DRAQ7 overcomes the complications associated with autofluorescence of doxorubicin during flow cytometric analyses

Abstract

Doxorubicin (DOX) is a widely used chemotherapeutic agent that induces apoptosis through DNA intercalation, topoisomerase II inhibition, and oxidative stress. However, its cardiotoxicity is a significant concern. Flow cytometry, particularly with Annexin V/propidium iodide (PI) or Annexin V/7AAD assays, is commonly used to assess DOX-induced cell death. A challenge arises from DOX’s intrinsic fluorescence, which interferes with accurate detection. To address this, we successfully used DRAQ7, a far-red dye, in HeLa and other cancer cell lines. As our study demonstrates DRAQ7 eliminates auto-fluorescence interference, this approach offers precise differentiation of apoptotic cells and improving experimental and therapeutic reliability.

Introduction

Doxorubicin (DOX), a universal chemotherapeutic drug, induces cell death primarily through apoptosis, a regulated form of programmed cell death [1, 2]. DOX-mediated cell death involves extensive DNA damage, primarily through DNA intercalation and the inhibition of topoisomerase II. Additionally, it generates oxidative stress by priming the production of reactive oxygen species (ROS), which further contributes to DNA damage. DOX usage has significantly improved the survival rates of cancer patients, often in combination with other chemotherapy drugs, enhancing its efficacy and applicability [3, 4]. However, a notable drawback of DOX is its potential for cardiotoxicity, which can lead to heart problems, including congestive heart failure [5, 6]. While the exact mechanisms of DOX-induced cardiotoxicity remains elusive, it remains a primary concern associated with DOX treatment.

Flow cytometry is a valuable instrumental approach that enables rapid detection and quantitative analysis of cells in a user-defined manner under control or treatment conditions, such as DOX-induced cell death [7–9]. Flow cytometric analysis of DOX-mediated cell death plays a crucial role in elucidating the molecular mechanisms and dynamics effects of DOX on cells and in evaluating its efficacy in cancer treatment along with its possible side effects. To this end, the Annexin V/propidium iodide (PI) or Annexin V/7-aminoactinomycin D (7AAD) assay is a flow cytometry method widely used to study DOX-induced cell death [10–12]. Annexin V, which binds to phosphatidylserine on the outer membrane of dying cells, is used in conjunction with a DNA-binding dye such as PI or 7AAD to completely determine the phenotypic features of cells in a population [13].

DOX has a natural ability to emit fluorescence when exposed to certain wavelengths of light [14, 15]. This intrinsic fluorescence can be useful for tracking DOX-coupled substances upon cellular uptake. However, auto-fluorescence of DOX can become a challenge in the context of flow cytometry, where precise detection of fluorescence is crucial. For example, the auto-fluorescence of DOX can interfere with the excitation or emission wavelenghts of other fluorochromes, complicating the accurate characterization of cellular features [16, 17]. This interference can lead to inaccurate signal measurement and misinterpretation of experimental results [18–22]. To avoid interference, it is vital to carefully select the fluorescent compounds or dyes with excitation and emission spectra that do not overlap significantly with those of DOX. This oversight is particularly critical as many current studies fail to account for this interference, resulting in compromised data accuracy and validity. The misinterpretation of experimental results due to such spectral overlap not only hinders scientific progress but also leads to potential misdirection in research efforts. The significance of this issue extends beyond mere experimental error; it affects the overall integrity of scientific findings, potentially impacting downstream applications and therapeutic developments. Inaccurate data can lead to false conclusions, misallocation of resources, and delays in critical advancements. Thus, addressing this gap by emphasizing the careful selection of appropriate dyes is essential to enhance the reliability and reproducibility of findings in this field. Researchers must be aware of the potential for DOX auto-fluorescence to confound their results and take proactive measures to mitigate this risk, ensuring the robustness and validity of their scientific investigations.

