Communications Biology

Fortilin deficiency induces anti-atherosclerotic phenotypes in macrophages and protects hypercholesterolemic mice against atherosclerosis

Article
Open access
Published: 11 July 2025

Fortilin deficiency induces anti-atherosclerotic phenotypes in macrophages and protects hypercholesterolemic mice against atherosclerosis

Communications Biology volume 8, Article number: 1040 (2025) Cite this article

4193 Accesses
1 Altmetric
Metrics details

Subjects

Cardiovascular biology

Abstract

In the atherosclerotic intima, macrophages (MΦ) perpetuate chronic inflammation and cholesterol accumulation. Fortilin, a 172-amino-acid multifunctional protein, is abundant in the atherosclerotic intima and promotes atherogenesis, but its mechanism has remained unclear. Herein, we report that fortilin in MФ (fortilin^MΦ) facilitates atherosclerosis by (a) enhancing MΦ survival, proliferation, and lipid uptake, leading to the accumulation of lipid-laden MФ in the intima and (b) inhibiting both the reverse transdifferentiation of MΦ into vascular smooth muscle cells (VSMCs) and the differentiation of mesenchymal stem cells (MSCs) into VSMCs. Mice lacking fortilin^MΦ under genetically induced hypercholesterolemia (fortilin^KO-MΦ-HC) exhibit drastically less atherosclerosis in their aortae compared to wild-type (fortilin^WT-MΦ-HC) controls. Imaging mass cytometry reveals that the intima of fortilin^KO-MΦ-HC mice contains fewer MФ but more VSMCs than that of fortilin^WT-MФ-HC mice. Cell-based assays reveal that fortilin deficiency in MΦ augments low-density lipoprotein (LDL)-induced apoptosis, suppresses proliferation and foam cell formation, and boosts TGF-β1 production. Fortilin-deficient THP1 MΦ transdifferentiate into VSMCs, and their conditioned medium causes MSCs to differentiate toward VSMCs in a TGF-β1-dependent fashion. Together, these findings suggest that fortilin^MΦ plays a complex facilitative role in atherogenesis and represents a viable molecular target for the treatment of atherosclerosis.

Similar content being viewed by others

Dehydrocorydaline maintains the vascular smooth muscle cell contractile phenotype by upregulating Spta1

Article Open access 20 January 2025

Vascular smooth muscle cells enhance immune/vascular interplay in a 3-cell model of vascular inflammation

Article Open access 23 September 2023

Protective effects of liraglutide on hypercholesterolemia-associated atherosclerosis involve attenuation of endothelial-monocyte adhesion through down-regulating the LOX-1/NF-κB signaling pathway

Article Open access 28 July 2025

Introduction

Atherosclerosis is the chronic build-up of plaque in the intima or the innermost layer of the arteries. Worldwide death due to atherosclerosis and its associated complications, including myocardial infarction (MI), stroke, and ruptured aneurysms, has surpassed that of every major disease, including cancer, infection, and trauma¹.

In the early stages of atherosclerosis, low-density lipoproteins (LDLs) in the plasma cross the endothelial barrier and enter the intima, where they become oxidized². Oxidized LDL (oxLDL) is a potent chemoattractant that recruits circulating monocytes to the intima and promotes their differentiation to macrophages (MΦ). MΦ in the atherosclerotic intima avidly take up oxLDL through scavenger receptor-A, cluster of differentiation 36 (CD36), and lectin-like oxLDL receptor (LOX)-1^3,4, transform themselves into foam cells (FCs), proliferate^5,6, and release a number of pro-inflammatory cytokines and chemokines⁷.

The slow progressive inflammatory process of atherosclerosis that continues for years could one day turn into a sudden, catastrophic, and potentially lethal condition in the coronary arteries (i.e., MI). During MI, the atherosclerotic plaque ruptures, leading to platelet activation, thrombotic occlusion of the coronary artery, and irreversible ischemic damage to the heart. A current consensus is that MΦ destabilize the plaques by producing metalloproteases and weakening the plaque surface and that vascular smooth muscle cells (VSMCs) stabilize the plaque by producing collagens and other extracellular matrices in the atherosclerotic intima⁸.

Fortilin, also known as translationally controlled tumor protein and histamine-releasing factor, is a ubiquitously expressed, highly conserved, 172-amino acid, 20-kDa protein that is present in normal cells in the nucleus⁹, cytosol⁹, and extracellular space¹⁰. Although it was originally cloned in 1988 by Gross et al. as a molecule abundantly expressed in tumor cells¹¹, the function of fortilin remained unknown until 2001 when we and others reported that fortilin blocks apoptosis^{9,12,13,14,15,16,17}. In addition to negatively regulating apoptosis, fortilin modulates a wide range of cellular functions and is implicated in cell cycle progression¹⁸, immunoglobulin (Ig)-E-mediated histamine release and allergic reactions^19,20, reactive oxygen species handling²¹, modulation of endoplasmic reticulum stress pathways²², and protein homeostasis (proteostasis)^21,23,24.

The role of fortilin in atherosclerosis remained unknown until recently when we generated a mouse strain that constitutionally lacks fortilin (fortilin^+/-) using homologous-recombination-based gene targeting strategies²⁵. Although we could not use fortilin^-/- mice because they are embryonically lethal^25,26, we placed fortilin^+/- and wild-type control (fortilin^+/+) mice on the Ldlr^–/–Apobec1^–/– hypercholesterolemic (HC) genetic background (fortilin^+/–&HC and fortilin^+/+&HC, respectively)²⁷, incubated them for 10 months on a normal chow diet, and assessed the degree and extent of atherosclerosis in the aortae. We found that fortilin^+/–&HC mice exhibited significantly less atherosclerosis in their aortae than did fortilin^+/+&HC mice²⁸. Because fortilin^+/-&HC mice had decreased fortilin expression in all organs, tissues, and blood, however, we did not know the cell type in which the lack of fortilin mediated the amelioration of atherosclerosis.

Because the atherosclerotic lesions of fortilin^+/–&HC mice contained significantly fewer MΦ than those of fortilin^+/+&HC mice²⁸ and because fortilin protects MΦ against apoptosis at least partly through Myeloid Cell Leukemia-1 (MCL-1)^23,29, we hypothesized that fortilin in MΦ (fortilin^MΦ) facilitates atherosclerogenesis. Using a mouse strain lacking fortilin^MΦ and on the HC genetic background, we herein show that fortilin^MΦ facilitates atherosclerosis. Strikingly, fortilin not only protects MФ against apoptosis but also promotes their proliferation and transformation into FCs, while preventing their transdifferentiation into VSMCs and inhibiting the differentiation of mesenchymal stem cells (MSCs) into VSMCs in a transforming growth factor beta 1 (TGF-β1)-dependent fashion, where fortilin suppresses the secretion of the anti-atherosclerotic cytokine TGF-β1 from MΦ. In contrast, the lack of fortilin^MΦ induces MΦ apoptosis in high oxLDL environments, decelerates MΦ proliferation and lipid uptake, increases TGF-β1 secretion, transdifferentiates MΦ themselves to VSMCs, and differentiates MSCs to VSMCs, together mitigating atherosclerosis.

Fortilin^MΦ is directionally pro-atherosclerotic through its diverse biological activities. Based on our results, we propose that fortilin is a key molecule in MΦ that drives HC-induced atherosclerogenesis.

Results

A new mouse strain lacking fortilin in MΦ was generated and placed on the HC genetic background

To test if fortilin^MΦ facilitates atherosclerosis, we first used the standard homologous recombination technique³⁰ to generate fortilin^flox/flox mice in which the fortilin gene was flanked by the LoxP sequence to allow for tissue-specific deletion (Fig. 1a-1). This technique is described in detail in the “Materials and Methods” section of the manuscript.

Fig. 1: Generation and characterization of a mouse strain that lacks fortilin in macrophages (MФ, fortilin^KO-MФ).

We crossed these mice (Fig. 1a-1) with C57BL/6 J mice overexpressing the Cre transgene under the control of the LysM enhancer/promoter (Fig. 1a-2a) to generate LysM-Cre^+/–fortilin^flox/flox mice (hereafter called fortilin^KO-MΦ mice), which is a mouse strain lacking fortilin in MΦ and other myeloid lineage cells (Fig. 1a-3a) but not in other cell types (Fig. 1a-3b). Their control was LysM-Cre^-/-fortilin^flox/flox mice (hereafter called fortilin^WT-MΦ mice) (Fig. 1a-2b). The peritoneal MΦ of the fortilin^KO-MΦ mice (MΦ^KO-fortilin) expressed barely detectable levels of fortilin, as determined by real-time quantitative reverse transcription polymerase chain reaction (RT-qPCR) (Fig. 1b) and western blot analysis (Fig. 1c, fortilin^KO-MΦ, arrow), whereas fortilin expression was similar in other organs between fortilin^KO-MΦ and fortilin^WT-MΦ mice (Fig. 1c). Fortilin^KO-MΦ mice appeared normal and fertile. There was no difference in survival between fortilin^KO-MΦ and fortilin^WT-MΦ mice, with median survival of 104.0 and 96.0 weeks, respectively (P = 0.24, Log-rank Mantel-Cox test, N = 24 and 28) (Fig. 1d). There was no difference in weight between fortilin^KO-MΦ and fortilin^WT-MΦ mice from ages 8 to 136 weeks when placed on the normal chow diet (Fig. S1a). We then placed both fortilin^KO-MΦ and fortilin^WT-MΦ mice on the HC genetic background by crossing them with Ldlr^–/–Apobec1^–/– mice^27,28,31 (hereafter called fortilin^KO-MΦ-HC (or simply KO) and fortilin^WT-MΦ-HC (or simply WT) mice, respectively) (Fig. 1e, f).

Fortilin^KO-MΦ-HC mice developed significantly less atherosclerosis than did fortilin^WT-MΦ-HC mice

We fed these mice a normal chow diet for 10 months and then subjected the entire aortae to en face and cross-sectional atherosclerosis assays. At the time of sacrifice, fortilin^KO-MΦ-HC mice were numerically mildly but statistically significantly heavier by 9.1% than fortilin^WT-MΦ-HC mice (Fig. 1g), although both strains had similar lipid profiles, including serum cholesterol levels (Fig. 1h–l; Fig. S1b–d). Fortilin is important for cell survival at the cellular level⁹, but the circulating blood of fortilin^KO-MΦ-HC and fortilin^WT-MΦ-HC mice did not differ in total white blood cell, monocyte, eosinophil, and basophil counts, although neutrophil and lymphocyte counts were modestly but significantly lower in fortilin^KO-MΦ-HC mice compared to fortilin^WT-MΦ-HC mice (Fig. S1e). Red blood cell and platelet counts did not vary significantly between fortilin^KO-MΦ-HC and fortilin^WT-MΦ-HC mice (Fig. S1f, g).

Strikingly, in this experimental system, the aortae of fortilin^KO-MΦ-HC mice exhibited 51.7% (Fig. 2a) and 33.0% (Fig. 2b) less atherosclerosis than those of fortilin^WT-MΦ-HC mice as determined by en face and cross-sectional assays, respectively (fortilin^WT-MΦ-HC vs. fortilin^KO-MΦ-HC mice = 12.13 ± 3.67 vs. 6.01 ± 3.44%, 50.5% reduction by the en face assay, P < 0.0001, n = 17 each; 0.162 ± 0.056 vs. 0.109 ± 0.031, 32.7% reduction by the cross-sectional assay, P < 0.026, n = 8 each; both by Student’s t test).

Fig. 2: Lack of fortilin in MФ ameliorates atherosclerosis.

Subsequently, we performed a quantitative multiplex analysis of the sera of these animals (N = 5–7 each) (Fig. 2c–g; Fig. S2a, b). We found that the sera from fortilin^KO-MΦ-HC mice had lower levels of inflammatory and pro-atherosclerotic cytokines, such as tumor necrosis factor α (TNF-α)³², interferon-γ (IFN-γ)^33,34, and interleukin 23 (IL-23)³⁵ (Fig. 2c–e), and higher levels of anti-inflammatory, anti-atherosclerotic cytokines such as TGF-β1 and TGF-β2^36,37 (Fig. 2f, g) and colony stimulating factor 3 (granulocyte (G)-CSF) (Fig. S2b)^38,39. Because TGF-βs are implicated in organ fibrosis⁴⁰, we measured the expression levels of fibrosis-related genes (type I collagen, Col1a1; type III collagen, Col3a1; and type IV collagen, Col4a1) in the lungs, kidneys, and livers of 20-week-old fortilin^WT-MФ-HC and fortilin^KO-MФ-HC mice. We found no significant differences between the two groups (Fig. S2c), which suggests that the degree of elevation in serum TGF-β1 and TGF-β2 levels observed in fortilin^KO-MФ-HC mice is unlikely to induce fibrosis-related genes in major organs. The serum levels of other pro-atherosclerotic cytokines, including eotaxin (CCL11, C-C motif chemokine ligand 11)⁴¹, IL-1α⁴², IL-6⁴³, and RANTES (CCL5)⁴⁴, were not significantly different between fortilin^WT-MФ-HC and fortilin^KO-MФ-HC mice (Fig. S2a). Interestingly, serum levels of some of the cytokines that are reported to be either anti-inflammatory or anti-atherosclerotic, such as IL-4⁴⁵ and IL-13⁴⁶, were significantly higher in fortilin^WT-MФ-HC than in fortilin^KO-MФ-HC mice (Fig. S2b). The serum level of IL-10, another anti-inflammatory cytokine⁴⁷, was also numerically higher—albeit not statistically significantly—in fortilin^WT-MФ-HC than in fortilin^KO-MФ-HC mice (Fig. S2b).

Imaging mass cytometry (IMC) analyses revealed that the atherosclerotic intima of fortilin^KO-MΦ-HC mice exhibited drastically fewer MΦ and more VSMCs

To further characterize the atherosclerotic lesions of fortilin^KO-MΦ-HC and fortilin^WT-MΦ-HC mice, we subjected the cross-sectioned aortic root tissue samples from these mice to IMC, which is a highly multiplexed, protein-level, single-cell-level, immunohistochemical analysis method^48,49. Fig. 3a–b show the workflow of this procedure, which is described in detail in the “Materials and Methods” section of the manuscript.

Fig. 3: Imaging mass cytometry (IMC) shows drastic reduction in MФ and increase in VSMCs in the atherosclerotic intima of fortilin^KO-MФ-HC mice compared with fortilin^WT-MФ-HC mice.

We first performed a cellularity assay in which we counted the number of cells localized within the intima using the Steinbock semiautomated cell segmentation strategy⁵⁰. We found that the atherosclerotic intima of fortilin^KO-MΦ-HC mice contained a smaller number of cells per unit surface area than that of fortilin^WT-MΦ-HC mice (WT vs. KO = 4823 ± 1192 vs. 2269 ± 1114 cells/mm², P = 0.004, Student’s t test) (Fig. 3d). We then performed a cell composition assay in which we pooled 1050 cells each from the intima of fortilin^WT-MΦ-HC and fortilin^KO-Φ-HC mice and, using the Uniform Manifold Approximation and Projection (UMAP) dimensionality reduction⁵¹, resolved the pooled cells into the three distinct cell types commonly identified in the atherosclerotic intima⁵²—endothelial cells (ECs, identified by CD31 positivity⁵³), MФ (identified by Mac2 positivity⁵⁴), and VSMCs (identified by alpha smooth muscle actin (αSMA)-positivity)⁵⁵) (Fig. 3e–g). We then determined the percentages of Mac2⁺αSMA^-CD31^-(MФ), Mac2^-αSMA⁺CD31^-(VSMCs), and Mac2^-αSMA^-CD31⁺(ECs) using the gating function of the Cytobank™ platform (Beckman Coulter Life Sciences, Carlsbad, CA, USA)^56,57, analyzing both the pooled sample set (Fig. 3e–h) and individual per-animal sample sets (Fig. 3i–l). We found that the atherosclerotic intima of fortilin^KO-MΦ-HC mice contained fewer MФ than that of fortilin^WT-MΦ-HC mice (Fig. 3e, h, i). Interestingly, the atherosclerotic intima of fortilin^KO-MΦ-HC mice contained more VSMCs compared to that of fortilin^WT-MΦ-HC mice (Fig. 3f, h, j), even after taking into account that the intima of fortilin^KO-MΦ-HC mice contained a lower total number of cells than that of fortilin^WT-MΦ-HC mice, in which the absolute numbers of VSMCs in the intima was greater in KO than WT mice (WT vs. KO = 159 VSMCs vs. 1069 VSMCs; Fig. 3d, f). ECs constituted a small portion of the intimal cells and were similarly represented in both fortilin^KO-MΦ-HC and fortilin^WT-MΦ-HC mice (Fig. 3g, h, k). The percentage of cells that did not express Mac2, αSMA, or CD31 (Mac2^-αSMA^-CD31^-) was higher in the intima of fortilin^KO-MΦ-HC mice than that of fortilin^WT-MΦ-HC mice (Fig. 3l). These data suggest that the lack of fortilin in MФ leads to an overall reduction in the cellularity in the intima (Fig. 3d) and a drastic change in its cell composition, with decreased MΦ and increased VSMC populations (Fig. 3h).

