The quest for new G-protein coupled receptor targets is challenging the limits of today's cell-based assays

Lori Valigra
Valigra is a freelance writer based in Cambridge, Mass. 

The strong kinship of cell-based assays with conditions in human cells has made them a mainstream tool for revealing the complexities of popular G-protein-coupled
click the image to enlarge 

This image shows the role of beta-arrestin in a model of GPCR desensitization and resensitization. (Source: Molecular Devices Corp.) 
receptor (GPCR) targets for drug discovery. Pharmaceutical companies expect GPCRs to average about 40% of their drug discovery efforts for years, even though kinases are coming on strong. The Human Genome Project identified upward of 1,000 GPCRs, 400 of which are nonolfactory and thus potentially druggable targets. Of those 400, about half are orphan GPCRs whose ligands have yet to be identified.

Drug companies like those numbers. GPCRs also are popular because they are membrane proteins located on the cell surface, they have second messenger amplification, and they are proven targets in a wide range of major disorders including heart disease, obesity, cancers, pain, and diabetes.

Several events drove the increased use of cell-based assays to study GPCRs. Around three years ago, once genomic data was available, many people became interested in orphan GPCRs, and the only way to look at orphan GPCRs is in cell-based assays, says Richard Eglen, PhD, chief scientific officer, DiscoveRx Corp., Fremont, Calif. The company offers HitHunter cyclic AMP assays that use enzyme fragment complementation. "The other factor is a lot of technologies became more robust and easier to handle on high-throughput automation systems for cell-based assays in general." Eglen adds that researchers also realized the complexity of GPCR signaling over what was thought to be a simple cyclic AMP or calcium pathway, so it was necessary to work with an intact cell.

Targeting higher-hanging fruit
With so much focus on GPCRs, many of the easier targets already are being researched, leaving only those which are more difficult to study and orphan GPCRs available. That means the limits of traditional approaches such as radioligand binding and calcium assays are being challenged. Large pharmaceutical companies and assay companies alike want to make incremental improvements to current assays and to add new technologies to get more functional and other data on GPCRs. They want to be able to work more easily with multiple assay technologies. Multiplexing and reverse transfection are some of the newer techniques being employed now. In the future, the move toward miniaturization, computer modeling, in silico screening, biosensors, and even GPCR crystallization will become more pronounced as those technologies mature. Crystallization, which could yield a wealth of information, has proven challenging because GPCRs are difficult to purify. Only one GPCR, the widely studied rhodopsin, has been crystallized to date.

"More and more of the low-hanging fruit in terms of the basic GPCR that expresses at high levels on the cell surface and signals through a very nice, clean calcium flux
click the image to enlarge 

This image shows cell arrays for GPCR structure-function analysis. Cells were reverse-transfected with a plasmid encoding a GPCR and detected using fluorescent imaging. The last three wells received cells but not DNA, so exhibit no fluorescence. After 24 hours of expression, live cells were imaged using Alpha Innotech's Array 7500i. (Source: Integral Molecular) 
assay, for example, is being fully exploited. The new targets are becoming more difficult to work with," says Benjamin Doranz, PhD, president and chief scientific officer of Integral Molecular Inc., Philadelphia. The company has both a cell-based reverse transfection technology and a lipoparticle alternative to a cell-based assay that uses critical parts of a GPCR, so it can be purified for use on biosensor. "We're already at the point where some GPCRs are very difficult to discover drugs against, and the traditional cell-based assays are not working. If you don't see those GPCRs failing, it's because they don't even make it into the pipeline. New cell-based assays will tackle those kinds of issues."

GPCRs are difficult to purify and solubilize, and working with them within cells is not always easy. Doranz says existing tools for working with GPCRs are limited compared to those for kinases, nuclear hormone receptors, and phosphatases, which generally are soluble and can be crystallized or their NMR structures can be derived.

There are even more basic issues. "Cell-based assays still are not what one would consider completely industrialized in a screening setting," says Deborah Brusini, business director of screening and research reagents for PerkinElmer Life and Analytical Sciences, Boston. "There still are a lot of issues around growing cells, keeping them standardized across several sites, doing it in high volume and manipulating them, and in particular trying to miniaturize these assays." Some companies are trying to create cells or cell lines that could be frozen, for example, then split up and standardized across multiple sites. PerkinElmer has not done this kind of work yet.

The company is focusing on making its new Lance TR-FRET (time-resolved fluorescence resonance energy transfer assay) technology an industrialized functional assay for big pharmaceutical and biotechnology companies that are running million-point screens and want to do so in very small volumes.

PerkinElmer also is researching aequorin, a technology that allows cells to be suspended in a mix rather than attached to a plate such as with a calcium flux assay. This month, the company plans to introduce an instrument called LumiLux Cellular Screening Platform that integrates the liquid-handling manipulations, injection of the compounds, and the detector that reads the reaction. Brusini said it is ideal for aequorin technology, because it will be easier for the customer to get cells into the plates to run their assays. "The technology also doesn't use as many consumables, and customers can reduce the time it takes to run these assays by about 30%, increase throughput, and get higher quality because they don't need to go through all manipulation steps with cells. It's potentially a big step up for big pharma because of the volumes they have to push through."

