Opposing functions of miR-155-5p and Socs1 drive vascular inflammation in diabetes-accelerated atherosclerosis

Abstract

Background

Diabetes accelerates atherosclerosis by driving persistent vascular inflammation. MicroRNA-155 (miR-155) is a post-transcriptional regulator of inflammatory genes, while suppressor of cytokine signaling 1 (Socs1) limits Janus kinase (JAK)/signal transducer and activator of transcription (STAT)-mediated cytokine responses. We explored how the imbalance between miR-155-5p and Socs1 contributes to atherosclerotic plaque progression in diabetes.

Methods

Apolipoprotein E knockout (ApoE-/-) mice were studied in two settings: age-dependent atherosclerosis progression under non-diabetic conditions, and streptozotocin-induced diabetes to model accelerated atherosclerosis. Diabetic mice received a miR-155-5p inhibitor, a Socs1-expressing adenovirus, or respective controls. Lesion size, composition, and gene expression were analyzed. Cultured vascular smooth muscle cells (VSMCs) and macrophages were transfected with miR-155-5p mimic/inhibitor and Socs1 siRNA/plasmid to assess inflammatory responses, phenotypes, and efferocytosis under diabetic-like conditions.

Results

During atherosclerosis progression, vascular miR-155-5p inversely correlated with Socs1 and positively with lesion size, while Socs1 correlated negatively with plaque burden. In diabetic mice, miR-155-5p inhibition reduced lesion area, lipid/collagen and macrophage/VSMC ratios, pro-inflammatory cytokines, M1 macrophages and synthetic VSMC markers, while increasing Socs1, M2 and contractile VSMC genes. Socs1 gene transfer reproduced these effects by reducing miR-155-5p and Stat1 expression, and lesion size. In vitro, miR-155-5p mimic suppressed Socs1, activated STAT1 and inflammatory phenotypes in macrophages and VSMCs, whereas miR-155-5p inhibition had opposite effects. Socs1 silencing amplified inflammation, and its overexpression counteracted miR-155-5p actions. Moreover, miR-155-5p inhibition reduced soluble Mer receptor tyrosine kinase (MerTK) in plaques and macrophages, indicating improved efferocytosis, whereas the mimic promoted macrophage MerTK shedding and impaired apoptotic cell clearance.

Conclusion

Reciprocal regulation between miR-155-5p and Socs1 influences vascular inflammation, phenotypic changes, and defective efferocytosis in a diabetic context. Targeting this axis may restore resolution mechanisms and enhance plaque stability in diabetes-associated vascular disease.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12933-026-03121-3.

Research insights

What is currently known about this topic?

Diabetes accelerates atherosclerosis via chronic vascular inflammation. MiR-155 promotes inflammatory signaling and macrophage activation. Socs1 restrains JAK/STAT-mediated cytokine responses.

What is the key research question?

Does dysregulation of the miR-155-5p/Socs1 axis drive plaque progression in diabetes?

What is new?

MiR-155-5p and Socs1 show an inverse expression pattern linked to atherosclerosis progression. MiR-155-5p inhibition and Socs1 induction reduce vascular inflammation, modulate vascular cell phenotypes, and stabilize plaques in diabetic mice. MiR-155-5p impairs efferocytosis via MerTK cleavage in macrophages.

How might this study influence clinical practice?

Modulation of the miR-155-5p/Socs1 axis may help limit vascular inflammation and atherosclerosis in diabetes.

Background

Diabetes mellitus is a chronic metabolic disorder that accelerates the development of atherosclerosis, resulting in earlier and more severe cardiovascular complications compared to non-diabetic individuals [1, 2]. Under diabetic conditions, the convergence of hyperglycemia, lipid imbalance, persistent inflammation and oxidative stress drives endothelial dysfunction, vascular inflammation, and maladaptive remodeling, thus leading to larger, more inflamed, and unstable atheroma plaques [3, 4].

Atherosclerosis progresses not only because of chronic vascular inflammation, but also due to defective resolution processes. Within the plaque, macrophage recruitment and activation, together with vascular smooth muscle cell (VSMC) phenotypic switching, sustain the production of chemokines and proinflammatory cytokines [5, 6]. In parallel, efferocytosis, the mechanism by which phagocytes engulf apoptotic cells, becomes defective in advanced plaques [7]. Ineffective clearance of apoptotic cells leads to secondary necrosis, amplifies inflammation, and contributes to necrotic core expansion and plaque instability [8, 9].

MicroRNAs (miRNAs) are short, single-stranded, non-coding RNAs that act as key post-transcriptional regulators of cell biology and inflammatory processes, including cardiovascular diseases [10–12]. Among them, miR-155 is strongly induced by inflammatory stimuli such as interferon-γ (IFNγ) and lipopolysaccharide (LPS), and it orchestrates multiple aspects of immune and vascular cell activation [13]. The miR-155 precursor produces two mature strands (miR-155-3p and miR-155-5p), the latter being the predominant and functionally active form [10, 14, 15]. In cardiovascular diseases, miR-155 modulates signaling pathways that control macrophage polarization, cytokine production, and VSMC phenotypic switching [16, 17]. In line with its pro-inflammatory role, therapeutic miR-155 inhibition has shown beneficial effects in several in vivo models of inflammatory injury, including ischemic stroke and neuroinflammatory disorders [18–20]. However, the relevance of miR-155 modulation in diabetes-accelerated atherosclerosis remains unclear.

MiR-155 regulates multiple target genes, and its expression is controlled by key inflammatory pathways such as nuclear factor-κB, mitogen-activated protein kinases, Toll-like receptor (TLR) and signal transducer and activator of transcription (STAT) signaling [21, 22]. Among its direct targets, suppressor of cytokine signaling 1 (Socs1) plays a central role in restraining cytokine-driven responses acting as a negative regulator of Janus kinase (JAK)/STAT signaling and limiting inflammatory activation [23, 24]. Reciprocal regulation between miR-155 and Socs1 has been reported in immune and inflammatory contexts [10, 25–27], and more recently in diabetic nephropathy [28]. Whether this regulatory mechanism operates in diabetic vascular injury and contributes to defective resolution remains unknown. Here, we investigated the role of the miR-155-5p/Socs1 axis in diabetes-associated atherosclerosis, combining in vivo and in vitro models to assess its impact on plaque progression, stability, and inflammatory-resolution pathways.

Material and methods

Experimental models and treatments

Animal procedures were performed according to the European Directive 2010/63/EU and approved by the Institutional Animal Care and Use Committee of IIS-FJD and Comunidad de Madrid (PROEX 217/19 and 128.4–23). Male Apolipoprotein-E deficient mice (ApoE-/-; Jackson Laboratory) were used in two experimental settings: (i) a non-diabetic, age-dependent model of atherosclerosis development (3 months, n = 5; 6 months, n = 6; 9 months, n = 4; 12 months, n = 4; 17 months, n = 5); and (ii) an intervention model of diabetes-accelerated atherosclerosis induced at 10–12 weeks of age (n = 46). Diabetes was induced through two consecutive intraperitoneal injections of streptozotocin (125 mg/kg in 10 mM citrate buffer; S0130, Sigma-Aldrich), to generate a model of rapid lesion progression under hyperglycemic and hyperlipidemic conditions [29, 30]. Body weight and non-fasting blood glucose (tail vein sampling; NovaPro glucometer; Nova Biomedical Iberia) were monitored weekly.

