Journal Logo

Original Articles

Cardioprotective Effect of Growth Differentiation Factor 15 Against Isoproterenol-Induced Cardiomyocyte Apoptosis via Regulation of the Mitochondrial Fusion

Zhang, Yan; Mei, Zhu; Jia, Xiaodong; Song, Haixu; Liu, Jing; Tian, Xiaoxiang

Editor(s): Xu, Tianyu; Fu, Xiaoxia

Author Information
doi: 10.1097/CD9.0000000000000051
  • Open




  • Growth differentiation factor 15 (GDF15) is a divergent member of the transforming growth factor-β superfamily. Serum GDF15 was recently observed to be remarkably increased in heart failure (HF) patients, while the effect of GDF15 on HF remained unclear. Here, we demonstrated that GDF15 is a protective factor for HF by reduced cardiomyocyte apoptosis in vivo and in vitro.


  • HF, following myocardial infarction, diabetes, hypertension, and aging, has become a leading global cause of morbidity and mortality. The level of GDF15 was found to be associated with the occurrence of HF, which might improve the apoptosis of cardiomyocytes. It suggested that the increase of GDF15 expression might be a molecular target of HF therapy.


Heart failure (HF) following myocardial infarction (MI), diabetes, hypertension, and aging has become a leading global cause of morbidity and mortality.[1] HF is characterized by pathological hypertrophy of cardiomyocytes (CMs), CM death, myocardial fibrosis, chamber dilation, and contractile dysfunction in response to pathological stimuli or genetic mutations.[2,3] These pathological processes eventually lead to reduced cardiac tissue compliance and the development of end-stage HF.[4,5]

There are very few effective therapeutic approaches to HF nowadays due to the limited ability to rescue cardiac contractili-ty.[6] Exploring the novel molecular targets of HF will be a helpful way to reduce mortality and improve patient quality of life. During HF development, an important initiated process is marked by the loss of CMs while the heart compensates for the loss of cardiac output by remodeling the myocardium.[7,8] Although the mechanisms underlying CMs loss are not fully understood, yet so far we know that apoptosis, oxidative stress, and inflammation cause changes in energy metabolism and contractile proteins.[9,10]

Growth differentiation factor 15 (GDF15), also termed macrophage inhibitory cytokine 1, is a divergent member of the transforming growth factor (TGF)-β superfamily.[11] GDF15 was originally reported as robustly expressed in the placenta and prostate tissue, while it was relatively low in other tissues.[12] Further studies revealed that GDF15 can be produced and secreted by many various cell types like macrophages, vascular smooth muscle cells, endothelial cells, and CMs in organs in response to pathological stress.[13–15] Serum GDF15 was recently observed to be remarkably increased in various cardiovascular diseases such as ischemia, MI, and HF implied that GDF15 might be a predictive biomarker for cardiovascular risks.[16–18] In particular, elevated GDF15 expression was found in response to increased reactive oxidative species and high levels of pro-inflammatory cytokines in HF patients.[19] However, it is obscure to tell whether GDF15 protects against or exacerbates cardiovascular disease.

In this study, using the serum of HF patients, isoproterenol (ISO)-induced mouse model of HF and primary CMs, we investigated GDF15 expression and secretion in pathological HF in vivo and in vitro, and we also identified the effects of GDF15 on CM apoptosis and mitochondrial function.

Materials and methods

Patient information

Between January 2017 and August 2018, we recruited consecutive patients attending the Department of Cardiovascular Medicine of General Hospital of Northern Theater Command. Eligibility requirements included an age of at least 18 years, voluntarily signed the informed consent form, and were willing to cooperate with clinical research, normal cognitive ability. Based on the latest HF guideline,[20] we divide patients into HF group (an ejection fraction (EF) of 35% or less) and non-HF group (an EF more than 35%). The exclusion criteria included: (1) acute coronary syndrome, (2) severe heart valve disease, (3) severe liver and kidney disease, (4) severe anemia, (5) end-stage chronic obstructive pulmonary disease, (6) acute asthma attack, (7) inflammatory diseases and sepsis, (8) trauma, (9) mental disorders, (10) rheumatoid arthritis, (11) systemic lupus erythematosus, (12) cancers, (13) pregnancy/lactating, and (14) any surgery in the preceding 8 weeks. A total of 114 patients (57 HF (HF group) and 57 without HF (non-HF group)) aged between 29 and 75 years were enrolled in the study. The baseline demographic data and clinical characteristics were recorded, an 18-lead resting electrocardiogram was performed, and New York Heart Association (NYHA) functional class was assessed. Propensity matching of the non-HF and HF group was performed 1:1 for the baseline demographic characteristic with nearest-neighbor matching. The variables used for propensity matching calculations were age, sex, and body mass index. Blood was drawn from a forearm vein and collected in an ethylenediaminetetraacetic acid-coated tube. Samples were centrifuged, and the serum was frozen at –80°C within 1 hour of sampling for further measurement of GDF15 concentrations. Human myocardial tissue was obtained from HF patients (n = 3), while unused healthy myocardia were from transplant donors (healthy hearts, n = 3).

