Cell-signalling research aims to understand how cells convert extracellular signals into the required cellular responses, and how these pathways can go awry and cause disease. With the emergence of cell signalling as a significant field in its own right over the past few years, a dizzying array of reagents, assays and technologies are now available to help researchers (see ‘In the market place’).

Cell-surface receptors

The large and diverse family of G-protein-coupled receptors (GPCRs) responds to a wide range of agonists, including amines, hormones, neurotransmitters and even light. This sheer diversity makes GPCRs a rich source of potential drug targets.

“We spent a lot of the 1990s, as did others, isolating novel genes encoding these GPCRs, which essentially had eluded classical pharmacology,” says Brian O'Dowd of the department of pharmacology at the University of Toronto, Canada. But many of these are still ‘orphan’ GPCRs, for which a natural ligand has not been identified.

To try to tap their therapeutic potential, O'Dowd's group has developed a multipurpose cellular assay (MOCA) that can be used to screen for compounds that activate or block any GPCR, and that needs no prior knowledge about the G proteins or second messengers involved. O'Dowd's lab jokingly refers to this as the ‘mother of all cellular assays’. The technology is being commercialized by Toronto-based PatoBios, of which O'Dowd is a co-founder.

The GPCR is genetically modified to incorporate a nuclear-localization sequence, which causes the receptor to be internalized and translocated to the nucleus, from which it is unable to recycle to the cell surface. The interaction of a ligand with the modified GPCR prevents translocation, and the receptor is retained on the cell surface. The different distributions of the receptor can be visualized by various methods such as fluorescent tags. The assay can also be adapted to detect ligand binding to the important family of transporter proteins involved in the re-uptake of neurotransmitters.

Activated receptors show up as dots. Credit: R.H. OAKLEY/NORAK BIOSCIENCES

Norak Biosciences of Research Triangle Park, North Carolina, has also developed a cell-based fluorescence assay for finding ligands for orphan GPCRs. Its Transfluor technology detects the binding of β-arrestin to the cytoplasmic part of the receptor, which occurs only after a ligand has bound. Activated receptors can then be detected and isolated. The assay has been validated on various image analysers, including the IN Cell Analyzer 3000 from GE Healthcare, Little Chalfont, UK, the ArrayScan from Cellomics, Pittsburgh, Pennsylvania, and the Opera high-throughput confocal microscopy platform from Evotec of Hamburg, Germany.

Blocking on cue

The intracellular interaction between the GPCR and its associated G protein is also a potential drug target. This is the approach of drug-discovery company cue BIOtech of Evanston, Illinois, which is also marketing its Minigene vectors. These deliver small peptides that block signal transduction through a specific G protein by competing for the site where the G protein would normally bind to the receptor. “You can turn off each G protein individually inside the cell,” says Annette Gilchrist, president and founder of cue, which licensed the technology from Northwestern University. The company offers both plasmid-based cDNA vectors for transient transfection and retroviral vectors for hard-to-transfect cells.

Pure and simple

Many potential drug targets, such as GPCRs and ion channels, are transmembrane proteins, and their isolation and study is difficult as they often need to remain embedded in a lipid membrane to maintain their structural integrity.

Cells treated with a cue BIOtech Gq Minigene vector (left) respond less strongly than normal cells (right) to stimulation by thrombin. Credit: CUE BIOTECH

By exploiting the fact that cell-membrane proteins are incorporated into the surface envelope of budding retroviruses, Integral Molecular of Philadelphia, Pennsylvania, has developed a way of isolating membrane proteins from cells while preserving their native structure. Cells producing non-infectious retroviral particles are also engineered to make high levels of the desired membrane protein. As the virus core buds from the cell it is surrounded by cell membrane enriched with the protein. The resulting Lipoparticles marketed by the company are non-infectious spheres (100–150 nm in diameter) of retroviral core protein surrounded by a phospholipid bilayer containing around 100 molecules of the desired membrane protein in its native conformation. The particles are stable when frozen and refrigerated. “Generally, the preparations that we make are greater than 100 picomoles of membrane protein per milligram of total protein,” says Benjamin Doranz, president, chief scientific officer and founding partner of Integral Molecular.

“The most complex protein that we've obtained is a 14-spanning membrane protein amino-acid transporter,” says Doranz. But there will be limits, he says, such as membrane proteins with extremely long cytoplasmic tails that interfere with viral assembly or those that do not traffic to the cell surface.

Lipoparticles can serve as a source of homogeneous and structurally intact membrane proteins for high-throughput screening, monoclonal antibody production and structure analysis using X-ray crystallography. They are also being paired with optical biosensors, such as the surface plasmon resonance (SPR)-based detectors developed by Biacore, of Stockholm, Sweden, for the kinetic analysis of membrane protein interactions with antibodies and ligands.

Characterizing kinases

Protein kinases are garnering increasing attention as drug targets, particularly in the light of the recent success of Gleevec, a tyrosine kinase inhibitor developed by Swiss-based pharmaceutical company Novartis, which was approved for the treatment of chronic myelogenous leukaemia in 2001. Gleevec represents a new class of drugs that disrupt components of the intracellular signalling pathways that cause cells to grow and divide uncontrollably, giving rise to cancers.

There are more than 500 protein kinase genes in the human genome. They catalyse the specific phosphorylation of proteins and play an essential role in many signalling pathways, including those involved in cell-cycle control.

