Homoplantaginin protects pancreatic tissue in severe acute pancreatitis mice via inhibiting ferroptosis by modulating the circDNMT3B/miR-20b-5p/SLC7A11 axis

Abstract

Background

Severe acute pancreatitis (SAP) is a critical gastrointestinal disorder associated with high mortality. Dysregulated ferroptosis, an iron-dependent form of cell death, is recognized as a pivotal mechanism driving SAP progression. Although Homoplantaginin (Homo) exhibits notable anti-inflammatory and antioxidant properties, its specific role in modulating ferroptosis during SAP remains unclear.

Objective

To investigate whether Homo alleviates SAP by inhibiting ferroptosis via the circDNMT3B/miR-20b-5p/SLC7A11 axis.

Methods

C57BL/6 mice (n = 10 per group) were randomly allocated into nine groups, including normal control, SAP model, and various intervention groups involving Homo treatment, circDNMT3B knockdown, and miR-20b-5p inhibition. SAP was induced via caerulein and LPS injections. Assessments included histopathological evaluation, serum analysis (amylase, cytokines IL-1β, IL-6, IL-18), measurement of ferroptosis markers (MDA, GSH, tissue iron), Western blotting for ferroptosis-related proteins (ACSL4, COX-2, SLC7A11, GPX4), and qRT-PCR for gene expression. The targeting relationship between miR-20b-5p and SLC7A11 was validated by dual-luciferase reporter assay.

Results

SAP mice demonstrated severe pancreatic injury and hallmarks of ferroptosis, including elevated inflammation, iron accumulation, increased ACSL4/COX-2 expression, and upregulation of miR-20b-5p, alongside suppressed SLC7A11 and GPX4. Homo treatment significantly mitigated pancreatic damage and reversed these ferroptosis-related changes. CircDNMT3B knockdown exacerbated SAP severity and attenuated the protective effect of Homo, indicating its dependence on circDNMT3B. Furthermore, the detrimental effects of circDNMT3B knockdown were partially rescued by concurrent inhibition of miR-20b-5p, suggesting that miR-20b-5p acts downstream of circDNMT3B within the protective pathway mediated by Homo. The dual-luciferase assay confirmed miR-20b-5p directly targets SLC7A11.

Conclusion

Our study demonstrates that Homo protects against SAP primarily by suppressing ferroptosis. Mechanistically, Homo upregulates circDNMT3B, which sequesters miR-20b-5p, thereby relieving its repression of SLC7A11, which in turn supports the GPX4-mediated antioxidant defense. These findings elucidate a novel molecular pathway in SAP-associated ferroptosis and highlight the circDNMT3B/miR-20b-5p/SLC7A11 axis as a promising therapeutic target.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12876-025-04600-9.

Introduction

Acute pancreatitis (AP) is a prevalent gastrointestinal disorder and the leading cause of acute abdomen-related hospital admissions. The global incidence of AP has risen steadily in recent years [1–6]. Although most patients experience a mild, self-limiting form of the disease, approximately 15% progress to severe acute pancreatitis (SAP). SAP is a critical condition marked by severe systemic symptoms, a high frequency of multiple organ dysfunction syndrome (MODS), and significant mortality [7]. The pathogenesis of SAP is multifactorial, involving a complex cascade of pathophysiological events [8, 9]. This complexity has hindered the development of targeted therapies. Consequently, current clinical management remains predominantly supportive, underscoring the urgent need to identify key molecular targets for effective intervention [10, 11].

Ferroptosis is an iron-dependent, regulated form of cell death driven by phospholipid peroxidation and the failure of the glutathione peroxidase 4 (GPX4) antioxidant system, which critically depends on cysteine availability [12, 13]. Accumulating evidence increasingly implicates ferroptosis in the pathogenesis of SAP [14–16]. Our previous findings are consistent with this view, having demonstrated characteristic molecular signatures of ferroptosis in SAP model mice, including depleted GPX4 levels and iron accumulation. These findings thereby highlight ferroptosis inhibition as a promising therapeutic strategy for SAP.

Concurrently, non-coding RNAs have emerged as pivotal regulators of inflammatory and cell death pathways. MicroRNAs (miRNAs) are deeply involved in diverse cellular processes and play crucial roles in inflammation, organ injury, and cancer [17–19]. Among them, circular RNAs (circRNAs) represent a unique class of non-coding RNAs characterized by high stability and tissue specificity. They function as competitive endogenous RNAs (ceRNAs) to bind miRNAs and modulate the expression of their target genes [20]. Notably, circDNMT3B (derived from the DNA methyltransferase 3B gene) has been identified as a molecular sponge for miR-20b-5p in various disease contexts. For instance, in sepsis models, the downregulation of circDNMT3B exacerbates intestinal mucosal barrier dysfunction, an effect counteracted by its ability to bind and inhibit miR-20b-5p [21]. Similarly, in diabetic retinopathy, reduced circDNMT3B expression impairs retinal vascular function via the miR-20b-5p/BAMBI axis [22]. These findings collectively suggest that the circDNMT3B/miR-20b-5p axis is a conserved regulatory module, prompting our investigation into its role in SAP.

Homoplantaginin (Homo), a natural flavonoid from traditional herbs such as Plantago asiatica, possesses notable anti-inflammatory and antioxidant properties [23]. Given this pharmacological profile, Homo represents a promising candidate for treating SAP, characterized by intense inflammation and oxidative stress. Although Homo has been reported to ameliorate the insulin-resistant inflammatory microenvironment in endothelial cells [24], its efficacy and mechanism of action in SAP remain unclear. Our bioinformatic analysis revealed a potential direct targeting relationship between miR-20b-5p and the 3’ UTR of SLC7A11, a key ferroptosis-related gene. Based on this finding and our preliminary data, we hypothesized that Homo mitigates SAP by inhibiting ferroptosis via the circDNMT3B/miR-20b-5p/SLC7A11 axis. This study was therefore designed to test this hypothesis and provide novel evidence and a theoretical foundation for Homo as a therapeutic agent for SAP.