To further validate the above IMC findings, we quantified the expression levels of markers for MΦ (Mac2), VSMCs (αSMA), and ECs (CD31) in cross-sectioned aortic root tissues from fortilin^WT-MΦ-HC and fortilin^KO-MΦ-HC mice, using four-color immunofluorescence (IF) staining (Fig. 3m). Mac2 expression was significantly higher in fortilin^WT-MΦ-HC mice compared to fortilin^KO-MΦ-HC mice (Fig. 3n, Mac2), whereas αSMA expression was significantly lower in fortilin^WT-MΦ-HC mice than in fortilin^KO-MΦ-HC mice (Fig. 3n, αSMA). CD31 expression was comparable between two groups (Fig. 3n, CD31). These findings (Fig. 3m, n) are congruent with the IMC data (Fig. 3e–k), further supporting the conclusion that fortilin deficiency in MФ reduces MΦ content while increasing VSMCs, without significantly impacting the EC population.

In addition to identifying a difference in the percentages of MΦ, VSMCs, and ECs between the intima of fortilin^KO-MΦ-HC and fortilin^WT-MΦ-HC mice (Fig. 3e–n), we also calculated the expression levels of proteins of interest and normalized them to the cross-sectional intimal area (Fig. S3a, c–p). For example, we calculated the fortilin expression index (E.I.^fortilin) for a sample by summing up all fortilin signals within intimal cells of the sample, dividing the result by the cross-sectional intimal area of the sample, and expressing it as an arbitrary unit (A.U.). We repeated the same calculation for other proteins of interest from the samples of fortilin^WT-MΦ-HC and fortilin^KO-MΦ-HC mice and compared the means of the two groups using Student’s t test. As expected, E.I.^fortilin was significantly lower in the intima of fortilin^KO-MΦ-HC mice than in that of fortilin^WT-MΦ-HC mice (P = 0.024, N = 6–9, Fig. S3a).

Furthermore, when comparing fortilin expression levels between fortilin^WT-MΦ-HC and fortilin^KO-MΦ-HC samples across the three gated cell types—MΦ (Fig. 3e), VSMCs (Fig. 3f), and ECs (Fig. 3g)—we found that fortilin expression was significantly lower in the MΦ of fortilin^KO-MΦ-HC compared to those in fortilin^WT-MΦ-HC samples (Fig. S3b, columns 1&2, WT vs. KO = 0.15 ± 0.19 vs. 0.085 ± 0.077, P < 0.0001, Kruskal-Wallis test, N = 750 and 139). In contrast, fortilin expression levels in VSMCs and ECs did not differ between fortilin^WT-MΦ-HC and fortilin^KO-MΦ-HC samples (Fig. S3b, columns 3–6). These results suggest that the lower E.I.^fortilin observed in the intima of fortilin^KO-MΦ-HC mice compared to fortilin^WT-MΦ-HC mice is primarily driven by the reduced fortilin expression in MΦ rather than in VSMCs or ECs.

Using the same algorithm employed to calculate E.I.^fortilin, we determined the E.I.s for αSMA (VSMCs, Fig. S3c), CD31 (ECs, Fig. S3d), CD45 (all leukocytes, Fig. S3e), Mac2 (MФ, Fig. S3f), CD11b (MФ, Fig. S3g), F4-80 (MФ, Fig. S3h), CD11c (dendritic cells, Fig. S3i), CD3 (T cells, Fig. S3j), CD19 (B cells, Fig. S3k), B220 (B cells, Fig. S3l), CD68 (MΦ, M1 type, Fig. S3m), CD38 (MΦ-M1, Fig. S3n), and CD206 (MΦ-M2, Fig. S3o). Overall, we observed that the lack of fortilin^MΦ decreased the infiltration of MΦ and other immune cells (T cells, B cells, dendritic cells) into the atherosclerotic intima (Fig. S3e–o).

The lack of fortilin promoted oxLDL-induced apoptosis, slowed proliferation, and mitigated FC formation in MΦ in cell-based assays

To explore the mechanism by which the lack of fortilin^MΦ decreases the MΦ population in the atherosclerotic intima, we first tested whether the lack of fortilin sensitizes MФ to oxLDL-induced apoptosis. We incubated peritoneal MΦ from fortilin^WT-MΦ (MΦ^WT-fortilin) and MФ^KO-fortilin mice with oxLDL and subjected them to a standard DNA fragmentation assay. We found that MФ^KO-fortilin underwent significantly more apoptosis than did MФ^WT-fortilin at 125, 250, and 500 µg/mL of oxLDL (Fig. 4a; P = 0.014, 0.046, and 0.046, respectively; Student’s t test).

Fig. 4: The lack of fortilin in MФ promotes oxLDL-induced apoptosis, slows cell proliferation, halts foam cell (FC) formation, and reduces CD36 expression.

Next, we used Crispr Cas9 methods⁵⁸ to delete fortilin in THP1 cells (THP1^KO-fortilin; control = THP1^WT-fortilin) and then characterized the cells. Both western blot analysis and immunofluorescence staining showed that THP1^KO-fortilin did not express any fortilin (Fig. S4a, b). We then subjected THP1^KO-fortilin and THP1^WT-fortilin to vehicle or 200 µg/mL of oxLDL and measured the degree of DNA fragmentation. The DNA fragmentation index was 41% higher in THP1^KO-fortilin than in THP1^WT-fortilin at baseline (Fig. 4b, oxLDL(–), P < 0.0001, Student’s t test). Upon oxLDL challenge, THP1^KO-fortilin exhibited a 112% higher DNA fragmentation index (Fig. 4b, oxLDL(+), P < 0.0001, Student’s t test), suggesting that the lack of fortilin not only increased the baseline apoptosis of THP1 cells but also sensitized THP1 cells to oxLDL-induced apoptosis.

Intriguingly, in the whole-animal atherosclerotic tissue of 10-month-old mice, fortilin^WT-MΦ-HC mice exhibited higher levels of activated caspase-3 than fortilin^KO-MΦ-HC mice (Fig. S3p), which was driven by an increased presence of activated caspase-3 in both MΦ and ECs (Fig. S3q, columns 1 vs. 2, 5 vs. 6). These data, when taken together with the cellular data, suggest that MΦ apoptosis may have occurred earlier and more extensively in fortilin^KO-MΦ-HC mice than in fortilin^WT-MΦ-HC mice and that the full and cumulative extent of MΦ apoptosis in fortilin^KO-MΦ-HC mice may not be fully captured at the late time point of 10 months.

To evaluate how the lack of fortilin impacts the proliferation of MФ, we first subjected THP1^KO-fortilin and THP1^WT-fortilin to the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS) assay⁵⁹. The growth index of THP1^WT-fortilin was greater than that of THP1^KO-fortilin at 24, 48, 72, and 96 h (Fig. 4c; P < 0.0001 for all four time points, Student’s t test). When assessing the incorporation of 5-bromo-2’-deoxyuridine (BrdU) into peritoneal MФ from fortilin^WT-MФ-HC and fortilin^KO-MФ-HC mice, we consistently found that the lack of fortilin in MФ significantly decreased BrdU incorporation into these cells (Fig. 4d, P = 0.0029, Student’s t test). These data suggest that the lack of fortilin^MΦ increases ox-LDL-induced apoptosis (Fig. 4a, b) and decreases proliferation of MΦ (Fig. 4c, d).

Although the lack of fortilin^MΦ did not impact the ability of MФ to migrate (Fig. S4c) or phagocytose (Fig. S4d), we found that the presence of fortilin was required for MФ to become FCs. Our in vivo FC formation assay showed that the lack of fortilin^MΦ drastically reduced the ability of MФ to become FCs (Fig. 4e; fortilin^WT-MΦ-HC vs. fortilin^KO-MΦ-HC = 12.6 ± 2.2 vs. 2.7 ± 2.7 (A.U.), P = 0.0013, Student’s t test). To further validate the role of MФ fortilin in FC formation, we subjected THP1^KO-fortilin and THP1^WT-fortilin to a FC formation assay using flow cytometry. Again, we found that the lack of fortilin^MΦ decreased FC formation by about 95% (Fig. 4f; THP1^WT-fortilin vs. THP1^KO-fortilin = 27.33 ± 0.52 vs. 1.24 ± 0.00, P < 0.0001, Student’s t test). Next, we tested the status of CD36, an oxLDL receptor⁶⁰, by stimulating THP1^WT-fortilin and THP1^KO-fortilin cells with oxLDL and subjecting their lysates to JESS™, a highly quantitative capillary-based western blot system (ProteinSimple®, San Jose, CA, USA)²⁴. We found that THP1^WT-fortilin expressed more CD36 than THP1^KO-fortilin in the absence of oxLDL (Fig. 4g, lanes 1–3 vs. 4–6) and that oxLDL stimulation significantly increased CD36 expression in THP1^WT-fortilin (Fig. 4g, lanes 1–3 vs. 7–9, quantified in the graph on the right) but not in THP1^KO-fortilin (Fig. 4g, lanes 4–6 vs. 10–12, quantified in the graph on the right). These results suggest that the lower CD36 expression contributed to the reduced oxLDL uptake and FC formation in THP1^KO-fortilin. Finally, we followed a protocol published by Salina et al.⁶¹ to perform a flow cytometry-based efferocytosis assay to evaluate the impact of fortilin on efferocytosis. Efferocytosis is the process by which dying cells are recognized, engulfed, and cleared by MΦ to protect the microenvironment against inflammation caused by the release of previously sequestered intracellular contents from the disintegrated apoptotic cells^62,63. We added CellTrace™ Far Red-labeled (red), apoptosing Jurket cells to CellTrace™ CFSE-labeled (green) THP1^WT-fortilin and THP1^KO-fortilin cells in two rounds, incubated them, stained them with Zombie NIR (near-infrared, for the identification of dead cells)(Fig. 4h-1), and determined the percentage of CFSE⁺ZombieNIR^-FarRed⁺ cells (living THP1 cells that have engulfed apoptotic Jurket cells) within the total CFSE⁺ZombieNIR^- cell population (living THP1) using a flow cytometer (Figs. 4h-2, S4e). We found that the engulfment indices were similar between THP1^WT-fortilin and THP1^KO-fortilin cells (Fig. 4h-3, THP1^WT-fortilin vs. THP1^KO-fortilin = 53.20 ± 11.63 vs. 55.10 ± 8.93%, N = 4, P = 0.8043, Student’s t test), suggesting that fortilin does not affect efferocytosis. In summary, these data suggest that the lack of fortilin^MΦ decreases the MΦ population in the atherosclerotic intima by (i) increasing MΦ apoptosis, (ii) decreasing MΦ proliferation, and (iii) decreasing FC formation through the downregulation of CD36, an oxLDL receptor⁶⁰, and that the lack of fortilin^MΦ does not impact (i) migration, (ii) phagocytosis, or (iii) efferocytosis.

The lack of fortilin reversely transdifferentiated MΦ to VSMCs

Although the cellular data explain why the lack of fortilin^MФ led to drastic reduction in MФ in the intima in the whole-animal experiment (Fig. 3e–i; Fig. S3f–h, m, n), they do not necessarily explain the increase in the absolute number of VSMCs in the intima of fortilin^KO-MФ-HC mice (Fig. 3f, h, j) compared with that of fortilin^WT-MФ-HC mice. VSMCs contribute to the thickening of the fibrous cap of plaques and are considered to be protective against plaque rupture in mature atherosclerotic lesions⁶⁴. Because MФ reportedly can transdifferentiate to myofibroblasts⁶⁵, we tested the hypothesis that the lack of fortilin^MΦ transdifferentiates MФ to VSMCs. We stimulated THP1^KO-fortilin (Fig. 5a, KO) and THP1^WT-fortilin (Fig. 5a, WT) cells with either vehicle or oxLDL and assessed the expression status of VSMC markers (αSMA, smooth muscle 22 alpha (SM22α), and calponin-1 (CNN1))⁶⁶ as well as MФ markers (Mac2 (also known as galectin-3)⁶⁷ and CD68⁶⁸), using both RT-qPCR (Fig. 5b–f) and western blotting (Fig. 5g). Regardless of oxLDL stimulation, both assays showed that THP1^KO-fortilin expressed higher levels of VSMC markers compared to THP1^WT-fortilin (Fig. 5b–d, g-B, g-C, g-D). THP1^KO-fortilin expressed lower levels of MΦ markers than THP1^WT-fortilin (Fig. 5e, f, g-E, g-F).

Fig. 5: The lack of fortilin reverses transdifferentiated MФ into VSMCs.

To further verify these findings, we transduced primary human MΦ (phMΦ) with lentiviral particles carrying either control short-hairpin RNA (shRNA^control) or shRNA^fortilin, selected the transduced cells (phMΦ^sh-control and phMΦ^sh-fortilin) for 3 days on puromycin, and subjected their RNA to RT-qPCR for the above markers of VSMCs and MΦ. We found that shRNA^fortilin effectively silenced fortilin mRNA expression (Fig. 5h) and that phMΦ^sh-fortilin expressed significantly higher levels of VSMC markers compared to phMΦ^sh-control (Fig. 5i–k), regardless of oxLDL stimulation. phMΦ^sh-fortilin expressed significantly lower levels of Mac2, regardless of oxLDL stimulation (Fig. 5l), and CD68, when stimulated by oxLDL (Fig. 5m), compared to phMΦ^sh-control. These data suggest that fortilin^MΦ is required to maintain the expression of MΦ markers in MΦ and that the lack of it causes the cells to lose these markers and instead express VSMC markers. It is plausible that fortilin sustains MΦ phenotypes and that fortilin deficiency in MΦ results in the reverse transdifferentiation of MΦ to VSMCs in the atherosclerotic intima. In fact, Feil et al. reported that the transdifferentiation of VSMCs to MФ occurs during atherogenesis⁶⁹.

Finally, cytokine profiling using a multiplex bead-based assay system (Miiliplex®) showed that two clones of THP1^KO-fortilin (KO^C & KO^G) secreted less pro-inflammatory cytokines (IL-1β, IL-2, IL-7, IL-8, IL-12, granulocyte-macrophage colony stimulating factor (GM-CSF), IFNγ, TNFα⁷⁰; Fig. S5a, b, d, e, f, g, h, i, respectively) and more anti-inflammatory cytokines⁷⁰ (IL-4, IL-5⁷¹, IL-10, and IL-13; Fig. S5j, k, l, m, respectively) into the medium than did THP1^WT-fortilin. Interestingly, THP1^KO-fortilin secreted more IL-6, a presumably pro-atherosclerotic cytokine⁷², than did THP1^WT-fortilin cells (Fig. S5c).

Fortilin-deficient MΦ secreted TGF-β and promoted the differentiation of MSCs to VSMCs

The increase in VSMC mass in the atherosclerotic intima of the fortilin^KO-MΦ-HC mice (Fig. 3f, h, j) may be due to both increased transdifferentiation of fortilin-null MΦ to VSMCs and increased differentiation of MSCs to VSMCs. To test the hypothesis that a cytokine(s) secreted by THP1^KO-fortilin drives the differentiation of MSCs to VSMCs, we cultured human umbilical cord-derived MSCs (UCDMSCs) with the conditioned medium (CM) from either THP1^KO-fortilin or THP1^WT-fortilin cells for 72 and 120 h and evaluated the expression levels of VSMC marker genes (αSMA, SMA22α, & CNN1) using RT-qPCR (Fig. 6a). We used TGF-β1 as the positive control for the differentiation of MSCs to VSMCs^73,74,75,76 and plain media as negative controls.

