Skip to main content
Lippincott Open Access logoLink to Lippincott Open Access
. 2026 Sep 3;46(10):e324585. doi: 10.1161/ATVBAHA.126.324585

MAP17 Orchestrates SGLT2-Dependent Glycolytic Reprogramming to Drive Trained Immunity and Accelerate Atherosclerosis

Jianjin Wu 1, Lei Wang 3, Fukang Zou 1, Fangbing Liu 1, Guanyu Fang 1, Kai Cheng 1, Haochen Gao 2, Kangkang Zhi 1,✉, Lefeng Qu 1,✉
PMCID: PMC13596194  PMID: 42689317

Abstract

BACKGROUND:

Atherosclerosis is driven by metabolic-immune crosstalk, in which trained immunity sustains vascular inflammation. MAP17 (membrane-associated protein 17), a redox- and metabolism-regulating adaptor protein, functions as a potential upstream driver of SGLT2 (sodium-glucose cotransporter 2). We aimed to determine whether MAP17 links hyperglycemia to glycolytic activation, inflammatory polarization, and plaque progression in atherosclerosis.

METHODS:

MAP17 expression and its correlations with clinical risk factors were analyzed in serum from 30 patients with atherosclerosis. A trained immunity model was established in bone marrow–derived macrophages via sustained high glucose and IFN-γ (interferon-γ)/lipopolysaccharide stimulation. Functional assays were performed after MAP17 overexpression/knockdown, SGLT2 silencing, or glycolysis inhibition.

RESULTS:

MAP17 was significantly coupregulated in patients with atherosclerosis, with the highest levels observed in those with concomitant diabetes or metabolic syndrome, and closely associated with elevated proinflammatory M1-like cytokines. Immunohistochemistry of carotid plaques confirmed its colocalization with SGLT2 within CD68+ macrophage-rich, lipid-laden, and inflamed regions. In bone marrow–derived macrophages, high glucose robustly induced MAP17 expression, which unidirectionally upregulated SGLT2, enhanced glycolytic flux, increased lactate production, and promoted M1-like polarization and foam cell formation. MAP17 knockdown markedly suppressed SGLT2 expression, glycolysis, and TNF-α (tumor necrosis factor-α)/IL (interleukin)-1β secretion, whereas MAP17 overexpression restored glycolytic activity, proinflammatory phenotype, and foam cell generation even in SGLT2-deficient cells. In diabetic chimeric Apoe−/− mice, MAP17 activation correlated with increased glycolytic marker expression, higher proinflammatory M1-like macrophage ratios, aggravated vascular inflammation, and greater plaque burden; these effects were mitigated by MAP17 or SGLT2 silencing, or by glycolysis inhibition.

CONCLUSIONS:

MAP17 is a key upstream controller of the SGLT2-glycolysis axis that promotes trained immunity and accelerates atherosclerosis. Targeting MAP17 may disrupt the metabolic-inflammatory feedback loop and represents a promising therapeutic strategy for diabetic atherosclerosis.

Keywords: atherosclerosis, glycolysis, oxidative stress, reactive oxygen species, trained immunity


What Are the Clinical Implications?

Our study identifies MAP17 (membrane-associated protein 17) as a previously unrecognized upstream regulator of the SGLT2 (sodium-glucose cotransporter 2)-glycolysis axis in macrophages, linking hyperglycemia-driven trained immunity to accelerated atherosclerosis. These findings extend the current understanding of diabetic atherosclerosis beyond conventional lipid-centered mechanisms by highlighting an immunometabolic pathway that may contribute to persistent vascular inflammation and plaque progression. Clinically, MAP17 may serve as a potential biomarker for identifying metabolically high-risk patients with atherosclerosis, particularly those with diabetes or metabolic syndrome. In addition, because MAP17 acts upstream of SGLT2 and promotes proatherogenic inflammatory programming, targeting MAP17 may represent a novel therapeutic strategy to complement existing metabolic and cardiovascular interventions. Although further validation in larger clinical cohorts and translational studies is required, these findings provide a rationale for refining risk stratification and developing mechanism-based therapies for diabetic atherosclerosis.

Atherosclerosis is a chronic inflammatory vascular disease and a leading cause of cardiovascular and cerebrovascular morbidity and mortality worldwide.1–3 It is characterized by persistent vascular inflammation, lipid deposition, and maladaptive tissue remodeling. A central driver of this process is the dynamic crosstalk between metabolic disturbance and immune activation. While both innate and adaptive immune cells contribute, macrophages are of particular importance due to their abundance in plaques and their ability to integrate metabolic signals with inflammatory responses.4–6

Recent advances reveal that macrophages can acquire trained immunity, a memory-like proinflammatory phenotype established after transient exposure to microbial or metabolic stimuli, which persists for weeks to months and primes cells for heightened responses to subsequent insults.7,8 This phenomenon is underpinned by metabolic and epigenetic reprogramming, enabling sustained cytokine production, oxidative stress, and tissue injury even in the absence of ongoing stimulation. Notably, trained macrophages frequently display rewired redox metabolism, with enhanced mitochondrial and NADPH (nicotinamide adenine dinucleotide phosphate) oxidase–derived reactive oxygen species (ROS) production.9,10 Beyond their direct oxidative damage, these ROS act as potent signaling molecules that amplify proinflammatory gene expression, stabilize hypoxia-inducible factors, and perpetuate the inflammatory-metabolic loop within atherosclerotic lesions.11

Among the upstream regulators potentially linking metabolic rewiring to trained immunity, MAP17 (membrane-associated protein 17) has emerged as an intriguing yet understudied candidate. MAP17 is a small, 17-kDa nonglycosylated transmembrane protein with minimal expression in normal tissues but marked overexpression in multiple human malignancies.12 Functionally, MAP17 perturbs cellular redox homeostasis, resulting in sustained ROS accumulation that both damages cellular components and activates signaling cascades, such as PI3K/AKT (phosphoinositide 3-kinase/protein kinase B),13 promoting PTEN (phosphatase and tensin homolog) oxidation, inhibiting apoptosis, and driving proliferation.

Beyond oncogenic contexts, MAP17 is essential in renal glucose handling. In the proximal tubule, it serves as a necessary activator of SGLT2 (sodium-glucose cotransporter 2), increasing transport activity by up to 2 orders of magnitude without affecting surface expression.14 Genetic evidence underscores this role: a patient with familial renal glucosuria lacking SGLT2 mutations was found to harbor a homozygous splice-site mutation in the MAP17 (PDZK1IP1) gene. MAP17 also interacts with PDZ-domain scaffolding proteins, including PDZK1 (PDZ domain-containing protein 1), NHE3 (sodium-hydrogen exchanger 3), and NHERF (Na+/H+ exchanger regulatory factor) family members, affecting transporter localization, signaling, and endocytic trafficking.15 Recent cryo-electron microscopy studies have resolved the human SGLT2-MAP17 complex, revealing MAP17’s binding to transmembrane helix 13 of SGLT2 and stabilization of conformations critical for sodium-coupled glucose symport and inhibitor binding.16,17

Given that enhanced glycolysis and dysregulated lipid metabolism are well-established hallmarks of atherogenesis, the metabolic functions of MAP17 provide a plausible mechanistic link to vascular disease. Elevated ROS production and a glycolytic bias, hallmark consequences of MAP17 upregulation, are known to trigger endothelial activation, facilitate monocyte recruitment, and polarize macrophages toward proinflammatory phenotypes, thereby accelerating plaque growth and destabilization. However, the role of MAP17 in trained immunity and atherosclerosis progression remains largely undefined. We therefore hypothesize that MAP17 serves as a key regulator of atherosclerosis. This study aims to define the role of MAP17 in macrophage immunometabolic reprogramming and determine how its redox and metabolic effects contribute to atherosclerosis progression under hyperglycemic-inflammatory conditions.

Materials and Methods

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Human Ethics and Consent to Participate Declarations

This study was approved by the Human Ethics Committee of Second Affiliated Hospital of Naval Medical University (ethics approval number: P2022015). All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Ethics Statement

This study was approved by the Animal Ethics Committee of Second Affiliated Hospital of Naval Medical University (ethics approval number: DL20220426).

Main Reagents and Antibodies

The main reagents and antibodies used in this study are given as follows: DMEM (11965092, Thermo Fisher Scientific), recombinant mouse IFN-γ (interferon-γ; 485-MI-100, R&D Systems) and IL (interleukin)-4 (404-ML-010, R&D Systems), lipopolysaccharide (L8880-10MG, Solarbio), dichloroacetate (347795-25G, Sigma-Aldrich), MAP17 shRNA (short hairpin RNA)/oeRNA and SGLT2 shRNA lentivirus (Yunzhou Biotechnology, Guangzhou, China), TRIzol reagent (15596026, Invitrogen), cDNA transcription kit (K1622, Thermo Fisher Scientific), radioimmunoprecipitation assay lysis buffer (89900, Thermo Fisher Scientific), anti-SGLT2 (ab306558, Abcam), anti-MAP17 (ab199540, Abcam), anti-β-actin primary antibodies (ab8227, Abcam), HRP (horseradish peroxidase)–conjugated secondary antibody (7074S, Cell Signaling Technology), enhanced chemiluminescence assay kit (34580, Thermo Fisher Scientific), cell counting kit-8 (CK04-05, Dojindo Molecular Technologies), dimethyl sulfoxide (D8418, Sigma-Aldrich), XFp Real-Time ATP Rate Assay kit (Agilent Seahorse Bioscience), fluorescein isothiocyanate-conjugated anti-F4/80 (F4/80-FITC; ab60343, Abcam), phycoerythrin-labeled anti-iNOS (inducible nitric oxide synthase-PE; ab310895, Abcam), allophycocyanin-conjugated anti-CD206 (cluster of differentiation 206-APC; ab270647, Abcam) antibodies, ELISA kit (Fanke Biotechnology, Guangzhou, China), PKH67 fluorescent cell linker kit (HR8659, Biorebo), Dil-acetylated LDL (low-density lipoprotein; L35354, Thermo Fisher Scientific), hematoxylin and eosin staining kit (G1120, Solarbio), Oil Red O (ORO) staining kit (G1261, Solarbio), and Masson trichrome staining kit (G1346, Solarbio).

