Tanshinone IIA and Cryptotanshinone Prevent Mitochondrial Dysfunction in Hypoxia-Induced H9c2 Cells: Association to Mitochondrial ROS, Intracellular Nitric Oxide, and Calcium Levels

The protective actions of tanshinones on hypoxia-induced cell damages have been reported, although the mechanisms have not been fully elucidated. Given the importance of nitric oxide (NO) and reactive oxygen species (ROS) in regulation of cell functions, the present study investigated the effects of two major tanshinones, Tanshinone IIA (TIIA) and cryptotanshinone (CT), on hypoxia-induced myocardial cell injury and its relationships with intracellular NO and ROS, calcium, and ATP levels in H9c2 cells. Chronic hypoxia significantly reduced cell viability which accompanied with LDH release, increase in mitochondrial ROS, intracellular NO and calcium levels, decrease in superoxide dismutase (SOD) activity, and cellular ATP contents. TIIA and CT significantly prevented cell injury by increasing cell viability and decreasing LDH release. The protective effects of tanshinones were associated with reduced mitochondrial superoxide production and enhanced mitochondrial SOD activity. Tanshinones significantly reduced intracellular NO and Ca2+ levels. ATP levels were also restored by TIIA. These findings suggest that the cytoprotective actions of tanshinones may involve regulation of intracellular NO, Ca2+, ATP productions, mitochondrial superoxide production, and SOD activity, which contribute to their actions against hypoxia injuries.


Introduction
It has been established that chronic hypoxia is associated with cardiac dysfunctions in certain pathological conditions such as ischemia reperfusion, myocardial infarction (MI), and hypertrophy [1]. Hypoxia causes changes of various cellular mechanisms related to mitochondrial dysfunction and oxidative stress [2]. Among these, hypoxia-induced changes of ROS and NO productions, intracellular calcium, and ATP levels may have particular importance, given the role of these molecules in regulation of cell functions in general [3]. For example, a recent study shows that hypoxia-increased mitochondrial superoxide anion (O 2 •− ), not cytosolic O 2 •− , plays an important role in hypoxia-induced cell apoptosis [4,5]. Studies have also found that excess NO production by hypoxia can result in mitochondrial ROS increase by inhibiting mitochondrial electron transport chain function, which in turn promotes peroxynitrite formation and cell apoptosis [6,7]. On the other hand, hypoxia may modulate NO production by regulating intracellular calcium which is important for Ca 2+ /calmodulin-dependent eNOS and nNOS activity, and NO increase in turn may inhibit mitochondrial complex IV [8]. This indicates an interaction among NO, ROS, intracellular calcium, and regulation of ATP synthesis in mitochondria. Understanding the relationship of these factors may help to interpret the mechanisms of cellular injury in hypoxia condition [9,10].
Tanshinones are a group of bioactive compounds isolated from Salvia miltiorrhiza (Danshen), a traditionally medicinal plant used in management of angina pectoris, atherosclerosis, 2 Evidence-Based Complementary and Alternative Medicine and MI [11]. Among these, tanshinone IIA (TIIA) and cryptotanshinone (CT) are two major bioactive tanshinones [12]. They have been reported to have actions against oxidative stress, myocardial infarction, and myocardial ischemia reperfusion injury [13]. For example, studies in vitro have revealed antioxidant actions of TIIA by attenuating intracellular ROS level and enhancing antioxidant enzymes activity [14,15]. TIIA and CT have also been shown to influence vasodilation by regulating NO and intracellular Ca 2+ levels in endothelial cells [16,17]. However, the actions of TIIA and CT on ROS and NO pathways under hypoxic conditions are still not clear. Thus, the present study was conducted to investigate the effects of TIIA and CT on hypoxia-induced cardiac injury and their regulations of intracellular NO, ROS, calcium levels, and ATP contents in H9c2 cells.

MTT Assay.
Cell viability was determined by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay as described previously with a modification [19]. The cells (1 × 10 4 cells/well) were seed in 96 wells. At the end of hypoxia period, MTT solution was added into plates at a final concentration of 0.5 mg/mL and incubated for 2 hr at 37 ∘ C. Then, the culture medium was discarded and 150 L DMSO was added to each well to dissolve dark blue formazan crystals. The absorbance was read at 570 nm using POLARstar OPTIMA microplate reader (BMG LabTech).

