The clinical significance of circulating glucose-regulated protein 78, Caspase-3, and C/EBP homologous protein levels in patients with heart failure

Abstract

Background and aims

The destruction of endoplasmic reticulum (ER) homeostasis leads to heart failure (HF), which further aggravates ER stress. Limited data are available on the levels of ER stress markers in HF patients in clinical practice. This study aimed to determine the clinical significance of the ER stress markers, glucose-regulated protein 78 (GRP78), Caspase-3, and C/EBP homologous protein (CHOP), in predicting HF and its severity.

Materials and methods

A total of 62 patients with HF and 44 healthy controls were enrolled in the study, and all participants were followed-up for 2 years.

Results

Serum GRP78, Caspase-3, and CHOP levels were significantly higher in patients with HF than those in healthy controls. The level of GRP78 increased with the severity of HF. GRP78 levels were negatively correlated with left ventricular ejection fraction, and positively correlated with N-terminal B-type natriuretic peptide, D-dimer, and lactic acid. Serum GRP78 and Caspase-3 levels showed moderate predictive values for HF patients.

Conclusion

ER stress markers, GRP78 and Caspase-3, had a certain predictive value in HF and can be used as serum biomarkers for the diagnosis of HF. Additionally, GRP78 showed a certain predictive value in HF severity.

Highlights

• •Serum GRP78, Caspase-3, and CHOP levels were significantly higher in HF patients.
• •Serum GRP78 and Caspase-3 levels showed moderate predictive values for HF patients.
• •The level of GRP78 increased with the severity of HF.
• •GRP78 levels were positively correlated with the indicators representing HF severity.
• •GRP78 levels could predict the severity of HF.

Abbreviations

ER

endoplasmic reticulum

GRP78

glucose-regulated protein 78

CHOP

C/EBP homologous protein

HF

heart failure

HFrEF

heart failure with reduced ejection fraction

HFpEF

heart failure with preserved ejection fraction

CAD

coronary artery disease

CVD

cerebrovascular disease

LVEDD

left ventricular end-diastolic diameter

LAH

left atrial hypertrophy

E/A

E/A ratio

E/e'

E/e' ratio

SPAP

systolic pulmonary artery pressure

r

correlation coefficient

1. Introduction

Heart failure (HF), which remains a major clinical and public health problem, is currently defined as “a clinical syndrome with symptoms and/or signs caused by a structural and/or functional cardiac abnormality and corroborated by elevated natriuretic peptide levels and/or objective evidence of pulmonary or systemic congestion” according to the European Society of Cardiology guidelines [1]. The prevalence of HF in developed countries is 1.5%–2.0%, and it accounts for the substantial morbidity and mortality worldwide [2].

Endoplasmic reticulum (ER), the site of production and folding of multiple proteins, is a multifunctional cellular organelle that plays a critical role in multiple physiological and pathological processes [3]. In the physiological state, the N terminal of three transmembrane proteins (inositol requiring protein 1α [IRE1α], activating transcription factor 6 [ATF6], protein kinase RNA-like endoplasmic reticulum kinase [PERK]) bind to glucose-regulated protein 78 (GRP78), which is the ER molecular chaperone that inhibits the activities of the three transmembrane proteins. After ER stress, errant proteins in the ER bind to GRP78, then GRP78 dissociated from three sensors, which in turn further activates downstream signaling pathway to clear errant proteins. GRP78 expression is upregulated following ER stress and plays pivotal roles in maintaining ER homeostasis [4].

Under specific conditions, ER stress is beneficial to the body. However, severe and prolonged ER stress results in cell death via apoptosis by up regulating C/EBP homologous protein (CHOP) [5]. CHOP is regarded as an important mediators of ER stress-induced apoptosis pathway [6], which acts on multiple targets to promote apoptosis [7]. Study have found that CHOP can induce apoptosis by promoting protein synthesis and oxidation in the stressed ER [8]. During ER stress, CHOP can inhibit autophagy and gradually initiate cell apoptosis [9]. Other study has shown that knockout of CHOP significantly attenuates ER stress-apoptosis in human colon cancer cells [10]. As mentioned above, CHOP plays an important role in ER stress-induced apoptosis.

