STRUCTURE OF BOVINE PANCREATIC TRYPSIN-INHIBITOR AT 125 K - DEFINITION OF CARBOXYL-TERMINAL RESIDUES GLY57 AND ALA58

Citation
S. Parkin et al., STRUCTURE OF BOVINE PANCREATIC TRYPSIN-INHIBITOR AT 125 K - DEFINITION OF CARBOXYL-TERMINAL RESIDUES GLY57 AND ALA58, Acta crystallographica. Section D, Biological crystallography, 52, 1996, pp. 18-29
Citations number
43
Categorie Soggetti
Crystallography,"Biochemical Research Methods",Biology
ISSN journal
09074449
Volume
52
Year of publication
1996
Part
1
Pages
18 - 29
Database
ISI
SICI code
0907-4449(1996)52:<18:SOBPTA>2.0.ZU;2-K
Abstract
The structure of bovine pancreatic trypsin inhibitor has been refined to a resolution of 1.1 Angstrom against data collected at 125 K. The s pace group of the form II crystal is P2(1)2(1)2(1) with a = 75.39(3), b =2 2.581(7), c = 28.606(9)Angstrom (cf. a = 74.1, b = 23.4, c = 28.9 Angstrom at room temperature). The structure was refined by restraine d least-squares minimization of Sigma w(F-o(2) - F-c(2))(2) with the S HELXL93 program. As the model improved, water molecules were included and exceptionally clear electron density was found for two residues, G ly57 and Ala58, that had been largely obscured at room temperature. Th e side chains of residues Glu7 and Arg53 were modelled over two positi ons with refined occupancy factors. The final model contains 145.6 wat er molecules distributed over 167 sites, and a single phosphate group disordered over two sites. The root-mean-square discrepancy between C( alpha atoms in residues Arg1-Gly56 at room and low temperatures is 0.4 Angstrom. A comparison of models refined with anisotropic and isotrop ic thermal parameters revealed that there were no significant differen ces in atomic positions. The final weighted R-factor on F-2 (wR(2)) fo r data in the range 10-1.1 Angstrom was 35.9% for the anisotropic mode l and 40.9% for the isotropic model. Conventional R factors based on F for F > 4 sigma(F) were 12.2 and 14.6%, respectively, corresponding t o 16.1 and 18.7% on all data. These large R-factor differences were no t reflected in values of R(free), which were not significantly differe nt at 21.5(5) and 21.8(4)%, respectively. These results, along with th e relatively straightforward nature of the refinement, clearly highlig ht the benefits of low-temperature data collection.