An alternative renin isoform is cardioprotective by modulating mitochondrial metabolism

Abstract The renin‐angiotensin system promotes oxidative stress, apoptosis, necrosis, fibrosis, and thus heart failure. Secretory renin plays a central role in these processes, initiating the generation of angiotensins. Nevertheless, alternative renin transcripts exist, which code for a cytosolically localized renin isoform (cyto‐renin) that is cardioprotective. We tested the hypothesis that the protective effects are associated with a beneficial switch of metabolic and mitochondrial functions. To assess H9c2 cell mitochondrial parameters, we used the Seahorse XF analyser. Cardiac H9c2 cells overexpressing cyto‐renin exhibited enhanced nonmitochondrial oxygen consumption, lactate accumulation, and LDH activity, reflecting a switch to more aerobic glycolysis known as Warburg effect. Additionally, mitochondrial spare capacity and cell respiratory control ratio were enhanced, indicating an increased potential to tolerate stress conditions. Renin knockdown induced opposite effects on mitochondrial functions without influencing metabolic parameters. Thus, the protective effects of cyto‐renin are associated with an altered bioenergetic profile and an enhanced stress tolerance, which are favourable under ischaemic conditions. Therefore, cyto‐renin is a promising new target for the prevention of ischaemia‐induced myocardial damage.


| INTRODUCTION
The renin-angiotensin system (RAS) promotes hypertrophy, increases oxidative stress, and exerts pro-inflammatory effects in the heart.
Renin is a secretory glycoprotein that generates angiotensin (ANG) I from its only known substrate, angiotensinogen. ANG I is then cleaved to ANG II, the effector peptide of the RAS. ANG II leads to apoptosis, necrosis, fibrosis, myocardial remodelling, and thus promotes cardiac failure. Correspondingly, inhibitors of the RAS are among the most potent drugs in the treatment of hypertension and cardiac failure, markedly increasing the life span of patients. 1 In addition to the classical renin transcript, alternative transcripts have been identified in several species including humans [termed exon (2)(3)(4)(5)(6)(7)(8)(9), exon(1A-9), renin-b, renin-c]. [2][3][4] These alternative transcripts are derived from the same renin gene, but lack exon 1. Exon 1 encodes for a signal sequence required for the cotranslational transport to the endoplasmatic reticulum (ER), and thus for the sorting of renin to the secretory pathway. All alternative renin transcripts are translated into a truncated prorenin. The protein, which we henceforth term cyto-renin, lacks the prefragment and the first The transcription of cyto-renin is under the control of an alternative promoter located in intron 1. 7 This promoter has very low basal activity, but is stimulated by glucose depletion in cardiac cells 7 as well as after myocardial infarction in the rat heart. 8 Cells overexpressing cyto-renin are more resistant against necrotic death under ischaemic conditions such as glucose depletion. 9 Moreover, overexpression of cyto-renin in the heart of transgenic rats reduces infarct size in ex vivo Langendorff preparations. 9 In the present study, we tested the hypothesis that cyto-renin overexpression exerts favourable effects on metabolic and mitochondrial functions, which may explain the reduced infarct size ex vivo or the protective effects in vitro. The investigations are based on measurements of oxygen consumption (OCR) and extracellular acidification rate (ECAR) using the Seahorse technology 10,11 as well as LDH activity, extracellular glucose and lactate contents. Thus, we were able to assess the impact of cyto-renin on the bioenergetic profile of H9c2 cardiomyoblasts. Specifically, we analysed the ratio between mitochondrial vs nonmitochondrial respiration, respiratory-dependent ATP synthesis, proton leak, and spare respiratory capacity after addition of specific inhibitors of the respiratory chain.

| Cell culture
The H9c2 cell line obtained from American Tissue Type Collection (Manassas, VA, USA) was cultured in DMEM medium supplemented with 100 U/mL penicillin, 100 μg/mL streptomycin and 10% foetal bovine serum in 75 cm 2 tissue culture flasks at 37°C in a humidified atmosphere of 5% CO 2 . Cells were fed every 3 days and subcultured after reaching around 80% confluence. H9c2 cells were transfected with a pIRES/Neo vector ± ren(2-9) cDNA as previously described. 6 Upregulation of ren(2-9) was 20-fold as determined by qRT-PCR analysis using the renin forward primer GCTCCTGGCAGATCACCAT and the reverse primer CCTGGCTACAGTTCACAACGTA.