Caspase is another important signaling pathway in ER stress-induced apoptosis besides CHOP. When excessive ER stress is triggered, Caspase-12 activated from ER into cytoplasm, then activates Caspase-9, following Caspase-3, ultimately triggering ERS-mediated apoptosis [11]. Based on previous literature, GRP78, CHOP and Caspase-12 are usually used as the biomarkers to respond to the ER stress level. Accumulating evidence suggests that abnormal cardiomyocyte death caused by ER stress is usually related to abnormal cardiac structure or function [12]. Studies have revealed a crucial role of ER stress, especially CHOP/Caspase-3 mediated apoptotic pathway signaling, in promoting cardiac hypertrophy and HF [3,[13], [14], [15]]. Caspase-3 participates in ERS-mediated apoptosis, especially in HF, and is considered as the ultimate common pathway member for different apoptotic cascades [11,16]. In addition, cardiomyocyte hypoxia and acute and chronic inflammatory responses in HF may lead to ER stress [17]. Therefore, ER stress may be a cause or consequence of HF. The destruction of ER homeostasis leads to HF, which further aggravates ER stress and forms a vicious circle [3]. Structural changes have occurred in the ER levels of myocardial cells in patients with HF [18]. Ni et al. reported that ER stress in cardiomyocytes was elevated in patients with HF, and alleviating ER stress may be an important mechanism underlying the therapeutic effect of β-AR blockers on HF [19].

Based on previous literature, GRP78, CHOP and Caspase-3 were selected as the ER stress biomarkers for studying heart failure in our present study. We speculated that the circulating levels of the three key proteins in patients with HF may be different from those in healthy individuals. Therefore, in the present study, we measured the levels of the three ER stress markers in patients with HF and correlated them with indicators reflecting the severity of HF to investigate the predictive value of ER stress biomarkers in HF and its severity. We aimed to assess the clinical significance of ER stress biomarkers and their diagnostic value in patients with HF.

2. Materials and methods

2.1. Subjects

A total of 62 patients with HF treated from June 2018 to June 2019 at the China-Japan Friendship Hospital were enrolled in the study. HF was clinically diagnosed and classified by two senior cardiologists separately according to the guidelines of the universal definition and classification of HF [1]. The exclusion criteria were as follows: pregnancy, acute coronary syndrome, pulmonary embolism, cerebrovascular disease, diabetic ketoacidosis, severely impaired renal function (estimated glomerular filtration <15 mL/min), abnormal hepatic function (transaminase levels > twice the upper limit of normal), severe mental illness, dementia, and a history of malignant tumors. A total of 44 sex- and age-matched healthy people from the physical examination center were enrolled as the control group. The study was approved by the Ethics Committee of China-Japan Friendship Hospital, and all participants included in the study signed the informed consent. Participants were followed-up by telephone interviews for 2 years after the day of enrollment.

2.2. Data collection and measurement of serum GRP78, Caspase-3, and CHOP levels

Baseline characteristics of the included subjects were obtained from the hospital's medical record system, including age, sex, body mass index (BMI), laboratory examination results, comorbidities, medications used, and treatment. Echocardiography was performed after admission. All-cause mortality within 2 years was obtained from medical records or family members. Blood samples from patients with HF were collected under standardized conditions immediately after admission. In brief, the samples were centrifuged at 3000×g for 15 min at 4 °C, and the supernatants were stored in aliquots at −80 °C until assayed. Serum GRP78, Caspase-3, and CHOP concentrations were measured using a commercial enzyme-linked immunosorbent assay kit (provided by BLUE GENE) according to the manufacturer's instructions.

2.3. Statistical analysis

Statistical analyses were performed using the SPSS software. Continuous data were analyzed using the Kolmogorov–Smirnov test to determine distribution. Differences between the groups were analyzed using the Independent-samples t-test when the data were normally distributed; otherwise, Mann–Whitney U tests were used. Categorical variables were presented as percentages of patients and analyzed using the Chi-square test. Correlations were performed using the Pearson rank correlation method when the data were normally distributed; otherwise, Spearman's rank correlation method was used. Multivariate regression analysis (multiple linear regression) was used for multivariate analysis. Receiver operating characteristic (ROC) curve analysis was performed to evaluate the accuracy of GRP78, Caspase-3, and CHOP levels in predicting HF and GRP78 levels in predicting mortality in patients with HF. Differences were considered statistically significant at p < 0.05. All graphs were analyzed using the GraphPad Prism or SigmaPlot software.

