In the southeastern corner of Washington, the Pacific Northwest National Laboratory is building what is essentially a massive scale factory for investigating proteins with a systems-biology-oriented framework.
To other labs, PNNL is known as the molecular biology laboratory with all the “toys” — an enviable set of tools and instruments to work with. It has collaborations with the Institute for Systems Biology as well as the Fred Hutchinson Cancer Research Center, both based in Seattle.
“We are exploring, taking a very rough cut of what is going on in a vast territory,” said Steven Wiley, chief scientist and director of the Biomolecular Systems Initiative at the Department of Energy laboratory in Richland, Wash.
Wiley spoke to BioCommerce Week in August, sitting in his large corner office in the PNNL facility next to the 560-acre Hanford Nuclear Reservation, which was used to produce plutonium during World War II, and which is scheduled for a billion-dollar federally mandated environmental cleanup. The building holding Wiley’s office is scheduled for demolition in 2009 to create a new set of facilities.
“Biology is still very primitive and rudimentary,” said Wiley, picking epidermal growth factor and its receptor system as an example. “We know a lot more about [EGF receptors], but nothing that allows [us] to predict. The next generation of scientists will take the parts we have learned about to make the transition to a predictive science. We want to know all the parts and translate the knowledge to a behavior of the system.”
At PNNL, this scientific effort has been fueled by at least $7 million a year in discretionary money over the last four years, said Wiley. This doesn’t count the millions of dollars in federal money invested to purchase the instruments that form a factory-like operation, the High Performance Mass Spectrometry Facility. This facility is largely based on seven mass spectrometry instruments, installed at a cost from $2 million and $4 million for each one of the instrument lines, coupled with electrospray ionization technology, four Fourier Transform ion cyclotron resonance spectrometers, including a massive 11.5-Tesla model, and fully integrated database capabilities.
The mass spec facility is performing proteomic analyses of whole cell lysates; proteomic analyses of whole cell lysates; quantitation using isotopically labeled growth media; targeted proteomics analyses of subcellular fractions; and nucleic acid analyses of RNA and DNA oligomers.
Wiley took his position as director in 2000, after 18 years as a professor in the University of Utah’s department of pathology, where he headed the cell imaging core facility.
“I came here and was given carte blanche to build a program in systems biology and exploit the national labs and implement what systems biology should look like — a multidisciplinary, team-oriented, big-scale science organization that can do things they can’t do in academia — like find a cancer cell, define the entire thing, and then find every gene, protein, and modification in it,” he said.
Wiley leads a team of 120 people, including 24 senior scientists. His BSI group is organized under the fundamental science unit of the 3,800-employee PNNL. PNNL has four directorates: fundamental science, national security, energy, and computer science and engineering.
To understand his group’s approach to molecular biology investigation, think chemical engineering, Wiley said. “We are strongly influenced by chemical engineers, and their sense of time and space,” he said. “The field has been developed to take processes in the lab and build them up to the scale of a chemical plant or an oil refinery.”
“The way that biology is done today is ethereal,” said Wiley. “We do one experiment at a time and the hope is that you can reproduce the experiment. What happens is that each one is done at once, and sequentially. Not only do you have the problem of doing the experiment but you have the problem of interpreting what changes are real and what isn’t real. The way around that is not to do them sequentially but to do them in parallel and you do all measurements on all cells simultaneously.”
No one else can do proteomics at higher sensitivity and the greatest coverage, he said. The group relies heavily on two technologies — high pressure liquid chromatography machines with columns of 20,000 pounds per square inch and a Fourier transform ion cyclotron resonance mass spectrometer. “We don’t run 2D gels, we just take proteins and shoot them in the mass spec, in work flows of 1,000 proteins per run,” Wiley said.
Nearly half of the laboratory’s mass spectrometers are automated to run 24 hours a day, and together produce some 50 to 100 million spectra a month.
The lab doesn’t take baby steps in its efforts.
“One project we have is investigating human response to EGF,” Wiley said. “We took large populations of cells under tightly controlled systems, stimulated them and tested them at eight time points, taking simultaneous measurements with flow cytometry, gene arrays, global proteomics, and 1,000 Western blots to generate a fairly substantial data set to put into structures.”
This data puts a premium on bioinformatics, he said.
“We can swamp the world with data, but you can’t get knowledge with data,” he said. “To do that, you have to put it in a structure.”
The laboratory is supported by PNNL’s supercomputer and an infrastructure that can handle the massive amounts of data produced by the group.
Still there are challenges.
“We don’t have the data to build a model that can predict,” said Wiley.
Does that mean even more data is needed? Yes, and it’s a trend that will continue for all molecular biology investigations, Wiley said.
“[Overall], the data in five years will be orders and orders of magnitude greater than all the data collected in the history of biology up to now,” Wiley said. “We struggle with how do we organize the information.”
The laboratory is developing a LIMS based on Nautilus technology from Thermo.
Earlier this month, PNNL received a $10.3 million biodefense contract from the National Institute of Allergies and Infectious Agents to identify the proteins that regulate the bacteria that cause salmonella poisoning, typhoid fever, and the monkey pox virus.
The five-year award is the Department of Energy lab’s third $10 million National Institutes of Health grant or contract in the past year and the second for Richard Smith, principal investigator and a Battelle Fellow.
What Smith would like, he told BioCommerce Week, is lots more throughput.
“We have higher throughput than any other lab in the world,” he said. “We’d like to have orders of magnitude high throughput than what we have today.”
The advantage Smith has is access to the engineers and designers of PNNL who can build complex new tools to streamline processes.
For sample preparation, the lab has developed a capillary liquid chromatography system.
“We have about 15 of them operating now that operate at very high pressures, about 10 times what a commercial system runs at,” he said. The systems cost about $80,000 to $90,000 each to develop, he said.
Additionally, the data management system the lab has developed allows investigators to not only look at the data coming from the mass spec labs, but also other data sets too.
The lab has, so far, been successful, he said.
“It’s a tough business, it takes a lot of different skills sets, a big team working together, something that hasn’t worked all that well in the academic labs where they do one thing, and do it well.”
Still, the lab is always seeking tools that it can add to increase its throughput. “If you have a toy that will get us there faster, we will talk to you,” said Karin Rodland, a scientist who joined the group in 2001 after working as an associate professor in the department of cell and developmental biology at Oregon Health Sciences University.
Fourier Transform Ion Cyclotron Resonance (FTICR) Mass Spectrometers
- 11.5-tesla, 205-mm bore mass spectrometer with Odyssey data station
- 9.4-tesla, 160-mm bore APEX III mass spectrometer with high-throughput capability
- 7-tesla, 150-mm bore mass spectrometer with Odyssey data station
- 3.5-tesla, 330-mm bore mass spectrometer with Odyssey data station
Ion Trap Mass Spectrometers
- Finnigan LCQ XPDECA Ion Trap mass spectrometer
- Two Finnigan LCQ Classic ion trap mass spectrometers
- Two Finnigan LCQ DUO ion trap mass spectrometers
- Finnigan MAT TSQ 7000 triple quadrupole mass spectrometer
- Sciex QSTAR quadrupole time-of-flight mass spectrometer