Polysaccharide from Angelica sinensis protects H9c2 cells against oxidative injury and endoplasmic reticulum stress by activating the ATF6 pathway

Short abstract

Objectives

Angelica sinensis exerts various pharmacological effects, such as antioxidant and anti-apoptotic activity. This study aimed to investigate the active ingredients in A. sinensis with antioxidant properties and whether A. sinensis polysaccharide (ASP) protects H9c2 cells against oxidative and endoplasmic reticulum (ER) stress.

Methods

The ingredients of A. sinensis and their targets and related pathways were determined using web-based databases. Markers of oxidative stress, cell viability, apoptosis, and ER stress-related signalling pathways were measured in H9c2 cells treated with hydrogen peroxide (H₂O₂) and ASP.

Results

The ingredient–pathway–disease network showed that A. sinensis exerted protective effects against oxidative injury through its various active ingredients on regulation of multiple pathways. Subsequent experiments showed that ASP pretreatment significantly decreased H₂O₂-induced cytotoxicity and apoptosis in H9c2 cells. ASP pretreatment inhibited H₂O₂-induced reactive oxygen species generation, lactic dehydrogenase release, and malondialdehyde production. ASP exerted beneficial effects by inducing activating transcription factor 6 (ATF6) and increasing ATF6 target protein levels, which in turn attenuated ER stress and increased antioxidant activity.

Conclusions

Our findings indicate that ASP, a major water-soluble component of A. sinensis, exerts protective effects against H₂O₂-induced injury in H9c2 cells by activating the ATF6 pathway, thus ameliorating ER and oxidative stress.

Introduction

Oxidative stress (OS) is a well-known factor that promotes apoptosis, and is implicated in the pathophysiology of several cardiovascular diseases, such as ischaemia/reperfusion injury, atherosclerosis, heart failure, and hypertension.^1–3 The endoplasmic reticulum (ER) plays an important role in maintaining intracellular homeostasis when cells are subjected to a variety of environmental and physiological stresses. OS is usually associated with increased formation of reactive oxygen species (ROS), which disturb cellular redox balance and homeostasis of the protein-folding environment of the ER, leading to ER stress.^4,5 To restore homeostasis of the ER, cells activate unfolded protein response (UPR) signalling pathways. These pathways are regulated by three primary ER-resident transducers called activating transcription factor 6 (ATF6), inositol-requiring enzyme 1α (IRE1α), and pancreatic eIF2-alpha kinase (PERK). ATF6, a 90-kDa glycoprotein (p90 ATF6), is an important transcriptional activator during ER stress.⁶ In the presence of misfolded proteins in the ER, the fully form of ATF6 shuttles to the Golgi apparatus, where it undergoes cleavage into its active 50-kDa form (p50 ATF6). Active p50 ATF6 then translocates into the nucleus and activates numerous genes encoding ER chaperones.⁶ IRE1α is a kinase and RNase that mediates selective splicing of X-box binding protein 1 (XBP1) mRNA.⁷ The spliced XBP1 mRNA encodes an active transcription factor that induces expression of genes encoding ER-associated degradation components.⁷ PERK stimulates phosphorylation of eukaryotic translation initiation factor 2α to reduce protein synthesis.⁸ The UPR signalling pathways function as protective responses in the early stage of cellular injury. However, under severe conditions, UPR activation is not sufficient to restore normal ER function, which subsequently results in activation of apoptosis.^6–8 The pro- and anti-apoptotic effects of the UPR are associated with the three ER transmembrane proteins of the ER, inducing expression of distinct sets of genes.⁹ Therefore, differential modulation of each UPR pathway may be an attractive therapeutic strategy for treating cardiovascular diseases.⁹ Several studies have shown that activation of ATF6 increases ER-resident protein expression to avoid the toxic effects of misfolded proteins, including 78-kDa glucose-regulated protein (GRP78),¹⁰ 94-kDa glucose-regulated protein (GRP94),¹¹ and protein disulphide isomerase family A, member 6 (PDIA6).¹² Additionally, activation of ATF6 also induces antioxidative stress genes encoding proteins outside the ER, such as catalase (CAT).¹³ These studies suggest that the ATF6 branch of UPR has an important protective function against ER stress and oxidative damage.¹⁴