3. Results

3.1. Baseline characteristics of the participants

The baseline characteristics of the participants are summarized in Table 1. Patients with HF had a mean age of 74.65 ± 1.66 years. Precisely, 58.1% of the patients were male. There were no differences between HF patients and healthy controls with respect to age, sex, BMI, triglycerides (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) levels (p-value not shown). Glucose and uric acid levels were higher in patients with HF than in healthy controls. Majority of the patients (64.5%) complained of New York Heart Association (NYHA) III or IV symptoms. Approximately half of the patients (56.5%) had heart failure with reduced ejection fraction (HFrEF).

Table 1.

Baseline Characteristics of the Study Population. Data expressed as mean ± standard error of the mean (SEM).

Characteristics	Controls (n = 44)	All HF patients (n = 62)	HFpEF (n = 27)	HFrEF (n = 35)	p valuea	Survival (n = 47)	Death (n = 15)	p valueb
Age (years)	70.80 ± 1.91	74.65 ± 1.66	77.85 ± 2.52	72.17 ± 2.14	0.031	74.13 ± 2.06	76.27 ± 2.37	0.993
Gender (M:F)	19:25	36:26	17:10	19:16	0.492	28:19	8:7	0.670
BMI (kg/m2)	23.66 ± 0.45	22.69 ± 0.36	22.22 ± 0.44	23.05 ± 0.54	0.259	22.63 ± 0.39	22.87 ± 0.87	0.967
Glucose (mmol/L)	6.30 ± 0.19	10.69 ± 0.76	8.64 ± 0.75	12.27 ± 1.14	0.031	9.51 ± 0.64	14.39 ± 2.17	0.033
TG (mmol/L)	1.49 ± 0.12	1.34 ± 0.09	1.17 ± 0.14	1.48 ± 0.13	0.009	1.34 ± 0.12	1.37 ± 0.14	0.336
TC (mmol/L)	4.29 ± 0.16	4.13 ± 0.16	3.72 ± 0.20	4.45 ± 0.24	0.028	4.03 ± 0.19	4.45 ± 0.35	0.281
LDL-C (mmol/L)	2.47 ± 0.10	2.66 ± 0.13	2.34 ± 0.15	2.91 ± 0.19	0.029	2.58 ± 0.14	2.92 ± 0.31	0.258
HDL-C (mmol/L)	1.28 ± 0.06	1.16 ± 0.03	1.16 ± 0.05	1.15 ± 0.04	0.882	1.11 ± 0.04	1.29 ± 0.06	0.025
Uric acid (μmol/L)	310.57 ± 18.48	433.03 ± 20.29	430.41 ± 32.19	435.06 ± 26.40	0.560	435.36 ± 22.69	425.73 ± 45.91	0.841
NYHA Ⅰ+Ⅱ:NYHA Ⅲ+Ⅳ	0	22:40	20:7	2:33	<0.001	22:25	0:15	0.001
HFpEF:HFrEF	0	27:35	27:0	0:35	<0.001	26:21	1:14	0.001
Heart rate (bpm)	72.91 ± 1.44	104.47 ± 3.17	95.48 ± 4.68	111.4 ± 3.96	0.011	102.40 ± 3.69	110.93 ± 6.01	0.252
hsCRP (mg/L)	2.5 ± 0	48.12 ± 6.37	43.57 ± 8.96	51.63 ± 8.90	0.537	44.56 ± 6.77	59.29 ± 15.72	0.349
NT-proBNP (pg/mL)	88.43 ± 17.18	11389.94 ± 1364.60	10517.26 ± 1947.68	12063.14 ± 1910.42	0.527	11065.02 ± 1575.72	12408.0 ± 2806.48	0.537
cTnT (ng/mL)	0.0088 ± 0.001	0.83 ± 0.31	0.15 ± 0.04	1.36 ± 0.54	0.041	0.30 ± 0.08	2.49 ± 1.19	0.004
D-dimer (mg/L)	0.41 ± 0.04	2.30 ± 0.31	2.04 ± 0.33	2.51 ± 0.48	0.798	2.10 ± 0.27	2.93 ± 0.94	0.336
Lactic acid (mmol/L)	0.63 ± 0.04	3.12 ± 0.43	1.82 ± 0.41	4.12 ± 0.64	0.005	2.07 ± 0.29	6.42 ± 1.17	<0.001
Pulmonary Edema (%)	0	51 (82.3%)	17 (62.9%)	34 (97.1%)	0.002	36 (76.6%)	15 (100%)	0.093
Hypoperfusion (%)	0	10 (16.1%)	1 (3.7%)	9 (25.7%)	0.047	2 (4.3%)	8 (53.3%)	<0.001
Comorbidities [N0. (%)]
CAD	0	29 (46.8%)	14 (51.8%)	15 (42.8%)	0.482	20 (42.6%)	8 (53.3%)	0.465
Atrial fibrillation	0	17 (27.4%)	8 (29.6%)	9 (25.7%)	0.732	15 (31.9%)	2 (13.3%)	0.284
Hypertension	0	40 (64.5%)	19 (70.3%)	21 (60%)	0.397	31 (65.9%)	9 (60%)	0.675
DM	0	27 (43.5%)	14 (51.8%)	13 (37.1%)	0.247	21 (44.7%)	6 (40%)	0.822
COPD	0	5 (8.1%)	1 (3.7%)	4 (11.4%)	0.524	2 (4.3%)	3 (20%)	0.160
Previous CVD	0	18 (29.0%)	8 (29.6%)	10 (28.5%)	0.927	14 (29.8%)	4 (26.7%)	1.000
Medication/Treatment [N0. (%)]
Vasoactive drugs	0	8 (12.9%)	0 (0%)	8 (22.9%)	0.023	0	8 (53.3%)	<0.001
Mechanical Ventilation	0	3 (4.8%)	0 (0%)	3 (8.5%)	0.336	0	3 (20%)	0.012
Follow-up outcome [N0. (%)]
Survival	44 (100%)	47 (75.8%)	26 (96.3%)	21 (60%)	0.001	47 (100%)	0	<0.001

