Stanford University scientists have developed a cell-based multiplexing technique that may be able to drop antibody and other reagent consumption in flow-based cellular analysis 100-fold, as well as substantially decrease the time it takes to acquire flow cytometric data.
The work on the technology, called fluorescent cell barcoding, also brings a dimension of high-content analysis to flow cytometry, and may eventually be attractive to industry flow-cytometry players that could further develop the technique for drug discovery applications or even to screen compounds and monitor drug efficacy using patient blood samples.
The method, which is described in the May issue of Nature Methods, was developed by Peter Krutzik, a graduate student in the laboratory of Garry Nolan, an associate professor of genetics at Stanford University. It builds on research conducted in Nolan's lab a few years ago, when he and colleagues developed a technique that used multiparameter flow cytometry to simultaneously detect activated kinase members in various cell signaling pathways within subpopulations of complex cell populations (see CBA News, 5/4/2004).
"It would be easy, even for a novice flow cytometry laboratory with standard equipment sitting around, to do up to 25 or 30 samples at once."
This technique already brought a multiplexed element to flow cytometric analysis because Nolan and colleagues were able to understand complex signaling pathways using flow cytometry data collected from cells that had been polychromatically labeled with several fluorophores.
"It was about having a new way to measure effects, and many of the effects of drugs we work on, of course, work on signaling pathways," Nolan said this week, referring to the previous work. "The most tenable measure of signaling is phosphorylation — that's the most stable and tenable immediate effect you can measure in terms of what a cell is thinking at any given point of time."
Nolan and colleagues later went on to show that the process could be automated — "up to 50 or 100 steps at a time," Nolan said — so that all the linkages could immediately be understood using computational analysis.
Now, in the most recent work, Nolan and Krutzik have upped the ante by developing a barcoding technique in which cell samples are pre-tagged with serial dilutions of anywhere from 5 to 7 fluorescence dye intensity levels, per dye color. This, in turn, allowed them to assign a "barcode" of sorts prior to flow cytometric analysis so that multiplexed measurements of single cells could be made in multiple cells using flow cytometry. The end result, in the case of three fluorescent tag colors, for instance, is the ability to analyze hundreds of samples simultaneously.
"It enables incredible assay-to-assay reproducibility, because basically you take all your samples, mix them in one, add the same amount of reagent, and then deconvolute the results and basically figure out what was what later," Nolan said. "It basically lets you run all of your controls, all of your experimental samples, all in the same tube."
Similar barcoding techniques have been applied to bead-based assays before, but never at the level of multiplexing found in this live-cell assay work, according to Nolan.
One of the major benefits of the technique is that if applied properly, it may be able to save core flow cytometry labs, or really any clinical or drug-discovery lab, a significant amount of money due to decreased reagent consumption.
"It would be easy, even for a novice flow cytometry laboratory with standard equipment sitting around, to do up to 25 or 30 samples at once," Nolan said. "When you're talking about reagents, such as antibodies, that could cost $200, $300, or even $1,000 for a big project — dropping that price down even tenfold basically means that things that only my laboratory can do, because we're well-funded for this area, now become possible for any laboratory." In fact, Nolan said that one of the reasons they developed the method is that reagent costs even for his lab were getting "completely out of control," despite vendor discounts.
And although the technique could be immediately beneficial to academic or other modestly funded laboratories, it should be applicable in industry endeavors, as well. "The way that this process works is the more samples you're doing at the same time, the cheaper it gets," Nolan said. It's really the principles of scale that actually work for us here. At its best, we can probably drop prices of antibodies and reagents 100- to 300-fold."
The most immediate application is for high-throughput drug-discovery; for instance, interrogating modifiers of kinase pathways in a highly multiplexed fashion, in the natural milieu of the cell, with the throughput that is naturally afforded by flow cytometry. The Nature Methods paper demonstrates proof of principle with a highly accurate screen of a small-molecule library for inhibitors of T-cell receptor and cytokine signaling.
A Link to Targeted Medicine?
The barcoding technique might also play a role in the development of targeted therapies. In an accompanying review article, James Jacobberger, an oncology professor at Case Western Reserve University, wrote that "part of the excitement among investigators pursuing cell-signaling cytometry is the possibility that unique patterns of signaling in precancer states and cancer may provide diagnostic information important for targeted therapy."
In addition, "the correlated patterns of phosphorylation-specific states of signaling molecules may provide pharmacodynamic measures in clinical trials, patient monitoring, and thus, 'personalized medicine,'" Jacobberger wrote.
"We've been doing this in primary cells, and it's so high-throughput that we're starting to apply it to patient samples," Nolan said. "What's the drug's effect on patient samples? You can't get any better than that —the on-target and off-target effects of three different drugs. Not just 'Hey, is the kinase on or is it off? But are several kinases on or several of them off?'"
Nolan said his lab has applied for patents on the method, and that they are likely to pursue commercialization. The work may also have attracted the attention of several stalwarts of the flow cytometry industry. Nolan also said that he has worked a lot in the past with Becton Dickinson, which has "expressed some interest" in the technique, and that another company, which he declined to identify, is also interested. Calls to BD seeking confirmation were not returned in time for this publication.
The next step for Nolan and colleagues, however, is to develop analysis software that can adequately analyze the highly multiplexed data that would be produced from such an assay.
"I've got two sub-groups in the lab working on different elements of how you get beyond standard flow cytometry software," Nolan said. "Immunologists are used to using fairly locked-in-place ways of viewing cell populations. But [we've] come at this thinking about how to express high-content information in a way that people can understand — in a visually compelling and interesting way that people can appreciate it."
— Ben Butkus ([email protected]eweb.com)