Science has intrigued Jonathan Wingfield, principal scientist, AstraZeneca, UK, since he was a boy, which is why he obtained a Ph.D. in microbiology. But something happened during his post-doc at Children's Hospital in Cincinnati, OH, that convinced Wingfield to reconsider his decision to pursue basic academic research, and ultimately led to his being honored as the 2015 SLAS Innovation Award winner.
"I remember vividly when I decided to go into drug discovery," Wingfield recalls. "I was doing basic research on transcription factors in the hospital's research lab. That day, I walked into the cafeteria—there was only one for both staff and patients—and sat down among all the children, so many of whom were seriously ill. I remember thinking to myself, 'I've had nine years of education, but I'm not making a difference. I have to do something more productive and helpful.' That was the turning point. Looking around at all those children made me want to get a job in a pharmaceutical company, where I could actually have some impact on the next generation of drugs that were coming through."
Wingfield eventually took a position at AstraZeneca, where he became more and more involved in the application of technology support to drug discovery R&D. "Working with technology, I have the ability to influence many different projects and potentially contribute something of value to all of them—with the focus of developing products that will ultimately end up in the clinic and be of real value to patients."
Understanding both the technologies used in drug discovery and the needs of the biology enabled Wingfield to envision a way to hasten the discovery process in a cost-effective way. "Having the ability to directly measure substrate-to-product conversion with label-free assays in a high-throughput environment has long been considered the 'Holy Grail' of drug discovery screening," he explains. But high-throughput screening for small-molecule modulators of soluble enzymes, for example, generally relies on tagging or labeling the molecular targets with fluorescent or luminescent assays. Such labeling may alter the behavior of the targets and compounds may simply interfere with the binding of an antibody to a substrate. Genuine hits may be hidden within the "noise" of false positives, and any hits that are identified need to be confirmed by other methods to ensure they are real.
Labeling also is costly, Wingfield observes. "Running high-throughput screens under 10 cents per well doesn't sound expensive, but if you run 30 HTS on a million samples a year the costs go up significantly. Much of the cost can be labeled reagents. Last year alone, we invested somewhere in the range of $140,000 in labeled reagents for a single high-throughput screen and that only gave us the capacity to screen a fraction of our compound library." If a high-throughput screening group has a certain amount of money budgeted annually, it can either run more samples against fewer projects, or more projects with fewer samples, he explains. "In an ideal world, we'd like to run every target against all the samples. To do that, we have to drive down the cost."
The answer, says Wingfield, is to find "a good, sensitive label-free technology that would be fast enough to use in a high-throughput screening environment, enabling us to have that direct measurement of substrate-to-product conversion in as simple a way as possible to minimize the number of artifacts and keep costs down."
Mass spectrometry (MS) is used across a wide range of drug-discovery processes at AstraZeneca, from immediate post-high-throughput screening to toxicity screening and screening of clinical samples, Wingfield says. As a label-free technology, it has a low risk of artifacts and a relatively low cost, because it does not require the use of complex—and expensive—assay reagents.
Mass spectrometry also can detect a wide range of end points, including proteins, lipids, sugars and DNA. "Traditionally we use isolated enzyme and substrate to build biochemical assays as a primary HTS; mix the two together in the presence of the required co-factors; and look for substrate-to-product conversion in the presence or absence of a particular compound. Then we move into a cellular environment, effectively looking for the same endpoint within the cell." Wingfield explains. "With a mass spectrometer, we can do the same cellular experiments, but also look for other endpoints. If we were interested in a compound that inhibits an aspect of metabolism, for example, we can measure its effects on lactate, fructose, glucose and other potential endpoints in a single sampling."
But from a high-throughput screening perspective, conventional mass spectrometry has limitations, Wingfield notes. "Traditional mass spectrometry relies on liquid chromatography (LC), which adds a significant time burden to the sampling process—anywhere from one to 15 minutes per sample. Screening a million samples would take about two years if you had only one LC/MS machine. One answer is to have rooms full of LC/MS instruments and reduce your time by processing samples in parallel. But of course, this creates a significant capital burden that includes servicing and maintenance costs, as well, and requires a reasonable amount of space." There are other forms of MS that can provide the throughput to meet HTS needs, such as matrix-assisted laser desorption ionization (MALDI); however, this also has limitations. When matrix is used to assist in desorption and ionization, the matrix itself can confound the data that is being generated.
The issue came to a head when AstraZeneca's high-throughput screening group was told it was moving to a new research center in Cambridge, UK. The new center will be smaller than the existing one, so buying more instruments wasn't an option, Wingfield explains. Also, the company was asking its team members to "drive technology development to enable scientific excellence," Wingfield says, "because we believe that through scientific excellence we will find the next generation of patient care and therapy." And so, following the late creative thinking guru Michael Vance's definition of innovation as "the creation of the new or the rearranging of the old in a new way," Wingfield and his colleagues took an "old" technology—acoustic droplet ejection—and figured out how to apply it in a new way.
