In this study, we report for the first time that MPP+-induced LMP is mediated by release of zinc from mitochondria. Many studies have demonstrated mediation of zinc in ROS neurotoxicity. In this study, we reported that MT-3 is the causative protein in H2O2 toxicity by release of zinc, however, it plays a protective role in MPP+ neurotoxicity by reduction of [Zn2+]i level. By demonstrating that zinc-overload or H2O2-induced cell death was increased in Rho 0 cells, where lysosomes, as well as mitochondria, were maintained at low levels, the importance of intracellular organelles such as mitochondria and lysosomes in regulation of zinc homeostasis was confirmed. These results provided meaningful information for use in development of therapeutics for PD and other diseases of the brain involving the contribution of zinc to neuronal death.
Based on our results, we have proposed a novel pathogenic mechanism leading to development of PD-related neurodegeneration. Mitochondrial dysfunction induced by MPP+ resulted in a rapid increase of ROS and the intracellular levels of labile zinc, as shown in the diagram (Fig. 8). An increase in the levels of ROS and zinc, in turn, led to the abnormal permeabilization of lysosomal membranes, resulting in lysosomal deficiency and neuronal death. This MPP+-induced pathogenic pathway was almost completely blocked in mitochondria-deficient Rho 0 CHO cells (Fig. 7, 8), therefore, mitochondrial dysfunction may be a primary cause of neurodegeneration in MPP+-treated cerebrocortical cultures. As a result of tracking the time during which there was a significant increase in ROS and zinc with the loss of MMP and development of LMP after treatment with MPP+, a decrease in the JC-1 and LysoTracker signals was observed from 4 h, a significant increase of Zinpyr-1 was observed after 10 h, and a difference in the DCF signal was observed after 12 h. However, the time points at which a significant change was observed could not be regarded as the order in which the change occurred within the cell. Because the range within which a meaningful difference is measured depends on the sensitivity of the fluorescence dye, we cannot say that the change is detected earlier than it is necessarily happening first. Antioxidants induced a decrease of DCF fluorescence (ROS) as well as an increase in the intracellular level of zinc and cytoplasmic release of cathepsin B, however the loss of MMP was not affected. That is, the mechanism of dissipated MMP occurs first and is not affected by antioxidants, and the remaining phenomena are mechanisms caused by generation of ROS. Because we compared the changes after treatment with inhibitor, we report that the change of ROS induces the release of zinc, which can cause LMP and result in lysosomal deficiency and neuronal cell death.
We previously demonstrated that rapid induction of NADPH oxidase activity occurs in zinc-induced neurotoxicity in a PKC-dependent manner [53–56]. Increasing the intracellular level of zinc can promote production of ROS through regulation of PKC and NADPH oxidase activity. In this study, we confirmed that ROS production could be improved by zinc. Administration of TPEN, a zinc chelator, with MPP+ results in a reduction of the DCF fluorescence signal by 50% (Fig. 2E), meaning that zinc additionally induces production of ROS. On the other hand, administration of Trolox with MPP+ resulted in complete suppression of the increase of [Zn2+]i level (Fig. 2D), therefore, the idea that generation of ROS precedes the release of zinc can be considered.
MPP+ is known to inhibit mitochondrial oxidative phosphorylation, therefore, MPP+-induced generation of ROS is easily understood; however, it is unclear how MPP+ can induce an increase in the level of cytosolic free zinc. There are two possible pathways for increasing the level of cytoplasmic zinc. First, because MPP+ induces the loss of MMP, leakage of zinc in the mitochondria into the cytoplasm may occur. The second possibility is that ROS induced the release of zinc bound to the proteins. When the expression of MT-3, a representative zinc-binding protein, was knocked out, a significant increase of cell death induced by MPP+ was observed, thus it is thought that the main source of [Zn2+]i was mitochondria rather than the release from the proteins induced by ROS.