| Extracellular glucose and lactate levels
After 24 hours incubation of untreated H9c2 cells, ren(2-9) cells or siRNA pretreated cells (scramble vrs. siRNA to renin) in 2 mL DMEM medium at 37°C, glucose and lactate levels of the cell culture supernatant were determined using the Biosens C line GF+ analyser (EKF Diagnostics, Germany). Glucose and lactate levels of the medium incubated without cells were measured simultaneously to calculate glucose consumption and lactate accumulation rates. An amount of 20 μL of the culture supernatants was incubated with 1 mL haemolysing solution using prefilled safe-lock reaction cups (EKF Diagnostics, Germany). Glucose consumption and lactate accumulation, respectively, were normalized to the cell numbers.

| LDH activity
Lactate dehydrogenase (LDH) activity was determined using the Cytotoxicity Detection Kit (Roche Applied Science, Mannheim, Germany) with slight modifications. Before measurements were performed according to manufacturer's instructions, 10 000 cells in 100 μL medium were seeded in 96-well plates and incubated for 24 hours at 37°C in a CO 2 incubator. Then, the cells were lysed by addition of 1% triton X-100 for 1 hour to detect total LDH activity.

| Seahorse XF analysis
The Seahorse XF analyser (Agilent Technologies, Seahorse Bioscience, Santa Clara, USA) was used to assess H9c2 extracellular acidification rate (ECAR, mpH/min) and cell mitochondrial oxygen consumption rate (OCR, pmol/min), including basal mitochondrial and nonmitochondrial OCRs, maximal respiration, proton leak, spare respiratory capacity, and ATP production. To determine optimal cell density for the experiments, different H9c2 cell numbers, ranging from 5000 to 180 000 cells/well, were seeded into the wells of a Based on the basal parameters, we determined the coupling efficiency of oxidative phosphorylation (CE) and the maximal cellular respiratory control (CRC max ). 10 CE was calculated as oligomycin-sensitive portion of mitochondrial respiration. CRC was derived from the ratio of FCCP-stimulated maximal respiration and oligomycininsensitive proton leak OCR. These parameters reflect electron transport and/or substrate oxidation as well as mitochondrial quality.

| Measurement of protein content
To control for variations in cell number due to different growth rates between H9c2 controls and ren(2-9)-expressing cells, we determined the protein contents per well after analysis of Seahorse experiments.

| Statistical analyses
The data presented are means ± SEM of independently performed experiments. Each of the Seahorse experiments comprises data of three wells per group. One-way ANOVA or two-way ANOVA with Bonferroni posttest analyses were performed as appropriate using GraphPad Prism (Graph Pad Software, La Jolla, CA, USA). Values of P < 0.05 were considered statistically significant.

