Eight-week-old male C57BL/6 mice were purchased from the Laboratory Animal Center, Chinese PLA General Hospital. The mice were housed in a temperature-controlled room under 12/12-h light/dark cycle and had free access to natural-ingredient food and water. CRS-3 model was established as previously described (Kelly, 2003; Sumida et al., 2015). In brief, AKI was induced through renal ischemia–reperfusion injury (IRI) through 35-min bilateral renal artery ischemia and 24- or 72-h reperfusion. To inhibit the activation of Grb2, a specific antagonist (ProbeChem, TB03, 1 mg/kg, i.p.) was used 30 min before renal IRI surgery. Heart tissues, blood, and kidney were collected after 24 or 72 h of renal IRI surgery (Kidder, 2014). Then, kidneys were immediately fixed in 4% paraformaldehyde for 24 h after IRI surgery.

Paraffin-fixed sections were stained with hematoxylin and eosin (H&E) or periodic acid–Schiff, and pathologic examinations were performed by interpreters who were blind to the patients’ identities. We examined pathologic characteristics including cell necrosis, loss of brush borders, cast formation, and tubular dilatation. We scored tubular injury semiquantitatively using a scale of –5, where 0, 1, 2, 3, 4, and 5 represent normal findings, ≤10, 11–25, 25–50, 50–75%, and >75%, respectively, based on previous studies (Cui et al., 2015; Wang et al., 2015). There were at least 10 random non-overlapping fields per animal for scoring.

Echocardiographic measurements were obtained as recommended by the American Society of Echocardiography and using echocardiography (Zhou et al., 2020). Briefly, the mice were lightly anesthetized with 1% pentobarbital until the heart rate stabilized at 400–500 bpm; then, both conventional two-dimensional images and M-mode images of the heart were acquired in a parasternal short-axis view. Vevo Analysis software was used to calculate left ventricular (LV) ejection fraction (EF), and fractional shortening (FS). All assessments were performed in a blinded manner (Zhou et al., 2018c).

Primary cultures of mouse ventricular cardiomyocytes were prepared from the ventricles of treated mouse as described previously (Jin et al., 2018). The cardiomyocytes were plated on laminin-coated glass cover or culture plates and incubated with Dulbecco’s modified Eagle’s medium supplemented with 20% fetal bovine serum and antibiotics. To investigate the roles of IL-6 and Grb2 in cardiomyocyte damage and mitochondrial bioenergetics impairment, HL1 cells were pretreated with recombinant mouse growth factor receptor-bound protein 2 (Grb2), IL-6 (biogot, 10 ng/ml, 24 h).

Mitochondrial membrane potential was stained using JC-1 (5 nM, 30 min; excitation/emission 543/560) (Jin et al., 2018). Afterward, the cells underwent trypsinization, and fluorescence was assessed under fluorescence microscopy using a Zeiss LSM-5, Pascal 5 Axiovert 200 microscope, using LSM 5 3.2 image capture and analysis software, with a minimum of 10 high-magnification fields (×40) per section and three to four thin sections per group. CCCP (50 μM) and oligomycin (10 μM) for 20 min were used as positive and negative controls for the mitochondrial membrane potential (ΔΨm) measurements (Li et al., 2018).

Cardiomyocytes were incubated for 30 min with Mitotracker Red FM (400 nM). Confocal image stacks were captured with a Zeiss LSM-5, Pascal 5 Axiovert 200 microscope, using LSM 5 3.2 image capture and analysis software and a Plan-Apochromat × 63/1.4 oil DIC objective as previously described (Zhou et al., 2020). The images were deconvolved with ImageJ. The average length of the mitochondria and the ratio of fragmented mitochondria were calculated as previously described (Zhou et al., 2018d).

Cells or isolated heart tissues were washed twice in cold phosphate-buffered saline and lysed in radioimmunoprecipitation assay lysis buffer with phenylmethylsulfonyl fluoride. Protein concentrations for the whole-cell lysates were determined via bicinchoninic acid assay, and equal amounts of each protein sample (25–30 μg) were separated on 8–14% sodium dodecyl sulfate–polyacrylamide gel at 100 V; then, a Turbo Transfer System (Bio-Rad, United States) was used for 7 min to transfer the separated proteins to a polyvinylidene difluoride membrane, and the proteins were blocked for 1 h at room temperature. The membranes were incubated with primary antibodies overnight at 4°C, washed three times with Tris-buffered saline–Tween 20 (TBST), incubated for 1 h at room temperature with a horseradish peroxidase-conjugated secondary antibody, and washed with TBST; then, the protein bands were visualized with ECL Plus as directed by the manufacturer’s instructions and developed on film. The antibodies used in our study were as follows: p-mTOR (CST, 1:1,000, #5536S), mTOR (CST, 1:1,000, #2983), p-AKT (CST, 1:1,000, #4060S), AKT (CST, 1:1,000, #4691S), Grb2 (abcam, 1:1,000, #ab32037), t-Drp1 (CST, 1:1,000, #5391), p-Drp1 (Thermo Scientific Fisher, 1:1,000, #PA5-64821), β-actin (proteintech, 1:10,000, #66009-1-Ig), and GAPDH (proteintech, 1:10,000, #66004-1-Ig).