Two weeks after induction, diabetic mice (blood glucose ≥ 19.4 mmol/L) were randomized into three groups: vehicle-treated control (Ctrl, saline, n = 8), mmu-miR-155-5p inhibitor (155-inh, 2 mg/kg, n = 7; miRCURY Locked Nucleic Acid (LNA) Power Inhibitor YCI0201878, Qiagen), and mmu-miR-155-5p negative control (NC, 2 mg/kg, n = 7; YI00199006-011-ADB, Qiagen) [28]. Treatments were injected intraperitoneally twice weekly for 6 weeks. Non-diabetic mice were included as baseline reference controls (n = 5). Aortic miR-155-5p expression was analyzed to confirm inhibitor efficacy.

In a separate cohort, diabetic mice received a single intravenous tail-vein injection of adenovirus expressing Socs1 (Ad-Socs1, n = 8; 1 × 10⁹ viral particles/g), a matched empty vector lacking the Socs1 insert (Ad-null, n = 6), or remained untreated (Ctrl, n = 6). The generation, characterization, and in vivo performance of this adenovirus system have been described previously [31, 32]. Aortic Socs1 and Stat1 levels were analyzed to confirm transgene expression and pathway modulation.

Mice were housed in ventilated cages under standard conditions in a temperature-controlled facility (20–22 ºC; 12 h light/dark cycle) with ad libitum access to chow diet and water. During follow-up, all STZ-treated mice developed severe hyperglycemia (blood glucose > 28.9 mmol/L) and received intermittent low-dose insulin (0.4 IU) to prevent excessive weight loss and mortality, without normalizing glycemia. At the end of the study, fasted mice were anesthetized (100 mg/kg ketamine and 15 mg/kg xylazine), perfused with saline via left ventricle, and euthanized. Blood samples were collected, allowed to clot, and centrifuged to obtain serum for determination of total cholesterol (STA-384; Cell Biolabs; 1–5 μL), triglycerides (11,528; Biosystems; 1–3 μL), aspartate aminotransferase (AST), and alanine aminotransferase (ALT) (Reflotron, Roche; 30 μL), according to the manufacturers’ instructions. Aortas were carefully dissected free of surrounding adipose tissue; the aortic root was processed for histology and the thoraco-abdominal aorta was snap-frozen for gene expression analysis.

Histological and immunohistochemical analysis

For histological analysis, serial 7-μm aortic root cross-sections were collected covering ≈1000 μm from valve leaflets. Oil-red-O (ORO)/hematoxylin staining was performed on alternate serial sections, while intervening sections were reserved for parallel histological analyses at comparable lesion levels. Atherosclerotic lesion size and neutral lipid content were quantified using image analysis software across serial sections along the aortic root. For each mouse, mean lesion area was calculated from sections containing the maximal lesion, and lipid content was expressed as the percentage of ORO-positive area relative to total lesion area [30, 32]. Collagen content was assessed in parallel sections by Sirius Red staining under polarized light [29]. In the non-diabetic model, relative plaque burden was assessed by normalizing lesion area to the total vessel cross-sectional area delineated by the external elastic lamina and expressed as a percentage.

For immunohistochemistry, sections were fixed in cold acetone (10 min), blocked for endogenous peroxidase (3% H₂O₂ in methanol, 30 min) and non-specific binding (6% host serum, 1 h), and incubated overnight at 4 ºC with primary antibodies against CD68 (1:400 dilution; Abcam, Cat# ab125212, RRID: AB_869007) and Mer receptor tyrosine kinase (soluble form, sMerTK; 1:50 dilution; R&D Systems Cat# AF591, RRID: AB_2098565). Detection was performed with biotinylated secondary antibodies (1:200 dilution; Jackson Immunoresearch), avidin–biotin complex reagent (PK-4000; Vector Laboratories), and chromogen substrates (3,3′-diaminobenzidine, ab64238, Abcam; or 3-amino-9-ethylcarbazole, K3461, Dako, Agilent Technologies), followed by hematoxylin counterstaining. Macrophage phenotypes were detected either by indirect immunofluorescence (M1 marker arginase (Arg) 2, Santa Cruz Biotechnology Cat# sc-20151, RRID:AB_2059089; M2 marker Arg1, Santa Cruz Biotechnology Cat# sc-18354, RRID:AB_2227469; 1:100 dilution) using Alexa Fluor 488/568 secondary antibodies (anti-rabbit, Thermo Fisher Scientific Cat# A-11011, RRID:AB_143157; anti-goat, Thermo Fisher Scientific Cat# A-11055, RRID:AB_2534102) [33] or by direct immunofluorescence (Arg1, Santa Cruz Biotechnology Cat# sc-393496 AF488, RRID:AB_2890065; Arg2, Santa Cruz Biotechnology Cat# sc-271430 AF546, RRID:AB_10648473; 1:50 dilution). VSMC content was assessed using a Cy3-conjugated antibody against α-smooth muscle actin (α-SMA; clone 1A4; 1:1000 dilution; Sigma-Aldrich, Cat# C6198, RRID: AB_476856) [30]. Samples were mounted with VectaShield Vibrance Antifade Medium containing DAPI (H-1800; Vector Laboratories). All histological analyses were performed in a blinded manner. Images were acquired with an Axioscope microscope (Carl Zeiss) under identical exposure. Positive staining was quantified within the intimal lesion of complete aortic root cross-section in at least two sections per mouse using Image-Pro Plus (Media Cybernetics) and expressed as percentage of positive area relative to total plaque area.

Cell cultures and transfections

Mouse aortic VSMCs (MOVAS cell line; ATCC-CRL-2797) were cultured in DMEM medium (D6546) containing 10% fetal bovine serum (FBS; F7524), 2% L-glutamine (G7513), 1% penicillin/streptomycin (P0781), and 0.4% G418 antibiotic (G8168), all from Sigma-Aldrich [34]. Mouse macrophages (RAW 264.7 cell line; TIB-71) were grown in DMEM medium supplemented with 10% FBS, 1% L-glutamine, and 1% penicillin/streptomycin. T cells (human Jurkat cell line; TIB-152) were cultured in RPMI-1640 medium (R0883; Sigma-Aldrich) supplemented with 10% FBS, 1% L-glutamine, and 1% penicillin/streptomycin. Cells were serum-deprived overnight (0–0.5% FBS) and stimulated with IFNγ (315–05, PeproTech) plus LPS (L2630, Sigma-Aldrich), using higher doses for short-term stimulation (200 ng/mL IFNγ plus 1 μg/mL LPS) and lower doses for prolonged stimulation (24 h; 20 ng/mL IFNγ plus 100 ng/mL LPS) to sustain inflammatory activation while preserving cell viability [28, 35]. For high-glucose conditions, cells were maintained for 24 h in low-glucose medium (D6046) and then exposed to 30 mM D-glucose (G7528, Sigma-Aldrich).

