High Throughput Screening Assessment of Reactive Oxygen Species (ROS) Generation using Dihydroethidium (DHE) Fluorescence Dye

Summary

This protocol describes a novel method to quantify intracellular reactive oxygen species (ROS) using dihydroethidium (DHE) as a fluorescence dye probe using a high-throughput screening approach. The protocol describes the methods for quantitative assessment of intracellular reactive oxygen species (ROS) in the three different hepatocellular carcinoma cell lines.

Abstract

Reactive oxygen species (ROS) play a key role in the regulation of cellular metabolism in physiological and pathological processes. Physiological ROS production plays a central role in the spatial and temporal modulation of normal cellular functions such as proliferation, signaling, apoptosis, and senescence. In contrast, chronic ROS overproduction is responsible for a wide spectrum of diseases, such as cancer, cardiovascular disease, and diabetes, among others. Quantifying ROS levels in an accurate and reproducible manner is thus essential to understanding normal cellular functionality. Fluorescence imaging-based methods to characterize intra-cellular ROS species is a common approach. Many of the imaging ROS protocols in the literature use 2'-7'-dichlorodihydrofluorescein diacetate (DCFH-DA) dye. However, this dye suffers from significant limitations in its usage and interpretability. The current protocol demonstrates the use of a dihydroethidium (DHE) fluorescent probe as an alternative method to quantify total ROS production in a high-throughput setting. The high throughput imaging platform, CX7 Cellomics, was used to measure and quantify the ROS production. This study was conducted in three hepatocellular cancer cell lines - HepG2, JHH4, and HUH-7. This protocol provides an in-depth description of the various procedures involved in the assessment of ROS within the cells, including - preparation of DHE solution, incubation of cells with DHE solution, and measurement of DHE intensity necessary to characterize the ROS production. This protocol demonstrates that DHE fluorescent dye is a robust and reproducible choice to characterize intracellular ROS production in a high-throughput manner. High throughput approaches to measure ROS production are likely to be helpful in a variety of studies, such as toxicology, drug screening, and cancer biology.

Introduction

Reactive oxygen species (ROS) are a group of naturally occurring, highly reactive, and temporally unstable chemical radicals formed as a part of the normal cellular metabolism in cells. ROS plays a key and essential role in the modulation of normal physiological and biochemical processes occurring in cells^1,2. The main source of ROS production in cells is from the mitochondrial electron transport chain (ETC) pathway as a part of the normal bioenergetic cycle. Significant additional sources of ROS production include enzymatic reactions such as cellular NADPH oxidases in cells. Metabolism of food molecules (e.g., glucose) occurs via the oxidative phosphorylation pathway in the mitochondrial matrix. A baseline level of ROS production is essential to regulate normal physiological cell signaling processes. Many key protein molecules that are part of the glucose metabolic signaling pathways (e.g., Akt and PTEN) are known to respond to intracellular ROS levels. Additionally, ROS are produced by various intracellular enzymes such as xanthine oxidase, nitric oxide synthase, and peroxisomal constituents as a part of the cellular enzymatic pathways^1,2. In contrast to the natural sources of ROS, certain environmental factors, such as xenobiotics, infectious agents, UV light, pollution, cigarette smoking, and radiation, also lead to excessive production of ROS, which are a key driver of intra-cellular oxidative stress^1,3. Elevated cellular oxidative stress can cause damage to native biomolecules inside a cell, such as lipids, proteins, and DNA, causing various diseases such as cancer, diabetes, cardiovascular disease, chronic inflammation, and neurodegenerative disorders^1,3,4. Therefore, accurate measurements of ROS are essential to understand the cellular mechanisms involved in oxidative stress-induced disease pathophysiology.

