Molecular and Cellular Response of the Myocardium (H9C2 Cells) Towards Hypoxia and HIF-1α Inhibition

Introduction Oxidative phosphorylation is an essential feature of Animalian life. Multiple adaptations have developed to protect against hypoxia, including hypoxia-inducible-factors (HIFs). The major role of HIFs may be in protecting against oxidative stress, not the preservation of high-energy phosphates. The precise mechanism(s) of HIF protection is not completely understood. Materials and Methods To better understand the role of hypoxia-inducible-factor-1, we exposed heart/myocardium cells (H9c2) to both normoxia and hypoxia, as well as cobalt chloride (prolyl hydroxylase inhibitor), echniomycin (HIF inhibitor), A2P (anti-oxidant), and small interfering RNA to beclin-1. We measured cell viability, intracellular calcium and adenosine triphosphate, NADP/NADPH ratios, total intracellular reactive oxidative species levels, and markers of oxidative and antioxidant levels measured. Results Hypoxia (1%) leads to increased intracellular Ca2+ levels, and this response was inhibited by A2P and echinomycin (ECM). Exposure of H9c2 cells to hypoxia also led to an increase in both mRNA and protein expression for Cav 1.2 and Cav 1.3. Exposure of H9c2 cells to hypoxia led to a decrease in intracellular ATP levels and a sharp reduction in total ROS, SOD, and CAT levels. The impact of hypoxia on ROS was reversed with HIF-1 inhibition through ECM. Exposure of H9c2 cells to hypoxia led to an increase in Hif1a, VEGF and EPO protein expression, as well as a decrease in mitochondrial DNA. Both A2P and ECM attenuated this response to varying degrees. Conclusion Hypoxia leads to increased intracellular Ca2+, and inhibition of HIF-1 attenuates the increase in intracellular Ca2+ that occurs with hypoxia. HIF-1 expression leads to decreased adenosine triphosphate levels, but the role of HIF-1 on the production of reactive oxidative species remains uncertain. Anti-oxidants decrease HIF-1 expression in the setting of hypoxia and attenuate the increase in Ca2+ that occurs during hypoxia (with no effect during normoxia). Beclin-1 appears to drive autophagy in the setting of hypoxia (through ATG5) but not in normoxia. Additionally, Beclin-1 is a powerful driver of reactive oxidative species production and plays a role in ATP production. HIF-1 inhibition does not affect autophagy in the setting of hypoxia, suggesting that there are other drivers of autophagy that impact beclin-1.


