As structural biology companies race ahead to obtain x-ray crystallography structures of promising drug targets, the biggest hurdle has traditionally been getting access to tunable x-ray sources at synchrotrons.
The capacity is there, said Lance Stewart, CEO of Emerald BioStructures, a subsidiary of MediChem Life Sciences. But the timing with which you can get access to this capacity is not very efficient.
That is beginning to change. Last week, the Commercial Collaborative Access Team (COM-CAT), a fee-for-service program, began offering access to the Advanced Photon Source (APS) at Argonne National Laboratory in Argonne, Ill. In addition, some structural biology companies have built, or are in the process of building their own beamlines at Argonne and other synchrotrons.
Until recently, companies that could not afford to build their own beamline had to compete with academic labs for synchrotron x-ray sources in general user programs. The peer-review process used to prioritize access could last up to six months, and proposals sent to the review committee tend to be peer-reviewed on the basis of scientific merit, putting purely technology-oriented companies at a disadvantage, said Kevin DAmico, vice president at Structural GenomiX (SGX) in San Diego.
The COM-CAT program is aimed at addressing companies appetite for beamtime. Last month, Advanced X-Ray Analytical Services (AXAS), another MediChem subsidiary, won the contract to operate COM-CATs beamline, which is owned by the Department of Energy and built with $8.7 million in Illinois state funds. According to Stephen Wasserman, AXAS vice-president, samples are usually processed within two weeks and turn-around times are often as short as a week.
Although only time will tell if COM-CAT provides enough beamline access to satisfy the desires of structural biology companies, Wasserman is confident the program will succeed. I think in the short term, and I suspect that will be a couple of years, we will be able to take care of our customers, he said. If we get to 100 users a year, I think we would consider this a success.
But just in case, Wasserman said there are ways to add incremental capacity to the beamline, such as improving automation. Additionally, AXAS has partnerships with other synchrotrons to provide access for its users during downtime periods at APS. Readily available beamtime, however, comes at a price: according to Stewart, fees are on the order of ten times higher at AXAS than at the national laboratories.
Syrxx, another San Diego-based structural biology company, decided to take another route by building its own beamline at the Advanced Light Source (ALS) in Berkeley, Calif. The company began collecting data in March 2001, and since then, has solved 40 protein crystal structures. We are shooting for 70 to 80 [protein structures] by the years end, said Nathaniel David, the companys director of business development.
But not even a private beamline presents the perfect solution. Automating the data-collection process, David said, made determining a protein structure about ten times more efficient, but Syrxx still has outstanding experimental needs. Syrxx shares its facility with the Genomics Institute of the Novartis Research Foundation, and the company is looking into buying another beamline, he added.
SGX, for its part, is constructing a $8 million private beamline at APS that will come online in about five weeks. The company has already determined over 50 protein structures, and is expecting to obtain additional structures of both protein and protein-small molecule complexes over the next few years. The capacity to solve one to two thousand such structures in a year is completely reasonable, said DAmico.
For other companies, collectively buying beamline access has presented another option. Currently, more than half a dozen pharmaceutical companies have access to two beamlines at APS via the Industrial Macromolecular Crystallography Association, and take turns running experiments every few weeks.
But if competition for access to synchrotron radiation has waned assuming a company can pay for it where is the bottleneck now? Getting the crystal, said Wasserman, and purifying proteins quickly enough. Over 60 percent of the major drug targets are membrane proteins, such as ion channels, receptors or pumps, which are just flat out difficult to express, purify and crystallize, Stewart said.