One way of evaluating the substrate selectivity and function of protein kinases in a high-throughput format is by peptide microarrays to which the kinases will bind selectively. Pepscan Systems of Lelystad, the Netherlands, launched its PepChip Kinase peptide microarray about a year ago. “Currently, we have 1,150 different substrates,” says Jos Joore, vice-president of array technology at Pepscan, all derived from a public database. The microarray can be used for substrate profiling of known and unknown kinases and for specificity testing of kinase inhibitors. The company is also turning its attention to proteases. Longer term, Joore says, it is trying to improve the specificity of its peptide substrates by providing them as constrained loops, or a combination of constrained loops, rather than linear peptides.

The porous PamChip surface provides a greater area for protein immmobilization. Credit: PAMGENE

The three-dimensional surface of the PamChip from PamGene of 's-Hertogenbosch, the Netherlands, is designed to allow peptide substrates to be deposited at higher concentrations than conventional arrays. “We can immobilize much more material per square millimetre than other flat materials,” says Rob Ruijtenbeek, PamGene's head of kinase research. A 500-fold increase in reactive surface compared with two-dimensional arrays is claimed.

At the heart of the system is PamGene's 5D-Pulse flow-through microarray technology, in which peptides are covalently immobilized using inkjet technology onto the porous microarray surface via the peptide amino terminus. Sample is then pumped back and forth through the porous material to facilitate mixing. The pumping-cycle frequency can be changed and detection is with fluorescent antibodies using a CCD camera/microscope. The system provides kinetic readouts in which substrate conversion is monitored over time; traditional array formats limit detection to a single time point.

In June, PamGene announced plans to join forces with Jerini Peptide Technologies (JPT) in Berlin, Germany, which allows PamGene to marry its microarray platform with JPT's comprehensive peptide sets for kinase profiling. Zeptosens of Witterswil, Switzerland, is similarly increasing the range of its arrays by selling antibodies developed by Cell Signaling Technology of Beverly, Massachusetts, for use with the Zeptosens planar waveguide detection protein microarray platform. With this technology, only fluorophores located at or near the surface of the waveguide are excited and signals from unbound molecules in the bulk solution are not detected.

According to Peter Oroszlan, director of business development at Zeptosens, this technology provides a significant increase in signal-to-background ratios, enabling detection of low-abundance proteins such as signalling molecules. “Our system allows you to measure 600 protein molecules in a spot, which corresponds to one zeptomole [10−21 moles],” he says. ZeptoMARK protein microarrays are available in capture or reverse-array formats (see Nature 429, 102; 2004), and applications include expression monitoring of proteins during drug profiling and pathway mapping, monitoring activation-state markers, such as phosphorylated proteins, and the study of disease progression.

Read-out

Biosensors for tagging proteins to monitor their activation levels and distribution continue to get more sensitive and versatile. The impact of green fluorescence protein (GFP) and its variants in studying cell signalling is immense. Biosensors that use GFP can, for example, detect conformational changes in proteins in response to ligand binding, changes in protein localization or changes in protein activity.

Lighting-up protein interactions: Klaus Hahn (left) and Alexei Toutchkine with the dye on which their biosensor is based.

New fluorescent dyes are also being developed. Klaus Hahn and his colleagues at the University of North Carolina School of Medicine, Chapel Hill, recently reported a biosensor that can visualize the natural dynamics of an unlabelled endogenous intracellular signalling protein, the GTPase Cdc42, in living cells. The sensor is composed of the Cdc42-binding fragment of the Wiskott–Aldrich syndrome protein (WASP), to which a novel merocyanine dye has been coupled. The dye is sensitive to changes in hydrophobicity that occur at the interface between the interacting proteins. Binding of WASP to GTP-activated Cdc42 causes the dye to fluoresce. The sensitivity provided by direct excitation of a novel fluorescent dye enables detection of protein activation at native levels.

One of the more novel detection platforms is quantum dots (see ‘Quantum dots show their true colours’, page 247); another is single-molecule photon stamping spectroscopy, which is being used to study the dynamics of the interactions of single proteins (see ‘Probing real-time protein interactions’).

LI-COR Biosciences of Lincoln, Nebraska, offers two-colour near-infrared fluorescence detection of signal transduction events. The firm originally applied its infrared detection technology to western blots, “but a western blot is not particularly convenient for looking at a pathway”, says Michael Olive, LI-COR's vice-president of research and development. The company has developed the In-Cell Western assay for quantifying proteins in fixed cultured cells in 96- or 384-well microplates in less time than conventional western blots by bypassing the need for lysate preparation and the use of gels and membranes. The use of two spectrally distinct near-infrared dyes effectively doubles the number of endpoints that can be analysed, enabling, for example, the measurement of both phosphorylated extracellular signal-regulated kinases (ERKs) and total ERK protein at the same time.

With slight tweaking, the ‘In-Cell Western’ can become the ‘On-Cell Western’. This was developed by James Wager-Miller in the department of anaesthesiology at the University of Washington in Seattle, who is using it to follow the internalization and recycling of GPCRs, in particular the cannabinoid receptor 1, to and from the cell surface. The hope is that this will lead to a better understanding of how these trafficking events can lead to the desensitization of cells following prolonged or repeated exposure to agonists.

The In-Cell Western is amenable to automation and this month the company will launch a new two-colour plate reader called Aerius, which automates the assay. This may take the technology “into the realm of lead validation”, says Olive, helping to prevent costly drug failures later on.

The next challenge for cell signalling will be to look at cellular behaviour on a global scale and for this further improvements in technology will be needed, along with better computational and mathematical tools for deriving information about complete signal-transduction networks.