Monotropein mitigates atherosclerotic plaque vulnerability through circ_0140736/miR-495/SIRPα-mediated efferocytosis

Summary

Efferocytosis is a biological process in which phagocytic cells engulf and degrade senescent, damaged, or dead cells to prevent secondary necrosis and inflammation. This process is implicated in various diseases, including atherosclerosis (AS). Enhancing efferocytosis efficiency may offer a promising therapeutic strategy for managing AS. Consequently, increasing attention is being paid to pharmacological agents that can promote efferocytosis. Monotropein (MON), the predominant iridoid glycoside in Morinda officinalis, has demonstrated anti-inflammatory and analgesic properties in previous studies. However, its role in regulating atherosclerotic plaque stability remains unclear. In this study, we show that MON enhances plaque stability in AS. The expression of circ_0140736 was found to be elevated in unstable atherosclerotic plaques. MON suppressed circ_0140736 expression and alleviated miR-495 sequestration via a competing endogenous RNA mechanism, thereby increasing miR-495 levels and downregulating signal regulatory protein α (SIRPɑ) expression. In conclusion, MON exerts its anti-atherosclerotic effects by promoting efferocytosis through the circ_0140736/miR-495/SIRPɑ axis, suggesting its potential as a promising therapeutic candidate for AS.

Graphical abstract

Highlights

• •Monotropein promotes plaque stability in ApoE^−/− mice
• •circ_0140736/miR-495/SIRPα axis mediates efferocytosis
• •Efferocytosis enhancement reduces atherosclerotic plaque vulnerability

Biological sciences; Immunology; Cell biology

Introduction

According to the World Health Statistics 2023, coronary heart disease and stroke are the two leading causes of death worldwide.¹ Atherosclerosis (AS) is a pathological basis of cardiovascular and cerebrovascular diseases, with unstable atherosclerotic plaques being the primary cause of acute stroke events.^2,3 Despite current international guidelines advocating lifestyle changes, risk factor control, and antiplatelet and statin therapy for AS treatment, its incidence remains high.^4,5,6 Therefore, the development of innovative therapeutic strategies for AS treatment is of utmost importance.

Monotropein (MON), extracted from Morinda officinalis, is a primary iridoid glycoside with pharmacological activities, including anti-osteoporotic, anti-inflammatory, and antioxidant effects.⁷ For instance, Shi et al. demonstrated that MON mitigates oxidative stress by activating the Akt/mTOR-mediated autophagy pathway, while Wang et al. found that MON alleviates dexamethasone-induced muscle atrophy through the AKT/mTOR/FOXO3a signaling cascade.^8,9 Thus, MON has potential as a pharmaceutical agent; however, its effects on AS remain unclear.

Efferocytosis is a process whereby phagocytes recognize and eliminate senescent and apoptotic cells, thereby preventing secondary necrosis and inflammation.^10,11 Impaired efferocytosis is the underlying cause of an increasing number of chronic inflammatory diseases. Multiple studies have suggested that efferocytosis is instrumental in AS progression.¹² Indeed, impaired efferocytosis in advanced atherosclerotic plaques contributes significantly to the formation of necrotic cores, resulting in secondary necrosis and inflammation induced by apoptotic cells.^13,14 Recent advancements in research on the molecular mechanisms of efferocytosis have revealed numerous potential targets for therapeutic interventions, offering novel treatment strategies for diseases characterized by deficiencies in efferocytosis and impaired inflammatory regression.^15,16,17 However, the clinical application of these therapies in AS remains unclear. In this study, we investigated the effect of MON on AS both in vitro and in vivo.

However, the precise molecular mechanism by which MON confers atheroprotection, particularly its influence on plaque stability, remains elusive. Moreover, it is completely unknown whether MON’s effects involve circular RNA (circRNA)/ microRNA (miRNA) regulatory axes, especially in the context of macrophage efferocytosis within atherosclerotic plaques. Given the pivotal role of efferocytosis in maintaining plaque integrity and the growing recognition of non-coding RNAs in cardiovascular disease, we hypothesized that MON enhances plaque stability by promoting macrophage efferocytosis via a specific circRNA/miRNA-dependent pathway. To test this hypothesis, we aimed to investigate the effect of MON on AS both in vitro and in vivo and to elucidate the underlying mechanism focusing on the circRNA/miRNA-regulated efferocytic pathway.

Results

MON attenuates AS progression in ApoE-deficient mice

To investigate the potential impact of MON (chemical structure shown in Figure S1) on AS, 7-week-old ApoE^−/− mice groups were fed either a normal chow diet (ND) or high-fat diet (HFD) for 4 weeks and subsequently treated with or without MON (80 mg/kg/d) while continuing ND/HFD for an additional 8 weeks (Figure 1A). These groups were classified as “ND,” “ND + MON,” “HFD,” and “HFD + MON.” Daily chow consumption did not differ significantly among the four groups. After 12 weeks of feeding with ND or HFD, no significant differences were detected in body weight or levels of total cholesterol, triglycerides, low-density lipoproteins, or high-density lipoproteins between the ND and ND + MON or HFD and HFD + MON groups (Figure S2). Additionally, we have confirmed that MON did not impact hepatic function, renal function, myocardial enzymes, or coagulation function in ApoE^−/− mice (Table S1). MON treatment does not induce adverse hematological effects (Figure S3). We evaluated the development of atherosclerotic plaques in both the aorta and aortic arch, and the successful induction of severe AS was confirmed by oil red O (ORO) staining in our model. Compared with the ND group, the HFD group exhibited a significant increase in both lesion size and lipid content. In contrast, a significant reduction in plaque formation was observed in the aorta and aortic sinus of the MON-treated group compared to that of the untreated group (Figures 1B–1E).

Figure 1.

Monotropein mitigated atherosclerosis progression in ApoE^−/− mice

(A) Flow chart depicting the experimental protocol used in the in vivo experiments.

(B and C) ORO analysis was performed to determine the proportion of lesion area relative to the total aortic area in ApoE^−/− HFD-fed mice and treated with or without monotropein. Scale bars, 2 mm.

(D and E) ORO analysis was conducted to evaluate the relative lesion area of the aortic root in ApoE^−/− HFD-fed mice, both with and without monotropein treatment. Scale bars, 100 μm. The data are expressed as the mean ± SEM. n = 8–10 for each group. ∗p < 0.05, ∗∗p < 0.01, using one-way ANOVA followed by Tukey’s post hoc test. MON, monotropein; ND, normal chow diet; HFD, high-fat diet; ANOVA, analysis of variance.

MON enhanced stability of atherosclerotic plaques in ApoE^−/− mice

Atherosclerotic plaques in the MON group exhibited smaller necrotic core areas than those without MON treatment (Figures 2A and 2B). Immunohistochemical analysis revealed a significant augmentation in both collagen content and fibrous cap thickness within the atherosclerotic plaques and a notable reduction in the MOMA (anti-mouse monocyte/macrophage antibody)-2⁺ area in the MON group (Figures 2A–2F). Accordingly, the plaque instability indices in the MON-treated group were significantly lower than those in the untreated group (Figure 2G). The plaque instability index was calculated using the following formula: (area of ORO⁺ staining) + (area of MOMA-2⁺ staining)/(area of α-SMA⁺ staining) + (area of collagen staining).¹⁸ The reduction in necrotic area and increase in fibrous cap thickness suggested that MON promotes plaque stability. Our analysis revealed that MON had the most significant effect on MOMA (Figure 2F); note: check Figure 1 for ORO staining. Next, to verify the reliability of the experimental data, we selected atorvastatin, a widely used, clinically approved drug, as a positive control. We found that the effect of MON on plaque progression and stability closely resembles that of atorvastatin (Figures S4 and S5).

Figure 2.

Monotropein enhanced stability of atherosclerotic plaques in ApoE^−/− mice

(A) Cross-sectional histological and fluorescence microscopy analyses were conducted on aortic roots obtained from ApoE^−/− HFD-fed mice, both with and without monotropein treatment. The samples underwent H&E, MASSON, α-SMA, or MOMA-2 staining to facilitate the identification of the necrotic core area. Scale bars, 100 μm.

(B–G) Quantification of average necrotic core area, fibrotic cap thickness, collagen content ratio, α-SMA-positive smooth muscle cells, and MOMA-2-positive macrophages in relation to the plaque area and instability index within the aortic root of ApoE^−/− HFD-fed mice with or without monotropein treatment. Data are expressed as mean ± SEM. n = 8–10 for each group. ∗p < 0.05 and ∗∗p < 0.01, using one-way ANOVA followed by Tukey’s post hoc test. MON, monotropein; ND, normal chow diet; HFD, high-fat diet; ANOVA, analysis of variance.