Due to the short timescales of ROS production and elimination inside cells, quantitative measurements of various ROS radicals remains a challenge. Methods such as electron paramagnetic resonance (EPR)⁵, high-pressure liquid chromatography (HPLC), and fluorescence probe-based imaging are used to measure the various cellular ROS⁶. While methods such as EPR and HPLC yield quantitatively accurate estimates, these methods involve the destruction of the cellular spatial morphology and are usually in the form of global and bulk measurements of a sample. In contrast, imaging-based methods such as fluorescence probe-based methods retain the cellular morphology and spatial context of the ROS generation. However, the specificity of various fluorescence probes for different types of ROS radicals has not been well-established^7,8. Several fluorescent probes such as dihydroethidium (DHE), dichlorodihydrofluorescein diacetate (DCFH-DA), dihydrorhodamine (DHR), dimethyl anthracene (DMA), 2,7 dichlorodihydroflurescein (DCFH), 1,3-Diphenylisobenzofuran (DPBF), and MitoSox are available for ROS detection commercially. In the past decades, DHE, MitoSox, and DCFH-DA are the commonly used fluorescent dyes to measure ROS in cells and tissues^8,9. DCFH-DA is a widely used dye for detecting intracellular H₂O₂ and oxidative stress. Despite the popularity of DCFH-DA, multiple previous studies have shown that it cannot be reliably used to measure intracellular H₂O₂ and other ROS levels^{8,9,10,11,12,13,14}.

In contrast, the fluorescent probe dihydroethidium (DHE) shows a specific response to the intra-cellular superoxide radical (O₂^-). While the superoxide radical is one of many of the ROS species observed in cells, it is an important radical involved in the reduction of transition metals, conversion to peroxynitrate, and formation of hydroperoxides, among other intracellular effects. DHE is quickly taken up by the cells and has a fluorescence emission in the red wavelength range¹⁵. Upon reaction with superoxide radical specifically, DHE forms a red fluorescent product, 2-hydroxy ethidium (2-OH-E⁺). Thus, DHE may be considered as a specific probe for superoxide detection. However, DHE can also undergo nonspecific oxidation with ONOO^- or OH^., H₂O_2, and cytochrome c to form a second fluorescence product, ethidium E^+, which can interfere with the measured 2-OH-E⁺ levels. However, these 2-OH E⁺ and E⁺ products, in combination, represent a major part of the total cellular ROS species observed inside a cell when stained with DHE. E⁺ intercalates into DNA, greatly enhancing its fluorescence^{8,9,10,11,13,14,15,16}. Since the fluorescence spectra of ethidium and 2-hydroxy ethidium only differ slightly, the majority of ROS levels seen in a cell secondary to superoxide production can be detected and measured using DHE fluorescence products. These ROS species are identified using 480 nm wavelength excitation and 610 nm wavelength emission^15,16,17.

In addition to choosing a specific fluorescent ROS detection probe, it is important to choose a sensitive method of detection to measure intracellular ROS. Accurate assessment of intracellular ROS levels is thus key to identifying disturbed redox balance states occurring in diseased cells or cells that have been exposed to various environmental stressors such as radiation, toxicological compounds, and genotoxic agents¹⁸. Since ROS is a commonly occurring phenomenon in cells that is responsible for regulating a variety of cell signaling activities, robust methods of detection of ROS are essential. To enable such high-throughput evaluation of ROS production within cells, this protocol uses a high-content screening (HCS) platform to measure the ROS species. The current protocol allows the high-throughput analysis of intracellular ROS production, which is of critical importance in many toxicology studies¹⁹. This protocol aims to provide an easy and versatile solution to detect and measure intracellular ROS in adherent hepatocellular carcinoma cells. The chemical reagents of H₂O₂ and menadione are used as potent ROS stimulators to measure the relative levels of ROS production in a controlled and high throughput setting. This protocol may be fine-tuned to measure ROS production in adherent and nonadherent cells under appropriate conditions, as necessary.