Clinical Sample Collection

From March 2022 to March 2023, atherosclerotic plaque specimens were collected from 30 patients with atherosclerosis who underwent carotid endarterectomy in our hospital. General clinical data, including fasting plasma glucose and HbA1c (hemoglobin A1c), were systematically recorded for all patients with atherosclerosis. Fasting plasma glucose was measured after an 8-hour fast using a standardized enzymatic method. According to the Gensini scoring system for assessing disease severity, the patients were divided into a mild-to-moderate group (n=19, Gensini score <80) and a severe group (n=11, Gensini score ≥80). In addition, peripheral blood samples were taken from these 30 patients with atherosclerosis and 30 sex- and age-matched healthy controls. Serum SGLT2 and MAP17 mRNA and protein levels were detected by quantitative reverse transcription-polymerase chain reaction (qRT-PCR) and Western blot, and the correlation between MAP17 and SGLT2 expression was assessed by the Pearson correlation analysis. The concentrations of inflammatory cytokines TNF-α (tumor necrosis factor-α), IFN-γ, and IL-4 in plaque tissues were determined using ELISA. Potential factors affecting atherosclerosis progression were identified by univariate analysis. This study complied with the Declaration of Helsinki and was approved by the Hospital Ethics Committee (approval number: P2022015), with informed and voluntary consent signed by all patients or their families. To ensure the stability of circulating mRNA, serum samples were centrifuged at 4 °C within 1 hour of collection. Total RNA was extracted using TRIzol LS reagent under strictly RNase-free conditions, followed by immediate reverse transcription to minimize transcript degradation.

Cell Culture

Bone marrow (BM)–derived macrophages (BMDMs) were obtained by flushing BM from mouse femurs and tibias. Cells were seeded in nontissue culture dishes and maintained in differentiation medium containing high-glucose DMEM (4.5-g/L glucose), 10% fetal bovine serum, 1% penicillin/streptomycin, 2-mmol/L L-glutamine, and 20% L929 cell-conditioned medium serving as a source of macrophage colony-stimulating factor. After 7 days of differentiation, adherent BMDMs were harvested and plated at a density of 2×105 cells/mL in complete DMEM supplemented with glucose (final concentration of 20 mmol/L), 10-ng/mL IFN-γ, 100-ng/mL lipopolysaccharide, and 10-ng/mL recombinant mouse IL-4 for 16 hours before use. Controls were simultaneously cultured in basal DMEM (5-mmol/L glucose) for 16 hours.

BMDM Metabolic Memory Model Establishment

BMDMs were differentiated with 5- or 20-mmol/L glucose as described above and cultured for 1, 2, 4, or 7 days. BMDMs were positively selected by adherence, transferred to complete DMEM with 5-mmol/L glucose, and incubated without disturbance for 48 hours to normalize glucose levels. BMDMs were then stimulated with 10-ng/mL IFN-γ and 100-ng/mL lipopolysaccharide in the medium for 16 hours to establish an in vitro metabolic memory model.18

Lentiviral Vector Construction and Grouping

Recombinant lentiviral vectors carrying MAP17 shRNA, MAP17 overexpression constructs, SGLT2 shRNA, and corresponding negative controls were commercially constructed by Fenghui Biotechnology (Changsha, China). Knockdown vectors were based on the pGMLV-hU6-MCS (multiple cloning site)-CMV-ZsGreen1-PGK-Puro-WPRE backbone, whereas overexpression vectors used the pLV-CMV-MCS-EF1-mCherry-T2A-Puro system. Viral titers ranged from 1.1×108 to 2.4×108 TU/mL. The shRNA target sequences used in this study are given as follows. Briefly, the MAP17 shRNA targeting sequence was 5’-GCTTTGCATGAACTGGAGAAT-3’ (targeting mouse Pdzk1ip1, NM_021551, Sigma-Aldrich), and the SGLT2 shRNA targeting sequence was 5’-GCCTATTTCCTGCTGGTCATT-3’ (targeting mouse Slc5a2, NM_133254, Sigma-Aldrich). For the overexpression construct, the full-length mouse Map17 cDNA (NM_021551) was subcloned into the MCS of the pLV-CMV-MCS-EF1-mCherry-T2A-Puro vector; an empty vector served as the negative control (NC [negative control]-oeRNA). For in vitro BMDM transduction, differentiated BMDMs were infected with corresponding lentiviral vectors according to the manufacturer’s instructions. The lentiviral vectors containing the ZsGreen1 fluorescent reporter gene were used to monitor successful transduction, which was evaluated by fluorescence microscopy at 72 to 96 hours after infection (Figure S1; Table S1). Following lentiviral infection, stably transduced BMDMs were selected with puromycin (1 μg/mL) for 5 days and subsequently maintained in complete medium for subsequent experiments. The efficiency of lentiviral-mediated gene modulation was confirmed by qRT-PCR and Western blot analysis. MAP17 and SGLT2 knockdown efficiencies were validated by reduced target gene expression levels compared with the NC-shRNA group (Figure 3A and 3B). In addition, MAP17 overexpression and its regulatory effect on SGLT2 expression were confirmed by qRT-PCR and Western blot analysis (Figure 4A and 4B). BMDMs were grouped into control, hyperglycemia, IFN-γ+lipopolysaccharide, IFN-γ+lipopolysaccharide+dichloroacetate (10 mmol/L, glycolysis inhibitor), NC-shRNA, MAP17 shRNA, SGLT2 shRNA, SGLT2 shRNA+NC-overexpression vector, and SGLT2 shRNA+MAP17 overexpression. After stable lentiviral transduction and selection, cells were used for subsequent experiments.

Figure 3.

Figure 3.

MAP17 (membrane-associated protein 17) silencing suppresses SGLT2 (sodium-glucose cotransporter 2) expression and glycolytic-inflammatory activation in macrophages. A, Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) assay for MAP17 and SGLT2 mRNA expression in bone marrow–derived macrophages (BMDMs) under glycolysis and SGLT2/MAP17 inhibition. B, Western blot (WB) assay for MAP17 and SGLT2 protein expression in BMDMs under glycolysis and MAP17/SGLT2 inhibition. C, Cell counting kit-8 (CCK-8) assay for survival rate of BMDMs under glycolysis and MAP17/SGLT2 inhibition. D through G, Detection of glycolysis-related indicators in BMDMs under glycolysis and MAP17/SGLT2 inhibition. H through K, Detection of BMDM M1- and M2-like polarization-related markers in BMDMs under glycolysis and MAP17/SGLT2 inhibition. L, Static adhesion assay of BMDMs under glycolysis and MAP17/SGLT2 inhibition. M, Oil Red O (ORO) staining for lipid secretion of BMDMs under glycolysis and MAP17/SGLT2 inhibition. Data are presented as mean±SD from 3 independent experiments. Statistical significance was analyzed using 1-way ANOVA followed by the Tukey post hoc multiple comparisons test. *P<0.01, **P<0.001, and ***P<0.0001.

Figure 4.

Figure 4.

MAP17 (membrane-associated protein 17) overexpression rescues SGLT2 (sodium-glucose cotransporter 2)–dependent glycolysis and inflammatory phenotype. A, Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) assay for MAP17 and SGLT2 mRNA expression in SGLT2-knockdown bone marrow–derived macrophages (BMDMs) after MAP17 overexpression. B, Western blot (WB) assay for MAP17 and SGLT2 protein expression in SGLT2-knockdown BMDMs after MAP17 overexpression. C, Binding domains between SGLT2 and MAP17 in humans and mice by structural bioinformatics via STRING. D, Co-immunoprecipitation (CO-IP) for binding detection between MAP17 and SGLT2. E, Cell counting kit-8 (CCK-8) assay for survival rate of SGLT2-knockdown BMDMs after MAP17 overexpression. F through I, Detection of glycolysis-related indicators in SGLT2-knockdown BMDMs after MAP17 overexpression. J through M, Detection of M1- and M2-like polarization-related markers in SGLT2-knockdown BMDMs after MAP17 overexpression. N, Static adhesion assay of SGLT2-knockdown BMDMs after MAP17 overexpression. O, Oil Red O (ORO) staining for lipid secretion of SGLT2-knockdown BMDMs after MAP17 overexpression. Data are presented as mean±SD from 3 independent experiments. Statistical significance was compared using the independent Student t test. *P<0.05, **P<0.01, and ***P<0.001.

qRT-PCR Assay

Total RNA was extracted from cells and tissues using TRIzol reagent and then reverse-transcribed with a cDNA reverse transcription kit. β-actin served as an internal reference for SGLT2 and MAP17 (Table 1). A standard SYBR-Green RT-PCR kit (A25742, Thermo Fisher Scientific) was used to detect mRNA expression, and the 2−ΔΔCt method was applied to calculate the multiple ratio relationship between experimental and control groups.

Table 1.

qRT-PCR Primer Sequences

graphic file with name atv-46-e324585-g003.webp

Western Blot Assay

Total protein was extracted from tissues using radioimmunoprecipitation assay lysis buffer. Lysates were centrifuged at 4 °C for 15 minutes at 12 000 rpm. After collecting the supernatant, protein concentration was determined using a bicinchoninic acid assay kit (23225, Thermo Fisher Scientific). Next, the proteins were separated by acrylamide gel electrophoresis and transferred onto polyvinylidene fluoride membranes. After blocking with 5% skim milk (1706404, Bio-Rad Laboratories), incubation was conducted with primary antibodies against MAP17, SGLT2, and β-actin overnight at 4 °C and then with HRP-conjugated secondary antibody. Finally, visual detection of immunoreactive proteins was performed using an enhanced chemiluminescence assay kit and an image analyzer ImageQuant LAS 4000 (GE Healthcare, Abingdon, United Kingdom). ImageJ (V1.8.0.112, National Institutes of Health) was utilized for the analysis of grayscale values in the images.