INTRODUCTION
At the most fundamental level, life can be thought of as a continual, creative process that generates tiny pockets of order in a universe that is growing, in aggregate, increasingly disordered. Because the natural tendency of things is to become disordered, maintaining the structures necessary to sustain life requires the constant influx of energy. In the Animalian world with which we concern ourselves, the most efficient source of energy is mitochondrial oxidative phosphorylation (1), in which reducing equivalents (electrons) are donated to the electron transport chain (ETC) where they create a proton gradient that drives ATP synthase and are ultimately accepted by oxygen (2). The presence of oxygen is therefore essential for efficient energy production, and deviations in this finely tuned process (both inadequate and excess oxygenation) can lead to both alterations in ETC function, loss of electrons, and production of free radicals which themselves can cause cellular damage (oxidative stress) in addition to decreasing ATP availability (3). Additionally, because the maintenance of transmembrane ion potentials and neurotransmitter reuptake are ATP-dependent processes, loss of ATP leads to excitotoxicity and ion shifts that ultimately increase intracellular calcium [Ca2+] i and lead to cell death (4,5).
Given the critical role of oxygen as an "electron sink" during oxidative phosphorylation, the ability to adapt to perturbations in oxygen availability is essential to survival. In order to protect against hypoxia, multiple adaptive mechanisms have developed over time, most notably through the hypoxia-inducible factors (HIF) but also through NF-κβ, the unfolded protein response, and mammalian target of rapamycin (mTOR) kinase (6)(7)(8). During normoxia, the alpha subunits of HIF proteins (there are three) are continuously degraded by prolyl hydroxylases (PHDs) (3,9,10). When intracellular oxygen drops below 10 mm Hg (11), PHDs become ineffective, allowing HIFs to accumulate and activate approximately 1,500 target genes [e.g., vascular endothelial growth factor (VEGF), erythropoietin (EPO), inducible nitric oxide synthase (iNOS), and lactate dehydrogenase A (2).] by binding to a hypoxia response element (HRE) (3).
Zhang et al. significantly increased our understanding of HIF-1α. Using mouse embryonic fibroblasts as a model, they found that knockout cell (KO) lines without HIF-1α consumed more oxygen and had higher ATP levels when compared to wildtype (WT) cells during both normoxia and hypoxia (1). Most interestingly, while WT cells exposed to hypoxia produced less reactive oxidative species (ROS) than normoxic WT cells, in KO cells exposure to hypoxia lead to an increase in ROS (1), suggesting that a major role of HIF-1α is in the attenuation of ROS production. Zhang et al. also demonstrated that hypoxia leads to mitochondrial autophagy and that this process is HIF-1 dependent and works through the BNIP3-Beclin-1-Atg5 pathway (1). Mitochondrial autophagy was thought to be a protective mechanism that leads to less ROS production. In addition, Hypoxia causes alteration of the cell calcium dynamics, which leads to alterations in many signaling pathways (12)(13)(14). A lack of Ca2+ signaling can trigger autophagy (15) and Ca2 + has been shown to regulate autophagy along with apoptosis (16)(17)(18). While Zhang et al. have made major contributions to our understanding of adaptive responses to hypoxia (as evidenced by the 2019 Nobel prize in medicine being shared by William Kaelin, Sir Peter Ratcliffe, and Gregg Semenza for their work in this field), important questions remain. Does HIF-1α exert an influence on [Ca 2+ ] i levels? Is the role of HIF-1 in attenuating oxidative stress universal across all cell types? How do antioxidant defenses (exogenous or endogenous) impact HIF-1α expression? Is HIF-1α the only pathway that drives autophagy via the BNIP3-Beclin-1-Atg5 pathway? The purpose of this study was to answer these questions. Our primary hypothesis was that in the setting of hypoxia, HIF-1α inhibition would lead to increased intracellular calcium-mediated by reduced expression of L-type voltage-dependent Ca channels.

Cobalt Chloride and Hypoxia Mediated HIF-1α Induction
To mimic hypoxic conditions, H9c2 cells were cultured at a density of 1 × 10 6 per ml on 100 × 30-mm Petri dishes/6 well plates as described above. In the CoCl 2 experimental group, H9c2 cells were cultured in a complete culture medium supplemented with 200µM CoCl2 alone or with ECM, A2P for 6 h, at atmospheric conditions of 95% air and 5% CO2 at 37 • C in a humidified incubator (21%O2). In the hypoxic experimental group, H9c2 cells were cultured under Hypoxia (1%O2) in a complete culture medium supplemented with ECM or A2P and were maintained in a modular incubator chamber flushed with a gas mixture containing 1%O2, 5% CO2, and 94% N2 at 37 • C for 24 h. For controls, H9c2 cells were cultured in a complete culture medium under normoxia conditions. After incubation, cells were harvested immediately for analysis or preserved at -80 • C freezer for future studies. Cell culture environment, reagent  concentrations and specificity of all experimental groups are listed in Table 1.

siRNA Transfection
H9c2 cells (1 × 10 6 cells/ml) were seeded in a 6-well plate containing a complete culture medium and cultured to 70-80% confluence before transfection. For knockdown of Beclin-1/Becn1 (Coiled-Coil, Moesin-Like BCL2-Interacting Protein), 10 µl of transfection reagent (TransIT-X2 R Dynamic Delivery System # MIR6000, Mirus) was diluted in 250 µl OptiMEM serum reduced medium, to this 25 nM of Beclin-1 siRNA was added and incubated at room temperature for 30 min to form transfection reagent: siRNA complex development. The transfection reagent: siRNA complex was then added to 70-80% confluence H9c2 cells (replaced with culture medium supplemented with 1%FBS) and incubated at 37 • C with 5% CO2 in an incubator for 1 h. After 1 h incubation, culture medium FBS was increased to 10% (complete culture medium), and these pre-Beclin-1-siRNA transfected cells were used for mimicking hypoxic conditions. siRNA and transfection reagents concentrations were used according to the manufacturer's instructions.

