Exp Eye Res. 2017 Aug;161:17-29. doi: 10.1016/j.exer.2017.05.011.

Streamlined duplex live-dead microplate assay for cultured cells.

Pfeffer BA1, Fliesler SJ2.

1 Department of Ophthalmology, University at Buffalo-State University of New York (SUNY), Buffalo, NY, USA; SUNY Eye Institute, Buffalo, NY, USA. Electronic address: brucepfe@buffalo.edu.
2 Department of Ophthalmology, University at Buffalo-State University of New York (SUNY), Buffalo, NY, USA; SUNY Eye Institute, Buffalo, NY, USA; Department of Biochemistry and Neuroscience Program, University at Buffalo- State University of New York (SUNY), Buffalo, NY, USA; Research Service, VA Western New York Healthcare System, Buffalo, NY, USA. Electronic address: fliesler@buffalo.edu.


A duplex fluorescence assay to assess the viability of cells cultured in multi-well plates is described, which can be carried out in the original culture plate using a plate reader, without exchanges of culture or assay medium, or transfer of cells or cell supernatant. The method uses freshly prepared reagents and does not rely on a proprietary, commercially supplied kit. Following experimental treatment, calcein acetoxymethyl ester (CaAM) is added to each well of cultured cells; after 30 min, the fluorescence intensity (emission λmax ∼ 530 nm) is measured. The signal is due to formation of calcein, which is produced from CaAM by action of esterase activity found in intact live cells. Since live cells may express plasma membrane multidrug transport proteins, especially of the ABC transporter family, the CaAM incubation is carried out in the presence of an inhibitor of this efflux process, thereby improving the dynamic range of the assay. Next, SYTOX® Orange (SO) is added to the culture wells, and, after a 30-min incubation, fluorescence intensity (emission λmax ∼ 590 nm) is measured again. SO is excluded from cells that have an intact plasma membrane, but penetrates dead/dying cells and can diffuse into the nucleus, where it binds to and forms a fluorescent complex with DNA. The CaAM already added to the wells causes no interference with the latter fluorescent signal. At the conclusion of the duplex assay, both live and dead cells remain in the culture wells and can be documented by digital imaging to demonstrate correlation of cellular morphology with the assay output. Two examples of the application of this method are provided, using cytotoxic compounds having different mechanisms of action.


Calcein AM; Cell viability assay; Fluorescence assay; Plate reader; Retinal cell line; Sytox Orange

PMID: 28572030



  1. Background and rationale

Cell culture models have wide application for the study of pathophysiological mechanisms underlying loss of cellular viability culminating in cell death, described variously as occurring via apoptosis, necrosis, and a variety of other more specialized categories (Méry et al., 2017; Galluzzi et al., 2018).  The characterization of molecular pathways and subcellular locales for different cell death pathways and modalities permits the selection of specific targets that can be assessed with appropriate and sensitive probes, yielding signals with measurable outputs.  For this reason, cell viability assays performed in vitro also are routinely employed in preclinical drug development, to determine the therapeutic index (the ratio of the concentration with highest tolerable impact on viability to that producing specific efficacy) of candidate drugs and other agents (Muller and Milton, 2012).  Therefore, there is an ongoing need for cell viability assays compatible with the throughput—including desired replicate samples—associated with advancement of screening library hits through lead optimization and later stage preclinical testing.  A detailed description is presented here of an optimized cell-based, rapid, sensitive, direct-read, bifunctional (duplex), multi-well viability assay, using the sequential application of two commercially obtained reagents corresponding to different relevant targets associated with cell death.  As shown by challenge using two cytotoxic molecules with different mechanisms of action (including the one exemplified below), tested on two representative cell lines (including the one exemplified below), dose-dependent viability results with extensive dynamic ranges and high Z-factors (Zhang et al., 1999) can be obtained.


