Materials and Reagents

15 or 50 ml centrifuge tubes
96-well microplates (ideally black-walled, clear bottom)
DMEM/F-12, no phenol red (Thermo Fisher Scientific, Gibco, catalog number: 21041025 ), store at 2-8 °C
PBS Tablets (Thermo Fisher Scientific, Gibco, catalog number: 18912014 ), store at room temperature (RT)
H₂DCFDA (Thermo Fisher Scientific, Life Technologies, catalog number: D399 ), store powder at -20 °C
Sulforhodamine B sodium salt (SRB) (Sigma-Aldrich, catalog number: S1402 ), store powder at RT
Trizma base (Sigma-Aldrich, catalog number: T1503 ), store powder at RT
Trichloroacetic acid (TCA) (Sigma-Aldrich, catalog number: T4885-1KG ), store powder at RT
Acetic acid (Sigma-Aldrich, catalog number: A6283 ), store liquid at RT
Extracellular matrix solution (e.g., thermally cross-linked ECM with cold 10 µg/ml Collagen I for fibroblasts, or 1% Engelbreth-Holm-Swarm (EHS) sarcoma from Matrigel or Geltrex for iPSC, neural stem cells, incubated at 37 °C for 30 min)
30% wt. Hydrogen peroxide
100% DMSO
MilliQ grade water
10 mM H₂DCFDA in DMSO (see Recipes)
10% w/v TCA in water (see Recipes)
0.004% w/v SRB in 10% w/v TCA (TCA-SRB) (see Recipes)
10 mM Trizma in water (see Recipes)

Equipment

Assumes availability of typical cell culture equipment (CO₂ incubator, centrifuge, biosafety cabinet, water bath, fridge and/or cold room, etc.)
BMG LABTECH POLARstar Omega plate reader or fluorescence plate spectrometer with similar fluorescence capacity in fluorescein channel and 494/522 nm and 565/586 nm
Multichannel pipette (~10 µl and ~100 µl range) and sterile reagent reservoirs
Plastic humidifier box
Corrosives waste container

Software

Microsoft Excel or other spreadsheet processing software

Procedure

Cell seeding
1. Coat 96 well plate plates with extracellular matrix solution if required for desired cell type.
2. Dissociate and count cells with cell counting device.
3. Seed at appropriate density in 100 µl in phenol red-free complete media overnight (typically 20,000-30,000 cells/cm²).
4. Incubate for up to 0.5 h at room temperature (~24 °C) to prevent uneven seeding.
5. Plate plate(s) within humidifier chamber in CO₂ incubator.

Dye loading and incubation
1. Prepare 20 µM H₂DCFDA in pre-warmed base medium (DMEM/F12, no phenol red) on day of experiment.
2. Empty plate into waste container (inversion and flicking can be utilized if cells are strongly adherent, otherwise carefully remove media with a multichannel, for suspension cells centrifuge prior to this).
3. Add 100 μl 20 µM H₂DCFDA to each well using a multichannel pipette.
4. Return plate to incubator for 30 min.
5. Empty plate into a waste container.
6. Add phenol-red free media and treatment including hydrogen peroxide as a positive control (e.g., 90 µl media, 10 µl compound).
7. Return plate to a humidified chamber in incubator for desired treatment period (e.g., 1 h or 24 h incubation).

Acquisition
1. Acquire with fluorescence plate spectroscopy with appropriate filter set closest to excitation 494 nm/emission 522 nm.
2. Empty plate and add 35 µl TCA-SRB (0.004% w/v SRB in 10% w/v TCA) (2-8 °C) for 15 min in a fridge or cold room. Empty plate into an appropriate corrosives waste container and wash with 200 µl of 1% acetic acid. A second wash may be required if the background is higher than expected, particularly with automated pipetting devices that cannot completely remove liquid from the corner of the well.
3. Empty plate and replace with 100 µl 10 mM Trizma base solution, incubate at ambient temperature for 5 min, and gently manually agitate plate for 5 s to redistribute dye following incubation.
4. Acquire by fluorescence plate spectroscopy with appropriate filter closest to excitation 565 nm/emission 586 nm.

Data analysis

Export data to spreadsheet format.
Deduct the average of cell-free controls from DCF and SRB data (at least duplicate).
Calculate percentage of SRB signal relative to untreated controls.
Calculate percentage of DCF signal relative to untreated controls.
Calculate ratio of DCF:SRB percentage per well to normalize ROS generation readout to cellular protein per well.
Statistical analyses include independent t-tests, 1-way or 2-way ANOVA with post-hoc multiple comparisons tests for multiple groups (Figure 4).