Protocol

1. Cell culture

1. Seed the test cells (HepG2, HUH7, and JHH4 hepatocellular carcinoma cells) into a 96-well plate at a seeding density of 10,000 cells/well in a final seeding volume of 200 µL per well.
   1. Before culturing the HepG2 cells, coat the 96 plate wells with type IV collagen (50 µg/mL) for 2 h duration at room temperature (RT). To avoid solidification of the stock collagen, place the stock solution into ice and subsequently, initiate the dilution process to the desired concentration.
   2. After a 2 h incubation period, aspirate off the excess collagen and wash the wells three times with the 1x PBS.
2. Cultivate the cells in Dulbecco's modified Eagle medium (DMEM) overnight at 37 °C and 5% CO₂ concentration in a humidified incubator.
3. Next day, or upon reaching the 80%-90% confluency, whichever is earlier, treat the cells with H₂O₂ (untreated, 250 µM, 500 µM, 750 µM, and 1000 µM) and Menadione (untreated, 25 µM, 50 µM, 75 µM, 100 µM), respectively for 30 min to induce the representative oxidative stress.
4. Alternatively, choose to test the cells with the desired test substance of choice capable of inducing oxidative stress within the cells.

2. Stock and dilute solution preparation for DHE staining of cells

1. Prepare DHE stock solution (5 mg/mL or 15.9 mM) by dissolving 5 mg of DHE into 1 mL of DMSO.
2. Dilute the DHE stock solution with double distilled autoclaved water to a final 100 µM concentration.
3. To avoid the freeze and thaw process of the dye (which may damage the fluorescence properties over time), prepare several aliquots in 1.5 mL centrifuge tubes with a concentration of 100 µM and store them at -20 °C.
4. To achieve a final working concentration of 10 µM, use an aliquot of the 100 µM DHE concentration and dilute it with pre-warmed DMEM media.
   NOTE: The working solution must be prepared fresh on the day of the experiment.
5. Vortex the working fluorescence dye solution for 5-10 s to ensure proper mixing of the dye.

3. DHE staining process

1. Remove the drug- or test-substance-containing media from each well and wash it gently once with either 1x PBS or DMEM.
   NOTE: When performing the addition or washing steps in the experiment, it is crucial to avoid direct contact between the cells and the pipettor. This can be achieved by gently pipetting the PBS along the sides of the wells to minimize any potential mechanical damage to the cells. Ensuring that the pipettor does not directly touch the cells will help maintain the cell integrity and viability for the duration of the experiment.
2. Add 100 µL of DMEM media containing 10 µM DHE (working solution) into each well and incubate at 37 °C for 30 min.
3. After 30 min incubation period, remove the DHE-containing media and wash each well gently three times with the 1x phosphate-buffered saline (PBS). This may be performed with a multi-channel pipettor to ensure rapid turnaround.
4. Add 200 µL of 1x PBS to each well with Hoechst 33342 nuclear staining dye for 10 min at RT.
5. Remove the Hoechst 33342 nuclear staining solution from the well and add 200 µL of 1x PBS to each well.
   NOTE: To obtain fixed cells, follow the same protocol as for the live cells, with one additional step of adding 4% PFA for 5 min after treatment and DHE staining. Remove the PFA solution and wash cells with the 1x PBS. Then, incubate the cells with the nuclear staining dye for 10 min.
6. Move the plate over to the high-content screening platform, fluorescence microscopy, or plate reader for image acquisition as quickly as possible.