To circumvent the interference of auto-fluorescence of DOX with other fluorochromes especially in cell death assays by flow cytometry, we propose the use of dyes that absorb light in the far-red region of the absorbance spectrum, typically with peak absorption wavelengths in the range of 650–750 nanometers (nm) [23–26]. This can help avoid potential interference from auto-fluorescence of DOX, particularly in the context of Annexin V/PI or Annexin V/7AAD staining. To this end, we tested the DNA-binding dye DRAQ7, which is efficiently excited at around 633 nm with an emission wavelength at approximately 695 nm, to examine DOX-induced cell death, in place of traditional PI or 7AAD staining. Our findings revealed that DRAQ7 dye effectively eliminated the interference imposed by auto-fluorescence of DOX, providing accurate measurements. We used DRAQ7 to accurately measure the rate of DOX-induced cell death in different cancer cell lines (HeLa, Me180, MCF7, HCC1143) in a dose- and time-dependent manner. This approach holds promise for precise and accurate measurement of DOX-induced cell death in research and clinical applications.

Materials and methods

Cell culture and drug treatments

HeLa cells (ACC 57) were obtained from DSMZ GmbH (Germany). MCF7 (HTB-22) and Me-180 (HTB-33) cells were purchased from ATCC (United States). HCC1143 cells were kindly gifted from Ayşe Elif Erson-Bensan from Department of Biological Sciences in METU (Ankara, Turkey). HeLa, MCF7, Me-180 and HCC1143 cells were cultured in RPMI 1640 (Gibco, United States), DMEM (Gibco, United States), McCoy’s 5A (Gibco, United States) and DMEM F:12 (Gibco, United States) basal media containing 2 mM L-glutamine, respectively. All media were supplemented with 10% fetal bovine serum (FBS) (Gibco, United States) and cells were cultured in an atmosphere of 5% CO₂ at 37°C.

HeLa cells were treated with doxorubicin (CST, United States) and cisplatin (Cayman Chemical Company, United States) in a dose- or time-dependent manner. The other cell lines were treated with doxorubicin at a single concentration and time point. Cell seeding was performed at a density of 150.000 cells/well in a 6-well plate format, one day prior to drug treatment. DMSO (0.2%) (v/v) was used as a negative control for cisplatin treatment.

Measurement of apoptosis by flow cytometry

At the end of the incubation period for each drug treatment, cells were trypsinized using 1X Trypsin-EDTA (0.025%) (Gibco, United States) and washed with 1X cold PBS (Gibco, United States). After centrifugation at 1500 rpm for 5 min at room temperature, cell pellets were resuspended in 1X Annexin binding buffer (Becton Dickinson, United States). Cells were stained with Annexin V-FITC (BioLegend, United States) in combination with either 7-AAD (BioLegend, United States) or DRAQ7 (BioLegend, United States). Single-stained controls, including Annexin V-FITC-only, 7-AAD-only, and DRAQ7-only samples, were employed for compensation. Unstained untreated cells were used to define background fluorescence and gating thresholds. In addition, DOX-treated unstained cells were included to evaluate DOX-associated autofluorescence/spectral interference across the relevant fluorescence channels. Following staining, cells were incubated for 15 min at room temperature in the dark.

Samples were acquired on a FACS Canto flow cytometer (Becton Dickinson, United States) using BD FACSDiva software (Version 6.1.3). The same acquisition settings were applied to all samples within each experiment. Annexin V-FITC, 7-AAD and DRAQ7 fluorescence were collected in FITC-A, PerCP-A and APC-Cy7-A detectors, respectively. The detector configuration included PE-Cy7 excitation at 488 nm with a 780/60 emission filter, APC excitation at 633 nm with a 660/20 emission filter, and APC-Cy7 excitation at 633 nm with a 780/60 emission filter. Detector voltages were kept constant across samples within each experiment. Uniform gating strategy, which included debris exclusion based on FSC-A and SSC-A properties and fluorescence gate definition using unstained and single-stained controls, was applied uniformly to all treatment conditions.

