Hesperetin derivative 2a inhibits lipopolysaccharide-induced acute liver injury in mice via downregulation of circDcbld2

Abstract

Acute liver injury (ALI) is a complex, life-threatening inflammatory liver disease, and persistent liver damage leads to rapid decline and even failure of liver function. However, the pathogenesis of ALI is still not fully understood, and no effective treatment has been discovered. Recent evidence shows that many circular RNAs (circRNAs) are associated with the occurrence of liver diseases. In this study we investigated the mechanisms of occurrence and development of ALI in lipopolysaccharide (LPS)-induced ALI mice. We found that expression of the circular RNA circDcbld2 was significantly elevated in the liver tissues of ALI mice and LPS-treated RAW264.7 cells. Knockdown of circDcbld2 markedly alleviates LPS-induced inflammatory responses in ALI mice and RAW264.7 cells. We designed and synthesized a series of hesperidin derivatives for circDcbld2, and found that hesperetin derivative 2a (HD-2a) at the concentrations of 2, 4, 8 μM effectively inhibited circDcbld2 expression in RAW264.7 cells. Administration of HD-2a (50, 100, 200 mg/kg. i.g., once 24 h in advance) effectively relieved LPS-induced liver dysfunction and inflammatory responses. RNA sequencing analysis revealed that the anti-inflammatory and hepatoprotective effects of HD-2a were mediated through downregulating circDcbld2 and suppressing the JAK2/STAT3 pathway. We conclude that HD-2a downregulates circDcbld2 to inhibit the JAK2/STAT3 pathway, thereby inhibiting the inflammatory responses in ALI. The results suggest that circDcbld2 may be a potential target for the prevention and treatment of ALI, and HD-2a may have potential as a drug for the treatment of ALI.

Introduction

The incidence of liver disease has gradually increased worldwide making it a global public health problem [1, 2]. Clinical studies have found that liver disease is a continuous and progressive pathophysiological process, and acute liver injury (ALI) is a key early stage that can be targeted to prevent exacerbation of the disease [3]. ALI is induced by a variety of liver damage factors, including viral infections, drugs and toxins [4, 5]. The occurrence and development of ALI are related to the inflammatory response, abnormal regulation of initial widespread immune activation and systemic inflammatory response syndrome. Liver macrophages are innate immune cells that play an important role in the regulation of ALI [6–8]. Under pathological conditions, liver-resident macrophages are activated through the secretion of inflammatory factors and chemokines into the microenvironment [9]. Such factors promote the recruitment of peritoneal macrophages and bone marrow mononuclear macrophages to the liver, which differentiate into inflammatory Ly6C high MoMϕs and aggravate liver tissue injury [3, 9]. Timely intervention during ALI can limit the extent and scope of liver injury and prevent ALI from developing into acute liver failure [10, 11]. Therefore, it is crucially important to explore the mechanisms of ALI occurrence and progression and to develop targeted drugs to treat liver diseases.

Circular RNAs (circRNAs) comprise a class of non-coding RNAs with covalently closed single-stranded loop conformations that are produced by direct backsplicing or exon skipping of precursor mRNAs. They lack 5′ and 3′ ends, which makes them exonuclease resistant and more stable than linear RNAs, resulting in longer half-lives in vivo [12–14]. In normal physiology, the expression of circRNAs is temporally and spatially controlled, while in certain disease states characteristic circRNA expression patterns are observed. This disease-specific expression can serve as an indicator for disease diagnosis and evaluation [15, 16]. In recent years, disrupted expression of numerous circRNAs has been found to be associated with liver diseases. For example, circGFRA1 is upregulated in hepatocellular carcinoma tissues and can promote hepatocellular carcinoma progression by modulating the microRNA-498/NAP1L3 axis [17]. Another study found that circFBXW4 is significantly downregulated in liver fibrosis, while overexpression of circFBXW4 can reduce fibrosis damage and inflammation in the liver [18]. circRNAs may therefore represent a novel biomarker and therapeutic target for hepatocellular carcinoma, hepatic fibrosis and nonalcoholic fatty liver disease [18–21]. To search for potential therapeutic ALI targets, we sequenced circRNAs from mouse primary macrophages and found that 91 circRNAs were dysregulated following lipopolysaccharide (LPS) treatment. Further screening of liver tissues from ALI model mice, revealed significant upregulation of circDcbld2, while downregulation of circDcbld2 in ALI mice inhibited inflammation. These results indicate that circDcbld2 is a potential target for alleviating liver inflammation and treating ALI.