4. Image acquisition and intensity measurement

1. Plate loading
   1. Open the Cellomics CX7 high-content screening (HCS) software for data acquisition.
   2. Carefully calibrate the imaging platform as per the manufacturer's specifications ahead of time with the specific 96-well plate of choice for use to avoid any hazy or blurry images in the data.
      NOTE: Multi-well plates from various vendors may have different material properties and can influence the quality of the final images obtained.
   3. Once calibrated, save the system parameters specific to the plate brand in the instrument database and reuse them for future data acquisitions.
   4. Carefully place the plate with the prepared samples onto the heated stage of the HCS imaging system. Make sure that the plate is securely positioned and in the correct orientation for imaging on the microplate holder (i.e., locate the label marked A1 on the Reader stage, then rotate the microplate so that the location of well A1 matches the corner with the sticker).
   5. Gently press down on the microplate to ensure it rests flat against the stage. Uneven leveling of the plate in the HCS system can lead to errors in image acquisition.
   6. After the plate is in place, press and hold the Ctrl key, then click Ctrl-OK in the Plate Load/Unload dialog in the software.
2. Selecting imaging parameters
   1. In the HCS imaging system, open Target Activation mode in the Cellomics bio-applications interface and create a new protocol using the two-channel setup protocol compatible with (a) nuclear staining and (b) DHE fluorescence probe data acquisition. In the current protocol, the filters 386_BGSRS_BGSRS, 480_BGS_RS, and 386_BGS_RS were used for (a) Hoechst 33342 nuclear-staining, (b) total ROS production, and (c) superoxide radical detection, respectively. The choice of these filters was driven by the specific overlap for the emission properties of each dyes (DAPI and DHE) used as a part of this protocol.
      NOTE: The naming convention of filters, e.g., 386-23_BGRFRN_BGRFRN, refers to the excitation of the sample with 386 nm light with a 23 nm bandwidth, followed by a dichroic filter that passes blue (B), green (G), red (R), far-red (FR), and near infra-red (N) light at specific wavelength ranges and the 386_BGS_RS refers to an emission filter that passes BGS and RS emission wavelength light after the dichroic.
   2. Specify the objectives for the image acquisition process within the software protocol interface, such as 10x or 20x magnification, as necessary.
   3. Choose the appropriate objective magnification depending on the desired level of imaging detail and resolutions necessary for the experiment. However, the resolution of each objective is fixed. Higher resolution detail will necessitate changing out the objective on the platform. In the current protocol, a 20x, 0.45 NA objective is used, which is a part of the high-content screening platform used in this protocol.
      NOTE: The 20x objective represents a good trade-off between imaging resolution detail and the speed of acquisition necessary to capture ROS changes over time. Higher magnification objectives may be chosen (e.g., 40X), however, this will lead to increased acquisition times.
   4. Specify the exposure settings of the image acquisition camera, such as the exposure time, camera binning, and z-stack interval, according to the specific requirements of the protocol.
      NOTE: The exposure time (often in milliseconds) varies based on the signal strength and sensitivity of the specific fluorophores used. This may also differ slightly from experiment to experiment based on the dye concentration and staining quality. In the current protocol, the parameters of acquisition are fixed as follows - Target % - 35%, image acquisition mode - 1104 x 1104 (with 2x2 binning), and objective 20x, 0.45 NA. These parameters may be set according to the specific experimental conditions.
3. Imaging configuration parameters
   NOTE: Once the imaging parameters are set, the image collection process may be initiated using the software interface.
   1. Image collection parameters
      1. Identify the primary tracking object of interest (the nucleus of the cells) using different algorithms available in the instrument (optics - isodata, fixed, and triangle).
         NOTE: The isodata method for the identification and marking of the primary object is used in this protocol.
      2. Segment the object (if any) either by shape or intensity parameter.
      3. Validate the primary object based on the desired size, shape, and intensity parameters unique to each cell type.
      4. Then, define the 'region of interest' (ROI) around this primary object.
         NOTE: The ROI is a key parameter of data acquisition that defines the location where the intensity of the fluorescence dye is identified to be consistent and repeatable across cell types. The ROI defines the key parameters (such as the intensity of the dye), which are measured for each primary tracking object (i.e., a cell) in different channels of the instrument. The ROI may be defined in the form of a circle or ring around the nucleus of each cell. The HCS software automatically obtains the intensity of the DHE fluorescence dye marker in channel 2 within the limit of the ROI for each cell that is segmented and analyzed in an automated manner.
      5. Set a 20 µm circle around the primary object (nucleus). The distance of 20 µm was chosen in this study based on the approximate sizes of the various cell lines used in this protocol (HepG2, JHH-4, and HUH-7). The 20 µm ring ROI around the cells obtained the fluorescence intensity in a consistent and repeatable manner.
         NOTE The key parameter in choosing an ROI is the ability to adequately cover the cell area; the size of the circular ROI depends on the cell size. Larger cells may require a larger ROI and vice-versa for a smaller ROI. These parameters must be determined ahead of time for each cell type and fluorescence dye of interest.
4. Population characterization and select features to store
   1. Once the various acquisition parameters are defined in the HCS imaging protocol, set the HCS system into the data acquisition mode.
   2. The image acquisition parameters for this protocol were set as follows: a scan limit of 1000 cells/well, with a minimum requirement of 10 objects (cells) per field.
      NOTE: The number of fields assessed for each well was determined based on the final cell count reaching the number 1000. The HCS system continues searching various locations within the well until it obtains the target population of 1000 cells/well. The acquisition speed thus depends upon the confluency and the total cell numbers present in each well of the 96-well plate. In the current protocol, the total- and average- intensity of channel 2 (representing the superoxide radicals and total ROS production) were collected from each well for further downstream analyses.
5. Image capture, processing, and analysis
   1. After adjusting and finalizing the image acquisition parameters, initiate the scanning process for each 96-well plate.
      NOTE: Cellomics CX7 imaging system enables high-content screening and automated analysis of samples of a variety of parameters, including ROS production within cells in a high-throughput manner. In addition to the ROS production, the HCS system is capable of providing additional valuable information about various parameters such as cell morphology, subcellular localization, and many other cellular features of interest. Once optimized for acquisition, the HCS imaging platform allows for comprehensive, qualitative, and quantitative characterization of cell parameters in a robust manner.
   2. After the scanning process is complete, launch the View Analysis software and export the various quantitative data parameters in a .csv format for additional analysis.
   3. Using third-party software, copy and paste the various quantitative values of interest for additional downstream analyses.
      NOTE: In the current protocol, GraphPad Prism software was used to analyze the DHE intensity values in response to induced oxidative stress due to H₂O₂ and menadione exposures.
6. Data and statistical analysis
   1. Calculate the mean average intensity of channel 2 (representing the total ROS values and superoxide radicals).
      NOTE: All the experiments to calculate the intensity values were repeated three times with 6 replicates per condition at each dosing level.
   2. Perform one-way ANOVA or t-test to measure the differences in the intensity values between various test conditions.