Ethical approval

The study complied with the 1964 Declaration of Helsinki and its later amendments and was approved by the ethics committees of the General Hospital of Northern Theater Command (No. K (2016)35-1). Informed consent was obtained from all participants in the study.

Serum GDF15 expression

Human or mouse GDF15 concentrations were determined in serum in single measurements using a quantitative sandwich monoclonal enzyme-linked immunosorbent assay (DGD150, R&D System, Inc., Minneapolis, Minnesota, USA).

ISO-induced HF mouse model

Male C57BL/6J mice (8–16 weeks old) were obtained from Nanjing Model Animal Center (Nanjing, China). The experiment grouping was carried out with litter and age matching. The mice (Ethics Approval No. 2017011032) were divided into 2 groups randomly. We number all mice, query 17 random numbers in the random number table, assign even random numbers to the experimental group, and assign mice with odd random numbers to the non-HF group. One was treated with ISO (30 mg/(kg·day)) to induce the HF mouse model (HF group, n = 12), and the other received saline for 28 days (non-HF group, n = 5). The cardiac EF values of the mice in the HF group were detected by echocardiography and EF < 35% was confirmed the HF. After 7 and 28 days of administration of ISO, the mice were euthanized respectively (n = 6). Serum was collected for the detection of GDF15 expression, and the fresh heart tissues were harvested. Half of the heart tissue was collected for histology and staining, and the other half was used for molecular analyses. Quantitative analysis was carried out using Image-Pro Plus software.


A Vevo2100 Imaging System (Toronto, Ontario, Canada) was used to perform transthoracic echocardiography on mice. The mice were continuously anesthetized with isoflurane (2% in air) and then fixed supine on a 37°C heating plate. The heart rate was maintained at 450 to 550 beats per minute. The scan head detected the long- and short-axis views of the heart. According to the image results, the mouse cardiac EF value and short-axis shortening rate were determined during systole and end-diastole.

CM culture

Adult mouse CMs were isolated by enzymatic dissociation using collagenase type II. Primary CMs were cultured in the presence of blebbistatin (a myosin II inhibitor) to block contraction and extend their survival during the culture as previously described.[21] We obtained the HL-1 cell line from the Chinese Academy of Sciences Cell Bank. Cells were cultured in Dulbecco's Modified Eagle Medium (Life Technologies Corporation, Carlsbad, California, USA) supplemented with 10% newborn bovine serum (Life Technologies Corporation). To mimic the cardiomyocytes’ impairment in vivo, ISO (0.5 mmol/L) was used to treat the primary CMs or HL-1 cells for 0, 24, and 48 hours.

Terminal deoxynucleotidyl transferase-mediated biotin-deoxyuridine triphosphate nick-end labeling (TUNEL) staining

Cardiac apoptosis was measured in heart tissue by TUNEL staining (BioVision, Milpitas, California, USA) according to the manufacturer's protocol.[22] In brief, formalin-fixed heart tissues were mounted on glass slides. The TUNEL reaction mixture containing terminal deoxynucleotidyl transferase and fluoresce-in-deoxyuridine triphosphate was added to each slide at room temperature for 30 minutes in a dark room. The slides were rinsed 3 times in phosphate-buffered saline for 5 minutes each and were counterstained with 4’,6-diamidino-2-phenylindole (DAPI, Vector Laboratories, Burlington, California, USA). The sections were examined under a fluorescence microscope (Zeiss, Optical Co. Ltd, Oberkochen, Germany) to quantify the percentages of apoptotic cells. Ten sections per heart were randomly selected for analysis, which was performed in a blinded manner. The number of positive cells and the total number of cells were counted in 3 fields at × 200 magnification. The results are presented as the ratio of positive to total cells.