Cell Viability Assay
Experimental groups (CoCl 2 and Hypoxia) exposed H9c2 cells viability was measured using the CellTiter-Glo R Luminescent cell viability assay kit (#G7570, Promega). Briefly, H9c2 cells were plated in 6-well plates and grown up to 70-80% confluence, then the plates were exposed to hypoxia or normoxia conditions, or Cobalt chloride as described in the methodology section. After incubation, H9c2 cells were incubated at room temperature for 15 min, and then, 100 µL of CellTiter-Glo reagent were added to cells and then placed on an orbital shaker for 2 min to induce cell lysis. Lysed cells were incubated at room temperature for 10 min to stabilize the luminescent signal and then using Synergy TM Multi-Mode Microplate Reader (BioTek, United States) luminescence was recorded. All the samples were read in duplicate. To represent the relative density of adhering cells (live cells) in culture dishes, we performed the crystal violet staining on CoCl2 and Hypoxia exposed H9c2 cells. In brief, after cells were exposed to hypoxic conditions, culture media was removed and the cells were washed with PBS (phosphate-buffered saline) and fixed for 20 min at RT with staining solution (0.05% w/v crystal violet, 1% formaldehyde, 1% methanol in 1x PBS). Fixed cells were rinsed with water and air-dried and images were obtained using a Nikon TS100 digital camera microscope.

Measurement of Intracellular Calcium (Ca2+) Levels
Intracellular calcium (Ca2+) levels in CoCl2 and Hypoxia exposed H9c2 cells were assessed using the Fluo-4 NW calcium assay kit (# F36206, Molecular Probes, Invitrogen), by following the manufacturer's protocol. Briefly, after cells were exposed to CoCl2 or hypoxia conditions, culture media was removed to avoid the source of baseline fluorescence then quickly 100-µl of Fluo-4 NW assay reagent was added and incubated for 30 min at 37 • C, then incubated for an additional 30 min in the dark at room temperature. The relative fluorescence units (RFU) emitted by the Fluo-4-NW dye were quantified using a Synergy TM Multi-Mode Microplate Reader (BioTek, United States), excitation at 494 nm, and emission at 516 nm.

Measurement of Cellular ATP Levels
Intracellular energy (ATP) levels were measured using the Bioluminescent Assay kit (MAK135, Sigma) according to the manufacturer's protocol and as previously described (26). The addition of luciferase and D-Luciferin (ATP assay buffer) to the cells allowed for the measurement of the luminescent intensity of the sample (which was proportional to the amount of ATP) using Synergy TM Multi-Mode Microplate Reader, BioTek, United States. All the samples were read in duplicate and purified ATP (Abcam, ab 83355) was used as a standard.

Measurement of Cellular NADP/NADPH Ratio
Changes in the NADP/NADPH ratio were evaluated using an NADP/NADPH -Glo assay kit (# G9081, Promega