Co-Immunoprecipitation

Cells were cultured and treated according to standard procedures. After collection, the cells were lysed, and total protein was extracted and quantified using the bicinchoninic acid assay kit. IP was performed following the instructions of the IP kit (88803, Thermo Fisher Scientific), wherein target-specific primary antibodies were used to immunoprecipitate the total protein extracts. After electrophoresis and membrane transfer of the resulting immune complexes, immunoblotting was conducted with corresponding secondary antibodies (anti-SGLT2, anti-MAP17) to detect binding.

Cell Viability Assay

Cells were seeded in 96-well plates at a density of 1×103 cells/well. After 72 hours of culture, 10-µL cell counting kit-8 reagent (5 mg/mL) was added to each well and incubated for 4 hours. After the supernatant was discarded, 100-µL dimethyl sulfoxide was added. The optical density was measured at a wavelength of 590 nm. The survival rate of the cells was calculated.

Real-Time ATP Metabolism Profiling

The Seahorse XFe 96 Real-Time ATP Rate Assay kit was used for extracellular flux analysis on BMDMs (8×104 cells/well; 5 replicates per biological sample according to the manufacturer’s instructions). Media were injected with different concentrations of glucose or control. After 7 measurements, oligomycin, 2-deoxy-glucose, FCCP, antimycin A, and rotenone from the XFp Real-Time ATP Rate Assay kit were sequentially injected for subsequent measurements. Glycolytic ATP (glycoATP) production rate and mitochondrial ATP production rate were automatically calculated using Wave Desktop software (Agilent version 2.6).

Macrophage Polarization Assay

Adherent cells were harvested and centrifuged (300×g–400×g, 5 minutes). After supernatant removal, the cells were resuspended in staining buffer and then washed 2× to 3×. Subsequently, the cells were incubated with an antibody blocking agent (anti-CD16/32 antibody) for 10 to 15 minutes to block nonspecific binding. Later, the cells were divided into groups for M1- and M2-like marker detection. M1-like marker antibodies (F4/80-FITC, iNOS-PE) and M2-like marker antibodies (F4/80-FITC, CD206-PE) were added, and the cell-antibody mixtures were gently mixed and incubated on ice in the dark for 30 to 60 minutes. After incubation, the cells were washed 2× to 3× with staining buffer to remove unbound antibodies, followed by centrifugation and supernatant removal. Afterward, the samples were loaded onto a flow cytometer (BD FACSCalibur, Becton, Dickinson and Company), and the data were analyzed using FlowJo software (FlowJo v10, Becton, Dickinson and Company). Finally, the proportions of F4/80+/iNOS+ (M1-like macrophage) and F4/80+/CD206+ (M2-like macrophage) were calculated by setting appropriate gates.

Inflammatory Factor Quantification

ELISA was used to measure proinflammatory M1-like cytokines, TNF-α (F0121), IL-1β (F2040), IL-6 (F2163), IL-12 (F30084), and IL-23 (F2018), and anti-inflammatory M2-like mediators, IL-10 (F2176) and TGF-β (transforming growth factor-β; F2686). The procedures were performed strictly according to the kit instructions. Briefly, the samples to be tested and controls were loaded into antigen-precoated microplates, sequentially followed by initial washing to remove unbound substances, addition of detection antibodies, secondary washing, substrate solution application for chromogenic development, and addition of stop solution to terminate the reaction. Finally, optical density was measured at 450 nm. The entire experiment required about 1.5 to 2 hours for execution.

Static Adhesion Test

Primary mouse dermal endothelial cells were cultured to confluence on glass coverslips in 12-well plates. Monolayers were activated with 10-ng/mL TNF-α (T0157, Sigma-Aldrich) for 16 hours. In addition, BMDMs from each group were labeled with green fluorescence using the PKH67 fluorescent cell linker kit according to the manufacturer’s instructions. Then, the labeled BMDMs were added to TNF-α-activated endothelial monolayers at a density of 5×104 cells per coverslip for 1-hour coculture. Afterward, the medium and nonadherent cells were removed via washes with PBS (P5493, Sigma-Aldrich), and adherent complexes were fixed in 4% paraformaldehyde. At last, 4 images per sample were acquired from each coverslip, and the number of adherent fluorescently labeled cells was counted and averaged. All displayed images were uniformly adjusted for brightness using ImageJ software (version 1.8.0.112, National Institutes of Health).

Quantitative Assessment of Macrophage Foam Cell Formation

Macrophages were incubated with Dil-acetylated LDL at a concentration of 25 µg/mL for 48 hours. After removing the culture medium and residual Dil-acetylated LDL, the cells were washed 3× with PBS and fixed in 4% paraformaldehyde. Lipid droplets were subsequently stained with ORO, and nuclei were counterstained with DAPI (4′,6-diamidino-2-phenylindole; D9542, Sigma-Aldrich). Phase-contrast images of ORO staining and fluorescent images of cell nuclei were captured (4 images per coverslip), and lipid content and cell count were quantified using ImageJ software. The lipid area was then normalized to the number of cells in each corresponding image, and the results were expressed as fold change over the control group in the experimental groups. The brightness of the displayed images was uniformly adjusted.

Animal Modeling and Grouping

Sixty male apolipoprotein E knockout (Apoe−/−) mice on a C57BL/6J background (18–20 g) were purchased from Kaixue Biotechnology (Shanghai, China). Age-matched male-specific pathogen-free-grade C57BL/6J mice were used as controls where appropriate. All mice were housed in a specific pathogen-free animal facility under controlled environmental conditions (temperature: 22±2 °C; relative humidity: 55±5%; 12-hour light/dark cycle), with free access to food and water. Mice were housed at 4 to 5 animals per cage with standard environmental enrichment. During the acclimatization period, mice were fed a standard laboratory diet (MD17122), mainly composed of corn flour, wheat flour, bran, soybean meal, soybean oil, fish meal, chicken meal, and vitamin-mineral premix. For atherosclerosis modeling, mice were switched to an atherogenic high-fat/high-cholesterol diet (D12108C, Research Diets, Inc, United States; equivalent domestic diet code: JKHFHC008), containing 20% protein, 40% carbohydrate, 40% fat, and 1.25% cholesterol. Animal modeling19: all mice used in the atherosclerosis and BM transplantation models were male. This approach was adopted to minimize the confounding influence of endogenous estrogens on vascular inflammation and lipid metabolism. BM was collected from donor mice and transduced ex vivo with lentiviral vectors carrying MAP17 shRNA, SGLT2 shRNA, MAP17 overexpression constructs, or corresponding negative controls according to the manufacturer’s instructions before transplantation. Before transplantation, successful lentiviral transduction of BM cells was monitored by GFP (green fluorescent protein)/ZsGreen fluorescence microscopy, and cells with detectable fluorescence signals were used for subsequent transplantation (Figure S1). Considering the limited tolerance of primary hematopoietic cells to antibiotic selection, puromycin selection was not applied to BM cells before transplantation to avoid affecting cell viability and engraftment capacity. Subsequently, 5×106 viable BM cells were intravenously injected into lethally irradiated recipient Apoe−/− mice (2×500 rads). Following transplantation, all recipient mice were placed on a high-fat diet for 12 weeks for atherosclerosis modeling, with untreated Apoe−/− mice serving as controls. Groups were set as follows: Apoe−/− (n=6), control donor (n=6), diabetic donor (n=12), dichloroacetate (n=6), NC-shRNA (n=6), MAP17 shRNA (n=6), SGLT2 shRNA (n=6), SGLT2 shRNA+NC-oeRNA (n=6), and SGLT2 shRNA+MAP17-oeRNA (n=6). Animal experiments were approved by the Institutional Animal Care and Use Committee of our hospital (approval number: DL20220426) and were conducted in accordance with relevant animal welfare regulations.

Terminal Sample Collection and Model Validation

At 12 weeks, the mice were euthanized with an overdose of anesthetic (50-mg/mL sodium pentobarbital), and whole blood was collected. Serum was analyzed for triglycerides, nonesterified fatty acids, LDL-C (LDL-cholesterol), and HDL-C (high-density lipoprotein-cholesterol) levels. The expression of SGLT2 and MAP17 in serum was determined by qRT-PCR and Western blot. Mice were perfused with PBS, followed by overnight fixation with 4% paraformaldehyde. The aortas were subsequently isolated and processed in 70% ethanol. For serum mRNA detection, blood samples were centrifuged at 4000 rpm for 10 minutes at 4 °C within 1 hour of collection. Total RNA was extracted from the serum using TRIzol LS Reagent (Invitrogen) under strictly RNase-free conditions to minimize degradation, followed by immediate reverse transcription. Samples were stored at −80 °C and avoided repeated freeze-thaw cycles.

Glycolytic, Inflammatory, and Adhesive Profiling

Lactate concentrations were determined using a lactate assay kit (MAK064, Sigma-Aldrich) according to the manufacturer’s instructions. Glucose concentrations were measured using a glucose assay kit (GAGO20, Sigma-Aldrich) according to the manufacturer’s instructions. Enzyme activity assay kits (Sigma-Aldrich) were used to quantify the activities of key glycolytic enzymes HK (hexokinase; MAK091), PFK (phosphofructokinase; MAK093), and PK (pyruvate kinase; MAK072). ELISA was used to assess the expression of inflammatory cytokines (TNF-α, IL-1β, IL-6, IL-12, IL-23, IL-10, and TGF-β), as well as the serum levels of adhesion molecules MCP-1 (monocyte chemoattractant protein-1), ICAM-1 (intercellular adhesion molecule-1), and VCAM-1 (vascular cell adhesion molecule-1).