Abbreviations are as follows: HF, Heart failure; HFpEF, Heart failure with preserved ejection fraction; HFrEF, Heart failure with reduced ejection fraction; M, Male; F, Female; BMI, Body Mass Index; TG, Triglycerides; TC, Total Cholesterol; LDL-C, Low-Density Lipoprotein Cholesterol; HDL-C, High-Density Lipoprotein Cholesterol; NYHA, New York Heart Association; hs-CRP, hypersensitive- C-reactive protein; NT-proBNP, N-terminal B-type natriuretic peptide; cTnT, cardiac troponin T; CAD, Coronary artery disease; DM, Diabetes mellitus; COPD, chronic obstructive pulmoriary disease; CVD, Cerebrovascular Disease; p value^a: HFpEF vs. HFrEF; p value^b: Survival vs. Death.

The mean heart rate of the patients was 104.47 ± 3.17 bpm. The mean hypersensitive C-reactive protein (hs-CRP) level was 48.12 ± 6.37 mg/L. Mean N-terminal pro-B-type natriuretic peptide (NT-proBNP) and cardiac troponin T (cTNT) were 11389.94 ± 1364.6 pg/mL and 0.83 ± 0.31 ng/mL, respectively. Mean D-dimer and lactic acid were 2.3 ± 0.31 mg/L and 3.12 ± 0.43 mmol/L, respectively. Most patients had pulmonary edema (82.3%), and 16.1% of the patients had hypoperfusion. Minority of the patients required vasoactive drugs (12.9%) or mechanical ventilation (4.8%). Overall, the biomarkers reflecting HF severity in the HF group were significantly higher than those in the healthy control group. Patients were characterized by relevant comorbidities such as coronary artery disease (46.8%), atrial fibrillation (27.4%), hypertension (64.5%), diabetes mellitus (DM) (43.5%), chronic obstructive pulmonary disease (COPD) (8.1%), and previous cerebrovascular disease (29.0%). During the 2-year follow-up period, 15 patients (24.2%) died. Overall, all healthy controls had no clinical evidence of vascular disease, diabetic disease, or COPD on careful history and routine laboratory tests.

There were differences with respect to age, glucose levels, and blood lipid levels between heart failure with preserved ejection fraction (HFpEF) and HFrEF groups. Parameters reflecting the severity of HF in the HFrEF group were significantly higher than those in the HFpEF group (heart rate, cTNT, lactic acid, and percentage of pulmonary edema, hypoperfusion, vasoactive drugs used, and non-survivors). Comorbidities did not differ between the HFrEF and HFpEF groups. Thus, the baseline of the HFrEF group was more serious than that of the HEpEF group. Moreover, most data such as age, sex, BMI, most blood lipid level, uric acid level, heart rate, and comorbidities did not differ between survivors and non-survivors. However, non-survivors were characterized by higher glucose, HDL-C, cTNT, and lactic acid levels. The percentage of other parameters reflecting the severity of HF in the non-survivors group was significantly higher than that in the survivors group (pulmonary edema, hypoperfusion, vasoactive drugs, and mechanical ventilation). Understandably, the deceased group was worse at the baseline.