To modulate miR-155-5p expression, cells were transfected for 18–24 h with mmu-miR-155-5p mimic (YM00470919, 10 nM), inhibitor (YCI0201878, 50 nM), or their respective negative controls (YM00479902-ABG and YI00199006-011-ADB; all from Qiagen) using Lipofectamine RNAiMAX (13778; Thermo Fisher Scientific) in OptiMEM medium (31985-070; Thermo Fisher Scientific) [28]. To modulate Socs1 expression, cells were transfected for 18–24 h with specific siRNA (siSocs1, 20 nM; 4390771, Ambion) or scramble negative control (siScr; 4390843) using Lipofectamine RNAiMAX, and with Socs1 overexpression plasmid (pSocs1, 2 μg) or empty vector (pNull) using lipofectamine 3000 (L3000, Thermo Fisher Scientific), as previously described [31, 32, 34]. Mock-transfected cells (transfection reagents only) were used as additional controls. Transfection efficiency was verified by qPCR of the corresponding target. After transfection, cells were either left untreated or exposed to inflammatory and hyperglycemic conditions.

Cell viability assay

Cell viability was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolum bromide tetrazolium assay (CyQUANT MTT Cell Proliferation Assay Kit; V13154, Invitrogen) [31]. Briefly, cells were transfected on 96-well plates (1 × 10⁴ cells/well) and cultured in medium containing 10% FBS. Non-transfected cells maintained in 20% FBS were used as positive viability control. Tetrazolium stock solution was added and incubated for 2–5 h, after which dimethyl sulfoxide was added to dissolve the formazan crystals. Absorbance was measured at 570 nm in quadruplicate, and cell viability was expressed as percentage of basal conditions.

Quantitative real-time PCR (qPCR) analysis

Total mRNA from aortas and cultured cells was extracted using TRI reagent (Thermo Fisher Scientific). Complementary DNA was synthetized using the High-Capacity cDNA reverse transcription kit (4368813, Applied Biosystems) and then analyzed by real-time qPCR using Premix Ex Taq (RR390; Takara). TaqMan mouse gene expression assays (Thermo Fisher Scientific; Table 1) were used, with 18S rRNA as endogenous control for normalization. For miRNA expression, cDNA was obtained using miRCURY LNA RT kit (339340) and then analyzed by qPCR using miRCURY LNA SYBR Green PCR kit (339345), miR155-5p-specific assay (YP02119303), and RNU6B (YP00203907) as the normalization control (all from Qiagen). qPCR data were normalized to the corresponding housekeeping gene and expressed as relative expression (arbitrary units) or as fold change relative to basal or control condition, as specified in each experiment.

Table 1.

TaqMan mouse gene expression assays used for qPCR analysis

Full name	Gene	Primer code
Actin alpha 2, smooth muscle	Acta2	Mm01546133_m1
Arginase 1	Arg1	Mm00475988_m1
Arginase 2	Arg2	Mm00477592_m1
CD68 antigen	Cd68	Mm03047343_m1
Chemokine (C–C motif) ligand 2	Ccl2	Mm00441242_m1
Chemokine (C–C motif) ligand 5	Ccl5	Mm01302428_m1
C-X-C motif chemokine ligand 10	Cxcl10	Mm00445235_m1
Eukaryotic 18S rRNA	18s	4310893E
Interleukin 1-beta	Il1b	Mm00434228_m1
Interleukin 10	Il10	Mm00439614_m1
Krüppel-like factor 4 (KLF4)	Klf4	Mm00516104_m1
Mannose receptor C type 1	Mrc1	Mm01329359_m1
Matrix metalloproteinase 2	Mmp2	Mm00439498_m1
Matrix metalloproteinase 9	Mmp9	Mm00442991_m1
Nitric oxide synthase 2, inducible	Nos2	Mm00440502_m1
Nitric oxide synthase 2, inducible	Nos2	Mm00440502_m1
Signal transducer and activator of transcription 1	Stat1	Mm00726417_s1
Suppressor of cytokine signaling 1	Socs1	Mm00782550_s1
Transgelin	Tagln	Mm00441660_m1
Tumor necrosis factor-alpha	Tnfa	Mm00443258_m1

Protein expression analysis

Cells were lysed in ice-cold buffer (10 mM Tris–HCl pH 7.4, 150 mM NaCl, 1% Triton X-100, 0.5% NP-40, 1 mM EDTA, 1 mM EGTA, 0.2 mM Na₃VO₄, 10 mM NaF, 0.2 mM PMSF, and protease inhibitors) [36]. Proteins (35 μg) were resolved on SDS-PAGE and immunoblotted for Socs1 (1:100 dilution, Thermo Fisher Scientific Cat# 38–5200, RRID:AB_2533372), phosphorylated STAT1 (P-STAT1, 1:500 dilution; Thermo Fisher Scientific Cat# 33-3400, RRID:AB_2533113), and β-actin (1:5000 dilution, Sigma-Aldrich Cat# A1978, RRID:AB_476692). Macrophage supernatants were concentrated with Amicon Ultra centrifugal filters (UFC503096; Sigma Aldrich) and immunoblotted for sMerTK (1:1000 dilution; RD Systems Cat# AF591, RRID: AB_2098565). Bands were quantified using ImageJ software, normalized to β-actin (Socs1 and P-STAT1) or Ponceau Red (sMerTK) and expressed as fold change relative to control conditions. In addition, sMerTK concentrations in macrophage supernatants were quantified by ELISA (R&D Systems, Cat# 300021181) following the manufacturer’s instructions.

Efferocytosis assay

An in vitro efferocytosis assay using apoptotic Jurkat T cells as targets was performed to assess macrophage efferocytic capacity [20, 37]. Macrophages were seeded in 8-well Lab-Tek II Chambers (154534PK, Thermo Fisher Scientific; 2 × 10⁴ cells/well) without additional coating, 24 h before the assay. T cells (2 × 10⁶ cells/mL) were rendered apoptotic by treatment with 1 μM staurosporine (62996-74-1; Thermo Fisher Scientific) for 90 min, and apoptosis was confirmed by flow cytometry using the Annexin V Detection Kit I (559763; BD Pharmingen). Apoptotic cells (10⁶ cells/mL) were labeled with 160 ng/mL pHrodo Red succinimidyl ester (P36600, Invitrogen; 30 min, RT), and macrophages with 0.4 μM CellTracker Green CMFDA (C705, Invitrogen; 30 min, 37 ºC). Apoptotic cells were then co-incubated with macrophages at a 5:1 cell ratio for 1 h to allow efferocytosis. Then, cells were washed with PBS, fixed in 4% paraformaldehyde, and mounted with antifade medium containing DAPI for fluorescence microscopy. Ten fields of view were analyzed per well (duplicate or triplicate wells per condition) using ImageJ. The efferocytotic index was calculated as the percentage of macrophages containing ingested apoptotic cells relative to the total number of macrophages per field.