Results

Dihydroethidium (DHE) is a superoxide-responsive fluorescence dye that provides specific information regarding the intracellular ROS states. DHE dye intrinsically emits blue fluorescence in the cytoplasm. However, upon interaction with superoxide radicals, it is transformed into 2-hydroxyethidium, which emits fluorescence in the red wavelengths (>550 nm) (Figure 1). DHE dye is easily transported into the cells and the nucleus. The fluorescence emitted can be visualized with a fluorescence microscopy setup commonly available in many labs. In the current study, the utility of DHE dye on multiple hepatocellular carcinoma cell lines - HUH7, JHH4, and HepG2 cells as a probe of ROS species generation was tested on a high-content screening platform. The cells were treated with two ROS-generating agents - (a) hydrogen peroxide and (b) menadione for 30 min in a dose-dependent manner (see Figure 2 and Figure 3 for dosing details) to induce oxidative stress in the cells followed by DHE dye labeling in a 96-well plate. Both H₂O₂ and menadione increased the fluorescence intensity dramatically in all three cell lines in a dose-dependent manner. Using the image segmentation and quantification algorithms within the high-throughput screening platform, the intensity of ROS-induced DHE fluorescence was quantified in 1000 cells per well across six replicates. Three independent replicates were measured, and the results were aggregated and analyzed. The Hoechst 33342 nuclear stain plays an important role in identifying the plane in which the adherent cells are present in each well. Additionally, the 2-OH-E⁺ and E⁺ fluorescence can be detected in both the nucleus and cytoplasm of the tested cells, respectively, on the high-throughput screening platform (see Figure 4).

