Many G-protein coupled receptors (GPCRs) trigger, upon binding of an agonist, a transient increase in intracellular calcium concentration. This variation acts as an internal secondary messenger and is an important modulator of many physiological mechanisms (reviewed by Rink (1990), Tsunoda (1993) and by Santella & Carafoli (1997)). Measurement of intracellular calcium concentration in cells expressing a GPCR can thus be used to monitor the efficiency of activation of a GPCR by various compounds known or suspected to be a ligand for this GPCR.
One of the methods of choice (reviewed by Mottheakis and Ohler, 2000) for such measurements is the use of cell lines expressing a GPCR and aequorin, such as described by Sheu et al (1993) or Button et al. (1993). In this system, cells are incubated with coelenterazine, which is the co-factor of aequorin. During this incubation, coelenterazine enters the cell (it is lipophylic and readily crosses the cell membrane) and conjugates with apoaequorin to form aequorin, which is the active form of the enzyme. When the cells are then exposed to an agonist of the GPCR, intracellular calcium concentration increases. This increase leads to the activation of the catalytic activity of aequorin, which oxidizes coelenterazine and yields apoaequorin, coelenteramide, CO2 and light. The intensity of light emission is proportional to the increase in intracellular calcium in the physiological range (Rizutto et al., 1995). Thus, in this system, measurement of light emission following agonist addition reflects its ability to activate the GPCR. Because light is emitted for only 20 to 30 seconds after activation of the GPCR, recording of the emitted light must be performed during the few seconds following agonist addition to the cells. This flash-type signal reflects the transient increase in calcium concentration following GPCR stimulation.
THE CELL LINES
Cell lines (usually CHO-K1, but other ones in some cases) were stably transfected with plasmids intended for expression of apoaequorin and of a GPCR. After selection with antibiotics, recombinant cells were subjected to a limit dilution, and clones expressing the G-protein coupled receptor and apoaequorin at a high level were selected. If the G-protein coupled receptor is not naturally coupled to a calcium signalling pathway, a universal coupling effector is coexpressed in order to redirect the coupling towards intracellular calcium release. This universal coupling effector is usually the Gα16 protein (Milligan et al., 1996), that was shown to be able to couple many GPCRs to the calcium pathway.
HOW TO PERFORM THE ASSAY
As explained above, detection of putative agonists by means of mammalian cell lines expressing apoaequorin and a GPCR requires the measurement of the emitted light to be performed just after placing the cells in contact with the supposed agonist. This can easily be measured at low throughput using a single-tube luminometer but until recently could not be considered for HTS because luminometer dispenser designs make it impossible to inject a different drug into each well.
Euroscreen has developed a method1 for performing highthroughput screening of drugs binding to GPCR by the use of mammalian cell lines expressing apoaequorin and a GPCR and by the use of a conventional luminometer. Following this method, the solutions to be tested for agonistic activities are placed in the wells of a 96-well plate. Cells expressing apoaequorin and a GPCR are detached from the culture plate (if they are adherent) and are incubated with coelenterazine to reconstitute active aequorin. These are then maintained in suspension with a stirrer and the cell suspension is injected, well by well, on the solutions of tested compounds. Light emissio n is then recorded usually for 20 seconds. By injecting the same cell suspension into each of the 96 wells, this method avoids the need to wash the dispenser(s) between each measurement and allows 96 measurements of agonist-induced aequorin light emission in 32 minutes with a single-dispenser luminometer. However, a six-detector MicroBeta JET can perform 96 measurements in six minutes or up to 10,000 measurements in one day.
Once the cells have been incubated with coelenterazine, they can be used for several hours and even several days for the measurement of agonist-induced increase of intracellular calcium concentration. A signal-tonoise ratio above 50 was currently obtained with this system of cell injection.