Materials and methods

Animals and ethical approval

Ninety specific pathogen-free (SPF) C57BL/6 mice (6–8 weeks old, weighing 24 ± 2.0 g, equal numbers of males and females) were obtained from the Experimental Animal Center of Guangxi Medical University (SCXK(Gui) 2020-0003). After one week of acclimatization under standard conditions (12-h light/dark cycle with free access to food and water), experimental procedures were initiated. All animal studies were conducted in accordance with protocols approved by the Institutional Animal Care and Use Committee of Guangxi Medical University (Approval No. 202401022).

Main reagents and materials

Chemicals and reagents

Homoplantaginin(Homo, purity ≥ 99.9%, Cat. No. HY-N1949) was purchased from MedChemExpress (Monmouth Junction, NJ, USA) and dissolved in dimethyl sulfoxide (DMSO, Sigma-Aldrich, St. Louis, MO, USA). Caerulein (Cat. No. C30598, Shanghai Aladdin Biochemical Technology Co., Ltd., China) and lipopolysaccharide (LPS, Cat. No. L2880, Sigma-Aldrich) were used to induce SAP.

Cell line and culture

The mouse pancreatic acinar cell line 266-6 (Cat. No. FH209) was obtained from FuHeng Cell Center (Shanghai, China). Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Thermo Fisher Scientific, Waltham, MA, USA, Cat. No. 11320033) supplemented with 10% fetal bovine serum (FBS, Thermo Fisher Scientific, Cat. No. F8687) and 1% penicillin-streptomycin solution (Thermo Fisher Scientific, Cat. No. 140675).

Oligonucleotides, Vectors, and transfection reagent

• The micrON mmu-miR-20b-5p mimic (Cat. No. miR1N0000001-1–5) and its negative control (micrON mimic NC, Cat. No. miR10003187-1–5) were synthesized by RiboBio (Guangzhou, China).

• The miR-20b-5p inhibitor (anti–miR-20b-5p) and its negative control (anti–miR-NC), which are chemically synthesized, HPLC-purified 2'-O-methyl-modified RNA oligonucleotides, were purchased from Kedi Biotechnology (Nanning, China). Their sequences (5'→3') are:

• anti–miR-20b-5p: mC mU mA mC mU mG mA mU mG mA mG mA mC mU mG

• anti–miR-NC: mC mA mG mU mA mC mU mU mU mU mU mA mA mA (The lowercase "m" denotes a 2'-O-methyl-modified ribonucleotide).

• The pmirGLO dual-luciferase reporter vector was from Promega (Madison, WI, USA).

• Adeno-associated viral vectors (pAAV-U6-shRNA-CMV-EGFP) encoding short hairpin RNA targeting circDNMT3B (shRNA-circDNMT3B) or a non-targeting control (shRNA-NC) were constructed by Heyuan Biotechnology (Shanghai, China).

• Lipofectamine 2000 Transfection Reagent was purchased from Invitrogen (Waltham, MA, USA).

Assay kits

Commercial kits for measuring malondialdehyde (MDA) and glutathione (GSH) levels, as well as enzyme-linked immunosorbent assay (ELISA) kits for interleukin (IL)−1β, IL-6, and IL-18, were obtained from Ruixin Biotechnology (Quanzhou, China). The tissue iron assay kit (Cat. No. E1046) was from Jiancheng Bioengineering Institute (Nanjing, China). RNA extraction kits (Cat. No. 9108 and 9109), the PrimeScript™ RT reagent kit for reverse transcription (Cat. No. RR092A), and qPCR kits (TB Green^® Premix Ex Taq™ II, Cat. No. RR820A; Mir-X™ miRNA First-Strand Synthesis Kit, Cat. No. 638313) were purchased from Kedi Biotechnology (Nanning, China). The Dual-Luciferase^® Reporter Assay System (Cat. No. E1910) was from Promega.

Antibodies

The following primary and secondary antibodies were used for Western blotting: rabbit anti-GPX4 (1:1,000, R38958), rabbit anti-SLC7A11 (1:1,000, R382036), rabbit anti-ACSL4 (1:1,000, A95060), rabbit anti-COX-2 (1:1,000, A59643), rabbit anti-GAPDH (1:10,000, R380626), and HRP-conjugated goat anti-rabbit secondary antibody (1:10,000, 511203).

Animal Grouping, model Replication, and administration

Ninety mice were randomly assigned to nine groups (n = 10): (1) Normal control (NC); (2) SAP model (SAP); (3) SAP + Homoplantaginin (SAP + Homo); (4) scrambled adenovirus control + SAP (shRNA-NC + SAP); (5) circDNMT3B-knockdown adenovirus + SAP (shRNA-circDNMT3B + SAP); (6) shRNA-NC + SAP + Homo; (7) shRNA-circDNMT3B + SAP + Homo; (8) shRNA-circDNMT3B + anti-miR-NC + SAP + Homo (shRNA-circDNMT3B + MIRi-NC + SAP + Homo); (9) shRNA-circDNMT3B + anti-miR-20b-5p + SAP + Homo (shRNA-circDNMT3B + MIRi-20b-5p + SAP + Homo). The sample size for each quantitative analysis is provided in the respective figure legends. A minimal reduction in sample size for certain molecular analyzes was occasionally necessitated by technical issues (e.g., insufficient protein yield or RNA degradation). No data points were excluded from the analysis arbitrarily, and all available data from successfully processed samples were included.

Following acclimatization, mice in groups 4–9 received an intraperitoneal injection of control or circDNMT3B-knockdown adenovirus (1 × 10^11 viral particles,100 µl) and were maintained for 3 weeks. Three days prior to SAP induction, groups 8 and 9 were pretreated with anti-miR-NC or anti-miR-20b-5p, respectively. Subsequently, SAP was induced in all groups except NC via seven hourly intraperitoneal injections of caerulein (50 µg/kg), with a concurrent LPS injection (10 mg/kg) at the final caerulein dose [25–27], and caerulein was dissolved in normal saline, while lipopolysaccharide (LPS) was dissolved in sterile phosphate-buffered saline (PBS). Beginning at 24 hours post‑modeling, mice in groups 3 and 6–9 were administered Homo (15 mg/kg in 10 mL/kg) via intravenous injection twice daily for three consecutive days [28], while control groups received saline vehicle. Homo was dissolved in a vehicle consisting of DMSO, PEG300, Tween-80, and saline at a ratio of 1:8:1:10. The special control groups (NC, SAP, shRNA-NC + SAP, and shRNA-circDNMT3B + SAP) received intravenous injections of an equal volume of saline via the tail vein. Homo intervention was administered for three consecutive days, after which the animals were anesthetized and samples were collected for analysis (as detailed in Sect. 2.4).