Exposure to H₂O₂ resulted in a statistically significant increment of the ROS-induced fluorescence across all the cell lines analyzed in a dose-dependent manner (see Figure 2; bottom panel). However, such a dose-dependent response was not seen in the case of menadione exposures. With menadione, the JHH4 cell line showed a dose-dependent increment in fluorescent intensity, while in HepG2 and HUH7 cell lines, a significant difference in fluorescence intensity was observed only at the higher menadione concentrations (see Figure 3; bottom panel). These differences in ROS generation responses could be due to the distinct mechanisms of ROS production of H₂O₂ and menadione. While H₂O₂ triggers ROS formation in cells in a more direct manner (e.g., through the action of antioxidant enzymes such as peroxidases and catalases), menadione is hypothesized to generate ROS species in an indirect manner through induced mitochondrial dysfunction. Due to the indirect mechanism of ROS production, DHE fluorescence may require more time to manifest with menadione. These findings indicate that ROS-dependent DHE fluorescence is sensitive to the duration of oxidative stress exposures, concentration, and cell types. Optimization of the experimental protocol for each unique cell type is thus a must. More importantly, this protocol demonstrates the feasibility of the use of DHE dye as a ROS detection agent in a high-throughput manner, which is of use in areas such as toxicology, cancer biology, and drug screening.

Figure 1: A schematic illustrating the pathways of oxidative conversion of DHE fluorescence molecule into 2-hydroxyethidium (2-OH-E⁺) and ethidium (E⁺), respectively. The fluorescence emission spectra of 2-OH-E⁺ and E⁺ overlap at an excitation of 480 nm. Please click here to view a larger version of this figure.

Figure 2: Hydrogen peroxide treatment increases intracellular ROS levels in a dose-dependent manner. Hepatocellular carcinoma cells - (A) HUH7, (B) JHH4, and (C) HepG2 were treated with H₂O₂ at doses of 250 µM, 500 µM, 750 µM, and 1000 µM for a period of 30 min. The cells were stained with DHE (10 µM) dye in culture media for an additional 30 min, followed by Hoechst 33342 nuclear staining. A total of 1000 cells were counted in each well. A linear, dose-dependent increase in the DHE fluorescence is observed across all three cell lines (top panel). Statistically significant increases in fluorescence are seen in all cell lines for dosing 500 µM and above (bottom panel). Each dataset is an average of six replicates in a 96-well plate obtained from three independent experiments (n = 3). Treatment conditions were compared to untreated samples using a one-way ANOVA test (**** - p < 0.0001, *** - p < 0.001; image scale bar - 50 µm) Please click here to view a larger version of this figure.

Figure 3: Menadione treatment increases intracellular ROS levels in a dose-dependent manner. Hepatocellular carcinoma cells - (A) HUH7, (B) JHH4, and (C) HepG2 were treated with menadione at doses of 25 µM, 50 µM, 75 µM, and 100 µM for a period of 30 min. The cells were stained with DHE (10 µM) dye in culture media for an additional 30 min, followed by Hoechst 33342 nuclear staining. A total of 1000 cells were counted in each well. A linear, dose-dependent increase in the DHE fluorescence is observed across all three cell lines (top panel). Statistically significant increases in fluorescence are seen consistently for the maximum dose (100 µM; bottom panel). Variability is observed for the remaining doses of menadione across different cell lines (JHH4 > HUH7 > HepG2). Each dataset is an average of six replicates in a 96-well plate obtained from three independent experiments (n = 3). Treatment conditions were compared to untreated samples using a one-way ANOVA test (**** - p < 0.0001, *** - p < 0.001; image scale bar - 50 µm) Please click here to view a larger version of this figure.

Figure 4: Single cell DHE fluorescence - A close-up view of the DHE fluorescence changes observed after treatment with (A) hydrogen peroxide and (B) menadione for a duration of 30 min. Cytoplasmic and nuclear changes of DHE fluorescence are observed in the different cells. The automatic segmentation mask generated by the high-content screening platform is seen as well. Image scale bar - 10 µm. Magnification - 20x. Please click here to view a larger version of this figure.

