A total of 50 patients aged ≥20 years and diagnosed with HF post-infarction without renal replacement therapy were enrolled between June and December, 2022 at the Department of Cardiology of the Qujing First People's Hospital, Qujing, China. Age- and sex-matched healthy adults from the outpatient department not diagnosed with cardiovascular diseases were included as the control group. Fasting serum samples were collected and stored at -80°C. The Ethics Review Board of Qujing First People's Hospital approved the study protocols (approval no. QJFH2021-017). All participants provided written informed consent prior to enrollment.

A total of 100 healthy male Sprague-Dawley rats (weighing 220-250 g; 8 weeks old) were purchased from the Animal Center of Kunming Medical University and kept at ~22°C and a 12-h light-dark cycle with free access to food and water. The study protocols were in accordance with the Guide for the Care and Use of Laboratory Animals and approved by the Animal Care and Use Committee of Kunming Medical University (Kunming, China). All relevant animal welfare considerations were taken, including efforts to minimize suffering and distress, the use of analgesics or anesthetics or special housing conditions, particularly for the AMI and UTMD procedures.

The left anterior descending coronary artery ligation was performed to induce AMI and establish a model of MI in rats, as previously described (16). An intraperitoneal injection of ML385 (30 mg/kg, HY-100523, MedChemExpress) was administered to inhibit nuclear factor erythroid 2-related factor 2 (Nrf2) activity, as previously described (17). All animals were randomly divided into the following experimental groups: The sham-operated (sham) group (n=8), the AMI group (n=36), the GFP + UTMD group (n=5), the AMI + FGF21 group (n=12), the AMI + UTMD@Vector group (n=3), the AMI + UTMD@ KLB group (n=12), the AMI + UTMD@KLB + FGF21 (n=12) group and the AMI + UTMD@KLB + FGF21 + ML385 group (n=12). During the surgery, 5, 3, 2, 3 and 3 rats died in the AMI, AMI + FGF21, AMI + UTMD@KLB, AMI + UTMD@ KLB + FGF21 and AMI + UTMD@KLB + FGF21 + ML385 groups, respectively, due to sudden cardiac death and severe heart failure. The mortality rate as was expected according to previous studies using rats following left anterior descending coronary artery (LAD) occlusion (18-20). Animal health and behavior were monitored every day for the first week, then once every 2 days. The pre-designed humane endpoints in the present study mainly were severe heart failure, loss of appetite and fragility for >24 h and an agonal stage. A total of 3 rats reached the humane endpoints and were therefore euthanized before the end of the study and were thus recorded as deaths due to heart failure.

In brief, the rats were placed in the supine position for endotracheal intubation and anesthetized by the inhalation of 2-3% isoflurane (induction, 3%; maintenance, 2%) using a small animal anesthesia machine (Harvard Apparatus). All animals received buprenorphine hydrochloride (0.1 mg/kg, subcutaneous injection) for pain management when establishing the model of MI. The surgical procedure involved a left thoracotomy to access the heart. A 5-0 Prolene suture with a needle was used to ligate the LAD. The ligation was performed ~2 mm below the left atrial appendage. The model of MI model was validated by observing distinct changes in the heart anatomy, and the whitening of the left ventricular anterior wall and apex. The rats in the sham group were only subjected to drilling underneath the LAD coronary artery but were not ligated. The rats were administered an intraperitoneal injection of the vehicle (0.9% saline, ST341, Beyotime Institute of Biotechnology) or FGF21 at 0.5 mg/kg/day. Recombinant human FGF21 (CSB-AP002471HU, Cusabio Technology, LLC) was expressed in E. coli and harvested by ion exchange and hydrophobic interaction chromatography, as previously described (21). In accordance with the AVMA Guidelines for the Euthanasia of Animals, the rats were euthanized by cervical dislocation under anesthetization with isoflurane inhalation (induction, 3%; maintenance, 2%). Animal death was verified by respiratory cessation and muscular tension disappearance. Subsequently, ~3 ml blood was rapidly obtained from the heart cavity to extract serum for use in enzyme-linked immunosorbent assay (ELISA). The heart tissues were then harvested for use in subsequent experiments. The duration of the animal experiment was 4 weeks, including the establishment of the model of AMI, the UTMD procedure and subsequent observation until euthanasia.