Western blotting

Equal amounts of protein extracted from heart tissues or cell lysates were separated by 10% to 15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred onto polyvinyli-dene fluoride microporous membranes (Merck Millipore, Burlington, Massachusetts, USA), and blocked in 5% milk prepared in Tris-buffered solution (pH 7.6) in preparation for Western blotting. The membranes were incubated with the primary antibodies against cleaved caspase-3, GDF15, cycloox-ygenase-2 (COX2), dynamin-related protein-1 (DRP-1), mitochondrial fission factor (MFF), mitochondrial fission 1 protein (FIS-1), mitofusin 2 (MFN2), and optic atrophy 1 (OPA-1; 1:1000, Cell Signaling, Technology Danvers, Massachusetts, USA), or glyceraldehyde 3-phosphate dehydrogenase (GAPDH; 1:5000; Sigma-Aldrich Co., St Louis, Missouri, USA) overnight at 4°C. They were incubated with horseradish peroxidase-conjugated anti-rabbit/mouse IgG (1:5000, Merck Millipore and Sigma-Aldrich Co.) for 1 hour at room temperature. Signals were detected with an enhanced chemiluminescence Western blotting system (Amersham; GE Healthcare Life Sciences, Massachusetts, USA).


For histopathological examination, the heart tissue was fixed in 4% paraformaldehyde and embedded in paraffin. The sections were stained with hematoxylin-eosin and Masson's trichrome stain to evaluate the level of fibrosis. The percentage of fibrosis in the left ventricle area was quantified by identifying and counting the number of blue-stained pixels using ImageJ software (National Institutes of Health, Bethesda, Maryland, USA).

Mitotracker and Immunofluorescence staining

For mitotracker staining, cells were stained with MitoTracker® Red FM (1,941,460, 100 nmol/L; Thermo Fisher Scientific, Waltham, Massachusetts, USA) for 15 to 30 minutes. Then, cells were fixed with 4% paraformaldehyde, incubated with the primary antibodies against GDF15 (ab39999; Abcam, Boston, Massachusetts, USA) overnight, and incubated with secondary antibodies (Thermo Fisher Scientific). Cell nuclei were stained with DAPI (Vector Laboratories, Burlington, California, USA). Samples were scanned using confocal microscopy (Zeiss).

RNA silencing and transfection

Small interfering RNAs (siRNAs) targeting mouse GDF15 and pcDNA3.1-GDF15 expressed vector were purchased from Ribo Biotechnology, Inc. (Guangzhou, China). Mouse primary CMs were transfected with 10 μmol/L siRNA using Lipofectamine RNA iMAX (13778150; Life Technologies Corporation) according to the manufacturer's instructions. Similarly, mouse primary CMs were transfected with 1 μg plasmid using Lipofectamine 2000 reagent according to the manufacturer's instructions.

Real-time polymerase chain reaction (PCR)

Total RNA was extracted from cells or aorta tissues with a RNeasy mini kit (Qiagen N.V., Hilden, Germany) and reverse transcribed with an iScript cDNA synthesis kit (Bio-Rad Laboratories, Inc., Hercules, California, USA). Real-time PCR was performed with the CFX96 real-time system (Bio-Rad Laboratories, Inc.). The primer sequences for mouse genes were as follows (5′–3′): GAPDH, (F) CTACCCCACGGCAAGTTCA, (R) CCAGTAGACTCCACGA CAAC; GDF15, (F) AGCCGA-GAGGACTCGAACTCAG, (R) GGTTGACGCGGAGTAGCAGCT. Target gene expression was normalized to that of GAPDH, and fold induction was calculated with the comparative ΔCT method and presented as the relative transcript level (2–ΔΔCT).


Continuous variables are represented by mean ± standard error of mean; enumeration data are represented by frequency and percentage (n (%)). Quantitative analysis was performed using Image-Pro Plus software (Media Cybernetics, Rockville, Maryland, USA). Differences were determined by 2-sided Student's t tests. GraphPad Prism 7.0 (GraphPad Inc., San Diego, California, USA) was used for statistical analysis. Comparisons between 2 groups were performed using unpaired Student's t tests and χ2 test, and Spearman's rank correlation coefficient was used for correlation analysis. One-way, 2-way, or repeated measures analyses of variance were followed by Holm-Sidak post hoc tests to identify statistically significant differences among groups. A P value < 0.05 was considered statistically significant.