Supplementary Figure 1: Imaging superoxide-driven DHE fluorescence in response to hydrogen peroxide. Hepatocellular carcinoma cells - (A) HUH7, (B) JHH4, and (C) HepG2 were treated with H₂O₂ at doses of 250 µM, 500 µM, 750 µM, and 1000 µM for a period of 30 min. The cells were then stained with DHE (10 µM) dye in culture media for an additional 30 min. Subsequently, cells were illuminated with 386 nm LED, and fluorescence imaging collected at >560 nm. No nuclear counterstaining was done to avoid spectral cross-talk. A dose-dependent increase in DHE fluorescence is observed at doses > 500 µM. UV-illuminated DHE fluorescence is expected to be due to the majority of superoxide radical production. Each dataset is an average of six replicates in a 96-well plate obtained from two independent experiments (n = 2). Treatment conditions were compared to untreated samples using a one-way ANOVA (**** - p < 0.0001, *** - p < 0.001; image scale bar - 50 µm) Please click here to download this File.

Supplementary Figure 2: Imaging superoxide-driven DHE fluorescence in response to menadione - Hepatocellular carcinoma cells - (A) HUH7, (B) JHH4, and (C) HepG2 were treated with menadione at doses of 25 µM, 50 µM, 75 µM, and 100 µM for a period of 30 min. The cells were then stained with DHE (10 µM) dye in culture media for an additional 30 min. Similar to hydrogen peroxide, a dose-dependent increase in DHE fluorescence was observed. Each dataset is an average of six replicates in a 96-well plate obtained from two independent experiments (n = 2). Treatment conditions were compared to untreated samples using a one-way ANOVA (**** - p < 0.0001, *** - p < 0.001; image scale bar - 50 µm) Please click here to download this File.

Discussion

In this study, a protocol to assess superoxide-driven intracellular reactive oxygen species (ROS) production using dihydroethidium (DHE) fluorescence dye was established on a high-content screening system. A majority of the current protocols available in the literature use the DCFH-DA as a fluorescence imaging probe to quantify ROS species. However, multiple studies have shown the DCFH-DA is not an ideal probe for the measurement of intracellular ROS. Various reasons postulated for the unsuitability of DCFH-DA as a probe include - (i) DCFH-DA does not exhibit a direct reaction with H₂O₂, which makes it an inappropriate dye for evaluating the intracellular H₂O₂ level. (ii) Several one-electron oxidizing species, such as OH^., NO₂^., and ONOO^-, oxidize DCFH to DCF^. (iii) transition metals, cytochrome c, and heme peroxidase can promote DCFH oxidation in the presence of oxygen and H₂O₂. (iv) Most importantly, DCFH-DA produces the intermediate product DCFH^.- / DCF., which, in the presence of oxygen, induces the production of additional superoxide. Dismutation with O₂^.- generates additional H₂O₂, which may result in an artifactual increase of the fluorescence intensity and erroneously elevated ROS levels. Multiple authors have deemed the use of DCFH-DA as an unreliable fluorescent probe for the detection of intra-cellular H₂O₂ and other ROS species^{8,9,10,11,12,13,14}.

In contrast to DCFH-DA, DHE reacts specifically with the superoxide radical, leading to the generation of a fluorescent product, 2-hydroxyethidium (2-OH E⁺). Non-superoxide ROS species can also react with DHE to generate a second fluorescent product, ethidium (E⁺). While the fluorescence spectra of 2-OH-E⁺ and E⁺ are relatively similar, these two molecules represent the end-products of the specific interactions between the intracellular ROS species and DHE and are responsible for the total fluorescence intensity observed within the cells due to oxidative stress. Some studies have indicated selective excitation at shorter UV range wavelengths (<400 nm) may be more specific for 2-OH ethidium excitation (as opposed to E⁺) and, thus, superoxide radical production^7,12,14. This was assessed by exciting the cells with 386 nm LED light alone. Relative to the 480 nm wavelength excitation, a reduced intensity of emission was observed at 560 nm wavelength (see Supplementary Figure 1 and Supplementary Figure 2). However, one cannot be certain that this is solely due to 2-OH Ethidium fluorescence alone and represents a potential shortcoming of the current approach. Other studies have shown UV excitation (e.g., 386 nm) can also result in simultaneous ethidium fluorescence emission, albeit at lower levels compared to 2-OH ethidium^7,15. Regardless, our protocol establishes DHE as a better dye alternative to measure the quantitative differences of the majority of intracellular ROS production using a high-content imaging and screening system.