Compensation was performed separately for the Annexin V-FITC/7-AAD and Annexin V-FITC/DRAQ7 panels using unstained and corresponding single-stained controls. DOX-treated unstained cells were additionally analyzed to assess DOX-derived fluorescence in the FITC-A, PerCP-A, and APC-Cy7-A detectors. Quantitative fluorescence analyses were performed by extracting fluorescence intensity values from the relevant channels. DOX-associated spectral interference was assessed by comparing DOX-treated unstained cells with untreated unstained controls. Signal separation between 7-AAD and DRAQ7 was evaluated using their respective detection channels, PerCP-A for 7-AAD and APC-Cy7-A for DRAQ7.

Statistical analysis

All experiments were carried out at least in three biological replicates. Student’s t test was used to analyze the data. P-Value of .05 was considered statistically significant unless indicated otherwise.

Results

Annexin V/7AAD profiles of DOX- and cisplatin-treated HeLa cells

Annexin V/7AAD staining is widely used to identify cell death [12]. We tested its accuracy in detecting doxorubicin (DOX)-induced cell death in HeLa cells by treating them with increasing DOX concentrations (0–8 µM) and staining with Annexin V-FITC and 7AAD (Fig. 1A). The results showed a significant decrease in the Annexin V−/7AAD− population and an increase in the Annexin V−/7AAD+ and Annexin V+/7AAD+ populations, indicating that Annexin V/7AAD staining failed to detect DOX-induced cell death accurately (Fig. 1B). In contrast, cisplatin-treated HeLa cells showed a clear dose-dependent increase in Annexin V+/7AAD− population and a corresponding decrease in viable cells (Annexin V−/7AAD−), confirming the accuracy of Annexin V/7AAD staining for cisplatin (Fig. 1C and D). This discrepancy suggests that DOX auto-fluorescence interferes with 7AAD detection. Further investigation revealed that DOX-treated cells and media displayed unexpected fluorescence in the PerCP channel dedicated to 7AAD detection, complicating data interpretation (Fig. 2A and B). To improve precision, we analyzed DOX-treated cells using various filters (PE-Cy7, APC, APC-Cy7). The majority of DOX-treated cells were detected in the PE-Cy7 channel (Fig. 2C), with minimal detection in the APC channel (Fig. 2D) and absence in the APC-Cy7 channel (Fig. 2E). This suggests that the APC-Cy7 channel can be used without interference from DOX auto-fluorescence. We propose using dyes which absorb light in far-red regions, such as DRAQ7 with an APC-Cy7 fluorochrome, along with FITC, to analyze DOX-induced cell death. Absorption spectrum analysis revealed an overlap between DOX and 7AAD (Fig. 2F) but not between DOX and DRAQ7 (Fig. 2G), indicating minimal interference between DOX and DRAQ7. These findings demonstrate that Annexin V/7AAD staining is unsuitable for DOX-treated experiments due to significant auto-fluorescence interference. Alternative staining methods, such as using DRAQ7 with APC-Cy7, should be employed to accurately measure apoptosis in the presence of DOX. The revelation of DOX auto-fluorescence, particularly in the PerCP channel, underscores the importance of using appropriate controls and background corrections in experiments involving DOX to ensure accurate and reliable data interpretation.

Figure 1.

Apoptotic profiles of HeLa cells treated with A. DOX and C. CP in a concentration-dependent manner for 16 h. B and D depict representative dot plot graphs of DOX and CP treatment, respectively. Q1, Q2, Q3 and Q4 represent percentages of dead (Annexin V −/7AAD +), late apoptotic (Annexin V+/7AAD+), live (Annexin V−/7AAD−) and early apoptotic (Annexin V+/7AAD−) cells, respectively. Error bars represent mean ± SD of three replicate samples (unpaired, two-tailed t-test). ns, non-significant, P > .05, **P ≤ .01, ***P ≤ .001, and ****P ≤ .0001.

Figure 2.