GDF15 highly expressed in serum and heart tissue of HF patients

The basic characteristics of both groups are shown in Table 1. We adjused the age, gender, BMI, blood glucose, blood lipid, and cardiovascular history but high-sensitivity C-reactive protein in non-HF patients. Serum GDF15 levels were increased robustly in HF patients compared to the non-HF group [Figure 1A]. Moreover, a significant positive correlation was found between the GDF15 expression and the NYHA class [Figure 1B]. Furthermore, immunohistochemical analysis of myocardial tissue obtained from HF patients and unused healthy myocardia from transplant donors (healthy hearts) demonstrated higher GDF15 levels in HF compared to healthy heart samples [Figure 1C].

Table 1 - Baseline characteristics of the non-HF controls and HF patients.
Patient characteristic Non-HF (n = 57) HF (n = 57) P
Age (years), mean ± SEM 59.30 ± 8.77 59.79 ± 8.61 0.763
Female, n (%) 22 (38.6) 23 (40.4) 0.848
Body mass index (kg/m2), mean ± SEM 26.64 ± 7.01 25.98 ± 5.40 0.602
Cardiovascular history or risk factors, n (%)
 Hypertension 16 (28.1) 16 (28.6) 0.953
 Diabetes mellites 6 (10.5) 9 (15.8) 0.406
Cholesterol (mmol/L), mean ± SEM
 Total 3.75 ± 1.80 3.54 ± 1.90 0.576
 Low-density lipoprotein 2.01 ± 1.00 2.01 ± 1.15 0.975
 Triglyceride 1.25 ± 0.56 1.53 ± 1.01 0.102
Fasting glucose (mmol/L), mean ± SEM 4.50 ± 2.71 5.43 ± 3.64 0.134
hsCRP (mg/L), mean ± SEM 0.01 ± 0.07 2.41 ± 5.09 0.001
Ejection fraction (%), mean ± SEM 55.83 ± 19.14 29.14 ± 4.93 <0.001
HF: Heart failure; hsCRP: High-sensitivity C-reactive protein; SEM: Standard error of mean.

Figure 1:
GDF15 is highly expressed in the serum and heart tissue of patients with HF. (A) Serum GDF15 levels in HF group (n = 57) and the non-HF group (n = 57). (B) Correlation analysis between GDF15 concentration and NYHA class (n = 114). (C) Immunohistochemical analysis demonstrated higher GDF15 levels in myocardial tissue obtained from HF group compared to non-HF group. GDF15: Growth differentiation factor 15; HF: Heart failure; NYHA: New York Heart Association.

Elevation of GDF15 occurs prior to heart remodeling in the ISO-induced mouse model

Echocardiography showed that EF and fractional shortening were significantly lower in ISO-treated mice after 4 weeks (P< 0.001) compared to saline-treated mice, which implied that ISO-induced heart dysfunction occurred in mice [Figure 2A and B]. Similarly, GDF15 secretion and expression continued to rise in both ISO-treated serum and heart tissue from 1 to 4 weeks compared to the saline-treated control group [Figure 2C and D]. Immunohistochemical analysis of myocardial tissue also demonstrated GDF15 elevation after 28 days in ISO-treated mice [Figure 2E]. This suggested that GDF15 upregulation occurs prior to heart remodeling in the ISO-induced mice model.

Figure 2:
GDF15 expression in the ISO-induced HF mouse model. HF group treated with ISO (30 mg/(kg·day)) for 7 days (n = 6) or 28 days (n = 6), non-HF group received saline for 28 days (n = 5). (A and B) EF and FS values in mice treated with ISO for 7 and 28 days or non-HF group for 28 days. (C) Serum GDF15 levels in ISO-induced mice after ISO treatment for 7 and 28 days or non-HF group for 28 days. (D) Western blotting for GDF15 expression in heart tissue from ISO-treated mice for 7 and 28 days or non-HF group for 28 days. (E) HE, Masson staining, and GDF15 expression in heart tissue with or without ISO administration. P < 0.01, P < 0.001 versus non-HF group. EF: Ejection fraction; FS: Fractional shortening; GADPH: Glyceraldehyde-3-phosphate dehydrogenase; GDF15: Growth differentiation factor 15; HE: Hematoxylin-eosin; HF: Heart failure; ISO: Isoproterenol.