Robust, reliable assays
The challenges to developing good assays include robustness, general applicability to various GPCR pathways and signal-to-noise issues, says Dominic Behan, PhD, co-founder, chief scientific officer, Arena Pharmaceuticals Inc., San Diego. Arena has its own constitutively activated receptor technology (CART) and melanophore assays that it uses in its drug discovery programs. The best assay for a particular GPCR depends on the receptor being targeted and which second messenger response to measure. "CART allows Arena to determine early on the G protein that a GPCR couples to and therefore the second messenger response that it transmits in a cell," Behan says. "That leads us to the choice of appropriate assays to adopt for that particular receptor system."

Arena's CART approach genetically modifies receptors so they can constitutively signal in the absence of a natural ligand stimulation. Once the receptors are altered they can be transfected into standard mammalian cell types such as 293 cells and screened for small molecules. Arena's other technology area, which Behan characterizes as a leap forward, is the melanophore system, which is so sensitive that it does not require genetic modification of the receptors to see constitutive activity.

Melanophores are frog skin cells. Depending on the second messenger signal produced within the cell, a pigment contained in melanosomes will disperse throughout the entire cell when the cyclic AMP levels rise, or aggregate toward the cell's center if the levels fall. It's a simple measurement of absorbance changes. "The system is very high throughput, robust and extremely reliable. It's universally applicable to many GPCR coupling pathways such as Gs, Gi, and Gq," Behan says. Arena can screen as many as 300,000 compounds in a week or two using a 394-well or 1,536-well format.

Targeting orphan GPCRs
Molecular Devices Corp., Sunnyvale, Calif., which made a name for itself with a number of cell-based assays including its fluorescence imaging plate reader (FLIPR) technology, recently adopted a complementary technique called Transfluor that it bought from Xsira Pharmaceuticals Inc., formerly Norak Biosciences, Morrisville, N.C. Whereas FLIPR has been widely used in high-throughput cell biology applications, Transfluor is used in imaging applications and is useful for orphan GPCRs.

FLIPR looks at live physiology, the signaling downstream from GPCRs as it is happening. The method has established the upper threshold of throughput with live cell
click the image to enlarge 

Novasite's screening system analyzes the GPCRs on thousands of individual cells per second, determining whether each cell is "on" or "off." The large number of discrete data points generated by this method yields signal to noise values of up to 80:1 and results in very small standard errors. (Source: Novasite Pharmaceuticals Inc.) 
assays on up to 75,000 wells per day, says Michael Sjaastad, PhD, director of marketing for imaging products at Molecular Devices. FLIPR is historically used to look at calcium signaling when a GPCR is activated, but Sjaastad says it would be useful to follow up that information with Transfluor technology to look at the desensitization process downstream of the calcium signaling. The company is currently working on ways to use the same cells on both the Transfluor and FLIPR instruments so it can get more comparable data. "We know which assays would be complementary to run on both," says Sjaastad.

Transfluor is a cell-based fluorescence bioassay based on the discovery that virtually all GPCRs, upon activation by ligand binding, undergo rapid deactivation or desensitization by a common pathway. An early step in the pathway is the binding of the cytoplasmic protein arrestin to the activated receptor. By attaching a fluorescent label to arrestin, it is possible to monitor arrestin translocation within the cell and thus detect the activation of any GPCR. The beta arrestin is labeled with green fluorescent protein (GFP), a protein in cells that also acts as the probe, so it is not necessary to buy a fluorescent dye to run the assay each time.

Transfluor also can help researchers improve their success rate with orphan GPCRs. Ralph Garippa, PhD, research leader, Roche Discovery Technologies, cell-based HTS and robotics group, Nutley, N.J., has looked at a number of different technologies to elaborate on orphan GPCRs. Orphan GPCR experimentation is difficult because it lacks positive controls. With known GPCRs, a peptide-based or small molecule-based ligand can be used as a trigger for an agonist or an antagonist experiment, says Garippa. "Without having that positive control, it makes assay development very difficult," he says. Transfluor takes advantage of a fairly universal process in GPCR signaling, the desensitization mechanism. "We felt that whether or not we knew what the endogenous ligand was or what the G-protein coupling was, that we could use this desensitization mechanism to identify small-molecule compounds or even peptides that would act at these orphan receptors."

Smaller for higher throughput
Invitrogen Corp. has a wide range of assays including GeneBLAzer, a functional assay that lends itself to miniaturized, high-throughput settings. It uses beta-lactamase as a reporter gene, and it can read out cyclic AMP or calcium signaling pathways in addition to some of the lesser used pathways in GPCR signaling. Its FRET-based
Allosteric and Dimeric GPCRs
Two emerging areas of G-coupled-protein receptor (GPCR) research are allosteric and dimeric GPCRs. "The allosteric field has come quite far in the past 5 to 10 years with being able to characterize modulators of GPCRs at allosteric sites," says Ralph Garippa, PhD, research leader, Roche Discovery Technologies, cell-based HTS and robotics group, Nutley, N.J. "The advantages [are] that you could find better specificity for your GPCR at an allosteric site because they're less conserved."