We used a blood collection tube to take samples of whole blood and let them stand at room temperature. The remaining steps were the same as those for the detection of BNP (RayBio, United States, EIAM-BNP), Troponin T (Signalway Antibody, United States, EK3212), and IL-6 (Jonln, China, JL20268) (Zhou et al., 2018b). ATP production was determined through the Enhanced ATP Assay Kit (Beyotime, China, Cat. No: S0027) according to the manufacturer’s protocol (Wang et al., 2020b).

Digested peptide mixtures were analyzed on an Orbitrap Fusion Lumos (Thermo Fisher Scientific) mass spectrometer interfaced with an Easy-nLC 1000 nanoflow liquid chromatography system (Thermo Fisher Scientific) with nano-spray ionization in positive ion polarity. Samples were dissolved with 50 μl of solvent A (0.1% formic acid in water), and 5 μl was loaded to a homemade trap column (100 μm × 2 cm) packed with C18 reverse-phase resin (particle size, 3 μm; pore size, 120 Å; SunChrom, United States) at a maximum pressure of 280 bar with 12 μl of solvent A and then separated on a 150 μm × 15 cm silica microcolumn (homemade, particle size, 1.9 μm; pore size, 120 Å; SunChrom, United States) with a gradient of 7–32% mobile phase B (100% acetonitrile and 0.1% formic acid) at a flow rate of 600 nl/min for 60 min. The MS analysis was performed in a data-dependent manner with full scans (m/z 350–1,500).

The potential functions of differentially expressed genes (DEGs) were analyzed by Database for Annotation, Visualization and Integrated Discovery 6.8. The Gene Ontology (GO) terms were considered to be significantly enriched when count in each term was >10 and p < 0.05. The related pathways of DEGs were identified using Kyoto Encyclopedia of Genes and Genomes (KEGG) and would be filtered based on the following criterion: p < 0.05.

We collected peripheral blood and centrifuged the specimens, storing serums at –80°C, followed by protein extraction. We measured 40 inflammation cytokines quantified by the Mouse Inflammation Array Q1 (RayBiotech, United States, QAM-INF-1) according to product instructions. A strict cutoff value is a fold change higher than 1.5 or lower than 0.67, with a P-value lower than 0.05. We analyzed proteomic results and visualized them with a heat map using Cluster and TreeView software.

Data distribution was evaluated with Shapiro–Wilk normality test. For normally distributed data, values are presented as mean ± standard error of the mean, and comparisons between two groups were performed with two-tailed Student’s t-test; ANOVA or repeated ANOVA and Tukey’s post hoc test were used for multiple comparisons or repeated measurements. Statistical analyses were performed with SPSS software (version 20.0), and P < 0.05 was considered statistically significant.

The mouse CRS-3 model was established as previously described (Sumida et al., 2015). The upregulation of BUN and Scr (Figures 1C,D) as well as damaged renal tubules (Figures 1A,B) indicated the development of AKI. Then, blood samples and heart tissues were isolated for 24 and 72 h after AKI. There was no significant difference in the heart tissue of mice at 24 h or 72 h after renal IRI. However, the swelling of cardiomyocytes appeared at 72 h after renal IRI (Figure 1E). To evaluate cardiac damage, myocardial injury markers such as blood BNP and Troponin T were determined through ELISA. Following AKI, the levels of BNP and Troponin T were significantly elevated when compared to that in the sham group (Figures 1F,G). Besides that, cardiac contractile function was detected with the help of echocardiography analysis. As shown in Table 1, compared to the sham group, the left ventricular systolic function was not altered after AKI, whereas cardiac diastolic function, as assessed by left ventricular mass (LVM) and E/A value, was interestingly significantly impaired after AKI. These data indicate that AKI appears to primarily blunt cardiac diastolic function instead of contraction index.

At the molecular level, cardiac contraction/relaxation mainly controlled the content of ATP in the myocardium. Interestingly, the levels of myocardial ATP were decreased to a greater degree in mice subjected to AKI than in sham-operated control mice, suggesting that cardiomyocyte ATP metabolism is disrupted as a result of AKI (Figure 2A). Cardiomyocyte ATP production is mainly generated by mitochondria which metabolize glucose into ATP, carbon dioxide, and water, a process that is called oxidative phosphorylation.

