NRVMs were obtained from 2 to 3 days old Sprague-Dawley rats. The isolation and culture were developed according to Galvez et al. (Galvez et al., 2001). Briefly, neonatal rats were decapitated, hearts extracted, and atria removed. Ventricles were washed in sterile Hank’s medium (Sigma-Aldrich, San Luis, MI, United States) at 37°C and submitted to enzymatic digestion. Cardiomyocytes were purified by a pre-pleated step for 2 h at 37°C to remove fibroblasts and afterward cultured in maintenance medium (DME:M199, 4:1; Sigma-Aldrich) supplemented with 15% fetal bovine serum and 0.1 mM bromodeoxyuridine (Sigma-Aldrich).

HUES9 human embryonic stem cells from Harvard Stem cells institute were differentiated into cardiomyocytes using a small molecule-derived approach (Hoes et al., 2018; Bomer et al., 2020). Briefly, HUES9 cells were maintained in Essential 8 medium (A1517001; Thermo Fisher Scientific, Waltham, MA, United States) before differentiation to cardiomyocytes was initiated, which was achieved by culturing HUES9 cells in RPMI1640 medium (21875-034, Life Technologies, Carlsbad, CA, United States) supplemented with 1x B27 minus insulin (Life Technologies, Carlsbad, CA, United States) and 6 μmol/L CHIR99021 (13122, Cayman Chemical, Ann Arbor, MI, United States). After 2 days, medium was refreshed with RPMI1640 supplemented with 1x B27 minus insulin and 2 μmol/L Wnt-C59 (5148, Tocris Bioscience, Bristol, United Kingdom). After 2 days, medium was changed to CDM3 medium (RPMI1640 plus 213 μg/mL ascorbic acid 2-phosphate plus 500 μg/mL recombinant human albumin). On day 8 after induction of differentiation, spontaneously contracting cardiomyocytes were observed, which were subsequently purified by replacing the medium to glucose-free CDM3 medium supplemented with 5 mmol/L sodium DL-lactate (CDM3L; Sigma-Aldrich). Ultimately, this resulted in >99% pure spontaneously beating cardiomyocyte cultures. The hES-CMs were suitable for experimentation ∼30 days after the differentiation protocol started.

Cardiomyocytes were stimulated with 10 nM recombinant human IGF-1 + 0.1% BSA (Thermo Fisher Scientific, Waltham, MA, United States) or 0.1% BSA in CDM3 or maintenance medium (hES-CMs or NRVMs respectively) between 15 min and 48 h depending on the experimental setup.

HEK-293T cells were cultured at 37°C, and 5% CO₂ in Dulbecco modified Eagle medium (DMEM; Thermo Fisher Scientific) supplemented with 10% fetal calf serum (Sigma-Aldrich). At 70% confluence, the cells were transfected with Fugene HD (Promega, Madison, WI, United States) together with pCMV∆8.91-transfer plasmid, VSV-G-packaging plasmid, and pLKO.1-plasmid expressing the genetically encoded mitochondrial Ca²⁺ indicator Mitycam provided by Dr. Adam Cohen (pMOS028, Addgene, plasmid #163046). The medium was replaced with CDM3 after 24 h and the supernatant containing the viral particles was harvested and filtered with a 0.45 nm filter after 48 h hES-CMs were incubated with CDM3 supplemented with 40% clean viral supernatant for 24 h. The next day, the medium was replaced by CMD3. After 48 h, cells were available for experiments.

Cardiomyocytes were lysed in RIPA buffer supplemented with protease inhibitors (Roche, Basile, Switzerland), phosphatase inhibitors (Sigma-Aldrich), sodium orthovanadate (Sigma-Aldrich), and phenylmethylsulphonyl fluoride (Roche). Subsequently, cells were centrifuged at 12,000 g for 10 min at 4°C. The supernatant was recovered, and the isolated protein was quantified using the BCA protein assay (Thermo Fisher Scientific). The protein extract was cooked at 95°C for 5 min in SDS-PAGE loading buffer (Thermo Fisher Scientific) and stored at −20°C.

Protein electrophoresis was carried out by standard methods. Briefly, SDS PAGE was loaded with 10 µg of protein extract. Proteins were separated by electrophoresis at 80 mV in running buffer and electro-transferred to a PVDF membrane at 0.45 A for 90 min in transference buffer. Primary antibodies were diluted in blocking buffer (TBS, 0.1% Tween, and 5% bovine serum albumin) in a 1:1,000 ratio and incubated overnight at 4°C. Membranes were washed in TBS-0.1% tween and incubated in blocking buffer with the secondary antibody (anti-mouse or anti-rabbit, Sigma-Aldrich) for 1 h in a 1:5,000 ratio. REVERT™ Total protein staining was used as loading control (LI-COR Lincoln, NE, United States). Western blot detection was carried out in a chemiluminescence detector by standard methods.