Hematoxylin and Eosin Staining

Aortic root tissues were sliced into 10-μm sections, fixed with 4% paraformaldehyde for 30 minutes, and washed with PBS 3× for 5 minutes each. They were then stained with hematoxylin for 4 minutes, washed with PBS 3× for 5 minutes each, and observed under a microscope (CX23, Olympus) to assess the intensity of nuclear staining. Based on this assessment, the sections were differentiated in hydrochloric acid-alcohol for several seconds, followed by immediate rinsing with tap water for 5 minutes after differentiation. Afterward, they were counterstained with eosin for 25 seconds, dehydrated 3× through an ethanol series, each time for 5 minutes, cleared 3× in xylene for 10 minutes each, and finally mounted with neutral resin for overnight drying. Representative sections were selected based on consistent anatomic landmarks of the aortic root, including the appearance of the 3 aortic valve cusps. Three random fields were selected, observed under a microscope at 40× magnification, and photographed to record the staining results.

ORO Staining

After the entire aortas were separated and fixed with 4% paraformaldehyde, periadventitial adipose tissue was meticulously removed under a microscope, and the aortas were longitudinally incised along the ventral midline with ophthalmic scissors. An ORO dye solution was prepared with the ratio of A1: A2=3: 2, and the opened aortas were immersed in the ORO dye solution for 15 minutes away from light. After staining, the aortas were rinsed with 60% isopropyl alcohol solution and 75% ethanol solution sequentially until the whole aortas were milky white. Then, the aortic root tissues were cut into 10-μm sections, and frozen sections were fixed with 4% paraformaldehyde for 15 minutes, followed by ORO staining, observation under a microscope, and photographing. The ORO-positive lipid deposition area was manually delineated for quantitative analysis, and the corresponding regions were outlined in representative images.

Masson Trichrome Staining

To quantify atherosclerotic plaque burden, the aortic roots fixed in paraformaldehyde (P6148, Sigma-Aldrich) were embedded in optimal cutting temperature compound and sectioned perpendicular to the aortic valve plane. Serial cryosections of the aortic root were prepared at a thickness of 10 µm, beginning from the level where all 3 aortic valve cusps became clearly visible. Sequential sections were collected at fixed intervals of 50 (Figure 5) or 80 µm (Figures 6 and 7). Sections with comparable anatomic landmarks, including the 3 aortic valve cusps and the maximal cross-sectional area of the aortic root, were selected for analysis across all mice to ensure consistent anatomic positioning among groups.

Figure 5.

Figure 5.

Diabetic bone marrow (BM)–derived trained immunity promotes MAP17 (membrane-associated protein 17)/SGLT2 (sodium-glucose cotransporter 2)–driven glycolysis and accelerates atherosclerosis (AS) progression. A, Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) assay for serum MAP17 and SGLT2 mRNA expression in diabetic BM-derived AS mice. B, Western blot (WB) assay for serum MAP17 and SGLT2 protein expression in diabetic BM-derived AS mice. C and D, Detection of glycolysis-related indicators in diabetic BM-derived AS mice. E, Detection of inflammatory cytokine levels in diabetic BM-derived AS mice. F, Detection of adhesion molecule levels in diabetic BM-derived AS mice. G, Hematoxylin and eosin (HE) staining of aortic root tissues from diabetic BM-derived AS mice. H, Oil Red O (ORO) staining of aortic root sections from diabetic BM-derived AS mice. I, ORO staining of entire aortas from diabetic BM-derived AS mice. J, Masson trichrome staining of aortic roots from diabetic BM-derived AS mice. K, Necrotic core area: quantification of the necrotic core size (×104 μm) within the aortic root sections (corresponding to G). L, Plaque area: quantification of the total ORO-positive plaque area (×104 μm2) in the aortic root (corresponding to H). Plaque composition and stability: (M) area ratio of plaque to vessel (%; corresponding to I) and (N) collagen volume fraction (%) quantified from Masson trichrome–stained sections (corresponding to J). Contour outlines indicate the regions used for quantitative analysis of plaque lesion area and necrotic core area. Data are presented as mean±SD (n=6 biologically independent mice per group) and were analyzed using 1-way ANOVA followed by the Tukey post hoc multiple comparisons test. Statistical significance is indicated as ns (nonsignificant; P>0.05); *P<0.01, **P<0.001, and ***P<0.0001. Quantitative analysis was performed on at least 3 representative sequential sections per mouse to ensure accuracy.

Figure 6.

Figure 6.

Targeting MAP17 (membrane-associated protein 17) significantly reduces glycolysis and inflammatory lesions in atherosclerosis (AS) mice. A, Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) assay for serum MAP17 and SGLT2 (sodium-glucose cotransporter 2) mRNA expression in diabetic bone marrow (BM)–derived AS mice after glycolysis and MAP17/SGLT2 inhibition. B, Western blot (WB) assay for serum MAP17 and SGLT2 protein expression in diabetic BM-derived AS mice after glycolysis and MAP17/SGLT2 inhibition. C, Detection of glycolysis-related indicators in diabetic BM-derived AS mice after glycolysis and MAP17/SGLT2 inhibition. D, Detection of inflammatory cytokine levels in diabetic BM-derived AS mice after glycolysis and MAP17/SGLT2 inhibition. E, Detection of adhesion molecule levels in diabetic BM-derived AS mice after glycolysis and MAP17/SGLT2 inhibition. F, Hematoxylin and eosin (HE) staining of aortic root tissues from diabetic BM-derived AS mice after glycolysis and MAP17/SGLT2 inhibition. G, Oil Red O (ORO) staining of aortic root sections from diabetic BM-derived AS mice after glycolysis and MAP17/SGLT2 inhibition. H, Masson trichrome staining of aortic roots from diabetic BM-derived AS mice after glycolysis and MAP17/SGLT2 inhibition. I, ORO staining of entire aortas from diabetic BM-derived AS mice after glycolysis and MAP17/SGLT2 inhibition. J, Necrotic core area: quantification of the necrotic core size (×104 μm) within the aortic root sections (corresponding to F). K, Plaque area: quantification of the total ORO-positive plaque area (×104 μm2) in the aortic root (corresponding to G). Plaque composition and stability: (L) area ratio of plaque to vessel (%; corresponding to I) and (M) collagen volume fraction (%) quantified from Masson trichrome–stained sections (corresponding to H). Contour outlines indicate the regions used for quantitative analysis of plaque lesion area and necrotic core area. Data are presented as mean±SD (n=6 biologically independent mice per group) and were analyzed using 1-way ANOVA followed by the Tukey post hoc multiple comparisons test. Statistical significance is indicated as ns (nonsignificant; P>0.05); *P<0.01, **P<0.001, and ***P<0.0001. Quantitative analysis was performed on at least 3 representative sequential sections per mouse to ensure accuracy.

Figure 7.

Figure 7.

MAP17 (membrane-associated protein 17) overexpression overrides SGLT2 (sodium-glucose cotransporter 2) knockdown effects, reinstating glycolysis and atherosclerosis (AS) progression. A, Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) assay for serum MAP17 and SGLT2 mRNA expression in diabetic bone marrow (BM)–derived AS mice after overexpressing MAP17 post-SGLT2 inhibition. B, Western blot (WB) assay for serum MAP17 and SGLT2 protein expression in diabetic BM-derived AS mice after overexpressing MAP17 post-SGLT2 inhibition. C, Detection of glycolysis-related indicators in diabetic BM-derived AS mice after overexpressing MAP17 post-SGLT2 inhibition. D, Detection of inflammatory cytokine levels in diabetic BM-derived AS mice after overexpressing MAP17 post-SGLT2 inhibition. E, Detection of adhesion molecule levels in diabetic BM-derived AS mice after overexpressing MAP17 post-SGLT2 inhibition. F, Hematoxylin and eosin (HE) staining of aortic root tissues from diabetic BM-derived AS mice after overexpressing MAP17 post-SGLT2 inhibition. G, Oil Red O (ORO) staining of aortic root sections from diabetic BM-derived AS mice after overexpressing MAP17 post-SGLT2 inhibition. H, ORO staining of entire aortas from diabetic BM-derived AS mice after overexpressing MAP17 post-SGLT2 inhibition. I, Masson trichrome staining of aortic roots from diabetic BM-derived AS mice after overexpressing MAP17 post-SGLT2 inhibition. J, Necrotic core area: quantification of the necrotic core size (×104 μm) within the aortic root sections (corresponding to F). K, Plaque area: quantification of the total ORO-positive plaque area (×104 μm2) in the aortic root (corresponding to G). Plaque composition and stability: (L) area ratio of plaque to vessel (%; corresponding to H) and (M) collagen volume fraction (%) quantified from Masson trichrome–stained sections (corresponding to I). Contour outlines indicate the regions used for quantitative analysis of plaque lesion area and necrotic core area. Data are presented as mean±SD (n=6 biologically independent mice per group) and were compared using the independent Student t test. Statistical significance is indicated as ns (nonsignificant; P>0.05); *P<0.01, **P<0.001, and ***P<0.0001. Quantitative analysis was performed on at least 3 representative sequential sections per mouse to ensure accuracy.

Three evenly spaced serial sections from the aortic root were selected for histological analysis. Masson trichrome staining was used to assess collagen deposition within atherosclerotic plaques. In brief, sections were refixed in Bouin solution at room temperature for 1 hour, rinsed in running tap water for 5 minutes, and then stained according to the manufacturer’s instructions, with aniline blue used to visualize collagen fibers. Subsequently, the sections were mounted with Neo-Mount mounting medium on xylene-wetted slides and sealed with coverslips. Plaque morphological characteristics were quantitatively analyzed using Image-Pro Plus software, version 6.0 (Media Cybernetics, Silver Spring, MD). Total plaque area, lipid-rich area, and collagen-positive area were measured, and collagen content was calculated as the percentage of collagen-positive area within the total plaque area. The mean value from the 3 sections was used for analysis for each mouse.