Circ_0140736 is upregulated in unstable plaques and downregulated by MON

We used high-resolution magnetic resonance imaging (HRMRI) and multi-slice computed tomography angiography (MSCTA) to evaluate the stability of plaques in patients with AS,¹⁹ and extensive validation was performed by histological analysis (Figures 3A and 3B). We performed RNA-seq on human carotid atherosclerotic plaque tissues from three patients with vulnerable plaque (VP) and three with stable plaque (SP) to profile circRNA expression (Figure 3C). The volcano plot derived from RNA-seq data identified circ_0140736 (circular RNA_0140736) as a significantly upregulated circRNA upon MON treatment (red point, labeled), consistent with its inclusion in the shortlist of eight circRNAs in Figure S6. Additionally, we conducted real-time quantitative PCR (RT-qPCR) to validate the eight plasma circRNAs that exhibited significant changes in a second cohort comprising healthy controls (HCs), patients with SP, and patients with VP. The plasma of patients with VP was demonstrated to exhibit the highest levels of circ_0140736 compared to that of patients with SP, and that plasma circ_0140736 levels were elevated in patients with SP relative to that of those in the control group (Figure 3D). The baseline and clinical characteristics of the study population are presented in Table S2. A comparison of various parameters, including age, sex, and lipid profiles, across the three groups did not yield any significant differences. To further investigate the correlation between circ_0140736 and progression of atherosclerotic plaques, we assessed its expression in ApoE^−/− HFD-fed mice. As anticipated, the plasma and plaque levels of circ_0140736 were observed to be elevated in HFD-fed ApoE^−/− mice, whereas MON treatment inhibited its expression (Figure 3E). The findings indicate that circ_0140736 potentially contributes significantly to the advancement and destabilization of atherosclerotic plaques, whereas MON demonstrates inhibitory effects on its expression.

Figure 3.

Monotropein alleviates atherosclerosis in ApoE^−/− mice through circ_0140736

(A) Type III plaque: eccentric plaque of the left internal carotid artery. T1WI and T2WI were isointense without enhancement; the surface of the plaque was smooth without ulceration. CTA indicated a CT value of <60 HU (Hounsfield units) (lipid plaque), no calcification, and intraplaque hemorrhage. JV: internal jugular vein.

(B) A type IV-V lesion in the internal carotid artery was observed, with histological examination detecting a lipid-rich necrotic core. On multicontrast-weighted MR, it was rich in a lipid necrotic core and showed mixed signal intensity on T1WI and T2WI, with uneven surfaces and slight enhancement. CTA revealed minimal calcification.

(C) Heatmap displaying the differential expression profiles of circRNAs, with red indicating upregulation and green indicating downregulation. The study included three patients with vulnerable plaque (VP) and three with stable plaque (SP).

(D) RT-qPCR was performed to investigate the expression of circRNAs in a cohort comprising 10 patients with VP, 10 patients with SP, and 10 healthy controls (HC). Eight candidate circRNAs were examined, of which circ_0140736 exhibited the highest expression level.

(E) Expression of circ_0140736 was analyzed in the plasma and aorta of atherosclerotic lesions in HFD-fed ApoE^−/− mice and ND-fed ApoE^−/−mice (control group) at 12 weeks using RT-qPCR. Monotropein was found to inhibit its expression. Data are expressed as the mean ± SEM. n = 8–10 for each group. ∗∗p < 0.01, ∗∗∗p < 0.001, ##p < 0.01, ###p < 0.01, using one-way ANOVA followed by Tukey’s post hoc test. MON, monotropein; HFD, high-fat diet; VP, vulnerable plaque; SP, stable plaque; HC, healthy control; ANOVA, analysis of variance.

MON alleviates AS in ApoE^−/− mice through circ_0140736

To investigate the impact of MON on AS progression mediated by circ_0140736, ApoE^−/− mice were subjected to an HFD for 4 weeks and were subsequently treated with or without MON (80 mg/kg/d). They were administered adeno-associated virus (AAV) RNA (sh-circRNA) or sh-negative control (NC) and maintained on an HFD for 8 weeks (Figure 4A). To assess the effectiveness of circRNA inhibition, we quantified its expression levels in both plasma and aortic samples and found that the expression levels of circRNA, as detected by RT-qPCR, were significantly decreased in the sh-circRNA-treated group compared to both the sh-NC- and PBS-treated groups (Figures 4B and 4C). We evaluated the development of atherosclerotic plaques in both the aorta and aortic arch, and the successful induction of severe AS was confirmed by ORO staining in our model. Compared to the group without MON, the MON group demonstrated a significant reduction in plaque formation in both the aorta and aortic sinus. Moreover, this effect was enhanced by sh-circRNAs (Figures 4D–4G). Meanwhile, we observed that the overexpression of circ_0140736 significantly reversed the effect of MON (Figure S7). These findings suggested that MON may attenuate the in vivo atherosclerotic burden through a circ_0140736-mediated mechanism.

Figure 4.

Monotropein inhibits plaque progression through circ_0140736 in ApoE^−/− mice

(A) Schematic of experimental design.

(B) Expression of circ_0140736 in the plasma and aorta of atherosclerotic lesions in ApoE^−/− mice treated with PBS, adeno-associated virus 2 RNA (sh-circRNA1, sh-circRNA2, and sh-circRNA3), or sh-NC was evaluated using RT-qPCR analysis. The results demonstrated that treatment with sh-circRNA significantly decreased the expression level compared to PBS or sh-NC treatment; sh-circRNA2 and sh-circRNA2 were used in subsequent experiments. (n = 5/group).

(C and D) ORO analysis was performed to determine the proportion of lesion area relative to the total aortic areas in ApoE^−/− mice with HFD, both with and without monotropein treatment. n = 9/group; scale bars, 2 mm.

(E and F) ORO analysis was conducted to assess the relative lesion area of the aortic root in ApoE^−/− mice fed with HFD, both with and without monotropein treatment, n = 9/group. Scale bars, 100 μm. ∗∗p < 0.01 and ∗∗∗p < 0.001. Direct comparison between the HFD+MON and HFD+MON+sh-circRNA groups revealed a statistically significant further reduction in plaque area (p < 0.01).

MON enhanced stability of atherosclerotic plaques in ApoE^−/− mice via circ_0140736

Atherosclerotic plaques in the MON group, compared to those without MON treatment, exhibited smaller areas of necrotic core, and this effect could be further enhanced by sh-circRNA (Figures 5A and 5B). Immunohistochemical analysis revealed a significant increase in collagen content and fibrous cap thickness, accompanied by a notable decrease in the area positive for MOMA-2 staining within the atherosclerotic plaques in the MON group. Moreover, this effect was further enhanced by sh-circRNAs (Figures 5A–5F). Therefore, compared to plaque instability indices in the untreated group, that of the MON-treated group was significantly reduced, and this effect was further enhanced by sh-circRNA (Figure 5G). Meanwhile, the overexpression of circ_0140736 reversed the effect of MON (Figure S8). The circ_0140736-dependent decrease in necrotic area and increase in fibrous cap thickness indicated that MON promoted plaque stability. The results of our analysis indicated that MON had the most significant impact on MOMA (Figure 5F). These experimental results showed that the impact of MON on the progression and stability of plaques is likely to be mediated by macrophages; note: check Figure 4 for ORO staining.

Figure 5.

Monotropein enhanced stability of atherosclerotic plaques in ApoE^−/− mice through circ_0140736

(A) Cross-sectional histological and fluorescence microscopy analyses were conducted on aortic roots obtained from ApoE^−/− HFD-fed mice either with or without monotropein treatment. Additionally, sh-circRNA was used for intervention. The samples were stained using H&E, MASSON, α-SMA, or MOMA-2. Scale bars, 100 μm.

(B–G) Quantification of average necrotic core areas, fibrotic cap thickness, collagen content ratio, α-SMA-positive smooth muscle cells, and MOMA-2-positive macrophages relative to the plaque area and instability index within the aortic root in ApoE^−/− HFD-fed mice, both with and without monotropein treatment. In addition, sh-circRNA was used as an intervention. ∗∗p < 0.01, ∗∗∗p < 0.001, n = 8–10 per group. The combination treatment (HFD+MON+sh-circRNA) resulted in a significantly greater increase in collagen content (p < 0.01) and efferocytosis index (p < 0.01) compared to monotropein treatment alone (HFD+MON).

circ_0140736 was upregulated in macrophages of unstable plaques

To obtain a more comprehensive understanding of the precise localization and potential functional significance of circ_0140736 in advanced AS, RNA-fluorescence in situ hybridization (FISH) analysis was performed on unstable plaques. Based on the combined immunostaining for circ_0140736 and Mac-3, most lesion macrophages expressed circ_0140736 (Figure 6A). In addition, a restricted population of smooth muscle cells (SMCs) expressed circ_0140736 (Figure 6B). RAW 264.7 cells were stimulated with oxidized low-density lipoprotein (ox-LDL). The expression levels of circ_0140736 significantly increased in a dose- and time-dependent manner following ox-LDL treatment (Figures 6C and 6D). The localization of circ_0140736 was predominantly observed in the cytoplasm rather than in the nucleus, as determined by in vitro RNA-FISH. This suggested that circ_0140736 primarily plays a functional role at the post-transcriptional level. In addition, a significant increase in circ_0140736 expression was observed post-ox-LDL treatment (Figures 6E and 6F).