Statistics

Results are presented as mean ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism v8 (GraphPad Software Inc). Data normality was assessed using the Shapiro–Wilk and Kolmogorov–Smirnov tests. Parametric tests (unpaired two-tailed Student’s test or one- or two-way ANOVA followed by Tukey’s post hoc test) were applied to normally distributed data, whereas non-parametric tests (Mann–Whitney or Kruskal–Wallis test) were used when normality was not met. Correlation analyses were performed using Pearson’s coefficient based on individual animal values. A p-value < 0.05 was considered statistically significant.

Results

miR-155-5p and Socs1 display reciprocal expression dynamics during atherosclerosis development

In non-diabetic ApoE-/- mice fed a standard chow diet (aged 3–17 months), aortic miR-155-5p expression increased in parallel with lesion development (Fig. 1A) and correlated positively with maximal lesion area (Fig. 1B). Conversely, Socs1 mRNA declined over time (Fig. 1C) and correlated inversely with both miR-155-5p (Fig. 1D) and lesion size (Fig. 1E). Relative plaque burden, normalized to vessel area, showed a similar age-dependent pattern and correlated positively with miR-155-5p levels and inversely with Socs1 mRNA expression (Supplementary Fig. S1).

Fig. 1.

Reciprocal association of miR-155-5p and Socs1 during atherosclerosis development and in cultured cells. A Temporal evolution of the aortic expression of miR-155-5p in non-diabetic ApoE-/- mice: 3 months, n = 5; 6 months, n = 6; 9 months, n = 4; 12 months, n = 4; 17 months, n = 5. B Pearson correlation analysis between miR-155-5p and atheroma lesion size. C Aortic Socs1 mRNA expression over time. D–E Correlation of Socs1 expression with miR-155-5p levels (D) and lesion size (E). F–H In vitro expression of miR-155-5p and Socs1 in murine VSMCs (n = 4 experiments) (F–G) and macrophages (n = 4–5 experiments) (H) after 1–24 h of stimulation with inflammatory (IFNγ 20 ng/mL plus LPS 100 ng/mL) or high-glucose (HG, 30 mM D-glucose) conditions. qPCR values were normalized to RNU6B or 18S rRNA, expressed as arbitrary units (a.u.) (A–E) or fold change relative to basal conditions (F–H), and presented as individual values with mean ± SD. *P < 0.05, **P < 0.01 versus the indicated time points

In cultured VSMCs and macrophages, miR-155-5p was upregulated from 1 to 24 h under inflammatory (IFNγ 20 ng/mL plus LPS 100 ng/mL; Fig. 1, F and H) or high-glucose conditions (30 mM glucose; Fig. 1, G and H). In contrast, Socs1 displayed a transient induction, peaking at 3 h under inflammatory stimulation and at 12 h under high-glucose conditions in VSMCs, with a similar early and transient response in macrophages.

Inhibition of miR-155-5p upregulates Socs1, attenuates atherosclerosis and enhances plaque stability in diabetic mice

To evaluate whether blocking miR-155-5p activity modulates plaque progression and composition under diabetic conditions, streptozotocin-induced diabetes was established in ApoE-/- mice at 10–12 weeks of age, when early atherosclerotic lesions are present. Treatment with a miR-155-5p inhibitor was initiated two weeks later to target diabetes-accelerated plaque progression. As expected, non-diabetic ApoE-/- mice included as baseline reference displayed lower lesion size than diabetic mice, confirming diabetes-accelerated atherogenesis (Fig. 2, A–C). Diabetic mice treated with miR-155-5p inhibitor showed significantly reduced lesion extension and maximal plaque area compared with vehicle and negative controls (55 ± 18% and 52 ± 20% reduction, respectively; Fig. 2, A–C). Body weight, blood glucose, serum lipids, and transaminases tended to be lower in inhibitor-treated mice but did not reach statistical significance (Table 2), indicating that vascular effects were not secondary to systemic metabolic changes or liver injury.

Fig. 2.

In vivo inhibition of miR-155-5p ameliorates atherosclerosis in diabetic mice. A Representative images at the level of maximal lesion (scale bars, 100 μm) of ORO/hematoxylin staining in aortic root cross-sections from non-diabetic ApoE-/- mice (ND, n = 5) and diabetic ApoE-/- mice treated with vehicle (Ctrl, n = 8), miR-155-5p inhibitor (155-inh, n = 7) or negative control (NC, n = 7). Lower panels show higher-magnification views of the boxed areas (l, lumen; m, media; a, atheroma). B Quantification of lesion area along serial aortic root sections from the valve leaflets. C Maximal lesion area per animal, calculated from sections containing the largest plaque. D Lipid content within atheroma plaques of diabetic groups. E Representative Sirius red-stained aortic sections from diabetic mice under polarized light (scale bars, 100 μm; dashed lines indicate atheroma area used for quantification). F Quantification of collagen content in atheroma plaques. G Lipid-to-collagen ratio. Data are presented as individual values with mean ± SD of the total number of animals per group. *P < 0.05, **P < 0.01, ***P < 0.001 vs. ND, Ctrl or NC groups

Table 2.

Metabolic parameters in diabetic mouse models with miR-155-5p or Socs1 intervention

	BW (g)	BG (mmol/L)	Chol (mmol/L)	TG (mmol/L)	AST (U/L)	ALT (U/L)
Model 1 groups
ND	25.0 ± 0.8 (24.4–25.7)	10.7 ± 0.7 (10.1–11.3)	8.7 ± 1.3 (7.6–9.8)	1.4 ± 0.1 (1.3–1.5)	129.6 ± 15.3 (115.0–147.0)	56.6 ± 6.7 (50.5–63.0)
Ctrl	22.3 ± 1.6a (21.4–23.7)	30.3 ± 2.9c (27.7–32.8)	12.5 ± 2.3c (10.6–12.8)	1.25 ± 0.16c (1.07–1.40)	163.9 ± 36.0 (128.3–194)	92.9 ± 27.5b (70.5–117.5)
NC	22.5 ± 1.6a (20.6–24.1)	29.2 ± 3.4c (27.3–32.7)	12.3 ± 2.3c (9.5–13.7)	1.37 ± 0.24c (1.15–1.57)	171.9 ± 33.3 (139.0–206.0)	87.9 ± 16.0a (75.0–107.0)
155-inh	20.4 ± 1.5c (19.1–21.7)	28.4 ± 2.6c (26.2–31.0)	10.7 ± 1.0c (10.4–11.3)	1.11 ± 0.17c (0.93–1.25)	143.3 ± 30.5 (119.0–167.0)	78.1 ± 9.0 (72.0–83.0)
Model 2 groups
Ctrl	22.7 ± 2.3 (21.3–25.0)	27.8 ± 4.0 (23.9–31.4)	11.9 ± 2.3 (10.1–14.0)	1.13 ± 0.33 (0.87–1.36)	135.5 ± 33.0 (118.0–152.3)	86.8 ± 19.1 (69.5–102.3)
Ad-null	23.0 ± 2.1 (21.0–25.3)	26.8 ± 2.6 (25.6–28.2)	11.7 ± 2.4 (9.9–13.3)	1.21 ± 0.36 (0.92–1.48)	146.5 ± 68.8 (96.8–221.3)	92.2 ± 17.5 (78.3–109.0)
Ad-Socs1	24.9 ± 2.4 (24.3–26.5)	26.8 ± 2.6 (25.6–28.2)	10.5 ± 1.6 (9.3–11.1)	0.98 ± 0.34 (0.75–1.16)	129.9 ± 16.6 (120.0–143.3)	81.0 ± 13.6 (69.8–94.8)