Hesperidin derivatives are traditional Chinese medicine monomer derivatives isolated from the peel of Rutaceae citrus plants (Citrus reticulata Blanco) [22]. They are flavonoid compounds that have various pharmacological effects, such as anti-tumor [23, 24], anti-inflammatory [25, 26] and anti-oxidation [27] actions and are promising drugs for the treatment of ALI. Our research group has long been committed to the study of hesperidin derivatives to treat liver diseases [28–30]. Small changes in hesperidin derivative chemical structure may lead to significant changes in biological activity. Therefore, we designed and synthesized a series of hesperidin derivatives against circDcbld2 and demonstrated that one hesperidin derivative, HD-2a, significantly inhibited the expression of circDcbld2.

In this study, we found that a novel circRNA, circDcbld2, was significantly upregulated in ALI, and was formed by reverse splicing of exons 2–3 of the host gene Dcbld2. Functionally, silencing circDcbld2 inhibited inflammation and alleviated hepatic injury in mice, indicating that circDcbld2 may be a biomarker for ALI progression. Among a series of hesperidin derivatives synthesized against circDcbld2, we found that HD-2a significantly inhibited the expression of circDcbld2 and played an anti-inflammatory and hepatoprotective role. Mechanistically, RNA sequencing (RNA-seq) indicated that the JAK2/STAT3 pathway plays a key role in HD-2a-mediated anti-inflammatory and hepatoprotective effects. In summary, we report the expression profile, function and inhibitor of circDcbld2 in ALI and suggest that circDcbld2 is a promising biomarker for ALI therapy.

Materials and methods

Animal studies

C57BL/6 male mice (6–8 weeks of age, weighing approximately 20–24 g) were supplied by the Experimental Animal Center of Anhui Medical University. Mouse ALI model was established as previously described [31]. Mice were randomly allocated to two groups, food and water were freely available throughout this study. LPS was dissolved in phosphate buffered saline (PBS). The control group was administered PBS, the ALI model group was injected intraperitoneally at a dose of 20 mg/kg LPS. Mice were sacrificed 6 h after LPS challenge. All animals were treated humanely, and all animal procedures met the relevant legal and ethical requirements according to the protocols (LLSC20220599) approved by the Institutional Animal Care and Use Committee of Anhui Medical University.

circDcbld2 knockdown

Luciferase-labeled specific liver tissue location of AAV8-circDcbld2 and vector were designed and synthesized by Hanbio Biotechnology (Shanghai, China). AAV and vector (1 × 10¹² vg/ml), diluted in saline, were injected into the tail vein of mice, respectively. Two weeks after AAV administration, mouse model of acute liver injury was established. Mice exposed to AAV delivery were anaesthetized with isoflurane and injected intraperitoneally luciferin solution. Images were acquired using an IVIS Lumina III Imaging System (Caliper Life Sciences, USA) to confirm the location of AAV-circDcbld2 on liver tissues. Mice were sacrificed after indicated treatment.

HD treatment

Mice were randomly divided into six groups (control, ALI, ALI + HD 25 mg/kg, ALI + HD 50 mg/kg, ALI + HD 100 mg/kg, ALI + silymarin 100 mg/kg). The control group was administered PBS, the ALI group was injected intraperitoneally at a dose of 20 mg/kg LPS, three HD groups received different doses of HD (25, 50, 100 mg/kg) by intragastric administration 24 h in advance, positive control group mice were exposed to silymarin (100 mg/kg) by intragastric administration 24 h in advance. Mice were sacrificed 6 h after LPS injection. The whole blood sample and liver tissue samples were harvested, liver tissues were paraformaldehyde-fixed and paraffin-embedded.

Cell culture and treatment

RAW264.7, the mouse macrophage cell line, were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, USA), 1% penicillin and 1% streptomycin. The cell culture flasks were all placed in a humidified 5% CO₂ incubator at 37 °C. The RAW264.7 cells were seeded in 6-well plates (NEST Biotechnology, China) 4 × 10⁵ cells per ml. LPS at concentration of 1000 ng/ml was added to the cell culture for 12 h.

RNA sequencing (RNA-seq)

The RNA of total samples was isolated and purified by TRIzol (thermofisher, 15596018). The amount and purity of total RNA was then quality-controlled by NanoDrop ND-1000 (NanoDrop, Wilmington, DE, USA) and RNA integrity was detected by Bioanalyzer 2100 (Agilent, CA, USA). Poly(A) RNA was subsequently purified by using PolyTtract mRNA Isolation System and used to generate cDNA libraries. Finally, all samples were sequenced using illumina Novaseq^TM 6000 (LC Bio Technology CO., Ltd. Hangzhou, China) inaPE150 mode according to standard operations.