The root of Angelica sinensis (Oliv.) Diels (Umbelliferae), known as Danggui in Chinese, has been used as a tonic agent in China for several years.¹⁵ Accumulating evidence obtained from experimental and clinical studies has highlighted the beneficial roles of A. sinensis in treating OS-related diseases.¹⁵ Our previous study showed that an ultra-filtration extract of Danggui Buxue Decoction, a medicinal formula primarily containing A. sinensis, protected neonatal rat cardiomyocytes from hydrogen peroxide (H₂O₂)-induced injury by exerting antioxidant and anti-apoptotic effects.¹⁶ Studies have been conducted to identify the active ingredients associated with the antioxidant effects of A. sinensis.¹⁵ However, completely determining different components in A. sinensis through traditional experimental methods is difficult. A functional network analysis has been proposed to examine the relationship between this herb and its targets and grouped pathways.^17–20 Several cancer studies have established the roles of network analysis in identifying active ingredients and their molecular mechanisms.^19,20

Therefore, we hypothesized that network analysis using data obtained from publicly available databases will help in systematically understanding the antioxidant effects of A. sinensis. A. sinensis polysaccharide (ASP), which is the primary water-soluble component of A. sinensis, was chosen for further experimental validations. Additionally, we determined the effect of ASP on ER stress and UPR pathways in H₂O₂-treated H9c2 cells.

Materials and methods

Candidate targets associated with OS injury

Candidate targets associated with OS injury were determined using the Comparative Toxicogenomics Database²¹ and Open Targets Platform.²² Known and predicted protein targets associated with OS injury were used as candidate targets. We also used PolySearch,² an online text-mining system, to identify reliable targets associated with OS injury.²³ Functional enrichment analysis of the screened targets was performed using Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis, which is available in the Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.8.²⁴

Determination of active ingredients in A. sinensis and their targets

Ingredients in A. sinensis and their potential targets were determined using the Traditional Chinese Medicine Integrative Database,¹⁸ Bioinformatics Analysis Tool for Molecular mechANism of Traditional Chinese Medicine,¹⁷ and PolySearch² tool. These targets were then analysed using KEGG pathway analysis that is available in the Database for Annotation, Visualization and Integrated Discovery. Pathways with a false discovery rate of P < 0.05 were considered significant.

Network construction and analysis

Active ingredients in A. sinensis were identified if their targets were enriched in the same pathways as OS injury-related candidate targets involved. An ingredient–pathway–disease network was constructed using Cytoscape software (Version 3.4.0, Cytoscape software, San Diego, CA, USA).

Experimental validation

The rat embryonic ventricular myocardial cell line H9c2 (China Center for Type Culture Collection; Beijing, China) was used to examine the effects of ASP in an in vitro experiment. H9c2 cells were cultured in high-glucose Dulbecco’s modified eagles medium (HyClone Laboratories, Logan, UT, USA) containing 10% newborn calf serum (Gibco, Auckland, New Zealand) and antibiotics (100 U/mL penicillin and 100 µg/mL streptomycin). H9c2 cells were treated with H₂O₂ and/or ASP (purity ≥90%; Solarbio, Beijing, China) after they reached 80% confluence.

Cell viability assay

H9c2 cells were seeded in 96-well plates (density, 1 × 10⁴ cells/well) and were treated with ASP (0–100 µg/mL) for 4 hours or H₂O₂ (0–400 µM) for 2–8 hours at 37°C in a humidified chamber. Cell viability was measured using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) assay (Solarbio). Briefly, MTT solution (final concentration, 0.5 mg/mL) was added to each well, and the cells were incubated at 37°C for 4 hours. Culture supernatant was then removed, and 150 µL dimethylsulfoxide was added to each well to dissolve formazan crystals. Optical density (OD) of each well was measured at 490 nm by using an infinite M200PRO microplate reader (Tecan, Männedorf, Switzerland). Cell viability was calculated using the following formula: experimental OD value/control OD value × 100.

OS analysis

OS was determined by analysing intracellular ROS production and malondialdehyde (MDA) and lactate dehydrogenase (LDH) levels after treatment with H₂O₂ and/or ASP. ROS generation in H9c2 cells was determined by performing flow cytometric analysis with an ROS assay kit (Beyotime, Shanghai, China). Briefly, H9c2 cells were harvested after treatment, washed twice with phosphate-buffered saline (PBS), incubated with 2,7-dichlorofluorescein diacetate (final concentration, 10 µM) diluted in a serum-free medium for 30 minutes, and washed three times with PBS. ROS levels in H9c2 cells were measured using a FACSVerse flow cytometer (BD Biosciences, San Jose, CA, USA). MDA and LDH levels were determined using commercially available detection kits (Jiancheng, Nanjing, Jiangsu, China) according to respective kit instructions.

