First Experiments on the New Protein Crystallography Beam Line 9-1, and Future Plans

- M. Soltis
- P. Kuhn
- P. Phizackerley

During the last two weeks of the run that ended August 31, 1996, we had the opportunity to conduct a number of experiments to test the new protein crystallography Beam Line 9-1 that has been under development at SSRL for the past three years (seeaccompanying article on Beam Line 9 in this Newsletter). Although conceptually similar to our existing protein crystallography Beam Line 7-1, Beam Line 9-1 provides a brighter x-ray beam and has an experimental enclosure and nearby setup space (for crystal mounting and data process- ing) that is considerably more spacious than Beam Line 7-1, making it more convenient to use.

The source of radiation for side station Beam Line 9-1 is a 3 mR fan of x -rays from an 8-period permanent magnet wiggler, operating at 2 T. In comparison to Beam Line 7-1, the horizontal fan is three times as wide, the wiggler has a higher field strength, and has twice the number of magnetic poles. This higher x-ray flux necessitated the use of a different arrangement of x-ray optics to Beam Line 7-1. The first optical element is a 1 m long Rh-coated water- cooled silicon mirror (inclined at 4.1 mR to the primary white x-ray beam) used to focus the beam in the vertical plane at the experimental camera and to reject higher energy components in the beam. The second optical element is a novel side-cooled silicon crystal monochromator used to both select the desired x-ray wavelength and to focus the beam in the horizontal plane.

As in the case of Beam Line 7-1, we have incorporated an experimental table that can be swung in the horizontal plane so that a limited range of x-ray wavelengths can be selected. However, the monochromator crystal has been manufactured to provide an optimized beam at the Se K-absorption edge, and was set to this wavelength for our initial trials. This wavelength was chosen because it provides a good compromise between low sample crystal absorption, and good spot-to-spot resolving power at the detector for large unit cells, while also providing for the possibility of enhanced anomalous scattering data when Se is incorporated into the protein.

When this station has been fully completed, it will be equipped with an advanced electronic area detector system mounted on a Kappa-geometry diffractometer. The detector will be mounted on (1) an offset stage permitting data collection to very high resolution, and (2) a long motorized bench so that very large unit cells can be resolved. Additional support equipment, such as a low temperature cryostat and a high-pressure gas cell, is also being provided.

Since we are currently working on the development of two new beam line stations for protein crystallography (Beam Line 9-1 and Beam Line 9-2), we have placed orders for two new state-of-the-art electronic area detector systems. The first is a large (total area = 188 mm x 188 mm) 2 x 2 matrix array CCD detector from Area Detector Systems Corp., and the second is a new large format (345 mm dia.) MAR-Research imaging plate data collection system with a readout that will be much faster than our existing 300 mm-diameter MAR-Research system. Both orders were placed early this year and we are eagerly awaiting their arrival. It is not yet certain which of these two systems will finally be chosen for use on Beam Line 9-1. For our initial experimental trials conducted in August 1996, we were very kindly loaned a new 300 mm diameter MAR-Research imaging plate detector system by the manufacturer and have been told that we may continue to use it until the new version of the detector has been delivered. The CCD system is being procured using NIH funding and the new MAR with DOE/OHER funding.

Beam Line 9-1 has been set up to provide a computational environment that allows the user to conduct a complete crystallographic experiment during experimental beam time. During the commissioning phase, the beam line and the MAR- Research detector were controlled by a Digital Alphastation running OpenVMS. The diffraction images were written through a network to a data disk residing on a 333 MHz Digital Alphastation running Digital UNIX. All data processing was performed locally on the UNIX machine. It was equipped with graphics to allow evaluation and analysis of the data and basic 2-D display of models and maps. We are in the process of acquiring an SGI R10,000 running Irix 6.2, equipped with 128 MB of memory, solid Impact graphics, dials, and crystal eyes for full 3-D stereo viewing of models and maps. In addition, several devices for data backup have been tested or are currently being tested: 4 mm DAT, 8 mm EXABYTE, digital linear tape (DLT), and a CD writer.

The VMS system mainly runs the software package MARControl, written by MAR-Research to control the detector, while the UNIX system currently runs the data reduction/analysis packages such as the CCP4 program suite, HKL and XtalView. We plan to install other crystallographic packages in the near future such as PHASES, XPLOR, and O.

Several experiments were carried out on Beam Line 9-1 at the end of the run in August as part of the commissioning phase for this new beam line and a number of these experiments are outlined here. Michael Stowell and co-workers, working in the laboratory of Douglas Rees at the California Institute of Technology, collected data from formaldehyde oxido reductase, a tungsten containing enzyme. Beam Line 9-1 produced 1.83 Å resolution data in comparison to 3.2 Å resolution data produced on their home x-ray source. Because the wavelength of the x-rays was relatively short, a high quality anomalous Patterson map was obtained that clearly showed the tungsten site. Phasing information from the tungsten site is currently facilitating structure solution. In addition, data were collected from a nitrogenase complex of the MoFe protein and a mutant of the iron protein. Their home x-ray source provided 4.0 Å data, yielding poor quality maps. A 2.5 Å resolution data set was collected on Beam Line 9-1 and the resulting maps are readily interpretable, and model building is underway.

Peter David and co-workers, working in the laboratory of Roger Kornberg at Stanford University, collected data from Yeast RNA polymerase II. This protein is an especially difficult system for data collection owing to the weakly diffracting crystals and the large unit cell (396 Å x 211 Å x 220 Å). However, exposure times were reduced by a factor of 5 compared to Beam Line 7-1, allowing comparably more data to be collected at higher resolution.

In a collaboration between Doug Freymann at the University of California San Francisco, and Peter Kuhn of SSRL, atomic resolution diffraction data were collected from Ffh, a bacterial homologue of the sequence recognition protein SRP54. The beam intensity was strong enough to collect data to 1.06 Å resolution using an exposure time of 6 min. per degree of crystal rotation. In addition, Peter Kuhn collected a high resolution data set from PNGase F, an enzyme responsible for deglycosylation of glycoproteins, complexed with the diacetylchitobiose core. The crystals diffracted to 1.45 Å resolution. The data produced good statistics (Rsym = 4.9%) and resulted in a high quality electron density map of the complex.

In a collaboration between Matt Redinbo and Wim Hol at the University of Washington, and Peter Kuhn and Mike Soltis (SSRL), diffraction data were collected on human topoisomerase I in complex with DNA. This crystal system has proven to be difficult for preparing heavy atom derivatives. Potential isomorphous derivative data were collected from crystals soaked in solutions of large metal cluster compounds. Diffraction data from these experiments gave excellent statistics, and the results are currently being examined. Potential isomorphous derivative data were also collected from crystals under Xe pressure at room temperature and at cryogenic temperature (~100 K). Data from these experiments were excellent and are likewise being evaluated. These crystals typically diffract to 5.0 Å using a home x-ray source. Data to 3.2 Å resolution were collected at room temperature and to 2.8 Å at cryogenic temperature on Beam Line 9-1.

Overall, experimental users reported collecting excellent quality data on Beam Line 9-1 as determined from data reduction statistics and data analysis. The final commissioning phase of the beam line will be performed during the next user run, and routine scheduling of users on this beam line may occur as soon as January 1997.

The Beam Line 9 project is being funded by the DOE Office of Health and Environmental Research (OHER). The project has also been accomplished in part through efforts of staff supported in part by the NIH NCRR SSRL Biotechnology Resource.

[SSRL Welcome Page ] [Research Highlights]  [Beamlines] [Accel Physics]  [User Admin]   [Announcements] 

December 2, 1996