For quantitative analysis, at least 3 representative sections from each mouse were analyzed using ImageJ software, and the mean value of multiple sections was used for statistical analysis to minimize regional variation. The plaque lesion area was defined as the intimal area occupied by atherosclerotic lesions. Necrotic core areas were defined as acellular regions within the plaques and were manually delineated based on morphological characteristics. The regions used for plaque lesion area and necrotic core quantification were outlined in representative images to illustrate the criteria for measurement.

DNA Methylation Assay

Total DNA was extracted from trained macrophages and BM progenitors. Targeted DNA methylation analysis of the promoters of proinflammatory (TNF-α, IL-1β, IL-6, IL-12, and IL-23) and anti-inflammatory (IL-10 and TGF-β) genes was performed using bisulfite treatment followed by methylation-specific polymerase chain reaction or bisulfite-sequencing polymerase chain reaction to assess the epigenetic imprinting associated with trained immunity.

Chromatin Immunoprecipitation Assay

Chromatin immunoprecipitation was performed to evaluate histone modifications. Briefly, cells were cross-linked with 1% formaldehyde and lysed. Chromatin was sheared by sonication and immunoprecipitated using antibodies against H3K4me3 (trimethylation of histone H3 at lysine 4), H3K27ac (acetylation of histone H3 at lysine 27; activating marks), and H3K9me3 (trimethylation of histone H3 at lysine 9; repressive mark). The enrichment of these modifications at the promoters of target cytokine genes was quantified by qPCR (quantitative polymerase chain reaction).

Statistical Analysis

The data were analyzed using SPSS 23 (IBM Corporation), and statistical graphs were generated using GraphPad Prism 8.0 (GraphPad Software, Inc). Percentages were used to represent categorical data, while mean±SD was used for continuous data. The normality of continuous variables was assessed using the Shapiro-Wilk test, and homogeneity of variance was evaluated using the Levene test. For data satisfying both normality and homogeneity of variance assumptions, the Student t test was used for comparisons between 2 groups, and 1-way ANOVA followed by the Tukey post hoc multiple comparisons test was performed for comparisons among >2 groups. When homogeneity of variance was violated, Welch correction was applied as appropriate. For nonnormally distributed data, nonparametric tests, including the Mann-Whitney U test or the Kruskal-Wallis test, were used. The correlation between variables was assessed using the Pearson correlation analysis. Statistical significance is indicated as ns (nonsignificant; P>0.05); *P<0.05; **P<0.01; ***P<0.001, and ****P<0.0001.

Results

MAP17 Is Upregulated in Atherosclerosis and Positively Correlates With SGLT2 Expression

Previous studies have implicated MAP17 and its downstream effector SGLT2 in the pathogenesis of atherosclerosis.11 To validate this, we first examined their expression in clinical atherosclerosis samples. qRT-PCR and Western blot analyses revealed that MAP17 was significantly upregulated in the serum of patients with AS, accompanied by a parallel increase in SGLT2 expression (n=30; Figure 1A through 1D). The Pearson correlation analysis further demonstrated a strong positive association between MAP17 and SGLT2 expression levels (Figure 1E), suggesting that MAP17 may transcriptionally or functionally regulate SGLT2. This coordinated upregulation indicates that MAP17 likely serves as an upstream activator that potentiates SGLT2 expression, thereby contributing to atherosclerosis progression through a synergistic immunometabolic mechanism.

Figure 1.

Figure 1.

Expression of SGLT2 (sodium-glucose cotransporter 2)/MAP17 (membrane-associated protein 17) and their correlation in atherosclerosis (AS). A and B, Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) assay for serum MAP17 and SGLT2 mRNA expression in patients with AS. C and D, Western blot (WB) assay for serum MAP17 and SGLT2 protein expression in patients with AS. E, The Pearson correlation analysis for correlation between MAP17 and SGLT2 in patients with AS. Data were compared using the independent Student t test. ***P<0.0001.

Diabetes History, Metabolic Syndrome, and Microinflammation Are Factors Influencing Atherosclerosis Progression

To systematically identify factors affecting atherosclerosis progression, we conducted a stratified cohort analysis of 30 patients with AS categorized by disease severity using the Gensini scoring system into mild-to-moderate and severe groups. Emerging evidence has demonstrated that microinflammatory response is an early event in atherosclerosis occurrence, wherein endothelial cell injury and dysfunction can trigger an inflammatory response, attracting monocytes and other immune cells to the injury site.20 This motivated our investigation into the relationship between inflammatory response and disease progression; thus, we detected the expression of inflammatory factors TNF-α, IFN-γ, and IL-4 in plaque tissues from both groups by ELISA. Univariate logistic regression models incorporating diabetes history, metabolic syndrome, and inflammatory biomarkers identified a history of diabetes, metabolic syndrome, and high levels of TNF-α, IFN-γ, and IL-4 as the factors influencing atherosclerosis progression (Table 2). This inflammatory signature supports that microinflammatory response is a driver of the occurrence and development of atherosclerosis. While HbA1c levels were comparable between the 2 groups, likely reflecting standardized long-term glucose-lowering therapy, fasting plasma glucose levels were significantly higher in the severe atherosclerosis group (P=0.045; Table 2).

Table 2.

Factors Influencing AS Progression by Univariate Analysis

graphic file with name atv-46-e324585-g009.webp

Hyperglycemia-Induced MAP17 Activation Drives Macrophage Glycolysis and Trained Immunity via SGLT2

Clinical profiling revealed that patients with diabetes and atherosclerosis exhibited elevated expression of MAP17, accompanied by increased SGLT2 levels and worsened disease severity. To investigate the mechanistic basis of these observations, we established an in vitro metabolic memory model using BMDMs stimulated under diabetic-mimicking conditions (high glucose+IFN-γ+lipopolysaccharide). Notably, high glucose exposure significantly upregulated MAP17 expression, which was followed by increased SGLT2 expression at both mRNA and protein levels, along with reduced cell viability. These effects were further amplified upon costimulation with IFN-γ and lipopolysaccharide (Figure 2A through 2C).

Figure 2.

Figure 2.

Hyperglycemia-induced MAP17 (membrane-associated protein 17) activation drives macrophage glycolysis and trained immunity via SGLT2 (sodium-glucose cotransporter 2). A, Quantitative reverse transcription-polymerase chain reaction (qRT-PCR) assay for MAP17 and SGLT2 mRNA expression in bone marrow–derived macrophages (BMDMs) under hyperglycemia and IFN-γ (interferon-γ)+lipopolysaccharide (LPS) stimulation. B, Western blot (WB) assay for MAP17 and SGLT2 protein expression in BMDMs under hyperglycemia and IFN-γ+LPS stimulation. C, Cell counting kit-8 (CCK-8) assay for survival rate of BMDMs under hyperglycemia and IFN-γ+LPS stimulation. D through G, Detection of glycolysis-related indicators in BMDMs under hyperglycemia and IFN-γ+LPS stimulation. H through K, Detection of BMDM M1- and M2-like polarization-related markers under hyperglycemia and IFN-γ+LPS stimulation. L, Static adhesion assay of BMDMs under hyperglycemia and IFN-γ+LPS stimulation. M, Oil Red O (ORO) staining for lipid secretion of BMDMs under hyperglycemia and IFN-γ+LPS stimulation. Data are presented as mean±SD from 3 independent experiments. Statistical significance was analyzed using 1-way ANOVA followed by the Tukey post hoc multiple comparisons test. *P<0.01, **P<0.001, and ***P<0.0001.

These results suggest that MAP17 may act as a glucose-responsive upstream regulator, initiating a cascade that activates SGLT2 and promotes macrophage metabolic reprogramming. To further validate this hypothesis, we comprehensively assessed cellular bioenergetics and immune function in BMDMs. Metabolic analysis showed a pronounced shift toward glycolysis under high glucose, as indicated by an elevated extracellular acidification rate and increased glycoATP production, coupled with a decreased oxygen consumption rate and diminished mitochondrial ATP output. These metabolic changes were significantly enhanced by IFN-γ+lipopolysaccharide stimulation (Figure 2D through 2G). Functionally, high glucose combined with inflammatory stimulation drove robust M1-like polarization of macrophages, evidenced by increased expression of iNOS and proinflammatory cytokines (TNF-α, IL-1β, IL-6, IL-12, and IL-23), alongside decreased CD206 expression and lower levels of anti-inflammatory mediators (IL-10 and TGF-β; Figure 2H through 2K). Furthermore, static adhesion capacity and lipid uptake were markedly increased under dual stimulation, indicating enhanced foam cell formation potential (Figure 2L and 2M). Collectively, these findings demonstrate that hyperglycemia-induced MAP17 upregulation serves as a key initiator of SGLT2 activation, promoting glycolytic flux, proinflammatory polarization, and macrophage-driven atherogenesis, thereby linking metabolic memory with trained immunity in atherosclerosis progression.

Bisulfite polymerase chain reaction and chromatin immunoprecipitation-qPCR analyses demonstrated that high-glucose-primed macrophages exhibited significant DNA hypomethylation at the promoters of proinflammatory genes (TNF-α and IL-6) and increased activating histone marks (H3K4me3 and H3K27ac). These epigenetic imprints persisted in BMDMs from diabetic donors even after 7 days of normoglycemic differentiation (Figures S3 through S7).

