Lenticular mitoprotection. Part A: Monitoring mitochondrial depolarization with JC-1 and artifactual fluorescence by the glycogen synthase kinase-3β inhibitor, SB216763

• Introduction
• Methods
• Results
• Discussion

Abstract

Purpose: Dissipation of the electrochemical gradient across the inner mitochondrial membrane results in mitochondrial membrane permeability transition (mMPT), a potential early marker for the onset of apoptosis. In this study, we demonstrate a role for glycogen synthase kinase-3β (GSK-3β) in regulating mMPT. Using direct inhibition of GSK-3β with the GSK-3β inhibitor SB216763, mitochondria may be prevented from depolarizing (hereafter referred to as mitoprotection). Cells treated with SB216763 showed an artifact of fluorescence similar to the green emission spectrum of the JC-1 dye. We demonstrate the novel use of spectral deconvolution to negate the interfering contributing fluorescence by SB216763, thus allowing an unfettered analysis of the JC-1 dye to determine the mitochondrial membrane potential.

Methods: Secondary cultures of virally transfected human lens epithelial cells (HLE-B3) were exposed to acute hypoxic conditions (approximately 1% O₂) followed by exposure to atmospheric oxygen (approximately 21% O₂). The fluorescent dye JC-1 was used to monitor the extent of mitochondrial depolarization upon exposure of inhibitor treatment relative to the control cells (mock inhibition) in atmospheric oxygen. Annexin V-fluorescein isothiocyanate/propidium iodide staining was implemented to determine cell viability.

Results: Treatment of HLE-B3 cells with SB216763 (12 µM), when challenged by oxidative stress, suppressed mitochondrial depolarization relative to control cells as demonstrated with JC-1 fluorescent dye analysis. Neither the control nor the SB216763-treated HLE-B3 cells tested positive with annexin V-fluorescein isothiocyanate/propidium iodide staining under the conditions of the experiment.

Conclusions: Inhibition of GSK-3β activity by SB216763 blocked mMPT relative to the slow but consistent depolarization observed with the control cells. We conclude that inhibition of GSK-3β activity by the GSK-3β inhibitor SB216763 provides positive protection against mitochondrial depolarization.

The loss of cellular respiration increases the levels of reactive oxygen species (ROS) in human lens epithelial cells (HLE-B3) [1]. A sufficient increase in ROS may also cause a collapse of the mitochondrial membrane potential (∆Ψ) in a process termed mitochondrial membrane permeability transition (mMPT) [2,3]. Dissipation of ∆Ψ prompts further disruption of the electron transport chain, decreasing the production of ATP, and increasing the formation of ROS [4,5] in a harmful downward cycle. The loss of ∆Ψ also leads to the release of apoptotic factors that can cause cellular dysfunction and cell death [6].

HLE-B3 cells have developed protective mechanisms to prevent the loss of ∆Ψ caused by mMPT during oxidative stress. mMPT is mediated via the opening of the mitochondrial permeability transition pore, a pore permeable to solutes of less than 1.5 kDa and sensitive to an accumulation of ROS [7-9]. Studies in the cardioprotection literature have shown that GSK-3β is a crucial enzyme involved in preventing the collapse of ∆Ψ through dynamic regulation of the opening and closing of the mitochondrial permeability transition pore [10,11]. Active GSK-3β allows the pore to open whereas the enzyme’s inactivation blocks the pore from opening. Mitochondrial protection (hereafter referred to as mitoprotection) dictates that impeding the opening of the permeability transition pore prevents mitochondrial depolarization, thus averting entry into the cell death pathway [12]. Recent studies involving pre-/post-conditioning ischemic reperfusion in mouse and rat cardiac myocytes have shown that under conditions of oxidative stress, inhibition of GSK-3β activity prevents the loss of ∆Ψ [12-14]. To date, no such studies linking GSK-3β with mMPT have been reported in an ocular system.

Currently, there have been no reported studies involving the use of SB216763 (a GSK-3β inhibitor) as it influences the mitochondrial membrane potential as analyzed with the potentiometric dye JC-1. In this report, we show that SB216763 contributes to the green emission spectrum thus contributing to a false result of depolarization. Our study describes the use of a technique that will enable the precise reconvolution of the proper contributions from JC-1 green and red emissions. The data demonstrate that preventing mitochondrial depolarization, via the use of SB216763, presumably due to blocking the opening of the mitochondrial membrane permeability transition pore, positively correlates with inhibiting GSK-3β enzymatic activity.

The influence of SB216763 on the pore as analyzed with JC-1 analysis has not previously been reported due to the emission spectrum of cells treated with SB216763 in the absence of JC-1 revealing a broad-spectrum over the range of 500–650 nm. In this study, we make novel use of spectral deconvolution based on experimental measurements, fluorophore reference spectra, and an algorithm for least-squares minimization to produce corresponding unmixed spectra. After deconvolution, the green/red intensity ratios (540/595 nm) provided by the JC-1 dye may be used to calculate the extent of mitochondrial depolarization.

The dynamics of the mitochondrial membrane permeability transition pore and its regulation regarding the opening and closing of the pore are tightly controlled by GSK-3β [10-12,18]. SB216763 is a potent and highly selective inhibitor of GSK-3β activity [19]. A role for GSK-3β in mitoprotection has been supported by the work of Juhaszova et al. [12], who concluded that “this enzyme, located proximally (in the sense of a signaling cascade) to the mitochondrial transition pore complex, acts as a master switch to convey a multiplicity of protective signals to their final point, the mitochondrial permeability transition pore.” Forster et al. [14] demonstrated that inhibition of GSK-3β by SB216763 prevented the oxidant-induced depolarization of ∆Ψ (presumably caused by the inhibition of the opening of the transition pore) in rat ventricular myocytes. The reintroduction of atmospheric oxygen after acute hypoxia led to a slow but inevitable loss of ∆Ψ (Figure 1E). SB216763 was used to demonstrate that inactivation of the enzyme prevents the loss of membrane potential (Figure 1E).