CMBs were synthesized using the thin-film hydration-sonic vibration method, as previously described (22). A mixture of chemicals, including dipalmitoyl-phosphatidylcholine (cat. no. 850355P-1G-A-3214, Avanti Polar Lipids, Inc.), 1,2-distearoyl-sn-glycerol-3-phosphoethanolamine-N-maleimide (DSPE-PEG, cat. no. 880160P-200MG-A-047, Avanti Polar Lipids, Inc.) and 3β-[N-(N',N'-dimethyl-aminoethane)-carbamoyl] cholesterol (DC-Chol, cat. no. 700001P-200MG-B-021, Avanti Polar Lipids, Inc.), was prepared and processed to form a thin lipid film (mass ratio, 5:2:0.5) (10). Glycerol solution (cat. no. 56-81-5, MillliporeSigma) was added, and octafluoropropane (C3F8, Foshan Huate Gas Co. Ltd.) gas was bubbled through the solution to generate gas-filled CMBs, widely used in clinical practice with favorable safety and performance (23). The CMBs were diluted to 1×10⁹/ml with PBS (C0221A, Beyotime Institute of Biotechnology), treated with ⁶⁰Co-γ irradiation (BFT-II Cobalt-60 gamma-ray Irradiation Device, MDS Nordion) and stored at 4°C. The particle size distribution and surface potential were determined using the Zetasizer Nano ZS system (Malvern Panalytical Ltd.) (10).

Plasmids (80 μg, Hanbio Biotechnology Co., Ltd.) expressing the GFP or KLB gene were incubated in a 100-μl CMB suspension (0.5×10⁹ MBs) for 15 min. The suspension was centrifuged at 400 × g for 5 min under 4°C to obtain plasmid-bound CMBs (10). The percentage of bound plasmid DNA and the payload mass of plasmid DNA in CMBs were calculated as follows: Percentage of bound plasmid DNA in CMBs=(DNA_total-DNA_free)/DNA_total ×100%; payload mass of plasmid DNA in CMBs=(DNA_total-DNA_free)/CMB number (10). The plasmid DNA was stained by 10 μg/ml propidium iodide solution (C0080 Beijing Solarbio Science & Technology Co., Ltd.) for 10 min at room temperature. The combination of plasmid DNA and CMBs was validated using a fluorescence microscope (DMI3000, Leica Microsystems GmbH) and a flow cytometer (BD Biosciences).

All animals were anesthetized via isoflurane inhalation (induction, 3%; maintenance, 2%) for the UTMD procedure. In brief, the rats were administered intravenous injections of KLB@CMBs (1.0-1.2 ml/h) via the tail vein, as previously reported (10). An M3S transducer (S5-1 probe, GE Healthcare) with a low 1-5 MHz frequency range was used due to the known 1-3 MHz resonant frequency of microbubbles (10,12,24). The parameters of the second harmonic mode were used and set as transmit 1.6 MHz and receive 3.2 MHz, mechanical index (MI, low 0.30; flash, 1.31), and -16 dB output power (10). The KLB delivery to the heart was repeated three time at 1-day intervals at 1, 3, and 5 days at 1 week following the induction of AMI (25).

H9C2 rat cardiomyoblasts (purchased from The Cell Bank of Type Culture Collection of the Chinese Academy of Sciences) were cultured in DMEM containing 10% (v/v) fetal bovine serum (cat. no. 0500, ScienCell Research Laboratories, Inc.) in a humidified incubator with 5% CO₂ and at 37°C. When the cell density reached 60%, the cells were transfected with a plasmid containing the KLB gene using Lipofectamine 3000^® reagent (L3000008, Invitrogen; Thermo Fisher Scientific, Inc.) following the manufacturer's protocols. The cells were cultured with or without FGF21 (2 μg/ml, CSB-AP002471HU, Cusabio Technology, LLC) for 24 h. Subsequently they were treated under hypoxic conditions (5% CO₂ and 1% O₂ at 37°C) to mimic AMI for 4 h and collected for further analysis.

As previously described, cardiac function was evaluated using an M-mode echocardiography system (S12-4 probe, EPIQ 5; Koninklijke Philips N.V.) 4 weeks following the induction of AMI (16). The rats were lightly anesthetized via isoflurane inhalation (induction, 3%; maintenance, 1.5%) for the echocardiography assessment. Left ventricular end-systolic diameter (LVESD) and left ventricular end-systolic diameter (LVESD) were measured to calculate the left ventricular ejection fraction (LVEF) and left ventricular fractional shortening (LVFS) using computer algorithms. Each diameter was obtained from at least four consecutive cardiac cycles and averaged.

