Establishment of VC with diabetic mouse

C57BL/6 mice (6–7 weeks old) were bought from Changzhou Cavens Lab Animal Corporation (Changzhou, Jiangsu, China). As reported, HFD prior to streptozocin (STZ) administration can induce insulin resistance (Sun et al. 2021). HFD combined with STZ is commonly used to establish T2DM mouse models (Guo et al. 2022). After 3 weeks of HFD (60% fat, TP23300, Trophic Animal Feed High-Tech Co., Ltd, Nantong, Jiangsu, China), mice were intraperitoneally injected with STZ (HY-13,753, MedChemExpress Corporation, Shanghai, China) at 40 mg/kg/day for 5 consecutive days, and other mice were injected with citrate as the control for STZ. Blood glucose levels (in the tail vein) were measured with a glucometer. The level of fasting blood glucose was detected, which was more than 16.7 mmol/L was determined as T2DM mice. T2DM mice were divided into three groups: the DM group (n = 5), the DVC group (n = 5), and the DVC + EX4 group (n = 5). At the fifth week, mice in the DVC and DVC + EX4 groups were injected subcutaneously with VD2 (5 µL/g) for 2 days (Guo et al. 2022; Platko et al. 2020), while mice in the DM group were injected subcutaneously with phosphate-buffered saline (PBS, 5 µL/g) as the control for VD2. Subsequently, mice in the DVC + EX4 group were injected intraperitoneally with EX4 (24 nmol/kg/day, HY-13,443, MedChemExpress Corporation, Shanghai, China) for 26 consecutive days (Zhang et al. 2015; Yamane et al. 2011), and other mice were injected with the same volume of PBS as the control for EX4. The control mice (n = 5) were fed with control fat diet (CFD, LAD3001G, Trophic Animal Feed High-Tech Co., Ltd, Nantong, Jiangsu, China), injected intraperitoneally with the same volume of citrate and PBS. All the mice were euthanized at the ninth week. Thoracic aortic tissues were harvested for detection.

Alkaline phosphatase (ALP) activity assay of aortic tissues

To measure calcium content, aortic tissues from the aortic arch to the iliac bifurcation were dissected and dried at 55 °C, and calcium was extracted with 10% formic acid overnight at 4 °C. Colorimetric quantification of calcium was achieved by a reaction with o-cresolphthalein, and total protein was determined by the Bradford protein assay (Luo et al. 2009).

Von Kossa and alizarin red S (ARS) staining of aortic tissues

Von Kossa and ARS staining of aortic tissues were performed using the calcium staining kit (Von Kossa method, G3282, Solarbio Life Sciences, Beijing, China) and alizarin red S solution (G1450, Solarbio Life Sciences, Beijing, China). Aortic tissues were fixed in 10% neutral formalin prior to dehydration and embedding. Sections were embedded in 95% ethanol, placed vertically, and air-dried thoroughly. Von Kossa staining was performed according to the manufacturer’s instructions. For the ARS staining, the sections were then placed in a vat containing ARS solution and stained for 5–10 min followed by a quick rinse in distilled water. The sections were dehydrated in a conventional transparent manner and embedded in resin.

Immunofluorescence staining

Aortic tissue was fixed in 4% paraformaldehyde, and permeabilized with 0.2% Triton X-100. Sections were stained for GLP-1R (green) with ab218532 (Abcam, Cambridge, UK) at 1/500 dilution, followed by AlexaFluor®488 goat anti-rabbit secondary antibody (ab150077, Abcam, Cambridge, UK) at 1/1000 dilution. Sections were also stained for α-SMA (red) with 67735-1-Ig (Proteintech, Wuhan, Hubei, China) at 1:200, followed by AlexaFluor®647 goat anti-mouse secondary antibody (ab150115, Abcam, Cambridge, UK). Nucleic acids were stained with 4’,6- diamidino‐2‐phenylindole (DAPI). Immunohistochemical signal intensity and positively stained field of tissue sections were evaluated using ImageJ software. Tomm20 antibody (ab283317, Abcam) and LAMP1 antibody (ab62562, Abcam) were used for cell immunofluorescence staining.

Human aortic smooth muscle cell (HASMC) culture and induction of osteogenic differentiation

HASMCs were purchased from Procell Technology (CP-H081, Wuhan, Hubei, China). They were cultured in the complete medium of HASMCs (CM-H081, Procell Technology, Wuhan, Hubei, China) in an incubator at 37℃ and under 95% air and 5% CO2 conditions. To induce osteogenic differentiation, HASMCs were treated with 10 mM β-GP (Ma et al. 2020) in the absence or presence of high glucose (30 mM) (Ghasempour et al. 2022). In the β-GP + EX4 group, EX4 (100 nM) was added to the cell culture medium for 14 days (Takaku et al. 2021), and the calcium deposition was visualized using a calcium assay kit (ab102505, Abcam, Cambridge, UK) and ARS staining.