MAP17 Silencing Suppresses SGLT2 Expression and Glycolytic-Inflammatory Activation in Macrophages

To further elucidate the role of MAP17 as an upstream regulator of glycolytic reprogramming, we investigated the effects of inhibiting MAP17, SGLT2, and glycolysis in BMDMs under high-glucose and IFN-γ+lipopolysaccharide stimulation (Figure 3A and 3B). Notably, MAP17 knockdown significantly suppressed SGLT2 expression, whereas SGLT2 knockdown had no effect on MAP17 levels, indicating a unidirectional regulatory relationship in which MAP17 positively regulates SGLT2 without reciprocal feedback. This finding positions MAP17 as a potential upstream activator within the glycolytic-inflammatory axis of trained immunity. Metabolic profiling further demonstrated that inhibition of glycolysis using dichloroacetate restored cell viability, characterized by reduced extracellular acidification rate and glycoATP production along with increased oxygen consumption rate and mitochondrial ATP generation (Figure 3C). Similar metabolic restoration was observed upon MAP17 or SGLT2 silencing, both of which effectively reversed the glycolytic shift induced by inflammatory stimulation (Figure 3D through 3G), suggesting that both molecules contribute to glycolytic activation in macrophages, with MAP17 acting as a primary initiator.

In parallel, immunophenotyping revealed that dichloroacetate, MAP17 shRNA, and SGLT2 shRNA all suppressed M1-like polarization and associated proinflammatory cytokines (iNOS, TNF-ɑ, IL-1β, IL-6, IL-12, and IL-23) while concurrently enhancing M2 polarization and anti-inflammatory markers (CD206, IL-10, and TGF-β; Figure 3H through 3K). Among them, SGLT2 knockdown exhibited the most potent effect on M1-like macrophage suppression, potentially reflecting its direct role in metabolic flux regulation, although MAP17 inhibition served as the upstream trigger. Finally, histopathologic assessment showed that all 3 interventions with dichloroacetate, MAP17 shRNA, and SGLT2 shRNA ameliorated atherosclerosis lesion-associated features, including inflammatory infiltration and foam cell accumulation. A distinct therapeutic gradient was observed, with SGLT2 knockdown demonstrating the strongest lesion attenuation, followed by dichloroacetate and MAP17 silencing (Figure 3L and 3M). These findings collectively support a mechanistic model in which MAP17 initiates SGLT2-dependent glycolytic activation, thereby driving proinflammatory macrophage programming and accelerating atherosclerotic lesion progression.

MAP17 Overexpression Rescues SGLT2-Dependent Glycolysis and Inflammatory Phenotype

Building upon the previously established unidirectional regulatory relationship, we next examined whether MAP17 overexpression could functionally rescue SGLT2 expression and its downstream effects. In SGLT2-silenced BMDMs, validation of silencing efficiency showed that SGLT2 shRNA achieved a 60% to 70% reduction in expression (Figure S8). MAP17 overexpression resulted in a partial rescue of the glycolytic phenotype by upregulating the residual SGLT2 fraction (Figure 4A and 4B), reinforcing the concept that MAP17 acts as a dominant upstream regulator of SGLT2. To further explore the mechanistic basis of this regulation, structural bioinformatics analysis using the STRING database (https://cn.string-db.org/) identified evolutionarily conserved binding domains between MAP17 and SGLT2 across both human and mouse species. This predicted interaction was experimentally validated by co-immunoprecipitation, which confirmed direct physical association (Figure 4C and 4D). These results suggest that MAP17 may regulate SGLT2 through protein-protein interactions, stabilizing or promoting SGLT2 expression and activity. Functionally, MAP17 overexpression reversed the impaired glycolytic profile seen in SGLT2-deficient cells, restoring extracellular acidification rate, glycoATP production, and mitochondrial respiration (Figure 4E through 4I). Immunophenotyping and metabolic readouts further demonstrated a shift from proinflammatory to balanced polarization states, with decreased M1-like macrophage markers and increased M2-like macrophage-associated factors (Figure 4J through 4M). In addition, MAP17 reexpression improved BMDM adhesive capacity and lipid-handling behavior, evidenced by reduced static adhesion and foam cell formation (Figure 4N and 4O). Collectively, these findings confirm that MAP17 not only regulates SGLT2 at the expression level but also functionally reinstates SGLT2-dependent immunometabolic activity, underscoring its upstream and central role in modulating glycolysis and macrophage-mediated atherosclerosis progression.

Diabetic BM-Derived Trained Immunity Promotes MAP17/SGLT2 Activation and Accelerates Atherosclerosis Progression

To validate the in vitro findings of MAP17-driven immunometabolic activation, we next used a diabetic BM transplantation model using Apoe−/− recipient mice. Following successful engraftment, we assessed serum expression of MAP17 and SGLT2 across experimental groups. Notably, mice receiving BM from diabetic donors exhibited markedly elevated MAP17 mRNA and protein levels, accompanied by a parallel increase in SGLT2 expression, compared with both nondiabetic donor and wild-type Apoe−/− controls (Figure 5A and 5B). These findings suggest that hyperglycemia-induced metabolic memory in donor BM cells triggers activation of the MAP17/SGLT2 axis in recipient macrophages.

To further investigate whether this memory inherited from BM progenitors confers a stable functional switch, we performed extracellular flux analysis on BMDMs differentiated in vitro for 7 days in a normoglycemic environment. Notably, BMDMs derived from diabetic donors maintained a pronounced trained metabolic phenotype, characterized by a significantly higher extracellular acidification rate compared with control groups (Figure S8A and S8B). Functional quantification confirmed a robust shift in ATP production, with a substantial increase in glycoATP and a concomitant reduction in mitochondrial ATP production (P<0.001), proving that the hyperglycemic imprinting results in a persistent and inheritable metabolic reprogramming (Figure S8C and S8D).

Metabolic profiling further revealed a pronounced glycolytic shift in the diabetic BM group, with significantly increased serum levels of lactate, glucose, and key glycolytic enzymes (HK, PFK, and PK), indicating sustained glycolytic reprogramming (Figure 5C and 5D). This was accompanied by an immunologic phenotype consistent with trained immunity, characterized by heightened secretion of proinflammatory cytokines (TNF-α, IL-1β, IL-6, IL-12, and IL-23) and suppression of anti-inflammatory mediators (IL-10 and TGF-β; Figure 5E). In addition, the diabetic BM group exhibited increased levels of adhesion molecules (MCP-1, ICAM-1, and VCAM-1), further reflecting enhanced monocyte recruitment and endothelial activation (Figure 5F). Histomorphometric analyses revealed an expanded necrotic core area with obvious inflammatory infiltration and significantly increased lipid deposition in both the aortic root and the entire aorta (Figure 5G, 5H, 5K, and 5L). Moreover, collagen accumulation in the aortic root was markedly elevated, indicating advanced plaque remodeling (Figure 5I, 5M, and 5N). Taken together, these in vivo results confirm that trained immunity induced by diabetic BM transplantation activates the MAP17/SGLT2 axis, which, in turn, drives glycolysis, inflammation, and exacerbates atherosclerotic lesion development, highlighting MAP17 as a key pathogenic effector in metabolically programmed vascular inflammation.

Targeting MAP17 Significantly Reduces Glycolysis and Inflammatory Pathology in Atherosclerosis Mice

To translate the cellular findings into an in vivo context, we evaluated the therapeutic impact of glycolysis-targeting interventions in atherosclerosis mice receiving diabetic BM transplantation. Mice were divided into 5 groups: diabetic donor, dichloroacetate, NC-shRNA, MAP17 shRNA, and SGLT2 shRNA. As expected, inhibition of glycolysis and gene silencing led to decreased MAP17 and SGLT2 expression at both mRNA and protein levels in treated groups compared with diabetic donor controls (Figure 6A and 6B). Metabolic profiling revealed that all 3 interventions with dichloroacetate, MAP17 shRNA, and SGLT2 shRNA significantly reduced glycolytic activity, evidenced by decreased levels of serum lactate, glucose, and glycolytic enzyme activity (Figure 6C). These results are consistent with prior in vitro findings and suggest that MAP17 plays a regulatory role in maintaining the hyperglycolytic state in atherosclerosis, at least in part through activation of SGLT2. Importantly, immune and inflammatory assessments demonstrated that inhibition of MAP17 or SGLT2 expression significantly decreased proinflammatory cytokine production and lowered adhesion molecule levels (Figure 6D and 6E), indicating that MAP17-SGLT2 signaling contributes not only to metabolic remodeling but also to immune activation in atherosclerosis. Histopathologic analysis further showed that inhibition of MAP17 or SGLT2, or glycolytic blockade with dichloroacetate, ameliorated atherosclerotic plaque features, including reduced necrotic core area and attenuated inflammatory infiltration (Figure 6F). Moreover, vascular remodeling indicators such as aortic lipid accumulation and collagen deposition were also significantly decreased following MAP17 or SGLT2 suppression (Figure 6G through 6M). Together, these results confirm that targeting MAP17, either directly or via its downstream effector SGLT2, can effectively suppress glycolysis, mitigate inflammation, and improve vascular pathology in atherosclerosis. MAP17 emerges as a central upstream node, offering a promising therapeutic target for disrupting glycolysis-driven trained immunity in diabetic atherosclerosis.

MAP17 Overexpression Overrides SGLT2 Knockdown Effects, Reinstating Glycolysis, and Atherosclerosis Progression

To further confirm the upstream regulatory role of MAP17, we conducted in vivo rescue experiments by overexpressing MAP17 in SGLT2-silenced diabetic BM donor-derived atherosclerosis mice. Consistent with in vitro findings, MAP17 overexpression effectively reinstated SGLT2 expression at both mRNA and protein levels (Figure 7A and 7B), demonstrating its ability to override SGLT2 knockdown. This molecular restoration was accompanied by a significant rebound in glycolytic activity, including elevated lactate production and glycolytic enzyme levels (Figure 7C). In addition, proinflammatory cytokines (TNF-ɑ, IL-1β, IL-6, IL-12, and IL-23) and adhesion molecules (MCP-1, ICAM-1, and VCAM-1) were markedly upregulated (Figure 7D and 7E), suggesting enhanced inflammatory activation. Histopathologically, MAP17 overexpression led to enlarged necrotic core areas, increased inflammatory infiltration in the aortic root (Figure 7F), and aggravated lipid and collagen deposition (Figure 7G through 7M), reflecting worsened lesion severity. Collectively, these results confirm that MAP17 acts as a dominant driver of SGLT2 expression and function in vivo, and reactivation of the MAP17-SGLT2 axis reinstates glycolysis and inflammation, ultimately promoting atherosclerosis progression. This further underscores MAP17 as a crucial upstream regulator capable of controlling metabolic memory, immune polarization, and vascular pathology in diabetic atherosclerosis.