Two independent experimental models are shown: 1) Non-diabetic mice (ND, n = 5) and diabetic mice treated with vehicle (Ctrl, n = 8), miR-155-5p negative control (NC, n = 7) or miR-155-5p inhibitor (155-inh, n = 7); 2) Diabetic mice untreated (Ctrl, n = 6), receiving empty vector (Ad-null, n = 6) or Socs1 adenovirus (Ad-Socs1, n = 8). Body weight (BW), blood glucose (BG), serum cholesterol (Chol), triglycerides (TG) and transaminases (AST, ALT) were determined at endpoint. Values are mean ± SD, with interquartile range (25th-75th percentile) in parentheses. ^aP < 0.05, ^bP < 0.01, ^cP < 0.001 versus ND

Histological analysis in diabetic mouse aortas revealed decreased lipid deposition (Fig. 2D) and increased collagen content in miR-155-5p inhibitor-treated mice (Fig. 2, E and F), leading to a lower lipid-to-collagen ratio (64 ± 22% reduction vs. control; Fig. 2G), consistent with enhanced plaque stability. Immunohistochemistry further demonstrated reduced CD68-positive cell content (Fig. 3, A and B) and increased neointimal VSMCs (α-SMA; Fig. 3, A and C) in plaques from inhibitor-treated mice, resulting in a decreased macrophage-to-VSMC ratio (72 ± 10% reduction vs. control; Fig. 3D). To gain further insight into the mechanisms involved, we quantified by qPCR the expression of inflammatory mediators and phenotypic makers of macrophages and VSMCs. Treatment with miR-155-5p inhibitor significantly reduced aortic miR-155-5p levels and increased Socs1 mRNA expression (Fig. 3E), with miR-155-5p correlating inversely with Socs1 (r = -0.6904, p = 0.0004) and positively with lesion area (r = 0.5966, p = 0.0034) (Supplementary Fig. S2A, B). This was accompanied by reduced expression of pro-inflammatory cytokines (Ccl2, Tnfa), the M1 macrophage marker Arg2, and the synthetic VSMC marker Klf4, together with upregulation of markers of anti-inflammatory M2 macrophages (Arg1) and contractile VSMCs (Acta2) (Fig. 3E). Consistently, immunofluorescence analysis revealed a shift in aortic macrophage polarization toward an M2 phenotype, as reflected by a lower Arg2/Arg1 ratio in inhibitor-treated mice (Fig. 3F). Importantly, negative control-treated mice showed parameters comparable to diabetic controls, confirming the lack of biological activity of the negative control. These results indicate that inhibition of miR-155-5p restores Socs1 expression, limits plaque growth and shifts macrophage and VSMC phenotypes toward a more stable, less inflammatory state.

Fig. 3.

miR-155-5p inhibition modulates vascular inflammation and cell phenotypes in vivo. A Representative images (scale bars, 50 μm; l, lumen; m, media; a, atheroma) of the immunodetection of macrophages (CD68 immunoperoxidase) and VSMCs (α-SMA immunofluorescence; dashed lines indicate lesion areas) in aortic root sections from diabetic mice treated with vehicle (Ctrl, n = 8), miR-155-5p negative control (NC, n = 7) or miR-155-5p inhibitor (155-inh, n = 7). Quantification of CD68 (B) and α-SMA (C) staining in plaques. D Macrophage-to-VSMC (CD68/α-SMA) ratio in plaques. E qPCR analysis of miR-155-5p, Socs1, inflammatory cytokines (Ccl2, Tnfa), macrophage polarization markers (Arg2 for M1, Arg1 for M2), and VSMC phenotype markers (Klf4 for synthetic, Acta2 for contractile) in mouse aorta. Values were normalized to 18S rRNA or RNU6B and expressed as arbitrary units (a.u.). F Immunofluorescence detection of macrophage markers (Arg2, M1 in green; Arg1, M2 in red; nuclear staining in blue) and quantification of M1/M2 ratio in atherosclerotic lesions. Data are shown as individual values with mean ± SD of the total number of animals per group. *P < 0.05, **P < 0.01, ***P < 0.001 vs. Ctrl or NC groups

Socs1 induction suppresses miR-155-5p and mitigates diabetic atherosclerosis

As a complementary approach, Socs1 expression was induced in diabetic ApoE-/- mice using a recombinant adenovirus, with an empty vector and untreated animals as control groups. Effective gene delivery and pathway modulation were confirmed by increased aortic Socs1 and reduced Stat1 expression (Fig. 4A), consistent with our previous observation in non-diabetic atherosclerotic models [32]. Importantly, aortic Socs1 levels were associated with reduced miR-155-5p, showing a significant inverse correlation between both parameters (Fig. 4, A and B). Consistently, Socs1 gene transfer significantly decreased atherosclerotic lesion size (47 ± 15% reduction vs. control and 52 ± 13% reduction vs. empty vector; Fig. 4C), without affecting metabolic parameters or transaminase levels (Table 2). Lesion size correlated negatively with Socs1 expression (r = − 0.6160, p = 0.0038) and positively with miR-155-5p (r = 0.6273, p = 0.0031) (Supplementary Fig. S2C–D). At the molecular level (Fig. 4D), Socs1 overexpression dampened inflammatory signaling by reducing cytokines (Ccl2, Tnfa), the macrophage marker Cd68, and M1- (Arg2, Nos2) and synthetic VSMC-associated (Klf4) genes, while promoting M2- (Arg1, Mrc1) and contractile VSMCs (Acta2) markers. These transcriptional changes were confirmed at the tissue level by immunofluorescence, showing a reduced aortic M1/M2 macrophage ratio in Socs1 adenovirus-treated mice (Fig. 4E). These findings indicate that Socs1 gene transfer in diabetic mice not only limits plaque growth but also reprograms VSMCs and macrophages toward a more reparative, anti-inflammatory profile.

Fig. 4.

