Succinate causes pathological cardiomyocyte hypertrophy through GPR91 activation

Activation of GPR91 by succinate causes cardiomyocyte hypertrophy likely due to long-term blood pressure counter regulatory mechanisms

Extracellular accumulation of up to millimolar levels of succinate is observed in pathophysiological conditions, such as ischemia [14],[15],[21]. Moreover, there is a close correlation between myocardial ischemia and hypertrophic cardiomyopathy [24],[25]. To investigate whether succinate causes cardiac hypertrophy, a succinate dose of 0.066 mg/kg, equivalent to that observed during ischemic events [15], was administered intravenously to rats, once a day, for 5 consecutive days. By using perfused-fixed hearts from PBS (control) and succinate treated animals, we evaluated changes in cardiomyocyte width from left cardiac ventricle sections in longitudinal orientation. Morphometric analysis showed a significant increase in average myocyte width of groups treated with succinate compared to controls, (12 ± 0.10 μm in control vs 14.7 ± 0.10 μm in succinate treated animals, p < 0.001), (Figure 1A). We also observed a significant increase in myocyte nuclear diameter in succinate treated groups (4.3 ± 0.12 μm in cardiomyocytes from control animals vs 5.6 ± 0.2 μm in succinate treated rats, p < 0.001), (Figure 1B), suggesting that high circulating succinate levels might cause cardiac hypertrophy. This finding was confirmed by evaluating the expression levels of genes expressed during cardiac hypertrophy. We detected upregulation of hypertrophic markers atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP), and β-myosin heavy chain (MYH7). Moreover we fond a significant increase in α-skeletal actin (α-SkA) mRNA levels, a known marker for pathological hypertrophy, in heart samples from succinate treated groups. We observed an increase of 160% in the expression level of ANP (a.u. = 100 in control cells vs. 260 ± 6.5% in cells from succinate treated rats, p < 0.001), an increase of 175% in the expression of BNP (a.u. = 100% in cells from control rats vs 275 ± 10% in cells from succinate treated rats, p < 0.001), an increase of 125% in the expression of MYH7 (a.u. = 100% in cells from control rats vs 225 ± 3% in cells from succinate treated rats, p < 0.001), and increase of 168% in the expression level of α-SkA (a.u. = 100 in control cells vs 268 ± 6.3% in cells from succinate treated rats, p < 0.001) in freshly isolated adult cardiomyocytes from succinate treated rats when compared to control animals (Figure 1C-F). Succinate is knows to activate the renin angiotensin system (RAS) [21], which can modulate blood pressure [26]. Since sustained increase in blood pressure is know to cause cardiac hypertrophy [27],[28], we investigated whether the hypertrophy induced by high levels of succinate in the blood stream was a consequence of succinate triggering changes in the arterial blood pressure. Under our experimental conditions, we found that the mean arterial blood pressure (MAP) level was unaffected after two days of succinate treatment, but slightly increased at day 4, and reverted to normal values on the final day of the experiment (day 3: 101.9 ± 0.78 mmHg in control rats, 109 ± 0.42 mmHg in control rats treated with losartan, 102 ± 1.52 mmHg for succinate-treated rats and 99 ± 3.85 mmHg for rats treated with succinate and losartan, day 4: 101.2 ± 2.86 mmHg in control rats, 95.82 ± 1.88 mmHg in control rats treated with losartan, 111.6 ± 2.60 mmHg in succinate-treated rats and 79.92 ± 2.48 mmHg in rats treated with succinate and losartan, day 5: 98.8 ± 2.94 mmHg in control rats, 90.43 ± 5.2 mmHg in control rats treated with losartan,105.3 ± 4.8 mmHg in succinate treated rats, 77.51 ± 5.79 mmHg in rats treated with succinate and losartan). These variations occurred without any change on the heart rate (Figure 2A-D), even though at the last experimental day, the serum concentration of succinate was significantly higher in treated animals compared to control (0 mM in control rats vs 0.9 ± 0.13 mM in succinate treated rats, p < 0.001), (Figure 2E). We noticed that increases in blood pressure induced by succinate were reverted by losartan, a well-known inhibitor of type I angiotensin-II receptor [29]. Additionally, the consequences of in vivo succinate exposure for