Anesthesia, blood collection and euthanasia

Twenty-four hours after the final treatment, mice were first anesthetized by intraperitoneal injection of a 1% sodium pentobarbital solution (50 mg/kg body weight). Under deep anesthesia, blood samples were collected from the retro-orbital plexus into EDTA-coated tubes. Plasma was then obtained by centrifugation at 1,000 × g for 15 min at 4 °C, aliquoted, and stored at −80 °C for subsequent analysis.

Immediately following blood collection, the anesthetized mice were euthanized by cervical dislocation to ensure a humane endpoint. Death was confirmed by the absence of breathing, heartbeat, and corneal reflex. The entire procedure, which minimized animal distress, was approved by the Institutional Animal Ethics Committee and performed in accordance with the American Veterinary Medical Association (AVMA) Guidelines for the Euthanasia of Animals.

Tissue harvesting

Following euthanasia, the pancreas was rapidly harvested via laparotomy. Each pancreas was divided into three parts in ice-cold saline: one part was fixed in 4% paraformaldehyde for histological examination; the second part was snap-frozen for protein and biochemical assays; and the third part was immersed in RNAlater, stored at 4 °C overnight, and then transferred to −80 °C for future RNA extraction.

Histopathological analysis by H&E staining

Pancreatic tissues fixed in 4% paraformaldehyde were processed routinely for paraffin embedding, sectioning, and hematoxylin and eosin (H&E) staining. The pathological morphology was observed under a light microscope. The severity of pancreatic injury was assessed semi-quantitatively by scoring three parameters—tissue edema, acinar necrosis, and inflammatory cell infiltration—on a scale of 0 to 4 for each, according to established criteria [29, 30] (detailed in Table 1). Higher total scores indicated more severe tissue damage.

Table 1.

Histopathological scoring criteria for pancreatic injury

Score	Edema	Acinar Necrosis	Inflammatory Infiltration
0	Absent	Absent	0–5 white blood cells (WBCs) per high-power field (HPF)
1	Diffuse expansion of lobular septa	1–4 necrotic cells per HPF	6–15 WBCs per HPF
2	Diffuse and extensive expansion of interlobular spaces	5–10 necrotic cells per HPF	16–25 WBCs per HPF
3	Diffuse expansion of acinar septa	11–16 necrotic cells per HPF	26–35 WBCs per HPF
4	Diffuse expansion of intercellular spaces	> 16 necrotic cells per HPF	Extensive necrosis with confluence > 35 WBCs per HPF, or confluent microabscesses

The degree of acinar cell necrosis and inflammatory infiltration was calculated as the average of every 10 fields of view under high-power field (HPF, ×200)

Western blot analysis

Protein expression levels of ACSL4, COX-2, SLC7A11, and GPX4 in pancreatic tissues were analyzed by Western blot. Total protein was extracted using RIPA lysis buffer and quantified with a BCA assay. Equal amounts of protein were separated by SDS-PAGE and transferred onto PVDF membranes. After blocking with 5% non-fat milk, the membranes were incubated overnight at 4 °C with specific primary antibodies, followed by incubation with an HRP-conjugated secondary antibody. Protein bands were visualized using an enhanced chemiluminescence (ECL) detection system and quantified with ImageJ software (National Institutes of Health, USA). GAPDH was used as an internal loading control for normalization.

Measurement of serum biomarkers

Serum levels of amylase (Amy) and the inflammatory cytokines IL-1β, IL-6, and IL-18 were quantified using commercial enzyme-linked immunosorbent assay (ELISA) kits according to the manufacturers’ instructions. The oxidative stress markers malondialdehyde (MDA) and glutathione (GSH) were measured using specific commercial biochemical assay kits. For all assays, standard curves were generated from the provided standards, and sample measurements were performed in duplicate.

Determination of pancreatic tissue iron content

The iron content in pancreatic tissues was determined using a commercial colorimetric assay kit (E1046, Nanjing Jiancheng Bioengineering Institute, China). Frozen tissue samples were weighed, homogenized on ice, and processed following the manufacturer’s protocol. The iron concentration was calculated based on the measured absorbance and a standard curve, then normalized to the tissue weight.

Quantitative Real-Time PCR (qRT-PCR) analysis

Total RNA was extracted from pancreatic tissues to evaluate the expression levels of circDNMT3B, miR-20b-5p, SLC7A11, and GPX4. After quantification, RNA was reverse transcribed into cDNA using specific kits. qPCR was performed using a SYBR Green system. All primer sequences are listed in Table 2. Gene expression was calculated using the 2^–ΔΔCt method, with U6 snRNA as the internal control for miR-20b-5p and GAPDH for circDNMT3B, SLC7A11, and GPX4.

Table 2.

