**1. Introduction**

Nearly 17.5 million people die each year from atherosclerosis-related diseases (31% of the global mortality). Of these, approximately 7.4 million died from coronary heart disease and 6.7 million from stroke [1]. Vulnerable atherosclerotic plaque rupture is the principal mechanism that accounts for myocardial infarction and stroke [2]. Therefore, there is a clinical need for plaque stabilization drugs.

Metformin, a biguanide, is the top choice of oral agent for the treatment of type 2 diabetes owing to its glucose-lowering effectiveness, safety, favorable effect on body weight, and low cost [3]. Moreover, metformin has been associated with decreased all-cause mortality and a reduced incidence of cardiovascular disease among patients with diabetes. A clinical trial investigated the effect of long-term metformin use and lifestyle at a diabetes prevention program and found that metformin was protective against atherosclerotic vascular disease early in diabetes development and potentially extended the range of this action to include high-risk male prediabetic subjects [4]. A recent meta-analysis showed

**Citation:** Chen, X.; Wang, S.; Xu, W.; Zhao, M.; Zhang, Y.; Xiao, H. Metformin Directly Binds to MMP-9 to Improve Plaque Stability. *J. Cardiovasc. Dev. Dis.* **2023**, *10*, 54. https://doi.org/10.3390/jcdd10020054

Academic Editors: Giovanni Cimmino and Plinio Cirillo

Received: 22 December 2022 Revised: 19 January 2023 Accepted: 27 January 2023 Published: 30 January 2023

**Copyright:** © 2023 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

the association between metformin and decreased cardiovascular mortality (95% CI, OR 0.44 [0.34−0.57]) or incidence of cardiovascular diseases (95% CI, OR 0.73 [0.59−0.90]) among patients with diabetes [5]. However, the molecular mechanism by which metformin improves atherosclerosis plaque stability remains unclear.

Vulnerable plaques are characterized by fragile, thin fibrous caps, massive lipid cores, intraplaque hemorrhage, immune activation, and increased levels of pro-inflammatory mediators (cytokines, chemokines, and matrix metalloproteinases) [6]. IL-6 stimulates the expression of adhesive molecules and results in an increase in the production and reactivity of acute phase indicators, such as C-reactive protein and TNF-α [7,8]. IL-18 and TNF-α are crucial for atherosclerotic plaque development and stability [9,10]. All of the cytokines mentioned above play a significant role in the formation and destabilization of atherosclerotic plaques. Mature plaques mainly comprise endothelial cells, vascular smooth muscle cells, macrophages, and fibrous caps containing extracellular matrix (ECM) components [11]. Among these components, the ECM is especially important for plaque stability [12]. Proteases have been implicated in the development and progression of atherosclerosis due to their ability to cause focal destruction of the ECM of blood vessels. Matrix metalloproteinase (MMP)-9, also known as gelatinase B, is a widely studied member of the MMP family. Histopathological studies have shown that MMP-9 is mainly distributed in the shoulder area, necrotic core, and fibrous cap area of atherosclerotic plaques, and the level and activity of MMP-9 in unstable plaques are higher than those in stable plaques [13–15]. Moreover, many studies have shown that high MMP-9 expression can be used as a predictor of atherosclerotic plaque instability, whereas its overexpression may lead to plaque instability [16–18]. Therefore, MMP-9 is a potential target for improving atherosclerotic plaque stability. However, whether metformin can target MMP-9 and inhibit its activity to stabilize plaque remains unclear.

Here, we report a novel mechanism by which metformin directly binds to MMP-9 and inhibits its activity to improve atherosclerotic plaque stability.

#### **2. Materials and Methods**

#### *2.1. Mice*

The investigations conformed to the US National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, revised 1996). Animal experiments were approved by the Committee of Peking University on Ethics of Animal Experiments (LA 2018-112) and conducted in accordance with the Guidelines for Animal Experiments, Peking University Health Science Center. Male ApoE knockout mice (ApoE- /-, C57BL/6J background) were purchased from Cyagen Biosciences Inc. (Suzhou, China) and used for the experiments. From 8 weeks of age, ApoE-/- mice were fed a high-fat, highcholesterol diet containing 40 kcal% fat and 1.25% cholesterol (D12108C; Research Diets, New Brunswick, NJ, USA) for 14 weeks. All mice were housed in a specific pathogen-free environment under a 12 h/12 h light-dark cycle.