| Nonmitochondrial O 2 consumption is increased in ren(2-9) cells
We used H9c2 cardiomyoblasts as an alternative model to study mitochondrial functions because unfortunately, the Seahorse system does not yield reliable results with primary adult rat cardiomyocytes     20). B, OCRs in cells after injection of rotenone (R) and antimycin A (AA) (cellular respiration) according to mitochondrial (blue) and nonmitochondrial components (yellow). C, OCRs in mitochondria after injection of oligomycin (Oligo) (mitochondrial respiration) according to ATP-linked respiration (orange) and proton leak (grey). D, Maximal OCRs obtained after injection of the uncoupler FCCP (maximal respiration) (green) and spare capacity (green shaded). Spare capacity was calculated by subtracting mitochondrial respiration from maximal respiration. E, Coupling efficiency (CE) of oxidative phosphorylation was determined as the oligomycin-sensitive portion of mitochondrial respiration. F, Maximal cell respiratory control (CRC Max ) was derived from the ratio of maximal respiration to oligomycin-insensitive proton leak OCR. Data represent mean ± SEM values with ***P < 0.001 or **P < 0.01 control cell lines vs ren(2-9) cells; #P < 0.05 total cellular respiration control cell lines vs the ren(2-9) cells Lastly, the mitochondrial DNA copy number related to the nuclear DNA content was not different between H9c2, pIRES F I G U R E 3 Bioenergetic profile of H9c2 cells is affected disadvantageously by renin knockdown. OCR was analysed after injection of inhibitors of the respiratory chain at indicated time-points. A, Real-time OCRs obtained from scramble controls (n = 9) and cells after knockdown of renin using 80 nmol siRNA (n = 9). B, Cellular respiration after injection of rotenone (R) and antimycin A (AA) representing mitochondrial (blue) and nonmitochondrial components (yellow). C, OCRs in mitochondria after injection of oligomycin (Oligo) (mitochondrial respiration) according to ATP-linked respiration (orange) and proton leak (grey). D, Maximal OCRs (green) measured after injection of the uncoupler FCCP and spare capacity (green shaded). Spare capacity was calculated by subtracting mitochondrial respiration from maximal respiration. E, Coupling efficiency (CE) of oxidative phosphorylation was determined as the oligomycin-sensitive portion of mitochondrial respiration. F, Maximal cell respiratory control (CRC Max ) was derived from the ratio of maximal respiration to oligomycin-insensitive proton leak OCR. Data represent mean ± SEM values with **P < 0.01 or *P < 0.05 scramble control vs siRenin-treated H9c2 cells. #P < 0.05 total cellular respiration scramble control vs siRenin-treated cells controls, and ren(2-9)cells, indicating similar mitochondrial contents in the different cell lines ( Figure 5C).

| Overexpression of ren(2-9) leads to enhanced extracellular lactate accumulation and LDH activity
The decreased ratio between mitochondrial and nonmitochondrial oxygen consumption in ren(2-9) cells might be due to a switch of energy metabolism from OXPHOS to enhanced glycolysis. In line with this hypothesis, the extracellular lactate accumulation after a 24-hour culture period was 2-fold higher in the medium of ren(2-