Figure 6.

Expression of circ_0140736 was observed in macrophages within unstable plaques

(A) RNA-fluorescent in situ hybridization (RNA-FISH) was performed on frozen sections of the aortic root from ApoE^−/− mice fed a high-fat diet for 12 weeks to detect circ_0140736 (labeled in green), while immunostaining for Mac-3 (labeled in red) was conducted. The arrows indicate the presence of macrophages expressing circ_0140736. Scale bars, 50 μm.

(B) RNA-FISH was performed on frozen sections of the aortic root from ApoE^−/− mice fed a high-fat diet for 12 weeks to detect circ_0140736 (labeled in red), while immunostaining for α-SMA (labeled in green) was conducted. The arrows indicate the presence of smooth muscle cells (SMCs) expressing circ_0140736. Scale bars, 50 μm.

(C and D) Expression of circ_0140736 in RAW 264.7 cells treated with ox-LDL was quantified using RT-qPCR at various concentrations (0, 25, 50, 100, and 150 μg/mL) and time intervals (0, 3, 6, 12, and 24 h).

(E and F) RNA-FISH was employed to detect the presence of circ_0140736 (labeled in red) in RAW 264.7 cells after ox-LDL treatment, while the nuclei were stained blue. U6 was used as a positive control and was primarily expressed in the nucleus. 18s was used as an internal control and was mostly expressed in the cytoplasm. U6 RNA, which is predominantly localized in the nucleus, was used as the positive control. The 18s RNA served as an internal control and was predominantly expressed in the cytoplasm. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, n = 6/group. Scale bars, 20 μm.

Combined effect of MON and circ_0140736 on macrophage efferocytosis in atherosclerotic plaques of ApoE^−/− mice

Impaired efferocytosis in advanced atherosclerotic plaques contributes significantly to the formation of necrotic cores, resulting in secondary necrosis and inflammation induced by apoptotic cells.²⁰ Intraplaque apoptotic cells were evaluated by detecting caspase-3 activation (Figure 7A), which demonstrated a decrease in the number of apoptotic cells within lesions treated with MON compared to those in other groups. To further confirm the association between the reduction in apoptotic cell count and enhanced efferocytosis, we assessed the quantity of “unbound” apoptotic (not co-stained with macrophages) and “bound” apoptotic bodies (co-stained with macrophage marker Mac-3), indicating the occurrence of efferocytosis activity (Figures 7A and 7B). The data are presented as the ratio of non-free to free apoptotic bodies, with higher values indicating increased efferocytosis. Our measurements demonstrated a significant augmentation in efferocytosis within the lesions of the MON-treated group compared to that of the untreated group. Moreover, this effect was enhanced by sh-circRNAs (Figures 7A and 7C), suggesting that MON exerts a beneficial effect on plaque stability by reducing the size of the necrotic core and inhibiting efferocytosis via circ_0140736. Subsequently, we analyzed proteins associated with efferocytosis, finding that MON had the most significant impact on signal regulatory protein α (SIRPα; Figures 7D and 7E). In addition, we found that MON directly enhances macrophage efferocytosis in vitro (Figure S9).

Figure 7.

Synergistic effect of monotropein and circ_0140736 on macrophage efferocytosis in atherosclerotic plaques of ApoE^−/− mice

(A) Representative images of aortic root sections were obtained, where apoptotic cells were labeled using cleaved caspase-3 (green), macrophages were identified by Mac-3 staining (red), and nuclei were counterstained with DAPI (blue). Scale bars, 50 μm.

(B) Data are presented as the proportion of cells showing positive staining compared to the total number of nuclei, ∗p < 0.05, ∗∗p < 0.01, n = 9/group.

(C) Each cleaved caspase-3-positive cell was categorized as either macrophage-associated (representing “not-free” apoptotic bodies) or non-macrophage-associated (representing “free” apoptotic bodies). Data are presented as the ratio of non-free apoptotic bodies to free apoptotic cells, with stars indicating free apoptotic bodies and arrows indicating not-free apoptotic bodies. ∗p < 0.05 and ∗∗p < 0.01; n = 9/group.

(D and E) Impact of monotropein on macrophage efferocytosis-related pathways, particularly the SIRPɑ pathway. ∗p < 0.05, n = 3/group.

Circ_0140736 upregulated anti-phagocytic molecule SIRPα expression in a transcription-independent manner

Studies have demonstrated that SIRPα acts as an inhibitory receptor for phagocytosis on the surface of phagocytic cells, binding to CD47 expressed on apoptotic cells to regulate the process.²¹ The reduction in SIRPα expression on phagocytes’ surface leads to an augmentation in phagocytosis.²² The protein levels of SIRPα were significantly upregulated in a dose- and time-dependent manner following ox-LDL treatment, which was consistent with the expression pattern of circ_0140736 post-ox-LDL treatment (Figures 8A and 8B). To investigate the potential direct involvement of SIRPα in the functional role of circ_0140736 during impaired efferocytosis, we conducted in vitro gain-of-function and loss-of-function experiments in macrophages to assess the impact of circ_0140736 on SIRPα. The protein levels of SIRPα were decreased upon siRNA transfection targeting circ_0140736 in the presence of ox-LDL group compared to the control group (Figure 8C). In contrast, the adenovirus-mediated overexpression of circ_0140736 significantly augmented ox-LDL-induced SIRPα expression (Figure 8D). Furthermore, we investigated whether circ_0140736 upregulates the expression of SIRPα through transcriptional mechanisms. Ox-LDL treatment had no impact on the mRNA levels of SIRPα in macrophages (Figure 8E). Additionally, the modulation of circ_0140736 expression did not lead to any significant changes in SIRPα mRNA levels (Figure 8F). Therefore, circ_0140736 may exert a regulatory influence on SIRPα expression independently of transcriptional mechanisms. Cytotoxicity assays showed that MON had no significant toxicity on macrophage viability or apoptosis at the tested concentrations, and it inhibited ox-LDL-induced apoptosis (Figure S10).

Figure 8.

Expression of SIRPɑ protein was upregulated by circ_0140736 following in vitro ox-LDL treatment

(A and B) Western blot analysis was performed to assess SIRPɑ protein expression in RAW 264.7 cells treated with ox-LDL at various doses (0, 25, 50, 100, and 150 μg/mL) and time points (0, 3, 6, 12, and 24 h). ∗p < 0.05, ∗∗p < 0.01, n = 6/group.

(C) Western blot analysis was performed to assess the expression of SIRPɑ protein in RAW 264.7 cells treated with ox-LDL, both with and without knockdown of circ_0140736, ∗p < 0.05, ∗∗p < 0.01, #p < 0.05 vs. ox-LDL, n = 6/group.

(D) Western blot analysis was conducted to evaluate the expression of SIRPɑ protein in RAW 264.7 cells treated with ox-LDL, both with and without circ_0140736 overexpression, ∗p < 0.05, ∗∗p < 0.01, #p < 0.05 vs. ox-LDL, n = 6/group.

(E) The mRNA expression of SIRPɑ in RAW 264.7 cells was assessed via RT-qPCR subsequent to ox-LDL treatment.

(F) The expression of SIRPɑ mRNA in RAW 264.7 cells with either downregulated or upregulated circ_0140736 was assessed using RT-qPCR analysis.