To evaluate the apoptosis of H9C2 cells or that of cells in the heart sections, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) staining (cat. no. A321, Elabscience) was used, as previously described (3). The apoptotic index was calculated by determining the ratio of TUNEL-positive nuclei to the total nuclei counts in five randomly selected fields per sample, as previously described (10).

For the assessment of ROS generation DHE staining was performed on fresh heart tissues (10). The heart sections were pre-treated with cold PBS for 10 min and exposed to 10 μmol/l DHE solution (cat. no. S0063, Beyotime Institute of Biotechnology) at 37°C in a dark environment, as previously described (10). Following a 30-min incubation at 37°C, the heart sections were subjected to triple PBS washes and were maintained in a wet condition until examination through a fluorescence microscope (Leica Microsystems GmbH).

Mitochondrial membrane potential was determined using the JC-1 staining assay kit (cat. no. C2005, Beyotime Institute of Biotechnology), as previously described (26). JC-1, a fluorescent dye, accumulated in the mitochondrial matrix in two forms: Aggregates (red fluorescence) in polarized mitochondria and monomers (green fluorescence) in depolarized mitochondria (10). The heart sections were treated with 10 μg/ml JC-1 solution at 37°C for 0.5 h in a light-protected environment, as previously described (27). Following three washes with PBS, the fluorescence intensity was observed using a fluorescence microscope (Leica Microsystems GmbH). To quantify the mitochondrial membrane potential, five random images were captured per sample and the fluorescence intensity of both the J-monomers and the J-aggregates was examined.

As previously described (28), the H9C2 cells were intervened and digested with pancreatic enzymes, and the cell suspension was then collected for analysis. The mitochondrial respiratory function was assessed using an XF24 Extracellular Flux Analyzer (Seahorse Bioscience) with a sequential titration of oligomycin, carbonyl cyanide 4-(trifluoromethoxy) phenylhydrazone (FCCP), rotenone and antimycin A (XF Cell Mito Stress Test kit, Seahorse Bioscience), as previously described (28).

Total antioxidant capacity was measured using a commercial kit based on the ferric-reducing ability of plasma method (S0116, Beyotime Institute of Biotechnology) according to the product manual. The capacity was quantified by measuring the absorbance at 593 nm using a microplate reader (Infinite F500, Tecan Group, Ltd.) and estimated as a percentage of the combined ferric reducing/antioxidant potency of the antioxidants in protein.

The heart tissues were fixed with 4% paraformaldehyde (cat. no. P0099, Beyotime Institute of Biotechnology) at room temperature for 72 h and subsequently sliced into sections measuring 5 μm in thickness. These sections were subjected to staining with hematoxylin and eosin [H&E; Right Tech, China] for 2 min at room temperature, and Masson's Trichrome Stain with celestine blue staining (cat. no. G1346, Beijing Solarbio Science & Technology Co., Ltd.) for 3 min at room temperature and Ponceau-acid magenta staining (cat. no. C0189, Beyotime Institute of Biotechnology) for 10 min at room temperature to observe and analyze the cardiac fibrosis morphology using microscope (Olympus IX70, Olympus Corporation) according to the manufacturer's instructions.

The MTT Cell Proliferation Assay kit (cat. no. C0009, Beyotime Institute of Biotechnology) was used to assess cell viability. The H9C2 cells (0.5×10⁴ per well) were cultured in 96-well plates and treated as needed (e.g., plasmid transfection, FGF21 treatment and hypoxia-induced stress), and were then incubated with MTT reagent for 4 h at 37°C (10). Additionally, lactate dehydrogenase (LDH) activity was measured in the serum or medium supernatant using a commercially available assay kit (cat. no. A020, Jiancheng Bioengineering Institute). A total of 50 μl diluted serum was added to 96-well plates and combined with 50 μl reaction mix [the mixed solution of reagent I and II (v/v: 5:1)] per well (10). The absorbance was then measured at 450 nm using a spectrophotometer (Evolution 60S, Thermo Fisher Scientific, Inc.), and LDH activity was calculated following the manufacturer's instructions (10).