Liver histological and immunohistological staining

Paraffin-embedded and 4% paraformaldehyde-fixed liver tissues were sectioned at 4 μm thickness for hematoxylin and eosin (H&E) staining and immunohistochemical (IHC) staining of F4/80 (Servicebio, China). Slides are scanned by a digital slide scanner (Pannoramic MIDI; 3DHISTECH Hungary) and analyzed by the CaseViewer software.

Immunofluorescence staining

Cultured RAW264.7 cells were induced by LPS (1000 ng/ml) and treated with siRNA-circDcbld2 or different concentrations of HD-2a to determine the expression of TNF-α in vitro. First, RAW264.7 cells were seeded in a six well plate and fixed with 4% paraformaldehyde for 15 min at room temperature. Then, RAW264.7 cells were blocked with 10% bovine serum albumin for 1 h at room temperature. Cell immunofluorescence staining was performed according to the above steps, while paraffin sections of liver tissue were treated with dewaxing and antigen repair before staining. In addition, anti TNF-α antibodies (1:5000; Abcam, UK) were added to RAW264.7 cells or liver tissues and incubated at 4 °C overnight. Next, corresponding secondary antibodies were conjugated with fluorescein isothiocyanate (1:200; ZSGB Bio, China) for 2 h without light at room temperature. Finally, the nuclei were counterstained by using 4′,6-diamidino-2-phenylindole (Beyotime, China). The positive expression of TNF-α was observed by using inversion fluorescence microscopy.

Transient transfection of RAW264.7 cells

When the density of RAW264.7 cells was between 50% and 60%, according to the manufacturer’s certificate, pcDNA-m-circDcbld2 (Hanbio Biotechnology, Shanghai, China) overexpressed plasmids and negative controls were transfected into activated RAW264.7 cells using Lipofectamine™ 3000 (Life Technologies, USA) during serum-free transfection. After 6 h transfection, the culture medium was replaced with fresh medium for additional 12 h incubation. Cells were collected after LPS and HD-2a treatment for Western blot, qRT-PCR and other experiments. All the experiments were carried out in triplicate. Small interfering RNA (siRNA) of circDcbld2 and negative control was synthesized by Hanbio Biotechnology (Shanghai, China). Transfection of siRNA-circDcbld2 and siRNA-NC to silence circDcbld2 expression was performed in a similar manner. Listed in the analysis Table 1 are the sequences used for transient transfections.

Table 1.

pcDNA and siRNA sequences used in this study.

Genes	Forward primer (5′-3′)	Reverse primer (5′-3′)
pcDNA-m-circDcbld2	atgactttttttttatacttcagGTGATGGATGTGGACACACTGT	taattcttttccttgcttcttacCTTCTTTATCTATAACTGAGTAT
siRNA-NC	UUCUCCGAACGUGUCACGUTT	ACGUGACACGUUCGGAGAATT
siRNA-circDcbld2	AGAAGGUGAUGGAUGUGGACAdTdT	UGUCCACAUCCAUCACCUUCUdTdT

DNA sequencing

The RNA was reverse-transcribed into cDNA using PrimeScript RT Master Mix (Takara, Japan). Polymerase chain reaction (PCR) was performed using 2×Taq Master Mix (Takara, Japan) from the manufacturer’s protocol. PCR products were identified by DNA sequencer (ABI3730XL, USA).

ELISA

Serum levels of MCP-1, IL-1β and TNF-α were detected by ELISA Kit (ABclonal, Wuhan, China) according to the manufacturer’s instructions. The absorbance of each well at 450 and 570 nm was measured by using a Thermomax microplate reader (BIO-TEKEL, USA). At least nine independent experiments were performed.

ALT/AST assays

The serum was collected with centrifugation from the whole blood sample with 1000 × g for 30 min. The activities of serum alanine aminotransferase/aspartate aminotransferase (ALT/AST) are measured by using the ALT/AST Assay Kits (Nanjing jiancheng Biotech, China) according to the manufacturer’s instructions.

CCK-8 assay

Cell count KIT-8 (CCK-8) assay was performed to determine the proliferation of RAW264.7 cells. The transfected cells were treated with 10 μl of CCK-8 in 96-well plate for 4 h at 37 °C in a cell incubator. Thereafter, 100 μl of DMSO was added and absorbance was measured at 452 nm. Cell viability (%) = (measured value − blank value)/(control value − blank value) × 100%. Blank values are those without CCK-8 dye added.