Given the indispensable role of mitochondria in regulating myocardial ATP supply, we question whether AKI is followed by a disruption of mitochondrial function. To achieve this aim, cardiomyocytes were freshly isolated from AKI-operated or sham mice. Mitochondrial membrane potential reduction is an early sign of cardiomyocyte metabolism disturbance. Thereby, JC-1 probe was applied to stain mitochondrial membrane potential in cardiomyocytes. Compared to that in cardiomyocytes isolated from sham mice, mitochondrial membrane potential was progressively repressed in cardiomyocytes isolated from AKI-operated mice (Figures 2B,C). Besides that, as a by-product of mitochondrial metabolism, the content of intracellular reactive oxygen species, as assessed by DCFH-DA probe, was drastically increased in cardiomyocytes isolated from AKI-treated mice when compared to that in cardiomyocytes from the sham group (Figures 2D,E).

Based on recent studies, mitochondrial morphology shift acts upstream of mitochondrial metabolism disruption. Excessive mitochondrial fragmentation formation promotes the dissipation of mitochondrial membrane potential, resulting into impaired oxidative phosphorylation as well as decreased ATP production. With the assistance of immunofluorescence using a mitochondria-specific probe, we observed a cleavage of mitochondria from an elongated network into small spheres or short rods in cardiomyocytes isolated from AKI-treated mice (Figures 2F,G). This morphological shift may be a result of increased mitochondrial fission because western blot analysis demonstrated that Drp1 phosphorylation, a classical marker of mitochondrial division, was significantly elevated in cardiomyocytes isolated from AKI-treated heart (Figures 2H,I). Taken together, our results confirmed that mitochondrial metabolism damage as well as mitochondrial morphology disturbance would be the intracellular molecular events underlying AKI-mediated myocardial damage.

To better clarify the molecular mechanism that is activated by AKI and participates into acute myocardial damage, we performed proteomic analysis through label-free technology to understand without bias protein expression in the heart following AKI. The results demonstrated that 448 proteins (p < 0.05, log2 fold change) were upregulated or downregulated in heart tissue (Figures 3A,B). KEGG analysis demonstrated that pyruvate metabolism, glyoxylate and dicarboxylate metabolism, starch and sucrose metabolism, and biosynthesis of amino acids were altered in the heart after AKI (Figure 3C). GO cellular component enrichment analysis demonstrated that 23 of the 448 proteins were enriched to mitochondria (Figure 3D). A biological process description of GO further illustrated that 91 of the 448 proteins were involved in intracellular metabolic events (such as cellular amino acid metabolic process, oxidative stress, small molecular catabolic process, and coenzyme metabolism process) (Figure 3D). These data suggest that cardiomyocyte metabolism or mitochondrial bioenergetics may be affected by AKI. With the help of STRING protein–protein interaction network analysis, we found that Grb2 is involved in the cross-talk and signal integration of various molecular events in the heart following AKI (Figure 3E). In fact, previous studies have also reported the key role played by Grb2 in regulating cancer metabolism and heart fibrosis (Zhang et al., 2003; Sun et al., 2019). Given these observations, we propose that Grb2 would be a potential pathological factor involving AKI-related myocardial damage. Western blots further showed that Grb2 was slightly increased 24 h after AKI and reached to statistical difference 72 h after AKI (Figure 3F). Taken together, myocardial Grb2 expression is significantly elevated by AKI, and this alteration may be a potential molecular mechanism underlying AKI-related myocardial damage.

To delineate the pathogenic mechanisms by which increased Grb2 promotes cardiac dysfunction, a specific antagonist of Grb2 (TB03) was injected into mice 30 min before renal ischemia–reperfusion surgery to fully block the activation of Grb2 during or after AKI. Besides that, Grb2 was mainly upregulated 72 h after AKI, and therefore this time point was used in the following functional assays. In AKI-operated mice, supplementation of TB03 largely reduced the levels of Troponin T and BNP (Figures 4A,B), suggesting that Grb2 inhibition was associated with cardioprotection after AKI. Besides that, left ventricular diastolic function, which was impaired by AKI, could be normalized by TB03 as evidenced by increased LVM and E/A value (Table 2). As expected, total ATP, which was repressed by AKI, could be increased to near-normal levels once TB03 was injected (Figure 4C).

Cardiomyocytes were also isolated in vitro from TB03-injected mice to observe mitochondrial function. AKI-mediated cardiomyocyte mitochondria cleavage or division could be reversed by TB03 (Figures 4D,E), suggesting that cardiomyocyte mitochondrial morphology could be normalized by Grb2 inhibition during AKI. Overall, these verification experiments uncover AKI-mediated cardiac dysfunction; cardiomyocyte metabolism disruption and mitochondrial dysfunction are induced by Grb2 upregulation.