Circ_0140736 regulated SIRPα expression by acting as a sponge of miR-495

Numerous studies have demonstrated that cytoplasmic circRNAs act as natural miRNA decoys, impeding post-transcriptional miRNA activity and attenuating the binding of endogenous miRNAs to target genes.^23,24 An online bioinformatics database (http://34.236.212.39/microrna/getGeneForm.do) was searched to identify potential miRNA recognition elements on circ_0140736. We detected one putative miR-495 (microRNA-495) site in both human and mouse sequences (Figure 9A). Additionally, miRBase analysis identified SIRPα as a putative target gene of miR-495 (Figure 9B). The sequences of circ_0140736 or SIRPα 3′ UTR, which contain the miR-495 binding site, were inserted downstream of the luciferase gene in the reporter plasmid for utilization in luciferase assays. Mutated versions of circ_0140736 (circ_0140736 mut) or SIRPα 3′ UTR (SIRPα 3′ UTR mut) were also included. The luciferase assay demonstrated a significant decrease in circ_0140736 expression in the miR-495 transfection group compared to that of the NC group in the wild-type cohort. However, the inhibitory effect was reversed upon binding site mutation, indicating a specific interaction between miR-495 and either circ_0140736 or SIRPα 3′ UTR (Figures 9C and 9D). To enhance comprehension of the interplay among circ_0140736, miR-495, and SIRPα, co-transfection experiments involving the SIRPα 3′ UTR, miR-495, and circ_0140736 plasmid were performed; luciferase assays were conducted to analyze their effects. The results indicated that circ_0140736 effectively mitigates the suppressive influence exerted by miR-495 on SIRPα expression (Figure 9E) and provided compelling evidence that circ_0140736 exerts regulatory effects on SIRPα expression through its interaction with miR-495. Finally, to further validate the sponge-like interaction between circ_0140736 and miR-495, RNA immunoprecipitation was performed on RAW 264.7 cell extracts using an antibody specific for Ago2, a crucial component of the RNA-induced silencing complex. We discovered that the levels of circ_0140736 and miR-495 RNA were elevated in the Ago2-associated miRNA ribonucleoprotein complexes (Figure 9F). Therefore, circ_0140736 acts as a competitive endogenous RNA for miR-495, leading to SIRPα expression upregulation.

Figure 9.

circ_0140736 functions as miR-495 sponge to regulate SIRPɑ expression

(A) StarBase prediction identified multiple potential miR-495 binding sites within the circ_0140736 sequence in both humans and mice.

(B) The analysis conducted by StarBase revealed that SIRPɑ is predicted to be a target gene of miR-495. The binding sites of miR-495 on SIRPɑ are visually presented.

(C) The cDNA sequence of circ_0140736 was cloned downstream of the luciferase gene (Luc-circ-Wt) and transfected into HEK293 cells along with a miR-495 mimic or negative control (NC). To minimize nonspecific binding, the miR-495 binding sites within circ_0140736 were mutated to generate Luc-circ-Mut. Luciferase activity was quantified 48 h post-transfection. ∗∗p < 0.01, n = 6/group.

(D) Similar to (C), HEK293 cells were co-transfected with miR-495 mimics or NC in combination with Luc-SIRPɑ-WT or MUT, and the luciferase activity was measured 48 h post-transfection. ∗∗p < 0.01, n = 6/group.

(E) HEK293 cells were co-transfected with Luc-SIRPɑ-WT and miR-495 along with circ_0140736 plasmid or vector to confirm the competing endogenous RNA activity of circ_0140736. Luciferase activity was measured 48 h post-transfection. ∗p < 0.05, n = 6/group.

(F) RNA-binding protein immunoprecipitation assays were performed using cell lysate input, normal mouse IgG, or anti-Ago2 to determine the relative expression levels of circ_0140736 and miR-495 in RAW 264.7 cells through RT-qPCR analysis. ∗∗p < 0.01, ∗∗∗p < 0.001, n = 6/group.

Discussion

Previous studies have demonstrated that MON, a major iridoid glycoside in M. officinalis, exhibits anti-inflammatory and analgesic effects.^7,25,26,27 However, its role in AS remains unclear. This study provides evidence that MON plays a role in attenuating AS progression and stabilizing AS plaques (Figures 1 and 2).

CircRNA is a single-stranded, covalently closed, non-coding RNA that primarily functions as a sponge for miRNAs, regulates protein transcription and post-transcriptional processes, interacts with RNA-binding proteins, and encodes a limited number of proteins.²⁸ Numerous studies have demonstrated that cytoplasmic circRNAs act as natural miRNA decoys, impeding post-transcriptional miRNA activity and attenuating the binding of endogenous miRNAs to target genes.²⁹ The distribution of circRNAs across various tissues and organs, coupled with their remarkable stability, resistance to degradation, and high conservation during evolution, endows them with immense potential as novel targets for gene therapy. Accumulating evidence strongly suggests a pivotal role for circRNAs in cardiovascular diseases.^30,31,32 Gao et al. demonstrated that circRNA CNEACR modulates necroptosis in myocardial cells by inhibiting Foxa2.³³ Zhang et al. identified that circRNA Galntl6 enhances stress-induced hypertension by upregulating Lig3 in the rostral ventrolateral medulla oblongata through miR-335.³⁴ However, the impact and molecular mechanisms of circRNAs in atherosclerotic plaque instability remain unclear. To assess whether MON affects atherosclerotic plaque vulnerability through circRNA-mediated mechanisms, HRMRI and MSCTA were employed to evaluate carotid plaque characteristics, and RNA-seq was performed to profile circRNA expression. The eight circRNAs with the highest expression levels were validated through PCR, revealing that circ_0140736 exhibited the highest expression level. MON suppressed the expression of circ_0140736 (Figure 3). Although detailed information on circ_0140736 is limited in the NCBI database, preliminary studies indicate that it is highly conserved across species. Our bioinformatics analysis revealed that circ_0140736 shares over 80% nucleotide sequence homology between humans and mice, suggesting that it may play a similar regulatory role in both species.³⁵ The effect of MON on the instability of AS plaques mediated by circ_0140736 was also elucidated (Figures 4 and 5). To investigate the impact of MON on atherosclerotic stability through circ_0140736, circ_0140736 was initially confirmed via FISH analysis to be predominantly expressed within plaque-associated macrophages. Furthermore, cellular FISH analysis indicated the cytoplasmic localization of circ_0140736 in macrophages, implying a post-transcriptional regulatory role (Figure 6).

Unstable carotid plaque is the primary etiology of ischemic cerebrovascular disease. Intraplaque hemorrhage, thin fibrous cap, and necrotic core are the fundamental pathological characteristics of plaque instability.^36,37,38 Biologically, efferocytosis refers to phagocytic engulfment and subsequent degradation of senescent, damaged, or deceased cells by specialized phagocytes to prevent secondary necrosis and inflammation.³⁹ However, in atherosclerotic plaques, the efficiency of efferocytosis is approximately 1/20^th of that observed in other body regions.⁴⁰ Therefore, improving the efferocytosis efficiency of macrophages has received increasing attention for AS treatment. Similar to Doran et al., we found that late plaque lipid core enlargement was caused by limited clearance, that is, exocytosis disorders.^10,41 Our analysis of previous studies revealed that MON affects SMCs, collagen fibers, macrophages, and lipid components (Figures 2 and 5). However, MON exhibited the most significant effect on macrophages and lipid components, substantially reducing the lipid core, which consists primarily of deceased cells. Does MON enhance the stability of AS plaques by promoting the efferocytosis of deceased cells?

Next, we performed caspase-3 staining to label the apoptotic cells within the plaques, whereas Mac-3 staining was used to identify the macrophages (phagocytes) present in the plaques. The co-localization of these two markers indicated that phagocytosis was responsible for the removal of apoptotic cells, whereas cells without co-localization were considered non-engulfed. Therefore, MON enhanced macrophage efferocytosis within the plaques (Figure 7). Subsequently, RT-PCR/WB (Western blot) analysis was conducted on the pathways associated with efferocytosis, and MON had the most pronounced impact on SIRPα (Figure 4). SIRPα is a phagocytic inhibitory receptor expressed on macrophage surfaces, which binds to the CD47 present on apoptotic cell surfaces, thereby exerting an anti-efferocytosis effect. The reduction in SIRPα expression on the phagocyte surface leads to an augmentation in phagocytosis, whereas increased expression suggests impaired efferocytosis.^42,43,44 In addition, we found that MON directly enhances macrophage efferocytosis in vitro (Figure S9). Our findings suggested that MON modulates macrophage efferocytosis within the plaque via circ_0140736, thereby enhancing AS plaques stability.

However, the mechanism whereby circ_0140736 regulates macrophage efferocytosis in plaques remains unclear. Previous studies have demonstrated that circRNAs participate in specific physiological and pathological processes through epigenetic, transcriptional, and post-transcriptional regulatory mechanisms.⁴⁵ We conducted an analysis of three databases (pprint, RsiteDB, and RNABindRPlus) and found no potential binding site between circ_0140736 and SIRPα. Additionally, co-immunoprecipitation experiments did not reveal any direct interaction between circ_0140736 and SIRPα. Through database analysis, we identified a simultaneous interaction between miR-495, circ_0140736, and SIRPα. Subsequently, we experimentally validated the regulatory role of circ_0140736 on SIRPα through miR-495 using dual-luciferase assays, RNA immunoprecipitation, and PCR/western blotting techniques (Figures 8 and 9).

