Scientists from Harvard Medical School have developed a method for performing fluorescence lifetime imaging with a multi-well plate reader and demonstrated how the technique can be used to measure protein-protein interactions and protein conformational changes in cells in a high-throughput manner.
The researchers are now hoping to commercialize the method as an industrial-scale drug-discovery tool, most likely by partnering with either Tecan or Edinburgh Instruments, the only companies that currently manufacture plate readers that can perform the technique. But at least one of those shops questions whether a market exists for the method.
At the heart of the assay, which is described in the September/October issue of the Journal of Biomedical Optics, is a combination of Förster resonance energy transfer (FRET) and time-domain fluorescence lifetime imaging (FLIM).
Basically, in FRET a researcher tags one molecule with a “donor” fluorophore and another molecule with an “acceptor” fluorophore, whose excitation wavelength strongly overlaps with the emission wavelength of the donor. Upon donor excitation, if the tagged molecules are very close to one another — less than 10 nm apart, as in a molecular interaction — the donor will excite the acceptor, and the donor gets dimmer while the acceptor gets brighter.
Additionally, researchers can use FLIM to tell if the donor has transferred its energy. In this case, the donor is excited with an ultra-short laser pulse, and its fluorescence decay is measured with extremely sensitive instrumentation that performs time-correlated single photon counting. If the tagged molecules are interacting and FRET is occurring, the donor’s fluorescence decays more quickly. In addition, the magnitude of this decay reveals the extent to which the molecules are interacting.
Each of these techniques is powerful on its own, according to Phill Jones, a neurology research fellow at Harvard Medical and Massachusetts General Hospital and first author of the study. But when combined they can provide an extremely detailed profile of a molecular interaction — not only whether two molecules are interacting, but how many and how closely they are interacting.
According to Jones, “this could be very important in disease processes because, for example, if you have a protein interaction that you wish to disrupt, you may want to know whether you’re reducing the number of molecules that are interacting, or if you’re making it a looser interaction,” he said.
“Similarly, the technique can be used to sensitively detect conformational changes in proteins that have been labeled with multiple fluorophores,” he added.
The Harvard researchers, as well as many others, have used this technique to analyze molecular interactions in cells with extreme sensitivity using a microscope. But in order to make the method useful for high-throughput drug discovery, it needs to be adapted to a well-plate format.
When Jones and colleagues started their work, only one plate reader existed – the Tecan UltraEvolution with an added fluorescence-lifetime module – that could make the measurements they needed. However, it was incapable of performing the relatively complex data analysis required.
Therefore, the Harvard team developed its own analysis program that enabled them to apply their technique on a well-by-well basis by, among other things, eliminating complex background auto-fluorescence and separating multiple fluorescence lifetimes.
“This kind of technique is already used in FLIM microscopy,” Jones said. “People do these cell studies – and there is quite a bit that has come out of our lab – where they use similar mathematical techniques to what I’ve done in this paper, but on a microscope.
“Obviously if you want to do high-throughput screening, you can’t be doing microscopy,” he added. “It involves finding cells, and making the measurement – it may take weeks to make these measurements. But with the plate reader, we’ve worked out how to do this on a high-throughput basis.”
The researchers tweaked their technique so that it allows an entire 96-well plate to be read in about 10 minutes. Furthermore, they used the method to detect the interaction in vitro between amyloid precursor protein and presenilin 1 (PS1), two proteins crucial in Alzheimer’s disease plaque formation; and to detect conformational differences between wild-type and mutant PS1 conformation.
FRETting Over Market Size
Jones told CBA News that he and his colleagues believe that the method will eventually be commercialized into a high-throughput drug-discovery tool. But first, the researchers must find a partner willing to take the plunge.
“I’ve spoken with Tecan about commercializing the method, and they’re very interested in the whole technique,” Jones said. “But we perhaps need to show more examples in print of how you can use the technique. I’m going to be publishing something soon where I’ll be using it to demonstrate the differences between isoforms of proteins of various disease processes. Hopefully that will attract more interest.”
Klaus Doering, a systems engineer for Tecan in Europe and Jones’ main contact in the company, this week told CBA News that it’s not a question of whether the technique works – it does, he said – but a question of whether there is a market for it.
According to Doering, Jones approached Tecan with the modifications and the idea of collaborating on commercializing the tool.
“At the moment, it seems to be difficult to establish the idea of fluorescence lifetime as a readout [method] in the pharmaceutical market.”
“Our response was that for our market field, it was too specialized,” Doering said. “At the moment, it seems to be difficult to establish the idea of fluorescence lifetime as a readout [method] in the pharmaceutical market.
“And then going a step forward [with] a very specialized system [with] multiwell-plate, fluorescence-lifetime, cell-based, FRET assays – all that together will be very difficult to establish in the market,” he added. “There will be two or three laboratories in the world that will happily adopt that. But for the broad market, it’s probably too early.”
Doering also said that Tecan would be happy to supply its fluorescence-lifetime UltraEvolution instrument to another entity, such as a start-up company, that might want to take the Harvard team’s software modifications and run with a new product. “I believe the idea is very good,” Doering said. “But Tecan is not a start-up business. We can’t quite do these very high-risk [ventures].”
Jones declined to disclose any specific alternative plans for commercializing the technique, but he may not be out of partnering options just yet. Although Tecan initially offered the only TCSPC-capable microplate reader on the market, another company out of the UK called Edinburgh Instruments – the type of start-up that Doering referred to – is currently beta-testing a dedicated TCSPC fluorescence lifetime plate reader called the NanoTaurus.
According to Edinburgh Instruments’ website, the system is “currently undergoing stringent validation tests” in collaboration with dye manufacturing, labeling, drug-development, and assay-development companies, and “will soon be commercially available.”
Edinburgh Instruments is one of three nascent Scottish biotech firms participating in the ITI Life Sciences program, a Scottish government-backed initiative designed to expand Scotland’s global presence in the life sciences marketplace. Last year, ITI Life Sciences announced that it had initiated a £3.7 million ($6.9 million at the time) research program to develop 3D cell-based drug-screening technology based in part on the technology used in Edinburgh’s NanoTaurus (see CBA News, 2/15/2005).
Jones said that his group is now working with the NanoTaurus, and that it is a better instrument than the Tecan. “It gives a much better signal to noise, and the background is much improved on it,” he said. To be fair, the UltraEvolution, unlike the NanoTaurus, is not specifically designed to accommodate the Harvard team’s relatively complex method.
“The reason this seems complicated at the moment is because it’s in the early stages,” Jones said. “It certainly can be adapted into something that’s higher throughput, and as easy to use as one of the fluorescence intensity plate readers that are on the market – but using a more sensitive technique with more information available.”