Novasite Pharmaceuticals Inc., San Diego, developed an integrated "single cell structure-function" approach to enable the discovery of allosteric modulators and difficult agonist drugs targeting GPCRs. John Ransom, PhD, vice president of biology, says the company's flow cytometry system can interrogate each cell individually, whereas with a FLIPR or a fluorometer, the entire population is averaged. "We've found a plate reader system like a FLIPR simple cannot detect a response in the cells if less than 25% of the cells are responding. But if you do a single-cell analysis and look at each individual cell, we can go down to 1% or less detection, and we can process thousands to tens of thousands of cells per second."

Alan Binnie, PhD, a distinguished scientist at Sanofi-Aventis, Tucson, Ariz., put the Novasite technology to a rigorous test several years ago with about 7,000 compounds, some targets and some controls. "I selected the absolute worst compounds we had as far as causing false positives and nonselectives, and embedded a set of control compounds. I was very surprised they were able to rapidly figure out where almost every one of the embedded control compounds was, as well as identify a new selective series, which we didn't expect to happen."

In terms of dimers, Dominic Behan, PhD, co-founder and chief scientific officer, Arena Pharmaceuticals Inc., San Diego, said his company licensed a technology from the University of Glasgow that can measure receptor dimerization. "It allows you to measure two receptors coming together by using various G protein chimeras, so we can measure homodimerization or heterodimerization using that assay system." He says it's still unclear what the overall implications of dimerization of GPCRs are to drug discovery and pharmaceuticals.
(fluorescence resonance energy transfer) substrate can load readily into living cells, and it gives a ratiometric readout that minimizes noise, says Bonnie Hanson, PhD, senior scientist, Invitrogen, Madison, Wis. "It really gives you very robust assays that can be miniaturized down to 3,456-well plates, and it's adaptable to high-throughput screening." Invitrogen also has a new substrate that gives a cytotoxicity readout with GeneBLAzer at the same time as the assay readout.

Like many of her industry colleagues, Hanson believes there will continue to be a mix of assays used in drug discovery and research. "With binding data, all you know is that your compound binds, but you don't know whether it's an agonist or an antagonist or an inverse agonist or partial agonist. That type of information needs cell-based assays."

No one-size-fits-all assays
Christine Williams, PhD, associate research fellow and section head of the hit discovery group at Pfizer Global Research and Development, Sandwich, UK, emphasizes that not all GPCRs are equal, so there's no one-size-fits-all assay or single philosophy used by her team, which has been developing assays for high-throughput screening for about seven years. Pfizer uses a lot of cell-based reporter gene assays at their Sandwich laboratories because they're cost effective, easy to automate, and it is easy to look at the correct biological signaling for the target of interest, Williams says.

"If you're just looking at the level of a ligand-binding assay, you're looking at the attraction of two molecules. Depending on your choice of ligands, you may miss things. Cell-based assays can be much more akin to the tissue- or in vivo -relevant system. They can find other types of compounds, for example, allosteric compounds acting at a slightly different site within the receptor." One example her lab is working on is the CCR5 receptor as a target for HIV-mediated virus entry. The lab developed a complex cell-based assay that uses a reporter gene that relies on fusion between two cell partners: one cell line expressing the GP120 and GP 41 viral glycoproteins and a second cell line containing the CCR5 and CD4 receptor mediators at that fusion event.

"You can actually look at expression of the reporter gene—in that case, beta-galactosidase—in response to the fusion between those two cell lines. This assay offered us significant advantage over a CCR5 binding assay." The assay enabled her lab to identify at least two potential more novel series than would have been possible with the binding assay that was run many years ago [J. Bradley et al., Journal of Biomolecular Screening, vol. 9, pp. 516-524 (2004)].

More reverse transfection data 
Integral Molecular is currently developing a newer approach called reverse transfection cell-array technology, a technique originally described by Massachusetts Institute of Technology researcher David Sabatini [J. Ziauddin, D. M. Sabatini, Nature, vol. 411, pp. 107-110 (2001)]. Integral Molecular president Doranz said his company devised its own approach, which is similar and involves introducing DNA into cells in a way done opposite to usual methods. That is, instead of adding DNA on top of cells, it puts DNA on a surface and then adds cells to the DNA. That way, the DNA will dry in a specific position, and arrays of DNA can be made on the surface with each encoding a different gene or variant of a gene. That means thousands of transfections can be done in parallel simply by adding cells, a dramatic increase in productivity and decrease in time over the current methods of transfections one by one.

The company is using the technology for structural activity relationship analysis. "We can take a membrane protein CXCR4 or CCR5 and mutate it a thousand different ways so that every amino acid is changed at least once. We put all those variants on a single array and make multiple copies of that array and ask questions about the structure and function of that protein," Sabatini says. "So we can get information about the contribution of every residue in the GPCR to that specific function or structure. This process enables you to do experiments you otherwise wouldn't be able to do."