Measurement of Cellular Hypoxia/Oxidative Stress Levels in Cells
Cellular hypoxia/oxidative stress levels were evaluated using a ROS-ID R hypoxia/oxidative stress detection kit (#ENZ-51042, Enzo Life Sciences). H9c2 cells were cultured and maintained as described above in 8-chambered cell culture slides (Falcon #08-774-208, Fisher Scientifics) and allowed to grow up to 70-80% confluence, then the cells were exposed to CoCl2 or hypoxia conditions. After cells were exposed to CoCl2 or hypoxia, culture media was removed and cells were washed gently twice with PBS then ROS-ID R hypoxia/oxidative stress detection solution was added and then cells were re-incubated under normal tissue culture conditions for 30 min. The detection solution was carefully removed, cells were washed twice with PBS, nuclei were stained with Hoechst 33342, then the slides were covered with a coverslip and a fluorescence microscope (Olympus, BX51) was used to capture the images using Green Filter for Oxidative stress and Texas Red Filter for Hypoxia detection. Data were normalized with nuclei stain (Hoechst 33342) fluorescence intensity and expressed as changes in fluorescence intensity.
mRNA Analysis: Reverse Transcription-Quantitative Polymerase Chain Reaction Total RNA was isolated from H9c2 cells in each experimental condition using a commercial total RNA isolation kit by following the manufacturer's protocol (RNeasy plus mini kit, Qiagen) and the extracted RNA was quantified using a spectrophotometer-NanoDrop (Thermo Fisher Scientific, Inc). Total RNA (2-5 µg) was reverse transcribed to cDNA using the iScript-Adv cDNA Synthesis Kit according to the manufacturer's recommendation (BioRad, United States). The cDNA (50-100 ng) was used for real-time PCR analysis in a final volume of 20 µl containing, iTaq universal SYBR R Green supermix (BioRad, United States) and specific gene primers (Supplementary Table 1). qPCR was performed using the CFX Connect Real-time PCR system (BioRad, United States). Fold changes in expression were calculated using the 2-Ct method (27). Each reaction was run in duplicate or triplicate and Hprt1, or α-Tubulin was used as a normalization control.

Autophagy Analysis
In addition to ATG5, Becn1 mRNA, and protein analysis, we also used Cyto-ID Autophagy detection assay (#ENZ-KIT175, Enzo Life Sciences, United States) to detect autophagy in CoCl2 or hypoxia treated H9c2 cells. Briefly, H9c2 cells were cultured in 8-chambered cell culture slides and exposed to CoCl2 or hypoxia conditions. After cells were exposed to CoCl2 or hypoxia, cells were incubated with fluorescent Cyto-ID dye at 37 • C for 30 min followed by nuclei stain-Hoechst 33342 for 5 min. After incubation, representative images were acquired using a fluorescence microscope (Olympus, BX51).

Mitochondrial DNA Quantification
Mitochondrial DNA (mtDNA) levels of the H9C2 cells were quantified by following previously reported methods with slight modification (26). H9C2 cell genomic DNA from each experimental condition was extracted using DNeasy Blood & Tissue kit (#69504, Qiagen). To estimate the amount of mtDNA relative to nuclear DNA (nDNA), we used two sets of primers encoding a mitochondrial gene (Mt_Nd1 and Mt_Nd6) and a nuclear gene Tubulin (Tuba1) (30). 100 ng of H9C2 cells DNA and iTaq universal SYBR R Green supermix (BioRad) were used for real-time PCR analysis. Gene expression (Fold changes) was calculated using the 2-Ct method (27), each reaction was run in duplicate or triplicate, and α-Tubulin was used as a normalization control.

