It has been an earnest desire of protein crystallographers to collect fast, accurate, high resolution diffraction data from protein crystals, preferably with exposure time as short as possible. In order to meet this challenge, a new type of Weissenberg camera has been developed for the recording of diffraction intensity from protein crystals using synchrotron radiation. The BL6A2 line has a plane-bending mirror designed by Y. Sato. The optical bench with triangular tilt-cut Si crystal monochromator was designed by N. Kamiya and was installed in the BL6A2 hutch. The Weissenberg camera was set on the 2θ arm of the optical bench. This camera can be used with Fuji Imaging Plate (IP) as an x-ray detector, and the reading out of the image from the IP is carried out by using BA100. The characterization of this system was carried out using the native crystal of chicken gizzard G-actin DNase I complex and its Yb3+, PCMB, indium, and FMA derivatives. Since these crystals are very sensitive for x rays, the resolution limit of the diffraction was 5 Å with a 4-circle diffractometer on a rotating anode x-ray generator (N. Sakabe et al., J. Biochem. 95, 887. This complex was crystallized in space group P2,2,2, with a=42.0, b=225.3, and c=77.4 Å. The data were collected with this system with the 430-mm radius cassette when Photon Factory was operated at 2.5 GeV and 270 mA and the wavelength λ=1.004 Å was chosen. In order to avoid overlapping of diffraction spots, oscillation angle range and coupling constant (degree/mm) were settled on the basis of simulation patterns of diffraction spots up to the maximum resolution to be measured considering the direction of the crystal axes, wavelength, radius of the camera, and mosaicness of the crystal. When the oscillation axis was a axis, the oscillation angle range was selected at either 10° (1°/mm) or 5° (0.5°/mm) depending on the density of reciprocal lattice points along the incident beam, and typical exposure time in each IP was 50 and 25 s, respectively. The exposure was stopped after 10 times oscillation. The total range of 117.5° was recorded on 16 sheets of IP with an overlapping range 0.5°. The data processing was carried out using program weis coded by T. Higashi. Two complete data sets along the a and c axes were collected using two crystals, independently, and the merge R(F2) for native crystal was 0.068. In order to know the feasibility of the data (F+−F−), Patterson maps were calculated with data of each derivative, and heavy atom vectors clearly appeared as prominent peaks in the Harker sections of the Patterson maps of both Yb3+ and PCMB derivatives. The heavy atom positions were obtained from the combinations of different Patterson and different Fourier maps, and were refined by using least-squares techniques. The final figure of merit up to 2.5 Å resolution was 0.61 with 22 700 reflections. The assignment of DNase I part on the electron density maps is progressing using the Nicholson molecular model referring to its structure reported by C. Oefner and D. Suck [J. Mol. Biol. 192, 605 (1986)]. In addition to promising results of ω-aminoacid pyruvate aminotransferase [N. Watanabe et al., Book of abstracts of second Japan-China Bilateral Symposium on Biophysics, p. 83 (1988)], these results support that this data collection system consisting of a new type of Weissenberg camera using SR, Fuji Imaging Plate, BA100, and program weis is one of the fastest and most accurate systems for protein crystallography in use today. We thank the Education Ministry and the Foundations of Yamada, Naito and Toray for the financial support of this project. We are grateful to Dr. K. Namba of ERATO for our use of BA100.
Read full abstract