LDH Release Measurement.
LDH release was determined by CytoTox 96 NonRadioactive Cytotoxicity Assay kit according to the manufacturer's instructions (Promega). After 8 hr hypoxia, the supernatant was collected and placed in 96 wells and 50 L of reconstitute substrate mixture was added in each well. After 30 mins incubation, 50 L of stop solution was added and absorbance was measured at 490 nm using Flexstation multiplate reader (Molecular Devices).

Cellular ATP Content
Measurement. Cellular ATP content was measured by ENLITEN ATP Assay System Bioluminescence Detection Kit according to the manufacturer's instructions (Promega). After hypoxia, The cells were washed with PBS and lysated, and supernatants were collected. Proteins (10 g/20 L) were added in white optiplate and initiated action by adding reconstituted reagent. Then, luminescence was measured in POLARstar OPTIMA microplate reader (BMG LabTech).
2.6. NADPH Oxidase Activity. NADPH oxidase activity was measured by lucigenin chemiluminescence as described previously with minor modification [20]. After 8 hr hypoxia, the cells were collected and centrifuged at 750 g for 10 mins at 4 ∘ C. The supernatant was discarded and the pellet was resuspended in lysis buffer (50 mM KH 2 PO 4 , pH 7.0, 1 mM EGTA, 10 g/mL aprotinin, 0.5 g/mL leupeptin, 1 g/mL pepstatin, and 0.5 mM PMSF). Next, the cells were homogenized by quick freeze and thaw step. 10 g of proteins were added in optiwhite 96-well plates with the 100 L assay buffer (50 mM KH 2 PO 4 (pH 7.0), 150 mM Sucrose, 100 M NADPH, and 1 mM EGTA). Then, the reaction was started by 5 M lucigenin. 5 M DPI was added as an inhibitor. Chemiluminescence was measured with POLARstar OPTIMA microplate reader (BMG LabTech).

Intracellular and Mitochondrial Superoxide Production.
Intracellular and mitochondrial superoxide production was measured by loading cells with 20 M dihydroethidium (DHE) and 2 M MitoSOX, respectively, by following method described previously with minor modification [21].     1 × 10 5 cells were transferred to 96-well plates, and then Fura-2AM fluorescence was obtained by alternate excitation at 340 and 380 nm and the emission was detected at 510 nm. The fluorescence maximum was determined by lysing cells with 0.2% Triton X-100 and fluorescence minimum was obtained by recording fluorescence following addition of 40 mM EDTA. The calcium concentration was calculated by equation according to what previously described [24].

Statistical Analysis.
Results were expressed as means ± SEM. Statistical differences among groups were analysed by one-way analysis of variance (ANOVA) using GraphPad Prism Software version 5.0. P ≤ 0.05 was considered significant.

Effects of TIIA and CT on Hypoxia-Induced Cell Injury.
Cells exposed to a 8 hr hypoxia exhibited a significant decrease in cell viability ( < 0.001), measured by MTT assay, which was significantly inhibited by pretreatment of TIIA and CT (3 M) (P < 0.01) (Figure 1(a)). 8 hr hypoxia also significantly increased LDH release (to 220.0%, P < 0.001), which was significantly inhibited by 3 M TIIA and CT (P < 0.001). The value of LDH release in TIIA-treated group was  significantly less than that of LDH release in CT group (P < 0.05) (Figure 1(b)). After 8 hr hypoxia, a significant reduction of the cellular ATP contents (by 20.9%, < 0.05) was observed compared to normoxia control, which was restored by pretreatment with TIIA. The cellular ATP contents in TIIA-treated group were significantly higher ( < 0.05) than those in CT group (Figure 1(c)).

Effects of TIIA and CT on Hypoxia-Induced Increase in
Mitochondrial Superoxide Production. Figure 4(a) illustrates the fluorescence images of cells stained with MitoSOX. When compared with normoxia group, distinct intensification in fluorescence was observed in hypoxia group. The cells exposed to hypoxia significantly increased MitoSOX fluorescence intensity to 154.3 ± 19.9 arbitrary units (a.u) while normoxia showed 88.2 ± 5.3 a.u. 1 hr pretreatment with 10 mM TEMPOL (SOD mimic), 10 M rotenone (complex I inhibitor), and 1 mM L-NAME (NO synthase inhibitor) significantly decreased the mitochondrial superoxide production. In the presence of TIIA and CT (3 M), the mitochondrial superoxide production was significantly reduced to 111.4 ± 30.2 a.u ( < 0.01) and 122.5 ± 29.7 a.u ( < 0.05), respectively (Figure 4(b)).