It is noteworthy that extensive pharmacological studies have established MON’s potent anti-inflammatory and anti-oxidative properties, primarily mediated through the inhibition of the NF-κB signaling pathway and the activation of the Nrf2/HO-1 axis, respectively.^7,26,27 These effects are well-known to exert beneficial impacts on AS by mitigating endothelial dysfunction, vascular inflammation, and oxidative stress. Our study unveils a novel, specific mechanism wherein MON promotes efferocytosis via the circ_0140736/miR-495/SIRPα axis, directly addressing the defective clearance of apoptotic cells within advanced plaques. We postulate that the plaque-stabilizing effect of MON likely arises from a concerted action of its multifaceted pharmacological properties. The newly identified pro-efferocytic mechanism may operate in parallel with, or be modulated by, its classic anti-inflammatory and anti-oxidative actions. For instance, the suppression of NF-κB signaling by MON could potentially influence the expression of efferocytosis-related genes, creating a more anti-inflammatory and pro-resolving macrophage phenotype. Conversely, efficient efferocytosis itself is a key anti-inflammatory and pro-resolving process. Future investigations are warranted to delineate the potential crosstalk between the circRNA-mediated efferocytic pathway and the canonical NF-κB/Nrf2 pathways in the context of MON’s anti-atherosclerotic action, which will provide a more integrated understanding of its therapeutic profile.

The atheroprotective effects of natural compounds like MON are often attributed to their broad anti-inflammatory and antioxidant properties. However, the precise molecular circuits, particularly those involving post-transcriptional regulators such as circRNAs and miRNAs, through which they modulate specific cellular processes like efferocytosis remain poorly delineated. While the role of circRNA/miRNA axes in AS pathogenesis—such as in lipid metabolism and inflammation—is increasingly recognized, their specific involvement in fine-tuning the efferocytic checkpoint, notably via the SIRPα “do not-eat-me” signal, represents a significant knowledge gap. Our study bridges this gap by identifying circ_0140736 as a key mediator of MON’s action. We demonstrate that MON downregulates circ_0140736, thereby relieving its sequestration of miR-495, which in turn leads to the suppression of its target SIRPα. This newly elucidated circ_0140736/miR-495/SIRPα axis provides a novel and specific mechanism by which MON enhances macrophage efferocytosis and plaque stability, moving beyond its previously characterized canonical pathways.

Our study demonstrated that MON, the principal constituent of iridoid glycosides derived from Morbiana officinalis, exerts inhibitory effects on the expression of circ_0140736 and attenuates the sponge-like adsorption of miR-495, thereby resulting in elevating the levels of miR-495. This subsequently leads to a reduction in SIRPα expression and facilitates macrophage efferocytosis. Consequently, it diminishes dead cell accumulation within the plaque, reduces necrotic core size, and ultimately mitigates atherosclerotic plaque instability. Our findings support the potential therapeutic efficacy of MON as a treatment for AS. To fully exploit its clinical application, large-scale, multicenter, collaborative clinical trials are necessary.

Limitations of the study

Several limitations of this study should be acknowledged. First, the RNA-seq data were derived from a relatively small cohort of human carotid plaque samples (VP, n = 3; SP, n = 3), and the functional validation was primarily performed in murine RAW 264.7 macrophages and ApoE^−/− mouse models. Whether the circ_0140736/miR-495/SIRPα axis operates similarly in human macrophages warrants further investigation using primary human cells. Second, although we demonstrated that MON promotes efferocytosis via this ceRNA (competing endogenous RNA) network, MerTK—a key efferocytosis receptor—also executes numerous other functions beyond apoptotic cell clearance, and the potential impact of MON on these MerTK-mediated processes remains unexplored. Third, sex differences were not examined, as only male ApoE^−/− mice were used. Given that efferocytosis and AS exhibit sexual dimorphism, validation in female animals is needed. Finally, the therapeutic efficacy and safety of MON require further evaluation in large-animal models and ultimately in clinical trials before translational application.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Dr. Ziming Ye (yeziming1620@163.com).

Materials availability

This study did not generate any unique reagents.

Data and code availability

• •All RNA-seq raw data can be accessed in the SRA public database (SRA: PRJNA1508780).
• •This paper does not report original code.
• •Any additional information for reanalyzing the data in this study is available from the lead contact upon request.

Acknowledgments

The research leading to these results received funding from the Joint Project on Regional High-Incidence Diseases Research of Guangxi Natural Science Foundation (grant no. 2024GXNSFAA010221), National Nature and Science Foundation of China (grant nos. 81901394, 82360456, 82060226, 82260240, and 81960220), Guangxi Nature and Science Foundation (grant nos. 2018GXNSFAA138010 and 2019GXNSFAA185029), as well as “Medical Excellent Award” funded by the Creative Research Development Grant from the First Affiliated Hospital of Guangxi Medical University (grant nos. 201907 and 202101). These funds played an important role in the study design, acquisition analysis, interpretation of data, and in drafting the manuscript.

Author contributions

Z.Y., W.G., and Y. Liu conceived the experiments and experimental plan. Z.W., B.H., W.Y., and S.X. performed the experiments. W.H. and Y. Lan collected and analyzed the data. Y. Liu and W.G. wrote the manuscript. Z.Y., Y. Liu, C.Q., and C.X. reviewed and edited the manuscript. All authors have read and approved the manuscript. All data were generated in-house, and no paper mill was used. All authors agree to be accountable for all aspects of work ensuring integrity and accuracy.