As previously reported [41], exposure to H2O2 within 20 min resulted in an increase of the level of labile zinc in lysosomes as visualized using FluoZin-3 fluorescence. At ~ 40 min after treatment with H2O2, the fluorescence zinc signal was lost from some zinc-containing vesicles, indicating LMP. Koh et al. also reported that an increase in the level of free zinc and LMP after treatment with H2O2 was only induced in wild-type astrocyte cultures but not in astrocytes cultured from the brains of MT-3-null mice, suggesting that MT-3 may be a potential source of zinc after H2O2 treatment [36]. In this study, we attempted to determine that MT-3 was the source of zinc after exposure to MPP+. We also observed conspicuous attenuation of H2O2-induced cell death in MT-3 KO astrocyte cultures, however, markedly increased MPP+-induced cell death was observed in astrocyte cultures lacking MT-3 (Fig. 5). Based on these results, we concluded that MT-3 did not act as a source of zinc in MPP+-induced cell death but rather bound to labile zinc, which was increased after treatment with MPP+, leading to regulation of the homeostasis of cytosolic zinc and suppression of cell death. Several studies have demonstrated the association of MT with neurotoxicity caused by MPTP. Dhannasekaran et al. demonstrated that the level of MT expression was reduced in the substantia nigra (SN) following injection of MPTP into mice, suggesting a possible association of dopaminergic neuronal death in SN with the level of MT [57]. Rojas and colleagues also showed that MPTP reduced the concentration of MT-1 and MT-2 in the striatum but not in other brain regions [58]. These results suggested that MT level may be associated with MPP+ neurotoxicity.
Because MT-3 plays a protective role in MPP+-induced cell death, we next examined the question of whether mitochondria may be a source of zinc using Rho 0 cells. First, a significant reduction of MPP+-induced cell death was observed in mitochondria-deficient Rho 0 CHO cells (Fig. 7B). No increase in ROS, [Zn2+]i, or LMP was observed in Rho 0 CHO cells after treatment with MPP+ (Fig. 7). In addition, we found that the basal level of ROS indicated by DCF fluorescence was reduced by ~ 50% in Rho 0 Cells compared to normal cells (Fig. 7C). This result was expected considering that mitochondria are the representative organelles involved in production of ROS in cells.
Of particular interest, when LysoTracker fluorescence representing organelles with low pH was detected, we found that the number of lysosome organelles in Rho 0 cells was also maintained at a level of ~ 50% compared to normal cells (Fig. 7E). Thus, while Rho 0 cells contain significantly fewer lysosomes that are capable of trapping labile zinc, zinc-mediated LMP and cell death were not observed in Rho 0 CHO cells. The reason is that generation of ROS, release of zinc, and induction of LMP are initiated by MPP+-induced mitochondrial dysfunction, thus these phenomena are not observed in mitochondria-deficient cells.
On the other hand, homeostasis of intracellular zinc is maintained at a high level in Rho 0 cells lacking mitochondria and lysosomes. When ZinPyr-1 fluorescence staining was performed for comparison of normal cells and Rho 0 cells, the fluorescence signal was two times greater in Rho 0 cells (Fig. 7D). Thus, a high basal level of labile zinc in the cytosol is maintained in Rho 0 cells, however, the level of free zinc is not increased after treatment with MPP+ (Fig. 7D), which confirms that the increase in the level of zinc induced by MPP+ is derived from mitochondria.
In addition, increased zinc-overload neurotoxicity following exposure to a high concentration of ZnCl2 was observed in Rho 0 CHO cells (Fig. 7F). In the case of cell death caused by influx of zinc from the outside, zinc homeostasis is regulated by mitochondria and lysosomes by uptake of excess zinc and suppression of zinc-induced cell death. Augmentation of cell death was also detected in Rho 0 CHO cells after treatment with H2O2 (Fig. 7F), confirming that deficiency of mitochondria may represent the reduction of zinc buffering capacity. Based on this result, we strongly suggest that intracellular organelles, including mitochondria and lysosomes, contribute to the maintenance of zinc homeostasis.