Detection of apoptosis

Apoptosis was evaluated by performing flow cytometric analysis and Hoechst 33342 staining. For performing flow cytometric analysis, the cells were washed twice with PBS and suspended in 100 µL binding buffer containing 5 µL annexin V-fluorescein isothiocyanate (FITC) and 5 µL propidium iodide (PI) (BD Biosciences) in the dark at room temperature. After 15 minutes, the cell suspensions were immediately analysed using a FACSVerse flow cytometer. Percentages of apoptotic cells were calculated as the ratio of annexin V-positive cells to the total cell population × 100. Hoechst 33342 staining (Beyotime) was performed to qualitatively analyse apoptotic cells. After ASP and/or H₂O₂ treatment, H9c2 cells were incubated with 10 µL 1× Hoechst 33342 at 37°C for 10 minutes, washed three times with PBS, and visualized using Olympus IX71 fluorescence microscopy (Olympus, Tokyo, Japan). Additionally, morphological changes in the cells were observed using the Olympus IX71 inverted microscope and photomicrographs were obtained at 10× magnification.

Small interfering RNA transfection

H9c2 cells were seeded in 24-well plates (4 × 10⁴ cells/well). After reaching approximately 50% confluence, the cells were transfected with a small interfering RNA (siRNA) against Atf6 or a non-targeting siRNA using Lipofectamine 2000 (Invitrogen Corporation, Carlsbad, CA, USA), according to the manufacturer’s instructions. The sequence of the Atf6 siRNA (siAtf6; Genepharma, Shanghai, China) was as follows: sense, 5'-GUGUGACUAAACCUGUUCUTT-3' and antisense, 5'-AGACUGAGAACUAGACAACTT-3'. The sequence of the non-targeting siRNA (Genepharma) was as follows: sense, 5'-UUCUCCGAACGUGUCACGUTT-3' and antisense, 5'-ACGUGACACGUUCGGAGAATT-3'. To assess the specific silencing effect of siAtf6, ATF6 protein expression was detected by immunoblot analysis.

Quantitative real-time RT-PCR

Total cellular RNA was extracted using Trizol reagent (Invitrogen) and the reverse transcript reaction was performed using PrimeScript RT reagent Kit (Takara Biotechnology, Dalian, China). The cDNA product was then subjected to quantitative real-time RT-PCR (qRT-PCR) analysis on a LightCycle 480 system (Roche Diagnostics, Burgess Hill, UK). The primers used were as below: for Atf6, 5'-CAGCAAAGACCATCATCATTCAG-3' and 5'-TTAGTCACACACAGTTTTCCGTTC-3'; for Cat, 5'-CTTTGAGGTCACCCACGATATT-3' and 5'-GTGGGTTTCTCTTCTGGCTATG-3'; and for β-actin, 5'-TGCTATGTTGCCCTAGACTTCG-3' and 5'-GTTGGCATAGAGGTCTTTACGG-3'. Reactions were performed in a 20 µL volume containing 10 µL 2 × SYBR Premix Ex Taq II (Takara Biotechnology), 0.8 µL forward primer, 0.8 µL reverse primer, 2.0 µL cDNA, and 6.4 µL dH₂O. The qRT-PCR conditions were 95°C for 30 s followed by 40 cycles with the following settings: 95°C for 5 s, 60°C for 25 s, and 72°C for 30 s. Relative expressions were calculated with normalization to β-actin values by using the 2^−ΔΔCt method.²⁵

Western blot analysis

H9c2 cells were washed twice with cold PBS and lysed in RIPA buffer (Solarbio), 1% phenylmethylsulfonyl fluoride (Solarbio), and 1% phosphatase inhibitor cocktail (Solarbio). Cell lysates were clarified by centrifugation at 13,000 × g for 10 minutes at 4°C. Protein concentrations were determined using a bicinchoninic acid protein assay kit (Solarbio). Samples were mixed with sodium dodecyl sulphate-polyacrylamide gel electrophoresis loading buffer (Solarbio) and boiled for 10 minutes. Samples were then separated by sodium dodecyl sulphate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes. The membranes were blocked for 2 hours in 5% skim milk and incubated overnight at 4°C with primary antibodies against Bax (Abcam, Cambridge, UK), Bcl-2 (R&D Systems, Minneapolis, MN, USA), C/EBP homologous protein (CHOP; Cell Signaling, Danvers, MA, USA), pro-Caspase-12 (Proteintech Group, Chicago, IL, USA), cleaved Caspase-12 (Proteintech Group), cleaved Caspase-3 (Cell Signaling), GRP78 (Abcam), GRP94 (Abcam), PDIA6 (Abcam), CAT (Proteintech Group), ATF6 (Proteintech Group), PERK (Proteintech Group), phospho-PERK (Beyotime), IRE1α (Novus Biologicals, Littleton, CO, USA), phospho-IRE1α (Novus Biologicals), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (ImmunoWay, Plano, TX, USA). The anti-ATF6 antibody specifically detects endogenous p90 and p50 ATF6 in cardiomyocytes.¹³ The membranes were then incubated with horseradish peroxidase-conjugated secondary antibody for 2 hours at 4°C and were examined using Immobilon Western Chemiluminescent HRP Substrate (Millipore, Billerica, MA, USA). Gel images were obtained using a Universal Hood II Imager (Bio-Rad, Hercules, CA, USA). Band densities were quantified using ImageJ 1.50e software (National Institutes of Health, Bethesda, MA, USA).