  1. Description and documentation of the viability assay

a General protocol

After cells of interest have been seeded in multi-well plates, and have attained the appropriate density and phenotype, they are treated with test reagents or are otherwise subjected to the desired experimental treatments; treatments are carried out using a simplified incubation medium (IM) containing serum at 1% (v/v) or less.  Without rinsing or transferring cells or culture medium, a 10x working stock of calcein acetoxymethyl ester (CaAM) is added to experimental wells, followed by a 30-min incubation, and fluorescence is measured using a plate reader.  The specific signal is due to the internalization of CaAM, and its conversion to fluorescent calcein via esterase activity, only within live cells (Bozyczko-Coyne et al., 1993).  A novel feature of the assay protocol is the inclusion of probenecid, an inhibitor of multidrug transporters such as ABCC1, with the CaAM incubation step, thereby preventing efflux of calcein and subsequent diffusion of signal (Homolya et al., 1993).

Next, fluorescence is again recorded 30 min after incubation with Sytox Orange® (SO), which is added without removal of the calcein reagent, as CaAM and SO exhibit good separation of their respective excitation and emission wavelength maxima.  SO forms a fluorescent complex with nuclear DNA of those cells which have lost plasma membrane integrity (Yan et al., 2000; Pierce et al., 2003), a further indicator of cell death, as a result of experimental treatments.


b Representative results: Figures 1 – 3

The glial cell line rMC-1, derived from rat retinal Müller cells that underwent immortalization using the SV-40 T-antigen (Sarthy et al., 1998), was generously provided by Dr. Vijay Sarthy (Northwestern University School of Medicine, Evanston, IL, USA).  During the course of investigations described here, this line was authenticated genomically, and also extensively characterized by analysis of expressed signature genes and proteins (Pfeffer et al., 2016).

Cells were treated with a dose range of staurosporine (Stsp), a non-specific protein kinase inhibitor that has been shown to be cytotoxic to a wide variety of cell types; in response to appropriate concentrations of Stsp, cell lines as well as more acute preparations of normal diploid cells, including neurons, undergo apoptosis within approximately 24 h (Seo and Seo, 2009).  Following 24 h exposure to Stsp, the cultured rMC-1 cells were then processed for the viability assays as described above, with the results depicted in Figs. 1 and 2.  Statistically significant decreases in cell viability were documented using both the CaAM and SO assays, for Stsp concentrations at 25, 125, and 500 nM; note that in contrast to the CaAM assay (Fig. 1), the SO assay signal increases in proportion to loss of viability (Fig. 2).  In the CaAM assay, at the highest concentration of Stsp tested, the quantitative loss of viability was more than 90% compared to the mean value for vehicle control (VC).  While the changes in viability of the rMC-1 cells generally appeared to show correspondence between the CaAM and SO assays as Stsp doses increased up to 125 nM, a “tapering-off” of signal intensity at 500 nM Stsp in the SO assay (Fig. 2) might suggest an apparent lack of correlation in terms of dose-responsiveness between the two assays; however, the cell-based mechanism responsible for this result is explained below (Contingencies).   Following the fluorescence measurements using the multi-well plate reader, digital images of the cells in representative wells were captured using an inverted microscope with phase-contrast optics (Fig. 3).





Figure 1:  Two 48-well plates of confluent rMC-1 cells (initially seeded at 40,000 cells/well, passage 30) were incubated in parallel for 24 h with a concentration range of staurosporine (Stsp) from 2.5 to 500 nM, or DMSO alone as vehicle control (VC).  Cell-free blank wells were also included.  Viability of the cells then was assessed by the sequential application of the CaAM and SO assays (results for the latter shown in Fig. 2, below).  In one plate probenecid was included in the CaAM assay medium.  Without probenecid (gray bars), the maximum fluorescence signal achieved (expressed as relative fluorescence units (RFU)) was ca. 3300 for VC, and the dynamic range between VC and the most toxic dose of Stsp (500 nM) only amounted to a 68.9% decrease in mean RFU. In contrast, the addition of probenecid apparently conserved the intracellular calcein signal, resulting in a maximum mean value of ca. 10000 RFU for VC, and extending the dynamic range to a decrease of 91.4% as a result of incubation with 500 nM Stsp. *P ≤ 0.05 vs. VC for the CaAM assay with probenecid (ANOVA); #P ≤ 0.05 vs. VC for the CaAM assay samples run without probenecid (ANOVA); §P ≤ 0.05 for the CaAM assay RFU values in the absence of probenecid vs. the matching dose of Stsp when probenecid was present (Student’s two-tailed t-test).  Error bars are [1.96 x standard error (SE)].