| DISCUSSION
The data presented here indicate that cardiac H9c2 cells overexpressing cyto-renin exhibit an altered bioenergetic profile, which is based on a switch to more aerobic glycolysis known as Warburg effect. Additionally, these ren(2-9) cells showed an enhancement of mitochondrial spare capacity which was reduced after renin knockdown. Unfortunately, the Seahorse system did not yield reliable results with primary adult rat cardiomyocytes (see results 3.1).
Therefore, we chose the H9c2 cardiomyoblast cell line despite being aware of its limitations with respect to its meaningfulness for the adult heart. However, as the aim of the study was to investigate the so far unknown functions of cyto-renin, we deem H9c2 cell line to be an appropriate model. Moreover, with respect to the protective effects of cyto-renin we already demonstrated such effects also in primary adult cardiomyocytes as well as in the isolated perfused heart preparations. 9 If sufficient substrates such as glucose, pyruvate, and glutamine were available, H9c2 cells carried out mitochondrial oxidative phosphorylation (OXPHOS) but to some extent also glycolysis as demonstrated by the mitochondrial oxygen consumption rate and the extracellular lactate accumulation. Thus, H9c2 cells seem to perform active glycolysis to support mitochondrial function in ATP synthesis.
The measurement of cellular oxygen consuming processes during coordinative inhibition of respiratory chain is the most useful test to estimate mitochondrial function. In ren(2-9) cells, the activity of nonmitochondrial oxygen consumers amounted to 40% of cellular oxygen uptake and this was twice as high as in control cells.
Nonmitochondrial oxygen consumers are, for example, the cytosolic NADPH oxidase, several desaturases and detoxification enzymes 10 but also extra-mitochondrial ATPases. 12 Herst et al 13 demonstrated that nonmitochondrial respiration can also be linked to cell surface oxygen consumption via trans-plasma membrane electron transport, leading to enhanced extracellular acidification. The authors hypothesized that the cell surface oxygen consumption indicates the extent of aerobic glycolysis, ie, glycolysis despite the presence of oxygen, known as Warburg effect. In line with this hypothesis, cyto-renin overexpressing cells exhibit increased extracellular lactate accumulation and cellular lactate dehydrogenase (LDH) activity together with reduced mitochondrial ATP-linked respiration. Thus, the enhanced glucose uptake could be a mechanism to compensate the limited ATP yield of only 2 ATP per glucose breakdown. All together, these data may indicate a shift from OXPHOS to more aerobic glycolysis, ie, lactate production. However, there are also data disclaiming this assumption. Thus, the extracellular acidification rate (ECAR) was reduced in ren(2-9) medium despite enhanced lactate accumulation. This can be explained by the fact that besides glycolytic lactate production a number of proton transporters in the plasma membrane as well as the generation of CO 2 in the tricarboxylic acid cycle take part F I G U R E 4 Influence of renin on the mitochondrial membrane potential. Mitochondrial membrane potential was determined as mean fluorescence intensity after addition of MitoTracker red CMXRos (A, B) or JC1 (C, D). Both mitochondrial probes were added either to H9c2, pIRES and ren(2-9) cells (each n = 10) or scramble controls and renin knockdown cells (n = [10][11][12][13][14]. Fluorescence intensity of the cells was measured by flow cytometry. Data represent mean ± SEM values with ***P < 0.001 or *P < 0.05 ren (2-9) cells or siRenin-treated cells to the corresponding controls WANKA ET AL.
| 5997 in the extracellular acidification. 14,15 Thus, conversion of one glucose molecule to lactate yields two molecules of lactateˉand two H + , while complete breakdown of glucose to CO 2 yields 6 HCO 3ˉa nd 6 H + . Therefore, the ECAR should be three times greater when glucose is converted completely to CO 2 . Because H9c2 cells performed both OXPHOS and glycolysis, the observed acidification rate of H9c2 cells should reflect both proton-generating pathways. Therefore, in ren(2-9) cells a preferential execution of aerobic glycolysis could indeed be associated with the observed reduced ECAR.
From a bioenergetic point of view, aerobic glycolysis is inefficient because of its limited ATP yield. Nevertheless, it may provide an advantage for the survival in a hypoxic environment as seen in cancer cells. 16 Also in neuronal cells, an upregulation of aerobic glycolysis resulted in a decreased mitochondrial respiration and, as a consequence, in the decreased production of reactive oxygen species (ROS) as well as an enhanced resistance to stress factors. 17 Vice versa, an age-dependent decrease of aerobic glycolysis was accompanied by an increased mitochondrial respiration and ROS production and led finally to neuronal cell death. 17 In this sense, in ren (2)(3)(4)(5)(6)(7)(8)(9) cells the switch between mitochondrial OXPHOS and aerobic glycolysis appears to be more balanced to avoid harmful generation of ROS during ischaemia/reperfusion and to maintain a high level of energetic plasticity for surviving. A dysfunctional heart usually lacks this metabolic plasticity. 18 In most cells, mitochondria operate at a basal level using only a small fraction of the available maximal energetic capability. Maximal oxygen consumption after uncoupling of the respiratory chain by FCCP reflects the maximum activity of electron transport and F I G U R E 5 Visualization of the mitochondrial network in H9c2 cell lines. Mitochondrial distribution in pIRES and ren(2-9) cells was evaluated by the labelling with MitoTracker red CMXRos probe (red) combined with the membrane dye wheat germ agglutinin (WGA, green) and the nuclear dye 4′,6′-diamidino-2-phenylindole (DAPI, blue) (A). Magnification amounted to 100x. Location of mitochondria in relation to the nucleus and the cell membrane was determined by the Line Profile software of the BZ II 9000 microscop (B) substrate oxidation. 