Drug developers have long faced a conundrum in choosing assays: the available tests for pharmacokinetics and toxicity are either fast, cheap, or predictive. Pick any two.
Traditionally, researchers have approached the problem by using relatively fast and inexpensive cellular assays to eliminate leads with blatant toxicities, then taking the remaining compounds into animal tests for a more nuanced view of their pharmacokinetics. Unfortunately, both steps of the process are fraught with pitfalls. Standard cellular assays may eliminate leads that would have turned out to be fine drugs, and even expensive animal tests provide only a hint of how a compound will behave in humans.
Now, progress in fields as diverse as computer science and stem cell culture have pushed more drug developers to consider a new path: high-throughput cellular assays that provide more information—and mimic human biology more closely—than some animal models. “I think this is something people started talking about several years ago, and recently I’m seeing it come into fruition more and more,” says Jim Cali, PhD, director of the assay design group at Promega in Madison, Wisc. Cali adds that “as we make it simpler and easier for people to make these measurements, and provide data that convince them that the [cellular assays are] reliable and predictive, they’re beginning to adopt these more and more.”
It’s all about content
Promega is certainly not the only vendor pushing cell-based assays. Indeed, Thermo Fisher Scientific’s Cellomics division in Pittsburgh, Pa. has promoted its own high content platform for years. However, the platform’s cost and complexity kept it out of the reach of many drug developers.
|
“Why isn’t everybody doing it? It’s not easy to use, it’s expensive, assay development takes some time, and then interpreting the [data] is a challenge for a scientist,” says Mark Collins, PhD, director of global marketing for Cellomics. In response to those complaints, the company has now introduced a streamlined version of the system called High Content 2.0. The new product line includes reagent kits, a simplified software interface, and redesigned hardware to make the most common types of fluorescence-based cellular assays easier. “The technological changes we made were directed at ease of use, robustness, and performance so that more people can gain the benefits of high-content cell imaging,” says Collins.
Among other changes, the new systems use LED lighting instead of a traditional arc lamp. “Instead of using white light, it generates light at the right color and wavelength for the biology that people want to do,” says Collins. He adds that the available wavelengths span about 85% of the experiments biologists are likely to do, without the variability of arc lamps. Basic researchers may still need the flexibility of the older platform, but the company is betting that many users will gladly trade flexibility for ease of use.
So far, that looks like a good bet. The first High Content 2.0 system the company introduced, called ToxInsight, has set toxicologists atwitter. “The reaction was great. The ToxInsight solution, with this simplified assay development workflow, stable solid state illumination, and speed, plus the informatics to make sense of the data … was kind of an answer to a prayer,” says Collins. Several pharmaceutical laboratories are now testing the system to validate it in their own toxicology labs, and Cellomics is introducing additional products for other user groups.
Shaping up the glutes
While easier high-content screening appeals to many toxicologists, new systems are unlikely to supplant more traditional cellular assays. For example, researchers have long used changes in glutathione levels as a surrogate marker for oxidative stress, a common feature of drug toxicity. However, conventional assays for glutathione use colorimetric or fluorescent markers, which can be cumbersome and error-prone.
|
“In those traditional assays, the process itself, given that it’s time-consuming, introduces artifacts that alter the actual concentration of glutathione,” says Promega’s Cali. Air oxidizes glutathione, so the longer a technician takes to perform the assay, the less accurate it becomes, and variations in technique from day to day or lab to lab can make the results hard to reproduce.
To address that problem, Cali and his colleagues developed a rapid, luciferase-based glutathione assay, which Promega now markets as GSH-Glo. “By shortening the procedure and simplifying it, we’ve turned it into an add-and-read style assay that’s amenable to high-throughput screening,” says Cali. The company’s researchers are also developing a companion assay that will measure both oxidized and reduced glutathione, providing additional details about the cell’s metabolic state.
Besides its expanding line of luciferase-based assays, Promega has also begun a marketing partnership with Celsis In Vitro Technologies of Baltimore, Md. Celsis harvests, tests, and cryopreserves primary human hepatocytes that drug development labs can use for pharmacokinetic and toxicity studies. Because they must be harvested from human livers and do not multiply in culture, primary hepatocytes are a scarce resource. As a result, scientists have been trying to optimize their hepatocyte assays to use as few cells as possible, while still achieving statistical significance.