Fig. 6: The lack of fortilin differentiates mesenchymal stem cells into VSMCs through TGFβ1.

We found that UCDMSCs expressed significantly higher levels of αSMA, SM22α, and CNN1 when cultured with the CM from THP1^KO-fortilin cells than with that from THP1^WT-fortilin cells at both 72 and 120 h (Fig. 6b–d; CM-THP1^KO-fortilin vs. CM-THP1^WT-fortilin, Student’s t test, P values shown in the figures). In addition, the expression levels of differentiation marker genes (CD45, CD34, and CD14) were significantly higher in UCDMSCs treated with the CM from THP1^KO-fortilin cells than with that from THP1^WT-fortilin cells at both 72 and 120 h, whereas the expression levels of stem cell marker genes (CD73, CD90, and CD105⁷⁷) were significantly lower in UCDMSCs treated with the CM from THP1^KO-fortilin cells than with that from THP1^WT-fortilin cells at both time points (Fig. 6e, f; Fig. S6a–d; CM-THP1^KO-fortilin vs. CM-THP1^WT-fortilin, Student’s t test, P values shown in the figures). Notably, the expression levels of the VSMC differentiation, and stem cell marker genes did not differ between UCDMSCs treated with medium only and those treated with the CM from THP1^WT-fortilin (Fig. 6e, f; Fig. S6a–d, M only vs. CM-THP1^WT-fortilin). This result suggests that the CM from THP1^WT-fortilin cells was incapable of inducing differentiation of UCDMSCs to VSMCs and that the CM from THP1^KO-fortilin cells induced differentiation of UCDMSCs to VSMCs.

Although these results suggested that a cytokine(s) secreted from THP1^KO-fortilin cells caused UCDMSCs to differentiate from the stem cell state into VSMCs, the specific cytokine responsible remained unidentified. TGF-β is produced by several different cell types, including ECs, VSMCs, MФ, platelets, and regulatory T cells (Treg)⁷⁸. Because serum TGFβ-1 and TGFβ-2 levels were higher in fortilin^KO-MФ-HC mice than in fortilin^WT-MФ-HC mice (Fig. 2f, g), because the atherosclerotic intima of fortilin^KO-MΦ-HC mice contained more VSMCs than did that of fortilin^WT-MΦ-HC mice (Fig. 3f, h, j), and because TGF-β promotes the differentiation of MSCs to VSMCs^73,74,75,76, we tested the hypothesis that TGFβ is that cytokine.

To evaluate this hypothesis, we first assessed whether MФ produce more TGF-β in the absence of fortilin than in its presence. We performed western blot on THP1^WT-fortilin and THP1^KO-fortilin cells with/without oxLDL stimulation and found that TGF-β1 expression was greater in THP1^KO-fortilin than in THP1^WT-fortilin cells and that the difference was greater with oxLDL stimulation (Fig. 5g-G). We then collected the CM of subconfluently seeded THP1^KO-fortilin and THP1^WT-fortilin cells stimulated by either vehicle or oxLDL and subjected them to enzyme-linked immunosorbent assays (ELISAs) of TGF-β1. We found that the CM from THP1^KO-fortilin cells contained more TGF-β1 than that from THP1^WT-fortilin cells, both with and without oxLDL stimulation (WT vs. KO = 311 vs 726 pg/mL for oxLDL (–); 400 vs. 1803 pg/mL for oxLDL (+), P < 0.0001 for both, Student’s t test) (Fig. 6g). We also found that the TGF-β1 release response from THP1^KO-fortilin increased more robustly upon oxLDL stimulation than that from THP1^WT-fortilin (oxLDL(–) vs. oxLDL(+) = 311 vs. 400 pg/mL for WT (28.6% increase, P = 0.271); 726 vs. 1803 pg/mL for KO (148.3% increase, P < 0.0001)) (Fig. 6g). These data suggest that fortilin deficiency causes MΦ to produce TGF-β1, especially in the presence of oxLDL, and they imply that fortilin deficiency has greater impact on the production of TGF-β1 from MΦ in the oxLDL-rich atherosclerotic vascular microenvironment than in the normal vascular microenvironment.

To definitively evaluate the above hypothesis, we then developed a TGF-β1 immunodepletion (ID) protocol in which CM from THP1^KO-fortilin and THP1^WT-fortilin, media alone, and media containing recombinant human TGF-β1 were incubated with α-TGF-β1 neutralizing monoclonal antibody. The immune complex was then precipitated by anti-mouse IgG-coated magnetic beads to generate immunodepleted CM and media (Fig. 6h). Western blot analysis confirmed that the TGF-β1 ID system was capable of depleting all detectable TGF-β1 (Fig. S6e, bands A1 vs. A2; A3 vs. A4; A5 vs. A6; and A7 vs. A8). Next, we incubated UCDMSCs for 120 h in one of the following conditions: media only, media containing TGF-β1, CM from THP1^KO-fortilin, or CM from THP1^WT-fortilin cells—each with or without TGF-β1 ID—and subjected their RNA to RT-qPCR to quantify VSMC marker gene expression (Fig. 6h). In the absence of ID, UCDMSCs robustly and significantly expressed VSMC marker genes (αSMA, SM22α, and CNN1) when incubated in media spiked with TGF-β1 compared to media alone (Fig. 6i–k-columns 1 vs. 3). In contrast, in the presence of ID, UCDMSCs failed to express any of the VSMC marker genes when incubated in TGF-β1-spiked media, further confirming the effectiveness of the ID system (Fig. 6i–k-columns 3 vs. 4). In this system, UCDMSCs expressed the VSMC marker genes when incubated in CM from THP1^KO-fortilin, but not in CM from THP1^WT-fortilin (Fig. 6i–k-columns 5 vs. 7), reinforcing the observations shown in Fig. 6b–d. Strikingly, ID of TGF-β1 completely abolished the ability of the CM from THP1^KO-fortilin cells to induce VSMC marker expression (Fig. 6i–k, columns 7 vs. 8). These data suggest that the induction of VSMC marker expression in UCDMSCs by CM from THP1^KO-fortilin is mediated by TGF-β1.

We also observed that both media spiked with TGF-β1 and CM from THP1^KO-fortilin cells without ID (columns 3 and 7, respectively), but not those with ID (columns 4 and 8, respectively), caused UCDMSCs to express the differentiation markers CD14 (Fig. S6f), CD34 (Fig. S6g), and CD45 (Fig. 6l) and that they prevented UCDMSCs from expressing the MSC markers CD73 (Fig. S6h), CD 90 (Fig. S6i), and CD105 (Fig. 6m). These findings suggest that TGF-β1 induces differentiation of UCDMSCs and, when taken together with the data from Fig. 6i–k, that the differentiation results in their transition to VSMCs.

Summary

The lack of fortilin^MΦ induces MΦ apoptosis (especially in high oxLDL environments), decelerates MΦ proliferation and lipid uptake, increases TGF-β1 secretion from MΦ, and TGF-β1-dependently transdifferentiates MΦ and differentiates MSCs into VSMCs. Collectively, fortilin^MΦ is pro-atherosclerotic (Fig. 7a) through its diverse biological activities, whereas the lack of fortilin^MΦ is anti-atherosclerotic (Fig. 7b).

Fig. 7: Profound phenotypic changes induced by fortilin in MΦ that make MΦ anti-atherosclerotic.

Discussion

We previously reported that fortilin global (constitutional) KO mice (heterozygous, fortilin^+/-) placed in the genetic background of the Ldlr^–/–Apobec1^–/– (HC) mice (fortilin^+/–&HC) developed less atherosclerosis than their control fortilin^+/+&HC mice²⁸. However, we still did not know which cell type(s) drove the anti-atherosclerotic phenotype of fortilin^+/-&HC mice. One of the most significant findings of the current study is that MΦ-specific fortilin KO mice (fortilin^KO-MΦ), when placed in the HC genetic background (fortilin^KO-MΦ-HC), developed drastically less atherosclerosis than their control fortilin^WT-MΦ-HC mice (Fig. 2a, b). This result strongly supports the premise that fortilin^MΦ drives atherosclerosis in the mouse model of atherosclerosis.

This reduction of atherosclerosis in fortilin^KO-MФ-HC mice was not due to favorable changes in the lipid profile, as lipid profiles of the two groups were similar (Fig. 1h–l; Fig. S1b–d). Body weights of fortilin^KO-MФ-HC mice were numerically mildly and statistically significantly greater than those of fortilin^WT-MФ-HC mice (Fig. 1g), making it unlikely that body weight played a role in the anti-atherosclerotic phenotype of fortilin^KO-MФ-HC mice as obesity promotes atherosclerosis⁷⁹. Complete blood counts showed no significant difference in white blood cell, monocyte, eosinophil, or basophil counts between the two groups (Fig. S1e). Although we detected a mild and statistically significant difference in neutrophil and lymphocyte counts between the two groups (Fig. S1e), it is unlikely that the small difference led to the drastically reduced atherosclerosis in fortilin^KO-MФ-HC mice. Erythrocyte and platelet counts were also similar between the two groups (Fig. S1f, g).

The mechanism by which fortilin deficiency in MΦ leads to the drastic reduction of atherosclerosis (Fig. 2a, b) was revealed by the IMC results, which showed a dramatic reduction of the MΦ population in the atherosclerotic intima in terms of both the absolute number and percentage of MΦ (WT vs. KO absolute number = 3439 vs. 300 cells/mm²; WT vs. KO percentage = 71.3% vs. 13.2%; these numbers were derived from the 1050 pooled cells from all samples; Fig. 3d, e, h, i; Fig. S3f–h, m, n). MΦ drives atherosclerogenesis: In the intima, MΦ take up oxLDL to become FCs, proliferate, produce pro-inflammatory cytokines to exacerbate local and systemic inflammation, and weaken the fibrous cap of the plaque by producing various proteases, ultimately leading to catastrophic plaque rupture and acute coronary syndrome⁸⁰. The idea that fortilin deficiency in MΦ results in the amelioration of atherosclerosis through robust reduction of the MΦ population in the atherosclerotic intima aligns with the current paradigm that reducing the MΦ population exerts plaque-stabilizing effects, at least in early-stage lesions⁶². Further investigation, however, is needed to elucidate the role of fortilin-deficient MΦ in the intima of the advanced atherosclerotic arteries.

Our cellular assays demonstrated that the depletion of the MΦ population in the atherosclerotic intima of fortilin^KO-MΦ-HC mice was multifactorial and due to (a) increased apoptotic MΦ death, especially upon oxLDL simulation (Fig. 4a, b), (b) decreased MΦ proliferation (Fig. 4c, d), and (c) decreased MΦ uptake of oxLDL (Fig. 4e & f), at least partially due to the downregulation of CD36, an oxLDL receptor⁶⁰, in fortilin-deficient MΦ (Fig. 4g). The cellular assays also suggested that fortilin deficiency is unlikely to impact efferocytosis (Figs. 4h, S4e), migration (Fig. S4c), or phagocytosis (Fig. S4d).

Fortilin deficiency in MΦ and resultant MΦ depletion in the atherosclerotic intima had a profoundly beneficial immuno-inflammatory impact at both the tissue and whole-animal levels. The absence of fortilin^MΦ drastically reduced the circulation of pro-inflammatory cytokines such as TNFα³², IFNγ^33,34, and IL-23³⁵ (Fig. 2c–e) while increasing the serum levels of the anti-inflammatory cytokines TGF-β1 and TGF-β2^36,37 (Fig. 2f, g). This shift likely disrupted—in fortilin^KO-MΦ-HC mice—the inflammation-perpetuating vicious cycle where tissue MΦ release pro-inflammatory cytokines into circulation that, in turn, recruits additional MΦ to the intima further amplifying local inflammation.

Importantly, while fortilin-deficiency in MΦ reduced their population in the intima and thereby substantially contributed to the reduction of systemic inflammation in whole animals (Fig. 2c–g), it also induced profound phenotypic changes in the MΦ themselves.

First, unlike THP1^WT-fortilin, fortilin-deficient MΦ (THP1^KO-fortilin) were incapable of producing pro-inflammatory cytokines (IL-1β, IL-2, IL-7, IL-8, IL-12, GM-CSF, IFNγ, TNFα⁷⁰; Fig. S5a, b, d, e, f, g, h, i, respectively). However, THP1^KO-fortilin secreted higher levels of anti-inflammatory cytokines⁷⁰ (IL-4, IL-5⁷¹, IL-10, IL-13, and TGF-β1^36,37; Fig S5j, k, l, m, Fig. 6g, respectively) into the medium compared to THP1^WT-fortilin.

Second, as demonstrated using both the THP1 human MΦ cell line (THP1^KO-fortilin and THP1^WT-fortilin) and phMΦ, fortilin deficiency induced the expression of VSMC marker genes (αSMA, SM22α, and CNN1; Fig. 5b–d, g-B–D, i–k) while suppressing MΦ marker gene expression (MAC2, CD68; Fig. 5e, f, g-E, g-F, l, m). These data suggest that fortilin-deficient MΦ transdifferentiate to VSMCs. However, further studies are needed to validate the notion that such transdifferentiation indeed occurs in fortilin-deficient MΦ in whole animals. For example, linage-tracing experiments could confirm that a subset of fortilin-deficient MΦ bearing a LyM-activated lineage marker differentiate into VSMCs in the atherosclerotic intima. Our cellular studies using both THP1 and phMΦ (Fig. 5), which support this possibility, provide a strong justification for conducting future in vivo lineage-tracing experiments.

The transdifferentiation of MΦ to VSMCs—a phenotypic transition characterized by the concurrent loss of MΦ markers (e.g., Mac2, CD11b) and the acquisition of VSMC markers (e.g., αSMA)—has been previously reported⁸¹. Zhuang et al. treated LysM-Cre;Rosa26^Td mice with angiotensin II and performed flow cytometry on single non-myocardial cells from the heart. Intriguingly, among the Td⁺ cells—representing the myeloid lineage cells with prior LysM expression—there were a population that expressed αSMA but lacked CD11b (Td⁺αSMA⁺CD11b^-). The findings suggest that cells of myeloid origin can transition to cells bearing only VSMC markers⁸¹, thereby supporting our current observation that MΦ can transdifferentiate into VSMCs (Fig. 5b-m).

Third, fortilin-deficient MΦ secrete TGF-β1 and direct the differentiation of MSCs into VSMCs (Fig. 6; Fig. S6). When UCDMSCs were cultured with conditioned media (CM) from THP1^KO-fortilin cells—but not from THP1^WT-fortilin cells—they gradually lost stem cell markers (CD73, CD90, and CD105; Fig. S6c, d; Fig. 6f), acquired differentiation markers (CD14, CD34, and CD45; Fig. S6a, b; Fig. 6e), and expressed VSMC markers (αSMA, SM22α, and CNN1; Fig. 6b–d). This differentiation was driven by TGF-β1 secreted by THP1^KO-fortilin cells because (a) THP1^KO-fortilin produced significantly more TGF-β1 than did THP1^WT-fortilin (Fig. 6g) and (b) immunodepletion (ID) of TGF-β1 from the CM of THP1^KO-fortilin nullified it ability to induce differentiation of UCDMSCs into VSMCs (Fig. 6i–k). The transdifferentiation of fortilin-deficient MΦ into VSMCs as well as the fortilin-deficient MΦ-induced differentiation of MSCs into VSMCs in the cellular system collectively manifested themselves in a significant increase in intimal VSMCs in the whole-animal system (WT vs. KO = 34 vs. 495 VSMCs/mm², P = 0.004; and 3.3% vs. 47.1%, P < 0.0001; respectively, Student’s t test; Fig. 3f, h, j).