10 This parameter is enhanced in ren(2-9) cells and decreased after renin depletion, indicating an influence of cytorenin on the electron transport chain activity and/or the substrate oxidation. Integration of the cell respiratory control ratio max (CRC max ) affirmed this statement. RCR max which is enhanced in ren(2-9) cells, too, is sensitive to changes in substrate oxidation and proton leak whereby proton leak was unchanged. The difference between the basal and the maximal respiratory activity represents the spare respiratory capacity (SRC). Thus, SRC indicates how close to its bioenergetics limit a cell is operating. In the heart, it has been shown that the SRC is depleted under severe stress conditions such as high workload during continuous catecholamine infusion, oxidative stress, or ischaemia. 11,19,20 This decrease is then connected with an increase of cell death. Based on these findings, the decreased SRC as detected in renin knockdown cells could be responsible for the enhanced cell death under basal conditions and after glucose deprivation. 9 Because, the SRC also represents the amount of oxygen consumption that is available for cells to use under conditions of increased ATP demand or during stress, it should support maintenance of organ function and cellular repair during such conditions. In this context, Dranka et al 21 hypothesize that a therapeutic increase of SRC could be protective in diseases such as hypertension or atherosclerosis that are associated with increased oxidative stress.
Furthermore, it is important to notify that mitochondria in excitable cells, such as cardiomyocytes, are exposed to high fluxes of calcium and other ions, where the proton gradient is utilized to remove toxic calcium concentrations from the cytosol.
Although SRC is a well-recognized phenomenon, data regarding its regulatory components or factors are limited. One potential source that determines levels of both maximal respiration and SRC is the substrate availability that is synchronized with the activity of electron transport chain. 22 Our initial investigations clearly show that maximal OCR and spare capacity were decisive influenced by the cell number seeded per well. Because increased cell density is, of course, associated with an increased consumption of substrates ie, glucose as well as oxygen the reduced maximal respiration may be due to H9c2 cardiomyoblasts the increased glycolytic flux could also deliver more pyruvate to mitochondria, thereby increasing the respiratory capacity. A sensitive parameter controlling the activity of the mitochondrial electron transport chain, the respiratory rate and the ATP synthesis is the mitochondrial membrane potential (ΔΨ m ), ie, the proton-motive force. 24 We determined ΔΨ m by the fluorescent probes JC1 and MitoTracker red CMXRos. 25,26 Surprisingly, we found different results using these probes. While control-and ren(2-9) cells showed similar ratios of orange/green JC1 fluorescence intensities, indicating similar level of ΔΨ m , the intensity of MitoTracker red fluorescence was significantly enhanced in ren(2-9) mitochondria. Up to now, the dual-colour (orange and green) ratiometric assessment of mitochondrial polarization state by JC1 is a more reliable indicator of ΔΨ m than other mitochondria tracking probes 27 implicating that the enhanced spare capacity in ren(2-9) cells did not be due to the ΔΨ m .
Besides ΔΨ m mitochondrial mass, morphology, or localization can contribute to changes in the fluorescence intensity of such probes.
Buckman et al 26 found that the loading of MitoTracker red is ΔΨ m dependent but is also influenced by the oxidant status. However, our preliminary data exclude a modified ROS level in ren(2-9) cells under basal conditions (data not shown). In other studies, enhanced MitoTracker red fluorescence intensity was associated with a hyperfusion of mitochondria 28 or an increase of mitochondrial mass. 29 From our analyses, we conclude that the enhanced MitoTracker red FLI in ren(2-9) cells was not influenced by an altered mitochondrial mass but could be due to a shift of mitochondria away from the nucleus. However, because nonyl-acridine orange labelling was subjected to a marked quenching effect, additional studies are necessary to verify this observation.
Altogether, our data demonstrate that the protective effects of cytosolic renin on the one side are associated with a shift from OXPHOS to more aerobic glycolysis. This improves the metabolic flexibility and therefore the ability to better cope with fluctuating nutrient and oxygen supplies as observed during ischaemia. On the other side, the increase of spare respiratory capacity could maintain sufficient functional mitochondria to avoid oxidative damage or calcium overload.
It is quite likely that cyto-renin compensates for high renin states in ischaemic conditions. Indeed, we previously showed that overexpression of secretory renin increased necrosis rate while overexpression of cyto-renin decreased necrosis rate in the same cell model. 6 However, in H9c2 cells even in the absence of secretory renin expression we observed markedly protective effects under glucose depletion. 9 Thus, cyto-renin may be a promising new target for the prevention of ischaemia-induced myocardial damage. These findings may be useful for the development of new therapeutic strategies.