#### *2.2. Carotid Collar Placement and Drug Treatment*

Male ApoE knockout mice (8 weeks of age, C57BL/6J background) were fed a highfat diet containing 40 kcal% fat and 1.25% cholesterol (D12108C; Research Diets, New Brunswick, NJ, USA) for 2 weeks. Carotid collar placement was performed 2 weeks later, and the operation process is briefly described as follows [19]: the mice were weighed, anesthetized by an intraperitoneal injection of 2% pentobarbital sodium (50 mg·kg<sup>−</sup>1), and their limbs were fixed on a thermostatic operating table. Erythromycin eye ointment was applied to the eyes of the mice to prevent dry eyes. Hair removal ointment was applied to remove neck and chest fur and fully expose the neck and chest surgical field. The epidermis was cut off at the median line of the neck using scissors, the right common carotid artery (RCCA) was bluntly separated with forceps, and the accompanying nerves and vessels were not damaged. A silicone collar with an inner diameter of about 0.3 mm (≈30% stenosis) was placed on the lateral side of the RCCA. The collar was fixed with a 6-0 silk thread

and sutured for disinfection. Meloxicam (1.5 mg·kg−1) was injected intraperitoneally for analgesia after surgery and resuscitated on heat mats. High-fat feeding was continued for more than 3 months until plaque formation. Subsequently, metformin (Sigma-Aldrich, St. Louis, MO, USA; 200 mg·kg−<sup>1</sup> body weight) or saline was administered by intragastric gavage for 14 consecutive days.

#### *2.3. Histopathology and Immunofluorescence*

The RCCAs from mice were harvested and embedded in an OCT compound (Lot# 4583; Tissue-Tek, USA). The OCT-embedded vascular tissue was sequentially sliced into slices approximately 6–8 μm thick using a microtome (Leica, Wetzlar, Germany), and placed on polylysine-coated glass slides. For all subsequent pathological staining (including immunofluorescence, oil red O, and Sirius red), 2–4 frozen sections of each vascular tissue with an interval of more than 50 μm were stained, and the average of the statistical values from the same sample was used as the final result [20].

To analyze plaque stability, serial sections (8 μm thick) were stained with picrosirius red to detect collagen deposition and oil red O to detect lipid deposition; both stains were analyzed by quantifying the positive area per total plaque area. Slices were incubated with primary antibodies against the macrophage marker CD-68 (1:50 dilution; Abcam, ab53444, Cambridge, UK) and smooth muscle cell marker α-SMA (1:50 dilution; Abcam, ab124964, Cambridge, UK), followed by incubation with fluorescence-conjugated secondary antibodies. The sections were mounted with 4', 6-diamidino-2-phenylindole (DAPI; Abcam, ab104139, Cambridge, UK) for nuclei visualization.

To further characterize the carotid arteries, slices were incubated with the following primary antibodies: anti-MMP-9 (1:50 dilution; Invitrogen, MA5-15886, Carlsbad, CA, USA), anti-active MMP-9 (1:50 dilution; NOVUS, NBP2-13173, Carlsbad, CA, USA), anti-MMP-2 (1:50 dilution; Abcam, ab92536, Cambridge, UK), anti-MMP-12 (1:50 dilution; Proteintech, 22989-1-AP, Rosemont, IL, USA), anti-IL-1β (1:50 dilution; Bioss, bs0812R, Peking, China), anti-IL-6 (1:50 dilution; Proteintech, 66146-1-Ig, Rosemont, IL, USA), and anti-TNF-α (1:50 dilution; Abcam, ab1793, Cambridge, UK), followed by incubation with fluorescenceconjugated secondary antibodies. The sections were mounted with 4', 6-diamidino-2 phenylindole (DAPI; Abcam, ab104139, Cambridge, UK) for nuclei visualization.