Primer information for RT-qPCR

Name	Primer sequence (5´−3´)
RT-mmu-miR-20b-5p	CTCAACTGGTGTCGTGGAGTCGGCAATTCAGTTGAGCTACCT
mmu‑miR‑20b‑5p‑F	ACACTCCAGCTGGGCAAAGTGCTCATAGTGC
SLC7A11-F	CATCATCATCGGCACCGTCAT
SLC7A11-R	AGCAGTTCCACCCAGACTCGA
circDNMT3B-F	CAAATCCCGTCTGCTTCCCT
circDNMT3B-R	AGTCACCAACAACAAGGGCA
GPX4-F	TGTGCATCCCGCGATGATT
GPX4-R	CCCTGTACTTATCCAGGCAGA
GAPDH-F	GCCAAAAGGGTCATCATCTCC
GAPDH-R	GTGATGGCATGGACTGTGGTC
U6-F	GGAACGATACAGAGAAGATTAGC
U6-R	TGGAACGCTTCACGAATTTGCG

Cell transfection and Dual-Luciferase reporter assay

The online database TargetScan (http://www.targetscan.org/vert_71/) was utilized to predict potential targets of miR-20b-5p. SLC7A11 was selected for further investigation in this study. Wild-type or mutant 3’-UTR fragments of SLC7A11, containing the putative miR-20b-5p binding site, were cloned into the pmirGLO dual-luciferase reporter vector. Prior to transfection, 266-6 cells were maintained in culture for at least 24 h and rinsed with phosphate-buffered saline (PBS, pH 7.4). Subsequently, using Lipofectamine 2000, the wild-type or mutant SLC7A11 3’-UTR luciferase reporter constructs were co-transfected with either the miR-20b-5p mimic or a negative control mimic (miR-NC) into the cells. The transfected cells were then cultured in appropriate medium at 37 °C under 5% CO₂. Transfection efficiency was verified by quantitative real-time PCR (qRT-PCR). Finally, luciferase activity was measured using the Dual-Luciferase Reporter Assay System (Promega) in strict accordance with the manufacturer’s protocol.

Statistical analysis

Statistical analyzes were performed using GraphPad Prism 10.0 software, with continuous data presented as mean ± standard deviation (mean ± SD); intergroup comparisons employed one-way ANOVA, followed by LSD-t tests for pairwise comparisons when significant differences were detected (P < 0.05), while non-parametric tests or appropriate data transformations were applied when homogeneity of variance or normal distribution assumptions were violated.

Results

Homo mitigates histopathological damage in SAP mice

Histopathological analysis of pancreatic sections revealed severe tissue damage in SAP mice compared to the NC group (Fig. 1). The total injury score was markedly elevated in the SAP group (9.2 ± 1.2) versus the NC group (0.8 ± 0.6; P < 0.001). Treatment with Homo substantially ameliorated this damage, reducing the total score by approximately 64% (3.3 ± 1.3; P < 0.01 vs. SAP group). Analysis of individual pathological parameters—edema, inflammatory cell infiltration, and necrosis—confirmed that the SAP group exhibited the most severe damage across all indices, while Homo intervention significantly improved each parameter (Table 3). These findings indicate that Homo effectively alleviates SAP-induced pancreatic tissue injury.

Fig. 1.

Homo ameliorates pancreatic histopathology in mice with severe acute pancreatitis (SAP). A Representative photomicrographs of H&E-stained pancreatic tissue sections from the indicated groups (scale bar = 50 μm). B Quantitative analysis of the total histopathological injury scores. Data are presented as mean ± SD (n = 10). ***P < 0.001 compared to the NC group; ##P < 0.01 compared to the SAP group

Table 3.

Histopathological assessment of pancreatic tissue across experimental groups. (mean ± SD, n = 10)

group	edema	Inflammation	necrosis	Total
NC	0.5 ± 0.53	0.3 ± 0.48	0 ± 0	0.8 ± 0.63
SAP	3.8 ± 0.42	3 ± 0.67	2.4 ± 0.52	9.2 ± 1.23
SAP + Homo	1.7 ± 0.67	1.1 ± 0.74	0.5 ± 0.53	3.3 ± 1.25
shRNA-NC + SAP	3.6 ± 0.52	2.92 ± 0.67	2.3 ± 0.48	8.9 ± 1.20
shRNA-circDNMT3B + SAP	3.9 ± 0.32	3.67 ± 0.49	3.4 ± 0.52	11 ± 0.94
shRNA-NC + SAP + Homo	1.2 ± 0.79	0.83 ± 0.72	0.4 ± 0.52	3 ± 1.33
shRNA-circDNMT3B + SAP + Homo	2 ± 0.67	1.92 ± 0.67	1.2 ± 0.42	4.7 ± 1.42
shRNA-circDNMT3B + MIRi-NC + SAP + Homo	2.1 ± 0.74	1.6 ± 0.7	1.4 ± 0.52	5.1 ± 1.73
shRNA-circDNMT3B + MIRi-20b-5p + SAP + Homo	1.5 ± 0.71	0.7 ± 0.48	0.3 ± 0.48	2.5 ± 1.08

Homo alleviates systemic Inflammation, oxidative Stress, and pancreatic injury in SAP mice

The systemic protective effects of Homo were assessed by analyzing key serum biomarkers. As shown in Fig. 2, SAP induction resulted in comprehensive pathophysiological disturbances relative to the NC group. Serum amylase (Amy) activity was significantly increased in the SAP group (1945 ± 147.1 U/L vs. 1074 ± 66.3 U/L in NC; P < 0.01), indicating severe pancreatic exocrine dysfunction. Concurrently, SAP mice exhibited markedly elevated levels of the pro-inflammatory cytokines IL-1β (117.1 ± 5.6 pg/mL), IL-6 (79.4 ± 5.2 pg/mL), and IL-18 (119.7 ± 6.9 pg/mL) (all P < 0.01 vs. NC). A pronounced oxidative stress imbalance was also evident, characterized by increased malondialdehyde (MDA; 11.7 ± 1.2 nmol/mL) and decreased glutathione (GSH; 2.3 ± 0.3 µmol/L) (both P < 0.01 vs. NC). Treatment with Homo significantly reversed these abnormalities, reducing Amy activity, cytokine levels, and MDA, while restoring GSH levels (all P < 0.05 vs. SAP). These results demonstrate that Homo effectively mitigates SAP-induced pancreatic damage, systemic inflammation, and oxidative stress.

Fig. 2.