cardiac function were further investigated by echocardiography experiment in the presence or absence of losartan. Table 1 shows that succinate increased cardiac output (52.17 ± 5 in control rats vs 68.03 ± 2.9 in succinate rats, p < 0.05), left ventricular end diastolic volume (LVd: 180 ± 9.7 in control rats vs 245.20 ± 10.11 in succinate rats, p < 0.01), stroke volume (SV: 136.7 ± 12.63 in control rats vs 167.5 ± 7.34 in succinate rats, p < 0.01), and left ventricular chamber dimension, at both systole (LVIDs: 3.02 ± 0.06 in control rats vs 3.96 ± 0.10, p < 0.01) and diastole (LVIDd: 6.15 ± 0.22 in control rats vs 6.99 ± 0.09, p < 0.01). Although losartan did not affect most evaluated parameters, it slightly attenuated succinate-induced increase in left ventricular chamber diameter during diastole (LVIDd: 6.99 ± 0.09 in succinate rats vs 6.50 ± 0.15 in succinate + losartan, p < 0.05). Real time PCR of hypertrophic markers from these experimental groups showed that losartan partially reverted the re-expression of ANP and MYH7 induced by succinate. We observed a decreased of 110.7% in the expression level of ANP (a.u. = 304 ± 4.33% in rats hearts treated with succinate vs. 193.3 ± 6.67% in rats hearts treated with succinate in the presence of losartan, p < 0.001), and a decreased of 38.7% in the expression of MYH7 (a.u. = 174 ± 3.05% in rats hearts treated with succinate vs 135.3 ± 2.9% in in rats hearts treated with succinate in the presence of losartan, p < 0.001), (Additional file 1: Figure S1A-B). Thus, these findings corroborate previous data showing that succinate activates RAS [21],[23],[30]. To address the role of GPR91 on succinate-induced cardiac hypertrophy, we performed echocardiogram using GPR91 knockout (GPR91 KO) mice. Table 2 shows that GPR91-KO mice have no alteration in echocardiographic parameters when compared to wild type. Despite the absence of GPR91, KO mice treated with succinate showed changes in left ventricular ejection fraction (LV-EF 44.33 ± 0.35 in wild mice vs 57.37 ± 0.77 in wild mice treated with succinate, p < 0.05, and 38.17 ± 1.4 in GPR91 KO mice vs 52.9 ± 4.7 in GPR91 knockout mice treated with succinate, p < 0.05), fractional shortening (LV-FS: 21.51 ± 0.23 in wild mice vs 29.86 ± 0.23 in wild mice treated with succinate, p < 0.05, and 18.05 ± 0.70 in GPR91 knockout mice vs 26.14 ± 2.97 in GPR91 knockout mice treated with succinate, p < 0.05), and systolic volume (LV-SV: 32.65 ± 0.65 in wild mice vs 22.13 ± 0.87 in wild mice treated with succinate, p < 0.01, and 31.43 ± 3.44 in GPR91 knockout mice vs 17.76 ± 1.36 in GPR91 knockout mice treated with succinate, p < 0.01). Importantly, succinate-induced increase in left ventricular posterior wall (LVPWs: 0.64 ± 0.09 in wild mice vs 1.05 ± 0.07 in wild mice treated with succinate, p < 0.05) was observed only in wild type mice, indicating that GPR91 is essential for succinate-induced hypertrophic effects on the heart. We also evaluated the expression levels of the aforementioned hypertrophic markers in cardiomyocytes isolated from GPR91 deficient animals. We found that in the absence of GPR91, injection of succinate was unable to induce the expression of ANP (a.u. = 100% in cells from WT mice vs 230 ± 15% in cells from succinate treated WT mice vs 15 ± 2% in cells from GPR91^−/− mice treated with succinate vs 120% in cells from GPR91^−/−, p < 0.001), BNP (a.u. = 100% in cells from WT mice vs 230 ± 15% in cells from succinate treated WT mice vs 23 ± 1% in cells from GPR91^−/− mice treated with succinate vs 116 ± 2% in cells from GPR91^−/−, p < 0.001), and MYH7 (a.u. = 100% in cells from WT mice vs 230 ± 15% in cells from succinate treated WT mice vs 12 ± 1.4% in cells from GPR91^−/− mice treated with succinate, vs 108 ± 4% in cells from GPR91^−/−, p < 0.001), (Figure 3A-C). The involvement of other key TCA cycle intermediate was not considered here since it was already demonstrated that succinate is the only agonist for GPR91 [21]. Taken together, these results show that high circulating levels of succinate lead to cardiac hypertrophy, through direct activation of GPR91. Nonetheless, the results also suggest that succinate-induced remodeling is not limited to its direct effects on the GPR91 in cardiac tissue, but might as well have its origin point in other organs besides the heart.