#### *2.4. Western Blotting*

Liver tissues and cell lines were lysed in a RIPA lysis buffer containing 1 mmol·L−<sup>1</sup> phenylmethanesulfonyl fluoride (Beyotime Institute of Biotechnology, Beijing, China) at 4 ◦C for 30 min. The lysates were then centrifuged at 15,000× *g* for 10 min at 4 ◦C and their protein concentrations were determined using the BCA Protein Assay (Beyotime Institute of Biotechnology, Beijing, China). Samples were mixed with 5× SDS loading buffer, boiled for 5 min, and 50 μg of total protein was subjected to SDS-PAGE in 10% gels and transferred to nitrocellulose membranes. After blocking, the membranes were incubated overnight at 4 ◦C with the following primary antibodies: anti-MMP-9 (1:1000 dilution; Invitrogen, MA5-15886, Carlsbad, CA, USA), anti-p-AMPK (1:1000 dilution; CST, #2535, Danvers, MA, USA), anti-AMPK (1:1000 dilution; CST, #2532, Danvers, MA, USA), and anti-GAPDH (1:5000; CST, #2118, Danvers, MA, USA). The membranes were washed with Tris-buffered saline/0.1% Tween 20 (TBST) and incubated with secondary antibodies for 1 h at 25 ◦C. Signals were detected using Pierce™ ECL Western Blotting Substrate (Thermo Fisher Scientific, Waltham, MA, USA). Protein levels were quantified by calculating the grayscale value of each band using ImageJ (version 1.43, National Institutes of Health, Bethesda, MD, USA) software.

#### *2.5. Matrix Metalloproteinases (MMPs) Activity Assay*

Matrix metalloproteinases (MMPs) activity in mouse serum and cell culture supernatant was measured using Invitrogen DQTM luciferase-conjugated gelatin substrate (D12054; Invitrogen, Carlsbad, CA, USA), a fluorescent substrate that can detect protease

activity with high sensitivity. The substrate consists of highly quenched fluorescein-labelled gelatin. After proteolytic digestion, the exhibited bright green fluorescence can be used to measure enzyme activity. Increased fluorescence intensity was monitored using a fluorescent microplate reader or fluorimeter. After receiving the cell supernatant, the cells were incubated with DQ gelatin, and a zinc-ion-containing buffer was added. After standing at room temperature and away from light for 24 h, the fluorescence intensity of each well was measured using a fluorescence microplate reader (TECAN, Männedorf, Switzerland).

#### *2.6. Molecular Docking and Dynamics Simulation*

The ligand metformin was processed using the Schrödinger 10.2 software (Schrödinger, LLC, NY, USA) LigPrep module. An OPLS3 force field was adopted for energy minimization. The crystal structure of MMP-9 was obtained from the RCSB Protein Data Bank. The crystallographic structure of 4WZV was prepared using the Protein Preparation Wizard module. A glide was applied to predict the potential binding mode of metformin with the MMP-9 protein. Following the docking results, an independent 50 ns molecular dynamics simulation was performed using Desmond. Na<sup>+</sup> and Cl<sup>−</sup> ions were each added at the physiological concentration of 0.15 mol·L−<sup>1</sup> to ensure the overall neutrality of the systems. Simulations were conducted using an OPLS3 force field and a TIP3P explicit solvent model. The final size of the solvated system was approximately 20,000 atoms. A 5 ps recording interval was selected, and the NPT ensemble was employed with a fixed temperature of 300 K and pressure of 1.01 bar. The analysis tool of the simulation interactions diagram was used to monitor ligand–protein interactions.

#### *2.7. Cell Culture, Plasmids, and Transfection*

HEK 293A cells were obtained from the Cell Resource Center, Peking Union Medical College (which is the headquarter of National Science & Technology Infrastructure– National BioMedical Cell-Line Resource, NSTI-BMCR). Cells were maintained at 37 ◦C, with 5% CO2 in DMEM supplemented with 10% FBS and 10<sup>4</sup> <sup>U</sup>·mL−<sup>1</sup> Pen/Strep. MMP-9 was overexpressed using an MMP-9-pcDNA3.1(+)-3Xflag plasmid synthesized by Ruibiotech (Beijing, China). Control plasmid did not contain sequences homologous to those of humans, mice, or rats. HEK 293A cells were seeded into 6-well plates (1.0 × 106 cells/well) for 24 h and transfected with MMP-9 or control plasmid using lipofectamine 3000 (Invitrogen, Waltham, MA, USA) for 24 h, according to the manufacturer's instructions. Furthermore, the transfected cells were incubated with metformin (Sigma-Aldrich, St. Louis, MO, USA; <sup>1</sup> <sup>μ</sup>mol·L−1) for an additional 24 h. For the degradation experiment, the transfected HEK293A cells were pretreated with metformin for half an hour and incubated with cycloheximide (CHX; MedChemExpress, HY-12320; 10 <sup>μ</sup>mol·L<sup>−</sup>1) to block protein synthesis for the indicated periods (0, 1, 2, 3 h). Lysates are harvested from the cells and analyzed by Western blotting.