Declaration of interests

The authors declare no competing interests.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work, the authors used DeepSeek solely for English language polishing and formatting standardization. After using this tool, the authors reviewed and edited the content as necessary and take full responsibility for the final publication.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Rabbit polyclonal anti-SIRPα antibody	Cell Signaling Technology	Cat#13379
Rabbit polyclonal anti-MerTK antibody	Cell Signaling Technology	Cat#9137S
Rabbit polyclonal anti-SR-B1 antibody	Abcam	Cat#ab21578
Rabbit polyclonal anti-Tim-4 antibody	Abcam	Cat#ab47637
Rabbit polyclonal anti-MFG-E8 antibody	Abcam	Cat#PA5-143884
Rabbit polyclonal anti-Gas-6 antibody	Abcam	Cat#ab67099
Anti-GAPDH antibody	Abclonal	Cat#ab8245
CD68-CoralLite antibody	Proteintech	Cat#PE-65187
Mouse monoclonal anti-MOMA-2 antibody	Abcam	Cat#ab33451
Mouse monoclonal anti-α-SMA antibody	Abcam	Cat#ab7817
Rabbit polyclonal anti-Caspase-3 antibody	Cell Signaling Technology	Cat#9662S
Rabbit polyclonal anti-Bax antibody	Cell Signaling Technology	Cat#2772S
Rabbit polyclonal anti-Cleaved caspase-3 antibody	Cell Signaling Technology	Cat#9661S
Biological samples
Human carotid atherosclerotic plaque (VP, n = 3)	Guangxi Medical University	N/A
Human carotid atherosclerotic plaque (SP, n = 3)	Guangxi Medical University	N/A
Human blood samples (41 AS patients, 20 healthy controls)	Guangxi Medical University	N/A
Chemicals, peptides, and recombinant proteins
Monotropein (≥98% purity)	Sigma-Aldrich	Cat#PHL80500; CAS: 5945-50-6
Atorvastatin	Sigma-Aldrich	Cat#BP1267
Oxidized LDL (ox-LDL)	Yiyuan Biotechnology	Cat#YB-002
DMEM/high glucose medium	Servicebio	Cat#G4510-500ML
Opti-MEM reduced serum medium	Gibco	Cat#31985070
Fetal bovine serum	ScienCell	Cat#0500
RIPA lysis buffer	Beyotime	Cat#P0013B
50× cocktail protease inhibitor	Servicebio	Cat#G3405-1ML
Critical commercial assays
miRNeasy Mini Kit	Qiagen	Cat#217004
PrimeScript™ RT Master Mix	Takara	Cat#RR036A
FastStart Essential DNA Green Master kit	Roche	Cat#06924204001
DNBSEQ-T7 RNA-seq	MGI Tech	DNBSEQ-T7
Duo-Lite Luciferase Assay kit	Vazyme	Cat#LRF-M02
5× HiScript II qRT SuperMix II	Vazyme	Cat#R223-01
2× ChamQ SYBR qPCR Master Mix	Vazyme	Cat#Q711
CCK-8 assay kit	Acmec	Cat#CP002
Annexin V-FITC Apoptosis Detection Kit	Thermo Fisher Scientific (eBioscience)	Cat#BMS500FI/300
RiboFECT™ CP Transfection Kit	RiboBio	Cat#C10511
FISH Kit	RiboBio	Cat#k11-7-001
riboSCRIPT Reverse Transcription Kit	RiboBio	Cat#C11027-2
Lipo3000	Thermo Fisher Scientific	Cat#L3000015
Oil Red O staining kit	Sigma-Aldrich	Cat#MAK194
Masson trichrome staining kit	Beyotime	Cat#C0189
EZ-Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit	Millipore	Cat#17-701
Deposited data
RNA-seq raw data	This paper; NCBI SRA	PRJNA1508780
Experimental models: Cell lines
Mouse: RAW264.7 macrophages	ATCC	Cat#TIB-71
Human: HEK 293T cells	Sigma-Aldrich	Cat#07161001
Experimental models: Organisms/strains
Mouse: ApoE-/- (C57BL/6 background, male, 7-week-old)	Vital River Laboratory Animal Technology	N/A
Mouse: C57BL/6J (for BMDM isolation, 5-week-old)	Vital River Laboratory Animal Technology	N/A
Oligonucleotides
Primers for circ_0140736, see Table S3	Janis Biotechnology	N/A
Primers for miR-495, see Table S3	Janis Biotechnology	N/A
Primers for SIRPα, see Table S3	Janis Biotechnology	N/A
Primers for GAPDH, see Table S3	Janis Biotechnology	N/A
Primers for U6, see Table S4	Janis Biotechnology	N/A
siRNA targeting circ_0140736 (si-MIAT-1)	RiboBio	Cat#circ_0140736
siRNA targeting circ_0140736 (si-MIAT-3)	RiboBio	Cat#circ_0140736
miR-495-3p mimic	RiboBio	Cat#miR-495 mimic
miR-495-3p inhibitor	RiboBio	Cat#miR-495 inhibitor
Negative control siRNA (si-NC)	RiboBio	Cat#si-NC
miR-149-5p mimic	RiboBio	Cat#miR-149-5p mimic
miR-149-5p inhibitor	RiboBio	Cat#miR-149-5p inhibitor
Digoxigenin-labeled circRNA probe	Servicebio	Cat#circRNA
miRNA probe	Servicebio	Cat#miRNA
Recombinant DNA
circ_0140736 overexpression plasmid (pmirGLO vector)	Beijing Qingke Biotechnology	N/A
circ_0140736 wild-type luciferase reporter plasmid	Beijing Qingke Biotechnology	N/A
circ_0140736 mutant luciferase reporter plasmid	Beijing Qingke Biotechnology	N/A
CD47 3'UTR wild-type luciferase reporter plasmid	Beijing Qingke Biotechnology	N/A
CD47 3'UTR mutant luciferase reporter plasmid	Beijing Qingke Biotechnology	N/A
pcDNA-MIAT (MIAT overexpression plasmid)	RiboBio	N/A
pcDNA-NC (empty vector control)	RiboBio	N/A
Software and algorithms
GraphPad Prism 9.0	GraphPad Software	https://www.graphpad.com
ImageJ	NIH	https://imagej.nih.gov/ij/
Agilent Feature Extraction Software	Agilent Technologies	https://www.agilent.com
FlowJo v10	FlowJo, LLC	https://www.flowjo.com
Photoshop	Adobe	https://www.adobe.com
RNAhybrid	Bielefeld University	https://bibiserv.techfak.uni
circMIR	N/A	https://circmir.org/
TargetScan	Whitehead Institute	https://www.targetscan.org/
circBase	N/A	https://circbase.org/
circBank	N/A	https://circbank.cn/
Nanodrop2000	Thermo Fisher Scientific	https://www.thermofisher.com
SPSS 17.0	IBM	https://www.ibm.com/spss

Experimental model and study participant details

Human subjects

A total of 41 patients diagnosed with carotid atherosclerosis (AS) and 20 healthy controls were enrolled in this study. Participants were consecutively recruited from the Department of Neurology and the Health Examination Center at Guangxi Medical University, Nanning, China, between June 2019 and June 2023. The 41 patients were further classified into a stable plaque (SP) group (n = 20) and a vulnerable plaque (VP) group (n = 21) based on HRMRI and MSCTA evaluations. Exclusion criteria included: (1) history of stroke, (2) hemorrhagic stroke, (3) non-atherosclerotic diseases (e.g., aneurysms, vasculitis, moyamoya disease, arterial dissection, reversible cerebral vasoconstriction syndrome, and vertebrobasilar insufficiency), (4) severe heart, lung, liver, or renal insufficiency, and (5) contraindications to magnetic resonance imaging (MRI) or multi-slice helical computed tomography angiography (MSCTA).

The baseline characteristics of the three groups are summarized in Table S2. The mean ages were 67.6 ± 8.15 years (control), 68.1 ± 6.98 years (SP), and 66.4 ± 6.95 years (VP), with no significant difference among groups (p = 0.7549). Male participants comprised 55.00% (11/20) of the control group, 50.00% (10/20) of the SP group, and 57.14% (12/21) of the VP group, also without significant difference (p = 0.8957). All participants were of Han Chinese ancestry. Written informed consent was obtained from all participants or their authorized representatives. The study was approved by the Ethics Committee of Guangxi Medical University, Nanning, China (Approval No. 202405698, approved on May 10, 2024), and was conducted in accordance with the principles of the Declaration of Helsinki.

Animal models

Male ApoE^-/^- mice (C57BL/6 background, 7 weeks old, weighing 18–22 g) were purchased from Vital River Laboratory Animal Technology (Beijing, China). Mice were housed under specific pathogen-free (SPF) conditions at 22 ± 2°C with a 12-hour light/dark cycle and free access to food and water. All animal experiments were conducted in strict accordance with the ARRIVE guidelines and followed the United Kingdom's Animals (Scientific Procedures) Act 1986 and EU Directive 2010/63/EU. The experimental protocol was approved by the Animal Care Committee of Guangxi Medical University (Approval No. 202405698, approved on May 10, 2024).

The ApoE^-/^- mouse on a high-fat diet is a well-established and widely used model for studying advanced atherosclerosis and evaluating therapeutic interventions. A total of 150 mice were used in this study. Sixty mice were randomly divided into four groups (n = 15 per group): (1) normal chow diet (ND), (2) ND + MON (80 mg/kg), (3) high-fat diet (HFD), and (4) HFD + MON (80 mg/kg). Mice in the HFD groups were fed a high-fat diet containing 40 kcal% fat and 1.25% cholesterol (D12108C, Beijing Huafukang Biotechnology Co., Ltd.) for 12 consecutive weeks. MON (80 mg/kg) was administered via intraperitoneal injection starting from week 2 after initiation of the HFD.

The remaining 90 mice were randomly divided into six groups (n = 15 per group): (1) HFD only, (2) HFD + MON, (3) HFD + short hairpin RNA targeting circular RNA negative control (HFD + sh-circ NC), (4) HFD + sh-circ RNA, (5) HFD + MON + sh-circ NC, and (6) HFD + MON + sh-circ RNA. For shRNA intervention, mice were intravenously injected via the tail vein with adeno-associated virus (AAV) carrying sh-circRNA or sh-circ NC at a dose of 4 × 10¹¹ viral genomes per mouse, administered starting from week 2 after initiation of the HFD.

Body weight was monitored throughout the experiment. Atorvastatin (Sigma-Aldrich) was used as a positive control to benchmark the efficacy of MON in mitigating atherosclerotic plaque vulnerability. The dose of MON (80 mg/kg) was determined based on our previous pilot study and literature review (Figures S11 and S12). After 12 weeks of feeding, all mice were sacrificed and tissues were harvested for subsequent analyses.

Only male mice were used in this study; therefore, the potential influence of sex on the observed effects was not evaluated. Further studies incorporating both sexes are warranted to assess the generalizability of these findings.

Method details

High resolution MRI (HRMRI) and MSCTA analysis

Carotid vessel wall imaging utilized a previously published multi-contrast protocol for effective plaque detection, incorporating T1-weighted, T2-weighted, and post-contrast T1- and T2-weighted sequences.⁴⁶ Each lesion was classified according to the modified American Heart Association classification for MRI.⁴⁷ Stable plaques (SP) were categorized as types I–III, whereas unstable plaques were classified as types IV–V or VI. MSCTA has rapid imaging speed and exceptional tissue resolution capability for evaluating the severity of arterial stenosis and vascular wall conditions.⁴⁸ Additionally, it facilitates the evaluation of plaque components such as fiber content, lipid deposition, and calcification by analyzing CT attenuation values. Meanwhile, the HRMRI technique has demonstrated a high level of accuracy in characterizing plaque morphology, composition, and surface condition . Two blinded, independent reviewers meticulously examined each axial image until a consensus was reached.