Succinate-induced cardiomyocyte hypertrophy involves activation of MEK/ERK1/2 and HDAC5 pathways

GPR91 can be coupled with different second messenger signaling systems, depending on the individual cell type where it is expressed. For instance, succinate uses both G_i/G_o and Gq/11 pathways to increase ERK1/2 phosphorylation and intracellular Ca²⁺[21]-[23],[31], which are, in fact, pathways involved in cardiac hypertrophy induced by G-protein coupled receptors [5],[13],[32],[33]. In order to gain further insight into the mechanisms by which succinate induces cardiomyocyte hypertrophy and the dependence of this process on GPR91, we used primary cultures of neonatal rat cardiomyocytes. We first investigated whether in vitro succinate treatment could also induce hypertrophy. For that, we tested several different concentrations of succinate, and measured cellular width after treatment with each of these concentrations (500 ± 12 μm² for control cells; 682 ± 20 μm² for cells treated with 25 mM succinate; 800 ± 15 μm² for 0.5 mM succinate, 1000 ± 15 μm² for 0.75 mM succinate, 1041 ± 25 μm² for 1 mM succinate, 1083 ± 13 μm² for 1.5 mM succinate, 1080 ± 10 μm² for 2 mM succinate, 1085 ± 20 μm² for 2.5 mM succinate, Additional file 2: Figure S2). We found that neonatal cardiomyocytes exposed to 1 mmol/L succinate for 36 hours had increased cell surface area when compared to controls (600 μm² ± 15 in cells from control vs 1100 μm² ± 20 in cells from succinate treated cardiomyocytes, p < 0.01, n = 60 cells), (Figure 4A-B). In addition, succinate treatment led to an increase in ANP expression levels (a.u. = 0.93 ± 0.06 in cells from control vs 1.3 ± 0.10 in cells from succinate treated cardiomyocytes, n = 3 independent experiments, p < 0.01), (Figure 4C-D), providing evidence for onset cardiomyocyte hypertrophy, an effect that was not observed upon knockdown of succinate receptor with GPR91 siRNA. Real time PCR analysis indicated that siRNA against GPR91 abolished the expression of succinate receptor (Figure 5A) and prevented the increase of hypertrophic markers induced by succinate: ANP (a.u. = 100% in control cells vs 180 ± 8% in succinate treated cells vs 100 ± 1.4% in cells with GPR91siRNA, p < 0.001) and BNP (a.u. = 100% in control cells vs 180 ± 8% in succinate treated cells vs 95 ± 1.4% in cells with GPR91siRNA, n = 3 independent experiments, p < 0.001), (Figure 5B-C). These findings show that succinate causes hypertrophy due to a direct activation of GPR91 in cardiomyocytes.

Because ERK1/2 phosphorylation is a classical MAPK pathway activated in cardiomyocyte hypertrophy [13],[32], and GPR91 activation is known to be associated with the ERK1/2 signaling cascade [21],[23],[31], we investigated whether succinate treatment alters ERK1/2 phosphorylation levels in cardiomyocytes as well. We found that exposing cardiomyocytes to succinate increased phosphorylation of ERK1/2, without affecting its expression levels (a.u. = 0.58 ± 0.02 in control cardiomyocytes vs 0.78 ± 0.02 in succinate treated cardiomyocytes, p < 0.01, n = 3 independent experiments). We also observed that efficient silencing of GPR91 prevented the increase in ERK1/2 phosphorylation upon succinate treatment (a.u. = 0.58 ± 0.02 in control cardiomyocytes vs 0.89 ± 0.05 in succinate treated cardiomyocytes vs 0.70 ± 0.03 in succinate treated cardiomyocytes transfected with GPR91siRNA, p < 0.01, n = 3 independent experiments), (Figure 6A-B). Moreover, the effects of succinate on cellular area and ANP expression were prevented by PD098059, an inhibitor of MEK1/2, (a.u. = 100% in control cells vs 180 ± 12% in cells treated with succinate vs 125 ± 5% in cells treated with succinate in the presence of PD098059, vs 109 ± 4% in PD098059 treated cells alone; (fluorescence intensity: a.u. = 62500 ± 500 in control cells vs 87500 ± 763 in cells treated with succinate vs 50500 ± 10000 in cells treated with succinate in the presence of PD098059, vs 71000 ± 577% in PD098059 treated cells alone; n = 45, p < 0.01), (Figure 6C-E). We observed a very low expression of ANP, close to the control level. Since PD098059 is specific to MAPK inhibition, this data may suggest the existence of another protein, downstream to the MAPK inhibition, that negatively feedbacks the pathway, inactivating other important intermediate or adding to MEK inhibition, contributing to the lower activity caused by PD098059 [34]-[36]. However, altogether these findings show that MEK/ERK cascade is a crucial downstream pathway of the hypertrophic effects induced by succinate.