Figure 2:  The Sytox Orange® (SO) assay values for the wells analyzed in Fig. 1 (above), in the presence of CaAM (and fluorescent intracellular calcein) + probenecid that remained in the wells after the CaAM assay was performed).  From 2.5 to 125 nM Stsp there was increasing toxicity as determined by this assay, with a leveling off of the SO fluorescence signal at 500 nM.  *P ≤ 0.05 vs. VC (Student’s two-tailed t-test).  For each dose of Stsp tested, there was no statistically significant difference between mean SO signal detected from wells previously treated with CaAM either with or without probenecid (results not shown).  Error bars as in Fig. 1.



Figure 3: Digitally captured micrographic images of rMC-1 cells treated with a dose range of staurosporine (Stsp).  Routinely observed morphology of this cell line at confluence is displayed for the vehicle control (VC) treatment condition (* = CaAM assay medium contained probenecid).  At the three highest concentrations of Stsp employed, corresponding to significant reductions in cell viability as measured in the duplex assay (Figs. 1 and 2, above), changes in cell morphology are observed that are hallmarks of loss of cell integrity associated with cell death, including:  plasma membrane blebbing (thin arrows); cell detachment from the substrate (stars), and retraction of the monolayer (arrowheads).   * = CaAM assay medium contained probenecid.  Phase-contrast optics.  Magnification bar = 100 microns.


c Contingencies

Both the CaM and SO signals are stable and robust enough to permit repeated plate reader measurements over a 10-min interval.  Since neither probe is cytotoxic themselves, following the assays and photographic documentation, the remaining viable cells may be used for additional assays, such as crystal violet staining (correlating with absolute cell number), 100% efficacy SO determinations, cell-based ELISA, or even protein or gene transcript analysis.

The use of cell culture media (including “growth,” maintenance, and IM) with reduced (1% (v/v) or less) serum (e.g., bovine calf serum as used here) is advantageous for optimal results using the duplex viability assay protocol:  it permits the testing of defined reagents promoting cell survival (such as antioxidants) accompanied by minimal confounding effects of undefined components in serum; the CaAM and SO assays are not compatible with high serum levels that contribute esterases and nucleases, respectively; and adverse effects of acute serum withdrawal are avoided.  Therefore, candidate cultured cells undergoing evaluation should be adapted to more defined media for the described protocols.

Time course experiments using the duplex assay are helpful to isolate differences in the progressive loss of cell viability for different cell types and different treatments.  In certain cases, with increasing time and/or magnitude of the experimental challenge, the investigators noticed a tapering-off or even more substantial decreases in SO assay signal.  This is most likely due the autonomous progression of cell death that occurs in cell culture, promoting loss of integrity of nuclear components as a result of endogenous nuclease activation, with eventual cleavage of DNA to fragments that bind SO with less efficiency, if at all (Frankfurt et al., 1996).


3 Applications

As discussed in the current paper, the protocol provided here may in fact distinguish between Stsp, as illustrated above, and an additional cytotoxic agent, cumene hydroperoxide, with respect to mechanism of action and cell death pathway invoked, since the treatment of rMC-1 cells with the latter did not give rise to a similar reduction of SO signal at higher concentrations.  The methodology outlined here has been exploited previously to demonstrate varied cytotoxic potencies and efficacies for a panel of 7-dehydrocholesterol-derived oxysterols (with cholesterol as a non-cytotoxic control), using the rMC-1 cells described above, and also a mouse photoreceptor cell line (661W), as well as normal diploid monkey retinal pigment epithelial cells (Pfeffer et al., 2016), providing insights regarding the pathophysiology of Smith-Lemli-Opitz syndrome (SLOS).  The assay protocol delineated here may be modified, by inclusion of inhibitors and other modulators of molecular and functional steps corresponding to specific pathways of cellular death and survival, to further elucidate disease mechanisms and treatments.  Given the interest in and urgency for developing pharmaceutical treatments that selectively target cancer cell molecular phenotypes while sparing normal cells and tissues, the methods described here will have utility for identifying small molecules and other agents with optimized therapeutic indices (e.g., Swinney, 2006).