Interpretations other than inhibiting GSK-3β resulting in preventing the opening of the mitochondrial permeability transition pore are equally plausible as an explanation for our data. JC-1 is a novel membrane-permeable cationic carbocyanine dye that accumulates to a high concentration in the mitochondria as J-aggregates and exhibits an emission maximum at about 590 nm. JC-1 may diffuse across the outer and inner mitochondrial membrane in response to the changes in the mitochondrial membrane potential. Diffusion to the cytoplasm from the mitochondria results in a lower concentration, where the dye exists as a monomer and yields green fluorescence with an emission maximum about 540 nm. Loss of JC-1 from the mitochondria is not inevitably a measure of opening of the membrane permeability transition pore. Rather, the loss is a measure of the decrease in the inner mitochondrial membrane potential, possibly due to the loss of integrity of the inner and outer mitochondrial membranes. That is, JC-1 is permeable to the mitochondrial membrane, independent of the transition pore, and the dye’s accumulation or loss does not necessarily depend on the state of the transition pore. Therefore, treatment with SB216763 might alternatively be explained as the drug’s ability to maintain the transmembrane potential of the inner mitochondrial membrane. Regardless of how one chooses to interpret the data, the fact remains that SB216763 elicited lenticular mitoprotection.

Past studies have suggested that the phosphorylation of GSK-3β indicates inactivation of the enzyme [20]. Therefore, a key question that we wished to address in our study was whether the presumed inhibition of GSK-3β activity would positively correlate with the prevention of opening the mitochondrial membrane transition pore. To evaluate whether SB216763 blocked mitochondrial depolarization, we performed a JC-1 assay. The fluorescence of the SB216763 inhibitor was broad and overlapped with the green (540 nm) and red (595 nm) peaks of JC-1 emission (Figure 1A). This significantly affected the signals detected from the cells, as seen in Figure 1C. The microscopy images were corrupted by the fluorescence of the SB216763 inhibitor, and we were not able to conclude about the eventual changes of green/red intensities of JC-1. Fortunately, the JC-1 emission consisted of two well-separated peaks (Figure 1B). Each measured spectrum presented in Figure 1C is composed of these two JC-1 peaks and the SB216763 broad spectrum. The spectral deconvolution described in the Methods section enables reconstruction of the JC-1 spectrum without SB216763 interference, as presented in Figure 1D. With this heightened sensitive technique, the changes in the JC-1 spectra of the control, mock-treated cells indicated a gradual but measured depolarization over 75 min of analysis, in the absence of SB216763 (Figure 1E). In the presence of SB216763, depolarization was effectively inhibited (Figure 1E). These results suggest that GSK-3β activity is a prerequisite for the normal functioning of the mitochondrial permeability transition pore to open and close. Furthermore, we suggest inhibiting GSK-3β activity from influencing the dynamics of the mitochondrial transition pore can prevent the loss of ∆Ψ (Figure 1E).

Others have previously reported that SB216763 displays artifactual fluorescence [21]. The emission spectrum of SB216763-treated cells illustrates a broad spectrum over the range of 500–650 nm (Figure 1A). Furthermore, Figure 2 explicitly demonstrates the green fluorescence of undissolved SB216763 particulate matter with cells treated with SB216763 in the absence of annexin V-fluorescein isothicyanate/propidium iodide. The particulates stick to the cell surface of the cells and are not incorporated within cells. Moreover, the lack of red cells with a green halo in either the SB + annexin V-fluorescein isothicyanate/propidium iodide–treated cells (Figure 2, second row) or annexin V-fluorescein isothicyanate/propidium iodide–treated cells in the absence of SB216763 (Figure 2, third row) indicates that apoptosis/necrosis is not occurring under the conditions of the experiment with either the control or SB216763-treated cells. The latter point is not trivial as it confirms that the technical manipulation of switching the cells from hypoxia to oxygen exposure (refer to Methods) is not, of itself, sufficient oxidative stress to elicit entry into the cell death pathway.

The noted green fluorescent aggregates (Figure 2, fourth row) were likely due to the poor solubility of the inhibitor, and therefore, the inhibitor may not have been effective at inhibiting GSK-3β activity. To confirm that SB216763 did, indeed, inactivate GSK-3β activity, we treated the HLE-B3 cells with SB216763 and verified the enzymes’ inability to phosphorylate its downstream substrate, glycogen synthase (GS). The failure to adequately phosphorylate GS indicated the inactivation of the active site of GSK-3β (manuscript in preparation). The inhibition of GSK-3β activity (as monitored by the failure to phosphorylate GS) positively associated with preventing mitochondrial membrane permeability transition. We further examined the influence of the inactivation of GSK-3β activity on another substrate, β-catenin. β-catenin is a prominent nuclear transcription factor whose activity is regulated by GSK-3β. When GSK-3β is active, it inactivates β-catenin by phosphorylation. Treatment of HLE-B3 cells with SB216763 resulted in decreased phosphorylated β-catenin, the consequence of which was the increased translocation of β-catenin (active form) to the nucleus (data not shown).

In conclusion, the studies demonstrated that GSK-3β plays a pivotal role in regulating the mitochondrial membrane permeability transition pore under conditions of cell culture in atmospheric oxygen. Additionally, these studies have shown a novel technique that can be used to monitor changes in JC-1 fluorescence while in the presence of the drug SB216763.