Additional important signaling cascades that are well-known to be involved in cardiomyocyte hypertrophy are the calcineurin-NFAT [2],[3] and the histone deacetylase (HDAC) [37] pathways, both activated upon an increase in intracellular Ca²⁺ signals [38]. We have previously demonstrated that succinate increases intracellular Ca²⁺ transients in cardiomyocytes [22]. Thus, we now investigated whether exposing cardiomyocytes to succinate would trigger the above-mentioned Ca²⁺-dependent hypertrophic signaling cascades. We found that succinate did not activate NFAT, represented by the absence of NFAT in the nucleus upon succinate treatment (Additional file 3: Figure S3A-B). Instead, succinate activated calmodulin kinase IIδ (CaMIIδ), (a.u. = 1.75 ± 0.06 in control cells, vs 2.30 ± 0.1 in succinate treated cells, n = 3, p < 0.01), (Figure 7A-B), and HDAC5 signaling pathway (Figure 7C-E). HDAC5 translocated from the nucleus to the cytosol upon succinate treatment, an effect that was prevented by KN93, an inhibitor of CaMKIIδ [39], (a.u. = 93000 ± 1528% in control cells, vs 46000 ± 3000% in succinate treated cells, vs 82000 ± 9000% in succinate and KN93 treated cells, vs 93000 ± 2000% in KN93 treated cells alone, p < 0.001, p < 0.01), (Figure 7C-D). The increase in cellular area induced by succinate was also prevented by KN93 treatment (a.u. = 100 ± 3.68% in control cells, vs 180 ± 5.45% in succinate treated cells, vs 117 ± 8.35% in succinate and KN93 treated cells, vs 114 ± 0.8% in KN93 treated cells alone, p < 0.001, p < 0.01), (Figure 7E). Part of KN93 effect on cellular area might be due to its unspecific effect as an antagonist. Validating this signaling pathway we further found that efficient silencing of GPR91 prevented both the translocation of HDAC5 from the nucleus to the cytosol (a.u. = 91667 ± 4410 in control cells, vs 88333 ± 4410 in GPR91 siRNA cells, vs 50667 ± 3480 in succinate treated cells, vs 89667 ± 333.3 in GPR91 siRNA and succinate treated cells, p < 0.001), and the increase in cellular area induced by succinate (a.u. = 102 ± 0.57% in control cells, vs 84 ± 0.57% in GPR91 siRNA cells, vs 156 ± 2% in succinate treated cells, vs 103.3 ± 8.8% in GPR91 siRNA and succinate treated cells, p < 0.001), (Additional file 4: Figure S4A-C). These results indicate that succinate causes cardiomyocyte hypertrophy through the activation of CaMKIIδ/HDAC5 pathway.

To verify whether succinate triggers similar hypertrophic signaling cascades in vivo as well, we intravenously administered 0.066 mg/kg succinate in 8-week-old rats. Succinate was injected for 5 days, once a day, and the expression pattern of phospho-ERK1/2 and changes in intracellular Ca²⁺ transients were evaluated in freshly isolated adult cardiomyocytes. We found that high circulating levels of succinate increased the expression of phospho-ERK1/2 compared to control cells (a.u. = 1 ± 0.3 in cells from control rats vs 2 ± 0.25 in cells from succinate treated cardiomyocytes, p < 0.01, n = 3), (Additional file 5: Figure S5A-B). We also found that administration of succinate for 5 consecutive days modulated Ca²⁺ transients in isolated ventricular myocytes loaded with the Ca²⁺ fluorescent probe fluo-4/AM. As expected, succinate increased the amplitude (3.8 ± 0.4 F/F₀ in cells from control rats vs 5.2 ± 0.6 F/F₀ in cells from succinate treated cardiomyocytes, p < 0.01, n = 40 cells), and decreased the decay rate of the Ca²⁺ transient, (950 ± 9 ms in cells from control rats vs 920 ± 7 ms in cells from succinate treated cardiomyocytes, p < 0.05, n = 40), (Additional file 6: Figure S6A-D). Collectively, these results show that succinate activates intracellular hypertrophic signaling cascades, both in vitro and in vivo.