Socs1 gene therapy reduces miR-155-5p and atherosclerosis in diabetic mice. A Expression of Socs1, Stat1 and miR-155-5p in aortas from diabetic ApoE-/- mice left untreated (Control, Ctrl; n = 6), treated with empty vector (Ad-null; n = 6) or Socs1 expressing adenovirus (Ad-Socs1; n = 8). qPCR values normalized by RNU6B (miRNA) and 18S rRNA (mRNA) are expressed as arbitrary units (a.u.). B Correlation between Socs1 and miR-155-5p expression across experimental groups. C Representative ORO/hematoxylin-stained aortic sections (scale bars, 50 μm; l, lumen; m, media; a, atheroma) and quantification of maximal lesion area per animal. D Gene expression analysis of inflammatory mediators, macrophage polarization markers, and VSMC phenotypic markers in aorta. E Immunofluorescence analysis of macrophage phenotypes in aortic sections (Arg2, red; Arg1, green; DAPI, blue; scale bars, 20 μm) and quantification of the M1/M2 ratio per mouse. Data are shown as individual values and mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 versus Ctrl or Ad-null

MiR-155-5p and Socs1 reciprocally modulate inflammatory and phenotypic programs in cultured VSMCs and macrophages

To elucidate the cellular mechanisms underlying the inverse association of miR-155-5p and Socs1, we performed in vitro studies in VSMCs and macrophages with targeted modulation of each factor. In murine VSMCs, transfection with a miR-155-5p mimic or inhibitor effectively modulated basal miR-155-5p levels (Fig. 5A) and the expression of contractile and synthetic genes (Fig. 5B), compared with respective negative controls, without affecting cell viability (Fig. 5C). Under inflammatory stimulation (IFNγ plus LPS), mimic-treated VSMCs downregulated contractile markers (Acta2, Tagln) and shifted toward a synthetic phenotype characterized by increased Klf4 marker, pro-inflammatory chemokines (Ccl2, Ccl5, Cxcl10) and matrix-remodeling enzymes (Mmp2, Mmp9). In contrast, the inhibitor attenuated these inflammatory responses and preserved a contractile state (Fig. 5D). Modulating Socs1 in VSMCs produced complementary results: siRNA-mediated Socs1 silencing induced miR-155-5p, inflammatory and synthetic markers (Fig. 5E), whereas Socs1 overexpression downregulated miR-155-5p and target genes (Fig. 5F).

Fig. 5.

In vitro modulation of the miR-155-5p/Socs1 axis in murine VSMCs. A–B qPCR analysis in control mock-transfected VSMCs (Ctrl) and transfected with miR-155-5p mimic (155-mim), miR-155-5p inhibitor (155-inh), or their respective negative controls (NC-mim and NC-inh) under basal (non-stimulated) conditions (n = 7 experiments). C Cell viability assay at 24 h (n = 4). Values expressed as % vs basal conditions. PC = positive control (20% FBS). D Expression levels of pro-inflammatory and phenotype markers in VSMCs transfected with mimic/inhibitor and stimulated for 2 h with 200 ng/mL IFNγ plus 1 μg/mL LPS (n = 6). E Gene expression analysis in VSMCs transfected with Socs1 siRNA (siSocs1) or scramble control (Scr) under basal conditions (n = 4). F Expression levels in VSMCs transfected with Socs1 overexpression plasmid (pSocs1) or empty vector (pNull) and stimulated (2 h, 200 ng/mL IFNγ plus 1 μg/mL LPS; n = 4). qPCR values normalized by RNU6B (miRNA) and 18S rRNA (mRNA) are expressed as arbitrary units (a.u.; A, B, E) or fold increases vs. basal (D, F). Data are shown as individual values with mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 versus control conditions

In macrophages, miR-155-5p inhibition increased Socs1 and reduced STAT1 phosphorylation, while the mimic had opposite effects (Fig. 6, A and B). At the transcriptional level, under inflammatory stimulation (IFNγ plus LPS), the mimic promoted an M1 pro-inflammatory profile (elevated Nos2 and Arg2), whereas the inhibitor favored M2 polarization (higher Arg1 and Mrc1) (Fig. 6C). Similar effects were observed under high-glucose conditions: miR-155-5p mimic increased pro-inflammatory markers (Nos2, Arg2 and Ccl2), while miR-155-5p inhibition enhanced M2-associated genes (Arg1, Mrc1); Il1b and Il10 followed the same directional pattern but did not reach statistical significance (Fig. 6D). Furthermore, Socs1 silencing enhanced miR-155-5p and M1-associated genes, whilst its overexpression increased M2 markers (Fig. 6, E and F). Interestingly, co-transfection experiments further support this functional link (Fig. 6G). Socs1 overexpression counteracted the miR-155-5p mimic-induced M1 profile (Nos2, Arg2), whereas Socs1 silencing blunted the miR-155-5p inhibitor-mediated induction of M2 markers (Arg1, Mrc1). These results demonstrate in vitro the reciprocal regulation of the miR-155-5p/Socs1 axis and its impact on inflammatory and phenotypic changes in vascular and immune cells, in agreement with our in vivo findings.

Fig. 6.

miR-155-5p and Socs1 regulate mouse macrophage phenotypes in vitro. A miR-155-5p expression in control mock-transfected macrophages (Ctr), transfected with miR-155-5p mimic (155-mim), inhibitor (155-inh), or respective negative controls (NC-mim and NC-inh) under basal (non-stimulated) conditions (n = 7 experiments). B Analysis of Socs1 and phosphorylated STAT1 (P-STAT1) proteins after transfection with mimic or inhibitor (n = 4–5). Representative immunoblots and quantification normalized to β-actin, shown as fold versus Ctrl. C–D Gene expression of macrophage polarization markers in mimic/inhibitor transfected cells stimulated for 24 h with 20 ng/mL IFNγ plus 100 ng/mL LPS (n = 7) (C) and high glucose (HG, 30 mM D-glucose; n = 4) (D). E Gene expression analysis in macrophages transfected with Socs1 siRNA (siSocs1) or scramble control (Scr) under basal conditions (n = 4). F Gene expressions in macrophages transfected with Socs1 overexpression plasmid (pSocs1) or empty vector (pNull) under stimulation (24 h, 20 ng/mL IFNγ plus 100 ng/mL LPS; n = 4). G Co-transfection experiments combining miR-155-5p mimic with Socs1 overexpression and miR-155-5p inhibitor with Socs1 silencing, to assess macrophage polarization markers (n = 4). qPCR values normalized to RNU6B (miRNA) and 18S rRNA (mRNA) are expressed as arbitrary units (a.u.; A, E) or fold increases vs. basal (C, D, F). Data are shown as individual values with mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 versus control conditions

miR-155-5p is a modulator of efferocytosis in atheroma plaques and in vitro

Efferocytosis, the process responsible for the clearance of apoptotic cells by phagocytes to resolve inflammation and maintain tissue homeostasis, is impaired in advanced atherosclerosis. This clearance depends on MerTK, a receptor tyrosine kinase that recognizes phosphatidylserine on apoptotic cells via Gas6 and Protein S. Proteolytic cleavage of MerTK generates a soluble form (sMerTK) that blocks ligand binding and hinders efferocytosis, thereby promoting plaque progression [8]. In diabetic ApoE-/- mice, treatment with miR-155-5p inhibitor significantly reduced sMerTK levels in aortic plaques, suggesting enhanced efferocytosis (Fig. 7, A and B). Moreover, in cultured macrophages, transfection with miR-155-5p mimic increased the release of sMerTK into the cell supernatant, as shown by Western blot (Fig. 7C). Consistently, ELISA analysis revealed that miR-155-5p mimic significantly enhanced sMerTK levels, whereas miR-155-5p inhibition reduced sMerTK release, both under basal and inflammatory conditions (Fig. 7D). In a functional assay of efferocytosis with apoptotic T cells, miR-155-5p mimic reduced the proportion of efferocytic macrophages, whereas miR-155-5p inhibition restored apoptotic cell uptake (Fig. 7E–F), indicating that miR-155-5p modulates intrinsic efferocytic capacity.

