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.
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.
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 cells1,2. 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 pathways1,2. 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 stress1,3. 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 disorders1,3,4. 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-established7,8. 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 tissues8,9. 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 levels8,9,10,11,12,13,14.
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., H2O2, 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 fluorescence8,9,10,11,13,14,15,16. 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 emission15,16,17.
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.
1. Cell culture
2. Stock and dilute solution preparation for DHE staining of cells
3. DHE staining process
4. Image acquisition and intensity measurement
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.
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 species8,9,10,11,12,13,14.
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 production7,12,14. 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 ethidium7,15. 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 cells8,9,13. 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 generation20,21. 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 H2O2. 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 protocol8,13. 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.
The authors have nothing to disclose.
RK and RRG were supported by a grant from the UNM Center for Metals in Biology and Medicine (CMBM) through NIH NIGMS grant P20 GM130422. RRG was supported by a pilot award from the NM-INSPIRES P30 grant 1P30ES032755. The imaging core support for the CX7 Cellomics instrument was provided through the AIM center cores funded by NIH grant P20GM121176. We would like to thank Dr. Sharina Desai and Dr. Li Chen for their invaluable assistance with technical issues related to the use of the CX7 Cellomics imaging platform.
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 |