One of the defining features of unstable, rupture-prone atherosclerotic plaques—identified by post-mortem and clinical imaging studies—is a thin or fragmented fibrous cap⁶⁴. The fibrous cap is the luminal layer of the atherosclerotic intima consisting of VSMCs and extracellular matrix overlying MΦ and a lipid core⁶⁴. The rupture of this cap in coronary arteries exposes blood stream to the highly thrombogenic lipid core, triggering platelet activation and aggregation, and initiation of the coagulation cascade and thrombus formation, ultimately resulting in MI⁸². An increased presence of VSMCs in the atherosclerotic intima is considered to confer stability to plaques⁶⁴. Our finding of a substantial increase of VSMCs in the atherosclerotic intima of fortilin^KO-MΦ-HC mice compared to fortilin^WT-MΦ-HC mice (Fig. 3f, h, j) suggests that targeting of fortilin^MΦ may stabilize plaques and make them less susceptible to the rupture. To test this hypothesis, however, further studies are needed to fully characterize plaque architecture in these mice, including the spatial distribution of VSMCs, as well as assessments of their proliferation, apoptosis, and senescence⁶⁴. Importantly, targeting of fortilin^MΦ may not be an effective strategy to mitigate neointimal proliferation following angioplasty or stent implantation in the atherosclerotic coronary arteries. This is because VSMCs constitute the majority of proliferating cells in the neointima^83,84, and the absence of fortilin in MΦ promotes both the transdifferentiation of MΦ (Fig. 5) and the differentiation of MSCs (Fig. 6) into VSMCs, potentially exacerbating neointimal growth.

IMC is a highly multiplexed immunohistochemical analysis method^48,49. It uses time-of-flight inductively coupled plasma mass spectrometry, which is capable of detecting dozens of protein markers simultaneously on a single tissue section. IMC achieves this by quantifying the abundance of metal isotopes tagged to antibodies and indexing them against their source location on the tissue section (Fig. 3a). Unlike single-cell RNA sequencing^85,86,87, IMC yields single-cell-level, protein-level data for both total and post-translationally modified proteins from the cells in their original locations (i.e., the intima or media). Using IMC, we obtained single-cell and protein-level data for over a dozen cell markers and intracellular proteins within the intima from the same tissue section. We subjected 1050 cells from each group to dimensionality reduction analyses using UMAP methods⁵¹ (Cytobank®), with the main goal of grouping MФ (Mac2⁺ cells), ECs (CD31⁺ cells), and VSMCs (αSMA+ cells)—the three most abundant cell types in the atherosclerotic intima—into distinct clusters (Fig. 3b). This particular strategy allowed us to perform a robust cell composition assay and to discover the unexpected increase in the VSMC population and decrease in the MΦ population in the intima of fortilin^KO-MΦ-HC mice compared to those of fortilin^WT-MΦ-HC mice (Fig. 3e, f, h, i, j).

In contrast to the drastic reduction of MΦ seen in the atherosclerotic plaques of fortilin^KO-MΦ-HC mice, we did not detect significant differences in circulating myeloid-derived cells (monocytes, eosinophils, or basophils) between fortilin^KO-MΦ and fortilin^WT-MΦ mice, except for a numerically mild but statistically significant reduction in neutrophils (Fig. S1e). This lack of difference persisted even though fortilin genes were deleted in the cell types in which the LysM promoter was activated (Fig. 1a). This finding might suggest that the biological consequences of fortilin deficiency are context dependent. In other words, the impact of fortilin deficiency may not be uniform across all cell types, as it is also impacted by the microenvironments in which they reside. For example, fortilin-deficient MΦ in the pro-inflammatory, pro-apoptotic (oxLDL-rich) microenvironment of atherosclerosis may undergo higher rates of apoptosis compared to fortilin-deficient monocytes and granulocytes, which circulate in the well-buffered, growth-factor-rich environment of whole blood.

For the mouse model of human atherosclerosis, we exclusively used Ldlr^–/–Apobec1^–/– (HC) mice, as they represent the most scientifically rigorous model of human familial HC with markedly elevated LDL levels^27,31. Although ApoE^–/– mice, which are a commonly used mouse model of human atherosclerosis, exhibit total cholesterol levels similar to those of Ldlr^–/–Apobec1^–/– mice (~560 mg/dL) and develop robust atherosclerosis on normal chow²⁷, most of their cholesterol resides in very low density lipoprotein (VLDL) and chylomicron, not in LDL. In contrast, most of the cholesterol in Ldlr^-/-Apobec1^-/- mice resides in LDL²⁷. In addition, ApoE deficiency can result in defective phagocytosis of apoptotic cells and increased inflammation⁸⁸.

In this study, we allowed fortilin^KO-MΦ-HC and fortilin^WT-MΦ-HC mice to develop atherosclerosis by feeding them for 10 months on a normal chow diet to obtain the most translationally relevant data. Although these mice develop atherosclerosis in an accelerated fashion within 3 months when placed on a high fat diet (HFD), the use of HFD inevitably leads to obesity, metabolic syndrome, alterations in gut microbiota⁸⁹, and excessive systemic inflammation⁹⁰, whereas human atherosclerosis typically develops over several decades in a low-grade chronic inflammatory microenvironment.

Additionally, we concluded all of the experiments by 10 months of age in these mice to mitigate the potential impact of senescence on the data. Mice begin to exhibit the senescence phenotype at 10–12 months of age, which was marked by the emergence of cells with high p16^Ink4a expression⁹¹. Several previous studies demonstrated that cellular senescence in MΦ, VSMCs, and ECs significantly contributes to the progression of atherosclerosis^92,93,94. Finally, in this context, the lower expression levels of CD38, which is a dual marker of pro-inflammatory^95,96,97 and senescent⁹⁸ MΦ, in the atherosclerotic tissue of fortilin^KO-MΦ-HC mice compared to fortilin^WT-MΦ-HC mice likely reflects the fact that fortilin deletion in MΦ reduced pro-inflammatory MΦ, considering that the experiment was concluded within 10 months of age.

We demonstrated that targeting fortilin^MΦ ameliorated atherosclerosis in mice without reducing serum cholesterol levels (Fig. 1h–l; Fig. S1b–d). Importantly, the MΦ-specific deletion of fortilin per se did not reduce mouse survival (Fig. 1d) or body weight (Fig. S1a), over the long term. Additionally, our previous work showed that constitutional knockdown of fortilin in all tissues in fortilin^+/–&HC mice mitigates atherosclerosis without reducing serum cholesterol levels or making the mice ill²⁸. These findings are significant because they suggest that fortilin, and fortilin^MΦ in particular, can be safely targeted as a novel, non-lipid, molecular target of atherosclerosis therapy. Further validation using a non-rodent model of human atherosclerosis, such as rabbits^99,100, would strengthen the evidence that fortilin—particularly fortilin^MΦ—promotes atherogenesis.

In clinical cardiology, the mainstay of atherosclerosis therapy has been lipid lowering using statins¹⁰¹, proprotein convertase subtilisin/kexin type 9 inhibitors^102,103, bempedoic acid¹⁰⁴, and ezetimibe¹⁰⁵. However, given that nearly 20% of patients admitted to US hospitals under the diagnosis of acute coronary syndrome had LDL levels <70 mg/dL (considered by many to be ideal levels)¹⁰⁶, it is unlikely that lipid-lowering therapy totally eliminates atherosclerosis and its complications. Although anti-inflammatory therapies using canakinumab (anti-IL-β1 antibody)¹⁰⁷, methotrexate¹⁰⁸, and colchicine¹⁰⁹ have been evaluated in clinical trials, they were always used with statins and associated with significant side effects, including fatal infections and sepsis¹⁰⁷.

As an alternative strategy, small-molecule fortilin inhibitors could effectively ameliorate atherosclerosis and enhance plaque stability by reducing the MФ population and increasing the VSMC population in the intima, suppressing both systemic and local inflammation without altering LDL levels. While these inhibitors can be selectively delivered to MΦ using nanoparticle-based¹¹⁰ or β-1,3-D-glucan particle-based¹¹¹ systems, they may also exert MΦ-specific anti-atherosclerotic effects without targeted delivery if they preferentially accumulate in MΦ, as observed with pexidartinib^112,113,114. The availability of fortilin^KO-MΦ-HC mice provides a valuable model to test the hypothesis that small-molecule fortilin inhibitors ameliorate atherosclerosis through selective inhibition of fortilin in MΦ.

Materials and methods

Animal study

We conducted all animal experiments using male mus musculus mice, 8 weeks of age or older, on the C57BL/6 J genetic background, and followed all relevant ethical regulations for animal use. Specifically, qualified vivarium staff and/or study scientists provided cage-side health monitoring and appropriate care throughout the studies to minimize pain and distress, including euthanasia when necessary, in accordance with the Organization for Economic Cooperation and Development (OECD) Humane Endpoints guidance. The pre-determined humane endpoints included rapid weight loss, severe lethargy or unresponsiveness, inability to eat or drink, persistent self-injury or abnormal posture, and labored breathing. Adverse events, whether expected or unexpected, were communicated and discussed among the animal care team and appropriate mitigation strategies were implemented. We euthanized mice first by CO₂ inhalation, followed by either exsanguination or a secondary CO₂ exposure, consistent with the recommendations of the American Veterinary Medical Association (AVMA) Panel on Euthanasia. The following sections describe individual experiments in detail.

Generation of macrophage (MΦ)-specific fortilin knockout (KO) mice

Generation of fortilin^flox/flox mice

We generated fortilin^flox/flox mice—in which the fortilin gene was flanked by the LoxP sequence to allow tissue-specific deletion—using the standard homologous recombination technique as previously described³⁰. We described the detailed process for the generation of fortilin^flox/flox mice in a previous publication²¹. Briefly, we first isolated a mouse BAC clone containing the full fortilin gene from the C57BL/6 J ES BAC clone library. We then constructed a fortilin targeting vector containing (a) the 4.8 kilobyte (kb) upstream genomic sequence, (b) the 3.6 kb genomic DNA sequence containing all six fortilin exons (exon 1–6), and (c) the 3.2 kb downstream genomic DNA sequence on the pVBFRTCKR targeting vector backbone²¹ that already contained (a) two LoxP sequences, (b) a phosphoglycerate kinase promoter-driven neomycin resistance gene flanked by two flippase recognition target (FRT) sequences, and (c) the thymidine kinase cassette.

We verified the integrity of the construct by extensive sequencing, linealized the vector using I-CeuI (R0699S, New England BioLabs, Ipswich, MA, USA), and electroporated it into C57BL6 embryonic stem cells (ESCs). We then identified the ESC clones with successful homologous recombination by subjecting the clones that survived in neomycin-containing medium to polymerase chain reaction (PCR) and Southern blotting (using radioactive DNA probes-1 and -2 after NdeI [R0111S, New England BioLabs] and NsiI [R0127S] digestion, respectively) analyses²¹. Next, we microinjected two of the mutated ESC lines (B43 and B61) into C57BL6 blastocysts and implanted them in pseudopregnant host mice to obtain chimeric mice. We crossed chimeric mice with C57BL6 mates and tested their offspring for germline transmission using PCR-based methods. Importantly, the resultant fortilin^flox/flox mice were fully in the C57BL/6 J genetic background from the beginning. Finally, we crossed fortilin^flox/flox mice with a transgenic flippase strain to remove the neomycin resistance gene cassette. Therefore, the fortilin^flox/flox mice we used in the current project did not contain the neomycin resistance gene cassette.

Generation of MΦ-specific fortilin KO mice

These mice were then crossed with C57BL/6 J mice overexpressing the Cre transgene under the control of LysM enhancer/promoter (Stock number 004781, Jackson Laboratory, Bar Harbor, ME, USA; backcrossed at least seven times to C57BL/6 J) to generate MΦ-specific deletion of fortilin as previously described¹¹⁵. We found that fortilin expression levels were equally low in both LysM-Cre^+/+fortilin^flox/flox mice and LysM-Cre^+/-fortilin^flox/flox mice (data not shown), suggesting that the removal of the knocked-in LoxP-fortilin-LoxP sequence occurred satisfactorily by heterozygous LysM-Cre transgene expression. We therefore performed all subsequent experiments using LysM-Cre^+/–fortilin^flox/flox (denoted fortilin^KO-MΦ hereafter) mice. The generation of fortilin^KO-MΦ and fortilin^WT-MΦ (LysM-Cre^–/–fortilin^flox/flox) mice was achieved by mating LysM-Cre^+/–fortilin^flox/flox and LysM-Cre^–/–fortilin^flox/flox mice, yielding fortilin^KO-MΦ and fortilin^WT-MΦ mice in a 1:1 ratio.

PCR-based genotyping procedure

We conducted genotyping of fortilin^KO-MΦ and fortilin^WT-MΦ mice on tail-derived genomic DNA using standard PCR-based methods. The presence of the LysM-Cre transgene was detected using the following primer sets: 5ʹ-CCCAGAAATGCCAGATTACG-3ʹ (Fig. 1a-2a, M1) and 5ʹ-CTTGGGCTGCCAGAATTTCTC-3ʹ (Fig. 1a-2a, M2). LysM-Cre^+/– mice yielded a 700 base pair (bp) amplified fragment, but LysM-Cre^–/– mice yielded no amplicons. The presence of the LysM wild-type allele was detected using the following primer sets: 5ʹ-TTACAGTCGGCCAGGCTGAC-3ʹ (Fig. 1a-2b, M3) and 5ʹ-CTTGGGCTGCCAGAATTTCTC-3′ (Fig. 1a-2b, M2). LysM-Cre^+/+ mice yielded no amplicon and LysM-Cre^–/– or LysM-Cre^+/– mice yielded a 350 bp amplicon. The presence of the LoxP-fortilin-LoxP knock-in construct was verified using the following primer sets: 5ʹ-TGGACC CTGACTTTCATCACCTC-3ʹ (Fig. 1a-1, F1) and 5ʹ-GTCATCTAACCTTACCCCAGTAAGC-3ʹ (Fig. 1a–1, F2). Fortilin^flox/flox mice yielded a 405 bp fragment and wild-type fortilin mice (fortilin^WT/WT mice) yielded a 280 bp fragment.

Survival and body weight analyses

We set aside 28 and 24 fortilin^WT-MФ and fortilin^KO-MФ male mice, respectively, and placed them on a normal chow diet for survival and body weight analyses. Up to two mice were housed in a cage. Mice were observed daily for appearance, behavior, and survival, and they were weighed every 4 weeks until all mice were dead.

Generation of fortilin^KO-MΦLdlr^–/–Apobec1^–/– (fortilin^KO-MФ-HC) and fortilin^WT-MΦLdlr^–/–Apobec1^–/– mice (fortilin^WT-MФ-HC)

To increase the serum concentration of low-density lipoprotein (LDL) to an atherosclerogenic level while consuming normal chow, we crossbred fortilin^KO-MΦ and fortilin^WT-MΦ mice with Ldlr^–/–Apobec1^–/–mice. Most of the cholesterol in these mice resides in LDL, not in chylomicrons or very-low-density lipoprotein (VLDL) as is the case for ApoE^–/– mice^27,31. We performed genotyping of Ldlr^–/–Apobec1^–/– mice on tail-derived genomic DNA using standard PCR-based methods. The status of the Apobec1 allele was assessed using the following primer sets: 5ʹ-TGAGTGAGTGGTGGTGGTAAAG-3ʹ and 5ʹ-CGAAATTCCTCCAGCAGTAAC-3ʹ. Apobec1^+/+ and Apobec1^+/− mice yielded a 475 bp amplified fragment, whereas Apobec1^−/− mice yielded no fragments. The status of the Ldlr allele was assessed using the following primer sets: 5ʹ-ACCCCAAGACGTGCTCCCAGGATGA-3ʹ and 5ʹ-CGCAGTGCTCCTCATCTGACTTGT-3ʹ. Ldlr^+/+ mice yielded a 383 bp fragment, whereas Ldlr^−/− mice yielded no fragments. We refer to fortilin^KO-MΦLdlr^–/–Apobec1^-/- and fortilin^WT-MΦLdlr^–/–Apobec1^–/– mice as fortilin^KO-MФ-HC (HC, hypercholesterolemic) and fortilin^WT-MФ-HC mice in this manuscript for the sake of simplicity.

