Investigation of Mitochondrial Metabolic Response to Doxorubicin in Prostate Cancer Cells: An NADH, FAD and Tryptophan FLIM Assay

Abstract

Prostate cancer (PCa) is one of the leading cancers in men in the USA. Lack of experimental tools that predict therapy response is one of the limitations of current therapeutic regimens. Mitochondrial dysfunctions including defective oxidative phosphorylation (OXPHOS) in cancer inhibit apoptosis by modulating ROS production and cellular signaling. Thus, correction of mitochondrial dysfunction and induction of apoptosis are promising strategies in cancer treatment. We have used Fluorescence Lifetime Imaging Microscopy (FLIM) to quantify mitochondrial metabolic response in PCa cells by tracking auto-fluorescent NAD(P)H, FAD and tryptophan (Trp) lifetimes and their enzyme-bound fractions as markers, before and after treatment with anti-cancer drug doxorubicin. A 3-channel FLIM assay and quantitative analysis of these markers for cellular metabolism show in response to doxorubicin, NAD(P)H mean fluorescence lifetime (τ_m) and enzyme-bound (a₂%) fraction increased, FAD enzyme-bound (a₁%) fraction was decreased, NAD(P)H-a₂%/FAD-a₁% FLIM-based redox ratio and ROS increased, followed by induction of apoptosis. For the first time, a FRET assay in PCa cells shows Trp-quenching due to Trp-NAD(P)H interactions, correlating energy transfer efficiencies (E%) vs NAD(P)H-a₂%/FAD-a₁% as sensitive parameters in predicting drug response. Applying this FLIM assay as early predictor of drug response would meet one of the important goals in cancer treatment.

Figure 1

Increase in the enzyme-bound fraction of NAD(P)H-a₂% upon doxorubicin treatment in PCa cells. Upper panel (a–f) LNCaP and lower panel (g–l) E006AA (AA) cells. PCa cells: LNCaPs (n = 26) with 4400–4890 ROIs, and E006AA cells (n = 28) with 5585–6372 ROIs, each ROI of 2 × 2 pixels, were analyzed for 5 time points (0, 15, 30, 45 & 60 min). The representative (a–e and g–k) color coded images and (f and l) the histograms showed increase in the enzyme-bound fraction of NAD(P)H-a₂% with time (0–60 min) upon treatment with 1 µM doxorubicin when compared with untreated 0 min time point control (Ctrl). Greater magnitude of response was seen in LNCaP vs E006AA cells (60% vs 23%) from 0 to 60 min.

Figure 2

Decrease in the enzyme-bound fraction of FAD-a₁% upon doxorubicin treatment in PCa cells. Upper panel (a–f) LNCaP and lower panel (g–l) E006AA (AA) cells. PCa cells: LNCaPs (n = 26) with 4400–4890 ROIs, and E006AA cells (n = 28) with 5585–6372 ROIs, each ROI of 2 × 2 pixels, were analyzed for 5 time points (0, 15, 30, 45 & 60 min). The representative (a–e and g–k) color coded images and (f and l) the histograms showed decrease in the enzyme-bound fraction of FAD-a₁% with time (0–60 min) upon treatment with 1 µM doxorubicin when compared with untreated 0 min time point control (Ctrl).

Figure 3

Increase in mitochondrial NAD(P)H-a₂/FAD-a₁ Redox Ratio upon doxorubicin treatment in PCa cells. Mitochondrial OXPHOS activity uses NAD(P)H (more enzyme bound-a₂%) and produces FAD (less enzyme bound-a₁%). The histograms showed increase in mitochondrial NAD(P)H-a₂/FAD-a₁ Redox Ratio in (a) LNCaP and (b) E006AA (AA) cells for the same ROIs (mentioned above) with time (0–60 min) upon treatment with 1 µM doxorubicin compared with 0 min untreated control (Ctrl). The data suggests that there is increased mitochondrial OXPHOS activity with doxorubicin treatment. Greater magnitude of response was seen in LNCaP vs E006AA cells (89% vs 38%) from 0 to 60 min.

Figure 4

Increase in E% with increase in mitochondrial NAD(P)H-a₂/FAD-a₁ Redox Ratio upon doxorubicin treatment in PCa cells. Correlation plots showed increase in median E% w.r.t. median NAD(P)H-a₂/FAD-a₁ redox ratio from control (Ctrl) to doxorubicin treatment groups with time (0–60 min) in both (a) LNCaP and (b) E006AA (AA) cells which suggests increased interaction of Trp (residues in proteins/enzymes) with NAD(P)H with increased mitochondrial OXPHOS activity.

Figure 5

Increase in mitochondrial ROS upon doxorubicin treatment. LNCaP cells were labeled with MitoSox Red. After imaging of the control cells for 0 min time point, 1 µM doxorubicin was added on stage and proceeded with time lapse imaging at 15 min intervals up to 60 min. Representative images of cells (n = ~100) (a) 0 min control (Ctrl), (b) increased ROS at 60 min doxorubicin treatment, (c) auto-contrasted image with clusters of adjoining cells (n = ~100) grouped together to represent individual ROIs as numbered, (d) were quantified and showed a 1.5-to 5.4-fold increase in ROS over time after doxorubicin treatment.

Figure 6

Increase in caspase-3 activity with time upon doxorubicin treatment in PCa cells. (a) LNCaP and (b) E006AA (AA) PCa cells were treated with 1 µM doxorubicin for different time points and assessed for the effector caspase-3 activity by DEVDase assay. There was no increase in caspase-3 activity in earlier time points 0 min–1 hr (Pre-Apoptotic Phase), which increased significantly ***p < 0.001 and ****p < 0.0001 at 12 hr and 24 hr (Apoptotic Phase) with doxorubicin treatment as compared to control group.

Figure 7

Median correlation of E% vs NAD(P)H-a₂%/FAD-a₁% an indicator of doxorubicin responsiveness. Upper panel (a–c) LNCaP and lower panel (d–f) E006AA (AA) PCa cells. Representative (a,b and d,e) NAD(P)H photon images show apoptotic cell morphology in (b) LNCaPs-doxorubicin “Responsive” cells as compared to the (a) LNCaP control cells; whereas the (d) E006AA control and (e) E006AA-doxorubicin “Slow Responders” look morphologically similar. The median correlation of E% vs NAD(P)H-a₂%/FAD-a₁% showed (c) decrease in E% in LNCaPs-doxorubicin “Responsive” cells and (f) no change in E% in E006AA-doxorubicin “Slow Responders” with increase in NAD(P)H-a₂%/FAD-a₁% ratio.