Three classical signaling pathways are under the control of Grb2 based on recent studies, including MAPK/ERK, MAPK/JNK, and PI3K/Akt (Zhang et al., 2003; Gui et al., 2012). Interestingly, Akt/mTOR pathway has been regarded as a key regulator of mitochondrial metabolism (Gan et al., 2016; Pan et al., 2018). Based on these findings, we asked whether Akt/mTOR pathway is the intracellular second messenger functioning downstream of Grb2 and thus affects mitochondrial metabolism in cardiomyocytes following AKI. To test our hypothesis, western blots were performed to analyze the alterations of Akt/mTOR pathway. As shown in Figures 5A–C, compared to the sham group, the expression of p-Akt was significantly downregulated in hearts from AKI-treated mice, an alteration that was followed by a decline in the expression of p-mTOR. Besides that, administration of TB03 significantly upregulated the levels of p-Akt and p-mTOR in hearts from AKI-treated mice. These data elucidate a possible link between Grb2 activation and Akt/mTOR pathway suppression in the context of AKI-mediated myocardial damage.

To better clarify the relationship between Grb2 and Akt/mTOR pathway, gain-of-function assay of Grb2 was performed in normal HL1 cardiomyocyte cell line using the recombinant mouse growth factor receptor-bound protein 2. Cardiomyocyte cultured with recombinant mouse growth factor receptor-bound protein 2 expressed less p-Akt and decreased p-mTOR when compared with the control group (Figures 5D–F). Furthermore, cardiomyocyte ATP production was also suppressed by recombinant mouse growth factor receptor-bound protein 2 (Figure 5G). In contrast, in TB03-pretreated cardiomyocyte, recombinant mouse growth factor receptor-bound protein 2 failed to repress the expression of p-Akt and p-mTOR (Figures 5D–F). Similarly, ATP production was also maintained in TB03-pretreated cardiomyocyte despite the administration of recombinant mouse growth factor receptor-bound protein 2 (Figure 5G). Overall, our results identify the Akt/mTOR pathway as one potential intracellular signal transducer of Grb2 upregulation in the pathogenesis of AKI-related cardiac damage.

The next experiments were performed to figure out the upstream regulator of Grb2 in AKI-mediated myocardial damage. Circulating inflammation cytokines have been reported to be the “connector” that is induced by AKI and acts as a contributor in triggering myocardial damage (Virzi et al., 2018; Clementi et al., 2019). Given that, we asked whether inflammation cytokines may be the upstream mediator of myocardial Grb2 upregulation after AKI. The Mouse Inflammation Array Q1 results demonstrate that serum cytokines were primarily altered at 24 h after AKI. Compared to the sham group, there were 10 cytokines significantly upregulated 24 h after AKI. The quantitative results demonstrated that IL-6, BLC, and TIMP-1 were the most pronounced cytokines upregulated by AKI (Figures 6A,B). Interestingly, IL-6 has been acknowledged as a critical inflammation parameter to evaluate the severity of AKI (Zhang et al., 2015; Andres-Hernando et al., 2017). More importantly, the causal role played by IL-6 in inducing Grb2 upregulation has been widely discussed (Lee et al., 1997; Bongartz et al., 2019). Therefore, AKI-mediated IL-6 may be a potential mechanism for Grb2 upregulation. In our study, ELISA assay further demonstrated that the concentration of IL-6 was up to approximately fourfold higher than those observed in the sham group (Figure 6C). Taken together, inflammation response, especially circulating IL-6 expression upregulation, may be an upstream signal for Grb2 upregulation.

Further supporting the inflammation cytokine array findings, cardiomyocytes were treated with exogenous IL-6 to observe the alterations of Grb2, Akt/mTOR pathway, and cardiomyocyte bioenergetics. After administration of IL-6, the expression of Grb2 was significantly elevated in cardiomyocyte, an effect that was accompanied with a drop in the levels of p-Akt and p-mTOR (Figures 7A–D). Besides that, cardiomyocyte total ATP production was inhibited (Figure 7E), mitochondrial membrane potential was repressed (Figures 7F,G), and mitochondrial fission was activated in response to IL-6 treatment (Figures 7H,I). The above-mentioned data validated that IL-6 could be an upstream inducer of AKI-related cardiomyocyte damage through upregulation of Grb2, inhibition of Akt/mTOR pathway, and suppression of ATP production. Interestingly, in si-Grb2-pretreated cardiomyocyte, IL-6 supplementation failed to affect Grb2 expression, Akt/mTOR pathway activation, ATP production, and mitochondrial morphology. Taken together, the deleterious action of AKI on cardiac dysfunction is working through IL-6 upregulation-mediated Grb2 activation, Akt/mTOR pathway inhibition, and cardiomyocyte bioenergetics impairment (Figure 8).