Increase in GDF15 is involved in the ISO-induced CMs apoptosis

As expected, ISO (0.5 mmol/L) administration significantly and time-dependently triggered the expression of both the oxidative stress marker protein COX2 and the apoptotic marker protein cleaved-caspase 3 in HL-1 cell line [Figure 3A and B]. GDF15 expression also dramatically increased in ISO-induced HL-1 cells at both the mRNA [Figure 3C] and protein levels [Figure 3A and B]. We also observed similar results in ISO-induced primary CMs [Figure 3D and E]. Meanwhile, TUNEL staining also showed that the number of apoptotic cell is increased in HL-1 cells when treated by ISO in time-dependent manner [Figure 3F and G]. This suggested that the ISO-induced increase of GDF15 expression is related to primary CMs damage.

Figure 3:
ISO increased GDF15 levels in CMs, and this was associated with oxidative inflammation and apoptosis. (A and B) Representative Western blotting and quantification for GDF15, COX2, and cleaved caspase 3 in ISO (0.5 mmol/L)-treated CMs over time. (C) GDF15 mRNA levels in ISO (0.5 mmol/L)-treated CMs over time. (D and E) Representative Western blotting and quantification for GDF15, COX2, and cleaved caspase 3 in ISO (0.5 mmol/L)-treated primary CMs. (F and G) TUNEL staining and quantification analysis for CMs apoptosis with or without ISO (0.5 mmol/L) treatment for 24 and 48 hours. Differences were determined by 2-sided Student's t tests. P < 0.05, P < 0.01, P < 0.001 versus control group. CM: Cardiomyocytes; COX2: Cyclooxygenase 2; GADPH: Glyceraldehyde-3-phosphate dehydrogenase; GDF15: Growth differentiation factor 15; ISO: Isoproterenol.

Suppression of GDF15 exaggerated ISO-induced CM apoptosis

As shown in Figure 4A–C, qPCR and Western blotting identified that GDF15 expression significantly reduced in CMs when transfected by siGDF15 compared to siControl. TUNEL staining demonstrated that reduction of GDF15 expression triggered apoptosis of CMs induced by ISO administration [Figure 4D]. Moreover, inhibition of GDF15 expression significantly enhanced the expression of COX2 and cleaved caspase-3 in ISO-induced CMs [Figure 4E–H]. In contrast, ISO-induced CMs apoptosis was significantly reduced when we overexpressed the GDF15 via pcDNA3.1-GDF15 plasmid transfection [Figure 4I–L]. The results suggested that GDF15 is protective against ISO-induced oxidative stress and apoptosis of CMs.

Figure 4:
Silencing GDF15 exaggerated ISO-induced CM damage. (A) GDF15 mRNA levels in cells transfected with GDF15 siRNA. (B and C) Western blotting and quantification for GDF15 in GDF15 siRNA-transfected cells. (D) Ratio of apoptosis in ISO-treated cells with or without GDF silencing for 24 hours. (E–H) Representative blotting and quantification for GDF15, COX2, and cleaved caspase 3 in ISO-treated cells with or without GDF15 silencing for 24 hours. (I–L) Representative blotting and quantification for GDF15, COX2, and cleaved caspase 3 in ISO-treated cells with or without pcDNA3.1-GDF15 transfection for 24 hours. Differences were determined by 2-sided Student's t tests. P< 0.01, P < 0.001 versus control group. P< 0.05 versus siControl group. CMs: Cardiomyocytes; COX2: Cyclooxygenase-2; GADPH: Glyceraldehyde-3-phosphate dehydrogenase; GDF15: Growth differentiation factor 15; ISO: Isoproterenol; siControl: siRNA for control; siGDF15: siRNA for GDF15.

Silencing GDF15 disrupted mitochondrial fusion and fission in ISO-induced CMs

Western blotting demonstrated that the mitochondria fission marker proteins DRP-1, FIS-1, and MFF were highly expressed in ISO-induced CMs concomitant with the increase of GDF15 [Figure 5A–C]. In contrast, the fusion marker proteins MFN2 and OPA-1 were reduced in ISO-induced CMs [Figure 5A–C]. MitoTracker staining showed that mitochondrial fission morphology was increased and fusion was reduced in ISO-induced CMs [Figure 5D]. Next, we analyzed mitochondrial marker protein expression with or without silencing of GDF15 in vitro. Inhibition of GDF15 increased MFF and DRP-1 expression but reduced levels of MFN2 and OPA-1 [Figure 5E–J]. MitoTracker and GDF15 co-staining also revealed that reducing GDF15 by silencing GDF15 enhanced mitochondrial fission in CMs [Figure 5K]. Collectively, the results suggested that GDF15 inhibited mitochondrial fission and promoted mitochondrial fusion.