Statistical analysis

Data are expressed as mean ± SD and were assessed using one-way ANOVA followed by Tukey’s post hoc test. H₂O₂ concentrations that induced 50% maximal inhibition (IC₅₀) were determined by performing non-linear regression analysis. All data were analysed using GraphPad Prism 5.01 software (GraphPad Software, San Diego, CA, USA). P < 0.05 was considered statistically significant.

Results

OS injury-associated candidate genes and KEGG analysis

A total of 144 significant genes were identified using the Comparative Toxicogenomics Database, Open Targets Platform, and PolySearch² databases. KEGG analysis showed that these candidate genes were primarily associated with inflammation, the metabolic process, signal transduction, and cellular processes, such as the tumour necrosis factor signalling pathway, nuclear factor-kappa B signalling pathway, non-alcoholic fatty liver disease, protein processing in the endoplasmic reticulum, apoptosis, and the PI3K-Akt signalling pathway.

Putative active ingredients in A. sinensis

To identify the putative active ingredients in A. sinensis, we first determined drug target interactions. A total of 163 ingredients were identified, of which 97 ingredients were screened because their targets could be predicted using the databases. We then mapped the corresponding targets of candidate active ingredients to KEGG pathways. We eliminated 71 ingredients that did not participate in any OS-related pathway. Finally, we constructed the ingredient–pathway–disease network that comprised 37 nodes (26 ingredients and 10 pathways associated with OS injury) and 61 edges (Figure 1). Of the 26 active ingredients, we selected ASP for further validation because its antioxidant role in myocardial cells has not been determined.

Figure 1.

The ingredient–pathway–disease network. The diamond nodes represent ingredients, the circular nodes represent pathways, and the square nodes represent diseases.

ASP inhibits H₂O₂-induced cytotoxicity in H9c2 cells

First, we examined the effects of H₂O₂ on H9c2 cell viability. H9c2 cells were treated with different concentrations (25, 50, 100, 200, 300, and 400 µM) of H₂O₂ for 2, 4, 6, and 8 hours. In the MTT assay, H₂O₂ treatment significantly decreased cell viability in a dose-dependent manner (P < 0.001) (Figure 2a). Based on the IC₅₀ value of H₂O₂ (291 µM) estimated from Figure 2a and similar methods described in a previous study,¹⁶ we used the H₂O₂ concentration of 300 µM and treatment duration of 6 hours in subsequent experiments. Second, we evaluated whether ASP protected H9c2 cells against H₂O₂-induced cytotoxicity. H9c2 cells were pretreated with different concentrations (6.25, 12.5, 25, 50, and 100 µg/mL) of ASP for 4 hours, followed by treatment with 300 µM H₂O₂ for 6 hours. We found that ASP pretreatment protected H9c2 cells against H₂O₂-induced cytotoxicity in a concentration-dependent manner (Figure 2b, P < 0.05 versus H₂O₂ alone). This protective effect of ASP peaked at a concentration of 50 µg/mL, which was then used in subsequent experiments. However, ASP pretreatment alone did not affect cell viability.

Figure 2.

ASP inhibits H₂O₂-induced cytotoxicity in H9c2 cells. Cell viability was determined by performing the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide assay. (a) H9c2 cells were treated with 25, 50, 100, 200, 300, and 400 µM H₂O₂ for 2, 4, 6 and 8 hours. (b) Cells were pretreated with 6.25, 12.5, 25, 50, and 100 µg/mL ASP for 4 hours, followed by incubation with 300 µM H₂O₂ for 6 hours. Data are expressed as mean ± SD (n = 3). #P < 0.05 versus control; *P < 0.05 versus H₂O₂ alone; &P < 0.001 over time. ASP, Angelica sinensis polysaccharide; H₂O₂, hydrogen peroxide.

