Solving Structures with
Collaborative Crystallography
At the BCSB, users solve protein structures at lightning speed
The Berkeley Center for Structural Biology’s Collaborative Crystallography (CC) program is making major advancements in solving protein structures, especially for users involved in high-throughput projects. The CC program is an NIH-funded, peer-reviewed service that allows external users to apply for both beam time and the support of a crystallographer to perform experiments and subsequent data analyses.
Banumathi Sankaran demonstrates the sample changer on Beamline 5.0.1.
Banumathi Sankaran has been providing user support under the guidance of Corie Ralston at the BCSB since 2005, and since she became research scientist in charge of the CC program in October 2008, she’s been aiding users in structure solution at an impressive rate. Under the oversight of Paul Adams, Acting Director of the Physical Biosciences Division and BCSB Head, the CC program has grown immensely. It started in May 2008 with four research groups; now 11 are actively participating, and Sankaran hopes this might increase.
“I am fully booked. In February, I have over 180 hours of beam time on Beamlines 5.0.1 and 5.0.2,” Sankaran says. She will be collecting for five different user groups and asserts that “they are all very happy.” Top-off has been very helpful and has significantly increased throughput. “Depending on the sample specifics,” she says, “screening a full puck (16 samples) can be done in an hour. The time required for data collection really depends on the sample, but typically I can screen, collect and process about 100 samples a day.”
The automated sample changer allows a user to efficiently and reproducibly mount protein crystals on the goniometer for diffraction analyses.
Sankaran assists with the macromolecular crystallography component of experiments, from basic data collection through structure solution and refinement. To date, 600 data sets have been collected in conjunction with the CC program. A majority of these are associated with a high-throughput project, done in collaboration with a group from Cambridge University, that seeks to develop novel inhibitors for proteins of pharmaceutical interest in micro-bacterium tuberculosis.
Since June 2009, 13 protein structures have been deposited into the RSCB Protein Data Bank (PDB); 15 more will be submitted by May.
Emerald BioStructures (EB), as part of the Seattle Structural Genomic Center for Infectious Diseases (SSGCID), has been actively participating in the CC program since November 2009. Since then, ten structures have been accepted into the PDB. “This is so fast,” Sankaran says. “We were able to solve one structure in just a week, from the time data was collected to submitting the structure.”
Jan Abendroth, Associate Director of Macromolecular Crystallography at EB, works with Sankaran to collect data. “At Emerald BioStructures we are extremely pleased with the CC program,” Abendroth says. “In the cooperation with Banu [Sankaran], the first shift yielded high-quality data sets, and the structures were deposited in the PDB shortly thereafter.”
This collaboration works toward SSGCID’s overall goal of solving structures that can aid the structure-guided design of new drugs against high-priority targets (see some structures solved at SSGCID). As such, target proteins are selected for their medical potential.
EB solves target structures for proteins from NIAID Category A-C (high priority) agents, as well as emerging and re-emerging infectious disease organisms. Included in this list are Giardia lamblia, a parasite that causes giardiasis, Ehrlichia chaffeensis, the causative agent of human monocytotropic ehrlichiosis, and Bartonella henselae, a proteobacterium that can cause endocarditis, bacillary angiomatosis, and cat-scratch disease–a torment to countless people with feline allergies, caused by the swelling of lymph nodes when bitten or scratched by a cat.
The crystal structure of glyceraldehyde-3-phosphate dehydrogenase from Bartonella henselae bound with NAD in the active site (PDB entry).
The crystal structure of nicotinate-nucleotide pyrophosphorylase from Ehrlichia chaffeensis (PDB entry).
SSGCID recently solved protein structures from the H1N1 virus. Researchers are now screening for small molecule fragments that bind H1N1 proteins, fragments that may one day result in a new class of compounds for the treatment of flu.
“The quality of data we receive is excellent,” Abendroth says. “Communication with Banu on the goals of data collection is also very efficient. Overall, the CC program has been very productive for us.”
This is Sankaran’s first time collecting data used in high-throughput settings, and she is pleased her collaborators can so rapidly determine structures and publish results. Prior to coming to the ALS in 2005, Sankaran worked with Christian Betzel at the DESY synchrotron in Germany. She then spent four years researching phasing methods at Brookhaven National Laboratory/NSLS with Zbigniew Dauter.
The CC program is advertised to, and accepts proposals from, an international research pool. It is popular among users working in high-throughput environments like structural genomics.
In other collaborations with the CC program, four new structures and five novel protein–ligand complexes have been solved. Collaborators in the CC program are mainly from the United States, but international groups from Bangalore (India) and Cambridge (UK) also participate.