4 References

Bozyczko-Coyne, D., B.W. McKenna, T.J. Connors and N.T. Neff (1993).  “A rapid fluorometric assay to measure neuronal survival in vitro.”  J Neurosci Methods 50(2):205-216 (http://dx.doi.org/10.1016/0165-0270(93)90009-g).

Frankfurt, O.S., J.A. Robb, E.V. Sugarbaker and L. Villa (1996).  “Monoclonal antibody to single-stranded DNA is a specific and sensitive cellular marker of apoptosis.” Exp Cell Res 226(2):387-397 (DOI:10.1006/excr.1996.0240).

Galluzzi, L., I. Vitale, S.A. Aaronson, J.M. Abrams, D. Adam, P. Agostinis, et al. (2018).  “Molecular mechanisms of cell death: recommendations of the Nomenclature Committee on Cell Death 2018.”  Cell Death Differ 2018 (DOI:10.1038/s41418-017-0012-4).

Homolya, L., Z. Holló, U.Z. Germann, I. Pastan, M.M. Gottesman, M.M. and B. Sarkadi (1993). “Fluorescent cellular indicators are extruded by the multidrug resistance protein.” J Biol Chem 268(29):21493-21496 (http://www.jbc.org/content/268/29/21493.long).

Méry, B., J.-B. Guy, A. Vallard, S. Espenel, D. Ardail, C. Rodriguez-Lafrasse, C. Rancoule and N. Magné (2017). “In vitro cell death determination for drug discovery:  A landscape review of real issues.”  J Cell Death 2017 (DOI:10.1177/1179670717691251).

Muller, P.Y. and M.N. Milton (2012).  “The determination and interpretation of the therapeutic index in drug development.”  Nat Rev Drug Disc 11(10):751-761 (DOI:10.1038/nrd3801).

Pfeffer, B.A., L. Xu, N.A. Porter, S. Ramachandran Rao and S.J. Fliesler, S.J. (2016).  “Differential cytotoxic effects of 7-dehydrocholesterol-derived oxysterols on cultured retina-derived cells: dependence on sterol structure, cell type, and density.” Exp Eye Res 145:297-316 (DOI:10.1016/j.exer.2016.01.016).

Pierce, M., C. Wang, M. Rebentisch, M. Endo, M. Stump and A. Kamb (2003).  “A highthroughput, homogeneous microplate assay for agents that kill mammalian tissue culture cells.”  J Biomol Screen 8(3): 283-291 (DOI:10.1177/1087057103008003006).

Sarthy, V.P., S.J. Brodjian, K. Dutt, B.N. Kennedy, R.P. French and J.W.Crabb (1998).  “Establishment and characterization of a retinal Müller cell line.” Invest Ophthalmol Vis Sci 39(1):212-216 (http://iovs.arvojournals.org/article.aspx?articleid=2181125).

Seo, S.R. and J.T. Seo (2009).  “Calcium overload is essential for the acceleration of staurosporine-induced cell death following neuronal differentiation in PC12 cells.” Exp Mol Med 41(4):269-276 (DOI:10.3858/emm.2009.41.4.030).

Swinney, D.C. (2006).  “Biochemical mechanisms of New Molecular Entities (NMEs) approved by United States FDA during 2001-2004:  mechanisms leading to optimal efficacy and safety.”  Curr Top Med Chem 6(5):461-478 (DOI:10.2174/156802606776743093).

Yan, X., R.C. Habbersett, J.M. Cordek, J.P. Nolan, T.M. Yoshida, J.H. Jett and B.L. Marrone (2000). “Development of a mechanism-based, DNA staining protocol using SYTOX Orange nucleic acid stain and DNA fragment sizing flow cytometry.” Anal Biochem 286(1):138-148 (DOI:10.1006/abio.2000.4789).

Zhang, J.-H., T.D.Y. Chung and K.R. Oldenburg, K.R. (1999). “A simple statistical parameter for use in evaluation and validation of high throughput screening assays.”  J Biomol Screen 4(2), 67-73 (DOI:10.1177/108705719900400206).