Macrophage junctional adhesion molecule-like (JAML) protein promotes NLRP3 inflammasome activation in the development of atherosclerosis

Animals

Male ApoE^−/− mice, aged 8 weeks, were acquired from Beijing Vital River Laboratory Animal Technology Co., Ltd. These mice were subjected to either a normal or HFD for 12 weeks. Macrophage-specific JAML knockout (JAML^M-KO) and knock‐in (JAML^M-KI) mice were generated using CRISPR/Cas9 and Cre/LoxP techniques by GemPharmatech Co., Ltd. and subsequently crossed with ApoE^−/− mice to generate JAML^M-KO/ApoE^−/− and JAML^M-KI/ApoE^−/− mice. All mice used in this study were male and aged 6-8 weeks. These mice were subjected to a HFD for 12 weeks. Then, JAML^WT/ApoE^−/− and JAML^M-KI/ApoE^−/− mice were fed with HFD and randomized to 0.9% saline group and MCC950 group (10 mg/kg) via intraperitoneal injection three times a week for 12 weeks [17]. They were provided with sufficient food and water ad libitum and maintained under a 12-h light/12-h dark cycle at a consistent ambient temperature of 23 °C. All animal experimental protocols adhered to the Management Rules of the Chinese Ministry of Health and were approved by the appropriate Ethics Committee (DWLL-2023-037).

The different mouse groups were randomly assigned and investigators were blinded to group allocation during the experiment and when assessing the outcome. The required sample size was calculated by assuming 80% power with significance level of 0.05. The numbers of animals used in each group was indicated in the respective figure legends.

Human atherosclerotic samples

Carotid artery plaque tissue was obtained via internal carotid endarterectomy, with the adjacent non-atherosclerotic section of the vessel wall serving as the control. All experimental protocols were approved by the Ethics Committee of Shandong University Qilu Hospital. Venous blood samples were collected from both healthy volunteers and patients, and PBMCs were isolated following the manufacturer’s instructions (Solarbio, P8680, China). Written informed consent was obtained from all the participants, and the study was conducted in accordance with the Declaration of Helsinki. To generate HMDMs, CD14(+) cells were isolated from the PBMCs using a CD14 MicroBeads (Miltenyi Biotec, 130-050-201).And isolated CD14(+) cells were cultured with Dulbecco’s Modified Eagle’s medium (DMEM; Gibco, USA), 10% FBS (Gibco, USA), 1% penicillin-streptomycin (Gibco, USA) and 20 ng/mL of M-CSF (Medchemexpress, China) for one week [18–20].

Data collection and bioinformatics analysis

The GSE43292 dataset was downloaded from the Gene Expression Omnibus database (https://ncbi.nlm.nih.gov/geo/). This dataset included human carotid endarterectomy samples from 34 participants; these were used to identify differentially expressed genes (DEGs) in human atherosclerotic plaques. Data analysis was performed using R project (R version 4.0.3). Firstly, the probe names in the gene expression profiles were converted into gene names according to platform annotation files and calculate the average value of genes corresponding to multiple probes.

Atherosclerotic lesion analysis

Mice were euthanized, and their aortas and hearts were immediately perfused with physiological saline. The entire aorta and heart were meticulously dissected under a microscope and immersed in 4% paraformaldehyde for 48 h. The aortas were then unfolded, stained with Oil Red O, and photographed using a digital camera. Hearts were dehydrated with gradient sucrose solutions and embedded in optimal cutting temperature compound (OCT, Sakura, Japan) for frozen sections (6 μm). Representative sections were stained with Oil Red O and H&E. Image Pro Plus software (Media Cybernetics, Inc., USA) was used to quantify the stained areas.

Cell culture

Mice were intraperitoneally injected with 1 ml sterile 6% starch solution. After 3 days, peritoneal macrophages were collected using ice-cold DMEM (Gibco, USA) through peritoneal lavage. The cell suspension was centrifuged at 800 rpm for 5 min, and the pellet was resuspended in fresh DMEM supplemented with 10% fetal bovine serum (Gibco, USA) and 1% Penicillin-Streptomycin (Gibco, USA). Macrophages were then treated with lipopolysaccharide (LPS; Sigma, USA) for 8 h at a concentration of 100 ng/ml, ox-LDL (Yiyuanbiotech, China) for 24 h at a concentration of 50 μg/ml, or Cholesterol crystals (CHO) for 24 h at a concentration of 1 mg/mL followed by treatment with adenosine 5’-triphosphate (ATP; Sigma, USA) for 30 min at a concentration of 5 mM.

The human embryonic kidney cells lines (HEK293T) were obtained from KeyGene BioTech (China) and cultured in DMEM medium supplemented with 10% FBS and 1% Penicillin-Streptomycin.