DOX autofluorescence displayed by (A) DOX-treated cells and (B) DOX-containing media without cells analyzed by flow cytometry analysis. DOX-treated cells have also been depicted by PE-Cy7/FITC (C), APC/FITC (D) and APC-Cy7/FITC (E) contour graphs. None of the samples were labeled by apoptotic markers and dyes. HeLa cells were treated with 4 µM DOX for 16 h. Absorption spectra of (F) FITC/7-AAD/DOX and (G) FITC/DRAQ7/DOX. Vertical dashed and straight lines represent the peak wavelength of DOX excitation and emission, respectively. The area under the dashed curved line represents excitation spectrum of the relevant fluorochrome while the color-filled area represents the emission spectrum. The spectra are distinguished according to their colors.

DRAQ7 as an alternative for DOX-induced cell death measurement

To evaluate the effectiveness of DRAQ7 in detecting DOX-induced cell death, HeLa cells were treated with DOX at concentrations ranging from 1 to 8 µM for 16 hours and analyzed using Annexin V/DRAQ7 staining (Fig. 3A and B). This approach demonstrated a clear, dose-dependent increase in the Annexin V+/DRAQ7- population and a corresponding decrease in viable cells. Time kinetics analysis (2–16 hours) further showed a progressive rise in Annexin V+/DRAQ7− cells over time (Fig. 3C and D), solidifying DRAQ7’s capability to accurately detect DOX-induced cell death. To validate DRAQ7’s effectiveness, it was compared with 7AAD. Monochrome staining with DRAQ7 and Annexin V revealed distinct and reliable results for DOX-treated cells, with minimal spectral overlap between DRAQ7 and DOX (Fig. 3E). In contrast, 7AAD showed significant overlap with DOX auto-fluorescence, leading to distorted data and poor detection in DOX-treated cells. The advantages of DRAQ7 were further demonstrated in DOX-treated various cell lines, including MCF7, HCC1143, and ME180. DRAQ7 consistently provided clear and reliable results, unlike 7AAD, which failed to distinguish cell death populations due to spectral overlap with DOX (Fig. 3F). These findings highlight DRAQ7’s effectiveness in accurately identifying different stages of cell death and discerning subtle changes in fluorescence signals. Overall, DRAQ7 stands out as a highly effective staining agent for analyzing DOX-induced cell death. Its minimal spectral overlap with DOX and reliable detection across multiple cell lines make it an alternative to 7AAD, ensuring precise and accurate cell death analysis in experimental settings.

Figure 3.

Apoptotic profiles of HeLa cells treated with doxorubicin in a (A, B) concentration- and (C, D) time-dependent manner for 16 h and 4 µM, respectively. Q1-1 represents live population (Annexin V−/DRAQ7−), as Q4-1 depicts early apoptotic population (Annexin V+/DRAQ7−). Q2-1 and Q1-1 show the late apoptotic (Annexin V+/DRAQ7+) and dead populations (Annexin V−/DRAQ7+), respectively. ns, non-significant, P > .05, **P ≤ .01, ***P ≤ .001, and ****P ≤ .0001. Flow cytometry analysis of (E) CP- and DOX-treated HeLa cells with monochromatic staining and (F) DOX-treated various cell lines with double staining. DOX was used at a concentration of 4 µM while CP concentration was 80 µM, for 16 h of incubation period. Error bars represent mean ± SD of three replicate samples (unpaired, two-tailed t-test). ns, non-significant, P > .05, *P ≤ .05, **P ≤ .01, ***P ≤ .001, and ****P ≤ .0001.

Discussion

This study contributes to the scientific literature by addressing a critical methodological challenge in fluorescence-based cell death assessment: the interference caused by doxorubicin (DOX)-derived fluorescence. This issue is particularly relevant for apoptosis and viability assays performed in DOX-treated cells, where the intrinsic fluorescence of DOX can overlap with commonly used detection channels and complicate accurate interpretation of flow cytometry data. In line with previous observations, our results showed that DOX generated a pronounced fluorescence signal in the PerCP-A detector, which was used for 7-AAD detection [16].