MPTP/MPP+-induced neuronal death has been shown to exhibit the typical apoptotic pathway [59]. Upregulation of Bax, JNK, and caspases, as well as downregulation of Bcl-2, Akt, ERK1/2, and GSK3β have been implicated in neurotoxic effects resulting from in vivo administration of MPTP in mice and in vitro treatment with MPP+ in different neuronal cell lines such as MN9D cells, SH-SY5Y cells, cerebellar granule neurons, cortical neurons, and midbrain dopaminergic neurons [60–62]. Increased levels of neuroinflammatory cytokines such as TNF-α and IFN-γ have also been reported. However, caspase-dependent apoptosis was not observed in the current cortical cultures (data not shown). This discrepancy may be due to the different cell types and concentration ranges of MPP+. Treatment with 300 µM MPP+ resulted in 80% neuronal death at 22 h, however, treatment with 10 ~ 50 µM MPP+ was usually administered in other groups showing cortical neuronal apoptosis. Miyara et al. reported on mechanistic differences between mild and acute MPP+ toxicity in SH-SY5Y cells. Mild exposure to MPP+ (10 ~ 200 µM for 48 h) induced a more slowly developing cell death with inhibited degradation of autophagosomes, whereas acute exposure to MPP+ (2.5 ~ 5 mM for 24 h) inhibited both degradation of autophagosomes and basal autophagy by irreversible lysosomal damage [63]. Therefore, development of mild and acute MPP+ models will be required in order to compare caspase relevance and study the role of zinc in autophagic flux.
According to increasing evidence, lysosomal depletion may lead to the pathogenesis of PD. An increase in the number of undegraded autophagosomes, a decrease in the number of lysosomes, and down-regulation of lysosomal-associated proteins, such as lysosomal-associated membrane protein-1 (LAMP-1), LAMP-2, and cathepsins, in dopaminergic neurons of postmortem brains from PD patients have been reported [64]. Treatment with Rapamycin and overexpression of TFEB, a representative transcription factor for lysosomal biogenesis, resulted in increased cell survival through lysosomal biogenesis in MPP+ treated dopaminergic neuroblastoma cells or MPTP-treated mice [65, 66]. Dehay et al. demonstrated that lysosomal depletion was secondary to LMP induced by increased levels of mitochondrial-derived ROS in MPP+-treated cells [43]. In this study, zinc was added as a critical mediator between mitochondrial damage and LMP. Although direct permeabilization of lysosomal membranes, leading to LMP, by pro-apoptotic protein BAX (BCL2-associated X protein) in MPP+-treated cells has been reported [67], the findings of our study demonstrated that zinc released from mitochondria induces LMP and resultant lysosomal depletion. Therefore, conduct of future studies will be required in order to examine the question of whether translocation and internalization of BAX into lysosomes are related to zinc.
The findings of the current study demonstrated the role of zinc as a link between mitochondrial dysfunction and lysosomal damage. MPP+, a mitochondrial toxin, induced loss of MMP and generation of ROS, which led to the release of labile zinc from mitochondria into the cytosol. The increase in the level of cytosolic zinc resulted in translocation of labile zinc into lysosomes, leading to LMP and resultant leakage of lysosomal proteases. This pathogenic mechanism involved direct induction of neuronal death as well as the steady loss of lysosomes, which can ultimately be linked to other pathological phenomena of PD, such as accumulation of a-synuclein and formation of Lewy bodies. We also reported that mitochondrial deficiency induced a shift in the balance of the number of lysosomes and zinc homeostasis, where cells were vulnerable to ROS- or zinc-mediated cell death. Understanding the role of zinc in neuronal death and lysosomal depletion and the development of a modulator for maintenance of zinc homeostasis is expected to provide a novel potential neuroprotective strategy in management of PD.