The serum levels of creatine kinase-myocardial band (CK-MB; E-EL-R1327c, Elabscience Biotechnology Co., Ltd.), cardiac troponin T (cTnT; E-EL-R0151c Elabscience Biotechnology Co., Ltd.), KLB (ml028418, Mlbio) and FGF21 (CSB-EL008627RA, Cusabio Biotechnology Co., Ltd.) were measured according to the instructions provided with the kits.

Total RNA was extracted using TRIzol reagent (cat. no. B0201, HaiGene) and converted into cDNA using the RT Easy II First Strand cDNA Synthesis kit (cat. no. 04379012001, Roche Diagnostics), as previously described (16). Followinh incubation for 60 min at 42°C, the reactions were terminated by heating at 70°C for 10 min. qPCR was performed using the Real-Time PCR Easy (HY-K0501SYBR-Green I, MedChemExpress), with the following procedures: Denaturation at 95°C for 10 min, amplification by 40 cycles of 95°C for 10 sec and 60°C for 30 sec. Gene expression was normalized to β-actin and calculated using the 2^-ΔΔcq method (29). The primer sequences are presented in Table SI.

Total proteins were extracted from the H9C2 cells or heart tissues using RIPA buffer containing protease (P0013C, Beyotime Institute of Biotechnology) and phosphatase inhibitors (P1045, Beyotime Institute of Biotechnology). The protein concentration was determined using BCA assay. Specifically, ~30 μg proteins per lane were loaded into 10% SDS-PAGE gels and transferred onto 0.22 μm PVDF membranes (MilliporeSigma). The membranes were blocked using 5% BSA (ST025, Beyotime Institute of Biotechnology) for 1 h at room temperature, and incubated using corresponding primary antibodies overnight at 4°C and HRP-linked secondary antibody for 1 h at room temperature (all antibodies used are listed in Table SII). Due to the close molecular weights of the target proteins, the protein expression signals on different membranes were detect with the same loading amount, and all other western blotting conditions for each protein were also the same excepting the primary antibody used. The density of immunoreactivity was assessed using a chemiluminescence ECL kit using a chemiluminescence system (Tanon 5200, Tanon Science & Technology Co., Ltd.) (30). ImageJ software (version v1.52a, National Institutes of Health) was used for the semi-quantification of protein expression.

Unless stated otherwise, quantitative variables are expressed as the mean ± standard deviations (SD). Statistical analyses and data visualization were conducted using SPSS 20.0 software (IBM Corp.) and GraphPad Prism (version 6.0; Dotmatics). The Shapiro-Wilk test was used to evaluate data distribution, and P>0.05 was considered as passing the normality test. As the clinical data in Fig. 1A did not meet the normal distribution (P<0.001 in the Shapiro-Wilk test), the non-parametric Mann-Whitney U test was adopted for the comparison of two groups here. Following normality tests, a two-tailed unpaired Student's t test was conducted for the comparisons of two groups, and one-way ANOVA with Tukey's post hoc test were used for the comparisons of multiple groups. The correlation between FGF21, KLB and myocardial injury biomarkers was examined using Spearman's correlation analysis. A P-value <0.05 was considered to indicate a statistically significant difference.

The serum levels of FGF21 were measured in patients with HF following MI and in age- and sex-matched healthy participants. The serum levels of FGF21 in the patients with HF were significantly increased compared with those in the healthy participants (128 vs. 111 pg/ml, Fig. 1A). Subsequently, AMI was induced in rats to provoke HF and the FGF21 serum levels were measured. Consistent with the results obtained in the patients, the rats with HF had higher serum levels of FGF21 compared with the control rats (Fig. 1B). However, the cardiac expression of the FGF21 receptor, KLB, not α-klotho, was significantly decreased in the rats with AMI compared with the sham-operated rats (Fig. 1C and D). The successful establishment of the murine model of MI was verified by measuring the decreased LVEF and LVFS according to the echocardiography (Fig. 1E-G). Furthermore, the correlation between the severity of myocardial injury and FGF21 or KLB levels was assessed (Fig. 1H-J). Although the levels of the markers of myocardial injury, CK-MB and cTnT positively correlated with the serum levels of FGF21 (r=0.663 and 0.738, respectively), there was an inverse correlation between the cardiac KLB content and the serum levels of FGF21 (r=-0.729). These data suggest that although the FGF21 levels significantly increase following AMI, the levels of its downstream KLB co-receptor in heart tissue are markedly decreased, which may be a critical factor that decreases the cardiac response to FGF21. Hence, promoting FGF21 sensitivity may be a potential strategy with which to enhance the therapeutic efficacy of FGF21.