A major strength of this study is that the improvement in flow cytometry interpretation was supported not only by representative dot plots but also by quantitative fluorescence measurements obtained from the original FCS data. DOX-treated cells stained with Annexin V-FITC/7-AAD or Annexin V-FITC/DRAQ7 were evaluated together with unstained, single-stained, and DOX-treated unstained controls. The detector/channel assignments used for these analyses are summarized in Supplementary Table S1, Section A. This approach allowed DOX-associated background fluorescence to be distinguished from dye-specific fluorescence signals.

Quantitative fluorescence assessment of the original FCS data was used to compare the behavior of the 7-AAD and DRAQ7 detection channels in DOX-treated samples. In our experimental setup, 7-AAD was evaluated in the PerCP-A detector, whereas DRAQ7 was evaluated in the APC-Cy7-A detector, a far-red detection channel. In the DOX concentration series, PerCP-A mean fluorescence in the 7-AAD panel increased from 21.8 in unstained cells to 219.1 in 1 µM DOX-treated cells and 314.1 in 8 µM DOX-treated cells, corresponding to approximately 10-fold and 14.4-fold increases, respectively. In comparison, APC-Cy7-A mean fluorescence in the DRAQ7 panel increased more modestly, from 39.8 in unstained cells to 133.6 in 1 µM DOX-treated cells and 157.9 in 8 µM DOX-treated cells. Gate-based analysis further supported this difference: in 1 µM DOX-treated samples, more than 95% of events shifted into the PerCP-high gate, whereas only approximately 6.4%–9.9% of events were detected in the APC-Cy7-high gate. These detector-specific fluorescence changes and gate-based comparisons are summarized in Supplementary Table S1, Section B and Section D.

Taken together, these quantitative comparisons indicate that the PerCP-A channel used for 7-AAD detection was more strongly affected in DOX-treated samples than the APC-Cy7-A channel used for DRAQ7 detection. Therefore, 7-AAD-based cell death measurements may be confounded by DOX-derived fluorescence if this interference is not carefully controlled. In contrast, DRAQ7 provided a more suitable alternative for detecting membrane-compromised cells under these experimental conditions, as its APC-Cy7-A detection channel showed comparatively lower DOX-associated signal. Together with the inclusion of unstained, single-stained, and DOX-treated unstained controls, these quantitative data support the use of DRAQ7 as a more reliable viability dye for DOX-induced cell death assessment. The control samples and corresponding event numbers are listed in Supplementary Table S1, Section B.

The significance of this study extends beyond HeLa cells, as the applicability of DRAQ7 was also examined in additional cell lines, including MCF7, HCC1143, and ME180. By systematically evaluating DRAQ7 performance in response to DOX dose and treatment duration, this study not only addresses a technical limitation of fluorescence-based apoptosis assays but also provides a practical framework for improving the accuracy of cell death detection in the presence of autofluorescent drugs. The combined use of gradient fluorescence assessment, absorption spectrum analysis, appropriate compensation controls, and quantitative FCS-based analysis provides a useful strategy for researchers facing similar challenges with DOX or other autofluorescent compounds.

Author contributions

Yusuf Cem Çiftçi (Conceptualization [lead], Data curation [equal], Formal analysis [lead], Investigation [equal], Methodology [equal], Writing—original draft [equal], Writing—review & editing [equal]), Ipek Erdogan Vatansever (Conceptualization [supporting], Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [equal], Writing—original draft [equal], Writing—review & editing [equal]), and Bünyamin Akgül (Conceptualization [equal], Funding acquisition [lead], Project administration [lead], Resources [lead], Supervision [lead], Writing—original draft [supporting], Writing—review & editing [equal])

Supplementary material

Supplementary material is available at Biology Methods and Protocols online.

Conflicts of interest

The authors declare no conflict of interest.

Funding

This study is funded by the Scientific and Technological Research Council of Turkey (TUBITAK Project No: 113Z371 to BA).

Data availability statement

The data underlying this article are available in the article and in its online supplementary material.