Ischemic diseases increase succinate blood level in patients

Materials and reagents

Cardiomyocyte primary culture extracting kit was obtained from Worthington Biochemical Corporation (Lakewood, USA). Dulbeccos’s Modified Eagle’s Medium (DMEM), penicillin, streptomycin, amphotericin and fetal bovine serum (FBS) were purchased from Gibco (Grand Island, USA). Cytosine β-D-arabinofuranoside (ARA-C), fibronectin, KN-93 (Ca²⁺/calmodulin-dependent protein kinase II inhibitor) and ultra-pure succinic acid were obtained from Sigma (St. Louis, USA), Fluo-4/AM, DAPI and secondary antibodies conjugated to Alexa-488, Alexa-633 and Silencer kit were from Ambion, Life Technologies (New York, USA). Trizol reagent was obtained from Invitrogen (Eugene, USA). Polyclonal anti- GAPDH, anti- HDAC5, and anti-ANP antibodies were from Santa Cruz Biotechnology (Santa Cruz, USA). Monoclonal anti – Phospho-p44/42 MAPK and polyclonal anti-MEK1/2 were purchased from Cell Signaling Technology (Boston, USA), monoclonal anti-CAMK2δ from Abnova (Taipei, Taiwan). Hydromount was from National Diagnostics (St. Louis, USA). PD- 98059 (MEK 1/2 innhibitor) was from Calbiochem (Germany). Enhanced chemiluminescence (ECL-plus Western Blotting Detection System) and peroxidase-conjugated antibodies were purchased from Amersham Biosciences (Buckinghamshire, UK). All other reagents were of the highest quality commercially available.

Human samples

Use of blood samples from patients with acute coronary disease, acute myocardial infarction and hepatic transplant as well as the term of free and informed consent, n: 03182712.2.0000.5149 and n: 00907612.0.0000.5149, respectively, were approved by the Ethics Committee in Research of UFMG-COEP.

Animals

In this study, we used adult (250 g) and neonatal (3–5 days old) Wistar rats, acquired from Centro de Bioterismo (CEBIO) of the Federal University of Minas Gerais. GPR91 ^−/− mice (C57BL/6 background) were provided by Novartis. Animals were maintained on a standard diet and housed under a 12-hour light–dark cycle. All animal experiments were performed in accordance to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH publication No. 85–23, revised 1996).

Genotyping of mice

Mice were characterized by PCR using genomic DNA extracted from tail biopsies. DNA was digested with the REDExtract-N-Amp Tissue PCR Kit (Sigma-Aldrich). Routine genotyping by PCR was performed using the following primers: The forward primer sequence was 5′ TTA CGC CAC TGG GAA CTG GA3′ and the reverse primer sequence was 5′ TTG ATG GCC TTC TGG GAA CA 3′[56]. Primers for GPR91 were designed using the software Primer3 based on the sequence deposited in the NCBI Nucleotide Bank NM_001001518.1.

Adult cardiomyocyte isolation

Adult ventricular myocytes were freshly isolated as previously described [59],[60] from adult Wistar rats previously treated with 0.066 mg/kg succinate intravenously. Cells were incubated in DMEM and the experiments were carried out at room temperature (22-24°C).