Quantification of mRNA by real-time quantitative reverse transcription PCR (RT-qPCR)

We previously described the methods of RT-qPCR¹⁶. For RNA analysis of mouse MΦ, we harvested peritoneal MΦ after inducing them in thioglycolate in fortilin^WT-MΦ-HC and fortilin^KO-MΦ-HC mice, as previously described¹¹⁶. We isolated RNA from the kidney, liver, and lung of fortilin^WT-MΦ-HC and fortilin^KO-MΦ-HC mice using the GeneJet™ RNA Purification Kit (Catalog #: K0731, Thermo Scientific, Waltham, MA, USA), according to the manufacturer’s instructions. We performed RT-qPCR in quadruplicate with 50 or 100 ng of total RNA, using either the TaqMan® RT-PCR Kit (ABI at Life Technologies, Grant Island, NY, USA) or the QuantiNova Probe RT-PCR Kit (Catalog #: 208352, Qiagen, Germantown, MD, USA), on a QuantStudio3 system (Applied Biosystems, Foster City, CA, USA). We used the following primer and probe sets (Integrated DNA Technologies, Coralville, IO, USA):

Mouse Fortilin—Forward: 5ʹ- TCCGACATCTACAAGATCCGG-3ʹ, Reverse: 5ʹ- ATCTTGCCCTCCACCTCCA-3ʹ, Probe 5ʹ-FAM- AGATCGCGG/ZEN/ACGGGCTGTGC-IAbkFQ-3′ where FAM = carboxyfluorescein, IAbkFQ = Iowa Black FQ, and ZEN™ = an internal quencher to enhance the quenching activity of the 3′ quencher Iowa Black FQ;
Mouse 18S ribosomal RNA (rRNA)—Forward: 5ʹ-GCCGCTAGAGGTGAAATTCT-3ʹ, Reverse: 5ʹ-TCGGAACTACGACGGTATCT-3ʹ, Probe: 5ʹ-JOEN-ACCAGAGCG/ZEN/AAA GCATTTGCCAAG-IAbkFQ-3ʹ where JOEN = 6-carboxy-4′,5′-dichloro-2′,7′- dimethoxyfluorescein;
Mouse Col1a1—Forward: 5ʹ-GAAACCCGAGGTATGCTTGA-3ʹ, Reverse: 5ʹ-GTTGGGACAGTCCAGTTCTT-3ʹ, Probe: 5ʹ-FAM-GTGCGATGACGTGCAATGCA-3IABkFQ-3ʹ
Mouse Col3a1—Forward: 5ʹ-CCCTTCATCCCACTCTTATT-3ʹ, Reverse: 5ʹ- GATCCTGAGTCACAGACACATATT-3ʹ, Probe: 5ʹ-FAM-TCATCTACG/ZEN/TTGGACTGCTGTGCC-IABkFQ-3ʹ
Mouse Col4a1— Forward: 5ʹ-GCCAAGTGTGCATGATGAGAA-3ʹ, Reverse: 5ʹ-TACAATGGGAGGGAGAAGAG-3ʹ, Probe: 5ʹ-FAM-AACTATGAT/ZEN/GCTCGCCTCTGCCAC-IABkFQ-3ʹ
Human 18S rRNA—Forward: 5ʹ- CTGAGAAACGGCTACCACATC-3ʹ, Reverse: 5ʹ-GCCTCGAAAGAGTCCTGTATTG-3ʹ, Probe 5ʹ-FAM-AAATTACCCACTCCCGACCCGG-IAbkFQ-3ʹ;
Human FORTILIN—Forward: 5ʹ-ATGACTCGCTCATTGGTGGAA-3ʹ, Reverse: 5ʹ-TGCTTTCGGTACCTTCGCCC-3ʹ, Probe 5ʹ-FAM-TGCCTCCGC/ZEN/TGAAGGCCC-IAbkFQ-3ʹ;
Human αSMA—Forward: 5ʹ-GAGGTATCCTGACCCTGAAGTA-3ʹ, Reverse: 5ʹ-AAGCTCGTTGTAGAAGGTGTG-3ʹ, Probe 5ʹ-FAM/TATCGAGCA/ZEN/CGGCATCATCACCAA/IABkFQ/-3ʹ;
Human SM22α—Forward: 5ʹ-TGGAGATCCCAACTGGTTTATG-3ʹ, Reverse: 5ʹ-CATCCCAGACCCTCACTTTATC-3ʹ, Probe 5ʹ-FAM-CCTTAGAAC/ZEN/AAGCCACCTCCCACC/IABkFQ-3ʹ;
Human CNN1—Forward: 5′-GTCAGCCGAGGTTAAGAACAA-3′, Reverse: 5′-CTTCAGGATCTCCCACACTTTC-3′, Probe 5′-FAM-CCCGGCCTG/ZEN/ACAAAGAAATTGGG-IAbkFQ-3′;
Human MAC2—Forward: 5′-CCTCGCATGCTGATAACAATTC-3′, Reverse: 5′-CTCATTGAAGCGTGGGTTAAAG-3′, Probe 5′-FAM-CGGTGAAGC/ZEN/CCAATGCAAACAGAA-IAbkFQ-3′;
Human CD68—Forward: 5′-ACGCAACTGGCTCAAAGA-3′, Reverse: 5′-TCCCAAAGTGCTGGGATTAC-3′, Probe 5′-FAM-AAAGAAAGC/ZEN/CGGGCATGACGG-IAbkFQ-3′;
Human CD45—Forward: 5′-CGGCTGACTTCCAGATATGAC-3′, Reverse: 5′-GCTTTGCCCTGTCACAAATAC-3′, Probe 5′-FAM-TCCAGAAAG/ZEN/GCAAAGCCAAATGCC-IAbkFQ-3′;
Human CD105—Forward: 5′-GCAGGTGTCAGCAAGTATGA-3′, Reverse: 5′-GAAAGAGAGGCTGTCCATGTT-3′, Probe 5′-FAM-TCAGCAATG/ZEN/AGGCGGTGGTCAATA-IAbkFQ-3′;
Human CD34—Forward: 5′-TAGCCAAGTCTGCCAACTATTC-3′, Reverse: 5′-CCAACATACCACCCTCCATTT-3′, Probe 5′-FAM-AAAGGCAGA/ZEN/CCCGAATGTACCAGG-IAbkFQ-3′;
Human CD14—Forward: 5′-CTCAGAGGTTCGGAAGACTTATC-3′, Reverse: 5′-TTCATCGTCCAGCTCACAAG-3′, Probe 5′-FAM-CAGCAGCAG/ZEN/CAACAAGCAGGAC-IAbkFQ-3′;
Human CD73—Forward: 5′-ACGTTCTGGAACCACGTATC-3′, Reverse: 5′-CAGGTTCTCCCAGGTAATTGT-3′, Probe 5′-FAM-TTGTTGCGT/ZEN/TCATCAATGGGCGAC-IAbkFQ-3′;
Human CD90—Forward: 5′-CGGAAAGGAGAAGAGAAGAACA-3′, Reverse: 5′-TCCCAAAGTGGTGGGATTAC-3′, Probe 5′-FAM-CCTCACAAC/ZEN/GCAAACCTACCCAGA-IAbkFQ-3′;

We used the 2^–ΔΔCT method to calculate the expression levels of a gene in question relative to the 18S rRNA levels in the sample.

Protein analyses using western blots

We performed SDS-PAGE and western blot analyses as described previously^{9,14,15,117,118}. All antibodies were used with appropriate IRDye680LT- or IRDye800CW-conjugated secondary antibodies (LI-COR, Lincoln, NE, USA). The signal intensities of proteins bands were quantified using either the Odyssey Infrared Imaging System (LI-COR) or ChemiDoc MP System (Bio-Rad, Hercules, CA, USA). When appropriate (Figs. S4a, 5g, S6e), we assessed total protein loading by adding 2,2,2-trichloroethanol (TCE) to a polyacrylamide gel at the final concentration of 0.5% (v/v) before polymerization. After standard SDS-PAGE, the gel was ultraviolet (UV)-irradiated on the Bio-Rad ChemiDoc MP Imaging System for 2 min and the image of proteins on the gel was electronically captured¹¹⁹. We also used glyceraldehyde 3-phosphate dehydrogenase (GAPDH) as the loading control. The following primary antibodies were used:

Rabbit anti-fortilin (MBL International, Woburn, MA, USA; Catalog #: PM017; 1:1,000 dilution; used for the experiments depicted in Fig. 1c);
Rabbit anti-fortilin (Abcam, Waltham, MA, USA; Clone EPR5540, ab133568; 1:2,000 dilution; Figs. 5g, S4a);
Mouse anti-GAPDH (Fitzgerald, Acton, MA, USA; Clone: 10R-G109a; 1:10,000 dilution; Fig. 1c); mouse anti-GAPDH (Santa Cruz Biotechnology, Inc. Dallas, TX, USA; Clone: 6C5; Catalog #: sc-32233; 1:1,000 dilution; Figs. 5g, S6e);
Rabbit anti-human calponin-1 (CNN1) mAb (Cell Signaling Technology (CST), Danvers, MA, USA; Clone D8L2T, Catalog #: 17819; 1:1,000 dilution; Fig. 5g);
Recombinant rabbit anti-human α-smooth muscle actin (SMA) antibody (Abcam, clone EPR5368, Catalog #: ab124964; 1:1,000 dilution; Fig. 5g);
Rabbit anti-human SM22α (Transgelin/TAGLN) polyclonal Ab (CST, Catalog #: 40471; 1:1,000 dilution; Fig. 5g);
Rabbit anti-human TGF-β1 polyclonal Ab (Abcam, Catalog #: ab92486; 1:1,000 dilution; Figs. 5g, S6e);
Rabbit anti-Mac2 (Galectin-3/LGALS3) mAb (CST, Clone: D4I2R, Catalog #: 8785; 1:1,000 dilution; Fig. 5g);
Rabbit anti-human CD68 mAb MultiMab™(CST, multiple clones, Catalog #: 86985; 1:1,000 dilution; Fig. 5g).

For the quantification of CD36 expression in THP1^KO-fortilin and THP1^WT-fortilin cells, we used JESS™, a highly quantitative capillary-based western blot system (ProteinSimple®, San Jose, CA, USA)²⁴. Briefly, we evaluated protein loading using the Luminol-Peroxide mix (ProteinSimple®, Catalog #: 043-311 and 043-379) on the total protein detection module of the system. We calculated CD36 expression indices (in arbitrary unit [A.U.]) by dividing the area under the curve of CD36 by the total proteins loaded in the same capillary using Compass Simple Western™ software (version 6.3.0.)²⁴. We used the following primary and secondary antibodies:

Primary antibody: Rabbit anti-CD36 polyclonal antibody (Novus, St. Louis, MO, USA; Catalog #: NB400-145; 1:50 dilution)
Secondary antibody: anti-rabbit secondary to horseradish peroxidase (HRP) mAb (ProteinSimple®, Catalog #: 042-206; 1:1 dilution).

Quantification of atherosclerosis

We quantified the degree and extent of atherosclerosis using both en face and cross-sectional planimetry assays, as previously described^28,120,121. Before starting the experiment, we calculated that at least 11 mice per group would be required for the en face assays. This estimate was based on a one-sample Z-test assuming α < 0.05 (type I error), power >0.95 (type II error), a standard deviation (s.d., σ) of 5.20, as observed in our previous published atherosclerosis studies²⁸ (Minitab®, State College, Pennsylvania, USA). In that study, the absolute difference (Δ) in Oil-Red-O (ORO)-positive atherosclerotic area between control and study groups was 9.19%²⁸. To be conservative, we powered the study to detect a Δ of 6% and included an estimated attrition rate of ~10% over the study period. In the actual experiment, we used 17 mice per group. We assigned animals to one of the two groups, based on their genotypes; therefore, randomization was not performed. We housed 8-week-old fortilin^WT-MФ-HC (N = 17) and fortilin^KO-MФ-HC (N = 17) mice individually in an air-conditioned room with a 12:12 h light-dark cycle with free access to normal chow diet (Lab Diet, St. Louis, MO, USA) and water until they were euthanized at 10 months of age. All animals completed the study and were included in the analyses. Animals were weighed at the time of euthanasia. After euthanasia, we performed cardiocentesis to obtain whole blood. A portion of the blood was subjected to automated complete blood count (CBC) with differential, while sera from the blood was subjected to (i) lipid profile assays at the Mouse Metabolic Phenotype Center at the University of Cincinnati^28,120 and (ii) cytokine assays using the Milliplex® (EMD MilliporeSigma, Burlington, MA, USA) and Luminex™ 200 (Austin, TX, USA) systems¹²². After obtaining the blood, we harvested the entire aorta en bloc from the left ventricular (LV) outflow tract to the aortic valve to the iliac bifurcation and divided it into two portions: (a) the proximal portion including the left ventricle and the first 1 mm of the ascending aorta and (b) the remainder of the ascending aorta, aortic arch, descending aorta, and proximal iliac arteries.

Cross-sectional atherosclerosis analysis

The first portion was snap-frozen in OCT compound (Tissue-Tek, Sakura-Finetech, Torrance, CA, USA) and subjected to cryosectioning. The sections were cut perpendicular to the axis of the aorta from the left ventricular outflow tract toward the ascending aorta with frequent monitoring of the anatomy using hematoxylin and eosin (H&E) staining and microscopy. The first section was collected and mounted onto a glass slide as soon as the three aortic cusps became clearly visible. We mounted two 7 µm thick sections per slide and prepared 20 slides per mouse. The first slide from each mouse was stained with H&E and subjected to planimetric analysis using ImageJ (NIH, Bethesda, MD, USA), and data were expressed in µm². After cross-sectional atherosclerosis analysis was completed, the tissue samples were thawed and fixed in 4% paraformaldehyde (PFA) and embedded in paraffin blocks for imaging mass cytometry (IMC). A tissue microarray containing all samples was then generated and subjected to IMC as described below.

En face atherosclerosis analysis

The latter part of the aortae were cleaned of extravascular tissue and fat under a stereomicroscope, opened, and pinned flat on a white surface. We fixed the samples with 10% (v/v) buffered formalin solution overnight, stained them with freshly prepared and filtered Oil-red-O (ORO) solution for 1 h, rinsed them twice with 78% methanol, mounted and dried them on glass slides, and scanned them in the tagged image file (TIFF) format using the ScanScope Slide Scanning System (Nikon, Melville, NY, USA). The planimetry of the entire surface area and ORO positive atherosclerotic lesion areas was performed on the scanned images using Sigma Scan Pro software (SYSTAT Software, San Jose, CA, USA). The atherosclerosis index was calculated by dividing the ORO positive area by the entire surface area, and results were expressed in percentages.

Blood chemistry and CBC

We sampled whole blood by cardiocentesis using a 25-gauge needle immediately after euthanasia by CO₂. Each sample was transferred into a Microtainer® tube containing lithium heparin (BD Biosciences, Franklin Lakes, NJ, USA) by removing the needle from the syringe, pouring the blood into the tube, and mixing it thoroughly with the lithium heparin. CBC with differential was obtained using the HEMAVET 950FS Hematology System (Drew Scientific, Dallas, TX, USA).

Lipid and lipoprotein assays

The concentrations of total cholesterol, triglycerides, phospholipids, and non-esterified fatty acids were determined using standard enzymatic methods as described previously^28,120, while size fractionation of lipoprotein was achieved using fast-performance liquid chromatography. Both assays were performed at the Mouse Metabolic Phenotype Center at the University of Cincinnati^28,120.