RNA extraction and real-time quantitative PCR

Total RNA was extracted from liver tissues and RAW264.7 cells using the TRIzol Reagent (Invitrogen, USA) and quantified by NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA). Then complementary DNA (cDNA) was achieved using Prime Script RT Master Mix (Accurate Biotechnology, China). Finally, Real-time quantitative PCR (qRT-PCR) analysis was performed using Premix EX Taq (Accurate Biotechnology, China) in CFX96 Real-time quantitative PCR system (Bio-Rad, USA). Primers sequences used in this study were listed in Table 2. The ratio for the mRNA interested was normalized with β-actin. The primers were synthesized by Sangon Biotech (Shanghai, China).

Table 2.

Primer sequences used in real-time PCR.

Genes	Forwad primer (5′–3′)	Reverse primer (5′–3′)
β-actin	TGCTTCACCACCTTCTTGAT	TGGAAAGCTGTGGCGTGAT
MCP-1	GGGCCTGCTGTTCACAGTT	CCAGCCTACTCATTGGGAT
TNF-α	CACCACCATCAAGGACTCAA	AGGCAACCTGACCACTCTCC
IL-1β	TGAGTGATACTGCCTGCCTG	CTTTGAAGTTGACGGACCC
circDcbld2	AGTCAGCAGAACGGAAATAGGCA	GTGCTGTTAGGATAGGTATGTGGGT

Western blot

The protein lysates from liver tissues and RAW264.7 cells were extracted with RIPA-buffer containing 1% PMSF on ice for 30 min, and then the samples were collected and centrifuged for 30 min at 12,000 r/min. The protein buffer was added to the up-clearing and denatured in a metal heater for 10 min, then electrophorised on 10% SDS PAGE gel and transferred to PVDF membrane. After the membranes were incubated with 5% milk for 2 h to block nonspecific binding, then the membranes were incubated with primary antibodies including JAK2 (1:2000, Abcam, UK), p-JAK2 (1:1000, Abcam, UK), STAT3 (1:1000, Abcam, UK) and p-STAT3 (1:1000, Bioss, China) overnight at 4 °C. Membranes were incubated with secondary antibody for 1 h at room temperature. Bands were visualized by enhanced chemiluminescence system (Bio-Rad, USA). Signal intensities of each Western blot were quantified by using the Image J software (NIH, Bethesda, MD, USA) and normalized to β-actin as internal control.

Statistical analysis

All data collected from this study were expressed as mean ± SD and analyzed using one-way analysis of variance (ANOVA), followed by Newman–Keuls post-hoc test (Prism 8.0 GraphPad Software, USA). *P < 0.05 indicated that the difference was statistically significant.

Results

circDcbld2 is upregulated in an ALI mouse model and LPS-induced RAW264.7 cells

Macrophages play an important role in the occurrence and development of ALI. Our previous analysis of circRNA expression in inflammatory macrophages showed abnormal circDcbld2 expression. To further study changes in the expression of circDcbld2 in LPS-induced ALI, we successfully established an LPS-induced ALI mouse model (Supplementary Fig. S1a–e). circDcbld2 (mmu_circ_0000693) is derived from the host gene Dcbld2. The genomic structure of the Dcbld2 locus suggests that circDcbld2 consists of 2 exons (exons 2 and 3) from the Dcbld2 locus located on chromosome chr16: 58424673–58433576 (Fig. 1a). qRT-PCR showed the levels of circDcbld2 expression to be significantly increased in liver tissues of ALI mice compared with those in normal mice (Fig. 1b). Consistent with this in vivo experiment, circDcbld2 was upregulated in LPS-induced RAW264.7 cells compared with normal RAW264.7 cells (Fig. 1c). These results suggested that circDcbld2 as a new potential target for regulating the inflammatory response and LPS-induced ALI.

Fig. 1. The abnormal expression of circDcbld2 is associated with the development of ALI in mice.

a Schematic diagram showed the genomic location and back-splicing pattern of circDcbld2. b qRT-PCR analyses of circDcbld2 RNA levels in liver tissues, n = 6. c qRT-PCR analyses of circDcbld2 RNA levels in RAW264.7 cells, n = 3. d Luciferase-labeled AAV-circDcbld2-KD vector. e The liver-tissue-specific location of AAV-circDcbld2-KD was shown by in vivo imaging analysis. f Levels of circDcbld2 in mice with AAV-circDcbld2-KD administration were detected by qRT-PCR, n = 6. g, i H&E staining and immunohistochemical staining of F4/80 in liver tissues. The positive staining areas were measured by Ipwin32 software, magnification, ×5, n = 6. (h) Serum concentrations of ALT and AST, n = 6. j mRNA levels of inflammatory factors IL-1β, TNF-α and MCP-1 were detected by qRT-PCR, n = 6. Data are represented as mean ± SD. ***P < 0.001. KD knockdown.

