A quantitative real-time approach for discriminating apoptosis and necrosis

Abstract

Apoptosis and necrosis are the two major forms of cell death mechanisms. Both forms of cell death are involved in several physiological and pathological conditions and also in the elimination of cancer cells following successful chemotherapy. Large number of cellular and biochemical assays have evolved to determine apoptosis or necrosis for qualitative and quantitative purposes. A closer analysis of the assays and their performance reveal the difficulty in using any of these methods as a confirmatory approach, owing to the secondary induction of necrosis in apoptotic cells. This highlights the essential requirement of an approach with a real-time analysis capability for discriminating the two forms of cell death. This paper describes a sensitive live cell-based method for distinguishing apoptosis and necrosis at single-cell level. The method uses cancer cells stably expressing genetically encoded FRET-based active caspase detection probe and DsRed fluorescent protein targeted to mitochondria. Caspase activation is visualized by loss of FRET upon cleavage of the FRET probe, while retention of mitochondrial fluorescence and loss of FRET probe before its cleavage confirms necrosis. The absence of cleavage as well as the retention of mitochondrial fluorescence indicates live cells. The method described here forms an extremely sensitive tool to visualize and quantify apoptosis and necrosis, which is adaptable for diverse microscopic, flow cytometric techniques and high-throughput imaging platforms with potential application in diverse areas of cell biology and oncology drug screening.

Figure 1

(a) U251 cells stably expressing the ECFP-DEVD–EYFP FRET probe (U251 caspase sensor) were developed as described. The microscopic images in brightfield and fluorescence channels are shown. (b) U251 caspase sensor cells were treated with doxorubicin (200 ng/ml) for 24 h. The ECFP–EYFP FRET channels and merged image of ECFP/EYFP channels are shown along with the DIC image. The respective image for the untreated control is also shown. (c) U251 caspase sensor cell line grown in 96-well glass-bottom plates were exposed to valinomycin (5 μM). The real-time imaging for caspase activation was evaluated using microscope as described in the 'Materials and Methods'. The merged images of DIC/ECFP/EYFP and ratio are shown from indicated time points. (d) U251 caspase sensor cell line grown in 96-well glass-bottom plates were exposed to H₂O₂ (2.5 mM). The real-time imaging for caspase activation was evaluated using microscopy as described in the 'Materials and Methods'. The merged images of DIC/ECFP/EYFP and ratio are shown from indicated time points.

Figure 2

(a) U251 caspase sensor cell line was transfected with DsRed targeted at mitochondria to generate stable cells expressing both the probe, U251 DEVD Mito Ds cells. Representative images in brightfield, ECFP, EYFP, DsRed channels and ratio mode of the stable cells are shown. (b) U251 DEVD Mito Ds cells were treated with caspase activating compound doxorubicin (200 ng/ml). The real-time imaging for caspase activation and Mito-DsRed fluorescence was carried out using fluorescence microscope at an interval of 10 min as described in the 'Materials and Methods'. The snapshot of merged channels of ECFP/EYFP/DsRed at 3 h, 6 h, 12 h and 24 h of treatment is shown. The caspase activation is reflected from the FRET loss and subsequent increase in blue channel fluorescence. (c) Images representing three distinct cell populations identified after doxorubicin (200 ng/ml) treatment of U251 DEVD Mito-DsRed cell: apoptotic cells with ECFP/EYFP ratio change and retaining red fluorescence, necrotic cells without any ECFP/EYFP fluorescence and retaining red fluorescence and live cells with intact ECFP/EYFP probe without ratio change and retaining red fluorescence. In the image, 'A' represents apoptotic cells, 'N' represents necrotic cells and 'L' represents the live intact cells. (d) U251 DEVD Mito Ds cells were either untreated or treated with caspase activating compound for 24 h. The cells were imaged with fluorescence microscope for ratio imaging and DsRed channel. The cells with caspase activation and necrosis were identified on the basis of ratio change and DsRed fluorescence as described. The quantitative necrosis and apoptosis from three different independent experiments are shown in the graph. The results represented are mean±S.D. (n=3).

Figure 3

(a) U251 DEVD Mito Ds cells were treated with CCCP (5 μM). The real-time imaging for caspase activation and Mito-DsRed fluorescence was carried out using confocal microscopy at an interval of 15 min as described in 'Materials and Methods'. The representative ECFP/EYFP ratio images and merged channels of ECFP/EYFP/DsRed are shown from indicated time points. (b) U251 DEVD Mito Ds cells were treated with necrosis-inducing agent H₂O₂ (2.5 mM). The real-time imaging for caspase activation and Mito-DsRed fluorescence was carried out using confocal microscopy at an interval of 15 min as described in 'Materials and Methods'. The representative ECFP/EYFP ratio images and merged channels of ECFP/EYFP/DsRed are shown from indicated time points.

Figure 4

(a) U251 DEVD Mito Ds cells were grown on 96-well glass-bottom plates and treated with different drugs. High-throughput imaging for ratio and Mito-DsRed was carried out as described using pathway bioimager. A representative montage image of ECFP–EYFP Ratio and a merged channel of ECFP/EYFP/DsRed and scatter plot generated after proper segmentation and analysis for untreated control and cisplatin (50 μg/ml)-treated samples is shown. (b) The merged images and scatter plot from high-throughput imager for cells treated with indicated drugs are shown. (c) U251 DEVD Mito Ds cells grown in 96-well glass-bottom plates were treated with necrosis-inducing agent, H₂O₂ (2.5 mM). Merged image and scatter plot from high-throughput imager for treated and control cells are shown.

Figure 5

(a) U251 DEVD Mito Ds cells were exposed to different drugs for 24 h. The cells were analyzed using flow cytometer for green fluorescence at 488 nm and ratio mode for FRET at 405 nm. The necrotic cells are gated based on Mito-DsRed fluorescence against green channel. The cell with increase in ratio represents cells with caspase activation. (b) The percentage of cells with apoptosis and necrosis were scored from the three different flow cytometry data for the indicated drugs. Results represented are mean±S.D. (n=3). (c) U251 DEVD Mito Ds cells were exposed to different drugs for 24 h. The cells were trypsinized and stained with Alexa Fluor 647 conjugated Annexin-V as described. The cells were analyzed for ratio change and annexin-V staining. The scatter plot of annexin-V versus ratio of the cells treated with indicated drugs are shown.

Figure 6

The schema summarizes the potentiality of the tool to discriminate the two forms of cell death: apoptosis and necrosis in a spatio-temporal manner. The apoptotic cells appear blue due to FRET loss, at the same time retaining the insoluble Mito-DsRed probe; whereas, in necrotic cells, the FRET probe is lost due to membrane permeabilization and are characterized by the presence of insoluble Mito-DsRed probe alone. The healthy cells retain both the intact FRET probe and the Mito-DsRed probe