ASP protects H9c2 cells from H₂O₂-induced apoptosis

To further evaluate the protective effects of ASP, we examined morphological changes in H9c2 cells that were treated with H₂O₂ and/or ASP by using an inverted microscope. We found that H₂O₂ induced cellular shrinkage and nuclear condensation, whereas ASP pretreatment partly ameliorated these morphological changes (Figure 3a). H₂O₂-treated H9c2 cells were then stained with Hoechst 33342 to determine changes in nuclear morphology. This staining showed chromatin condensation and nuclear fragmentation, which are typical morphological changes associated with H₂O₂-induced apoptosis. However, ASP pretreatment alleviated these nuclear morphological changes induced by H₂O₂ (Figure 3b). We performed flow cytometric analysis to quantify the percentage of apoptotic cells. H₂O₂ treatment significantly increased the percentage of apoptotic cells to 41% ± 4% (P < 0.05 versus control), whereas ASP pretreatment significantly decreased the percentage of apoptotic cells to 26% ± 2% (Figure 3c, d; P < 0.05 versus H₂O₂ alone).

Figure 3.

ASP prevents H₂O₂-induced apoptosis of H9c2 cells. H9c2 cells were pretreated with 50 µg/mL ASP for 4 hours, followed by incubation with 300 µM H₂O₂ for 6 hours. (a) Cellular morphology was observed using an inverted microscope. (b) Fluorescence photomicrographs of H9c2 cells stained with Hoechst 33342. (c and d) Apoptosis was assessed by performing flow cytometric analysis after annexin V-FITC/PI double staining, and the number of annexin V-positive cells was quantified. (e) Western blot images of cleaved Caspase-3, Bax, and Bcl-2. (f and g) Densitometry analysis of cleaved Caspase-3 expression and the Bcl-2/Bax ratio. GAPDH was used for normalization. Data are expressed as mean±SD (n = 3). #P <0.05 versus control; *P <0.05 versus H₂O₂ alone. ASP, Angelica sinensis polysaccharide; H₂O₂, hydrogen peroxide; fluorescein isothiocyanate/propidium iodide (FITC/PI), GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Finally, we examined expression of apoptotic regulatory proteins to confirm the protective effect of ASP against H₂O₂-induced apoptosis. H₂O₂ treatment resulted in significantly higher cleaved Caspase-3 levels and the Bax/Bcl-2 ratio compared with those in untreated control cells (both P < 0.05). However, this upregulation was significantly reduced by ASP pretreatment (Figure 3e–g, P < 0.05 versus H₂O₂ alone).

ASP attenuates OS in H9c2 cells

Flow cytometric analysis showed that H₂O₂-treated H9c2 cells showed a greater right shift of mean fluorescence value than did untreated cells, which indicated an increase in intracellular ROS levels. As expected, ROS accumulation was significantly reduced in ASP-pretreated cells (Figure 4a, b; P < 0.05). We also examined the levels of other OS markers in H9c2 cells and found that ASP pretreatment significantly reduced MDA and LDH levels in H₂O₂-treated cells (Figure 4c, d; both P < 0.05).

Figure 4.

ASP alleviates oxidative stress in H9c2 cells. H9c2 cells were pretreated with 50 µg/mL ASP for 4 hours, followed by incubation with 300 µM H₂O₂ for 6 hours. (a and b) ROS accumulation was assessed by performing flow cytometric analysis. (c) Decreased MDA content in H9c2 cells by ASP after exposure to H₂O₂. (d) Inhibition of LDH release in H9c2 cells by ASP after exposure to H₂O₂. Data are expressed as mean ± SD (n = 3). #P < 0.05 versus control; *P < 0.05 versus H₂O₂ alone. ASP, Angelica sinensis polysaccharide; H₂O₂, hydrogen peroxide; ROS, reactive oxygen species; MDA, malondialdehyde; LDH, lactate dehydrogenase.

ASP alleviates ER stress in H9c2 cells

ER stress is considered one of the mechanisms contributing to ROS-mediated cellular apoptosis. Of the ER stress-associated apoptotic molecules, CHOP and cleaved Caspase-12 are the main mediators of apoptosis. Western blot analysis showed that H₂O₂ treatment resulted in significantly higher CHOP and cleaved Caspase-12 levels in H9c2 cells compared with untreated control cells (both P < 0.05), whereas ASP pretreatment reduced the H₂O₂-induced increase in the levels of these proteins (Figure 5a–c; both P < 0.05). These results together with the results of the ingredient–pathway–disease network analysis suggested that ASP pretreatment alleviated ER stress in H₂O₂-treated H9c2 cells.

Figure 5.