The echocardiographic parameters of HF are presented in Table 2. As shown, the left ventricular ejection fraction (LVEF) was, by definition, significantly different between the HFrEF and HFpEF groups. Other significant differences were found in left ventricular end-diastolic diameter (LVEDD) and systolic pulmonary artery pressure (SPAP), both of which were higher in HFrEF, which is a distinctive feature of the disease. Moreover, a significantly decreased LVEF was observed in non-survivors compared to survivors. The mean SPAP was also higher in the non-survivors group than in the survivors group.

Table 2.

The echocardiographic parameters of heart failure patients. Data expressed as mean ± standard error of the mean (SEM).

Parameters	HFpEF (n = 27)	HFrEF (n = 35)	p valuea	Survival (n = 47)	Death (n = 15)	p valueb
LVEDD (mm)	49.85 ± 1.10	55.17 ± 1.35	0.005	53.15 ± 1.16	51.93 ± 1.62	0.591
LVEF (%)	58.52 ± 1.36	35.23 ± 1.37	<0.001	48.81 ± 1.88	34.60 ± 2.97	0.000
LAH (%)	21 (77.8%)	26 (74.3%)	0.750	38 (80.9%)	9 (60%)	0.195
E/A	<1: 12	<1: 7	0.101	<1: 17	<1: 2	0.421
	>1: 11	>1: 16		>1: 19	>1: 8
	>2: 0	>2: 3		>2: 2	>2: 1
	Unimodal: 4	Unimodal: 9		Unimodal: 9	Unimodal: 4
E/e'	20.05 ± 1.02	22.52 ± 1.16	0.123	21.05 ± 0.86	22.52 ± 0.92	0.441
SPAP (mmHg)	34.0 ± 2.61	38.0 ± 1.38	0.038	34.45 ± 1.45	41.93 ± 3.19	0.018

Abbreviations are as follows: HFpEF, Heart failure with preserved ejection fraction; HFrEF, Heart failure with reduced ejection fraction; LVEDD, Left ventricular end-diastolic diameter; LVEF, Left ventricular ejection fraction; LAH, Left atrial hypertrophy; E/A, E/A ratio; E/e', E/e' ratio; SPAP, systolic pulmonary artery pressure. P value^a: HFpEF vs. HFrEF; p value^b: Survival vs. Death.

3.2. Serum GRP78, Caspase-3, and CHOP levels in HF patients and healthy controls

As shown in Fig. 1A, circulating GRP78 levels in the HF group were significantly higher than those in the controls (2.64 ± 0.3 vs. 1.02 ± 0.08 ng/mL, p < 0.001). In addition, circulating Caspase-3 levels and CHOP levels in HF patients were all significantly higher than those in controls (5.83 ± 0.15 vs. 3.15 ± 0.23 ng/mL, p < 0.001; 85.11 ± 14.36 vs. 50.84 ± 5.52 pg/mL, p = 0.009), Fig. 1E and I. Subgroup analysis of HF patients showed that circulating GRP78 levels in HFrEF were significantly higher than those in HFpEF patients (3.24 ± 0.48 vs. 1.87 ± 0.24 ng/mL, p = 0.031), Fig. 1B. Similarly, circulating GRP78 levels in the NYHA class III + IV group and non-survivors group showed the same increased trend when compared to the NYHA class I + II group (3.2 ± 0.42 vs. 1.62 ± 0.28 ng/mL, p = 0.002) and survivors group (3.33 ± 0.51 vs. 2.42 ± 0.36 ng/mL, p = 0.038), respectively, Fig. 1C and D. Circulating Caspase-3 levels were slightly higher in the subgroup of HFrEF, NYHA class III + IV group, and non-survivors group when compared to the corresponding controls (5.9 ± 0.21 vs. 5.75 ± 0.21 ng/mL; 5.94 ± 0.2 vs. 5.64 ± 0.2 ng/mL; 5.93 ± 0.41 vs. 5.8 ± 0.15 ng/mL), although the difference did not reach statistical significance, Fig. 1F, G, 1H. Circulating CHOP levels were elevated in HFrEF patients compared to HFpEF (98.05 ± 24.65 vs. 68.33 ± 7.88 pg/mL), Fig. 1J. Furthermore, circulating CHOP levels were elevated in patients with NYHA class III + IV compared to those with NYHA class I + II (95.31 ± 21.59 vs. 66.56 ± 9.48 pg/mL), Fig. 1K. Non-survivors at admission showed higher CHOP levels compared to survivors (141.1 ± 55.19 vs. 67.24 ± 6.03 pg/mL), Fig. 1L; however, the difference was not statistically significant with respect to the three subgroups of CHOP. Additionally, levels of GRP78, Caspase-3, and CHOP in patients with pulmonary edema were higher than those in patients without pulmonary edema (2.8 ± 0.35 vs. 1.9 ± 0.5 ng/mL, p = 0.182; 5.95 ± 0.17 vs. 5.28 ± 0.28 ng/mL, p = 0.021; 86.87 ± 17.18 vs. 76.96 ± 15.7 pg/mL, p = 0.761).