3.5.
Effects of TIIA and CT on SOD Activity. Cytosolic SOD activity in hypoxia group (78.8 ± 4.1%) did not significantly ( > 0.05) change compared to normoxia control (81.3 ± 2.1%). In contrast, the cells in hypoxia showed a significant decrease in mitochondrial SOD activity by 12.6% compared to normoxia control. In the presence of TIIA and CT (3 M), the cytosolic SOD activity did not significantly change compared to hypoxia control. The decrease in mitochondrial SOD activity by hypoxia was restored when the cells were pretreated with 3 M TIIA or 3 M CT. There was no statistically significant difference in mitochondrial SOD activity between tanshinones treated groups and normoxia control ( Figures  5(a) and 5(b)).

Effects of TIIA and CT on Hypoxia-Induced Increase in
Intracellular Nitric Oxide Production. Figure 6(a) illustrates the fluorescence images of cells stained with DAF-2. When compared with normoxia control, distinct intensification in DAF-2 fluorescence was observed in hypoxia group. The quantitative values of the florescence intensity of images were presented in Figure 6(b). Pretreatment with TIIA and CT significantly decreased the intracellular NO production. There was no statistical difference between tanshinonestreated groups and normoxia control. L-NAME (1 mM, NOS inhibitor) significantly ( < 0.01) reduced the intracellular NO production compared to hypoxia control.

Discussion
The main finding of the present study is that TIIA and CT protect against chronic hypoxia-induced H9c2 cells injury by restoring cellular ATP contents, decreasing mitochondrial superoxide, intracellular NO, and calcium levels in H9c2 cells. This is consistent with previous observations that hypoxia-induced apoptosis was associated with ROS, NO, and calcium in myocardial cells [5,25]. Additionally, cellular ATP contents, NO, and calcium are closely associated in mitochondrial ROS production and this suggests that chronic hypoxia-induced cell damages are related to mitochondrial dysfunction.