Homo ameliorates SAP-induced systemic abnormalities. Serum levels of amylase (Amy), inflammatory cytokines (IL-1β, IL-6, IL-18), and oxidative stress markers (MDA, GSH) were measured to assess pancreatic exocrine function, systemic inflammation, and redox status. A Expression level of amylase in plasma supernatant; B Expression level of IL-1β; C Expression level of IL-6; D Expression level of IL- 18; E Expression level of MDA; F Expression level of GSH. Data are presented as mean ± SD (n = 6). **P < 0.01 vs. NC group; #P < 0.05 vs. SAP group

Homo reverses ferroptosis in the pancreas of SAP mice

The protein expression of key ferroptosis regulators in pancreatic tissue was assessed by Western blot (Fig. 3). Compared to the NC group, SAP induction significantly suppressed the expression of the ferroptosis-inhibitory proteins SLC7A11 (0.49 ± 0.20 vs. 1.10 ± 0.13) and GPX4 (0.66 ± 0.15 vs. 1.07 ± 0.12), while upregulating the ferroptosis-promoting proteins ACSL4 (1.19 ± 0.19 vs. 0.43 ± 0.22) and COX-2 (1.08 ± 0.22 vs. 0.47 ± 0.19) (all P < 0.05). Treatment with Homo effectively reversed these alterations, restoring the expression of SLC7A11 (1.23 ± 0.23) and GPX4 (1.19 ± 0.12) and suppressing the elevated levels of ACSL4 (0.42 ± 0.22) and COX-2 (0.49 ± 0.22) (all P < 0.05 vs. SAP group). These results indicate that Homo inhibits ferroptosis in SAP by coordinately modulating the expression of critical ferroptosis pathway proteins.

Fig. 3.

Homo modulates the expression of ferroptosis-related proteins in the pancreas of SAP mice. A-D Representative Western blot bands (left) and quantitative analysis (right) of ACSL4 (A), COX-2 (B), SLC7A11 (C), and GPX4 (D) protein expression in pancreatic tissue. Data are presented as mean ± SD (n = 8). *P < 0.05 vs. NC group; #P < 0.05 vs. SAP group

To further evaluate ferroptosis levels in pancreatic tissue, we measured Fe²⁺ concentrations. As shown in Fig. 4, the Fe²⁺ content was significantly elevated in the SAP group (10.99 ± 2.55 µmol/L) compared to the NC group (2.54 ± 0.89 µmol/L; P < 0.01), indicating pronounced iron accumulation. Treatment with Homo markedly reduced this accumulation, as evidenced by the decreased Fe²⁺ levels in the SAP + Homo group (3.85 ± 0.96 µmol/L; P < 0.01 vs. the SAP group).

Fig. 4.

Homo reduces ferroptosis-related iron accumulation in the pancreas of SAP mice. The Fe²⁺ content in pancreatic tissue was measured using a colorimetric assay kit. Data are presented as mean ± SD (n = 8). **P < 0.01 compared to the NC group; ##P < 0.01 compared to the SAP group

Collectively, these findings demonstrate that Homo inhibits ferroptosis and exerts a protective effect against SAP through a multi-faceted mechanism: it upregulates the anti-ferroptotic proteins SLC7A11 and GPX4, downregulates the pro-ferroptotic proteins ACSL4 and COX-2, and ultimately reduces both iron deposition and lipid peroxidation.

Homo modulates the circDNMT3B/miR-20b-5p/SLC7A11/GPX4 axis in SAP mice

To investigate the underlying molecular mechanisms, we measured the mRNA expression levels of related genes in pancreatic tissues from each group using qRT-PCR (Fig. 5). The results showed that compared with the NC group, the SAP group exhibited a significant upregulation of miR-20b-5p expression, while the mRNA levels of SLC7A11 and GPX4 were markedly downregulated. Although circDNMT3B expression showed a decreasing trend, the difference was not statistically significant. Homo intervention produced notable effects: compared with the SAP group, the SAP + Homo group demonstrated significantly suppressed expression of miR-20b-5p, while the mRNA expression of circDNMT3B, SLC7A11, and GPX4 was markedly reversed (all P < 0.05). These mRNA-level changes were consistent with protein detection results, collectively suggesting that Homo may activate the SLC7A11/GPX4 axis by upregulating circDNMT3B and inhibiting miR-20b-5p.

Fig. 5.

Homo modulates the expression of the circDNMT3B/miR-20b-5p/SLC7A11/GPX4 axis in SAP mice. The mRNA expression levels of (A) circDNMT3B, B miR-20b-5p, C SLC7A11, and D GPX4 in pancreatic tissue were determined by qRT-PCR. Data are presented as mean ± SD (n = 6). ^nsP > 0.05, *P < 0.05 vs. NC group;# P < 0.05 vs. SAP group

To further validate the direct regulatory role of miR-20b-5p on SLC7A11, we performed predictions using bioinformatics databases such as TargetScan and identified a conserved binding site for miR-20b-5p in the 3’-UTR region of the SLC7A11 gene (Fig. 6A). We then constructed luciferase reporter vectors containing either the wild-type (WT) or mutant (MUT) binding site of the SLC7A11 3’-UTR. Dual-luciferase reporter assays revealed that overexpression of miR-20b-5p significantly inhibited the luciferase activity of the SLC7A11-WT reporter vector but had no significant effect on the activity of the SLC7A11-MUT vector (Fig. 6B), confirming a direct targeting relationship between miR-20b-5p and SLC7A11. Subsequently, qRT-PCR was performed to determine the SLC7A11 mRNA expression in different groups of 266-6 cells, which revealed a decrease in SLC7A11 mRNA expression in the miR-mimic group (P < 0.01, Fig. 6C). Similarly, Western blot analysis also indicated that the expression of SLC7A11 protein was significantly downregulated when miR-20b-5p was overexpressed (P < 0.01, Fig. 6D). Overall, the above results demonstrate that miR-20b-5p can target and inhibit the expression of SLC7A11 in 266-6 cells.