There is evidence to indicate that FGF21 activates intracellular protection signals against hypoxia-induced cellular injury (9). However, our previous observation suggests that the insufficiency of the cardiac KLB co-receptor may impair the pharmacological action of FGF21 (Fig. 1C and D). Further experiments were performed. First, the transfection of plasmid containing the KLB gene significantly induced KLB overexpression in H9C2 cells under normal conditions or hypoxia, as verified by measuring its mRNA and protein levels (Fig. S1). As was expected, the cellular activity of H9C2 cells decreased after 3 h under hypoxic conditions in vitro (Fig. 2A). However, FGF21 treatment or KLB overexpression alone could not substantially reverse the decreased cellular activity of H9C2 cells under hypoxic conditions. When the cells overexpressing KLB were treated with FGF21, cellular injury was reduced (Fig. 2A). The present study also assessed oxidative stress, LDH leakage, and cell death that reflect hypoxia-induced myocardial injury. The overexpression of KLB enhanced the effects of FGF21, reducing ROS generation and cell injury induced by hypoxia (Fig. 2B-D). As shown by JC-1 staining, the hypoxia-induced decrease in the mitochondrial membrane potential was partly attenuated by FGF21 treatment, and KLB overexpression significantly enhanced the protective effects of FGF21 on the mitochondria (Fig. 2E). In summary, these findings suggest that KLB overexpression optimizes the effects of FGF21 on cardiomyocytes in response to hypoxia in vitro.

CMBs were synthesized to carry plasmids expressing the KLB gene according to published procedures (10) (Fig. 3A). Dynamic light scattering revealed that he CMBs had a particle diameter of 1,482±303 nm and a zeta potential of +24.2±4.5 mV (Fig. 3B). The carrying capacity of the CMBs for the plasmids was estimated by the addition of 40 μg plasmid DNA to the CMBs, revealing the maximized carrying capacity of 0.5×10⁹/CBMs (Fig. 3C). Fluorescence microscopy revealed the stabilized KLB@CMBs with the KLB-expressing plasmid labeled by propidium iodide (Fig. 3D). According to the results of flow cytometric analysis, the binding rate of CMBs to the plasmid reached 55.2% (Fig. 3E). The synthesized KLB@ CMBs that were prepared had a good stability within at least 6 h under room temperature (Fig. S2). The distribution of CMBs was visualized in the heart with echocardiography, demonstrating that the UTMD-triggered CMBs burst and released the KLB-expressing plasmid in myocardial tissue (Fig. 3F). At 1 week after the UTMD-mediated KLB delivery, GFP fluorescence was detected in the heart tissue (Fig. 3G). These data confirm that the UTMD delivery of the KLB gene to the heart is feasible and appropriate for use in following experiments.

At 1 week after establishing the rat model of MI, the UTMD-mediated KLB delivery was repeated to the heart tissues three times. Subsequently, cardiac KLB expression was assessed, revealing that the UTMD technique significantly increased the KLB protein levels in myocardial tissue (Fig. 4A). In addition, cardiac structure and function were examined using echocardiography at 4 weeks following the induction of MI to determine whether the UTMD-mediated KLB delivery affects cardiac remodeling. Indeed, while the intravenous administration of FGF21 did not significantly improve LVEF and LVFS in the rats post-MI, the UTMD-mediated cardiac delivery of KLB significantly enhanced these parameters, amplifying the effects of FGF21 on heart dysfunction (Fig. 4B and C). Moreover, H&E staining revealed that the infarct border zone exhibited an improved tissue structure and lower injury scores in the FGF21-treated rats with KLB@CMBs delivered by UTMD (Fig. 4D and E). Masson's staining was also performed on the cardiac sections to evaluate the effects of combined treatment with FGF21 and the UTMD-mediated KLB delivery on the degree of fibrosis. The administration of FGF21 alone did not significantly reduce the infarct size, while the cardiac-specific KLB delivery partially improved collagen fiber alignment in the infarct area (Fig. 4F and G). These data indicated that the overexpression of KLB in the heart using the UTMD technique increased FGF21 sensitivity and attenuated adverse ventricular remodeling following MI.