Cells were lysed with TRIzol^® (Invitrogen, Carlsbad, CA, United States) following the manufacturer’s instructions. The mRNA was quantified using a NanoDrop 2000 machine (Thermo Fisher Scientific). The cDNA was synthesized with the Quantitect RT kit (Qiagen N.V., Hilden, Germany) following the manufacturer’s instructions. All primers were designed in Primer-Blast software (NCBI, Bethesda, MD, United States) and internally validated. qPCR was carried out using the SYBR^® Green Master Mix (Bio-Rad, Hercules, CA, United States) in the thermocycler CFX384 Touch Real-Time PCR Detection System (Bio-Rad).

Cardiomyocytes were incubated with 20 nM tetramethylrhodamine ethyl ester (TMRE⁺, Invitrogen; excitation/emission spectra: 553/577) in Tyrode medium at 37°C for 30 min. Cardiomyocytes were mounted in the Olympus IX-71 inverted microscope DeltaVision Elite. Baseline fluorescence was recorded for ∼200 s and subsequently, exposed to 10 mM Carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP) for ∼1,000 s. The noise signal was subtracted. TMRE⁺ fluorescence was expressed as F—F_min (relative fluorescence unit, RFU).

hES-CMs were seeded in Seahorse assay plates at a density of 0.25 x 10⁵ cells/well. 24 h before the experiment, cardiomyocytes were stimulated with IGF-1 or IGF-1 + 10 µM ru360 (Merck, Darmstadt, Germany). Oxygen consumption rate was determined by the Seahorse XF Cell Mito Stress Test programmed in the extracellular metabolic flux analyzer Agilent Seahorse XF96 (Agilent Technologies, Santa Clara, CA, United States). 20 min before the assay, CDM3 medium was replaced by Seahorse assay medium XF (Agilent Technologies) supplemented with 10 mM glucose and 1 mM sodium pyruvate at 37°C without CO₂ control. The standard stress protocol was performed (Hoes et al., 2018), and the oxygen consumption rate was normalized by protein content determined by BCA protein assay.

Cell viability was determined using the Real-Time-Glo™ MT Cell viability assay Kit (Promega, Madison, WI, United States) following the manufacturer’s instructions. Briefly, the NanoLuc^® enzyme and the viability substrate were 1x diluted and equilibrated at 37°C in CDM3 medium or CDM3 medium +10 µM ru360 (experimental media). hES-CMs were seeded in a clear bottom 96 well plate. CDM3 medium was replaced by the corresponding experimental medium and the luminescence was recorded from 0 to 48 h in the Synergy H1 plate reader (BioTek, Winooski, VT, United States). The signal was corrected by the protein content in each well.

NRVMs were seeded in gelatin-coated coverslips in a maintenance medium and stimulated with IGF-1 for 48 h. Subsequently, NRVMs were fixed in PBS + 4% paraformaldehyde for 15 min. Afterward, cells were incubated with rhodamine-phalloidin (1:500, Thermo Fisher Scientific) and Hoescht (1:1,000, Thermo Fisher Scientific) for 1 h at room temperature. Coverslips were mounted in slides with DAKO mounting medium (Agilent Technologies). The images were acquired using the Zeiss LSM 700 laser scanning confocal microscope. The relative cell area was quantified using the software Fiji. 50 to 100 cells were analyzed per condition.

All the results were expressed as mean ± SE from at least three independent assays. For values with normal distribution and equal variances was used t-test or one-way ANOVA was followed by the Tukey post-hoc test for multiple comparisons. For comparing groups without normal distribution was used U Mann-Whitney test or the Kruskal–Wallis test followed by the Dunn post-hoc test for multiple comparisons. A p-value < 0.05 was considered for statistical differences. The software GraphPad Prism 9 (GraphPad Software Inc., San Diego, CA, United States) was used for data analysis and visualization.

To study mitochondrial Ca²⁺ handling under conditions of physiological hypertrophy, we studied Ca²⁺ uptake kinetics of NRVMs in response to IGF-1 stimulation. We first corroborated the activation of IGF-1 receptor-dependent signaling in these cells and observed a time-dependent increase in AKT phosphorylation (Supplementary Figure S1). IGF-1 also promoted a clear hypertrophic response in NRVMs (Supplementary Figure S2). To assess the effect of IGF-1 on mitochondrial Ca²⁺ uptake we made use of the Ca²⁺ indicator Rhod2, restricting the fluorescence signal into the mitochondria by a permeabilization step with saponin (Figure 1A). IGF-1-treated NRVMs significantly improved their Ca²⁺ uptake capacity after an extracellular pulse of 2 µM free Ca²⁺ (Figures 1B, C), observed as higher maximal fluorescence and area under the curve than the control condition (Figure 1D). As a control, we blocked mitochondrial Ca²⁺ uptake with the MCU inhibitor ruthenium red (RuRed), confirming the role of the MCUC on the observed effect (Figures 1C, D). The rise in mitochondrial Ca²⁺ concentration activates the pyruvate dehydrogenase (PDH) phosphatase, which dephosphorylates and activates the PDH complex (Denton, 2009). We found that IGF-1 stimulation significantly reduced the PDH phosphorylation by 0.55 ± 0.04-fold over the control condition (Figure 1E).