Figure 5:
GDF15 regulated the balance of mitochondrial fission and fusion. (A–C) Western blotting and quantification for mitochondrial marker proteins MFN2, OPA-1, FIS-1, MFF, and DRP-1 in CMs treated with ISO for 0, 24, and 48 hours. (D) MitoTracker staining of CMs with or without ISO treatment. (E–J) Expression and quantification of mitochondrial marker proteins MFN2, OPA-1, FIS-1, MFF, and DRP-1 in ISO-treated CMs with or without GDF15 silencing. (K) MitoTracker and GDF15 co-staining in CMs with or without GDF15 silencing. Differences were determined by 2-sided Student's t tests. P < 0.01, P < 0.001 versus siControl group without ISO treatment or 0 hour. P < 0.05 versus siControl group without ISO treatment or 0 hour. CMs: Cardiomyocytes; DRP-1: Dynamin-related protein-1; FIS-1: Fission 1 protein; GADPH: Glyceraldehyde-3-phosphate dehydrogenase; GDF15: Growth differentiation factor 15; ISO: Isoproterenol; MFF: Mitochondrial fission factor; MFN2: Mitofusin 2; OPA-1: Optic atrophy 1; siControl: siRNA for control; siGDF15: siRNA for GDF15.


The present study proves that the elevation of GDF15 in CMs can prevent the apoptosis of ISO-treated CMs. The results also demonstrate that GDF15 might regulate the balance of mitochondrial fusion and fission, which is involved in GDF15 regulation in HF in vivo and in vitro.

HF is a complex syndrome associated with low-grade chronic inflammation that leads to deleterious cardiac death and remodeling, but the underlying mechanisms are poorly under-stood.[23] CM apoptosis under abnormal metabolism is a key initiating factor leading to myocardial damage and cardiac failure.[24,25] GDF15 is a distant member of the TGF-β family and involved in various biological processes including inflammation, cell cycling, and apoptosis.[26] Interestingly, consistent with our findings, several studies have reported that GDF15 is robustly increased in serum in HF,[16–19] which indicates that the GDF15 has myocardial detrimental effects. Although there is prognostic value of measuring GDF15 in HF, its major functions are not completely understood. This encouraged us to study the potential of GDF15 in protecting against apoptosis and associated mechanisms. First, we evaluated the GDF15 expression and secretion in an ISO-induced mouse model of HF. Similar to the previous studies, GDF15 expression was significantly increased in failing heart tissue in mice, and the level correlated with worse cardiac dysfunction as determined by echocardiography.[16–19] GDF15 elevation was also detected in ISO-treated cultured CMs. Again, GDF15 upregulation was related to the increase of CMs apoptosis. Inhibition of GDF15 exaggerated CMs apoptosis with or without ISO treatment, implying a protective effect of GDF15 against apoptosis.

GDF15 is considered as a mitochondrial cytokine (mitokine) and is induced by mitochondrial dysfunction.[17,18,27,28] Mitochondrial dysfunction has been implicated in various diseases ranging from metabolic disorders to cancer and age-related conditions such as HF.[29] Here, we wondered whether GDF15 regulated mitochondrial dysfunction in failing CMs. To dissect the mechanisms by which GDF15 participates in CMs apoptosis, we assessed mitochondrial morphology in CMs with or without ISO treatment using Mito Tracker staining. As we expected, more mitochondrial fissions were detected in ISO-treated CMs associated with enhanced GDF15 expressions. Western blotting revealed that the levels of fission proteins increased, but those of fusion proteins decreased in ISO-treated CMs. Inhibition of GDF15 triggered fission protein expressions in CMs with or without ISO administration. Altering the function of the respiratory chain enzyme subunits could be relevant to the pathogenesis of cardiac hypertrophy.[30,31] To confirm whether mitochondrial fission changes in the context of oxidative stress, we further detected COX2 expression in CMs with or without GDF15 expression. As expected, silencing GDF15 significantly enhanced COX2 expression, which also implied mitochondrial dysfunction.


In conclusion, our results indicate that GDF15, as a mitochon-drial fusion-regulated protein, is upregulated in an early HF model to prevent CM apoptosis of heart in vivo and in vitro, suggesting that it might be a potential therapeutic target for HF research. There were several limitations in our study. First, we only compared the correlation between the expression level of GDF15 and the severity of HF in patients, and did not do further research on the mechanism of population specimens. Secondly, we should further establish GDF15 genetically modified animal models to clarify its role in the development of HF in the body.