Serum cytokine assays

The mouse serum cytokines and chemokines were simultaneously measured as described previously¹²² and using a MILLIPLEX^® MAP Mouse Cytokine/Chemokine Magnetic Bead Panel according to manufacturer’s instructions (Catalog#: MCYTOMAG-70K, MilliporeSigma). Briefly, the diluted serum samples, standards, and controls were mixed with assay buffer and premixed 32-plex beads in a 96-well plate. The plate was then incubated at 4 °C overnight with shaking. The well contents were removed, and the plate was washed twice with wash buffer. The mixture of detection antibodies was then added to each well and the mixture was incubated for 1 h at room temperature, followed by the addition of streptavidin-phycoerythrin. After a 30 min incubation, the plate was washed twice and sheath fluid was added to each well prior to analysis using the Luminex 200™ System (Luminex-DiaSorin, Saluggia, Italy) (Fig. 2c–e; Fig. S2a, b). TGF-β1 and β2 concentrations in the mouse sera were measured in the same way using a MILLIPLEX® TGFβ Magnetic Beads 3 Plex Kit (Catalog #: TGFBMAG-64K-03, MilliporeSigma) (Fig. 2f, g). Cytokines and chemokines contained in the conditioned media (CM) from THP1^KO-fortilin and THP^WT-fortilin human monocyte-macrophage cell lines were analyzed using a MILLIPLEX^® MAP Human Cytokine/Chemokine Magnetic Bead Panel according to manufacturer’s instructions (Catalog #: HSTCMAG28SPMX13, MilliporeSigma) and as described above (Fig. S5a–m).

Enzyme-linked immunosorbent assay (ELISA)

We quantified human TGF-β1 in the CM of THP1^WT-fortilin and THP1^KO-fortilin cells (Fig. 6g) using a TGF-β1 ELISA kit (Invitrogen Thermo Fisher, Waltham, MA, USA; Catalog #: BMS249-4) as described previously¹¹⁹ and following the manufacturer’s instructions.

IMC

Tissue preparation

The workflow of IMC is shown in Fig. 3a, b and in a previously published report¹²³. After sacrifice, we embedded the very proximal portion of the ascending aorta, including the left ventricular outflow tract, in the OCT compound and snap-froze it on dry ice before subjecting it to cryosectioning and cross-sectional atherosclerosis assays. Subsequently, we thawed the tissue samples, fixed them in 4% PFA, and embedded them in paraffin blocks to generate a formalin-fixed paraffin embedded (FFPE)-tissue microarray (TMA) for IMC.

IMC antibody testing, conjugation, validation, and titration

We tested all antibodies listed below using immunofluorescence on samples of mouse spleen, liver, heart, and atherosclerotic aorta. We selected antibodies that yielded specific and high-intensity signals in a pattern that is appropriate for the given antigen and conjugated them to their respective metals using the MaxPar X8 Multimetal Labeling Kit (Fluidigm, San Francisco, CA, USA). Next, we tested these metal-conjugated antibodies in IMC using the mouse tissues and confirmed that each antibody continued to generate specific signals. We also generated an FFPE TMA that included all available mouse aorta samples. Finally, we subjected the FFPE TMA to staining with all metal-conjugated antibodies and visually examined each channel to ensure that the dilution used in the staining produced specific signals with the appropriate signal-to-noise ratio. In some cases, we increased the dilution of the antibody that exhibited a significant signal spillover to the adjacent channels. We repeated the process until we obtained strong and specific signals with an acceptable signal spillover from the adjacent channels. We used the following metal-conjugated antibodies for tissue staining (conjugated metal is indicated in the parenthesis; *, pre-made and validated at the Single Cell Imaging and Mass Cytometry Analysis Platform at McGill University; **, custom conjugated):

anti-(α-)Ly6G (¹⁴¹Pr) Ab*, Clone 1A8, Abcam (Catalog #: ab210204);
α-Ki67 (¹⁴³Nd) Ab*, Clone SP6, Abcam (Catalog #: ab16667);
α-B220 (¹⁴⁴Nd) Ab*, Clone RA3-6B2, Abcam (Catalog #: ab233273);
α-pSTAT (¹⁴⁵Nd) Ab**, Clone 15H13L67, Invitrogen (Catalog #: 700349);
α-CD31 (¹⁴⁶Nd) Ab*, Clone RM0032-1D12, Abcam (Catalog #: ab56299);
α-CD45 (¹⁴⁷Sm) Ab*, Clone D3F8Q, CST (Catalog #: 98819);
α-p53 (¹⁴⁸Nd) Ab**, Clone EPR20416-120, Abcam (Catalog #: ab246550);
α-CD11b (¹⁴⁹Sm) Ab*, Clone EPR1344, Abcam (Catalog #: ab209970);
α-CD19 (¹⁵⁰Nd) Ab**, Clone EPR23174-145, Abcam (Catalog #: ab245235);
α-CD3e (¹⁵²Sm) Ab**, Clone CAL57, Abcam (Catalog #: ab237721);
α-CD8α (¹⁵³Eu) Ab*, Clone D4W2Z, CST (Catalog #: 60168)
α-CD11c (¹⁵⁴Sm) Ab*, Clone D1V9Y, CST (Catalog #: 39143)
α-TGF-β1 (¹⁵⁵Gd) Ab**, Clone EPR21143, Abcam (Catalog #: ab229856);
α-FoxP3 (¹⁵⁸Gd) Ab*, Clone D6O8R, CST (Catalog #: 72338);
α-F4/80 (¹⁵⁹Tb) Ab*, Clone BM8, Abcam (Catalog #: ab16911);
α-fortilin (¹⁶¹Dy) Ab**, Clone EPR5540, Abcam (Catalog #: ab133568);
α-p-Smad3 (¹⁶³Dy) Ab**, (Clone EP823Y), Abcam (Catalog #: ab52903);
αiNOS (¹⁶⁴Dy) Ab*, Polyclonal, Abcam (Catalog #: ab15323);
α-CD68 (¹⁶⁵Ho) Ab**, Clone EPR23917-164, Abcam (Catalog #: ab283667);
α-Mac2 (¹⁶⁷Er) Ab**, Clone M3/38, Cedarlane, Burlington, Ontario, Canada (Catalog #: CL8942AP);
α-CD206 (¹⁶⁹Tm) Ab*, Polyclonal, Abcam (Catalog #: ab64693);
α-CD3 (¹⁷⁰Er) Ab*, Clone Cd3-12, Abcam (Catalog #: Ab255972);
α-CD38 (¹⁷¹Yb) Ab**, Clone EPR21079, Abcam (Catalog #: ab216343);
α-activated caspase 3 (¹⁷²Yb) Ab*, Clone 5A1E, CST (Catalog #: 9664);
α-MHC-II (¹⁷⁴Yb) Ab*, Clone 28148, Abcam (Catalog #: ab25228);
α-αSMA (¹⁷⁵Lu) Ab, Clone 1A4, Abcam (Catalog #: ab7817).

Tissue staining

For the final staining, we generated two back-to-back sections from the FFPE TMA within a week of the actual staining. One was subjected to standard H&E staining to determine the regions of interest (ROI), while the other one underwent IMC metal labeling as previously described¹²³. For metal labeling, we subjected the FFPE TMA to deparaffinization and heat-mediated antigen retrieval on the Ventana Discovery Ultra auto-stainer platform (Roche Diagnostics, Indianapolis, IN, USA) using EZ Prep solution (Roche Diagnostic, Catalog #: 950-102) and ULTRA Cell Conditioning 1 solution (Roche Diagnostic, Catalog #: 950-224), respectively. After incubating the sample at room temperature in Dako Serum-free Protein Block solution (Agilent, Santa Clara, CA, USA; Catalog #: X090930-2) for 45 min, we incubated it at 4 °C overnight with a cocktail of the 26 primary antibodies, all of which were conjugated to their respective metals and added to Dako Antibody Diluent solution at the optimized dilutions. The next day, we counterstained the sample with Cell-ID Intercalator-Iridium (Fluidigm), rinsed it in distilled water, air-dried it, and stored it at 4 °C until image acquisition.

Selection of ROI

Using H&E staining, we determined the ROI containing the lumen, intima, and media of each aorta section, which we used to set the ROI coordinates for IMC.

Image acquisition

First, we obtained optical images of the slides using Hyperion software and, based on the previously selected ROI on the H&E slides, selected the areas for UV laser ablation. Using the established protocol at the McGill University Goodman Cancer Institute Single Cell Imaging and Mass Cytometry Analysis Platform (Montreal, Canada), we acquired images using the Hyperion Imaging System (Fluidigm) by inserting the stained slide into the Hyperion laser ablation chamber and allowing the system to generate UV laser pulses at a frequency of 200 Hz, which were directed to the pre-selected ROI¹²⁴. Each UV laser pulse ablated a 1 µm² area of tissue within the ROI that was located at a certain coordinate. The pulse evaporated and ionized the area, and the fume was vacuum carried to a Helios time-of-flight mass cytometer (Fluidigm), where the amounts of metal-tagged antibodies in the fume were quantified and the signals were assigned to the corresponding coordinate.

IMC data processing, image segmentation, pixel classification, and cell segmentation

We used the Fluidigm/Standard BioTools, Inc. MCD™ viewer to convert text format files generated during data acquisition into tiff image files. We then used CellProfiler (Broad Institute of the Massachusetts Institute of Technology, Boston, MA, USA)¹²⁵ to identify individual nuclei, define cell borders, and generate cell masks using the Steinbock protocol⁵⁰. Focusing only on the cells contained in the intima, we determined the median signal intensity of each antigen for the cell and exported the results in two comma-separated-values (CSV) tables—one for cells from KO mice and the other for cells from WT mice. Each row of the CSV tables represented a single cell, and the columns contained the information specific to the cell, including the signal intensities of various antigens divided by the surface area of that cell. Although the tissue samples were stained by 26 different metal-conjugated antibodies, we found that the signal intensities were adequate only for CD31, CD45, CD11b, fortilin, αSMA, Mac2, F4-80, CD11c, CD3, CD19, CD68, CD206, CD38, and B220. One stained sample (M387-1, WT) exhibited severe staining irregularity and was removed from the analysis. In two samples (M228-1, KO; M396-2, KO), part of the atherosclerotic intima detached from the rest of it, and we removed them from the analysis.

Cellularity analysis

We determined the number of nuclei within the atherosclerotic intima of each animal and divided it by the cross-sectional area of the intima to calculate the cellularity index (cells/mm²).

Single cell dimensionality reduction analysis

To determine the composition of the atherosclerotic intima, we first uploaded the CVS tables—Steinbock-WT and Steinbock-KO—to the Cytobank server (Beckman-Coulter, Indianapolis, IN, USA) and converted them to flow cytometry standard (fcs) files. To visualize the three predominant cell types of the atherosclerotic intima (MФ, VSMCs, and ECs), we performed a dimensionality reduction analysis on the pooled WT and KO intimal cells (1050 cells each) using the Uniform Manifold Approximation and Projection (UMAP) algorithm with the parameters set as follows: the number of neighboring sample points = 100; the effective minimum distance between embedded points = 0.6; UMAP channels = αSMA (VSMCs), Mac2 (MФ), and CD31 (ECs); and event sampling = equal (1050 cells for WT and KO each). We visualized the data using “En Fuego” as a plot color scheme and stacked dots as a dot plotting method with the dot size of 2. We then calculated the percentage of ECs (CD31⁺), MФ (Mac2⁺), and VSMCs (αSMA⁺) for pooled WT and KO intimal cells using Gating Editor of the Cytobank platform.

Cell profiling analysis

In addition to the above analysis of the pooled cells, we also performed per-animal analyses. In this assay, we summed all signals within cells and divided them by the area of the intima to calculate the expression index (E.I.), as shown in the following equation:

$$E.I.\left(A.U.\right)=\sum \left(S1+S2...{Sn}\right)\div\,A$$

where S1 represents the signal intensity of a certain antigen in cell number one; A is the cross-sectional area of the intima that contains cell number one through number n; and all cells from an E.I. come from a single animal.

Immunofluorescence staining

We performed four-color immunofluorescence staining to visualize MΦ, VSMCs, and ECs in the atherosclerotic intima and to validate the IMC findings. Formalin-fixed, paraffin-embedded (FFPE) aortic tissue slides from male fortilin^WT-MΦ-HC and fortilin^KO-MΦ-HC mice (age 29–34 weeks, n = 5 per group) were first baked at 60°C for 1 h, followed by deparaffinization in xylene. Antigen retrieval was performed by incubating the slides in ULTRA Cell Conditioning 1 (CC1) solution (Roche Diagnostic, Catalog #: 950-224) diluted 1:1 in ddH₂O, at 95°C for 36 min in the EZ-Retriever® tissue-staining microwave (BioGenex, Fremont, CA, USA), then allowing the slides cool to room temperature. We then blocked non-specific binding with Dako Serum-free Protein Block (Agilent, Santa Clara, CA, USA; Catalog #: X090930-2) for 30 min at room temperature. Tissues were incubated overnight at 4°C in a humidified chamber with the following mAbs, each diluted in Dako Antibody Diluent (Agilent, Catalog #: S202230-2):

(1)
α-Mac2 mAb conjugated to Alexa Fluor 647 (Clone M/38; BioLegend, Catalog #: 125408; 1:100 dilution)—visualized using a red laser (638 nm) and Cy5 filter, assigned magenta.
(2)
α-αSMA mAb conjugated to Alexa Fluor 488 (Clone 1A4; BioLegend, Catalog #: 614854; 1:100 dilution)—visualized using a green laser (488 nm) and FITC filter, assigned green.
(3)
α-CD31 mAb conjugated to Alexa Fluor 555 (Clone D8V9E; Cell Signaling Technology, Catalog #: 89266; 1:100 dilution)—visualized using a yellow laser (561 nm) and TRITC filter, assigned red.
(4)
4’,6-diamidino-2-phenylindole (DAPI; Invitrogen, Catalog #: 62248; 1: 1,000 dilution)—visualized using a UV laser (405 nm) and DAPI filter, assigned blue.

The following day, we washed the slides in PBS, stained nuclei with DAPI, washed them again in PBS, and mounted the coverslips using ProLong™ Gold Antifade Mountant (Thermo Fisher Scientific, Catalog #: P36930). We visualized and captured the fluorescence signals using a Nikon A1R confocal microscope controlled by NIS-Elements AR software (version 5.11.01, Nikon, Tokyo, Japan) employing appropriate lasers and filter sets for each fluorophore as described above. We analyzed the image in ImageJ (v1.54 f NIH, Bethesda, MD, USA)¹²⁶ to quantify the expression levels of Mac2, αSMA, and CD31 within the atherosclerotic intima defined as the region of interest (ROI). We calculated the expression index for each cell marker by dividing the total area of pixels with signal intensity above the background threshold by the total ROI area, multiplying by 100, and expressing the result in arbitrary units (A.U.).

Cell culture and cell lines

Mouse peritoneal MΦ

We induced peritoneal MΦ by intraperitoneally injecting 1 mL of 3% thioglycollate (Fluka-MilliporeSigma, Catalog #: B2551, Lot #: BCBL8096V) into fortilin^WT-MΦ (Fig. 4a, S4c), fortilin^KO-MΦ (Fig. 4a, S4c), fortilin^WT-MΦ-HC (Fig. 4d, e), or fortilin^KO-MΦ-HC (Fig. 4d, e) mice, incubating them for 4 days, sacrificing them, and collecting the cells from the abdominal cavity using a Pasteur pipette under sterile conditions. We seeded the cells on cell culture plates in high-glucose Dulbecco’s Modified Eagles Medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (FBS, MilliporeSigma, Catalog #: F4135), 1 x GlutaMAX™ (Gibco, Waltham, MA, USA; Catalog #: 35050079), and 1% penicillin-streptomycin in a humidified incubation chamber with 5% CO_2. We allowed the cells to attach to the plastic plate for 4 h, gently washed off non-adherent cells with phosphate buffered saline (PBS) and cultured them in the same DMEM for the oxLDL-induced DNA fragmentation assay, BrdU assay, and migration assay.

Mouse bone marrow (BM) derived macrophages (BMDMΦ)

We generated BMDMФ from the BM of fortilin^WT-MΦ and fortilin^KO-MΦ mice as previously described¹²⁷. Briefly, we flushed out BM from the femur and tibia bones from the 6–8-week-old mice using cold PBS with 2% FBS. After passing BM cells through a 21-gauge needle several times and through a 70 µm cell strainer, we added three volumes of 0.8% NH₄Cl solution (Stem Cell Technology, Vancouver, BC, Canada), incubated the mixture on ice for 10 min, and spun down the cells by centrifugation. We resuspended and seeded the cells in the BDBMФ growth medium consisting of Iscove’s Modified Dulbecco’s Medium, 10% FBS, and 10 ng/ml MΦ colony stimulating factor (M-CSF). We changed the medium using fresh BMDMФ growth medium on day 3. On day 7, we performed flow cytometry on the cells to ensure that they were expressing both CD11b and F4/80 on their surface. We used the cells to perform the phagocytosis assay. We maintained these cell lines in DMEM supplemented with 10% FBS, 1 x GlutaMAX™, and 1% penicillin-streptomycin in a humidified incubation chamber with 5% CO₂.

