Sept 15, 2005 | Can cell systems biology rescue drug discovery? That's the ambitious title to a June 2005 Perspective in Nature Reviews Drug Discovery  written by Eugene Butcher, a professor of pathology at Stanford University and cofounder of BioSeek, a small company that thinks the answer to that question, at least in part, is yes.
Founded in late 2000, BioSeek is the brainchild of Butcher and Ellen Berg, also a cofounder and its chief scientific officer. Berg led the efforts to develop BioMAP - biologically multiplexed activity profiling - the company's core cell-based assay technology. The basic idea is to put a small number of primary human cell types, say endothelial cells and blood mononuclear cells, into a small number of different environments, each with a different blend of activators for known pathways, and then screen compounds in each.
"The reality is that drugs work in a very complex setting," says Berg. "Instead of the typical screening drug discovery where the approach is to isolate a pathway, isolate the target, and use that for screening drugs, we turn on many pathways at once. Then we perturb them with the drug and measure a set of readouts. It turns out that changes in the protein level in the assay are surprisingly reproducible and robust, and it's due to the reality that complex systems actually behave quite reproducibly."
Somewhat unusually, the impetus for BioSeek's approach was not derived from ongoing work in an academic lab or other company. Rather, Berg had been mulling over how to improve assay effectiveness. Prior to founding BioSeek, she was at Protein Design Labs and involved in discovery and development of antibodies for treatment of inflammatory diseases. Berg holds patents in inflammation biology and cell adhesion.
'Poised for Growth'
Writing in Nature Biotechnology in October 2004 , Butcher, Berg, and Eric Kunkel, head of biology and assay development at BioSeek, described a panel of just four different cell cocktails able to reflect the distinctive contributions of many targets relevant to inflammation. Profiles of as few as 24 to 40 proteins (cytokines, chemokines, etc.), they reported, could discriminate most pathways and mechanisms affected by known modulators of inflammation and an array of other drugs and pathways.
|Algorithms Are Part of the|
Two ordinary dual-processor Dell
servers handle all of the computation
and database management necessary
to make sense of BioSeek's assays.
Now, after roughly five years of platform development and building a database of assays, the Burlingame, Calif.-based BioSeek seems poised for growth. In March, the company received a milestone payment from Dynavax Technologies for an ongoing collaboration in which BioMAP was used to characterize the biological activity of a novel family of TNF-α inhibitors. Dynavax is researching orally available small molecules for chronic inflammatory diseases.
CEO Peter Staple says BioSeek has more than 15 collaborations under its belt, including three with top 10 pharmas. After "working in garages on friends' and family money" for a couple of years, the company raised venture capital twice - once in early 2002 and again in 2003 - for a total of $14 million. Staple says he plans to cautiously seek more funding. The company's headcount remains a lean 14, with the research staff divided evenly between "biologists, assay people, and computational IT."
The low burn rate and doubling of revenue last year are propelling BioSeek rapidly towards a cash-neutral basis, says Staple. Like most young biotechs, BioSeek's revenue comes largely from fee-for-service work and R&D collaborations, but the company has internal drug development aspirations.
Says Staple: "I just want to emphasize we're really focused on compounds. We're not starting from looking at pathways and targets in the abstract. We're looking at, from a practical standpoint, the area pharma is most interested in, which is how to identify compounds that have the most chance of success because they have a greatβ activity."
BioSeek's Scientific Advisory Board (SAB) is also impressive, including, for example, Lee Hood, founder of the Institute for Systems Biology, and Judah Folkman of Harvard Medical School, a pioneer in anti-angiogenesis research. Berg says Folkman was "very helpful when we were first developing the assays, correlating what we found in vitro to in vivo." Butcher chairs the SAB.
| ||Ellen Berg on Systems Biology|
Read on for Ellen Berg's view of
the systems biology landscape.
The really hard work was designing the assays, which took about two-and-a-half years. Along the way, BioSeek screened thousands of compounds. "We've been able to identify profiles that tell us about mechanisms of action, so when we look at a new compound, we can make the interpretation immediately. The other advantage is [that] in complex systems, where you have multiple pathways stimulated, all the pathways are 'talking' to each other, and so these assays are surprisingly sensitive to many different mechanisms of action," she says.
Learning to work effectively with primary human cells was another difficult but necessary challenge, since cells cultivated for years in incubators have often lost important regulatory pathways, says Berg.
"The hard part is finally done," agrees Berg. BioSeek now has 15 high-throughput assays that cover a lot of the biology and many tissues and disease states. "The easy part now is the profiling, and we're running up to 100 compounds a week, in multiple systems, so you get an entire profile from your compound on a weekly basis."
Sometimes the assays turn up pleasant surprises. One client was using a pathway-based screen to select compounds for an inflammation therapeutic area, with a known inflammatory pathway. They'd gotten active hits but didn't understand the mechanism of action. BioSeek profiled the compounds through its core set of assays and blasted the profiles against its database.
"We discovered their compounds were not specific for that particular pathway, but appeared to be similar to a cancer compound, suggesting a cancer target. When we got back to them with the data, they were at first very skeptical because they were an inflammation group and, of course, they were looking for the inflammation targets," says Berg.
But the skepticism disappeared when the client's oncology group at a different location actually had the screens for this cancer target up and running and confirmed that the compound was active against that cancer target. "Once you do something like that for a partner, their eyes open, and they recognize the value of the kind of screening we do that assesses many targets across multiple therapeutic areas," says Berg.