Figure 4. Example of relative ROS generation. iPSC-derived astrocytes were treated with solvent (0.2% DMSO), 100 µM H₂O₂, or proteasomal inhibitor 0.1 µM MG132 for 24 h. MG132-induced ROS was unable to be detected under treatment conditions by intracellular DCF microscopy assay. Mean of independent replicates is presented with SD error bars (n = 3). Significant differences identified by one-way ANOVA followed by a post-hoc Holm-Sidak test. *p < 0.05.

Notes

Include at least 2 technical duplicates, empty well control, and positive control well.
As a general rule each plate requires ~12 ml of 20 µM H₂DCFDA, 6 ml of 0.004% SRB + 10% TCA, 22 ml of 1% acetic acid and 12 ml of 10 mM Trizma base.
The concentration of H₂O₂ for a positive control should be determined by a cytotoxicity assay that does not result in total cell death (ideally > 50%), we have found 100 µM to be suitable for fibroblasts and human iPSC-induced astrocyte cultures but would vary with cell type and duration of analysis.
A parallel plate with cell-free controls is essential if there is substantial fluorescence contribution in the DCF channel by test substance added. Otherwise a single column of cell-free wells can be utilized for convenience.
The period of time to allow for cell adherence post-seeding can be confirmed by microscopy to avoid the full 0.5 h room temperature incubation time.
The humidifier chamber advised allows for usage of all wells without concern of evaporation for up to several days and simply involves stacking microplates on a plastic container with a platform of elevation (such as an empty pipette tip box) above a volume of water (Walzl et al., 2012). Parafilm which also allows carbon dioxide and oxygen exchange is a cost-effective means of reducing evaporation if no humidifier chamber is used. Replacing edge wells with a water reservoir is a common technique but is not necessarily effective for longer incubation periods and results in a loss of ~37.5% of assayable well space.
Thermal gradients can be avoided by usage of metal blocks that fit beneath 96-well plates, however this can be cumbersome for scalability. Microplates can be loaded with H₂DCFDA at ambient temperature instead of prewarmed media for large scale handling. The overnight incubation mitigates the duration of uneven heating that occurs during transition from plate from ambient temperature conditions to incubator.
Cold TCA can be replaced by 4% paraformaldehyde (PFA) in PBS which is more routinely available to laboratories performing immunocytochemistry, however PFA fixation and SRB staining cannot be combined as SRB only binds to basic protein amino acid residues effectively under mild acidic conditions (Skehan et al., 1990).
For batch processing of plates, the 15 min TCA-SRB step can be replaced with an overnight duration period instead to reduce inter-well variability.
Prior publications describing the SRB protocol typically involve 3-5 wash steps with acetic acid. We have found minimal difference in outcome by staining in low volume followed by single high-volume wash.
Loading of H₂DCFDA can be replaced by incubation with low concentrations of H₂DCFDA (≤ 1 µM) without a wash step however cytotoxicity of probe dependent on cell type must be considered.
This protocol is intended for adherent cell types.
Although more laborious, in plate spectrometry format the SRB assay is less susceptible to artefactual increase in signal per cell under apoptotic conditions than nucleic acid groove/intercalating probes. Nucleic acid binding dyes such as Hoechst 33342 result in overestimation of viable cell population (Figure 5).

Figure 5. Example of artefactual nuclear dye plate spectroscopy scoring of apoptotic cell populations. iPSC-derived astrocytes were treated with solvent (0.2% DMSO) or 100 µM H₂O₂ for 24 h. Mean of technical duplicates is presented with SD error bars. Representative nuclear staining with Hoechst 33342 (10 µM) of control (a) in comparison to DNA-condensed apoptotic bodies that provide higher signal per cell leading to overestimation of viable cell population (b) (scale bar = 30 µm).
For normalization to cell number or confluency, microscopic scoring of individual cells can be faciliated by a SRB cellular mask prior to, or restained after solublisation, along with nuclear scoring by non-overlapping dyes such as Hoescht 33342, DAPI or SYTOX Green. The cellular mask and total cellular protein data provided by SRB allow for correlation of total cellular protein, cellular population and nuclei which may be applicable toxicity and genotoxicity studies, and provide alternative normalization methods where applicable (Figure 6). Efficient microscopic data acquisition and analysis requires use of automated microscopy platforms and software where available.