Fortilin-deficient THP1 cells and their control (THP1^KO-fortilin and THP1^WT-fortilin)

THP1 cells whose fortilin was knocked out (THP1^KO-fortilin) by targeting exon 3 of the fortilin gene using the standard CRISPR-Cas9 system^58,128 and the exon-targeting single guide RNA (TTTCGGTACCTTCGCCCTCG) were obtained from Synthego (Redwood City, CA, USA). The final clones were characterized using indel analysis, western blot analysis, and immunofluorescence staining. For the indel analysis, purified genomic DNA from THP1^KO-fortilin clonal and homozygous KO cells were amplified with PCR using the following primers flanking the target cut site:

Forward PCR Primer: 5’–TATACCCACTGCGAAAGAACCT–3’;
Reverse PCR Primer: 5’–CGTGAACTATTGGCGTGGAAG–3’

The gel-purified amplicons were subjected to Sanger sequencing using the forward PCR primer. Sequences around the target cut site were then evaluated using the ICE CRISPR Analysis Tool (Synthego) to show the insertion of one nucleotide ( + 1) at the site. A mock-transfected pool of THP1 cells was provided by Synthego and used as the control (THP1^WT-fortilin). Both THP1^WT-fortilin and THP1^KO-fortilin human monocyte-MΦ cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium supplemented with 1x GlutaMAX™, 1.0% penicillin-streptomycin, and 10% FBS. For all experiments described in this manuscript, these cells were differentiated fully to MΦ by incubating them in the above medium with 5 ng/mL phorbol 12-myristate-13-acetate (PMA, Sigma, Catalog #: 8139) for 48 h, as described previously¹²⁹.

Human umbilical cord-derived mesenchymal stem cells (UCDMSCs)

UCDMSCs were obtained from the ATCC (Manassas, VA, USA; Catalog #: PCS-500-010) and cultured in the complete mesenchymal stem cell medium (“complete MSC medium”)—consisting of Mesenchymal Stem Cell Basal Medium (ATCC, PCS-500-030) supplemented with the Mesenchymal Stem Cell Growth Kit (ATCC, Catalog#: PCS-500-040), resulting in final concentrations of 2% FBS, 5 ng/mL human recombinant (rh) FGF basic, 5 ng/mL rh FGF acidic, 5 ng/mL rh epidermal growth factor, 2.4 mM L-alanyl-L-glutamine, and 100 units/mL penicillin-streptomycin—in flasks coated with 0.04% gelatin (MilliporeSigma, Catalog #: G1393)¹³⁰. UCDMSCs were passed every other day at a ratio of 1:2. For all assays, we used cells below passage 15.

Immunocytochemistry

For immunocytochemical characterization of THP1^WT-fortilin and THP1^KO-fortilin (Fig. S4b), we seeded the cells into 4-chamber slides (Thermo Fisher Scientific Nunc Lab-Tek II, Catalog #: 154526PK), treated them with PMA (100 ng/mL) for 48 h, and fixed them with 4% PFA for 20 min. The cells were permeabilized with 0.1% Triton-X-100 in PBS for 15 min and blocked using 10% goat serum in PBS with Tween for 1 h. Rabbit anti-fortilin antibody (1:500 dilution, Abcam, Catalog #: ab133568) was incubated with cells overnight at 4°C. After incubation with the primary antibody, Alexa Fluor 488 goat anti-rabbit IgG (F + L) (1:2000 dilution, Invitrogen, Catalog#: 2179202) was applied for 1 h. The nucleus was stained with 4’,6-diamidino-2-phenylindole (DAPI). Imaging was performed using a Nikon A1R confocal microscope controlled by NIS-Elements AR image acquisition software (version 5.11.01, Nikon, Tokyo, Japan).

DNA fragmentation assay

The amount of DNA fragmentation, which is a marker of apoptosis, in peritoneal MΦ (Fig. 4a) and THP1^WT-fortilin and THP1^KO-fortilin (Fig. 4b) was quantified using an ELISA kit (Cell Death Detection ELISA PLUS Cat # 11774425001, Roche, Indianapolis, IN, USA) following the manufacturer’s instructions and as previously described^16,22. Briefly, cells were treated with various concentrations (0, 125, 250, or 500 µg/mL for peritoneal MΦ; 200 µg/mL for THP1^WT-fortilin and THP1^KO-fortilin) of oxLDL (Kalen Biomedical, Montgomery Village, MD, USA) and lysed by lysis buffer provided in the kit. Twenty microliters of each cell lysate and 80 μL of each assay buffer containing anti-histone-biotin and anti-DNA-peroxidase were then dispensed into streptavidin-coated wells of a 96-well plate in five sets for peritoneal MΦ and in triplicate for THP1^WT-fortilin and THP1^KO-fortilin cells. After incubation and wash, captured nucleosomes were detected using the 2,2’-azino-di (3-ethylbenzthiazoline-6-sulfonic acid method), with signal intensity measured at 405 nm (A405). The DNA fragmentation index was calculated as (A405^study – A405^media) / (A405^{positive control} – A405^media) × 100 and expressed as arbitrary unit (A.U.), where A405^study represents A405 of the study cell sample, A405^media is theA405 of the medium only sample, and A405^{positive control} is the A405 of medium containing DNA-histone-complex provided in the kit.

BrdU proliferation assay

We induced peritoneal MФ in fortilin^WT-MΦ-HC and fortilin^KO-MΦ-HC mice as described above. BrdU incorporation into MФ was determined using a BrdU Cell Proliferation Kit (Calbiochem, San Diego, CA, USA; Catalog #: QIA58, Lot #: D00161810) according to the manufacturer’s instructions. Briefly, 2 × 10⁴ cells from each group were seeded in 6 wells of a 96-well cell culture plate and allowed to adhere to the well. Cells were labeled with BrdU for 24 h before they were fixed and denatured using the fixative/denaturing solution provided in the kit. Cells were then incubated with 1x anti-BrdU antibody for 1 h and, after being washed, they were incubated with goat anti-mouse IgG antibody conjugated to HRP for 30 min. Bound anti-BrdU antibody-anti-mouse-IgG complex was detected by the HRP substrate tetra-methylbenzidine and quantified using a SpectraMax spectrophotometer (Molecular Devices, Sunnyvale, CA, USA) at wavelengths of 450–540 nm. The BrdU index of MФ from fortilin^KO-MФ mice was calculated as A450 – A540^{fortilinKO-MФ} / A450 – A540^{fortilinWT-MФ}.

MTS proliferation assay

THP1^WT-fortilin and THP1^KO-fortilin cells were seeded (1 × 10⁴ per well, 8 wells each) in 96-well plates in 100 μL of medium and incubated for 24, 48, 72, and 96 h at 37 °C. The mass of living cells was measured using the MTS method according to the manufacturer’s instructions (CellTiter96^®_Aqueous, Cat #G3582, Promega, Madison, WI, USA). Briefly, 100 μL of MTS reagent were added directly to the culture medium at the four time points. The cells were incubated at 37°C for 2 h in a humidified, 5% CO₂ atmosphere before absorbance at 490 nm (A490) was recorded using a 96-well plate reader. The MTS growth index was calculated as (A490^{THP1-WT-fortilin} – A490^Media at a given time point) / (A490^{THP1-WT-fortilin} – A490^Media at time zero) or (A490^{THP1-KO-fortilin} – A490^Media at a given time point) / (A490^{THP1-KO-fortilin} – A490^Media at time zero).

FC formation assays

In vivo FC formation assay using ORO staining

Fifteen-week old male fortilin^WT-MФ-HC (N = 4) and fortilin^KO-MФ-HC (N = 4) mice were placed on a HFD (Clinton-Cybulsky High Fat Rodent Diet with Regular Casein and 1.25% added Cholesterol (D12108Ci: Irradiate-3), Research Diets, Inc., New Brunswick, NJ, USA) for 4 weeks before they were intraperitoneally injected with 1 mL of 3% thioglycollate (Fluka). Four days later, peritoneal MФ were harvested from the peritoneal cavity and allowed to attach to tissue culture plates. For ORO staining, 1 ×10⁵ MФ from each animal were seeded in duplicate on a round cover slip within a well of a 24-well plate and incubated in DMEM with 20% FBS overnight at 37°C in a humidified incubation chamber with 5% CO₂. The next morning, cells were fixed using Perfusion Fixative (Electron Microscopy Sciences, Hatfield, PA, USA; Catalog #: 1224SK, containing PFA 4% and sucrose 4% in PBS 0.1 M, pH 7.2) for 15 min at room temperature. After cells were washed, 200 µL of propylene glycol were added to each well for 5 min before cells were stained with 200 µL of 0.5% ORO in propylene glycol solution for 10 min at 60 °C. Cells were then rinsed with 500 µL of 85% propylene glycol solution and washed twice with 500 µL of ddH₂O. The coverslip with stained cells was then mounted on Superfrost-plus microscope slides (Thermo Fisher, Catalog #: 12-550-15) using PermaFlour™ Aqueous Mounting Medium (Thermo Fisher, Catalog #: TA-030-FM). The FC index (arbitrary unit, A.U.) was calculated as (the number of ORO-positive MФ) / (the number of total MФ) × 100. At least 126 cells were counted in a field. Eight randomly selected fields were counted in one slide/well, and two slides were counted per animal.

FC formation assay using flow cytometry

We performed a flow cytometry-based FC formation assay as described previously¹³¹ with modifications. Briefly, THP1^WT-fortilin and THP1^KO-fortilin cells were incubated for 8 h with either (i) 10 µg/mL of 3,3’- dioctadecyloxacarbocyanine (DiO)-labeled medium oxLDL (Kalen Biomedical, Catalog #: 770282-9) in RPMI-1640 medium with 10% (v/v) lipoprotein deficient serum (LPDS) or (ii) RPMI-1640 medium with 10% (v/v) LPDS at 37°C. Cells were then washed twice with PBS and resuspended in Flow Cytometry Staining Buffer (eBioscience, San Diego, CA, USA; Catalog #: 00-4222-26). DiO-oxLDL uptake was analyzed by flow cytometry on a Canto II Flow Cytometer (Becton Dickinson, Franklin Lake, NJ, USA) using a blue laser (488 nm) and the 530/30 bandpass. The data were analyzed using FlowJo Software (Becton Dickinson). The FC index was calculated as the mean of median fluorescence intensity of DiO-oxLDL-treated cells divided by that of medium-treated cells, and values were expressed as arbitrary units (A.U.). The experiment was performed in three independent sets, and the result was presented as mean with standard deviation (s.d.).

Migration assays

We performed in vitro migration assays as previously described¹³² with modifications. Briefly, peritoneal MФ from fortilin^WT-fortilin (N = 3) and fortilin^KO-fortilin (N = 3) mice were induced by intraperitoneal injection of 3% thioglycollate. Two thousand each of peritoneal MФ suspended in 25 µL of DMEM without FBS were seeded on a pore area in the middle plate of the ChemoTx® Cell Migration System (Neuro Probe, Inc., Gaithersburg, MD, USA; Catalog #: 101-5), which consists of an upper lid, middle plate with a polycarbonate track-etch membrane with pore size of 5 µm, and a lower chamber. Next, 30 µL of DMEM with 1% mouse serum from 8-month-old HC male Ldlr^-/-Apobec1^-/- mice were aliquoted into the lower chambers of the system. Two identical plates were generated. The cells were incubated at 37°C for 14 h. After incubation, all cells on the lower surface of the membrane of the first plate were scraped off, leaving the cells that attached to the membrane but did not migrate through the membrane. All cells on the upper surface of the membrane of the second plate were scraped off, leaving the cells that had migrated through the membrane. Remaining cells were fixed with 4% PFA, permeabilized with 70% methanol, and stained with hematoxylin solution. The number of cells on the upper surface of the membrane (Cells^top) and Cells^bottom were counted from three randomly selected fields under a light microscope. The migration index (%) was calculated as Cells^bottom / (Cells^top + Cells^bottom) × 100 for MФ induced from each mouse. On average, 109 cells were counted per field.

Phagocytosis assays

Phagocytosis assays were performed as previously described¹³³ using BMDMФ from fortilin^WT-MФ (BMDMФ^WT-fortilin) mice (N = 6) and fortilin^KO-MФ (BMDMФ^KO-fortilin) mice (N = 5). Zymosan A Bioparticles conjugated to Alexa Fluor 594 and opsonized by purified rabbit polyclonal IgG antibodies (Thermo Fisher-Molecular Probes, Catalog #s: Z23374 and Z2850, respectively) were added to the culture medium (RPMI-1640 with 10% FBS) of BMDMФ in the wells of a 24-well plate at a multiplicity of infection of 10. The plate was centrifuged at 450 g for 2 min to allow the bioparticles to attach to BMDMФ. After the treatment, BMDMФ were incubated in a humidified 37°C, 5% CO₂ incubator for 1 h before phagocytosis was stopped by ice-cold PBS. The cells were then fixed directly in the wells with 4% PFA, washed twice with PBS, permeabilized with permeabilization solution (0.5% Triton X-100 and 1% FBS in PBS), stained with fluorescein isothiocyanate-labeled phalloidin (Sigma-Aldrich, St. Louis, MO, USA; Catalog #: P5282) and then with 1 µg/mL of DAPI, washed again, and briefly air dried before phagocytosed particles were visualized under a fluorescence microscope. At least 650 cells were counted per field, and three randomly selected fields were examined per sample. A phagocytosis index (A.U.) was calculated as the number of red particles per 100 BMDMФ.

MΦ transdifferentiation assay

THP1 human monocyte-MΦ cells

To test the possibility that fortilin deficiency in MΦ causes MΦ to transdifferentiate into VSMCs, we seeded 0.5 ×106 THP1^WT-fortilin and THP1^KO-fortilin cells in triplicate in 6-well dishes, differentiated them using 5 ng/mL of PMA, incubated them for 8 h in the absence or presence of 50 µg/mL of oxLDL, harvested the cells, isolated RNAs, and subjected them to RT-qPCR for the markers of VSMCs and MΦ (Fig. 5b–f), using the protocol described above. We also subjected 30 µg of proteins isolated from the same cells to western blot analyses to validate the findings of RT-qPCR using the protocol described above (Fig. 5g).

Primary human MΦ (phMΦ)

To further validate the findings from THP1 cells, we first acquired phMΦ (human M1 MΦ [hMDM-GMCSF(-)], Catalog #: C12916, Sigma Aldrich) and cultured them according to the manufacturer’s instructions. To knockdown fortilin in phMΦ, we transduced the cells with lentivirus short hairpin RNA targeting fortilin (shRNA^fortilin) particles (MISSION®, Catalog #: TRCN0000147481, 0000180005, 0000183718, 0000413426, and 0000435136, Sigma Aldrich) or with control lentivirus particles (shRNA^control)(MISSION®, Catalog #: SHC016V) following manufacturer’s instructions at a multiplicity of infection of 3. We then selected transduced cells by incubating them in medium containing 100 μg/mL of puromycin for 3 days. Next, we seeded 1.0 × 103 phMΦ transduced with either shRNA^fortilin (hereafter, phMΦ^sh-fortilin) or shRNA^control (phMΦ^sh-control) in triplicate in 6-well dishes, incubated them for 8 h in the absence or presence of 50 µg/mL of oxLDL, harvested them, isolated total RNAs, and subjected it to RT-qPCR for the markers of VSMCs and MΦ using the protocol described above.

