Retention of parental line characteristics in OVCAR-8-DsRed2 and NCI/ADR-RES-EGFP fluorescent lines. (A) Representative confocal images (×40) show fluorescence levels of OVCAR-8-DsRed2 and NCI/ADR-RES-EGFP cells cultured in the presence (left panels) and absence (right panels) of G418 selection after 72 h. (B) Growth curves of OVCAR-8 parental (yellow squares), OVCAR-8-DsRed2 (red squares), NCI/ADR-RES parental (light green triangles), and NCI/ADR-RES-EGFP (dark green triangles) lines show the fluorescent cells retain parental growth kinetics through day 6 (n = 2 for all lines; error bars represent SEM). (C) Relative mRNA expression levels of ABCB1, ABCC1, and ABCG2 (blue, yellow, and green columns, respectively) were determined by quantitative real-time RT-PCR and normalized against PMCA4 control gene expression. Fold changes of RT-PCR Cp values for ABCB1, ABCC1, and ABCG2 transcript expression between each fluorescent line and its parental and opposing fluorescent lines show typical expression patterns. (D) Comparative expression of P-gp in the parental and fluorescent lines is shown by western blot analysis with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as a control.

CaAM and mitoxantrone accumulation in fluorescent lines shows both retain parental efflux activity. Flow cytometry analysis of (A) OVCAR-8-DsRed2 and (B) NCI/ADR-RES-EGFP lines shows accumulation of CaAM with and without CsA inhibition of P-gp. OVCAR-8-DsRed2 and NCI/ADR-RES-EGFP cells are shown without addition of drug (orange line), with CaAM alone (purple line), and with both CaAM and CsA (blue line). The NCI/ADR-RES-EGFP line naturally fluoresces in the FL1 channel and so exhibits a much higher background level of green fluorescence. Representative confocal images (×40) show co-cultures of fluorescent lines incubated with (C) mitoxantrone only (40 μM) or (D) mitoxantrone (40 μM) and CsA (10 μM). (C) Mitoxantrone-dosed cultures show the accumulation of drug in OVCAR-8-DsRed2 cells (magenta merge) and effective clearance from NCI/ADR-RES-EGFP cells. (D) Inhibition of P-gp by CsA eliminates this selective efflux of mitoxantrone (magenta and white merge, respectively), as both lines show similar drug accumulation.

Compatibility assessment of the cell lines with a high-throughput laser-scanning microplate cytometer. (A) Schematic of a 96-well plate plated for cytotoxicity determination, showing the distribution scheme for red and green cells (see Materials and Methods). Examples of plate “heat maps” are shown denoting relative cell counts on a single plate on both (B) red (OVCAR-8-DsRed2) and (C) green (NCI/ADR-RES-EGFP). The Acumen software assigns a color to each well based on the range of cell counts observed on each plate (bright green corresponding to the highest value and black to the lowest). (D) Representative single-well false-color image of co-cultured OVCAR-8-DsRed2 and NCI/ADR-RES-EGFP cells imaged by microplate cytometer. Sample histograms of co-cultured (E) red and (F) green cells detected on their respective fluorescent channels demonstrate the gating used to define viable cells. Population data for single-well co-cultures treated with no drug (dark gray), 31.25 nM Taxol (gray), or 500 nM taxol (light gray) are overlaid, with “viable cell” definition windows highlighted in green and red. All cells displaying fluorescence intensities greater than 16,500 fluorescence units (FLU) (the maximum detectable value), a population later determined to be apoptotic, are grouped together on the far right side of the histograms (denoted by *). (G) False-color well images (from row D of the plate shown in A–C) demonstrate the serial decrease in red cells within co-cultures in 96-well plates dosed with increasing Taxol concentrations: OVCAR-8-DsRed2 by MTT assay (dark red), in monoculture (red), and in co-culture (pink) and NCI/ADR-RES-EGFP in MTT assay (dark green), in mono-culture (green), and in co-culture (light green). (H) Comparison of cytotoxicity curves of Taxol against OVCAR-8-DsRed2 and NCI/ADR-RES-EGFP lines generated by both MTT assay and Acumen in 1,536-well plates: OVCAR-8-DsRed2 mono-culture (dark red) and co-culture (red) and NCI/ADR-RES-EGFP mono-culture (dark green) and co-culture (light green), with mono-culture and co-culture averages (solid line and dashed line, respectively).

Correlation of viability analysis by cell fluorescence and viability dye accumulation. Percentages of nonviable OVCAR-8-DsRed2 cells treated with 1:2 titrations of Taxol were determined by Acumen microplate cytometry using both fluorescence gating as described in Materials and Methods (white bars) and accumulation of Sytox Orange cell-impermeant viability dye (black bars)—which fluoresces in the deep red—to determine the accuracy of Acumen gate-based viability measurements. Viabilities determined by both methods were found to correlate closely at each concentration point, with no statistically significant difference in cell count at any taxol concentration (n = 3; error bars represent SEM), suggesting the Acumen fluorescence thresholds used to count viable cells (and exclude apoptotic subpopulations) provide an accurate high-throughput assessment of cell viability.