Fig. 6.

miR-20b-5p targets and silences SLC7A11 in 266-6 cells. A The binding sites between miR-20b-5p and SLC7A11 3’UTR predicted by a bioinformatics database; B The luciferase activity of SLC7A11-WT and SLC7A11-MUT detected via dual-luciferase reporter assay; C SLC7A11 mRNA expression in 266-6 cells (miR-NC and miR-20b-5p mimic) determined through qRT-PCR; D SLC7A11 protein expression in 266-6 cells (miR-NC and miR-20b-5p mimic) detected by western blot. **p < 0.01,^nsP > 0.05 vs. miR-NC group

Genetic knockdown of circDNMT3B abrogates the protective effects of homo in SAP

Our findings establish that Homoplantaginin (Homo) alleviates SAP by suppressing ferroptosis. To determine whether circDNMT3B is required for this protection, we constructed a circDNMT3B-knockdown SAP model via adenoviral delivery. Successful knockdown was first confirmed by qPCR, showing significantly lower circDNMT3B expression in the shRNA-circDNMT3B + SAP group than in the shRNA-NC + SAP group (Fig. 7A, P < 0.05). Crucially, the knockdown remained effective under Homo treatment, as evidenced by the persistent reduction of circDNMT3B in the shRNA-circDNMT3B + SAP + Homo group compared to the shRNA-NC + SAP + Homo controls (P < 0.05). Consistent with its proposed role as a miR-20b-5p sponge, circDNMT3B knockdown led to a compensatory upregulation of miR-20b-5p (Fig. 7B, P < 0.05) and a concomitant decrease in the mRNA levels of the ferroptosis suppressors SLC7A11 and GPX4 (Fig. 7C, D, P < 0.05) in the presence of Homo. These data strongly support the hypothesis that circDNMT3B is functionally required for the protective effect of Homo.

Fig. 7.

Knockdown of circDNMT3B attenuates Homo-mediated regulation of the miR-20b-5p/SLC7A11/GPX4 axis. The relative expression levels of (A) circDNMT3B, B miR-20b-5p, C SLC7A11, and D GPX4 in pancreatic tissue were quantified by qRT-PCR. Data are presented as mean ± SD (n = 6). *P < 0.05, **P < 0.01 vs. shRNA-NC + SAP group; #P < 0.05vs. shRNA-NC + SAP + Homo group

We then assessed the functional consequence of circDNMT3B loss on Homo’s efficacy. Histopathological analysis revealed that circDNMT3B knockdown significantly attenuated the protective effects of Homo. H&E staining (Fig. 8A) showed more severe acinar necrosis and inflammatory infiltration in the shRNA-circDNMT3B + SAP + Homo group compared to the shRNA-NC + SAP + Homo control. Corresponding pathological scores (Fig. 8B) confirmed significantly elevated necrosis and inflammation (P < 0.05). This diminished protection was further validated systemically and molecularly. As shown in Fig. 9, serum levels of amylase, IL-1β, IL-6, IL-18, and MDA were significantly higher in the shRNA-circDNMT3B + SAP + Homo group, while GSH levels were lower (all P < 0.05). Western blot analysis (Fig. 10) demonstrated that circDNMT3B knockdown under Homo treatment led to decreased SLC7A11 and GPX4 protein expression and increased ACSL4 and COX-2 levels (all P < 0.05). Furthermore, pancreatic Fe²⁺ content (Fig. 11) was significantly higher in the shRNA-circDNMT3B + SAP + Homo group (P < 0.05). Collectively, these data demonstrate that circDNMT3B knockdown effectively reverses the therapeutic benefits of Homo against SAP, functionally establishing circDNMT3B as a critical upstream mediator of Homo’s action.

Fig. 8.

circDNMT3B knockdown exacerbates pancreatic injury and attenuates the protective effect of Homo. A Representative H&E-stained sections of pancreatic tissue (scale bar: 50 μm). B Quantitative analysis of total histopathological injury scores is shown in the lower panel. Data are presented as mean ± SD (n = 6). *P < 0.05, **P < 0.01 vs. shRNA-NC + SAP group; #P < 0.05 vs. shRNA-NC + SAP + Homo group

Fig. 9.

circDNMT3B knockdown exacerbates systemic manifestations and attenuates the therapeutic effects of Homo in SAP mice. Serum levels of pancreatic injury marker (amylase, Amy), inflammatory cytokines (IL-1β, IL-6, IL-18), and oxidative stress markers (MDA, GSH) were measured. A Expression level of amylase in plasma supernatant; B Expression level of IL-1β; C Expression level of IL-6; D Expression level of IL- 18; E Expression level of MDA; F Expression level of GSH. Data are presented as mean ± SD (n = 6). *P < 0.05, **P < 0.01 vs. shRNA-NC + SAP group; #P < 0.05vs. shRNA-NC + SAP + Homo group

Fig. 10.

circDNMT3B knockdown disrupts the protein expression of ferroptosis regulators and attenuates Homo-mediated correction. Western blot analysis of ferroptosis-related proteins (A) ACSL4, B COX-2, C SLC7A11 and D GPX4 in pancreatic tissue. Representative bands are shown in the left panels, with quantitative analysis presented on the right. Data are expressed as mean ± SD (n = 6). *P < 0.05, **P < 0.01 vs. shRNA-NC + SAP group; #P < 0.05 vs. shRNA-NC + SAP + Homo group

Fig. 11.

circDNMT3B knockdown exacerbates iron accumulation and attenuates the iron-resolving effect of Homo in SAP mice. Ferrous iron (Fe²⁺) content in pancreatic tissue was measured using a colorimetric assay. Data are presented as mean ± SD (n = 6). *P < 0.05, **P < 0.01 vs. shRNA-NC + SAP group; # P < 0.05 vs. shRNA-NC + SAP + Homo group

Quantitative correlation analysis of upstream and downstream molecules in the Homo-Regulated circDNMT3B/miR-20b-5p axis

To precisely assess the regulatory impact of Homo mediated through circDNMT3B, we conducted a fold-change analysis of mRNA and protein expression levels of key pathway molecules. As summarized in Table 4, Homo treatment induced marked upregulation of SLC7A11 and GPX4 mRNA levels compared to the SAP model group—increasing by 9.1-fold and 12.1-fold, respectively. In contrast, their corresponding protein levels exhibited significant yet comparatively moderate increases of 2.5-fold and 1.8-fold. This expression pattern indicates that under SAP conditions, SLC7A11 and GPX4 mRNAs are likely subject to strong post-transcriptional repression, possibly by miR-20b-5p. By alleviating this repression, Homo promotes a pronounced rebound in mRNA accumulation, which in turn facilitates the synthesis of functional proteins. To further establish the pivotal role of circDNMT3B, we examined the consequences of its knockdown. Relative to the shRNA-NC + SAP + Homo group, circDNMT3B knockdown led to coordinated downregulation of both mRNA and protein levels of SLC7A11 and GPX4, with reductions occurring in similar proportions. Notably, GPX4 mRNA expression fell sharply to 0.2-fold, while its protein level decreased to 0.56-fold. This near-proportional decline in both molecular tiers contrasts clearly with the rebound pattern seen upon repression relief, functionally affirming that circDNMT3B acts as an essential upstream regulator through which Homo sustains the mRNA stability and protein expression of its downstream targets.