Figure 6. Example SRB cytoplasmic and nuclear mask. iPSC-derived astrocytes were fixed and stained as described by TCA-SRB (a), before staining with 10 µM LCS1 dye in H2O (b). Overlay with phase contrast image (c) (scale bar = 100 µm).

Recipes

10 mM H₂DCFDA
27.6 mg per 1 ml DMSO
Store at -20 °C and use within 3 months of preparation
Each 1 ml is sufficient for ~450 x 96-well microplate assays
10% TCA
Dissolve 10 g TCA per 100 ml MilliQ grade water
0.004% w/v SRB in 10% TCA
Dissolve 4 mg SRB per 100 ml 10% TCA
10 mM Trizma base (121.14 g/mol)
Dissolve 121.14 mg Trizma per 100 ml of MilliQ grade water

Acknowledgments

This research was funded by philanthropic donations and a grant from the Illawarra Health and Medical Research Institute (IHMRI). L.O. is supported by a National Health and Medical Research Council (NHMRC) of Australia Boosting Dementia Research Leadership Fellowship (APP1135720). The H₂DCFDA assay is based on a method in a previously listed oxidative stress-related publication (Han et al., 2009). The SRB protocol is based on findings of a previously established cytotoxicity protocol (Skehan et al., 1990). We thank Dr Reece Gately for helpful discussions.

Competing interests

The authors declare they have no competing interests.

References

Baker, W. L. (1991). Potential interference of hydrogen peroxide in the 2,2'-bicinchoninic acid protein assay. Anal Biochem 192(1): 212-214.
Han, F., Fan, L., Wang, X. and Li, W. (2012). Sulforhodamine B restaining as a whole-cell label allows visualizing one more fluorochrome and its application in assaying protein nucleocytoplasmic distribution. Cytometry A 81(6): 532-540.
Han, Y. H., Moon, H. J., You, B. R. and Park, W. H. (2009). The effect of MG132, a proteasome inhibitor on HeLa cells in relation to cell growth, reactive oxygen species and GSH. Oncol Rep 22(1): 215-221.
Hayashi, K., Yonemura, S., Matsui, T. and Tsukita, S. (1999). Immunofluorescence detection of ezrin/radixin/moesin(ERM) proteins with their carboxyl-terminal threonine phosphorylated in cultured cells and tissues. J Cell Sci (Pt 8): 1149-1158.
Kim, H. M., Han, S. B., Kim, M. S., Kang, J. S., Oh, G. T. and Hong, D. H. (1996). Efficient fixation procedure of human leukemia cells in sulforhodamine B cytotoxicity assay. J Pharmacol Toxicol Methods 36(3): 163-169.
Lundholt, B. K., Scudder, K. M. and Pagliaro, L. (2003). A simple technique for reducing edge effect in cell-based assays. J Biomol Screen 8(5): 566-570.
Paige Davis Volk, A. and Moreland, J. G. (2014). Chapter Thirteen: ROS-Containing Endosomal Compartments: Implications for Signaling. In: P. Michael Conn(Ed). Methods in Enzymology Academic Press.
Shellman, Y. G., Ribble, D., Yi, M., Pacheco, T., Hensley, M., Finch, D., Kreith, F., Mahajan, R. L. and Norris, D. A. (2004). Fast response temperature measurement and highly reproducible heating methods for 96-well plates. Biotechniques 36(6): 968-972, 974-966..
Skehan, P., Storeng, R., Scudiero, D., Monks, A., McMahon, J., Vistica, D., Warren, J. T., Bokesch, H., Kenney, S. and Boyd, M. R. (1990). New colorimetric cytotoxicity assay for anticancer-drug screening. J Natl Cancer Inst 82(13): 1107-1112.
Walzl, A., Kramer, N., Mazza, G., Rosner, M., Falkenhagen, D., Hengstschläger, M., D., r Schwanzer-Pfeiffer and Dolznig, H. (2012). A simple and cost efficient method to avoid unequal evaporation in cellular screening assays, which restores cellular metabolic activity. Int J Appl Sci Tech 2.
Wojtala, A., Bonora, M., Malinska, D., Pinton, P., Duszynski, J. and Wieckowski, M. R. (2014). Methods to monitor ROS production by fluorescence microscopy and fluorometry. Methods Enzymol 542: 243-262.