Fig. 7.

Effect of miR-155-5p on efferocytosis in vivo and in vitro. A Representative images (scale bars, 50 μm; l, lumen; m, media; a, atheroma) of sMerTK immunodetection in aortic sections from diabetic mice treated with vehicle (Ctrl, n = 8), miR-155-5p negative control (NC, n = 7) or inhibitor (155-inh, n = 7). B Quantification of sMerTK staining in atheroma plaques. C Immunoblot analysis of sMerTK in macrophage culture supernatants following transfection with miR-155-5p mimic or negative control (n = 6). Ponceau Red staining of the same membrane is shown as loading control. Relative sMerTK levels normalized to control (mock-transfected cells). D ELISA quantification of sMerTK concentration in supernatants from macrophages transfected with miR-155-5p mimic, inhibitor and respective negative controls and maintained under basal and stimulation (24 h, 20 ng/mL IFNγ plus 100 ng/mL LPS) conditions (n = 7). E Functional efferocytosis assay in macrophages transfected with miR-155-5p mimic, inhibitor and negative controls under basal conditions (n = 6). Representative higher-magnification images (scale bars, 20 μm) showing apoptotic T cells (red) engulfed by macrophages (green), indicated by arrows. F Quantification of efferocytosis index on lower-magnification fields, expressed as percentage of macrophages containing ingested apoptotic cells relative to total macrophages. Data are shown as individual values with mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 vs. control or negative control

Discussion

This study supports the role of the miR-155-5p/Socs1 axis as a modulator of vascular inflammation, plaque progression, and stability in diabetes-accelerated atherosclerosis. Using complementary in vivo miR-155-5p loss-of-function and Socs1 gain-of-function approaches in diabetic ApoE-/- mice, together with mechanistic studies in VSMCs and macrophages, we provide evidence that reciprocal regulation between miR-155-5p and Socs1 modulates inflammatory signaling, vascular cell plasticity, and efferocytosis. These findings suggest how dysregulated miRNA-target interactions may impair resolution pathways in diabetic atherosclerosis.

We first observed an inverse association between miR-155-5p and Socs1 during atherogenesis in non-diabetic mice, with miR-155-5p increasing in parallel with plaque burden, and Socs1 declining. This pattern is consistent with previous evidence that miR-155 directly targets Socs1 and amplifies cytokine-mediated signaling in inflammatory settings [10, 14, 18, 27, 38]. MiR-155 levels rise during disease progression [26, 39–41], and decrease during atherosclerosis regression [42], and elevated circulating miR-155 has been linked to disease severity in patients with coronary artery disease [11, 16, 17]. Conversely, Socs1 expression is reduced in atherosclerotic lesions, consistent with sustained JAK/STAT activation [43]. In line with evidence from autoimmune and inflammatory models [38], our data support a dynamic dysregulation of the miR-155-5p/Socs1 axis during vascular inflammation and lesion progression.

Given that diabetes accelerates atherosclerosis [4], we next examined the miR-155-5p/Socs1 axis in a diabetic context and performed functional studies to assess its impact on disease outcome. In diabetic mice, miR-155-5p inhibition restored Socs1 expression and was associated with reduced lesion size and increased plaque stability, including lower lipid and inflammatory cell content and higher collagen and VSMCs. These effects were accompanied by downregulation of inflammatory cytokines and markers of M1 macrophages and synthetic VSMCs, together with induction of M2 and contractile markers. In parallel, Socs1 gene transfer elicited protective vascular changes and reduced miR-155-5p levels. Similar anti-inflammatory effects of either miR-155 inhibition or Socs1 activation have also been reported by our group and others in normoglycemic mouse models of atherothrombosis [26, 32, 36, 44, 45]. Importantly, these vascular effects were not associated with changes in systemic metabolic parameters or liver injury, suggesting a predominant vascular contribution. Our findings are consistent with a bidirectional regulatory loop in diabetes whereby miR-155-5p promotes and Socs1 restrains vascular inflammation, consistent with prior observations in diabetic kidney disease [28].

The role of miR-155 has proven context-dependent. In non-diabetic models, its genetic deletion affects macrophage and endothelial cell function in atherosclerotic lesions [39–41], whereas myeloid deficiency can increase lipid accumulation and cytokine release [46–48]. Our results suggest that, in the diabetic milieu of hyperglycemia and chronic inflammation, miR-155-5p exerts predominantly pro-atherogenic effects, and its inhibition improves markers of plaque stability. Differences among studies likely reflect disease stage and cell-specific actions of miR-155, as suggested by integrative analysis predicting divergent roles across early and advanced lesions [49]. Our study also supports a role for Socs1, a central negative regulator of JAK/STAT signaling, as a mediator of the vascular effects of miR-155-5p inhibition in diabetic atherosclerosis. Other reports similarly reveal a reciprocal association between miR-155-5p and protective pathways, including the circadian regulator BMAL1 (brain and muscle ARNT-like protein-1), which supports endothelial cell survival [50]; HMG box-transcription protein 1, involved in limiting foam cell formation [51]; the transcriptional repressor B-cell leukemia/lymphoma 6 (Bcl6), which restrains macrophage-driven inflammation [40, 52]; and the adipokine adipolin, which promotes liver X receptor α-mediated cholesterol efflux [53]. Together, these studies reinforce the concept that excessive miR-155-5p activity can disrupt anti-inflammatory and pro-resolving mechanisms, including Socs1, thereby favoring plaque progression.

Our in vitro results help clarify the cellular mechanisms activated by inflammatory and high-glucose conditions relevant to diabetes. In VSMCs, miR-155-5p mimic promoted a synthetic phenotype with pro-inflammatory markers and matrix-remodeling enzymes, while its inhibition promoted contractile markers. In macrophages, miR-155-5p activated STAT1 and induced M1 markers, whereas its inhibition upregulated Socs1 and favored M2 polarization. These findings are consistent with previous reports identifying miR-155-5p as a driver of inflammatory macrophage activation, foam cell formation, dysregulated lipid handling [39, 40, 54] and vascular cell dysfunction [55, 56], including its induction downstream of inflammatory transcription factors such as STATs [14, 21]. Notably, Socs1 overexpression reduced STAT1 activation and was accompanied by decreased miR-155-5p levels both in vivo and in vitro, supporting an indirect regulatory mechanism whereby attenuation of STAT1-dependent inflammatory signaling limits miR-155-5p induction, rather than a direct transcriptional effect of Socs1 on the miRNA [14, 21, 28]. Consistently, Socs1 silencing recapitulated the pro-inflammatory effects of miR-155-5p, while Socs1 overexpression counteracted them. These findings support a model in which miR-155-5p induction and Socs1 suppression converge to sustain maladaptive vascular inflammation in diabetes.

