure contains all residues of AKT1 from 2–429 except for 45–48, 114–144 (the linker region), 189–198 (the αB and αC helices) and 299–312 (the C-terminal end of the activation loop). The model also has 21 ordered water molecules and a single copy of Inhibitor VIII. The R-factor of the final model is 0.245 with an Rfree value of 0.307. 312 residues (98.7%) lie in the “most-favored” or “additional” regions of the Ramachandran plot, with 2 residues (0.6%) in the “generously allowed” region and 2 (0.6%) in the “disallowed” regions. All figures were generated using Pymol [31] .
The crystals belonged to space group P21 with unit cell dimensions of a = 49.31 Å, b = 69.94 Å, c = 61.85 Å, β = 100.6°.
The crystal structure was solved by molecular replacement, with all calculations performed using the program Molrep [27] (link). The molecular replacement calculations were performed in two steps: In the first step, a search model consisting of residues 147–440 of the inactive AKT2 kinase domain (PDB code: 1MRV) was used in a standard rotation function/translation function calculation, resulting in a single solution with an R-factor of 0.505 (similar searches with an active conformation of AKT1 kinase domain failed to find a reliable solution). The quality of this molecular replacement solution was improved slightly by brief crystallographic refinement to 2.8 Å resolution in Refmac5. In the second step, a search model consisting of residues 2–106 of the unliganded AKT1 PH domain (PDB code: 1UNP) was used in a rotation function/phased translation function search, using phase information calculated from the coordinates of the AKT2 kinase domain solution. A single PH domain plus kinase domain solution was found with an R-factor of 0.417. This combined solution was subjected to multiple cycles of refinement in Refmac5 to 2.7 Å resolution [28] (link), followed by model rebuilding in the program O [29] (link). The final round of model rebuilding was guided by the use of simulated annealing composite omit maps followed by crystallographic refinement in CNX 2005 [30] (link) to generate the final molecular model.
The final structure contains all residues of AKT1 from 2–429 except for 45–48, 114–144 (the linker region), 189–198 (the αB and αC helices) and 299–312 (the C-terminal end of the activation loop). The model also has 21 ordered water molecules and a single copy of Inhibitor VIII. The R-factor of the final model is 0.245 with an Rfree value of 0.307. 312 residues (98.7%) lie in the “most-favored” or “additional” regions of the Ramachandran plot, with 2 residues (0.6%) in the “generously allowed” region and 2 (0.6%) in the “disallowed” regions. All figures were generated using Pymol [31] .
Free full text: Click here