HIF-1α and Intracellular Ca 2+
Exposure of H9c2 cells to hypoxia led to an increase in intracellular Ca 2+ levels (Figure 1). Both A2P (antioxidant) and ECM (HIF-1 inhibitor) prevented this response. Interestingly, in cells exposed to siRNA inhibitor for beclin-1, exposure to hypoxia lead to a reduction in Ca 2+ levels compare to controls. Exposure of H9c2 cells to hypoxia-mimetic agent cobalt chloride (CoCl 2 ) and hypoxia (1%O 2 ) did not change the Cav1.1 protein expression significantly (Figure 2). However, compare to controls Cocl2 + ECM and 1%O2 exposed h9c2 cells showed increased mRNA levels and decreased protein levels with Beclin1/becn1inhibition cells grown under hypoxia (Figure 2). Exposure of H9c2 cells to hypoxia and Cocl2 exposure significantly modifies the Cav1.2 (CACNA1C) and Cav1.3 (CACNA1D) expression at mRNA and protein levels and these responses were blocked in presence of A2P and ECM (Figures 3, 4). Hypoxia increased the Cav1.2 expression at mRNA and protein level. Cav1.2 mRNA levels are increased in Cocl2 + ECM exposed cells and decreased in Beclin-1 inhibition cells grown under hypoxia and normoxia. In addition compared to hypoxia Beclin-1 inhibition cells grown under hypoxic conditions expressed significant decrease in Cav1.2 mRNA and the protein expression levels (Figure 3).
Hypoxia increased the Cav1.3 expression both at mRNA and protein level. In addition, Beclin-1 inhibition significantly increased the Cav1.3 mRNA levels both in normoxia and in hypoxia cells, and the protein expression levels were not changed. Hypoxia cells, Cav1.3 mRNA levels were decreased in 1%O2 + A2P, 1%O2 + ECM and in the Beclin-1 inhibition cells grown under Hypoxia. Protein levels also decreased in 1%O2 + A2P, 1%O2 + ECM, and in Beclin-1 inhibition cells grown under Hypoxia (Figure 4). These data suggest that in the setting of hypoxia, HIF-1 inhibition could lead to increased intracellular calcium and that, this is mediated by reduced expression of L-type voltage dependent Ca ++ channels (Figures 1-4 and in Supplementary Figures 2A-F) in accordance with our primary hypothesis. In addition, the poor correlation between expression levels of mRNA and protein change possibly varied post-transcriptional mechanisms.
in ATP with hypoxia. Though the impact of hypoxia on total ROS was reversed with HIF-1 inhibition through ECM (1%O2 vs. 1% O2 + ECM) interestingly, in normoxia, cobalt chloride (prolyl hydroxylase inhibitor) led to reduced ATP levels, similar to those seen during hypoxia and an increase in ROS (Control vs. Cocl2). In cells exposed to beclin-1 inhibitor, ATP levels increase during hypoxia compared to hypoxia cells (no beclin-1 inhibitor). Most notably, beclin-1 inhibition led to a profound increase in total ROS levels during normoxia, which was only partially attenuated during hypoxia ( Figure 6A). In addition to the increased total ROS levels in Cocl2 and Beclin1 inhibition cells, we observed oxidative stress markers; total protein carbonyl content ( Figure 6B) and Lipid peroxidation ( Figure 6C) levels were significantly increased. The anti-oxidant marker SOD and Catalase levels were also considerably decreased in Hypoxia and in Cocl2 exposed H9C2 cells (Figures 6D,E). We observed that both A2P (anti-oxidant, oxidation-resistant derivative of ascorbic acid) and ECM (HIF-1α inhibitor) attenuated the hypoxia (1%O2) and hypoxiamimetic agent (Cocl2) disturbed cells Oxidative and anti-oxidant responses to varying degrees (Figure 6).
In addition, we observed that Hypoxia leads to decreased NADP/NADPH levels ( Figure 5B) which were not affected by A2P but appeared to be amplified with ECM. Exposure of cells to both CoCl 2 and hypoxia leads to an increase in nitroreductase activity (hypoxia detector) as well as reactive species (Supplementary Figure 3). During exposure to CoCl 2 , the addition of the antioxidant A2P and HIF-1α inhibitor ECM leads to a reduction in nitroreductase activity as well as ROS. In cells exposed to hypoxia, A2P and ECM also lead to a reduction in nitroreductase activity but no influence on ROS levels.

Impact of Anti-oxidants and HIF-1α Inhibitors on HIF-1α Expression
Exposure of H9c2 cells to hypoxia led to an increase in HIF-1α mRNA ( Figure 7A) and protein expression (Figure 7B). Compared to hypoxia, in cells exposed to hypoxia + A2P lead to a reduction in HIF-1α expression both at mRNA, and at protein. In hypoxia (compared to no HIF-1 inhibition cells) HIF-1 inhibition through ECM significantly downregulates the H9c2 cell's Hif-1α expression both at mRNA, and protein levels (Figures 7B,C). These data suggest that both A2P (anti-oxidant), and ECM (HIF inhibitor), inhibit the hypoxia-induced HIF-1α expression in myocardial (H9c2) cells. In addition compared to hypoxia, beclin-1 inhibition cells grown under hypoxic conditions also expressed significant decrease in HIF-1α mRNA and the protein expression levels (Figures 7A-C).