The lack of intermediate drivers of ROS generation (like those seen in DCFH-DA) is postulated to be a major advantage in the use of DHE to evaluate intracellular ROS production. However, this mechanism of DHE fluorescence is not firmly established yet. In instances where accurate quantification of the ROS levels is necessary, one is advised to use additional orthogonal methods of quantitative validation such as HPLC and/or liquid chromatography-mass spectrometry (LC-MS) methods to accurately estimate the levels of 2-OH-E⁺ within the cells^8,9,13. However, one must be aware that HPLC and LC-MS methods necessarily involve aggregation and solubilization of the cells for evaluation. Thus, the spatio-temporal information encoded in a fluorescence-based imaging method as in the current protocol would be lost in HPLC and LC-MS and represents a major advantage of imaging-based ROS assessment methods. An additional advantage of the red-shifted fluorescence of DHE is the longer wavelength light (in red) is less toxic to cells due to the lower energy carried by red light.

The dose-dependent nature of DHE fluorescence to measure the relative intracellular ROS changes was tested using menadione and hydrogen peroxide as oxidative stress inducers. Menadione and hydrogen peroxide were chosen as the test substances due to the variable mechanisms of ROS and oxidative stress generation^20,21. While ROS production due to menadione exposure is indirect (through mitochondrial dysfunction), H₂O₂ produces oxidative stress in a more direct manner. Increasing concentrations of menadione led to enhanced DHE fluorescence production in a linear, repeatable, and dose-dependent manner. Similar to menadione, H₂O₂ is a well-known inducer of ROS capable of generating hydroxyl radicals (OH^.) through Fenton chemistry in a direct manner²². A linear and dose-dependent response was similarly observed in DHE fluorescence due to H₂O₂ exposure. Due to the selectivity of DHE for the superoxide radical specifically, it does not directly react with the H₂O_2. Instead, the intracellular superoxide radicals produced initially react with H₂O₂ to form secondary ROS species, such as hydroxy radicals via the Haber-Weiss and Fenton reactions, which are then detected by DHE fluorescence²³. The fluorescence emission spectrum of DHE in response to two different oxidizing substances by measuring the majority of intracellular ROS production is established in the current protocol^8,13. These results establish the utility of fluorescent DHE dye as a better alternative for the detection and quantification of intracellular ROS levels using high-throughput screening approaches.

The superior selectivity, sensitivity, and optical properties of DHE make it a valuable alternative for investigating ROS production and oxidative damage in a high-throughput manner, as established in the current protocol. Future studies in our lab will explore the role of DHE as a ROS marker in the cross-reactivity between ROS and cellular signaling mechanisms under various physiological and pathological conditions in a high-throughput setting.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
1.5 mL centrifuge tubes	VWR	20170-038
96- well plate	Corning Costar	07-200-90
Cellomics Cx7	ThermoFisher	HCSDCX7LEDPRO
Collagen	Advanced Biomatrix	5056
DHE (Dihydroethidium)	ThermoFisher	D1168
DMEM	Sigma	6046
FBS	VWR	97068-085
GraphPad Prism	GraphPad	Version 6.0
HepG2 cell line	ATCC
Hoechst	ThermoFisher	33342
HUH7 cell line	ATCC
Hydrogen Peroxide	Sigma	88597
JHH4 cell line	ATCC
Menadione	Sigma	M5625