Acoustic ejection technology was being used for various aspects of compound management at the company for close to a decade, "so we see it as something of an old technology," Wingfield says. But that technology has several advantages. First and foremost, "the speed, accuracy, precision and robustness of acoustic dispensers have been proven." In addition, it enables researchers to work from a relatively small volume and with a range of buffers. "We can fire acoustic droplets from a 384-well plate at a very high frequency, even if we only have a working volume of 2-3 microliters. This means sampling rates around one sample per second are possible."
The "aha" moment that led to the idea of putting the two technologies—mass spectrometry and acoustic droplet ejection—together came when a mass spectrometrist working in AstraZeneca's compound management section was looking at images generated by Labcyte's acoustic technology. He noted that the ejected droplets looked "surprisingly like a standard mass spec electrospray," Wingfield recalls. "Seeing a striking similarity between the two gave us the idea of combining them. In principle, that integration would result in a system capable of delivering about 4000 data points per hour into a high-sensitivity, label-free detector. It could enable sampling from 1536-well plates and reduce the total assay volume required to less than five microliters."
To develop the new system, AstraZeneca spearheaded a three-way collaboration that included Wingfield and his colleagues, the Labcyte team for its acoustic technology and a team from Waters for its mass spectrometry expertise. The collaborators began work on the new platform in 2013, and have since developed several iterations of a prototype acoustic source linked to a mass spectrometer. (Listen to Wingfield's award-winning presentation for details). Since March 2014, the group has achieved the following:
The three-way collaboration is an unusual one for AstraZeneca. "We don't normally go down the road of creating new technologies, but the company recognized that working this way gave us an opportunity to do some really great, innovative science—and that great science will lead to the discoveries that will ultimately result in the next generation of patient care," Wingfield says. "We have a team of experts in instrumentation working with a team of experts in the acoustic space, and our team at AstraZeneca brings the application piece to the table. It's a collaborative model that's very different from simply hiring vendors and letting it go at that. And, as a result, we're in a very good position to build an instrument that can have an impact in the field of drug discovery, and potentially areas beyond the life sciences, as well."
The team's focus for 2015 is on biochemical high-throughput screening. To validate the application, Wingfield and his colleagues will do a pilot study to compare results of a traditional high-throughput screen with the mass spectrometry endpoint. Some 60,000-70,000 compounds will be run through two types of biochemical assays—an indirect labeled assay and a label-free MS assay. "The results should give us a good benchmark," Wingfield says. "If the pilot is successful, we will look to implement a full high-throughput screen early next year."
In addition, the team has started to use the new technology to perform quality control on the company's compound collection. "To validate the compound structures, we need to get samples—effectively, chemical structures in a solvent—into the mass spectrometer quickly, and analyze their mass spectrometry signatures. It's a good starting point because it's something we can do relatively easily without a lot of resource requirements."
From there, the group is likely to move to a cell-based assay with a more complex biological matrix and a simple analyte endpoint. "If we can achieve it, the step after that would be to move on to metabonomics, where we will be looking for a complex set of analytes within a complex biologic matrix," Wingfield explains. Other drug-discovery applications could include ADME/Tox, protein ligand and protein/protein interactions and early kinetics.
Wingfield and his colleagues are open to additional partners and collaborators. "Even though it is still the very early days for this project, part of why we went to SLAS2015 was to let people know what we're doing so they can share in the work and contribute to it if they want to," Wingfield emphasizes. "We have the basic platform, and now we'd like to enable academics and others who are interested in the system to access the platform, so we can all work together to identify where the most benefit could be delivered from this technology.
"The ability to load samples into a mass spectrometry detector at a high rate from significantly reduced assay volumes has significant potential not only within drug discovery but other areas of industry," Wingfield continues. "Dynamic fluid analysis—the ability of the acoustic injector to adjust automatically to varying viscosities and surface tensions of the sample—allows the generation of droplets from a wide range of fluids including blood, plasma, cell culture medium, acid digests and chemical syntheses. So, we envision lots of potential applications."
Now On-Demand at SLAS.org: Novel Acoustic Loading of a Mass Spectrometer—Towards Next Generation High-Throughput MS Screening
Jonathan Wingfield's 2015 SLAS Innovation Award-winning presentation is available to the scientific community at SLAS.org. Wingfield discusses the benefits and challenges of developing and testing the new technology, as well as plans for future applications. SLAS members can view for free.
In the SLAS Journals
Be the first to know when the JBS Special Issue on Advances in Mass Spectrometry within Drug Discovery by guest editors Jonathan Wingfield and Ian Wilson becomes available at JBS Online. Sign up for e-Alerts today!
Learn more about ADE when the JALA Special Issue on Advancing Scientific Innovation with Acoustic Droplet Ejection becomes available. Sign up for e-Alerts today!
May 22, 2015