Evidence-Based Complementary and Alternative Medicine 9
Myocardial hypoxia is a main cause of cardiac dysfunction due to its triggering cell injury, apoptosis, and/or necrosis [1,26]. The present study showed that the main cause of cell injury or death under the chronic hypoxia condition was associated with mitochondrial dysfunction with accompanying LDH release and cellular ATP depletion, which is consistent with previous reports [5,27]. The protective actions of tanshinones against chronic hypoxia-induced cell injury indicate that these compounds may conserve mitochondrial function. Previous studies have reported cardioprotective effects of TIIA on H 2 O 2 -induced cell injury [28] and doxorubicin-induced cell apoptosis [29] in neonatal cardiomyocytes by protecting DNA integrity mitochondrial proteins and reducing intercellular ROS production. Similarly, the antiapoptotic effect of CT has been shown previously with actions of preventing mitochondrial-dependent apoptosis in nitric oxide induced neuroblastoma cells apoptosis [30]. The increase of ATP level by TIIA may be related to its protection of mitochondrial electron transport chain (ETC) function as ATP is mostly generated by oxidative phosphorylation, a process translocating protons by complex I/III/IV and subsequently uptake of the protons by ATP synthase accompanying the synthesis of ATP, in mitochondria ETC [31]. The finding that CT was less effective than TIIA in restoring cellular ATP contents may be related to a previous observation that CT enhanced AMP-activated protein kinase (AMPK) [32], as it has been known that AMPK is associated with energy homeostasis, mitochondrial function, and cell survival [33]. It will be interesting to investigate further the effects of tanshinones on AMPK activity in chronic hypoxia condition.
Previous studies on hypoxia-induced ROS generation have shown conflicting results. This could be due to a confusion of ROS examined (cytosolic and mitochondria). It has been shown that hypoxia decreased cytosolic superoxide generation but increased mitochondrial superoxide generation [34]. Consistent with this, a significant decrease in cytosolic superoxide generation, but increase in mitochondrial superoxide generation after hypoxia, was observed in the present study. Decreased cytosolic superoxide generation may be associated with lower oxygen level during hypoxia condition and/or decreased NADPH oxidase activity [2,21]. Interestingly, the activity of cytosolic antioxidant enzyme superoxide dismutase was not significantly changed after hypoxia, indicating that this cytosolic antioxidant enzyme may not play a major role in cell injury and death pathway during chronic hypoxia. On the other hand, there was a significant increase of intracellular hydrogen peroxide/peroxynitrite production, as indicated by DCFH-DA fluorescence probe, suggesting that a mitochondrial-derived ROS component may be involved as shown by the effects of complex I and III inhibitors (rotenone and antimycin A).
The present result is in line with a previous report showing that increased NO and ONOO − generations resulted in enhanced mitochondrial superoxide generation by blocking mitochondrial electron transport chain [35]. NO can act as a physiological regulator of respiration by reversibly inhibiting cytochrome c-oxidase at the low concentrations (nanomolar). However, at higher concentrations NO can oxidize ubiquinol of ubiquinol-cytochrome c-reductase (Complex III) to increase unstable ubisemiquinone, which produces superoxide by univalent electron transfer to O 2 [36]. Additionally, exposure to higher concentrations of NO can increase peroxynitrite formation which causes an inhibition of mitochondrial respiration at multiple sites (complex I, complex II, cytochrome c oxidase, the ATP synthase, aconitase, MnSOD, and creatine kinase) [37]. This implies that restoring electron transport chain function by reducing NO production, in addition to antioxidant enzyme activity, may help to reduce mitochondrial superoxide production during chronic hypoxia condition.
Since cytosolic ROS may not play major role in hypoxiainduced cell damages, it is not surprising to observe the lack of effect of TIIA and CT in intracellular ROS and NADPH oxidase activity. The important finding in this study is that the mitochondria superoxide generation was increased by hypoxia. This increase was significantly inhibited by TIIA and CT treatments, indicating that mitochondrial ROS plays a major role in cell damage-induced hypoxia. Interestingly, NO synthase inhibitor L-NAME also significantly inhibited mitochondrial superoxide generation, which suggests that endogenous NO may regulate the ROS production in hypoxia condition. It is possible that ROS may be generated from mitochondrial nitric oxide syntheses which may be uncoupled under hypoxic condition [38]. The observation of increase in mitochondrial superoxide dismutase activity and decrease in intracellular NO level by TIIA and CT in the present study is consistent with previous studies showing actions of tanshinones on regulating NO level and SOD activity in H 2 O 2 -induced cell injury and inflammationinduced cell death in endothelial cells [16,39].
Interestingly, TIIA and CT significantly decreased intracellular NO and mitochondrial superoxide generations, but not peroxynitrite/hydrogen peroxide levels. Previous studies using the same DCF-DH probe found that TIIA significantly inhibited ROS generation induced by doxorubicin [14,15]. However, it is not clear if the ROS in those studies is peroxynitrite/hydrogen peroxide specific as no specific inhibitors were used to validate the species of ROS observed. One possible explanation is that ROS labelled with DCF-DH may mainly be peroxynitrite as specific peroxynitrite inhibitor MnTBAP markedly reduced ROS generation (about 70%) in the present study. Thus, TIIA and CT may have a capacity to direct ROS production from peroxynitrite to hydrogen peroxide, as both compounds showed no significant effects on peroxynitrite/hydrogen peroxide production; even they significantly reduced NO and superoxide productions which theoretically should reduce peroxynitrite formation. Partial supporting evidence is that TIIA and CT increased mitochondrial SOD activity, which may result in increase in hydrogen peroxide formation. Further study is required to confirm this hypothesis. The effects of tanshinones on other antioxidant enzymes such as glutathione peroxidase and catalyse were not examined in this study which also requires further investigation.
Changes in intracellular calcium level during hypoxia are important in mitochondrial functions, especially in mitochondrial membrane permeability transition pore opening [40]. Decreased intracellular ATP by hypoxia can decrease cellular pH by glycolysis activation and this elicits imbalancing in intracellular ion exchange and, subsequently, increases in intracellular calcium level [41]. This is consistent with the present finding showing that hypoxia-induced ATP depletion was accompanied with increased intracellular calcium level. The finding of inhibition of intracellular calcium by TIIA and CT is consistent with previous reports in neonatal cardiomyocytes and rat coronary artery [42,43]. The increased intracellular calcium level is likely due to ATP depletion caused by hypoxia. Thus, it is possible that TIIA may regulate intracellular calcium through affecting ATP level. However, CT reduced intracellular calcium without affecting ATP levels. This suggests that other mechanisms such as endoplasmic reticulum-related stress, which also regulate intracellular calcium production [44], may also be involved. Additionally, studies have shown that increased intracellular calcium may increase mitochondrial ROS production [45,46]. Therefore, tanshinones may have multiple targets of reducing mitochondrial ROS production.
In summary, the findings from the present study indicate that TIIA and CT protect H9c2 cells via preserving mitochondria function by reducing excess production of mitochondrial superoxide, SOD activity, intracellular NO, and calcium levels and restoring cellular ATP contents. These molecular mechanisms may be involved in the cardioprotective actions of TIIA and CT in hypoxic injuries.