ASP alleviates endoplasmic reticulum stress and promotes ATF6 activation and ATF6-induced protein expression in H9c2 cells treated by H₂O₂. (a) Western blot analysis images of CHOP and Caspase-12. (b and c) Densitometry analysis of CHOP and cleaved Caspase-12 expression. (d) Western blot analysis images of p90 ATF6, p50 ATF6, GRP78, GRP94, PDIA6, CAT, IRE1α, p-IRE1α, PERK, and p-PERK. (e–l) Densitometry analysis of relative protein levels normalized to GAPDH levels. H₂O₂ induced p50 ATF6 and its downstream proteins, but not p-IRE1α and p-PERK were increased by ASP. (m and n) Real-time PCR for quantification of Atf6 and Cat mRNA expression. ASP pretreatment increased Atf6 and Cat mRNA expression levels in H9c2 cells treated by H₂O₂. Data are expressed as mean ± SD (n = 3). #P < 0.05 versus control; *P < 0.05 versus H₂O₂ alone. ASP, Angelica sinensis polysaccharide; H₂O₂, hydrogen peroxide; ATF6, activating transcription factor 6; CHOP, C/EBP homologous protein; p90 ATF6, 90-kDa ATF6; p50 ATF6, 50-kDa ATF6; GRP78, 78-kDa glucose-regulated protein; GRP94, 94-kDa glucose-regulated protein; PDIA6, protein disulphide isomerase family A, member 6; CAT, catalase; IRE1α, inositol-requiring enzyme 1α; p-IRE1α, phosphorylated-IRE1α; PERK, pancreatic eIF2-alpha kinase; p-PERK, phosphorylated-PERK, GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

ASP preferentially activates UPR signalling pathways in H9c2 cells

Western blot analysis showed that H₂O₂ treatment activated three UPR signalling pathways (ATF6, IRE1α, and PERK) compared with untreated control cells (P < 0.05 versus control) (Figure 5d–l). Compared with H₂O₂ alone, pretreatment with ASP further enhanced p50 ATF6 expression (P < 0.05). However, there was no significant effect of pretreatment with ASP on phosphorylation of IRE1α and PERK, which suggested that ASP could selectively activate the ATF6 branch in the UPR. To verify activation of the ATF6 branch, we evaluated the expression of the well-known ATF6-induced proteins GRP78, GRP94, PDIA6, and CAT. Western blot analysis showed that GRP78, GRP94, PDIA6, and CAT levels were further significantly elevated following ASP pretreatment of H₂O₂-treated H9c2 cells (all P < 0.05). ASP alone had no effect on the three UPR signalling pathways. Additionally, qRT-PCR analysis showed that ASP pretreatment resulted in significantly higher mRNA expression levels of Atf6 and Cat in H₂O₂-treated H9c2 cells (Figure 5m, n; both P < 0.05 versus H₂O₂ alone). Taken together, these data suggested that ASP promoted activation of the ATF6 pathway to protect H9c2 cells from H₂O₂-induced injury.

ATF6 plays a pivotal role in ASP-induced cytoprotection

To further confirm the involvement of the ATF6 pathway in ASP-induced protective effects in H9c2 cells, we transfected H9c2 cells with siAtf6 to silence ATF6 expression. Western blot analysis showed significant downregulation of ATF6 expression in siAtf6-transfected H9c2 cells compared with H9c2 cells transfected with a nontargeted siRNA (siCon) and non-transfected control cells (Figure 6a, b; both P < 0.05). The MTT assay showed that Atf6 knockdown diminished the protective effect of ASP in H₂O₂-treated H9c2 cells (Figure 6c, P < 0.05 versus siCon + H₂O₂ + ASP)_. Flow cytometric analysis for quantifying the percentage of apoptotic cells showed that the beneficial effects of ASP were reduced in the presence of siAtf6 (Figure 6d, e). Similar results were also obtained in OS-related marker analysis (Figure 6f–h). These results indicated that the ATF6 pathway played a pivotal role in ASP-induced cytoprotection.

Figure 6.