HIF-1α and the Beclin-1 Pathway
Exposure of H9c2 cells to hypoxia led to an increase in both ATG5 mRNA ( Figure 8A) and protein expression ( Figure 8B). Both A2P and ECM attenuated this response to varying degrees at mRNA and no change was observed in protein expression of hypoxia (1%O2 + A2P) group, this could be due to post-transcriptional mechanisms. In cells exposed to siRNA inhibitor for beclin-1, ATG5 mRNA and protein levels were increased compared to controls in normoxia (21%O2). Compared to hypoxia cells ATG5 mRNA, Beclin-1 inhibition cells grown under hypoxia expressed significantly less ATG5 mRNA and protein levels (Figures 8A,B). In addition to ATG5 levels, the Cyto-ID Autophagy detection marker staining shown in Supplementary Figure 2G, visually (in cells) demonstrates the impact of CoCl2, hypoxia, A2P, and ECM on autophagy (Mean Fluorescence intensity levels were shown in Supplementary Figure 2H).

H9c2 cells grown under hypoxia expressed decreased intracellular Calcium ([Ca 2+ ]i) levels
the ATP levels in H9c2 cells (Figure 5A), and increased the oxidative stress (ROS) levels under normoxia and noticeably alter the H9c2 cells oxidative and antioxidant levels (Figure 6). NADP/NADPH ratio decreased in hypoxia, and Beclin-1 silenced H9c2 cells (Figure 5B). HIF1α and downstream targets VEGF and EPO protein expression levels (Figures 7D,E) and Hypoxia/Oxidative stress ratio levels (Supplementary Figure 3) are decreased in cells cultured with A2P (antioxidant). Hypoxia, Cocl2 exposure, and Beclin1 inhibition increased the autophagy protein ATG5 expression levels (Figure 8) and decreased the Becn1 expression (Figure 9) and mitochondrial DNA levels of H9c2 cells (Figure 10). The major molecular and cellular changes in Myocardium (H9c2) cells in response to chemical (CoCl2) and low oxygen-induced hypoxia were summarized in Figure 11.  (31,32) and PC12 (33) cells exposed to hypoxia as well as the intact rat brain (34). Li et al. also found that HIF-1α inhibition led to a decrease in the Cav1.2 gene (which encodes for an L-type voltage-dependent Ca 2+ channel) which is normally known to be increased during hypoxia (32). They did not report the [Ca 2+ ] i levels in the HIF-inhibited cells (24). Presumably in the setting of low oxygen, increased [Ca 2+ ] i is maladaptive, and understanding whether or not HIF-1α plays a role in Ca 2+ regulation is important. Similar to what other authors have reported, we found that hypoxia exposure (1%O 2 ) leads to a significant increase in [Ca 2+ ] i (Figure 1) and decreased cell viability (p < 0.019, Supplementary Figures 1A,B). Like Li et al., we found that inhibition of HIF-1α with ECM, attenuated Cav 1.2 (and Cav 1.3) expression, but we also observed a reduction in [Ca 2+ ] i with hypoxia + ECM. Given the clear adverse impact of excessive [Ca 2+ ] i on cell survival and the general prosurvival response of HIF-1α, this result is inquisitive. While not statistically significant, there was also a trend toward increased [Ca 2+ ] i in normoxic cells exposed to the PHD inhibitor, CoCl 2 . Additionally, normoxic cells exposed to the HIF-1α inhibitor ECM also produced a significant decrease in [Ca 2+ ] i . Taken together, these results suggest that HIF-1α leads to increased [Ca 2+ ] i , although there must be other mechanisms involved as the CoCl 2 group exhibited a similar increase in HIF-1α as the hypoxic group but did not yield an appreciable increase in

HIF-1α and Intracellular Ca 2+
Interestingly, while uncontrolled [Ca 2+ ] i accumulation is thought to represent a "final common pathway" for hypoxic cell death (35), there is some evidence that moderate increases in [Ca 2+ ] i are protective, at least in hypoxia-tolerant organisms (36). Furthermore, increased [Ca 2+ ] i may play an essential role in both hypoxic and anesthetic preconditioning (37). Though L-type Ca2+ channels are more sensitive to hypoxia, changes in cellular Ca2 + levels will also be associated with the levels of other types of Ca2+ channels, such as the T-type Ca2+ channel (Cav3.1-3) (38). Sarcoplasmic/endoplasmic reticulum calcium ATPases (SERCA), Phospholamban (PLN) (39,40), sodiumcalcium exchange, plasma membrane Ca2 + ATPase (41,42), will likely affect the Ca2 + levels along with extrinsic and intrinsic stimuli (43). More work is needed to better understand the role of HIF-1α in modulating the intracellular concentration of this important cation, especially in intact animals.