Reagents

MON (≥98% purity; Sigma-Aldrich Chemical Corporation, St Louis, MO, USA) was dissolved in 0.9% normal saline. DMEM/high glucose medium, digoxigenin-labeled circRNA probe, 50× cocktail protease inhibitor, and miRNA probe were procured from Servicebio (Wuhan, China). The 5× HiScript II qRT SuperMix II, 4× gDNA wiper mix, 2× ChamQ SYBR qPCR Master Mix, and Duo-Lite Luciferase Assay kit were obtained from Vazyme Biotech Co., Ltd. (Nanjing, China). Opti-MEM reduced plasma medium was sourced from Gibco (Grand Island, NY, USA). Fetal Bovine plasma was purchased from ScienCell (Los Angeles, CA, USA). Color-mixed protein marker 180 and GAPDH were acquired from Abclonal (Wuhan, China). The RIPA lysis buffer was purchased from Beyotime (Shanghai, China). Lipo3000 was purchased from Thermo Fisher Scientific (Waltham, MA, USA). RiboFECT™ CP Transfection Kit, FISH Kit, and riboSCRIPT Reverse Transcription Kit were procured from RiboBio Co., Ltd. (Guangzhou,China).

Aorta staining with Oil Red O

The tissues surrounding the aorta were meticulously dissected under a microscope using microscopic tweezers and microscissors, followed by precise cutting along the medial edge of the aortic arch. The excised aortae were immersed in Oil Red O dye for 30 min to facilitate staining, after which they were gently rinsed in 75% alcohol for 10 s, washed with phosphate-buffered saline (PBS) for 20 s, and mounted onto a wax plate using acupuncture needles against a black background to facilitate precise photographic documentation. Photoshop software was used to analyze both the percentage of Oil Red O-positive areas and total aortal surface area.

Quantification of atherosclerotic lesion size in the aortic sinus

Atherosclerotic lesions in the aortic sinus were quantified as previously described with minor modifications.⁴⁹ In brief, serial paraffin sections (4-5 μm thick) were cut transversely from the aortic root, using the aortic valve leaflets as the anatomical landmark to ensure a consistent starting point across all animals. To obtain an unbiased measure of lesion burden, multiple sections collected at systematic, equal intervals (every 5th section, representing a 20-25 μm interval) were analyzed for each mouse. Selected sections were stained with Hematoxylin and Eosin (H&E) or Oil Red O (ORO). The cross-sectional lesion area (defined as the intimal area within the internal elastic lamina) in each section was outlined using ImageJ software. The mean lesion area for each animal was calculated by averaging the areas from all analyzed sections. The lesion volume for the sampled region was estimated by multiplying the mean lesion area by the total length represented. All analyses were performed by investigators blinded to the experimental groups.

Hematoxylin and eosin staining of tissue sections

The sections were air-dried for 30 min and subsequently rinsed thrice with distilled water. Hematoxylin staining was performed for 5 min, followed by a 30-s water rinse The slide was briefly immersed in a solution of 1% ethanol hydrochloride for 2 s before being washed with water for 1–2 s. Tap water was used to restore the blue color of the slide for approximately 5–10 s, after which the slide was rinsed again with water for an additional 2 s. Eosin staining lasted between 2–3 min, followed by rinsing for 2-min. The slices were subsequently immersed in a series of reagents, including: (i) 70% absolute ethanol (5 s), (ii) 80% absolute ethanol (10 s), (iii) 90% absolute ethanol (30 s), (iv) 95% absolute ethanol (1 min), (v) absolute ethanol I (4 min), (vi) absolute ethanol II (5 min), (vii) xylene I (5 min), and (viii) xylene II (5 min). Finally, the slides were sealed and observed under a microscope.

Tissue sections were stained with Masson

Drying: After examination under a standard microscope, intact frozen sections were meticulously selected and allowed to equilibrate at room temperature for 10 min.

Nucleus staining: Tissue sections were immersed in hematoxylin staining solution for 5 min. Differentiation: Tissue sections were immersed in a solution of ethanol hydrochloride for 3–10 s, followed by thorough rinsing under running water for 10 min. Blue return: Subsequently, the tissue sections were transferred to ammonia water to induce the desired blue coloration and then rinsed again with running water for 10 min. Wash: The tissue sections were rinsed with double-steamed water for 1 min. Ponceau staining: The tissue sections were immersed in a solution containing Ponceau acid fuchsin for 10 min. Rinsing: The tissue sections were sequentially immersed in an aqueous acetic acid solution, followed by an aqueous phosphomolybdic acid solution, and again in an aqueous acetic acid solution for 1 min. Connective tissue staining: Tissue sections were immersed in a solution of aniline blue for 2 min. Rinsing: Subsequently, the tissue sections were immersed in a solution of acetic acid for 1 min. Dehydration and transparency: The tissue sections were sequentially transferred to 95% ethanol (I) for 5 s, followed by exposure to 95% ethanol (II), absolute ethanol (I), and absolute ethanol (II) for 10 s to facilitate dehydration. Xylene (I) and xylene (II) were used consecutively with an immersion time of 2 min each to achieve transparency of the tissue sections. Neutral gum was used as the sealing agent. Observation and statistics: An inverted phase-contrast microscope was used for observation and image capture. Photoshop software was used to quantify the stained and total plaque areas, while GraphPad Prism 6 software was used for analysis and visualization.

RNA fluorescence in situ hybridization (FISH) of tissue sections

The sections were dried for 20 min at room temperature in a fume hood and washed thrice with PBS buffer in an immunohistochemistry chamber. RNA FISH was performed using a digoxigenin-labeled circRNA probe according to the manufacturer's instructions for the FISH kit. After sealing, the slices were analyzed using a confocal microscope.

Immunofluorescence staining of tissue sections

The sections were air-dried for 20 min at room temperature in a fume hood, followed by three washes in a PBS buffer-filled immunohistochemistry box. Subsequently, the sections were immersed in an antigen-retrieval solution within an immunohistochemical box and subjected to antigen retrieval at 95°C for approximately 30 min in a constant-temperature tank. After equilibrating the sections to room temperature and blotting them dry with a filter paper, they were circled using an immunohistochemical pen. Subsequently, the slices were placed in a light-proof humid chamber, and 10 μL of membrane permeabilization solution was added to the circle. The membranes were subsequently permeabilized and blocked at room temperature for 30 min, and the primary antibody was appropriately diluted (1:50) with the immunofluorescent antibody and carefully applied to the immunohistochemical circle. The slides were then placed back into a humid chamber and incubated overnight at 4°C for optimal binding. Subsequently, the sections were washed thrice with a TBST solution for 5 min each to ensure the thorough removal of unbound antibodies. Next, fluorescent secondary antibodies were meticulously incubated in darkness at 37°C for 2 h to facilitate specific labeling. Then, the sections were washed three times with TBST buffer for 5 min each before proceeding to DAPI staining, which was performed for 15 min to visualize cellular nuclei. Finally, after washing three times washes with TBST buffer for 5 min each, the sections were air-dried on filter paper, and carefully sealed and observed under a confocal microscope.

SiRNA and miRNA mimic transfection

The siRNA targeting circRNA were obtained from Guangzhou Ruibo Biotechnology Co., LTD. The transfection procedure was performed as follows: Raw 264.7 macrophage cells were evenly seeded in six-well cell culture plates 1 d before transfection, with a density of approximately 40–50% under microscopic observation to ensure that the cells reached a confluence of 50–60% on the day of transfection. The transfection system was prepared by gently mixing 60 μL of 1× riboFECT™ CP Buffer, 7.5 μL of a solution containing siRNA at a concentration of 20 μM, and 6 μL of riboFECT™ CP Reagent in each well. The mixture was incubated at room temperature for 15 min before the addition of an appropriate volume of Opti-MEM reduced plasma medium, and then gently mixed. The liquid was dispensed evenly into each well, and the cells were cultured in an incubator and subsequent experiments were conducted 24 h following siRNA transfection.

Plasmid transfection

Full-length circRNA sequences were obtained from the circBase and circBank databases. The circRNA overexpression plasmid and corresponding control plasmid were designed and constructed by Beijing Aoke Dingsheng Biotechnology Co., Ltd The transfection procedure was conducted as follows: Raw 264.7 macrophage cells were evenly seeded in six-well cell culture plates 1 d before transfection at a density of approximately 40–50% under microscopic observation to ensure the cells achieve 50–60% confluence on the day of transfection. The transfection system for each well was prepared by combining 3.75 μL of Lipo3000, 5 μg of either the circRNA overexpression or control plasmid, and 200 μL of Opti-MEM reduced plasma medium. After gentle mixing, the mixture was incubated at room temperature for 10 min before the addition of an appropriate amount of Opti-MEM reduced plasma medium. The mixtures were gently mixed and evenly distributed in each well. After incubating for 6-h in the cell culture incubator, the cells were subsequently replaced with a regular complete medium. Subsequent experiments were conducted following further incubation for 24 h.