Knockdown of circDcbld2 alleviated LPS-induced inflammation and liver injury in ALI mice

We further evaluated the role of circDcbld2 in ALI model mice. Luciferase-labeled AAV8-circDcbld2 was injected into the tail vein of mice, and empty vector was used as a control (Fig. 1d). Liver tissue-specific circDcbld2-knockdown (KD) was confirmed by in vivo imaging (Fig. 1e), and the circDcbld2-knockdown efficiency was confirmed in mouse liver tissue (Fig. 1f). Compared with control mice, hematoxylin and eosin (H&E) staining showed attenuation of hepatocyte necrosis and disordered liver tissue structure and F4/80 immunohistochemistry showed reduced inflammatory cell infiltration in circDcbld2-KD-treated ALI model mice (Fig. 1g, i). Consistently, the serum levels of both alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were substantially lower in circirDcbld2-KD-treated mice, indicating reduced liver damage (Fig. 1h). In addition, circirDcbld2-KD reduced the mRNA levels of the inflammatory factors, IL-1β, TNF-α, and MCP-1, in ALI model mice (Fig. 1j). Collectively, these data showed that liver-specific knockdown of circDcbld2 suppresses inflammation and liver tissue damage.

circDcbld2 regulates inflammatory responses of RAW264.7 cells

We then investigated the biological function of circDcbld2 in RAW264.7 cells. First, we constructed a small interfering RNA (siRNA) to target the circDcbld2 looped junction (siRNA-circDcbld2) and demonstrated its silencing efficiency after transfection of RAW264.7 cells (Fig. 2a). When RAW264.7 cells were transfected with siRNA-circDcbld2, qRT-PCR showed that mRNA levels of IL-1β, TNF-α, and MCP-1 were significantly decreased compared with the LPS+si-NC group (Fig. 2b). Conversely, overexpression of circDcbld2 promoted inflammatory responses in RAW264.7 cells (Fig. 2c, d). Furthermore, ELISA revealed that downregulation of circDcbld2 in RAW264.7 cells significantly inhibited the secretion of inflammatory cytokines (Fig. 2e). The immunofluorescence signal of TNF-α was significantly decreased in circDcbld2-silenced LPS-induced RAW264.7 cells compared with that in LPS-induced RAW264.7 cells (Fig. 2f). These results confirmed that circDcbld2 is an important regulator of LPS-induced liver injury and inflammatory response. We therefore propose that drugs to regulate circDcbld2 should be urgently developed to reduce the inflammatory response in ALI.

Fig. 2. Inhibition and overexpression of circDcbld2 regulated LPS-induced inflammatory responses in RAW264.7 cell.

a siRNA was designed specially targeted to the back-splicing site of circDcbld2. b Levels of circDcbld2 and mRNA levels of IL-1β, TNF-α and MCP-1 in RAW 264.7 cells transfected with Si-circDcbld2 were detected by qRT-PCR, n = 3. c Establishment of pcDNA-circDcbld2 overexpressing RAW264.7 cells. d Levels of circDcbld2 and mRNA levels of IL-1β, TNF-α and MCP-1 in RAW264.7 cells with pcDNA-circDcbld2-OE administration were detected by qRT-PCR, n = 3. e ELISA analysis of circulation level of pro-inflammatory cytokines TNF-α, IL-1β and MCP-1 in RAW264.7 cell supernatant, n = 3. f Immunofluorescence analysis of TNF-α (green) expression in RAW264.7 cells, and the positive staining areas were measured by Ipwin32 software. Representative views were presented, magnification, ×20. Data are represented as mean ± SD. **P < 0.01, ***P < 0.001.

HD-2a effectively inhibits circDcbld2

After determining that circDcbld2 regulates the occurrence of ALI, we focused on identifying compounds that regulate circDcbld2. Among a series of drugs designed and synthesized against circRNAs, hesperetin derivative 2a (HD-2a) exhibited strong anti-inflammatory activity. Figure 3a shows the chemical formula of HD-2a. Cell Counting Kit-8 assay results showed that 8.0 μM HD-2a had a limited suppressive effect on the viability of RAW264.7 cells (Fig. 3b). We next showed that different safe doses of HD-2a effectively inhibited circDcbld2 expression and that different doses of HD-2a reversed LPS-induced upregulation of circDcbld2 in mice and RAW264.7 cells (Fig. 3c, e). Transfection of siRNA-circDcbld2 as a control showed that HD-2a effectively inhibited the expression of circDcbld2 (Fig. 3d). The above studies confirmed that HD-2a has potential as an inhibitor of circDcbld2. Therefore, the role and mechanism of action of HD-2a in inhibiting the expression of circDcbld2 in LPS-induced liver injury and inflammation are worthy of further investigation.