“A few years ago people were reluctant to think about using primary hepatocytes even in the 96-well format, [but] that’s become routine,” says Cali, adding that “we’re now moving to 384-well formats, and the good folks at Celsis IVT are really spearheading that.”
Blood work
Hepatocytes aren’t the only human cells ADME/Tox labs are interested in. Myelosuppression, or blood cell toxicity, limits the dosing and potential for many marketed drugs, and kills many promising leads at the expensive preclinical or clinical trial stage. Unfortunately, developing a cell-based assay for myelosuppression has not been easy.
“The cells that we’re looking at, red blood cells, white blood cells, and platelets, have different lifespans,” explains Emer Clarke, PhD, chief scientific officer for ReachBio in Seattle, Wash. That variation means that cell culture systems are poor surrogates for real blood-producing bone marrow. Human bone marrow is relatively easy to get, but because of individual variations, researchers need to use marrow that’s been thoroughly tested. ReachBio specializes in exactly that, either performing myelosuppression assays on prequalified bone marrow as a contract research organization (CRO) or selling the prequalified marrow to companies that want to do the testing in-house.
|
In either case, bone marrow assays are not amenable to the kind of miniaturization that researchers are exploring for hepatocytes. That’s because the basic assay for myelosuppression is a colony-forming test. “We want the numbers of colonies that we’re looking at in a control versus a treated [dish] to be statistically relevant,” says Clarke, adding that “we like to get about 80 to 100 colonies per dish, and set up all the assays in triplicate per condition.” Those requirements have so far restricted myelosuppression assays to 35 mm dishes, and Clarke sees little prospect for pushing them into 96-well or smaller plates.
Besides testing for myelosuppression in drug candidates, some pharmaceutical companies are also using ReachBio’s assays to screen potential leukemia treatments, by distinguishing between cancerous and normal blood cell colonies.
Many companies are also interested in testing drugs in blood cells from other species. “If your toxicity tests are all done in mouse, but we’ve shown a differential between human and mouse, that information may also be taken into context as you move forward to a clinical trial on humans,” says Clarke.
Tracing the heart line
While donor-derived primary human cells are currently the backbone of many cell-based assay programs, researchers ultimately hope to replace that scarce resource with a renewable one, using human embryonic stem cells (hESC). Though stem cell science is still in its infancy, some tool companies are already on the verge of offering hESC-based assays to drug developers.
Using NIH-approved hESC lines from Geron in Menlo Park, Calif., scientists at GE Healthcare are now deriving human cardiomyocytes for drug toxicity assays. “We successfully brought their protocol and their cells to Cardiff [Wales, U.K.] in November of last year, and ever since then we’ve been growing the cells on a large industrial scale,” says Stephen Minger, PhD, research and development director for cell technologies at GE Healthcare in Cardiff.
Minger’s team can now grow large batches of hESC cells in defined medium, stimulate them with growth factors that simulate the development of the heart, and harvest functional cardiomyocytes. They then dissociate and freeze the cardiomyocytes. Even after months of cold storage, researchers can re-plate the frozen cells and watch them form a beating, heart-like muscle layer.
“We’ve kind of come up with a product that fits very neatly into [drug developers’] workflow. They can thaw the cells on Friday, and by Monday, Tuesday, or Wednesday they’re ready to use,” says Minger. The company plans to launch the cells in early 2011, allowing toxicologists to screen human cardiomyocytes as a routine part of early lead development.
Regardless of the specific strategy they advocate, developers of the new cellular assays agree that the field is poised to overcome some of the pharmaceutical industry’s thorniest problems. “I worked in Big Pharma for 15 years, [and] we developed a lot of drugs that cured mice,” says Thermo’s Collins. He adds that “implementing toxicity assays earlier in the drug discovery process [will have] an enormous impact.”
About the Author
Originally trained as a microbiologist, Alan Dove has been writing about science and its interfaces with industry and government for more than a decade.
Filed Under: Drug Discovery