HIF-1α and the Balance Between ATP and Reactive Oxidative Stress
The relationship between O 2 availability, ROS/RNS, oxidative stress, and high energy phosphates (ATP) is unclear. Zhang et al. exposed mouse embryonic fibroblast (MEFs) to hypoxia (1%O 2 ) and found that hypoxic MEFs deficient in HIF-1α exhibited higher levels of O 2 consumption, ATP, and ROS than wild-type (WT) cells exposed to 20% O 2 (1). It is worthy of notice how the cells in a low oxygen environment would have produced more ATP via oxidative phosphorylation than normoxic cells? The key to understanding this lies in the relationship between O 2 and oxidative stress. Acute exposure to hypoxia increases ROS production, primarily through complex III (44)(45)(46). A fundamental role of HIF-1α may therefore be to serve as an "energy brake break", protecting cells from oxidative stress by actively shutting down oxidative phosphorylation (1).
We also observed that the specific Oxidative stress marker; total protein carbonyl content and Lipid peroxidation levels were markedly increased both in Hypoxia and in Cocl2 exposed cells compared to controls. Antioxidant (A2P) and HIF-1a inhibition (ECM) cells showed low Lipid peroxidation levels compared to Hypoxia and in Cocl2 exposed cells. Total Protein carbonyl (PC) content was also reduced with A2P treatment in Cocl2 exposed cells (Cocl2 + A2P), but these PC levels were not significantly changed in hypoxia + A2P and hypoxia + ECM exposed cells. The decreased anti-oxidant markers SOD and Catalase levels were significantly increased in A2P treated myocardium cells compared to Hypoxia and Cocl2 exposed cells. While, Hif1 inhibition (1%O2 + ECM) increased the cells total SOD levels compared to Hypoxia cells; however, Catalase levels were not significantly increased to control levels with HIF-1a inhibition (1%O2 + ECM). This difference can likely be attributed to differences in technique (knockout versus inhibition with ECM).
With no interference on HIF-1α, hypoxia does lead to an overall decrease in ROS despite increasing superoxide production at Complex III (45). However, despite the clear increase in ROS levels when HIF-1α activity is inhibited by ECM in the setting of hypoxia (Figure 6), we found inconsistent changes in oxidative stress. Interestingly, our work suggests that ECM may impact HIF-1α expression itself. ECM is known to bind to the hypoxia response element and this is the primary mechanism by which it interacts with HIF-1α in cancer cell lines (47). However, some authors have demonstrated that ECM can lead to an increase in HIF-1α mRNA and protein expression during normoxic conditions (47). In H9c2 cells, we observed a decreased HIF-1α mRNA and protein expression in ECM exposed cells in the presence of hypoxia. This difference may be because the majority of research on ECM/HIF-1 has been focused on cancer cell lines [MCF-7 (human mammary tumor), U251 (human glioma), and HepG2 (hepatocellular carcinoma) cells] (47), which may behave differently than non-cancerous cardiomyocytes.
We found that the addition of the anti-oxidant A2P to normoxic H9c2 cells exposed to CoCl 2 resulted in decreased ROS and HIF-1α expression. Similarly, the addition of A2P to hypoxic H9c2 cells also resulted in decreased HIF-1α expression, with no detectable change in ROS levels. Part of this discrepancy may be the relatively low ROS levels detected in hypoxia, as well as different sources of ROS -CoCl 2 leads to peroxide overproduction, whereas hypoxia leads to superoxide production (45). Interestingly, the addition of A2P to H9c2 cells did not affect [Ca 2+ ] i in the setting of normoxia (with CoCl 2 ) but significantly reduced [Ca 2+ ] i in the setting of hypoxia, which might be protective. Given the role of [Ca 2+ ] i in cell death as well as the possible role of moderate increases in [Ca 2+ ] i on preconditioning, further investigation seems warranted. We are not aware of any reports of anti-oxidants impacting [Ca 2+ ] i in the setting of hypoxia, although there is evidence that antioxidants improve skeletal function during exposure to hypoxia (57).