This work was supported by the National Natural Science Foundation of China (82070308 to Dr. Liu, and 82070875 to Dr. Tian,) and the Shenyang Science and Technology Project (19-112-4-052 to Dr. Tian, and 19-112-4-056 to Dr. Liu).

Conflicts of Interest



[1]. Tomasoni D, Adamo M, Lombardi CM, et al. Highlights in heart failure. ESC Heart Fail 2019;6(6):1105–1127. doi: 10.1002/ehf2.12555.
[2]. Bacmeister L, Schwarzl M, Warnke S, et al. Inflammation and fibrosis in murine models of heart failure. Basic Res Cardiol 2019;114(3):19. doi: 10.1007/s00395-019-0722-5.
[3]. Suthahar N, Meijers WC, Silljé H, et al. From inflammation to fibrosis-molecular and cellular mechanisms of myocardial tissue remodelling and perspectives on differential treatment opportunities. Curr Heart Fail Rep 2017;14(4):235–250. doi: 10.1007/s11897-017-0343-y.
[4]. DeBerge M, Shah SJ, Wilsbacher L, et al. Macrophages in heart failure with reduced versus preserved ejection fraction. Trends Mol Med 2019;25(4):328–340. doi: 10.1016/j.molmed.2019.01.002.
[5]. Tromp J, Westenbrink BD, Ouwerkerk W, et al. Identifying pathophysi-ological mechanisms in heart failure with reduced versus preserved ejection fraction. J Am Coll Cardiol 2018;72(10):1081–1090. doi: 10.1016/j.jacc.2018.06.050.
[6]. Murphy SP, Ibrahim NE, Januzzi JL Jr. Heart failure with reduced ejection fraction: a review. JAMA 2020;324(5):488–504. doi: 10.1001/jama.2020.10262.
[7]. Nabben M, Luiken JJFP, Glatz JFC. Metabolic remodelling in heart failure revisited. Nat Rev Cardiol 2018;15(12):780. doi: 10.1038/s41569-018-0115-8.
[8]. Akhmedov AT, Rybin V, Marín-García J. Mitochondrial oxidative metabolism and uncoupling proteins in the failing heart. Heart Fail Rev 2015;20(2):227–249. doi: 10.1007/s10741-014-9457-4.
[9]. Kohlhaas M, Maack C. Interplay of defective excitation-contraction coupling, energy starvation, and oxidative stress in heart failure. Trends Cardiovasc Med 2011;21(3):69–73. doi: 10.1016/j.tcm.2012.03.002.
[10]. Thandavarayan RA, Chitturi KR, Guha A. Pathophysiology of acute and chronic right heart failure. Cardiol Clin 2020;38(2):149–160. doi: 10.1016/j.ccl.2020.01.009.
[11]. Bootcov MR, Bauskin AR, Valenzuela SM, et al. MIC-1, a novel macrophage inhibitory cytokine, is a divergent member of the TGF-beta superfamily. Proc Natl Acad Sci U S A 1997;94(21):11514–11519. doi: 10.1073/pnas.94.21.11514.
[12]. Gerstein HC, Pare G, Hess S, et al. Growth differentiation factor 15 as a novel biomarker for met for min. Diabetes Care 2017;40(2):280–283. doi: 10.2337/dc16-1682.
[13]. Chung HK, Ryu D, Kim KS, et al. Growth differentiation factor 15 is a myomitokine governing systemic energy homeostasis. J Cell Biol 2017;216(1):149–165. doi: 10.1083/jcb.201607110.
[14]. Hsu JY, Crawley S, Chen M, et al. Non-homeostatic body weight regulation through a brainstem-restricted receptor for GDF15. Nature 2017;550(7675):255–259. doi: 10.1038/nature24042.
[15]. Wollert KC, Kempf T, Lagerqvist B, et al. Growth differentiation factor 15 for risk stratification and selection of an invasive treatment strategy in non ST-elevation acute coronary syndrome. Circulation 2007;116(14):1540–1548. doi: 10.1161/CIRCULATIONAHA.107.697714.
[16]. Lok DJ, Klip IT, Lok SI, et al. Incremental prognostic power of novel biomarkers (growth-differentiation factor-15, high-sensitivity C-reactive protein, galectin-3, and high-sensitivity troponin-T) in patients with advanced chronic heart failure. Am J Cardiol 2013;112(6):831–837. doi: 10.1016/j.amjcard.2013.05.013.