Analysis of fluorescence within the OVCAR-8-DsRed2 and NCI/ADR-RES-EGFP cell lines. (A) Both fluorescent lines were examined by flow cytometry to determine the percentage of fluorescent cells: 100% NCI/ADR-RES-EGFP (green line) and 100% OVCAR-8-DsRed2 (red line). Comparison between FSC and fluorescence channel data determined that 98.84% and 98.67% of NCI/ADR-RES-EGFP (B, FL1) and OVCAR-8-DsRed2 (C, FL2) cell populations, respectively, exhibited fluorescence in their respective channels. (D) Confocal images of fluorescent co-cultures stained with 4′,6-diamidino-2-phenylindole (DAPI) nuclear stain demonstrate the absence of viable nonfluorescent cells. Co-cultures were stained with 2 μg/ml DAPI in 1× PBS and incubated for 20 min prior to imaging. Fields shown in images are at ×40 and are representative of the entire co-cultures.

Drug sensitivities (IC₅₀, in μM; mean ± SD) of fluorescent OVCAR-8 and NCI/ADR-RES lines are comparable to parental lines as determined by MTT cytotoxicity assay. Cytotoxicity data were generated comparing the sensitivity of the OVCAR-8-DsRed2 and NCI/ADR-RES-EGFP lines to the parental OVCAR-8 and NCI/ADR-RES lines. Cells were treated with NSC73306, a P-gp-selective compound, or doxorubicin, a well-characterized P-gp substrate, for 72 h, and their viability was assayed by MTT (n = 3 for each line/drug pair). (A) IC₅₀ and RR values (fold sensitivity of high P-gp-expressing line compared with low P-gp-expressing line) show the fluorescent cell pair retains parental levels of sensitivity to both paclitaxel and NSC73306. Representative confocal images (×40) show co-cultures of fluorescent lines dosed with (B) paclitaxel (0, 50, and 100 nM) and (C) NSC73306 (0, 2, and 5 μM).

Discrimination of mixed OVCAR-8-DsRed2 and NCI/ADR-RES-EGFP populations using flow cytometry. Populations of both fluorescent lines were mixed at known proportions and analyzed by flow cytometry on the (A) FL1 (green) and (B) FL2 (red) channels: 100% OVCAR-8-DsRed2 (red line), 75% OVCAR-8-DsRed2–25% NCI/ADR-RES-EGFP (orange line), 50% OVCAR-8-DsRed2–50% NCI/ADR-RES-EGFP (gold line), 25% OVCAR-8-DsRed2–75% NCI/ADR-RES-EGFP (blue line), and 100% NCI/ADR-RES-EGFP (green line). Histograms of both cell lines are directly correlated to the number of cells present, and both are clearly distinguishable on their respective channels. Red fluorescence bleed-through on the green channel allows for direct quantification of both lines simultaneously using the FL1 channel. (C) Density plot data shows that the cell lines scatter distinctly. When gated, these populations of DsRed2 (red dotted lines, 49.99% of total cell count) and EGFP (green dotted lines, 48.96% of total cell count) cells directly overlay with the peaks seen in ungated 50% DsRed2–50% EGFP (blue solid lines) data on both the (D) green and (E) red channels, again allowing for quantifiable discrimination of the two lines by flow cytometry analysis.

Spectra of the emission (Em.) wavelengths of EGFP and DsRed2 proteins reveal the spectral window available for the inclusion of a third fluorophore excited at 488 nm. Fluorophores useful for indicating cell viability, substrate efflux, and inhibition or as markers for immunostaining may be employed. Examples of fluorophores that emit within this region that are also P-gp substrates are shown below: 7-aminoactinomycin D (7AAD), LDS 751, and mitoxantrone. Inhibition of P-gp would lead to increased cellular accumulation, and therefore greater fluorescence intensity, in the region described.

Morphological comparison of OVCAR-8/NCI/ADR-RES parental and fluorescent cell line pairs. Similarly confluent cultures of both parental (left panels) and fluorescent (right panels) OVCAR-8/NCI/ADR-RES cell lines were imaged by brightfield and ultraviolet (UV) (fluorescent lines only) microscopy to demonstrate morphological differences inherent to each pair. Both NCI/ADR-RES cell lines exhibit a larger cell size than their respective OVCAR-8 lines. Cell pairs were imaged on a Leica Microsystems (Bannockburn, IL) DM IL inverted microscope at ×10 using a QImaging (Surrey, BC, Canada) MicroPublisher version 3.3 RTV camera.