Previous research has demonstrated that FGF21 regulates mitophagy, mitochondrial dynamics and mitochondrial metabolism to alleviate mitochondrial stress injury (4). Therefore, the present study examined whether the FGF21 regulation of mitochondrial function is dependent on KLB in myocardial tissue. As shown by DHE staining, whereas the administration of FGF21 did not significantly decrease myocardial ROS generation, the KLB gene delivery to the heart markedly reinforced the effects of FGF21 on myocardial ROS production post-infarction (Fig. 5A). In agreement with this finding, KLB expression was not affected by FGF21 administration alone (Fig. 5B). Enhanced ATP generation reflects the re-establishment of mitochondrial energy homeostasis impaired by mitochondrial dysfunction post-infarction. As was expected, treatment with FGF21 slightly recovered the ATP levels in the heart tissues, while the UTMD-mediated KLB delivery significantly increased the ATP generation post-infarction (Fig. 5C). Hence, KLB overexpression, but not FGF21 administration, recovered cardiac ATP production following AMI, and the combined interventions largely improved mitochondrial energy. On the whole, these data indicated that the UTMD-mediated cardiac delivery of KLB promoted FGF21 sensitivity, recovering mitochondrial homeostasis and alleviating adverse cardiac remodeling.

The present study then aimed to investigate how the combined FGF21 and KLB interventions mitigate oxidative stress and safeguard mitochondria post-infarction. Nrf2 antioxidant signaling is a critical protective pathway against intracellular oxidative injury (10). Thus, the present study assessed whether combined FGF21 and KLB@CMBs activate the Nrf2-mediated antioxidant pathway to improve heart dysfunction. Indeed, FGF21 administration and KLB delivery significantly sensitized the heart to FGF21 treatment and increased Nrf2 expression, which was inhibited by ML385, an NRF2 inhibitor (Fig. 6A). By contrast, the protein expression of kelch-like ECH-associated protein 1 (KEAP1), an inhibitor of Nrf2 binding, did not significantly decrease after the combined FGF21 and KLB interventions (Fig. 6A). Therefore, the combined interventions may prevent heart failure post-infarction by promoting Nrf2 expression and not reducing KEAP1 expression (Fig. 6A). As was expected, KLB delivery and FGF21 treatment increased the expression of the Nrf2 target proteins, heme oxygenase 1 (HO-1), NAD(P) H quinone dehydrogenase 1 (NQO1) and glutamate-cysteine ligase modifier subunit (GCLM) (Fig. 6B). Conversely, the Nrf2 inhibitor, ML385, significantly abolished these effects and reduced the expression of these antioxidant enzymes (Fig. 6B). Furthermore, it was verified that the ML385 injection significantly reduced the protective effects of combined FGF21 and KLB@CMB treatment on cardiac dysfunction in rats post-infarction (Fig. 6C-E). Finally, DHE staining was performed to assess superoxide anion free radicals, demonstrating that oxidative stress was moderately attenuated by KLB cardiac delivery, and the Nrf2 inhibitor compromised the antioxidant effect of KLB in the heart post-infarction (Fig. 6F and G). Moreover, the total antioxidant capacity in the heart tissues increased by the combination of FGF21 and KLB@CMB treatment was attenuated by the inhibition of Nrf2 (Fig. S3).

The present study further evaluated the effects of Nrf2 inhibition and the FGF21-KLB-mediated activation on mitochondrial quality. The combined interventions improved the loss of mitochondrial membrane potential in the myocardial tissues post-infarction, and this effect was attenuated by the inhibition of Nrf2 (Fig. 7A and B). The mitochondrial transmembrane potential reflects mitochondrial function and impaired mitochondria are the primary source of ROS production. The mitochondrial oxygen consumption rate (OCR) revealed that FGF21-KLB boosted the basal respiration level, partly reduced by Nrf2 inhibitor treatment in the H9C2 cardiomyoblasts (Fig. 7C). In addition, the synthetic compound, FCCP, uncoupled mitochondria respiration from oxidative phosphorylation, indicating a similar difference in the measurement of maximal OCR (Fig. 7C-G). These results imply that combined GFG21-KLB interventions improve mitochondrial function in H9C2 cardiomyoblast cells under hypoxic conditions, and this effect is compromised by Nrf2 inhibition.