Primary culture of neonatal cardiomyocytes

Cardiomyocytes were isolated from hearts of 3–5 days old Wistar rats according to the manufacturer’s instructions (Worthington Biochemical Corporation). The rats were anesthetized with sodium pentobarbital (50 mg/kg body weight intraperitoneally), and the beating hearts were removed surgically. Cells were resuspended in Dulbeccos’s Modified Eagle’s Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), 100 units/ml penicillin, 100 μg/ml streptomycin and 0.25 μg/ml anfotericin-b. Cardiomyocytes were plated into fibronectin-coated culture dishes or flasks, and incubated at 37°C in a 5% CO₂ incubator. Two days after plating, cells were rinsed with DMEM and fed for another 24 h with regular culture medium, now including 20 μg/ml cytosine β-D-arabinofuranoside (ARA-C). The ARA-C was added to the culture medium for 48 h to inhibit growth of non-cardiomyocyte cells. The medium was then replaced with FBS-free DMEM with or without succinate (1 mmol/L) for 12 h and subsequently changed to 10% FBS-containing medium with or without succinate (1 mmol/L) for 24 h. Cardiomyocytes were used in experiments at the fourth day of culture. At this point, there were approximately 5-6 × 10⁴ cardiomyocytes per culture dish, which comprised approximately 95% the total cell population [22],[59],[61]. The cells were then used for immunofluorescence, western blot or qRT-PCR analysis.

Preparation of siRNA

Potential target sites within the GPR91 gene were selected and then searched with NCBI Blast to confirm specificity for the receptor. The siRNAs for GPR91, and a siRNA containing the same nucleotides for GPR91 but in a scrambled sequence were prepared by a transcriptional-based method using the Silencer kit, according to the manufacturer’s instructions. The sense and antisense oligonucleotides of siRNA were, respectively: 5′ AAT CTC TAA TGC CAG CCA ATT CCT GTC TC 3′ and 5′ AAA ATT GGC TGG CAT TAG AGA CCT GTC TC 3′. For siRNA studies, day 4 neonatal cardiomyocyte cultures were treated with 100 nM of each siRNA [22]. We used single wall carbon nanotubes (CNT) to deliver siRNA and silence GPR91, as previously described [62]. Cardiomyocytes were incubated at 37°C in an atmosphere of 5% CO₂ for 48 hours prior to use.

Western blotting

Cardiomyocytes were harvested as described and protein content was quantified according to Bradford protein assay. For ANP, 50 μg of whole cell proteins were separated by 12% SDS-PAGE. For GAPDH detection, mouse monoclonal anti-GAPDH antibody was used at a dilution of 1:1500. For ERK 1/2 detection, a rabbit polyclonal antibody was used at dilution of 1:500. For ANP detection rabbit polyclonal anti-ANP was used at a dilution of 1:200. For Phospho-p44/42 MAPK (ERK1/2) detection, a rabbit monoclonal antibody was used at a dilution of 1:1000. For CaMKIIδ detection, a mouse monoclonal anti- antibody was used at a dilution of 1:500. The antibody incubation proceeded for 2 h at room temperature. After washing, blots were incubated in HRP-conjugated goat-anti-mouse or rabbit IgG1 secondary antibody at a dilution of 1:5000 at room temperature for 1 h. Immunodetection was carried out using enhanced chemiluminescence [22],[59].

Immunofluorescence

Confocal immunofluorescence was performed as described [22],[59]. Briefly, cardiomyocytes were seeded onto 6 well culture dishes, treated with succinate 1 mM and 36 hours later were fixed with 4% paraformaldehyde, permeabilized with PBS 1X/Triton 0.5% and non-specific binding was blocked (PBS, BSA 10%, Triton 0.5%, goat serum 5%) for 1 hour. Cells were then incubated with anti-ANP (1:10), anti α-actinin (1:150), anti-HDAC5 (1:50) or with anti-NFAT (1:50) for 2 hours at room temperature. This was followed by incubation with specific secondary antibodies conjugated with Alexa-Fluo 488 or 633 (1:500) for 1 hour. Images were obtained using Zeiss LSM 510 confocal microscope (Thornwood, USA) [22],[59].

Measurement of intracellular Ca²⁺

Intracellular Ca²⁺ was monitored in individual cardiomyocytes by line scanning and time lapse confocal microscopy as described previously [22],[59],[60]. Briefly, adult cardiomyocytes were incubated with Fluo-4/AM (6 μmol/L) for 30 minutes at 37°C. Coverslips were transferred to a perfusion chamber on the stage of a Zeiss LSM510 confocal microscope. Cells were electrically stimulated at 1 Hz to produce steady-state conditions. Fluo-4 was excited at 488 nm and observed at 505–550 nm. Increases in Ca²⁺ were expressed as percent increase in Fluo-4 fluorescence over baseline [22],[63],[64].