To generalize the results found in NRVMs, we replicated our findings in hES-CMs. Similar to NRVMs, IGF-1 induced AKT phosphorylation in these cells (Supplementary Figure S1). To measure the mitochondrial Ca²⁺ uptake, hES-CMs were infected with a lentiviral vector containing the genetically encoded Ca²⁺ indicator Mitycam, whose expression is restricted to mitochondria. As shown in Figure 1F, the Mitycam signal has an almost identical fluorescence pattern to TMRE⁺, a potentiometric dye that accumulates within the mitochondrial matrix. We promoted Ca²⁺ release from the sarcoplasmic reticulum by using electric field stimulation in presence of β-adrenergic agonist isoproterenol (Iso). We observed a drop in Mitycam fluorescence during electric stimulation, which was more pronounced at the end of the protocol in the cells treated with IGF-1 (Figures 1G–I), suggesting a greater capacity for Ca²⁺ uptake. As well as in NRVMs, IGF-1 stimulation significantly reduced the PDH phosphorylation (∼43% y ∼41% at 6 and 24 h of incubation, respectively) compared to non-treated cells (Figure 1J), indicative of a higher PDH activity.

Overall, these data suggest that IGF-1 enhances mitochondrial Ca²⁺ handling and pyruvate metabolism in both cellular models, indicating a conserved response in both species.

The limiting pathway of mitochondrial Ca²⁺ influx is the MCUC. Its function is mainly regulated by changes in the expression of the MCUC subunits affecting its Ca²⁺ sensitivity, gating, or the stability of the complex (Figure 2A) (Feno et al., 2021). In hES-CMs, IGF-1 did not change either MCU mRNA (Figure 2B) or protein levels (Figure 2C) at any of the evaluated time points. However, compatible with a higher MCU conductance, we observed a significant reduction of MICU1 mRNA levels at 24 and 48 h (0.59 ± 0.05 and 0.61 ± 0.04-fold over control conditions, respectively) and increased mRNA levels of MCUR1 at 6 h (1.41 ± 0.1-fold over control condition) (Figure 2D).

Next to MCUC composition, Ca²⁺ transport also depends on an electric gradient between the inter-membrane space and the mitochondrial matrix, known as mitochondrial membrane potential (∆Ψ_mt). This is the main driving force for Ca²⁺ influx through the MCUC (Figure 2A) (Nicholls and Crompton, 1980). Therefore, an increment of ∆Ψ_mt will promote a higher MCU-mediated Ca²⁺ transport. To test this hypothesis, we determined the basal ∆Ψ_mt in TMRE⁺-loaded cardiomyocytes and found that, indeed, IGF-1 increased the basal ∆Ψ_mt by 1.5 ± 0.09-fold over the control condition (Figure 2E). Altogether, these results indicate that IGF-1 enhances mitochondrial Ca²⁺ handling by a mechanism that combines changes in the MCUC and the driving force for mitochondrial Ca²⁺ influx.

The elevation of mitochondrial Ca²⁺ concentrations promotes the activity of PDH and the TCA cycle, indirectly increasing oxygen consumption rate (OCR) and ATP synthesis (Williams et al., 2015). Based on this premise, we investigated the effect of IGF-1 on the OCR of hES-CMs, using an extracellular metabolic flux analyzer. As expected, IGF-1 increased basal (1.37 ± 0.10-fold), ATP-linked (1.45 ± 0.14-fold), and maximal respiration (1.29 ± 0.09-fold), including the proton leak (1.71 ± 0.12-fold) (Figures 3A, B). To confirm whether the mitochondrial Ca²⁺ could play a role in the IGF-1-mediated mitochondrial respiration, we evaluated the OCR in hES-CMs co-incubated with the specific MCU inhibitor ru360. Remarkably, ru360 completely prevented the changes in OCR induced by IGF-1, showing similar values as the control condition (Figures 3A–C), except for the proton leak, which remained high (1.48 ± 0.15-fold over the control condition). It is important to note that in all evaluated time points, ru360 did not reduce cell viability (Supplementary Figure S3). Altogether, these results suggest that IGF-1 increases mitochondrial respiration by a mechanism dependent on MCU-mediated Ca²⁺ uptake.