Preparation of cholesterol crystals

Cholesterol powder (Sigma, USA) was dissolved in 95% ethanol and heated to 60 °C, then the sample was filtered through filter paper and crystallized at room temperature. The sample were sterilized by filter sterilizing, autoclaving and ground to make the size within 1-10 μm prior to use. The obtained Cholesterol crystals (CHO) can be stored at -20 °C for spare [21, 22].

Immunofluorescence

After rehydration and blocking, sections were incubated overnight at 4 °C with primary antibodies, including MOMA-2 (ab33451), IL-1β (ab234437), CD31 (ab222783), F4/80 (14-4801-82), JAML (21302-1-AP), and NLRP3 (AG-20B-0014-C100). The next day, the sections were incubated with secondary antibodies, including Alexa Fluor^® 594 (ab150156 and ab150120) and Alexa Fluor® 488 (ab150117, ab150165, and ab150061) for 30 min at room temperature. After washing with 1× phosphate buffered saline (PBS), nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI; ab104139) and visualized using a fluorescence microscope (DS-Ri2, Nikon, Japan). Non-specific immunoglobulin (Ig) G and PBS were used as negative controls.

Transcriptome RNA sequencing

Total RNA from macrophages was isolated using the Total RNA Isolation kit (Tianmo #TR205-200, Genstone, China). The RNA quality was assessed using an Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA, USA). Strand-specific RNA sequencing libraries were constructed using the VAHTS Total RNA-seq (H/M/R) Library Prep Kit (Vazyme, China), and sequencing was performed using Illumina NovaSeq 6000 (Illumina, USA). DEG analysis between macrophages from JAML^WT and JAML^M-KO mice was performed using the edgeR package.

Co-IP

Cells were lysed in radioimmunoprecipitation assay (RIPA) buffer containing protease inhibitors (CW2200) at 4 °C. The lysate was centrifuged at 13,000 rpm for 15 min at 4 °C, and the supernatant was collected. Antibodies or IgG (30000-0-AP) was added to the supernatant and incubated for 2 h at 4 °C. Protein A/G PLUS-Agarose (sc2003) was then added and incubated overnight at 4 °C. The following day, beads were washed five times with lysis buffer (CW2334), treated with sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) Loading Buffer (CW0027) at 99 °C, and processed for western blot analysis.

LC-MS/MS analysis for JAML-interacting proteins

Protein samples collected from JAML co-immunoprecipitates were subjected to LC-MS/MS analysis, with IgG used as the negative control. The samples were electrophoresed, and the gels were analyzed using an LC-MS/MS system with a Q Exactive HF-X mass spectrometer (Thermo Fisher Scientific, Shanghai, China). Data were analyzed using Proteome Discoverer 2.0 software (Thermo Fisher Scientific, Shanghai, China) against the Mus_musculus_uniprot_20230307 database.

Statistical analysis

All statistical analyses were conducted using GraphPad Prism 8 software (GraphPad Software, USA). Data are expressed as mean ± standard error of the mean (SEM), and each experiment was replicated at least three times independently. For normally distributed data, comparisons between two groups were made using a t-test, while comparisons among multiple groups were made using one-way analysis of variance. For non-normally distributed data, the rank sum test was employed. Statistical significance was set at p < 0.05.

Increased JAML expression during atherosclerosis

First, we re-analyzed data from a previously published genome-wide expression study of the human carotid atheroma dataset (GSE43292). Our findings revealed a significant upregulation of JAML expression in human atherosclerotic plaques (Fig. 1A). Following this, we constructed an atherosclerotic mouse model. As shown in Fig. 1B, AS plaques were significantly increased in the mice fed with a 12-week high-fat diet (HFD), as well as significantly increased the expression of JAML compared to that in the control group (Fig. 1C). Immunofluorescence analysis further indicated a high degree of colocalization between JAML and the macrophage marker F4/80 appears to be much higher than that between JAML and CD31 [23] (Fig. 1D, E). These findings suggest that JAML protein is highly expressed in macrophages during atherosclerosis. We further examined the expression level of JAML in peripheral blood mononuclear cells (PBMCs) and human monocyte-derived macrophages (HMDMs) and found that JAML expression was significantly elevated in both cell types of patients with CAD, with the highest level in HMDMs of CAD patients (Fig. 1F, G). Taken together, these findings suggested that JAML was involved in the pathogenesis and progression of atherosclerosis.

