Abstract
by Kelly J. Cassutt, Ph.D

We present a multiplexing method for HTS screening: loading the same cells using both the UV dye Fura-2 AM and No Wash Membrane Potential Dye. Multiplexing assays measure orthogonal cell responses (calcium mobilization and membrane potential) using one volume of library compound. FDSS6000 imaging-based plate reader for cellular assays (Hamamatsu Photonics Systems, Bridgewater, NJ) uses Xenon lamps, enabling the use of fluorescent dyes with excitation wavelenghts from UV to visible. System optics read up to two excitation wavelengths and two emission wavelengths in the same assay. Current HTS strategies use one dye per assay, such as Fura-2-AM, Fluo-3-AM, or Fluo-4-AM (calcium mobilization), or DisBAC2(3) (membrane potential).

Figure 1. Same cells dye loaded using both calcium and membrane potential dyes. Fluorescence of calcium mobilization (a) and membrane potential (b). Arrow marks injection time. For both panels agonist 1 (Green) induces calcium mobilization and membrane potential depolarization. Agonist 2 (Orange) induces calcium mobilization and membrane potential hyperpolarization. Blue traces are buffer only. (n=4/group, subtract bias reading 10, negative control correction algorithm on). Click here to enlarge.

Introduction
Multiplexing in cell-based HTS (simultaneously collecting two unique results from double fluorescent dye loaded cells) is available using the FDSS6000. The science supporting multiplex assays is established, with several groups reporting results using double-dye loaded cells for studying either calcium channels or G-protein coupled receptors for calcium mobilization and membrane potential, including Fura-2AM/DisBac2(3),(1) Indo-1AM/DisBac2(3),(2) Fura-2AM/DiBac4(3)(3) and Fluo-4AM/Di4 ANEPPS.(4) These assays use single channel spectrofluorometers for readout. Spectrofluorometers provide UV to IR excitation wavelengths for the listed dye pairs, not transferable to laser-based HTS platforms (monochromatic excitation only) or LED based HTS platforms (UV power not sufficient). Xenon lamps provide the same wavelengths and power as spectrofluorometers.

This protocol provides for simultaneous loading of cells using Fura2-AM (Invitrogen Corp., Carlsbad, CA) and No-Wash Membrane Potential Dye (Molecular Devices Corp., Sunnyvale, CA) to read calcium mobilization and membrane potential changes for two agonists and two antagonists.

Materials and methods
A recombinant adherent cell line was plated and incubated overnight at 37 C, 5% CO2. Cells were extensively washed and left with a 30 ml residue. Fura-2 AM was diluted with an equal volume of 20% pluronic acid then diluted to 3.6 mM. To each well we added 15 ml for a final dye concentration of 1.2 mM. Cells were incubated with Fura-2 AM for 1 h, RT. Following washing cells were left with a 30 ml residual volume; to the cells we added 15 ml No Wash Membrane Potential Dye, to 0.5 3 stock concentration. After a 15 min incubation RT cells were loaded onto the FDSS6000.

Instrument optics were set to sequentially read ex 380 nm/em 540 nm for calcium mobilization (Fura-2 AM) and ex 480 nm/em 540 nm for membrane potential, 200 ms integration time each pair, 1.2 sec cycle time. (N.B. -No signal crosstalk is expected as signals are collected sequentially not simultaneously). For Fura-2 at ex380 em540 calcium mobilization decreases fluorescence; with membrane potential depolarization increases fluorescence signal and hyperpolarization decreases fluorescence signal. Following a baseline reading 5 ml of agonist (103) was added at 20 ml/s at a height of 2 mm above the cells (agonist). For antagonism studies compounds were pre-incubated with cells (5 ml at 103 to 45 ml cells); 5 ml of agonist (113) was added at 20 ml/s at a height of 2 mm above the cells.

Figure 2. Figure 2. Same cells dye loaded using both calcium and membrane potential dyes. Agonist potency, Hill Slope depends on readout. Calcium mobilization EC50 is 17 times more potent than membrane potential EC50. Click here to enlarge.

Figure 3. Same cells dye loaded using both calcium and membrane potential dyes. Compounds show (a) similar antagonist IC50 potency (calcium mobilization) yet (b) disparate antagonist potency IC50 (membrane potential). Click here to enlarge.

Results
Multiplexing differentiates agonist responses
The results in Figure 1 compare measurements of calcium mobilization and membrane potential using two agonists. As expected both agonists mediate calcium mobilization (Figure 1a); interestingly there is diametric difference in membrane potential responses as Agonist 1 hyperpolarizes cells while Agonist 2 depolarizes cells (Figure 1b). Taken together the results suggest multiplexing provides more information about agonist effects on cells than calcium mobilization or membrane potential alone.

Multiplexing differentiates agonist potency
Figure 2 presents Agonist 1 dose response curves measuring both calcium mobilization and membrane potential. There is over a 17 fold difference in EC50 depending on readout. Hill Slope is nearly 5 times greater for membrane potential (Hill Slope =5) as compared to calcium mobilization (Hill Slope =0.9). These differences, which suggest two mechanisms of action for Agonist 1, are identified using multiplexing, arguing that for completeness assay development should investigate both calcium mobilization and membrane potential.

Multiplexing differentiates antagonist potency
Results in Figure 3 compare two antagonists against Agonist 1 at EC50 (membrane potential) and EC90 (calcium mobilization). Both compounds show equivalent antagonist potency IC50 (about 3 fold difference) when measured using calcium mobilization. By contrast there is a nearly 8,000 fold difference in antagonist potency IC50 when membrane potential readout is used. These results show how multiplexing using an HTS platform can clearly differentiate antagonist potency, in the same cells, at the same time.

Summary
Simultaneous calcium mobilization and membrane potential data collection has no impact on assay cost apropos cells, plates, compounds, and run time. Assay development is minimal as both No Wash Membrane Potential Kit5 and Fura-2 AM(6,7) are well established dyes for FDSS6000. Not washing cells after Fura-2 AM incubation also works well, and can be useful in reducing assay prep time and costs.

Hamamatsu Corp.
908-231-0960
www.hamamatsu.com

References
1. Montero, M. et al. Activation by Chemotactic Peptide of a Receptor-operated Ca2+ entry Pathway in differentiated HL60 cells. J. Biol. Chem. 269:29451-29456 (1994).
2. Kremer, S.G. et al. Simultaneous fluorescence measurement of calcium and membrane potential responses to endothelin. Am. J. Physiol. 263(Cell Physiol. 32):C1302-C1309 (1992).
3. Laskey, R.E et al. Calcium entry-dependent oscillations of cytoplasmic calcium concentration in cultured endothelial cell monolayers. Proc. Natl. Acad. Sci. 89:1690-1694 (1992).
4. Johnson, P.L. et al. Errors caused by combination of Di-4 ANEPPS and Fluo 3 / 4 for simultaneous measurements of transmembrane potentials and intracellular calcium. Ann. Biomed. Eng. 27:563-571 (1999).
5. Baxter, D.F. et al. A Novel Membrane Potential-Sensitive Fluorescent Dye Improves Cell-Based Assays for Ion Channels. J. Biomol Screen. 7(1):79-85 (2002).
6. Sumichika, H. et al. Identification of a Potent and Orally Active Non-peptide C5a Receptor Antagonist. J. Biol. Chem. 277:49403-49407 (2002).
7. Iguchi, H. et al. SOX6 Attenuates Glucose-stimulated Insulin Secretion by Repressing PDX1 Transcriptional Activity and is Down-regulated in Hyperinsulinemic Obese Mice. J. Biol. Chem. 280:37669-37680 (2005).