SUITABLE PERKINELMER LIFE SCIENCE INSTRUMENTATION
COST OF THE METHOD
The assay costs are minimal. In a typical experiment, 50 x 106 cells are resuspended in 10 ml of buffer and coelenterazine is added at a final concentration of 5 μM. After reconstitution of active aequorin, the cells are diluted 10 times before use, and 50 μl of the dilution (i.e. 25 000 cells) are injected per well. The coelenterazine loading of cells in suspension and at high density avoids the big consumption of coelenterazine that would be necessary if the cells were loaded while attached to the plates. This renders the method very cheap. The cost of coelenterazine is only 0.014 US$/well (approximately 0.015 Euros or 0.025 DM). The volume of the cell suspension necessary to fill the dead volume of the injectors of the luminometer is usually between 1.5 mL (single injector devices) and 6 ml (6 detector MicroBeta JET); thus it costs between 0.4 and 1.70 US$ to fill the dead volume.
Plates: Either solid white or white-side / clear bottom 96-well plates can be used. The approximate cost per well is 0.04 US$/well
Cell s: only 25, 000 cells per well are required, this is an approximate cost of 0.01 US$/well.
Total cost per measurement is:
ADVANTAGES OF THE AEQUORIN-BASED ASSAY:
High signal-to-noise ratio (≥ 50).
Robust assay: everything at room temperature, a single preparation of cells can be used for several hours or even several days.
Homogeneous assay (no separation step) thus limiting the handling time and device requirements to perform the assay.
Low cost (10 times less expensive than luciferase assay).
Universal coupling to calcium for most of the receptors through co-expression of an adequate signalling protein. This allows screening of a target without previous knowledge of the specific signalling pathway.
Short exposure time of cells to compounds (≤ 30 seconds compared to several hours for transcription-based assays): allows testing of the activity of cytotoxic compounds. avoids metabolization of the tested compounds.
Ca++ mobilization is an early event following the binding of an agonist. No artefact due to the transcription of a reporter gene (no bell-shape curve for dose-effect measurements).
EC50 values are similar to those obtained by classical functional assays.
This functional screening assay has been tested with success on receptors belonging to different families including
Cannabinoid receptor CB1
Chemokine receptors: CCR1, CCR2, CCR3, CCR5, CCR6, CCR8, CXCR4
Corticotropin-releasing factor CRF2 Histamine H1 receptor
Melanin-concentrating hormone (MCH) receptor (SLC-1, HNOP801)
Melanocortin receptor MC4
Nociceptin re ceptor ORL-1
Orexin receptors 1 and 2 Purinergic P2Y2
Serotonin receptors : 5-HT1B, 5-HT2B
This system can be used to detect either agonists or antagonists. Surmountable antagonists as well as non-surmountable antagonists are detected.
With a throughput of up to 1000 assays per hour, it is suitable for large-scale drug screening.
EXAMPLES OF RESULTS
This functional screening assay has been tested with several kinds of G-protein coupled receptors. EC50 values are similar to affinity values obtain by radioligand binding and/or other functional assays.
Human chemokine CCR5 receptor:
A CHO cell line expressing the chemokine CCR5 receptor, the Gα16 coupling protein and apoaequorin was established. Cells were prepared as described above and were injected onto a series of dilutions of known ligands placed in the wells of a 96-well plate. The emitted light was immediately recorded for 30 seconds. After reading the first well, cells were injected into the next well and emitted light was recorded, etc. For each plate, a series of curves representing the intensity of the emitted light as a function of time for each well was displayed (figure 2). The intensity of the emitted light was integrated over 30 s using the MicroLumat Winglow software, yielding for each well one value representative of the emitted light, and hence of the stimulation of the CCR5 receptor by the agonist present in the well. These values can be plotted against the logarithm of the ligand concentration to generate a doseresponse curve as shown in figure 3. These allow the determination of half-maximal response doses (EC50) for each ligand. For the generation of these data, 288 measurements were performed in less than 3 hours using the MicroLumat single-detector, singleinjector luminometer.