Macrophage-specific JAML deletion attenuates atherosclerosis development

To investigate the role of macrophage-derived JAML in atherosclerosis, we engineered macrophage-specific JAML-knockout mice. We crossed JAML^flox/flox mice with Lyz2-Cre mice to generate JAML^flox/flox:Lyz2-Cre⁺ mice (hereafter referred to as JAML^M-KO mice) and JAML^flox/flox:Lyz2-Cre⁻ littermates (hereafter referred to as JAML^WT mice) as control groups. These were further bred with ApoE^−/− mice to obtain JAML^M-KO/ApoE^−/− and JAML^WT/ApoE^−/− mice, respectively (Fig. 2A, B). Following a HFD regimen, a significant reduction in plaque area was observed in JAML^M-KO/ApoE^−/− mice compared to that in controls (Fig. 2C, D). Histological analyses, including hematoxylin-eosin (H&E), Oil Red O, and MOMA-2 staining of the aortic root, showed a significant decrease in lipid accumulation and macrophage infiltration in JAML^M-KO/ApoE^−/− mice (Fig. 2E, F). Despite these differences, no significant changes in heart rate, body weight, blood pressure, or lipid profile were observed between the two groups (Fig. 2G and Table S2). Thus, these data demonstrate that macrophage-specific JAML deficiency mitigates the progression of atherosclerosis.

Macrophage-specific JAML overexpression exacerbates atherosclerosis

To further elucidate the role of macrophage-derived JAML in atherosclerosis in vivo, we generated macrophage-specific JAML-overexpressing mice (JAML^M-KI mice) and verified overexpression efficiency (Fig. 3A, B). Upon induction of atherosclerosis through HFD feeding, analyses of atherosclerotic plaques revealed a significant increase in plaque area, lipid deposition, and macrophage infiltration in JAML^M-KI/ApoE^−/− mice (Fig. 3C-E). Notably, JAML overexpression did not influence heart rate, body weight, blood pressure (Table S3), or lipid profiles (Fig. 3F). Collectively, these results suggest that JAML overexpression in macrophages accelerates the progression of atherosclerosis.

JAML deletion inhibits NLRP3 inflammasome activation

To elucidate the molecular mechanisms underlying the role of JAML in atherosclerosis, we isolated macrophages from JAML^M-KO and JAML^WT mice and stimulated them with LPS or ATP for transcriptomic sequencing. The results demonstrated that 478 genes upregulated and 391 genes downregulated in JAML^M-KO macrophages compared to controls (Fig. S1A). The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and gene set enrichment analyses (GSEA) revealed a significant association of JAML with the NOD-like receptor signaling pathway [24] (Fig. S1B, C). Notably, differential gene expression analysis between the JAML^WT and JAML^M-KO groups highlighted a significant alteration in NLRP3 expression, a key member of the NOD-like receptor family, following JAML knockout (Fig. S1D).

The NLRP3 inflammasome is a multi-protein complex composed of NLRP3, ASC, and caspase-1, which mediates the activation of caspase-1. P10 cleaves pro-IL-1β and pro-IL-18 to produce mature IL-1β and IL-18, potentially contributing to a variety of inflammatory diseases, including atherosclerosis [25, 26]. To determine whether JAML affects NLRP3 inflammasome activation, we assessed the expression of NLRP3 inflammasome-related factors. Following 12 weeks of HFD feeding, we observed a marked reduction in NLRP3 and IL-1β expressions in the aortic roots of JAML^M-KO/ApoE^−/− mice (Fig. 4A-C). Additionally, the expression of NLRP3 inflammasome-associated proteins, including NLRP3, pro-IL-1β, IL-1β, and P10, was significantly lower in JAML^M-KO/ApoE^−/− mice than in controls (Fig. 4E), and the production of NLRP3 inflammasome-dependent cytokines, such as IL-1β and IL-18, was also decreased in the serum of JAML^M-KO/ApoE^−/− mice (Fig. 4D). However, no statistically significant differences were observed in pro-caspase-1 and ASC protein expression levels, which are involved in inflammasome assembly (Fig. 4D, E).

It has been reported that the stimulation of LPS, ox-LDL and Cholesterol crystals (CHO) may contribute to the activation of NLRP3 inflammasome [21, 27, 28]. Therefore, we tested a variety of stimuli by western blot, and the results of our experiment showed that NLRP3 inflammasome activation was only possible after ATP was added after LPS, ox-LDL or CHO addition (Fig. 5A). Next, we evaluated the molecular alterations after stimulation by LPS + ATP and ox-LDL + ATP. The mRNA and protein expression levels of NLRP3 inflammasome-related factors were decreased in LPS and ATP-stimulated JAML^M-KO macrophages (Fig. 5B, E). Consistent with these findings, IL-1β, P10 and IL-18 protein levels in supernatant from JAML^M-KO macrophages were also significantly decreased (Fig. 5C). Additionally, IL-1β and IL-18 concentrations in the supernatants of JAML-depleted macrophages were significantly reduced (Fig. 5D). These data indicate that JAML deficiency in macrophages markedly reduces the activation of the NLRP3 inflammasome in atherosclerotic plaques.