Our study identifies miR-155-5p as a regulator of efferocytosis, a key pathway in inflammation resolution and homeostasis. Efficient clearance of apoptotic cells, mainly mediated by MerTK, prevents necrotic core expansion and preserves plaque integrity [5, 8]. We found that miR-155-5p inhibition in diabetic mice reduced sMerTK levels in plaques, whereas miR-155-5p overexpression in macrophages increased sMerTK release and impaired apoptotic cell engulfment. These results extend prior evidence that inflammation-induced proteolytic cleavage of MerTK cleavage leads to defective efferocytosis in plaques [8, 9, 57]. Because sMerTK reflects receptor shedding rather than transcriptional regulation, our data support an indirect mechanism whereby miR-155-5p enhances inflammatory signaling pathways that favor MerTK cleavage. In this context, cytokine-driven STAT activation has been shown to promote ADAM17-mediated MerTK shedding [58], providing a possible mechanistic link between miR-155-5p induction and the increased sMerTK levels observed in our study. Consistent with this, miR-155 has been reported to impair efferocytosis through additional inflammatory mechanisms, including repression of Bcl6 in lesional macrophages [52] and via TLR4-dependent pathway during ischemic stroke [19]. Since impaired efferocytosis contributes to plaque vulnerability and acute cardiovascular events [7], restoring this process by miR-155-5p blockade may represent a dual strategy to suppress inflammation, promote resolution, and enhance plaque stability.

Beyond mechanistic interest, our work provides preclinical evidence with translational implications. In this study, in vivo LNA-based miR-155-5p inhibition reduced atherosclerotic lesions in diabetic mice without affecting metabolic parameters, supporting the emerging concept of miRNA-targeted therapies in cardiovascular disease [11, 12]. Likewise, Socs1 induction provided vascular protection by lowering miR-155-5p and inflammation, consistent with previous studies on the beneficial effects of Socs1-based therapies (e.g. gene transfer and peptidomimetics) in metabolic and cardiovascular diseases [28, 32, 36, 45]. Among these strategies, miR-155-5p inhibition appears more readily translatable, as LNA-based antimiRs such as cobomarsen have entered clinical evaluation in oncologic disorders (NCT02580552; NCT03837457) [16], whereas Socs1 gene therapy will likely require advances in targeted delivery. Both approaches converge on restoring JAK/STAT balance and may complement current therapies. In this line, lifestyle and nutritional interventions have also been reported to modulate miR-155-5p/Socs1 axis, potentially reducing the burden of diabetes and other non-communicable chronic diseases [59].

Our study has some limitations. Experiments focused on male mice, and potential sex-related differences in miRNA regulation remain to be defined. Systemic delivery may have off-target effects, and future studies will require cell-specific targeting approaches. STZ-induced diabetes in ApoE-/- mice does not fully model obesity-associated type 2 diabetes, although it reliably reproduces hyperglycemia-driven atherosclerosis acceleration. Direct in vivo disruption of the miR-155-5p/Socs1 interaction using target-site blockers was not examined, although this strategy has been shown feasible for other miR-155-5p targets in inflammatory models [54]. Additional miRNAs may also regulate Socs1 in atherosclerosis, acting in parallel or cooperatively with miR-155-5p, such as miR-30a-5p [60]. Lipid transporters and endothelial cell-specific effects were not assessed, despite evidence that miR-155 influences cholesterol efflux [39] and endothelial function [61]. Finally, while our data implicate miR-155-5p in defective efferocytosis via MerTK, other clearance pathways, such as integrin-mediated uptake and other receptor tyrosine kinases (e.g. Tyro3 and Axl), may also contribute. These limitations point to important avenues for future mechanistic and translational studies of the miR-155-5p/Socs1 axis.

Conclusion

Our results support that miR-155-5p and Socs1 exert opposing influences on vascular inflammation and plaque remodeling in diabetes. Their reciprocal regulation is associated with changes in VSMC phenotype, macrophage polarization, and efferocytosis, thereby influencing plaque stability (Fig. 8). This combined in vivo and in vitro evidence identifies the miR-155-5p/Socs1 axis as a candidate target in diabetic atherosclerosis, warranting further evaluation in long-term preclinical models before clinical translation.

Fig. 8.

Schematic model of the reciprocal regulation between miR-155-5p and Socs1 and the downstream effects on inflammatory, phenotypic and efferocytosis-related pathways in diabetic atherosclerosis

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

α-SMA/Acta2

α-Smooth muscle actin/Actin alpha 2, smooth muscle

Arg

Arginase

ApoE

Apolipoprotein E

ALT

Alanine aminotransferase

AST

Aspartate aminotransferase

Bcl6

Repressor B-cell leukemia/lymphoma 6

Ccl

Chemokine (C–C motif) ligand

Cxcl

C-X-C motif chemokine ligand

FBS

Fetal bovine serum

IFNγ

Interferon-γ

IL

Interleukin

JAK

Janus kinase

Klf4

Krüppel-like factor 4

LPS

Lipopolysaccharide

MerTK (s)

Mer receptor tyrosine kinase (soluble)

miR/miRNA

MicroRNA

Mrc1

Mannose receptor C type 1

Mmp

Matrix metalloproteinase

Nos2

Nitric oxide synthase 2, inducible

ORO

Oil-Red-O

qPCR

Quantitative real-time PCR

SD

Standard deviation

Socs

Suppressor of cytokine signaling

STAT

Signal transducer and activator of transcription

Tagln

Transgelin

TLR

Toll-like receptor

TNFα

Tumor necrosis factor-α

VSMC

Vascular smooth muscle cells

Author contributions

M.K. and I.H.R. performed in vitro and in vivo experiments, analyzed the data, and contributed to manuscript preparation. I.P. and R.A.M. conducted in vivo and histological analyses. J.E. contributed to conceptualization, discussion, and funding. I.L. and O.L.F. participated in study conceptualization, methodology, and manuscript revision. C.G.G. conceived and supervised the study, analyzed and interpreted the data, and wrote the manuscript. All authors read and approved the manuscript.

Funding

This study was funded by MICIU/AEI/10.13039/501100011033 and “ERDF/EU” (grant PID2021-127741OB-I00 to C.G.G.), Instituto de Salud Carlos III (grant PI23/00119 to J.E.), and Conacyt-Mexico (FOP02-2022-02 #321869 to O.L.F.).

Data availability

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

Declarations

Competing interests

The authors declare no competing interests.