CHO-CCR5-Gα16-apoaequorin cells were injected onto increasing (from column 1 to 12) concentrations of RANTES (rows A and B, same dilutions in duplicate), MIP1α (rows C and D, duplicates), MIP1β (rows E and F, duplicates), and another analog of RANTES (rows G and H, duplicates). The emitted light was recorded for 30 seconds and the intensity of the emitted light was plotted as a function of time. The scaling is the same for all the graphs presented here.
Agonist-induced light emission in CHO cells expressing the CCR5 receptor
Human chemokine CCR3 receptor:
A K562 cell line expressing the chemokine CCR3 receptor, the Gα16 coupling protein and apoaequorin was established. Dose-response curves were obtained as described above and results are shown in figure 4.
Human chemokine CCR8 receptor:
CHO cells expressing the chemokine CCR8 receptor were transfected with a plasmid encoding the Gα16 coupling protein and apoaequorin and stable transfectants were selected with antibiotic. The mix of clones obtained from this transfection was used to generate a dose-response curve with I-309 as described above. Results are shown in figure 5.
Human serotonin 5HT2B receptor:
A CHO cell line expressing the serotonin 5HT2B receptor, the Gα16 coupling protein and apoaequorin was established.
A. Detection of agonists
Dose-response curves for agonists of the 5HT2B receptor were obtained as described above and results are shown in figure 6A.
B. Dectection of antagonists
For t he detection of antagonists, CHO-5HT2B- Gα16-aequorin cells were incubated with various concentrations of the compounds to be tested. An agonist (α-methyl-5-HT) at a single concentration was then injected on the mixture of cells + antagonist and the emitted light was recorded and integrated for 30 seconds. Non-surmountable antagonists (Methiothepin, Methysergide) as well as surmountable antagonists (Mesulergine, Mianserin, Ketanserin) prevented, in a dosedependent manner the activation of the 5- HT2B receptor by α-methyl-5-HT (Figure 6B). Similar results were obtained with other antagonists (Ritanserin, rauwolscine, Yohimbine, Spiperone, SB206553 not shown).
PRODUCTS AND SERVICES PROVIDED BY EUROSCREEN
Custom cloning of the G-protein coupled receptor and establishment of a stable cell line, expressing the receptor and apoaequorin.
Selection of the most appropriate calciumcoupling system, if necessary.
Pharmacological characterization of the receptor, development and characterization of the functional assay in a 96-well plate.
A control cell line expressing apoaequorin only is provided (on request) to validate the specificity of the functional response.
Button, D. and Brownstein, M. (1993) Aequorin-expressing mammalian cell lines used to report Ca2+ mobilization. Cell Calcium 14, 663-671.
Milligan, G., Marshall, F., and Rees, S. (1996) Gα16 as a universal G protein adapter: implications for agonist screening strategies. TIPS 17, 235-237.
Mottheakis, L.C. and Ohler, L.D. (2000) Seeing the light: calcium imaging in cells for drug discovery. Drug Discovery Today: HTS supplement 1, (1) 18- 19.
Rink, T.J. (1990) Receptor-mediat ed calcium entry. FEBS Lett. 268, 381-385.
Rizzuto et al. (1995) Photoprotein-mediated measurement of calcium ion concentration in mitochondria of living cells. Methods in Enzymology 260, 417-428.
Santella, L., Carafoli, E. (1997) Calcium signaling in the cell nucleus. FASEB J. 11, 1091-1109.
Sheu, Y.-A., Kricka, L.J., Pritchett, D.B. (1993) Measurement of intracellular calcium using bioluminescent aequorin expressed in human cells. Analytical Biochemistry 209, 343-347.
Tsunoda, Y. (1993) Receptor-operated Ca2+ signaling and crosstalk in stimulus secretion coupling. Biochim. Biophys. Acta 1154, 105-156.