Fig. 1.

Serum GRP78, Caspase-3 and CHOP levels in patients with heart failure and healthy controls. (A, E, I) Serum GRP78, Caspase-3, CHOP levels in patients with heart failure (HF) compared to controls; (B, F, J) Serum GRP78, Caspase-3, CHOP levels in patients with heart failure with reduced ejection fraction (HFrEF) compared to heart failure with preserved ejection fraction (HFpEF); (C, G, K) Serum GRP78, Caspase-3, CHOP levels in patients with HF according to NYHA class (NYHA III/IV vs. I/II); (D, H, L) Serum GRP78, Caspase-3, CHOP levels in patients with HF according to survival during 2-year follow-up (Death vs. Survival).

3.3. Correlation with associated baseline characteristics

The correlations of GRP78, Caspase-3, and CHOP with clinical and laboratory parameters of patients were evaluated. Univariate analysis showed that high levels of GRP78 or Caspase-3 were correlated with higher glucose levels (correlation coefficient [r] = 0.233, p = 0.029; r = 0.298, p = 0.004). The data are presented in Supplementary Table 1. None of the biomarkers were significantly influenced by age, BMI, TG, TC, LDL-C, HDL-C, or uric acid (all p = non-significant [NS], data shown in Supplementary Table 1).

The above results indicated that the levels of circulating GRP78 were related to the severity of HF. To further verify this conclusion, we conducted a correlation analysis between GRP78 levels and the indicators that represent the severity of HF (such as heart rate, hs-CRP, NT-proBNP, cTNT, D-dimer, lactic acid, LVEDD, LVEF, and SPAP). As shown in Table 3, The level of GRP78 was positively correlated with NT-proBNP (r = 0.345, p = 0.006), D-dimer (r = 0.301, p = 0.017), and lactic acid (r = 0.269, p = 0.034). The level of GRP78 was also positively correlated with heart rate, hsCRP, cTNT, and SPAP, although the correlation was not statistically significant (data not shown). Additionally, the level of GRP78 was negatively correlated with the LVEF (r = −0.409, p = 0.001). Fig. 2 shows the graph of the correlation between GRP78 levels and LVEF (Fig. 2A), NT-proBNP (Fig. 2B), D-dimer (Fig. 2C), and lactic acid (Fig. 2D). GRP78 levels were not influenced by age, BMI, glucose, TG, TC, LDL-C, HDL-C, or uric acid (all p = NS, data shown in Table 3).

Table 3.

The correlation of GRP78 with associated baseline characteristics.

Variable	r	p
Age	0.019	0.886
BMI	0.027	0.837
Glucose	0.093	0.473
TG	0.026	0.843
TC	−0.178	0.166
LDL-C	−0.169	0.188
HDL-C	−0.127	0.324
Uric acid	0.065	0.618
Heart rate	0.152	0.238
hsCRP	0.243	0.057
NT-proBNP	0.345	0.006
cTnT	0.032	0.805
D-dimer	0.301	0.017
Lactic acid	0.269	0.034
LVEDD	−0.054	0.679
LVEF	−0.409	0.001
sPAP	0.021	0.871

Abbreviations are as follows: HF, Heart failure; r: correlation coefficient.

Fig. 2.

Correlation between serum GRP78 levels and the indicators representing the severity of heart failure. (A) The relationship between Left ventricular ejection fraction (LVEF) and serum GRP78 levels in patients with heart failure (HF); (B) The relationship between N-terminal B-type natriuretic peptide (NT-proBNP) and serum GRP78 levels in patients with HF; (C) The relationship between D-dimer and serum GRP78 levels in patients with HF; (D) The relationship between Lactic acid and serum GRP78 levels in patients with HF.