[17]. Kempf T, Björklund E, Olofsson S, et al. Growth-differentiation factor-15 improves risk stratification in ST-segment elevation myocardial infarction. Eur Heart J 2007;28(23):2858–2865. doi:10.1093/eurheartj/ehm465.
[18]. Kempf T, von Haehling S, Peter T, et al. Prognostic utility of growth differentiation factor-15 in patients with chronic heart failure. J Am Coll Cardiol 2007;50(11):1054–1060. doi: 10.1016/j.jacc.2007.04.091.
[19]. Baggen VJ, van den Bosch AE, Eindhoven JA, et al. Prognostic value of N-terminal pro-B-type natriuretic peptide, troponin-T, and growth-differentiation factor 15 in adult congenital heart disease. Circulation 2017;135(3):264–279. doi: 10.1161/CIRCULATIONAHA.116.023255.
[20]. Abraham WT, Psotka MA, Fiuzat M, et al. Standardized definitions for evaluation of heart failure therapies: scientific expert panel from the Heart Failure Collaboratory and Academic Research Consortium. Eur J Heart Fail 2020;22(12):2175–2186. doi: 10.1002/ejhf.2018.
[21]. Fujita T, Ishikawa Y. Apoptosis in heart failure. – the role of the β-adrenergic receptor-mediated signaling pathway and p53-mediated signaling pathway in the apoptosis of cardiomyocytes. Circ J 2011;75(8):1811–1818. doi: 10.1253/circj.cj-11-0025.
[22]. Kuo HF, Liu PL, Chong IW, et al. Pigment epithelium-derived factor mediates autophagy and apoptosis in myocardial hypoxia/reoxygenation injury. PLoS One 2016;11(5):e0156059. doi: 10.1371/journal.pone.0156059.
[23]. Gladden JD, Chaanine AH, Redfield MM. Heart failure with preserved ejection fraction. Annu Rev Med 2018;69:65–79. doi:10.1146/annurev-med-041316-090654.
[24]. Gao G, Chen W, Yan M, et al. Rapamycin regulates the balance between cardiomyocyte apoptosis and autophagy in chronic heart failure by inhibiting mTOR signaling. Int J Mol Med 2020;45(1):195–209. doi: 10.3892/ijmm.2019.4407.
[25]. Takemura G, Kanamori H, OkadaH etal. Anti-apoptosis in nonmyocytes and pro-autophagy in cardiomyocytes: two strategies against post-infarction heart failure through regulation of cell death/degeneration. Heart Fail Rev 2018;23(5):759–772. doi: 10.1007/s10741-018-9708-x.
[26]. Wesseling M, de Poel JHC, de Jager SCA. Growth differentiation factor 15 in adverse cardiac remodelling: from biomarker to causal player. ESC Heart Fail 2020;7(4):1488–1501. doi: 10.1002/ehf2.12728.
[27]. Kumar AA, Kelly DP, Chirinos JA. Mitochondrial dysfunction in heart failure with preserved ejection fraction. Circulation 2019;139(11):1435–1450. doi: 10.1161/CIRCULATIONAHA.118.036259.
[28]. Kolwicz SC Jr, Purohit S, Tian R. Cardiac metabolism and its interactions with contraction, growth, and survival of cardiomyocytes. Circ Res 2013;113(5):603–616. doi: 10.1161/CIRCRESAHA.113.302095.
[29]. Bellumkonda L, Tyrrell D, Hummel SL, et al. Pathophysiology of heart failure and frailty: a common inflammatory origin? Aging Cell 2017;16(3):444–450. doi: 10.1111/acel.12581.
[30]. Kim TY, Wang D, Kim AK, et al. Metabolic labeling reveals proteome dynamics of mouse mitochondria. Mol Cell Proteomics 2012;11(12):1586–1594. doi: 10.1074/mcp.M112.021162.
[31]. Haileselassie B, Mukherjee R, Joshi AU, et al. Drp1/Fis1 interaction mediates mitochondrial dysfunction in septic cardiomyopathy. J Mol Cell Cardiol 2019;130:160–169. doi: 10.1016/j.yjmcc.2019.04.006.

Growth differentiation factor 15; Heart failure; Apoptosis; Mitochondrial

Copyright © 2022 The Chinese Medical Association, published by Wolters Kluwer Health, Inc.