JAML overexpression promotes NLRP3 inflammasome activation

To further demonstrate the relationship between JAML and the NLRP3 inflammasome, we examined the activation of the NLRP3 inflammasome following overexpression of JAML. JAML^M-KI mice exhibited elevated protein expression levels of NLRP3 and IL-1β in aortic root plaques compared to those in control mice (Fig. 6A-C). In vitro, we used an adenovirus to overexpress JAML in macrophages (Fig. 6D). As shown in Fig. 6E-G, macrophages overexpressing JAML displayed increased mRNA and protein expression levels of NLRP3, pro-IL-1β, IL-1β, and P10 compared to those in controls, including IL-1β, P10, and IL-18 in cell supernatant. Furthermore, JAML overexpression led to increased levels of the inflammatory cytokines IL-1β and IL-18. Taken together, these results support the role of JAML in positively regulating NLRP3 inflammasome activation during atherosclerosis. Stimulation of ox-LDL + ATP was also evaluated, and the results indicated that JAML knockdown (Fig. S1A, B) and overexpression (Fig. S2C, D) decreased and increased the activation of the NLRP3 inflammasome, respectively, compared with the corresponding Control groups.

To further validate our hypothesis, MCC950, a NLRP3 inflammasome inhibitor, was used [17, 18, 29]. We compared atherosclerotic lesion progression between JAML^WT/ApoE^−/− mice and JAML^M-KI/ApoE^−/− mice after 12 weeks of feeding a HFD and intraperitoneal injection of MCC950. Collectively, the results suggested that MCC950 reversed the effects of JAML overexpression on AS progression (Fig. S4). The above results further indicated that atherosclerosis-promoting effects of JAML by activating NLRP3 inflammasome.

JAML promotes NLRP3 inflammasome activation via the NF-κB pathway

Next, we investigated the molecular mechanisms through which JAML regulates NLRP3 inflammasome activation in macrophages. NLRP3 inflammasome activation is triggered by exogenous pathogen-associated molecular patterns and endogenous damage-associated molecular patterns through the activation of the NF-κB signaling pathway, which induces the transcription of pro-IL-1β and NLRP3 [26, 30]. We stimulated macrophages with LPS and assessed the activation of the NF-κB signaling pathway. Compared to the control group, JAML knockout significantly inhibited LPS-induced phosphorylation of IKKα/β, IKBα, and NF-κB p65 (Fig. 7A, B). Additionally, JAML overexpression promoted the activation of the NF-κB pathway (Fig. 7C, D). In addition, we further validated the activation of NF-κB with ox-LDL stimulation, and the results were consistent with the above (Fig. S3). The application of BAY 11-7082, an NF-κB pathway inhibitor, reversed JAML-induced activation of the NLRP3 inflammasome (Fig. 7E, F). These findings demonstrate that JAML promotes NLRP3 inflammasome activation via the NF-κB pathway.

JAML interacts with PKM2 to regulate the activation of the NF-κB signaling pathway

To elucidate the mechanism through which JAML activates NF-κB, we conducted a search for JAML-interacting proteins using co-immunoprecipitation (Co-IP) coupled with liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. PKM emerged as one of the top JAML-binding proteins and was selected for further investigation (Table 1). Among the two isoforms of PKM [31], only PKM2 plays a pivotal role in regulating inflammation in macrophages [32]. The interaction between JAML and PKM2 was confirmed through Co-IP (Fig. 8A). Next, we determined the domains of JAML and PKM2 that were responsible for their interaction. Based on previously reported structural characteristics of JAML and PKM2 [5, 33, 34], truncated fragments encoding different domains of these proteins were constructed (Fig. 8B) and transfected into 293 T cells. The intracellular domain (ICD) of JAML was found to interact with the A + B domain of PKM2 (Fig. 8C, D).

Phosphorylation of PKM2 at Tyr105 prevents the formation of the PKM2 tetramer and promotes the formation of an inactive dimer, facilitating its translocation to the nucleus and binding to the NF-κB p65 subunit [35–37]. We assessed the protein levels of phosphorylated PKM2 (p-PKM2) and total PKM2 in macrophages, as well as the distribution of PKM2 in the cytosol and nucleus. JAML deficiency did not affect the total PKM2 protein expression (Fig. 8E) but significantly reduced p-PKM2 levels, PKM2 nuclear expression, and the binding of PKM2 to p65 (Fig. 8F-J). Next, we transfected macrophages with PKM2-specific siRNA (Fig. 8K) and observed that PKM2 knockdown reversed the activation of NF-κB signaling and NLRP3 inflammasome induced by JAML overexpression (Fig. 8L-N). These data suggest that JAML regulates the activation of the NF-κB signaling pathway through its interaction with PKM2 in macrophages.