3.4. Assessment of GRP78, Caspase-3, and CHOP levels for diagnostic accuracy of HF and GRP78 level for predicting mortality of HF

To evaluate the diagnostic accuracy of GRP78, Caspase-3, and CHOP in HF, ROC curve analyses were performed as described by Delong [20]. As shown in Fig. 3, the area under the curve (AUC) for Caspase-3 was 0.876 (95% confidence interval [CI] = 0.796 to 0.956, p < 0.0001), while AUCs for GRP78 and CHOP were 0.777 (95% CI = 0.687 to 0.867, p < 0.0001) and 0.649 (95% CI = 0.544 to 0.754, p = 0.009), respectively. As a positive control, the AUC for NT-proBNP was 0.999 (95% CI = 0.999 to 1.001, p < 0.0001). In addition, the serum Caspase-3 level at 4.027 ng/mL was found to be 100% sensitive and 80% specific for distinguishing healthy participants from HF patients. Furthermore, the serum GRP78 level was found to be 66% sensitive and 89% specific at 1.529 ng/mL.

Fig. 3.

ROC curve analysis chart of Caspase-3, GRP78 and CHOP levels; HF group vs. control group.

Since the serum GRP78 levels were significantly higher in deceased HF patients than in those who survived during the 2-year follow-up, we performed ROC analysis to evaluate the predictive ability of GRP78 level for mortality in HF patients. As shown in Fig. 4, the AUC for GRP78 was 0.679 (95% CI = 0.516 to 0.843, p = 0.038), while the AUCs for lactic acid and cTNT were 0.838 (95% CI = 0.691 to 0.986, p < 0.0001) and 0.745 (95% CI = 0.598 to 0.893, p = 0.004), respectively. The serum GRP78 level at 3.755 ng/mL was found to be 53% sensitive and 90% specific for predicting mortality in patients with HF.

Fig. 4.

ROC curve analysis chart of GRP78, cTNT and lactic acid levels; death vs. survival.

4. Discussion

Evidence has revealed that ER stress is associated with various cardiovascular diseases, such as hypertension, ischemic heart disease, atherosclerosis, cardiomyopathy, cardiac hypertrophy, and HF [3,15,21]. In recent years, the relationship between ER stress and HF has become a hot topic in the field of HF research [14,15,22]. These studies confirmed a close correlation between ER stress and HF; however, published literature on the circulating levels of ER stress biomarkers in patients with HF is limited.

ER stress leads to the decrease of protein inflow into the ER, and increases the expression of gene encoding chaperone GRP78 to mitigate the impact of increased unfolded protein load [23]. It is known that GRP78 inhibits apoptotic signaling and protects the cell from apoptosis induced by ER stress. Research has shown that GRP78 expression is upregulated following ER stress and has protective effects on injured glial cells [24]. GRP78 accumulation can inhibits ER stress-induced apoptosis in HEK293 cells [25]. CHOP, a member of the C/EBP family of transcription factors, is expressed at low levels under physiological conditions. However, CHOP is significantly induced in response to ER stress and participates in the expression of downstream apoptosis-related genes [5,26]. According to previous literature, ER stress-induced apoptosis can be accompanied by the increase of GRP78 and CHOP. Shakubatrox valsartan sodium protects cardiomyocytes by inhibiting the ER stress of cardiomyocytes and down regulating the expression of GRP78 and CHOP, thereby reversing ventricular remodeling and improving cardiac function. Based on previous literature, we speculated that the activation of cardiac ER stress in patients with HF inducing cardiomyocyte apoptosis, accompanied by the increase expression of GRP78, CHOP and Caspase-3. Our results showed significantly elevated levels of GRP78, Caspase-3, and CHOP in HF patients compared to healthy controls, which confirmed our hypothesis. Long term or severe ER stress maintains high CHOP levels to promote apoptotic signaling that induces activation of Caspase-3 and cell death by apoptosis [27]. Therefore, the changing trends of Caspase-3 and CHOP should be consistent, and our research has also confirmed this point, that is, the levels of Caspase-3 and CHOP in patients with HF are elevated compared to those in healthy controls.