Fig. 3. HD-2a inhibited circDcbld2 expression to alleviate liver injury in ALI mice.

a Molecular structural formula of Hesperetin derivative 2a (HD-2a). b Cells vitality of different concentrations of HD-2a on RAW264.7 cells by CCK-8 analyzed, n = 3. c RNA level of circDcbld2 in RAW264.7 cells with different doses of HD-2a, n = 3. d qRT-PCR analyses of circDcbld2 RNA levels in RAW264.7 cells with HD-2a treatment and siRNA-circDcbld2 transfection, n = 3. e RNA level of circDcbld2 in liver tissues from mice with different doses of HD-2a, n = 6. f Liver function was assessed by serum levels of ALT and AST in mice, n = 6. g Paraffin-embedded sections of liver tissue from six group mice stained with H&E staining. Representative views were presented, magnification, ×5. Data are represented as mean ± SD. **P < 0.01, ***P < 0.001.

Hepatoprotective effects of HD-2a in ALI mice

To further verify the efficacy of HD-2a in ALI, we administered different doses of HD-2a to ALI model mice. Serum levels of ALT and AST, biochemical indicators of liver injury, were significantly increased in LPS-treated mice compared with the control group, but were significantly decreased after pretreatment with HD-2a (25, 50 and 100 mg/kg) (Fig. 3f). To assess histopathology, H&E staining showed that the liver tissue of mice in the normal group was arranged in an orderly manner, while that of the ALI model group exhibited liver cord structure disorder, with inflammatory infiltration, hepatocyte space enlargement and other features of liver injury (Fig. 3g). HD-2a treatment (25, 50 and 100 mg/kg) mitigated LPS-induced liver damage in a dose-dependent manner.

Anti-inflammatory effects of HD-2a on ALI mice

Real-time quantitative PCR was used to detect the RNA levels of circDcbld2 and related inflammatory cytokines in the liver of mice in each group. This confirmed that HD-2a can inhibit the expression of circDcbld2, thereby inhibiting the production of TNF-α, IL-1β and MCP-1 (Fig. 4a). Meanwhile, immunohistochemistry showed that F4/80⁺ macrophage infiltration was significantly reduced in ALI mice pretreated with HD-2a (Fig. 4b). These results indicated that HD-2a inhibits circDcbld2, reducing liver inflammation in ALI mice. Next, we showed that the serum levels of three pro-inflammatory factors, TNF-α, IL-1β and MCP-1, were significantly reduced in the HD-2a-treated group compared with the ALI group (Fig. 4c). Double-immunofluorescence staining produced consistent results; compared with liver sections from ALI mice, those from HD-2a-treated mice showed significantly decreased macrophage infiltration and TNF-α fluorescence (Fig. 4d). In summary, we demonstrated that HD-2a can inhibit the expression of circDcbld2 to reduce inflammation and protect the liver. HD-2a is therefore a promising compound for targeting circDcbld2 to treat ALI.

Fig. 4. Anti-inflammatory effects of HD-2a on mice with acute liver injury.

a qRT-PCR analyses of MCP-1, TNF-α and IL-1β mRNA levels in liver tissues from mice, n = 6. b Immunohistochemical staining of F4/80, the positive staining areas were measured by Ipwin32 software. Representative views were presented, magnification, ×5. c ELISA analysis of circulation level of pro-inflammatory cytokines MCP-1, TNF-α and IL-1β in serum, n = 6. d Double-immunofluorescence analysis of F4/80 (red) and TNF-α (green) expression, the positive staining areas were measured by Ipwin32 software. Representative views were presented, magnification, ×10. Data are represented as mean ± SD. ***P < 0.001.

Anti-inflammatory effects of HD-2a on LPS-induced RAW264.7 cells

Consistent with our in vivo experiments, HD-2a inhibited the expression of inflammatory cytokines in LPS-induced macrophages. As shown in Fig. 5a, 2 μM HD-2a has limited anti-inflammatory effects, while 4 μM and 8 μM HD-2a can effectively inhibit LPS-induced inflammatory response. Therefore, we selected 4 μM HD-2a for subsequent in vitro experiments. ELISA showed that the levels of inflammatory factors (IL-1β, TNF-α and MCP-1) were significantly reduced in the culture medium of HD-2a-pretreated RAW264.7 cells compared with LPS-induced RAW264.7 cells, indicating that the secretion of related inflammatory factors was inhibited (Fig. 5b). Additionally, immunofluorescence revealed that LPS-induced TNF-α signaling was significantly attenuated in RAW264.7 cells pretreated with HD-2a (Fig. 5c). Furthermore, transfection of siRNA-circDcbld2 as a control showed that HD-2a had a strong anti-inflammatory effect as a circDcbld2 inhibitor (Fig. 5d). In contrast, overexpression of circDcbld2 significantly reversed the anti-inflammatory effects of HD-2a (Fig. 5e), indicating that the anti-inflammatory effects of HD-2a are mediated by inhibiting circDcbld2.