Discussion

Atherosclerosis is a chronic inflammatory disease underpinning cardiocerebrovascular morbidity and mortality,21 arising from the intricate interplay between metabolic dysregulation and immune activation. While traditional pathogenic models have focused on lipid metabolism disorders, vascular inflammation, and smooth muscle cell proliferation,22 our findings, together with emerging literature, identify MAP17 as a pivotal upstream regulator in the metabolic-immune crosstalk that accelerates atherosclerosis progression.23 MAP17, a small transmembrane protein minimally expressed under physiological conditions, is markedly upregulated in various pathological contexts, where it perturbs cellular redox balance, enhances ROS production, and reprograms metabolic pathways toward glycolysis.24,25 Such metabolic shifts are increasingly recognized as central to both endothelial dysfunction and macrophage-driven inflammation, providing a mechanistic bridge between metabolic stress and immune activation in vascular disease.

In this study, we provide the first systematic evidence linking MAP17 upregulation to atherosclerosis in both clinical and experimental settings. Clinical data revealed significantly elevated MAP17 expression in patients with AS, with a strong positive correlation to SGLT2 levels,26 as well as associations with diabetes, metabolic syndrome, and increased circulating proinflammatory cytokines (TNF-α, IFN-γ, and IL-4). These comorbid conditions are known accelerators of atherosclerosis, in part due to sustained microinflammatory states27 and hyperglycemia-induced endothelial injury and oxidative stress.28 Our stratified analysis suggests that MAP17 expression could serve as a biomarker reflecting a metabolically and immunologically primed vascular state, particularly in individuals with persistent metabolic stress. The stability of MAP17 and SGLT2 mRNA in serum is likely attributable to their encapsulation within extracellular vesicles, shielding them from endogenous RNase activity.

Within this pathogenic landscape, trained immunity emerges as a critical mechanistic link. Unlike vascular smooth muscle cells, macrophages possess high plasticity and the capacity for epigenetic and metabolic reprogramming in response to environmental cues. Trained immunity describes the phenomenon whereby innate immune cells, particularly monocytes/macrophages, acquire a heightened and persistent proinflammatory phenotype after an initial stimulus.29 Persistent hyperglycemia, as occurs in diabetes and metabolic syndrome, not only injures the endothelium but also programs monocytes/macrophages toward long-lasting inflammatory readiness through metabolic rewiring.30 Previous studies have documented trained immunity within atherosclerotic plaques under metabolic stress,8,31 with hyperglycemia acting as both a priming and sustaining factor.32 Our data integrate MAP17 into this framework: MAP17 upregulation in macrophages under hyperglycemic conditions appears to act upstream in the metabolic rewiring cascade, augmenting glucose uptake and glycolytic flux while promoting inflammatory polarization.

Mechanistically, we identified SGLT2 as a key downstream effector of MAP17. Functional assays demonstrated a unidirectional regulatory relationship, with MAP17 knockdown reducing SGLT2 expression, whereas SGLT2 silencing did not affect MAP17. This aligns with prior evidence that MAP17 physically associates with and functionally augments SGLT2 activity,33 amplifying glucose transport capacity and supporting glycolytic metabolism under stress. In our in vitro trained immunity model, mimicking the atherosclerosis microenvironment with sustained high glucose and IFN-γ/lipopolysaccharide costimulation, MAP17 overexpression led to marked increases in SGLT2, glycolytic rate, and M1-like polarization. These effects were coupled with elevated ROS production, consistent with MAP17’s known redox-modulatory properties.

The pathogenic interplay between MAP17 and SGLT2 can be conceptualized as a feedforward loop: hyperglycemia induces MAP17 overexpression; MAP17 enhances SGLT2 expression and glucose uptake; increased glycolysis sustains proinflammatory cytokine production; and inflammatory signaling further boosts MAP17 expression. This loop reinforces metabolic adaptation and inflammatory training in macrophages, thereby fueling atherogenesis. Disruption of any node, through MAP17 shRNA, SGLT2 knockdown, or glycolytic inhibition (dichloroacetate), attenuated M1-like polarization, reduced cytokine secretion, and decreased plaque burden. Notably, MAP17 overexpression could partially rescue the effects of SGLT2 inhibition, underscoring its upstream regulatory dominance within this axis. Our findings suggest that the effects of MAP17 are at least partially mediated through the SGLT2-glycolysis axis. We acknowledge that MAP17 may also influence atherosclerosis through SGLT2-independent pathways, such as other glucose transporters or T-cell-mediated responses. Our in vivo BM transplantation experiments further support the concept that MAP17-driven immune memory contributes to atherosclerosis. Diabetic Apoe−/− mice receiving BM from diabetic donors exhibited increased MAP17/SGLT2 expression, elevated glycolytic and inflammatory markers, and exacerbated plaque pathology, consistent with the transfer of a trained immunity phenotype 57. Therapeutic interventions targeting this axis showed that SGLT2 silencing exerted the strongest inhibitory effect on glycolysis and inflammation, followed by dichloroacetate and MAP17 silencing. Although MAP17 knockdown yielded slightly less suppression than SGLT2 silencing, its ability to regulate SGLT2 expression positions it as a potentially more upstream and versatile target.

From an immunometabolic perspective, MAP17’s dual capacity to modulate redox homeostasis and metabolic flux situates it at a nexus between environmental stressors and macrophage inflammatory programming. By amplifying glucose uptake via SGLT2 and sustaining glycolytic metabolism, MAP17 facilitates the epigenetic and bioenergetic underpinnings of trained immunity. This mechanism offers a plausible explanation for the heightened atherosclerosis risk in patients with metabolic syndrome or diabetes, where MAP17 overexpression may act as both a sensor and amplifier of metabolic-inflammatory stress. This study has several limitations: (1) the clinical cohort size was modest; (2) only male mice were utilized to ensure mechanistic clarity; (3) plaque burden in the aortic root was quantified using 3 representative sequential frozen sections per mouse rather than a larger series of serial sections across the entire aortic root; although this approach was selected to maintain consistency across multiple histological stainings under limited frozen tissue availability, it may not fully capture the complete distributional trend of lesion morphology; and (4) while SGLT2 is a key mediator, the contribution of other MAP17-related metabolic transporters warrants further investigation. Future studies should incorporate a more extensive serial-sectioning strategy for more comprehensive histological assessment, together with single-cell transcriptomic and metabolomic approaches to delineate MAP17/SGLT2 spatiotemporal expression across macrophage subpopulations, identify key downstream proatherogenic effectors, and further refine the therapeutic potential of MAP17-targeted interventions beyond diabetic atherosclerosis.

Conclusions

Collectively, our findings define MAP17 as a master regulator of macrophage immunometabolic reprogramming in atherosclerosis, acting in part through SGLT2-mediated glycolysis to sustain trained immunity and inflammatory activation. Targeting the MAP17-SGLT2 axis may disrupt the self-reinforcing metabolic-inflammatory circuit that underlies plaque progression, offering a promising avenue for therapeutic intervention in high-risk metabolic and cardiovascular disease states.

Article Information

Author Contributions

J. Wu and L. Wang contributed equally to the conceptualization and design of the study. J. Wu was primarily responsible for data collection and analysis, while L. Wang played a key role in interpreting the results and drafting the manuscript. F. Zou, F. Liu, and G. Fang both contributed to data analysis and interpretation, and participated in writing and revising the manuscript. They also provided valuable feedback on the study’s methodology and analysis. K. Cheng and H. Gao contributed to the project’s overall management and provided critical feedback on the study’s design and implementation. They also contributed to writing and revising the manuscript. K. Zhi and L. Qu contributed to the overall management of the project and provided critical feedback on the design and implementation of the research. All authors played significant roles in the development of the study and approved the final version of the manuscript.

Sources of Funding

This research is supported by the National Natural Science Foundation of China (grant 82370498), the National Natural Science Foundation of China (grant 82400555), the Youth Start-Up Fund of Naval Medical University (grant 2023QN091), and the Key Project of Planning Subject of Shanghai Higher Education Association (grant 2QZD2425).

Disclosures

None.