Atf6 knockdown diminishes ASP-induced cytoprotection. (a and b) H9c2 cells were transfected with a nontargeted siRNA (siCon) and a siRNA directed against Atf6 (siAtf6). Non-transfected control cells are also shown. Western blot and densitometry analyses show significant Atf6 knockdown in H9c2 cells transfected with siAtf6. (c–h) H9c2 cells were transfected with siAtf6 or siCon, followed by treatment with 300 µM H₂O₂ for 6 hours with or without 50 µg/mL ASP pretreatment. (c) Results of the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide assay showing that Atf6 knockdown diminished the protective effect of ASP in H₂O₂-treated cells. (d and e) Apoptosis was assessed by performing flow cytometric analysis and annexin V-positive cells were quantified. (f) Intracellular ROS levels were assessed by flow cytometry. (g) MDA content in H9c2 cells. (h) LDH release from H9c2 cells. Data are expressed as mean ± SD (n = 3). #P < 0.05 versus siCon group; $P < 0.05 versus control group; *P < 0.05 versus siCon + H₂O₂ group; &P < 0.05 versus siCon + H₂O₂ + ASP group. siAtf6, siRNA directed against activating transcription factor 6; H₂O₂, hydrogen peroxide; ASP, Angelica sinensis polysaccharide; ROS, reactive oxygen species; MDA, malondialdehyde; LDH, lactate dehydrogenase.

Discussion

A. sinensis has been used in more than 80 composite formulae of traditional Chinese medicines.¹⁵ A. sinensis exerts various pharmacological effects, including antioxidative, anti-inflammatory, anti-atherosclerotic, antihypertensive, and immunomodulatory effects.¹⁵ However, the main active ingredients and signalling pathways underlying the beneficial effects of A. sinensis against H₂O₂-induced injury remain unclear. Analysis of publicly available web-based databases can provide insight on potential active ingredients of herbs.^17–20 In the present study, we identified 26 active ingredients and their corresponding biological pathways by analysing multiple web-based integrated databases. The 26 active ingredients were classified as phthalides, organic acids, and polysaccharides.¹⁵ The individual ingredients of A. sinensis acted on the same pathways, thus exerting a synergistic therapeutic effect against oxidative injury. Although these computational analyses identified active ingredients of A. sinensis, the accuracy of the target prediction tools may affect efficacy of the obtained results. To confirm the efficiency of our screening protocol, we selected ASP for further validation.

Previous studies have shown that ASP possesses different pharmacological activities and has relatively low toxicity.²⁶ ASP exerts hepatoprotective effects by regulating amino acid, energy, and lipid metabolism.²⁷ ASP exerts neuroprotective effects by limiting cerebral ischaemia/reperfusion injury,²⁸ increasing microvessel number, and improving blood flow after cerebral ischaemia.²⁹ The antioxidant activity of ASP has been tested in in vivo and in vitro experimental models. Zhuang et al. found that ASP protected chondrocytes from H₂O₂-induced injury by exerting antioxidant, anti-apoptotic, and anti-inflammatory effects.³⁰ Lei et al. showed that ASP attenuated an H₂O₂-induced increase in apoptosis and ROS levels in PC12 neuronal cells, and enhanced the activity of antioxidant enzymes in cortical tissues of rats with focal cerebral ischaemia.²⁹ Additionally, ASP treatment reduced the senescence of endothelial progenitor cells by relieving OS.³¹ However, the effect of ASP on oxidative injury in cardiomyocytes and mechanisms underlying this effect are unclear. H9c2 is a clonal cell line derived from BD1X rat ventricle and exhibits many of the properties of cardiac muscle, such as electrophysiology, ion channels, and receptors.^32,33 H₂O₂-induced model of H9c2 cells has been frequently used to determine the pathogenesis of cardiac oxidative injury and its treatment strategy.³⁴ In the present study, we found that treatment with 300 µM H₂O₂ was optimal for inducing oxidative damage in H9c2 cells. Moreover, we found ASP exerted maximal protective effects at a concentration of 50 µg/mL. H9c2 cells treated with H₂O₂ alone showed increased ROS, MDA, and LDH levels, which indicated induction of oxidative injury. However, this increase in ROS, MDA, and LDH levels was attenuated in ASP-pretreated H9c2 cells. Moreover, ASP pretreatment increased the level of the antioxidant enzyme CAT in H₂O₂-treated cells, which indicated that ASP functioned as a ROS scavenger. These results suggest that ASP has antioxidant effects in H9c2 cells.