Fig. 5. HD-2a attenuates inflammatory responses in LPS-treated RAW264.7 cells.

a Real-time quantitative PCR analyses of MCP-1, TNF-α and IL-1β mRNA levels in RAW264.7 cells with HD-2a treatment, n = 3. b ELISA analysis of circulation levels of pro-inflammatory cytokines MCP-1, TNF-α and IL-1β in cell supernatant, n = 3. c Immunofluorescence analysis of TNF-α (green) expression in RAW264.7 cells, and the positive staining areas were measured by Ipwin32 software. Representative views were presented, magnification, ×20. d Real-time quantitative PCR analyses of MCP-1, TNF-α and IL-1β mRNA levels in RAW264.7 cells with HD-2a treatment and siRNA-circDcbld2 transfection, n = 3. e mRNA levels of MCP-1, TNF-α and IL-1β in RAW264.7 cells with HD-2a and pcDNA-circDcbld2 administration were detected by qRT-PCR, n = 3. Data are represented as mean ± SD. ***P < 0.001.

HD-2a inhibits circDcbld2 to suppress JAK2/STAT3 signaling

To elucidate the mechanism by which HD-2a inhibits circDcbld2, we extracted liver tissue RNA from the ALI model and HD-2a treatment groups for RNA-seq analysis. After HD-2a pretreatment, 1830 genes were significantly downregulated and 1743 genes were significantly upregulated in LPS-induced ALI mice (Fig. 6a). We also found that HD-2a reduced the mRNA levels of many inflammatory cytokines and chemokines, such as IL-1β, IL-6, CCL2, CCL4, and CCL7 (Fig. 6b). Subsequently, we conducted Gene Ontology analysis of differentially expressed genes (Fig. 6c), which showed that inhibition of circDcbld2 by HD-2a was correlated with the inflammatory response (Fig. 6d). Previous studies have shown that the anti-inflammatory and hepatoprotective effects of hesperidin derivatives are closely related to the JAK-STAT signaling pathway, and KEGG enrichment analysis also indicated involvement of the JAK-STAT signaling pathway (Fig. 6e). Western blotting showed that HD-2a inhibited the phosphorylation of JAK2 and STAT3 in LPS-induced ALI (Fig. 6f and Supplementary Fig. S2a). Consistently, circDcbld2 knockdown significantly inhibited p-JAK2 and p-STAT3 protein levels in LPS-induced ALI mice (Fig. 6g and Supplementary Fig. S2b, c). In contrast, overexpression of circDcbld2 significantly reversed the inhibitory effect of HD-2a on JAK2/STAT3 phosphorylation (Supplementary Fig. S2d). In summary, our results indicate that LPS may activate the JAK2-STAT3 pathway to induce liver injury and that HD-2a may reduce inflammation and protect the liver by preventing circDcbld2-JAK2-STAT3.

Fig. 6. HD-2a inhibited circDcbld2 to suppress JAK2/STAT3 signaling.

a Volcano plot showing gene expression alterations in LPS-induced ALI treated with HD-2a. b A histogram showing inflammatory cytokine and chemokine genes in LPS-induced ALI treated with HD-2a. c GO Enrichment BubblePlot demonstrates the overall enrichment of differentially expressed genes in Biological Process, Cellular Component and Molecular Function. d The first 20 biological processes with the smallest P-value were mapped. e KEGG pathway analysis showed that the JAK-STAT signaling pathway participates in HD-2a mediated regulation of LPS induced ALI. f Immunoblottings analysis of JAK2, p-JAK2, STAT3 and p-STAT3 protein expression in liver tissues from HD-2a treated mice. g Immunoblottings analysis of JAK2, p-JAK2, STAT3 and p-STAT3 protein expression in liver tissues from mice exposed to AAV-circDcbld2-KD infection. Data are represented as mean ± SD.

Discussion

circRNAs are covalently closed and highly stable RNAs, and their specific temporal and spatial expression in disease makes them promising markers of disease diagnosis and evaluation [32]. The roles of circRNAs in pathogenesis and as potential drug targets have attracted much attention. In this study, the level of circDcbld2 in the liver of mice with ALI was significantly upregulated compared with that in normal mice. In addition, circDcbld2 knockdown inhibited liver tissue structure disorder, hepatocyte injury, and inflammatory cell infiltration, while circDcbld2 overexpression enhanced the degree of LPS-induced liver injury. In summary, circDcbld2 is a promising biomarker and therapeutic target for ALI.