Table 4.

Fold-change analysis of mRNA and protein expression of key molecules

Comparison Group	Molecule	mRNA Fold Change	Protein Fold Change	Biological Interpretation
SAP+Homo vs SAP	SLC7A11	9.1	2.5	Post-inhibition mRNA rebound with effective protein synthesis
	GPX4	12.1	1.8	Post-inhibition mRNA rebound with effective protein synthesis
shRNA-circDNMT3B+SAP+Homo vs shRNA-NC+SAP+Homo	SLC7A11	0.65	0.6	Synchronous reduction upon loss of upstream regulation
	GPX4	0.2	0.56	Synchronous reduction upon loss of upstream regulation

Collectively, these quantitative findings support a model wherein Homo, via circDNMT3B, exerts its regulatory influence predominantly at the mRNA stability level, thereby governing subsequent protein production.

miR-20b-5p inhibits the protective effect of homo by reversing circDNMT3B knockdown

To ascertain whether circDNMT3B exerts its function by sequestering miR-20b-5p, we administered a miR-20b-5p inhibitor (anti–miR-20b-5p) or its negative control (anti–miR-NC) to circDNMT3B-knockdown SAP mice concurrently treated with Homo.

qPCR analysis confirmed that inhibition of miR-20b-5p effectively counteracted the molecular alterations caused by circDNMT3B knockdown. Compared with the anti-miR-NC control group, miR-20b-5p expression was significantly suppressed (Fig. 12A, P < 0.05), which was accompanied by a marked recovery in the mRNA levels of its downstream targets, SLC7A11 and GPX4 (P < 0.05). This molecular rescue translated into improved histopathology (Fig. 12B). H&E staining revealed that miR-20b-5p inhibition significantly ameliorated pancreatic tissue architecture, reducing acinar necrosis and inflammatory infiltration. Semi-quantitative scoring confirmed substantial reductions in both necrosis and inflammation scores (P < 0.05). The therapeutic benefits were further evidenced by the reversal of systemic and oxidative stress markers (Fig. 12C). Inhibition of miR-20b-5p significantly lowered serum levels of amylase, IL-1β, IL-6, IL-18, and MDA, while elevating GSH content (all P < 0.05). Measurement of pancreatic Fe²⁺ content (Fig. 12D) demonstrated that miR-20b-5p inhibition significantly reduced abnormal iron deposition (P < 0.05), providing functional phenotypic evidence of suppressed ferroptosis. Western blot analysis provided conclusive protein-level validation (Fig. 13). Ultimately, under circDNMT3B-knockdown conditions, miR-20b-5p inhibition significantly upregulated SLC7A11 and GPX4 protein expression and downregulated ACSL4 and COX-2 (all P < 0.05).

Fig. 12.

Inhibition of miR-20b-5p rescues the protective effect of Homo impaired by circDNMT3B knockdown. A The relative mRNA expression levels of miR-20b-5p, SLC7A11, and GPX4 in pancreatic tissue were determined by qRT-PCR. B Representative H&E-stained sections of pancreatic tissue (scale bar: 50 μm) and the corresponding quantitative histopathological injury scores. C Serum levels of amylase (Amy), inflammatory cytokines (IL-1β, IL-6, IL-18), and oxidative stress markers (MDA, GSH). D Fe²⁺ content in pancreatic tissue was measured by a colorimetric assay. All data are presented as mean ± SD (n=6). *P < 0.05 vs. shRNA-circDNMT3B+MIRi-NC+SAP+Homo group

Fig. 13.

miR-20b-5p inhibition restores the balance of ferroptosis-related proteins following circDNMT3B knockdown. Representative Western blots and quantitative data demonstrating that inhibition of miR-20b-5p upregulates SLC7A11 and GPX4 protein levels while downregulating ACSL4 and COX-2 in pancreatic tissue of the indicated groups. A Expression level of ACSL4 protein in pancreatic tissue; B Expression level of COX-2 protein in pancreatic tissue; C Expression level of SLC7A11 protein in pancreatic tissue; Expression level of GPX4 protein in pancreatic tissue.Data are mean ± SD (n = 6). ^nsP > 0. 05,*P < 0.05 vs. shRNA-circDNMT3B + MIRi-NC + SAP + Homo group

In summary, specific inhibition of miR-20b-5p fully rescues the deleterious phenotypes induced by circDNMT3B knockdown. This functionally establishes miR-20b-5p as the key downstream effector of circDNMT3B, conclusively verifying the core regulatory axis: Homo → circDNMT3B → miR-20b-5p → SLC7A11/GPX4.

Discussions

SAP a condition with high mortality and limited treatment options, remains an unmet clinical challenge. This study identifies the natural flavonoid Homo as a potent therapeutic agent against SAP. We elucidate a novel mechanism of action, demonstrating that Homo alleviates pancreatic injury by suppressing ferroptosis through the newly identified circDNMT3B/miR-20b-5p/SLC7A11 signaling axis.

Ferroptosis is characterized by intracellular divalent iron (Fe²⁺) overload, which drives extensive lipid peroxidation and the collapse of the glutathione-dependent antioxidant defense system, particularly the key enzyme GPX4 [12, 13]. Accumulating evidence underscores the pivotal role of ferroptosis in the pathogenesis of SAP [14, 31, 32]. It contributes to disease progression through a dual mechanism: localized ferroptosis in pancreatic tissue directly induces acinar cell damage and necrotic expansion, while the release of damage-associated molecular patterns (DAMPs) from ferroptotic cells amplifies systemic inflammatory cascades [15]. Consequently, targeting ferroptosis has emerged as a promising therapeutic strategy for SAP.