PD 98059 (MEK 1/2 inhibitor), KN93 (Ca²⁺/calmodulin-dependent kinase II inhibitor) and losartan (AT1 angiotensin II receptor antagonist)

10 μmol/L of PD or 1 μmol/L KN93 [39] was added combined of 1 mmol/L succinate to the cells. Losartan was administrated in the drinking water, once a day, for 5 days at doses of 200 mg/l [65].

Intravenous administration of succinate and measurement of arterial blood pressure

Animals were housed in a temperature-controlled room under a 12 hour light–dark cycle with water and standard rodent chow available ad libitum. Rats and mice under ketamine and xylazine anesthesia had the femoral vein catheterized for intravenous injection of succinate (0.066 mg/kg in rats and 0.039 mg/kg in mouse) or PBS. Blood pressure was measured in unanesthetized animals by the tail-cuff method, MAP was recorded in rats for 5 days; at least 15 measurements were made daily with a BP-2000 blood pressure analysis tail-cuff System (isitech Systems). The mean systolic blood pressure and pulse rate were taken for each animal [21],[66].

Real-time PCR

Total RNA was isolated from adult and neonatal cardiomyocytes using TRIzol and cDNA synthesized using SuperScript II kit (Invitrogen). DNA templates were amplified by real time PCR on the StepOnePlus™ Real-Time PCR Systems (Applied Biosystems, CA) using the SYBR green method, as described. Sequence of the primers used are: ANP FW: 5′-GGATTTCAAGAACCTGCTAGA-3′ and RE 5′-CTTCATCGGTC TGCTCGCTCA-3′; BNP5′AGACCACCGCTC TTGTGTGTG-3′ and RE 5′ CTGACC CATACCTACCATGACACC-3′; MHC and GPR91 FW: 5′-TTACGCCACTGGGAACT GGA-3′ and RE 5′-TTGATGGCCTTCTGGGAACA-3 [22],[61],[67]: FW: 5′-CAG

GCGGTGCTGTCTCTCTAT-3′ and RE: 5′-GGCAGGGCATAACCCTCATA-3′ for α-SkA [61]. Experiments were performed in triplicate for each data point.

Tissue preparation

After administration for 5 days succinate the rats were weighed, anesthetized (i.p. sodium pentobarbital 80 mg/BW plus ketamine chloridrate 10 mg/BW) and perfused intracardially with Ringer’s solution followed by 10% neutral buffer formalin (NBF). After fixation, the heart was removed, weighted and the relative organ weight was calculated per 100 g of body weight. Fragments of the heart were embedded in paraffin, sectioned (5 μm) and mounted in silanized glass slides. For histological and morphometric studies, the sections were stained with hematoxylin and eosin or Masson’s trichrome [68].

Morphometry

The quantitative analysis of myocyte cross-sectional height and nuclear diameter were measured in histological sections by using computer-assisted image analysis and the Scion Image software (http://www.scioncorp.com), as previously described [68]. For this purpose, digital images were obtained with a Nikon Coolpix 995 digital camera (Nikon Instruments Inc., Melville, USA).

HPLC analysis

Serum samples from humans were collected to determine succinate using HPLC as previously described [69]. Mice samples were collected under CO₂ euthanasia, while human samples were collected from conscious individuals. All subjects had documented coronary atherosclerosis. Accurate details about the inclusion and exclusion criteria for this trial were previously published. All human subjects gave written informed consent, and the study was approved by the Institutional Review Committees at all sites.

Echocardiography

Animals were anaesthetized using a nose cone with isoflurane at 5% for one minute and the maintenance dose was 1.25%. The anterior chest was shaved, and the animals were placed in supine position on an imaging stage equipped with built-in electrocardiographic electrodes for continuous heart rate monitoring and a proper heating pad to avoid hypothermia. In vivo cardiac function was assessed noninvasively using a high-frequency, high-resolution echocardiographic system consisting of a VEVO 2100 ultrasound

Machine equipped with a 30–40 MHz bifrequencial transducer (Visual Sonics, Toronto, Canada). High-resolution images were obtained as previously described [70]-[73].

Statistical analysis

Results are expressed as mean values ± S.E., except where otherwise noted. Prism (GraphPad Software, San Diego, CA) and Image J (NIH; Bethesda, MD) were used for data and image analysis, respectively. Statistical significance was tested using One-way ANOVA followed by Bonferroni test, and p value < 0.05 was taken to indicate statistical significance.