In addition, our study showed that serum levels of GRP78, Caspase-3, CHOP increases with the severity of HF (although the difference between the latter two did not reach statistically significance). In patients with severe HF, more serious ER stress-induced apoptosis in the heart occurred, thus the expression of CHOP/Caspase-3 in the apoptosis signal pathway was higher. Meanwhile, the compensatory expression of GRP78 which has protective effect was also increased to protect the myocardium.

Our research revealed elevated levels of GRP78 and CHOP, which is consistent with the conclusions of previous studies [5,26,28]. However, Sabirli et al. found that GRP78 and CHOP levels were lower in the HF group than in the control group [29], which is contrary to our conclusion. Differences in the study population and measurement methods may partly explain the different conclusions of Sabirli's study. In many diseases related to ER, the expression of GRP78 and CHOP is increased. For instance, higher circulating GRP78 and CHOP concentrations in patients with Type 2 DM, elevated expression of GRP78 in COVID-19 autopsies, abnormal level of GRP78 in patients with obstructive sleep apnea [[30], [31], [32]]. This may be due to ER stress accompanied by the increased expression of GRP78 on the cell surface and subsequent secretion into the blood. Additionally, other ER stress factors such as PERK, IRE1α, ATF6 and Caspase-12 could be studied in the future study besides the three key biomarkers (GRP78, CHOP, Caspase-3) in patients with HF.

Our findings indicate that ER stress biomarkers, especially GRP78 and Caspase-3, have predictive value in HF and can be used as biomarkers of HF, while the diagnostic accuracy of CHOP for HF was lower. Small sample size and large standard deviation of CHOP between the two groups may partly explain the weak correlation between CHOP and HF diagnosis accuracy. Of course, there is no doubt that NT-proBNP is a recognized biomarker of HF compared to GRP78 or Caspase-3 [33].

Additionally, during the 2-year follow-up, more deaths occurred in the HFrEF, NYHA III + IV, and pulmonary edema groups (compared with their respective control groups). The results of the ROC curve suggested that the accuracy of using GRP78 to predict the mortality of HF was slightly lower than that of lactic acid and cTNT levels; however, the AUC difference between them did not reach statistical significance.

Heart failure, with relatively complex pathophysiological mechanism, still lacking of targeted prevention and treatment measures. In this study, we confirmed that ER stress-mediated apoptosis play an important role in the occurrence and development of HF through the detection of ER stress markers. In the future research, exploring the specific mechanism of ER stress in HF, trying to intervene in various target of the ER stress pathway, and regulating ER stress-mediated apoptosis may be a new potential target for the treatment of HF.

This study has the following limitations. First, our sample size was limited to 106 participants in one institution; large multicenter studies and larger populations are needed to verify these findings in the future. Second, our study did not evaluate the clinical significance of ER stress biomarkers in the prognosis of HF, that is, the response of GRP78 to HF treatment and whether GRP78-guided treatment can bring additional benefits to patients with HF is unclear. Finally, the current research is limited to theory, and it is necessary to explore specific methods to achieve therapeutic value in the future research.

5. Conclusion

In summary, our results found that the levels of circulating GRP78, Caspase-3, and CHOP in patients with HF were significantly higher than those in the control group. Combined with the analysis results of the ROC curve, we infer that ER stress markers, GRP78 and Caspase-3, can be used as serum biomarkers for the diagnosis and differential diagnosis of HF, and provide a new approach for the treatment of HF. In addition, the levels of GRP78 in the HFrEF, NYHA class III + IV, and non-survivors groups were significantly higher than those in the corresponding control groups, and the level of GRP78 was positively correlated with the indicators representing the severity of HF, such as NT-proBNP and D-dimer. These results suggest that GRP78 can be used as a serum biomarker for HF to determine the severity of the disease. The results of the ROC curve analysis suggested that GRP78 could predict death in HF to a certain extent. These findings support the potential therapeutic value of targeting ER stress to reduce or reverse cardiac remodeling that is present in HF.

Author contribution statement

Xuecheng Zhao: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Da-Qi Zhang: Conceived and designed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Rongjing Song: Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data.

Rong Wang: Performed the experiments; Contributed reagents, materials, analysis tools or data.

Guoqiang Zhang: Conceived and designed the experiments; Wrote the paper.

Funding statement

This work was supported by the grants from National Natural Science Foundation of China (grant numbers. 81871600) and project supported by Hainan Province Clinical Medical Center.

Data availability statement

The data that has been used is confidential.

Declaration of interest's statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Additional information

Supplementary content related to this article has been published online at [URL].

Appendix A. Supplementary data

The following is the Supplementary data to this article.