Key words: cAMP • GPCR • enzyme fragment complementation • LEADseeker • assay miniaturization
G protein-coupled receptors (GPCRs) are the largest family of transmembrane receptors and are responsible for the transduction of a diverse range of extracellular signals. Up to 40% of clinically marketed drugs are active at this receptor family; however, these drugs exhibit their activity at less than 10% of known GPCRs. Thus, sensitive high-throughput GPCR screening assays are required to associate the many orphan GPCRs with disease to potentially identify novel pharmaceutical agents.
Here, we describe an assay for screening at GPCRs based on enzyme fragment complementation (EFC) technology and cyclic adenosine 3’, 5’ monophosphate (cAMP) measurement. The data show the cAMP II HitHunter Chemiluminescence Assay has a sensitivity of ~ 1 nM when carried out in 1536-well microplates and imaged using the LEADseeker™ Multimodality Imaging System.
LEADseeker Multimodality Imaging System 18-1140-71
cAMP II HitHunter Chemiluminescence 90-0032-03
Assay, 1536 wells
Other materials required
Phosphate buffered saline (PBS) solution
Dulbecco’s Modified Eagle’s Medium
Greiner Bio-one Lumitrac™ 200 microplates, 1536-wells
Biomek™ FX Laboratory Automated Workstation (Beckman Coulter)
Chinese hamster ovary (CHO-K1) cells
384-well Library of Pharmacologically Active Compounds (LOPAC) (Sigma)
384-well autofluorescence/colored compound plates (ChemBridge Corporation)
Each assay involved the addition of DMSO solution (2 µl, 1% final concentration), cAMP standard (2 µl, 1–770 nM), or, cell culture suspension (2 µl) prepared in physiological buffer such as Krebs-Henseleit buffer, PBS, or serum-free media (e.g. Dulbecco’s Modified Eagle’s Medium) without phenol red, and EFC detection reagents (2 ×2 µl) containing labeled cAMP (enzyme donor [ED] fragment), inactive EFC enzyme (enzyme acceptor [EA] fragment), cAMP monoclonal antibody, and cell lysis reagent. Assay plates were incubated overnight before measurement on the LEADseeker Multimodality Imaging System.
A standard curve for the cAMP II HitHunter Chemiluminescence Assay was carried out on Greiner Bioone Lumitrac 200 1536-well microplates and measured on the LEADseeker Multimodality Imaging System (2 ×2 binning with coincidence averaging) in luminescence mode (Fig 1a). The assay had a comparable limit of detection and working range as a model cAMP scintillation proximity assay (SPA) (RPA559) also carried out in 1536 wells with imaging using the LEADseeker instrument (Fig 1b). Sensitivity of detection was determined with assays carried out in PBS or Krebs-Henseleit buffer using a well-established method involving c AMP measurement at two standard deviations above the zero-dose of cAMP and interpolation from the standard curve (1). Using this method, the sensitivity of the cAMP II HitHunter Chemiluminescence Assay was ~ 1 nM.
DMSO and ethanol tolerance
DMSO and ethanol are commonly used to dissolve pharmaceutical agents. Dilutions of DMSO and ethanol were prepared in assay buffer to give a range of concentrations from 0.1% (v/v) to 5% (v/v) in the wells. The assay was shown to be highly tolerant (up to 2% v/v) of both solvents (Fig 2).
Response to agonist stimulation
Experiments with stimulated cells established in culture were carried out as follows: Chinese hamster ovary (CHO-K1) cells were seeded (2 µl, 106 cells/ml) into 1536-well cluster plates and stimulated with forskolin (2 µl, 1–100 µM) for 30 min (37 ºC, 95% humidity, 5% CO2), before addition of the cAMP EFC detection reagents (2 ×2 µl). All dispensing steps were carried out robotically either on a Biomek FX Laboratory Automated Workstation or on a CyBi-well system.
The results from forskolin-stimulated CHO-K1 cells are shown in Figure 3. The EC50 was 17.7 mM.
Assay performance was monitored by the use of forskolinstimulated CHO-K1 cells and Z’ was calculated as described by Zhang et al (2). Z’ values were comparable to those obtained using a cAMP SPA carried out on 1536-wells (Fig 4). Calculated Z’ values for the cAMP II HitHunter Chemiluminescence Assay were 0.61 and 0.64 for the SPA method.