The SAP mouse model established in this study recapitulated not only the expected severe pancreatic injury and systemic inflammation but also a distinct ferroptotic signature, including dysregulation of key proteins (downregulated GPX4 and SLC7A11; upregulated ACSL4 and COX-2) and biochemical markers (elevated Fe²⁺ and MDA; depleted GSH), consistent with findings reported in other organ injury models [15]. Critically, intervention with Homo effectively normalized these ferroptosis-related abnormalities and significantly mitigated histopathological damage, identifying ferroptosis inhibition as a core mechanism underlying its protective action against SAP.

Homo, a key bioactive flavonoid derived from Plantago asiatica, possesses well-documented anti-inflammatory and antioxidant properties. While its direct application in SAP had not been previously reported, these established pharmacological activities prompted our investigation into its potential therapeutic value. Our results confirm that Homo effectively mitigates the inflammatory cytokine storm and alleviates oxidative stress in SAP. This aligns with prior mechanistic studies: for instance, Fan et al. demonstrated that Homo activates the AMPK/TFEB pathway to promote autophagy and alleviate vascular endothelial inflammation [33], while Meng et al. reported that Homo activates the Nrf2 pathway to counteract oxidative stress in endothelial cells [34]. Additionally, its anti-inflammatory action is linked to the suppression of the NF-κB signaling pathway [35]. These documented mechanisms provide a plausible explanation for our observed reductions in IL-1β and IL-6, as well as the restoration of GSH and reduction of MDA in the SAP model.

Beyond these known activities, our study unveils a novel, previously unrecognized mechanism: Homo’s ability to regulate ferroptosis through a circRNA-mediated competing endogenous RNA (ceRNA) network. We specifically identified the circDNMT3B/miR-20b-5p/SLC7A11 axis. Under SAP conditions, circDNMT3B expression is suppressed, diminishing its capacity to act as a molecular sponge for miR-20b-5p. This leads to an accumulation of free miR-20b-5p, which post-transcriptionally represses SLC7A11, thereby impairing the GPX4 system and exacerbating ferroptosis. Our functional experiments robustly support this axis: circDNMT3B knockdown not only aggravated SAP severity but also significantly attenuated the therapeutic efficacy of Homo. Crucially, the detrimental effects of circDNMT3B knockdown were effectively rescued by inhibiting miR-20b-5p, which restored SLC7A11/GPX4 expression and Homo’s protective action. To mechanistically confirm the direct targeting relationship central to this axis, we performed dual-luciferase reporter assays. The results demonstrated that miR-20b-5p specifically binds to the 3’ untranslated region (3’UTR) of SLC7A11, as evidenced by a significant reduction in luciferase activity of the wild-type reporter, an effect that was abolished upon mutation of the predicted binding site. This evidence functionally establishes circDNMT3B as a central regulator of ferroptosis by sequestering miR-20b-5p, thereby alleviating its direct repression of SLC7A11. Our findings exemplify the ceRNA regulatory paradigm highlighted in other pathological contexts [36].

Limitations and future perspectives

While this study establishes a novel link between circRNA‑mediated ceRNA mechanisms and ferroptosis in SAP, several limitations should be acknowledged. First, the therapeutic assessment was conducted at a single dose of Homo (15 mg/kg), leaving the dose‑response relationship and optimal dosing regimen uncharacterized. Second, the absence of a circDNMT3B overexpression model, due to technical challenges in circular RNA vector construction, limits bidirectional functional verification. Third, the dynamic expression profile of circDNMT3B during SAP progression remains unclear, and potential synergistic effects involving pathways such as Nrf2 warrant further investigation. Finally, it is important to recognize the inherent limitations of animal models: murine systems cannot fully recapitulate human SAP pathophysiology, the upstream regulatory mechanisms of circDNMT3B in humans remain undefined, and the bioavailability and pharmacokinetic properties of Homo itself require careful consideration. While this study establishes the therapeutic potential of Homo against established SAP, future investigations are warranted to fully delineate its scope and mechanism. Therefore, studies with larger clinical cohorts are needed to further validate the efficacy and safety of Homo and its targeted pathway in human SAP. To address the remaining issues, future work should focus on: combining circDNMT3B overexpression to confirm its molecular sponge function; systematically evaluating Homo across a range of doses (e.g., 5, 15, and 30 mg/kg) in vivo to determine its therapeutic window and comparing its efficacy in preemptive (prevention) versus post-onset (treatment) regimens; and assessing its translational potential in clinically relevant models such as human-derived pancreatic organoids.

In conclusion, our findings suggest that the natural compound Homo may confer protection against SAP. We propose a model in which this protection is achieved, at least in part, through a pathway involving the upregulation of circDNMT3B. Our data are consistent with a mechanism whereby circDNMT3B acts as a sponge for miR-20b-5p, potentially leading to the upregulation of SLC7A11 and the subsequent inhibition of ferroptosis (Fig. 14). These insights contribute to a better understanding of SAP pathogenesis and highlight the circDNMT3B/miR-20b-5p/SLC7A11 axis as a potential target for future therapeutic exploration.

Fig. 14.

Schematic mechanism of homo alleviating severe acute pancreatitis (SAP) by suppressing ferroptosis via the circDNMT3B/miR-20b-5p/SLC7A11-GPX4 axis

Authors’ contributions

Zheng Li: Writing-review and editing, data planning, conceptualization, writing-review and editing, supervision. Xiao Teng: Writing —— original draft, visualization, form analysis, data planning, conceptualization. ShanFeng Sheng: Visualization, validation, supervision. Mingyuan Pan: Visualization, validation, supervision. Yuyang Li: Visualization, validation, supervision.

Funding

This work was supported by the Guangxi Natural Science Foundation [2023GXNSFAA026206].

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This animal experiment was approved by the Guangxi Medical University Experimental Animal Ethics Committee, with approval number: 202401022.

Consent for publication

Data availability statements:

Competing interests

The authors declare no competing interests.