HIF-1α and the Beclin-1/Autophagy Pathway
In the setting of hypoxia, we found that beclin-1 inhibition led to decreased ATG5 mRNA and protein expression, and we observed a significant decrease in autophagy, consistent with Zhang et al. Surprisingly, in the setting of normoxia, beclin-1 inhibition led to increased ATG5 mRNA and protein expression, and we observed a significant increase in autophagy. To determine whether mitochondrial autophagy does play a role in controlling ROS production, we measured total ROS levels in the setting of normoxia and hypoxia without and with beclin-1 inhibition and found that in both scenarios, beclin-1 inhibition leads to a substantial increase in total ROS levels. This strongly suggests that beclin-1 plays a prominent role in ROS attenuation independent of its role in mitochondrial autophagy. Most importantly, we found that while hypoxia leads to an increase in autophagy, which can be attenuated with the addition of antioxidants, HIF-1 inhibition does not affect this process at all despite the fact that beclin-1 reduces ATG5 mRNA and protein expression, suggesting that other non-HIF-1 factors drive the Beclin-1/autophagy pathway.
Placing our study in the context of previously published work, our results suggest that HIF-1α exerts several effects in H2c9 cells exposed to hypoxia, including upregulation of both Cav 1.2 and 1.3, which results in an increase in intracellular Ca ++ . Increased ATG5 and autophagy through a beclin-1 independent mechanism and reduced ATP and ROS production likely through a general slowdown in oxidative phosphorylation (Figure 12).
Echinomycin (ECM) appears to have a direct impact on HIF-1α expression in H2c9 cells and is not limited to its interaction with the hypoxia response element (HRE).

Limitations
This study has several limitations. First, it was a study of cell lines, which means that the observed results are not directly applicable to intact organisms. Second, we used only one cell type (H9c2), which we selected because of the importance of the myocardium in long-term clinical outcomes in humans exposed to stress. Third, we only tested one anti-oxidant agent and one HIF-1α inhibitor. Though, we used pre-designed specificityenhanced SignalSilence R Beclin-1 siRNA II (reduces off-target effects), not using a separate control siRNA is possibly a limitation. The data are shown in Supplementary Figure 2

Further Directions
Our results pose several questions that need answers through further work -how are antioxidants (e.g., A2P) able to exert an impact on HIF-1α protein expression without affecting detectable ROS levels? How ATG5 is upregulated despite the fact that beclin-1 is not upregulated in H9c2 cells exposed to hypoxia? How does beclin-1 mediate ROS production independent of autophagy? Involvement of PINK1/Parkin pathway in hypoxiainduced autophagy, studies with specific autophagy inhibitors (e.g., bafilomycin, chloroquine, etc.), in H9c2 cells. In addition, are moderate intracellular calcium increases during hypoxia exposure truly mediated by HIF-1α and upregulation of Cav 1.2 and if so, does this protect against cell death? What is the role of SERCA, PLN, Na + /Ca2 + exchangers, and plasma membrane Ca2 + ATPase on H9c2 cell's intracellular Ca2 + levels in Hypoxia?

CONCLUSION
Though additional confirmatory studies are needed, in the setting of hypoxia, HIF-1α upregulates both Cav 1.2 and 1.3 expression that results in an increase in intracellular Ca ++ , increases ATG5 and autophagy through a beclin-1 independent mechanism, and reduces ATP and ROS production in H2c9 cells.

DATA AVAILABILITY STATEMENT
The original contributions presented in this study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

AUTHOR CONTRIBUTIONS
HO: study design, the performance of the study, data analysis and interpretation, and manuscript writing and preparation. ML: data analysis and interpretation and manuscript writing. RT: design the study, data analysis and interpretation, and manuscript writing and preparation. All authors contributed to the article and approved the submitted version.

FUNDING
This study was funded by the Department of Anesthesiology at the University of Virginia.

ACKNOWLEDGMENTS
We wish to thank Navya Atluri, Department of Anesthesiology, University of Virginia, Charlottesville, VA, for her assistance with the study.