RNA extraction

Total RNA extraction from tissue: After autoclave sterilization in a glass homogenizer, the tissue was pre-cooled on ice. The vascular tissue was then placed in a glass homogenizer, and 1 mL of total RNA extraction reagent was added. It was fully ground on ice until no visible tissue mass remained and retained on ice for 30 min. Total RNA was extracted from the cells by adding an appropriate amount of total RNA extraction reagent based on cell density, followed by gentle agitation and incubation on ice for 30 min. The aforementioned liquid was transferred to a 1.5 mL EP tube devoid of enzyme, followed by the addition of 200 μL chloroform per mL of total RNA extraction reagent. The EP tube was vigorously shaken for 2 min to ensure complete mixing between the chloroform and total RNA extraction reagent, and then left on ice for 15 min until phase separation occurred. Then, it was centrifuged at 12,000 rpm for 15 min at 4°C. The upper colorless liquid, approximately 350 μL, was aspirated carefully using a 100 μL pipettor gun, then it was transferred to a new enzyme-free EP tube, and an equal volume of isopropanol was added, gently inverted to mix, incubated on ice for 10 min, and centrifuged at 12,000 rpm for 10 min at 4°C. The supernatant was discarded, followed by a wash with 1 mL of absolute ethanol. The mixture was then centrifuged at 12,000 rpm for 5 min in a refrigerated centrifuge at 4°C. The liquid was then discarded, and the EP tube was placed on an ultraclean table for drying. To dissolve the RNA, 10–20 μL of DPEC water was added, and subsequently extracted RNA concentration and purity were determined using Nanodrop2000. The RNA was diluted according to its concentration.

Real-time quantitative PCR (RT-qPCR)

Reversed transcription was performed using the PrimeScript TM RT Master Mix (Takara, Shiga, Japan) to synthesize cDNA. A RT-qPCR assay was conducted using the FastStart Essential DNA Green Master kit (Roche, Shanghai, China). The reference gene used in this study was GAPDH. Relative gene expression was determined using the 2-ddCt method. Sequences of PCR primers are provided in Tables S1 and S2 (Janis Biotechnology Company, Nanning, China).

Western blot analyses

The preparation of protein extracts from cultured cell samples was conducted following established protocols. The protein extracts were separated using 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, followed by western blotting employing rabbit polyclonal anti-SIRPα and -MerTK antibodies (Cell Signaling Technology, Inc., Danvers, MA, USA), as well as rabbit polyclonal anti-SR-B1, -Tim-4, -MFG-E8, and -Gas-6 antibodies (Abcam, Waltham, MA, USA). The chemiluminescent visualization was conducted utilizing the ECL Plus Western Blot Detection System (Amersham Biosciences, Chicago, IL, USA). To ensure accurate comparison, the reference proteins (e.g., β-actin, GAPDH) were detected on the same membrane as the test proteins.

Dual-luciferase reporter experiment

Construction of the dual-luciferase plasmid: The pmirGLO vector, a dual-luciferase reporter gene vector, was custom-designed and synthesized by Beijing Qingke Biotechnology Co., Ltd. Binding sites were predicted using the RNAhybrid, circMIR, and TargetScan databases, and mutant sequences were designed accordingly. Wild-type or mutated circRNA sequences were inserted into the pmirGLO vector to construct dual luciferase reporter plasmids. Two potential binding sites exhibiting a high probability of interaction between miR-495 and the circRNAs were predicted computationally. The specific nucleotide sequences corresponding to these binding sites, including any mutation sites, are provided below. HEK 293T cells were co-transfected with the plasmid along with either miR-495 mimic or control miR-NC. The experimental procedure has been previously described.

Luciferase activity detection: The Duo-Lite Luciferase Assay kit was allowed to equilibrate to room temperature, the cell medium was discarded, and the cells were washed twice with PBS. Subsequently, 300 μL of 1× cell lysis buffer was added to lyse the cells for 5 min, after which they were transferred to a new EP tube after thorough aspiration. The EP tube was then centrifuged at room temperature at 12,000 rpm for 2 min, and the supernatant was carefully decanted. For each well, 20 μL of supernatant and 100 μL of luciferase substrate were added, and firefly luciferase reporter gene activity was promptly detected using a microplate reader after gentle agitation. Subsequently, 100 μL of Renilla substrate working solution was swiftly added to each well, followed by thorough mixing and immediate detection of Renilla luciferase reporter gene activity using a microplate reader. Luciferase activity was calculated and analyzed using the following formula: luciferase activity = (experimental group firefly luciferase/experimental group Renilla luciferase)/(control group firefly luciferase/control group Renilla luciferase).

Cell RNA FISH experiment

Raw 264.7 cells were evenly distributed in confocal dishes, followed by RNA fluorescence in situ hybridization of circRNAs using a digoxigenin-labelled probe and the TSA chromogenic method developed by the Servicebio Company for detection.

RNA sequencing and bioinformatics analysis

RNA Extraction and Quality Control:Total RNA was extracted from human carotid atherosclerotic plaque tissues (three Vulnerable Plaque [VP] and three Stable Plaque [SP] samples) using TRIzol reagent according to the manufacturer’s instructions. Residual genomic DNA was removed by DNase I digestion. RNA purity was assessed by A260/A280 ratio using a NanoDrop One^c spectrophotometer (Thermo Fisher Scientific), and RNA integrity was verified by 1.5% agarose gel electrophoresis. RNA concentration was quantified using a Qubit 3.0 fluorometer with the Qubit RNA Broad Range Assay kit . Library Construction and Sequencing :Two micrograms of total RNA per sample were subjected to ribosomal RNA depletion using the Ribo-off rRNA Depletion Kit. Strand-specific cDNA libraries were constructed using the KC-Digital™ Stranded mRNA Library Prep Kit for Illumina . This kit incorporates unique molecular identifiers (UMIs, 8 random bases) into pre-amplified cDNA molecules, enabling the elimination of PCR and sequencing duplication artifacts. Library fragments ranging from 200 to 500 bp were size-selected, quantified, and sequenced on a DNBSEQ-T7 sequencer (MGI Tech Co., Ltd.) with paired-end 150 bp (PE150) reads. Read Preprocessing and UMI Deduplication :Raw sequencing reads were processed using Trimmomatic (version 0.36) to remove adapter sequences and low-quality bases. Clean reads were subsequently deduplicated using an in-house UMI-based pipeline. Reads sharing identical UMI sequences were clustered, and within each cluster, pairwise alignment was performed. Reads with ≥95% sequence identity were grouped into sub-clusters, from which a consensus sequence was derived via multiple sequence alignment. This procedure effectively eliminated errors and biases introduced by PCR amplification or sequencing. Read Alignment and Quantification :Deduplicated reads were aligned to the human reference genome (GRCh38/hg38) using STAR with default parameters. Gene-level read counts were obtained using featureCounts . Expression levels were normalized as reads per kilobase per million mapped reads (RPKM). Differential Expression Analysis :Differential expression analysis between VP and SP groups was performed using the edgeR package in R. Genes with a |log₂(Fold Change)| ≥ 1 and a false discovery rate (FDR)-adjusted p-value < 0.05 were considered significantly differentially expressed. CircRNA Identification and Annotation :CircRNA candidates were identified from the RNA-seq data using CIRCexplorer2 . Back-spliced junction reads were detected from STAR-aligned BAM files. Only circRNAs supported by at least two unique back-spliced reads and present in at least two samples per group were retained for downstream analysis. Identified circRNAs were annotated against the circBase database (https://www.circbase.org/).

In vitro efferocytosis assay

In vitro efferocytosis assay was performed as previously described with modifications. Briefly, cells were induced to undergo apoptosis by ultraviolet irradiation (254 nm) and subsequently labeled with CFSE (Thermo Scientific). Monotropein-treated or control macrophages were co-cultured with the labeled apoptotic cells at a ratio of 1:5 for 4 hours. Following co-culture, unengulfed cells were removed by washing with PBS. The macrophages were collected and stained with CD68-CoralLite. Flow cytometry was then performed to determine the efferocytosis rate of macrophages by quantifying the percentage of CFSE-positive macrophages.

Quantification and statistical analysis

The above experiments were independently replicated at least three times. Data were analyzed and graphically represented using Graphpad Prism 9.0 professional software. Student's T test was used to compare the data between two groups, whereas one-way ANOVA or two-way ANOVA was used to compare multiple groups of data. Statistical results were presented as Mean±standard error (Mean±SEM), with significance indicated by p <0.05 (∗), p <0.01 (∗∗), p <0.001 (∗∗∗), p <0.05 (#), and p <0.01 (##),p <0.001 (###).