Supplemental Material

Table S1–S2

Figures S1–S8

Full Unedited Western Blots

Major Resources Table

ARRIVE Guidelines

Supplementary Material

Nonstandard Abbreviations and Acronyms

BM
bone marrow
BMDM
bone marrow–derived macrophage
CD206
cluster of differentiation 206
DAPI
4′,6-diamidino-2-phenylindole
GFP
green fluorescent protein
glycoATP
glycolytic ATP
HbA1c
hemoglobin A1c
HDL-C
high-density lipoprotein-cholesterol
HK
hexokinase
HRP
horseradish peroxidase
IFN-γ
interferon-γ
IL
interleukin
iNOS
inducible nitric oxide synthase
LDL
low-density lipoprotein
LDL-C
low-density lipoprotein-cholesterol
MAP17
membrane-associated protein 17
NHE3
sodium-hydrogen exchanger 3
ORO
Oil Red O
PFK
phosphofructokinase
PK
pyruvate kinase
qRT-PCR
quantitative reverse transcription-polymerase chain reaction
ROS
reactive oxygen species
SGLT2
sodium-glucose cotransporter 2
TGF-β
transforming growth factor-β
TNF-α
tumor necrosis factor-α
*

J. Wu and L. Wang contributed equally and share first authorship.

References

  • 1.Libby P. The changing landscape of atherosclerosis. Nature. 2021;592:524–533. doi: 10.1038/s41586-021-03392-8 [DOI] [PubMed] [Google Scholar]
  • 2.Falk E. Pathogenesis of atherosclerosis. J Am Coll Cardiol. 2006;47:C7–12. doi: 10.1016/j.jacc.2005.09.068 [DOI] [PubMed] [Google Scholar]
  • 3.Kong P, Cui ZY, Huang XF, Zhang DD, Guo RJ, Han M. Inflammation and atherosclerosis: signaling pathways and therapeutic intervention. Signal Transduct Target Ther. 2022;7:131. doi: 10.1038/s41392-022-00955-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Netea MG, Dominguez-Andres J, Barreiro LB, Chavakis T, Divangahi M, Fuchs E, Joosten LAB, van der Meer JWM, Mhlanga MM, Mulder WJM, et al. Defining trained immunity and its role in health and disease. Nat Rev Immunol. 2020;20:375–388. doi: 10.1038/s41577-020-0285-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Netea MG, van der Meer JW. Trained immunity: an ancient way of remembering. Cell Host Microbe. 2017;21:297–300. doi: 10.1016/j.chom.2017.02.003 [DOI] [PubMed] [Google Scholar]
  • 6.Hajishengallis G, Li X, Mitroulis I, Chavakis T. Trained innate immunity and its implications for mucosal immunity and inflammation. Adv Exp Med Biol. 2019;1197:11–26. doi: 10.1007/978-3-030-28524-1_2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Udell JA, Jones WS, Petrie MC, Harrington J, Anker SD, Bhatt DL, Hernandez AF, Butler J. Sodium glucose cotransporter-2 inhibition for acute myocardial infarction: JACC review topic of the week. J Am Coll Cardiol. 2022;79:2058–2068. doi: 10.1016/j.jacc.2022.03.353 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Dong Z, Hou L, Luo W, Pan LH, Li X, Tan HP, Wu RD, Lu H, Yao K, Mu MD, et al. Myocardial infarction drives trained immunity of monocytes, accelerating atherosclerosis. Eur Heart J. 2024;45:669–684. doi: 10.1093/eurheartj/ehad787 [DOI] [PubMed] [Google Scholar]
  • 9.Murugathasan M, Jafari A, Amandeep A, Hassan SA, Chihata M, Abdul-Sater AA. Moderate exercise induces trained immunity in macrophages. Am J Physiol Cell Physiol. 2023;325:C429–C442. doi: 10.1152/ajpcell.00130.2023 [DOI] [PubMed] [Google Scholar]
  • 10.Zhou J, Liu W, Zhao X, Xian Y, Wu W, Zhang X, Zhao N, Xu FJ, Wang C. Natural melanin/alginate hydrogels achieve cardiac repair through ROS scavenging and macrophage polarization. Adv Sci (Weinh). 2021;8:e2100505. doi: 10.1002/advs.202100505 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Fang S, Sun S, Cai H, Zou X, Wang S, Hao X, Wan X, Tian J, Li Z, He Z, et al. IRGM/Irgm1 facilitates macrophage apoptosis through ROS generation and MAPK signal transduction: Irgm1+/- mice display increases atherosclerotic plaque stability. Theranostics. 2021;11:9358–9375. doi: 10.7150/thno.62797 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Okitsu H, Tsuboi M, Nakajima S, Takahashi E, Hakoshima A, Ishimaru S. A case report of fibromatosis originating from the right anterior scalene muscle. Nihon Kyobu Geka Gakkai Zasshi. 1990;38:1506–1510. [PubMed] [Google Scholar]
  • 13.Guijarro MV, Link W, Rosado A, Leal JF, Carnero A. MAP17 inhibits Myc-induced apoptosis through PI3K/AKT pathway activation. Carcinogenesis. 2007;28:2443–2450. doi: 10.1093/carcin/bgm154 [DOI] [PubMed] [Google Scholar]
  • 14.Coady MJ, El Tarazi A, Santer R, Bissonnette P, Sasseville LJ, Calado J, Lussier Y, Dumayne C, Bichet DG, Lapointe JY. MAP17 is a necessary activator of renal Na+/glucose cotransporter SGLT2. J Am Soc Nephrol. 2017;28:85–93. doi: 10.1681/ASN.2015111282 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Villa-Bellosta R, Barac-Nieto M, Breusegem SY, Barry NP, Levi M, Sorribas V. Interactions of the growth-related, type IIc renal sodium/phosphate cotransporter with PDZ proteins. Kidney Int. 2008;73:456–464. doi: 10.1038/sj.ki.5002703 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Niu Y, Liu R, Guan C, Zhang Y, Chen Z, Hoerer S, Nar H, Chen L. Structural basis of inhibition of the human SGLT2-MAP17 glucose transporter. Nature. 2022;601:280–284. doi: 10.1038/s41586-021-04212-9 [DOI] [PubMed] [Google Scholar]
  • 17.Hiraizumi M, Akashi T, Murasaki K, Kishida H, Kumanomidou T, Torimoto N, Nureki O, Miyaguchi I. Transport and inhibition mechanism of the human SGLT2-MAP17 glucose transporter. Nat Struct Mol Biol. 2024;31:159–169. doi: 10.1038/s41594-023-01134-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Riccardi G, Giosue A, Calabrese I, Vaccaro O. Dietary recommendations for prevention of atherosclerosis. Cardiovasc Res. 2022;118:1188–1204. doi: 10.1093/cvr/cvab173 [DOI] [PubMed] [Google Scholar]
  • 19.Raitakari O, Pahkala K, Magnussen CG. Prevention of atherosclerosis from childhood. Nat Rev Cardiol. 2022;19:543–554. doi: 10.1038/s41569-021-00647-9 [DOI] [PubMed] [Google Scholar]
  • 20.Doran AC. Inflammation resolution: implications for atherosclerosis. Circ Res. 2022;130:130–148. doi: 10.1161/CIRCRESAHA.121.319822 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Fan J, Watanabe T. Atherosclerosis: known and unknown. Pathol Int. 2022;72:151–160. doi: 10.1111/pin.13202 [DOI] [PubMed] [Google Scholar]
  • 22.Pedro-Botet J, Climent E, Benaiges D. Atherosclerosis and inflammation. New therapeutic approaches. Med Clin (Barc). 2020;155:256–262. doi: 10.1016/j.medcli.2020.04.024 [DOI] [PubMed] [Google Scholar]
  • 23.Rizzo MR, Di Meo I, Polito R, Auriemma MC, Gambardella A, di Mauro G, Capuano A, Paolisso G. Cognitive impairment and type 2 diabetes mellitus: focus of SGLT2 inhibitors treatment. Pharmacol Res. 2022;176:106062. doi: 10.1016/j.phrs.2022.106062 [DOI] [PubMed] [Google Scholar]
  • 24.Dong F, Li R, Wang J, Zhang Y, Yao J, Jiang SH, Hu X, Feng M, Bao Z. Hypoxia-dependent expression of MAP17 coordinates the Warburg effect to tumor growth in hepatocellular carcinoma. J Exp Clin Cancer Res. 2021;40:121. doi: 10.1186/s13046-021-01927-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Huang Z, Chen J, Liu S, Xiang X, Long Y, Tan P, Fu W. MAP17 is a novel NASH progression biomarker associated with macrophage infiltration, immunotherapy response, and oxidative stress. J Inflamm Res. 2025;18:601–619. doi: 10.2147/JIR.S497737 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Yang L, Zhang X, Wang Q. Effects and mechanisms of SGLT2 inhibitors on the NLRP3 inflammasome, with a focus on atherosclerosis. Front Endocrinol (Lausanne). 2022;13:992937. doi: 10.3389/fendo.2022.992937 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Steinvil A, Berliner S, Bromberg M, Cohen M, Shalev V, Shapira I, Rogowski O. Micro-inflammatory changes in asymptomatic healthy adults during bouts of respiratory tract infections in the community: potential triggers for atherothrombotic events. Atherosclerosis. 2009;206:270–275. doi: 10.1016/j.atherosclerosis.2009.01.045 [DOI] [PubMed] [Google Scholar]
  • 28.Batty M, Bennett MR, Yu E. The role of oxidative stress in atherosclerosis. Cells. 2022;11:3843. doi: 10.3390/cells11233843 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Geckin B, Konstantin Fohse F, Dominguez-Andres J, Netea MG. Trained immunity: implications for vaccination. Curr Opin Immunol. 2022;77:102190. doi: 10.1016/j.coi.2022.102190 [DOI] [PubMed] [Google Scholar]
  • 30.Bekkering S, Dominguez-Andres J, Joosten LAB, Riksen NP, Netea MG. Trained immunity: reprogramming innate immunity in health and disease. Annu Rev Immunol. 2021;39:667–693. doi: 10.1146/annurev-immunol-102119-073855 [DOI] [PubMed] [Google Scholar]
  • 31.Edgar L, Akbar N, Braithwaite AT, Krausgruber T, Gallart-Ayala H, Bailey J, Corbin AL, Khoyratty TE, Chai JT, Alkhalil M, et al. Hyperglycemia induces trained immunity in macrophages and their precursors and promotes atherosclerosis. Circulation. 2021;144:961–982. doi: 10.1161/CIRCULATIONAHA.120.046464 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Zhong C, Yang X, Feng Y, Yu J. Trained immunity: an underlying driver of inflammatory atherosclerosis. Front Immunol. 2020;11:284. doi: 10.3389/fimmu.2020.00284 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Capuano A, Clementi E, Paolisso G. Editorial: clinical prospective of SGLT2 inhibitors in atherosclerosis. Front Cardiovasc Med. 2022;9:1040649. doi: 10.3389/fcvm.2022.1040649 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.


Articles from Arteriosclerosis, Thrombosis, and Vascular Biology are provided here courtesy of Wolters Kluwer Health

RESOURCES