The abnormal expression of circRNAs is closely related to the occurrence and development of complex diseases, such as nonalcoholic fatty liver disease, acute and chronic liver injury and hepatocellular carcinoma [33–35]. The development of circRNA-targeting drugs may provide effective strategies for the treatment of related diseases. For example, GA-RM/GZ/PL ameliorates NAFLD-induced lipid metabolism disorders by targeting liver cells and causing the overexpression of circRNA_0001805, the release of glycyrrhizic acid, and a reduction in liver lipid accumulation [36]. Consistent with these results, we have designed and synthesized a series of drugs targeting circDcbld2, and found that HD-2a can effectively inhibit the expression of circDcbld2. HD-2a reduced the inflammatory response in the liver and maintained the normal structure of the hepatic lobule, while overexpression of circDcbld2 reversed the therapeutic effect of HD-2a, indicating that HD-2a has an anti-inflammatory and hepatoprotective role by specifically inhibiting circDcbld2. Therefore, HD-2a has the potential to be developed as an inhibitor of circDcbld2 to prevent and treat ALI.

We have previously focused on the development of hesperidin derivatives and their mechanism of action in liver diseases, and we have found that the anti-inflammatory and hepatoprotective effects of hesperidin derivatives are closely related to the JAK-STAT pathway. We showed that the inhibition of HD-12 in LPS-induced M1 macrophages was related to the JAK2/STAT3 pathway [37]. We also confirmed that HD-14 blocks the JAK1-STAT1 pathway by activating PPARγ to treat CCl₄-induced ALI [29]. In addition, KEGG enrichment analysis has highlighted involvement of the JAK-STAT signaling pathway in ALI. Importantly, HD-2a significantly inhibited the phosphorylation of JAK2 and STAT3 in ALI mouse liver tissue and LPS-treated RAW264.7 cells. In addition, the phosphorylation and activation of JAK2 and STAT3 were significantly reduced by blocking the expression of circDcbld2, whereas the phosphorylation levels of JAK2 and STAT3 were increased after circDcbld2 overexpression.

The JAK/STAT signaling pathway is commonly associated with inflammatory diseases, immune diseases, and tumors, including ALI and interstitial lung disease [38, 39]. Some studies have shown the JAK-STAT signaling pathway to have a hepatoprotective effect. Interestingly, we found that hesperidin derivatives block part of the JAK-STAT signaling pathway and can inhibit macrophage activation and polarization to exert anti-inflammatory effects. A review of the literature found that blocking the JAK-STAT pathway produces anti-inflammatory effects in part by affecting the activation status of pro-inflammatory macrophages by down-regulating interferon signatures and IL-6 expression [38]. It is worth noting that macrophages can differentiate into pro-inflammatory M1 macrophages and anti-inflammatory M2 macrophages depending on their surrounding microenvironment and cytodynamic stimulation. Activation of M1 macrophages is usually the result of TLR4 and/or IFN signaling, which affects the JAK1/JAK2/P-STAT1 pathway by activation of the IFN receptor. IL-10 signaling drives differentiation of M2 anti-inflammatory macrophages through the IL-10 receptor-JAK1/Tyk2/P-STAT3-related pathway [40]. In summary, different stimulators activate different receptors in the JAK-STAT pathway, leading to their various effects in different cells. Further studies are needed to elucidate the specific mechanisms involved.

Although this study provides compelling evidence to support the treatment of ALI with HD-2a via downregulation of circDcbld2 and blocking activation of the JAK/STAT signaling pathway, the therapeutic effects of HD-2a may also involve mechanisms other than the JAK/STAT pathway. In the future, we will focus on key genes and signals downstream of the circDcbld2/JAK2/STAT3 axis, and investigate the relevant mechanisms in a mouse model of ALI.

Conclusion

In this study, we found that circDcbld2 was significantly upregulated in a mouse model of ALI. LPS-induced upregulation of circDcbld2 promoted liver inflammation and aggravated liver injury. As shown in Fig. 7, HD-2a inhibited the expression of circDcbld2 to reduce liver injury via the JAK2/STAT3 pathway.

Fig. 7.

Hesperetin derivative attenuates LPS-induced acute liver injury and inflammation by inhibiting circDcbld2.

Author contributions

LJS, XC, SZ designed the manuscript and performed the experiments; JJX, XFL, SXD analyzed the data; YLY, JYL contributed all samples, reagents and materials; JL contributed to all aspects of this study, data interpretation, and revised the manuscript for publication. All authors have revised and approved the final manuscript.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41401-023-01171-x.