Knockdown of GLP-1R in HASMCs

Small interfering RNA against GLP-1R (Si-Glp 1r) and the control siRNA were synthesized by RiboBio Technology, Guangzhou, Guangdong, China. Cell transfection was performed using Lipofextamin®RNAiMAX (Thermo Fisher Scientific, Waltham, MA, USA). Briefly, cells were seeded to reach 80% confluence at the time of transfection. Lipofextamin®RNAiMAX reagent was diluted in Opti-MEM® medium (Thermo Fisher Scientific, Waltham, MA, USA). Subsequently, siRNAs were also diluted in Opti-MEM® medium. The diluted siRNAs were added to the diluted Lipofextamin®RNAiMAX reagent (1:1 ratio). They were incubated at room temperature for 5 min. The siRNA-lipid complex was then added to HASMCs. After 48 h, transfected HASMCs were harvested for the induction of osteogenic differentiation and β-GP/EX4 treatments. Knockdown of AMPKα1 (Si-AMPKα1) in HASMCs was performed as described above.

Western blotting

Cells were lysed on ice for 30 min in RIPA buffer (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 1 mM protease inhibitor phenylmethylsulfonyl fluoride (PMSF, Gibco, Waltham, MA, USA). Samples were boiled in protein loading buffer for 10 min at 100 °C in a metal bath, and equal amounts of proteins were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gel and transferred to polyvinylidene fluoride (PVDF) membranes (Millipore, Bedford, USA). The membranes were then blocked in TRIS-buffered saline with 0.5% Tween 20 (TBST) containing 3% bovine serum albumin (BSA) for 1 h at room temperature, followed by incubation with primary antibodies at 4 °C overnight. Secondary antibodies (Goat Anti-Rabbit IgG H&L HRP, ab6721, 1:2000) were incubated for 1 h at room temperature, and the membrane signals were visualized using a chemiluminescent horseradish peroxidase (HRP) substrate reagent (Bio-Rad, Hercules, CA, USA), and images were captured using a Tanon5200 imaging system (Biotanon, Shanghai, China). β-Actin or Tomm20 was used as a control for cell lysate and mitochondrial fractions, respectively. The primary antibodies were shown as follows: anti-RUNX2 (1:200, 20700-1-AP Proteintech, Wuhan, Hubei, China), anti-BMP2 (1:1000, 15544-1-AP, Proteintech), anti-LC3B (1:1000, 14600-1-AP, Proteintech), anti-p62 (1:10000, ab109012, Abcam), anti-PINK1 (1:500, 23274-1-AP, Proteintech), anti-Parkin (1:1000, 14060-1-AP, Proteintech), anti-pAMPKα (T172) (1:1000, ab133448, Abcam), anti-AMPKα (1:1000, ab32047, Abcam), anti-pULK1 (S555) (1:1000, #5869, Cell Signaling Technology, Danvers, MA, USA), and anti-ULK1 (1 µg/mL, ab167139, Abcam).

Mitochondrial function test

Mitochondrial membrane potential was detected with tetramethylrhodamine, methyl ester (TMRM, Thermo Fisher Scientific, Waltham, MA, USA) probe, and mitochondrial superoxide levels were detected with MitoSOX red (Thermo Fisher Scientific, Waltham, MA, USA). TMRM was diluted at 10 µM in media (10×) and aliquoted for single use. MitoSOX was also aliquoted for single use (Little et al. 2020). Relative fluorescence intensity (RFI) was calculated and analyzed using ImageJ software.

Transmission electron microscopy (TEM)

Sections were fixed in 2.5% glutaraldehyde. Postfixation was performed in 1% (v/v) osmium tetroxide in PBS for 2 h at 4℃. Sections were then dehydrated through an acetone gradient and embedded in Araldite. Ultrathin sections were cut with a Leica Ultracut R ultramicrotome (Leica, Germany) and stained with lead citrate and uranyl acetate. Sections were observed under a JEM-1230 transmission electron microscope (JEOL, Japan).

Mitochondrial fraction protein extraction

Mitochondrial fractions in mouse aortic tissue were isolated using the Mitochondria Isolation Kit for Tissue (89,801, Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions. Mitochondrial fractions in HASMCs were isolated using the Mitochondria Isolation Kit for Cultured Cells (89,874, Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer’s instructions.

Knockdown of AMPKα1 in vivo

Adeno-associated virus (AAV) 9 was used as the knockdown vector. The AMPKα1 knockdown vector (AAV-sh- AMPKα1) and the negative control (NC) vector (AAV-sh-NC) were established by Hanbio Technology (Shanghai, China). One day before VD2 injection, the AAV vectors (1011 vector genome copies/mouse) were administered by tail vein injection. Other treatments were the same as the described above. All the mice were euthanized at week 9. Thoracic aortic tissue was harvested for detection.

Statistical analysis

All data are presented as mean ± standard deviation (SD). SPSS 26.0 (IBM, Armonk, NY, USA) and GraphPad Prism 9 were used for data analysis. ImageJ software was used for image analysis. The difference between two groups was compared by using the Student’s t-test. The difference between multiple groups was compared by using the one-way analysis of variance (ANOVA) followed by the Tukey’s post hoc test. Differences were considered statistically significant when P < 0.05.