Standard curves and tests involving isoproterenol, adrenaline, nor adrenaline, salbutamol, terbutaline, and fenoterol (10-9 to 10-4 M) induced cAMP generation from A431 cells were carried out using the cAMP II HitHunter Chemiluminescence Assay. Standards (2 µl, 1–770 nM) or cells, prepared in Krebs-Henseleit buffer (2 µl, 5 ×105 cells/ml), were incubated at room temperature (RT) for 30 min in the presence of target agonist, before addition of the EFC detection reagents. Plates were imaged for 30 s on the LEADseeker Multimodality Imaging System.
There was a dose-dependent increase in cAMP generation with the β2 adrenoreceptor agonist isoproterenol (EC50 = 4.3 nM), adrenaline (EC50 = 24.2 nM), noradrenaline (EC50 = 482.9 nM), salbutamol (EC50 = 255.9 nM), terbutaline (EC50 = 2142 nM), and fenoterol (EC50 = 221.5 nM) (Fig 5). EC50 values were broadly similar to those reported previously when SPA and fluorescence polarization techniques were used to measure cAMP induction with several of these agonists (3).
The response of A431 cells to isoproterenol was clearly inhibited by the b2 adrenoreceptor antagonist propranolol (IC50 = 12.99 nM) (Fig 6). Briefly, A431 cells (2 µl, 5 ×105 cells/ml) were pre-incubated (30 min, RT) with propranolol (1–300 nM) before addition of isoproterenol (2.5 µM, 30 min, RT), and EFC reagents (2 ×2 µl). Plates were incubated overnight before imaging on LEADseeker Multimodality Imaging system for 30 s.
Here we describe an approach that employed twenty 96-well compound library plates reformatted and dispensed into five 384-well source plates consisting 2 ×384 Library of Pharmacologically Acti ve Compounds (LOPAC™, Sigma) library plates and 3 ×384-well autofluorescence/colored compound plates (a custom set prepared by The ChemBridge Corporation, San Diego, CA, USA), giving 6400 tests in total, assayed over 5 ×1536-well plates.
The source 384-well plates contained each compound nominally at a concentration of 1 mM in DMSO (100%). For the assay, these agents were further diluted 1:25 with PBS, before a further dilution of 1:4 in the assay well, to give a final concentration of 10 µM in 1% DMSO. All dispensing was carried out robotically as described previously using CyBi-well.
Briefly, A431 cells (5 ×105 cells/ml) were harvested and resuspended in Krebs-Henseleit buffer. Two microliters were dispensed into 1536-well plates and stimulated (30 min, RT) with the target compounds, before addition of the 2 ×EFC detection reagents. Plates were incubated overnight before imaging on the LEADseeker instrument as described previously. Controls included cAMP standard curves (two per 1536-well plate), forskolin (0.78–200 µM) and isoproterenol (1–10 µM) stimulated cells, and unstimulated cultures.
The standard curve results are shown in Figure 7.
The results from the agonist screen are shown in Figure 8. The normalized response was derived from values obtained from negative control compounds. A test compound was considered to be a hit, when there was a significant increase in cAMP levels greater than three standard deviations above the mean (5.78% of LOPAC compounds were considered hits). Interference of the test compounds on the assay had been shown previously to be 3.6% with the ChemBridge library and 1.8% with the LOPAC compounds, used at the same concentration as above (data not shown).
A technique is described here for the direct measurement of cAMP in cell cultures. The assay has a wide dynamic range, is simple, very sensitive (~ 1 nM), reliable, and compared to other methods for cAMP measurement, is highly convenient. Typically only four dispensing steps (including cellular and compound additions) are required, thus reducing technical intervention and overall total assay time. The method is ideally suited to automation and miniaturization on 1536well cluster plates. The homogeneous nature of the assay means there is no need for separation of bound and free antigen and the non-radioactive approach eliminates the handling and disposal of radioactivity. Combined with imaging on the LEADseeker instrument platform, the assay has been shown to be well suited for high-throughput screening for compounds acting at GPCRs, at which assay results may be obtained from approximately 8000 data points in less than 5 min.
1. Davies, C. Concepts. In: The Immunoassay Handbook, 2nd Edition, (Wild, D., ed.) Nature Publishing Group (2001).
2. Zhang, J.H., et al., A simple statistical parameter for use in evaluation and validation of high-throughput screening assays. J. Biomol. Screen., 4: 67-73 (1999).
3. Horton, J.K., Capper, S.J., Price Jones, M., and Hughes, K.T. Assays for Drug-Screening Applications and Research, in The Immunoassay Handbook, 3rd Edition, (Wild, D., ed.), Elsevier, pp. 854–884 (2005).
back to top