ROS directly damage cell structure, nucleic acids, lipids, and proteins, and trigger a series of events to exacerbate cellular injury.^1–4 In the present study, KEGG analysis of candidate OS-associated genes showed that these targets were mostly involved in intracellular signalling cascades associated with inflammation, metabolism, and apoptosis. We found that H9c2 cells exposed to H₂O₂ showed a significant increase in cellular apoptosis, as shown by morphological changes, Hoechst 33342 staining, and fluorescein annexin V-FITC/PI double labelling. However, ASP pretreatment significantly reduced the proportion of apoptotic cells. These observations are consistent with the results of previous studies in other cell types showing the anti-apoptotic effect of ASP.^29,30

Recently, excessive generation of ROS has been proposed to affect ER homeostasis and interfere with correct protein folding, ultimately initiating ER stress-mediated apoptotic pathways to aggravate cardiomyocyte injury.^4,5 The pro-apoptotic transcription factor CHOP is strongly induced under conditions of severe ER stress.³⁵ Activation of CHOP alters apoptosis-related proteins, such as Bax and Bcl-2.³⁵ Caspase-12 is another major mediator of ER stress-induced apoptosis.³⁶ Caspase-12 activation indirectly activates cytoplasmic Caspase-3, which is a major effector protease that induces cell death.³⁶ We found that CHOP, cleaved Caspase-12, Bax, Bcl-2, and cleaved Caspase-3 levels were significantly increased in H₂O₂-treated H9c2 cells. In contrast, ASP pretreatment decreased the expression levels of these apoptosis-related proteins. These results indicate that the protective role of ASP against H₂O₂-induced injury involves inhibition of ER stress-mediated apoptosis.

Initially, ER stress triggers an adaptive UPR process to restore ER homeostasis. However, if ER stress is not resolved, the UPR represses the adaptive response and triggers apoptosis. Three ER transmembrane proteins, namely ATF6, IRE1α, and PERK, mobilize the UPR by targeting different genes.⁹ Differences in structure and activation kinetics of these three ER transmembrane proteins allow differential regulation of individual components of the UPR to alleviate ER stress.⁹ Some studies have examined the strategy to preferentially activate UPR signalling pathways. Wang et al. found that ginkgolide K, a natural product isolated from Ginkgo biloba, protected the heart against tunicamycin-induced ER stress by selectively activating the IRE1α signalling branch.³⁷ Selective activation of the PERK branch plays an important role in restoring cellular homeostasis against β-amyloid-induced ER stress.³⁸ In contrast to the IRE1α and PERK branches, the ATF6 pathway of the UPR exerts a major role in responding to ROS-mediated ER stress during myocardial ischaemia/reperfusion.¹⁴ Upregulated ATF6 expression exerts protective effects against ischaemia/reperfusion injury in in vivo and ex vivo models and in cultured cardiomyocytes.³⁹ Studies using a transgenic mouse model have shown that selective activation of the ATF6 branch of UPR protects the heart from ischaemia/reperfusion damage.^13,40 Once activated, ATF6 binds the ER stress response element, resulting in increased expression of ATF6-induced ER proteins, such as GRP78, GRP94, and PDIA6.^10–12 Previous studies reported that preinduced GRP78 protected cardiomyocytes from OS-induced cell death¹⁰ and that GRP94 overexpression in H9c2 cells reduced stress-induced cell death.¹¹ Gain- and loss-of-function studies have shown that PDIA6 protects cardiomyocytes against ischaemia/reperfusion-induced cell death by enhancing ER protein folding.¹² Transgenic mice that expressed a mutant KDEL receptor, which decreased ER-resident protein levels, were associated with increased ER stress and dilated cardiomyopathy.⁴¹ Moreover, a study on Atf6 gene deletion in the mouse heart showed that endogenous ATF6 had an important protective function against ischaemia/reperfusion injury because it was required for induction of numerous antioxidant proteins, such as CAT.¹³ CAT neutralizes large quantities of ROS generated in various cellular locations by converting H₂O₂ into O₂ and water, thereby alleviating OS injury.⁴² In the present study, ASP pretreatment activated ATF6 and its downstream genes in H₂O₂-treated H9c2 cells, as shown by conversion of p90 ATF6 to p50 ATF6 and increased levels of the ATF6 target proteins GRP94, GRP78, PDIA6, and CAT. We further found that knockdown of Atf6 by using siAtf6 partially abrogated the protective role of ASP in H9c2 cells treated by H₂O₂. The results of the siRNA experiments support a causal connection between ATF6 activation and resistance to H₂O₂-induced injury. Although further animal studies are required, our data provide new insight on the molecular mechanism underlying the cardioprotective effects of ASP.

In summary, our findings indicate that ASP, the major water-soluble component of A. sinensis, exerts its protective effects against H₂O₂-induced injury in H9c2 cells by selectively activating the ATF6 pathway, thereby ameliorating ER stress and oxidative stress.

Ethics and consent statements

This study did not use animal models and involve humans, their data, or tissue. The H9c2 cell line in the present study was obtained from the China Center for Type Culture Collection, where collection of cell culture was approved by the Institutional Animal Care and Use Committee.

Declaration of conflicting interest

The authors declare that there is no conflict of interest.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.