Generation of CM from THP1^WT-fortilin and THP1^KO-fortilin cells

We generated CM from THP1^WT-fortilin and THP1^KO-fortilin cells by seeding 1 × 10⁶ cells in each well of a 6-well plate, differentiating them in the medium containing 5 ng/mL of PMA, washing them with PBS, incubating them in DMEM supplemented with 50 µg/mL of oxLDL for 8 h. After transferring the media to microfuge tubes, we centrifuged the samples at 1300 x g for 5 min. We transferred the media to sterile 15 mL tubes and used them immediately for the experiments without freezing and thawing.

Immunodepletion of TGF-β1

We first covalently coupled mouse α-TGF-β mAb (R&D Systems, Minneapolis, MN, USA; Clone 1D11, Catalog #: MAB1835-100) to tosylactivated magnetic beads (Dynabeads® M-280; Thermo Fisher, Catalog #: 14203) following manufacturer’s instructions. We then immunodepleted TGF-β1 from the CM from THP1^WT-fortilin and THP1^KO-fortilin as previously described¹¹⁹. Briefly, we mixed 1 mL of CM with 1 mL of α-TGF-β-mAb-coated magnetic beads (20 mg/mL), incubated the mixture for 24 h at 4 °C, immobilized the beads with a magnet, and transferred the supernatants to fresh microcentrifuge tubes. We then filter-sterilized the immunodepleted CM using a 0.2 µm sterile syringe filter (Fisherbrand™Syringe Filters, Catalog #: 09-719 C) before adding them to UCDMSCs. For the negative control, we used RPMI-1640 medium supplemented with 10% LPDS and 0.1% penicillin/streptomycin/amphotericin B solution (10 units/mL, 10 µg/mL, and 25 ng/mL, respectively). For the positive control, we used the same medium with human recombinant TGF-β1 at 5 ng/mL (R&D Systems; Catalog #: 240-B-010/CF).

Mesenchymal stem cell differentiation assays

For the time course analysis of the differentiation of UCDMSCs, we seeded 2 × 10⁵ cells into each well of a 12-well plate in triplicate in the complete MSC culture medium described in the Human umbilical cord-derived mesenchymal stem cells (UCDMSCs) section. The next morning, we washed the cells with PBS and started to incubate them in (a) medium only (RPMI media supplemented with 10% LPDS and 50 µg/mL oxLDL), (b) medium + 5 ng/mL of TGF- β1, (c) CM from THP1^WT-fortilin, or (d) CM from THP1^KO-fortilin for up to 120 h. We harvested the cells at 0, 72, and 120 h, extracted the RNAs, and subjected them to RT-qPCR using an appropriate primer and probe set for a given gene of interest.

To assess the impact of TGF-β1 on the differentiation of UCDMSCs, we seeded 1 × 10⁵ cells into each well of 6-well dishes in triplicate and incubated them in (a) medium only without TGF-β1 immunodepletion (ID), (b) medium only with TGF-β1 ID, (c) medium + 5 ng/mL of TGF-β1, (d) medium + 5 ng/mL of TGF-β1 with TGF-β1 ID, (e) CM from THP1^WT-fortilin, (f) CM from THP1^WT-fortilin with TGF-β1 ID, (g) CM from THP1^KO-fortilin, or (h) CM from THP1^KO-fortilin with TGF-β1 ID for 120 h. We then harvested the cells, extracted the RNAs, and subjected them to RT-qPCR using an appropriate primer and probe set for a given gene of interest.

Efferocytosis assays

Efferocytosis is the phagocytotic removal of apoptotic cells by MΦ. We evaluated the role of fortilin in efferocytosis using the protocol described by Salina et al.⁶¹. Briefly, we first labeled 1 × 10⁷ of THP1^WT-fortilin and THP1^KO-fortilin cells with CellTrace™ carboxyfluorescein succinimidyl ester (CFSE) (Invitrogen™-ThermoFisher, excitation/emission = 492/517, laser used = 488 nm [blue], bandpass filter = 530/30 nm [FITC]), following the manufacturer’s instructions. After washing, we seeded 1×10⁶ cells per well in 6 well-plates and allowed them to fully differentiate into MΦ in the presence of 50 ng/mL of PMA for 24 h at 37°C and 5% CO₂. Next, to generate labeled apoptotic cells (ACs), we induced apoptosis in Jurkat cells by subjecting them to UV irradiation (50 mJ/cm² for 30 min), labeled them with CellTrace™ Far Red (Invitrogen™-ThermoFisher, excitation/emission = 648/658, laser used = 640 nm [red]; bandpass filter = 660/20 nm [APC]) following the manufacturer’s instructions, and incubated them for 4 h at 37 °C and 5% CO₂ to allow apoptosis to progress. We then collected, washed, and resuspended the ACs in fresh RPMI medium at a final concentration of 5 × 10⁶ cells/mL. We added ACs to THP1^WT-fortilin and THP1^KO-fortilin cells at a 1:1 THP1-to-AC ratio, incubated the cells for 18 h, washed off the unengulfed ACs, added a second round of ACs to the THP1 cells and incubated them for 2 h. After washing, we harvested the THP1 cells with a cell scraper, stained them with Zombie NIR (BioLegend, San Diego, CA, USA; excitation/emission = 715/743, laser used = 640 nm [red], 780/40 nm [APC/Cy7]) for 10 min in the dark, and fixed them in 2% PFA before washing and resuspending the cells in PBS for flow cytometry analyses. In flow cytometry, we first excluded debris based on morphology profiles (FSC-A vs. SSC-A), non-singlet cells using FSC-A vs. FSC-H, and nonviable cells (ZombieNIR⁺) using SSC-A vs. APC/Cy7-A. We then gated on CFSE-labeled THP1 cells (CFSE⁺, FITC) and visualized them on a FITC-A vs. APC-A plot where CFSE⁺FarRed⁺ cells represented THP1 cells that had engulfed ACs. We calculated the efferocytosis index as the percentage of CFSE⁺FarRed⁺ cells out of total CFSE⁺ cells, expressed as a percentage. We performed the experiment four independent times.

Ethics statement

We have complied with all relevant ethical regulations for animal use. This study was carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All experimental protocols involving animals were registered with and approved by the Institutional Animal Care and Use Committees (IACUCs) of the University of Texas Medical Branch and the University of Washington prior to study initiation.

Statistical analysis and reproducibility

The degree of the spread of data was expressed by the standard deviation (±s.d.). We used the two-tailed unpaired t test (Student’s t test) to compare the means of two groups when data of both groups were normally distributed, while we used the Mann–Whitney test when the data were not normally distributed. To compare the means of three or more groups, we used one-way analysis of variance (ANOVA) with Fisher’s or Tukey-Kramer’s pairwise comparison (Minitab® or GraphPad Prism [Sandiego, CA, USA]). For survival analysis of fortilin^KO-MФ and fortilin^WT-MФ mice, we constructed a Kaplan-Meier curve using GraphPad Prism. The log-rank Mantel-Cox test was then used to determine whether a statistical difference existed between the two survival curves as previously described¹³⁴. P < 0.05 was considered to be statistically significant. P < 0.10 was considered to show a trend toward statistical significance. Both Prism Software 10.1.2 and Minitab® 17 or 21 were used for statistical analyses. The numbers of biological replicates used in each experiment, along with the statistical test used to analyze the particular experiment, are provided in the manuscript. No data were excluded unless outliers were identified and verified by the outlier tests (Minitab® 17/21 or GraphPad Prism). Although the researchers were not blinded to allocation during experiments and readout evaluation, all readouts from the experiments were predetermined, highly objective, and obtained according to validated protocols.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The data supporting the findings of this study are available within the paper and its Supplementary Information files. The numerical source data can be found in Supplementary Data. All relevant data are available from the authors upon reasonable request.

References

Collaborators GBDCoD. Global, regional, and national age-sex-specific mortality for 282 causes of death in 195 countries and territories, 1980-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet 392, 1736–1788 (2018).
Article Google Scholar
Rader, D. J. & Pure, E. Lipoproteins, macrophage function, and atherosclerosis: beyond the foam cell?. Cell Metab. 1, 223–230 (2005).
Article CAS PubMed Google Scholar
Tontonoz, P., Nagy, L., Alvarez, J. G., Thomazy, V. A. & Evans, R. M. PPARgamma promotes monocyte/macrophage differentiation and uptake of oxidized LDL. Cell 93, 241–252 (1998).
Article CAS PubMed Google Scholar
Chinetti-Gbaguidi, G. et al. Human atherosclerotic plaque alternative macrophages display low cholesterol handling but high phagocytosis because of distinct activities of the PPARgamma and LXRalpha pathways. Circ. Res 108, 985–995 (2011).
Article CAS PubMed PubMed Central Google Scholar
Robbins, C. S. et al. Local proliferation dominates lesional macrophage accumulation in atherosclerosis. Nat. Med 19, 1166–1172 (2013).
Article CAS PubMed PubMed Central Google Scholar
Parks, B. W. & Lusis, A. J. Macrophage accumulation in atherosclerosis. N. Engl. J. Med 369, 2352–2353 (2013).
Article CAS PubMed PubMed Central Google Scholar
Frostegard, J. et al. Cytokine expression in advanced human atherosclerotic plaques: dominance of pro-inflammatory (Th1) and macrophage-stimulating cytokines. Atherosclerosis 145, 33–43 (1999).
Article CAS PubMed Google Scholar
Libby, P. & Hansson, G. K. From focal lipid storage to systemic inflammation: JACC review topic of the week. J. Am. Coll. Cardiol. 74, 1594–1607 (2019).
Article CAS PubMed PubMed Central Google Scholar
Li, F., Zhang, D. & Fujise, K. Characterization of fortilin, a novel antiapoptotic protein. J. Biol. Chem. 276, 47542–47549 (2001).
Article CAS PubMed Google Scholar
Sinthujaroen, P. et al. Elevation of serum fortilin levels is specific for apoptosis and signifies cell death in vivo. BBA Clin. 2, 103–111 (2014).
Article PubMed PubMed Central Google Scholar
Gross, B., Gaestel, M., Bohm, H. & Bielka, H. cDNA sequence coding for a translationally controlled human tumor protein. Nucleic Acids Res. 17, 8367 (1989).
Article CAS PubMed PubMed Central Google Scholar
Tuynder, M. et al. Biological models and genes of tumor reversion: cellular reprogramming through tpt1/TCTP and SIAH-1. Proc. Natl. Acad. Sci. USA 99, 14976–14981 (2002).
Article CAS PubMed PubMed Central Google Scholar
Yang, Y. et al. An N-terminal region of translationally controlled tumor protein is required for its antiapoptotic activity. Oncogene 24, 4778–4788 (2005).
Article CAS PubMed PubMed Central Google Scholar
Graidist, P., Phongdara, A. & Fujise, K. Antiapoptotic protein partners fortilin and MCL1 independently protect cells from 5-fluorouracil-induced cytotoxicity. J. Biol. Chem. 279, 40868–40875 (2004).
Article CAS PubMed Google Scholar
Zhang, D., Li, F., Weidner, D., Mnjoyan, Z. H. & Fujise, K. Physical and functional interaction between myeloid cell leukemia 1 protein (MCL1) and Fortilin. potential role MCL1 a fortilin chaperone. J. Biol. Chem. 277, 37430–37438 (2002).
CAS PubMed Google Scholar
Chen, Y. et al. Physical and functional antagonism between tumor suppressor protein p53 and fortilin, an anti-apoptotic protein. J. Biol. Chem. 286, 32575–32585 (2011).
Article CAS PubMed PubMed Central Google Scholar
Amson, R. et al. Reciprocal repression between P53 and TCTP. Nat. Med 18, 91–99 (2011).
Article PubMed Google Scholar
Yarm, F. R. Plk phosphorylation regulates the microtubule-stabilizing protein TCTP. Mol. Cell Biol. 22, 6209–6221 (2002).
Article CAS PubMed PubMed Central Google Scholar
Kashiwakura, J. C. et al. Histamine-releasing factor has a proinflammatory role in mouse models of asthma and allergy. J. Clin. Invest 122, 218–228 (2012).
Article CAS PubMed Google Scholar
MacDonald, S. M., Rafnar, T., Langdon, J. & Lichtenstein, L. M. Molecular identification of an IgE-dependent histamine-releasing factor. Science 269, 688–690 (1995).
Article CAS PubMed Google Scholar
Chattopadhyay, A. et al. Fortilin potentiates the peroxidase activity of Peroxiredoxin-1 and protects against alcohol-induced liver damage in mice. Sci. Rep. 6, 18701 (2016).
Article CAS PubMed PubMed Central Google Scholar
Pinkaew, D. et al. Fortilin binds IRE1α and prevents ER stress from signaling apoptotic cell death. Nat. Commun. 8, 1–16 (2017).
Article CAS Google Scholar
Liu, H., Peng, H. W., Cheng, Y. S., Yuan, H. S. & Yang-Yen, H. F. Stabilization and enhancement of the antiapoptotic activity of mcl-1 by TCTP. Mol. Cell Biol. 25, 3117–3126 (2005).
Article CAS PubMed PubMed Central Google Scholar
Nakashima, M. et al. Fortilin binds CTNNA3 and protects it against phosphorylation, ubiquitination, and proteasomal degradation to guard cells against apoptosis. Commun. Biol. 8, 1 (2025).
Article CAS PubMed PubMed Central Google Scholar
Koide, Y. et al. Embryonic Lethality of Fortilin-null mutant mice by BMP-pathway overactivation. Biochim Biophys. Acta 1790, 326–338 (2009).
Article CAS PubMed PubMed Central Google Scholar
Chen, S. H. et al. A knockout mouse approach reveals that TCTP functions as an essential factor for cell proliferation and survival in a tissue- or cell type-specific manner. Mol. Biol. Cell 18, 2525–2532 (2007).
Article CAS PubMed PubMed Central Google Scholar
Powell-Braxton, L. et al. A mouse model of human familial hypercholesterolemia: markedly elevated low density lipoprotein cholesterol levels and severe atherosclerosis on a low-fat chow diet. Nat. Med. 4, 934–938 (1998).
Article CAS PubMed Google Scholar
Pinkaew, D. et al. Fortilin reduces apoptosis in macrophages and promotes atherosclerosis. Am. J. Physiol. Heart Circ. Physiol. 305, H1519–H1529 (2013).
Article CAS PubMed PubMed Central Google Scholar
Reynolds, J. E. et al. Mcl-1, a member of the Bcl-2 family, delays apoptosis induced by c-Myc overexpression in Chinese hamster ovary cells. Cancer Res. 54, 6348–6352 (1994).
CAS PubMed Google Scholar
Cogliati, S. et al. Mitochondrial cristae shape determines respiratory chain supercomplexes assembly and respiratory efficiency. Cell 155, 160–171 (2013).
Article CAS PubMed PubMed Central Google Scholar
Dutta, R., Singh, U., Li, T. B., Fornage, M. & Teng, B. B. Hepatic gene expression profiling reveals perturbed calcium signaling in a mouse model lacking both LDL receptor and Apobec1 genes. Atherosclerosis 169, 51–62 (2003).
Article CAS PubMed Google Scholar
Canault, M. et al. Progression of atherosclerosis in ApoE-deficient mice that express distinct molecular forms of TNF-alpha. J. Pathol. 214, 574–583 (2008).
Article CAS PubMed Google Scholar
Whitman, S. C., Ravisankar, P., Elam, H. & Daugherty, A. Exogenous interferon-gamma enhances atherosclerosis in apolipoprotein E-/- mice. Am. J. Pathol. 157, 1819–1824 (2000).
Article CAS PubMed PubMed Central Google Scholar
Buono, C. et al. Influence of interferon-gamma on the extent and phenotype of diet-induced atherosclerosis in the LDLR-deficient mouse. Arterioscler Thromb. Vasc. Biol. 23, 454–460 (2003).
Article CAS PubMed Google Scholar
Subramanian, M., Thorp, E. & Tabas, I. Identification of a non-growth factor role for GM-CSF in advanced atherosclerosis: promotion of macrophage apoptosis and plaque necrosis through IL-23 signaling. Circ. Res. 116, e13–e24 (2015).
Article CAS PubMed Google Scholar
Grainger, D. J., Mosedale, D. E., Metcalfe, J. C. & Bottinger, E. P. Dietary fat and reduced levels of TGFbeta1 act synergistically to promote activation of the vascular endothelium and formation of lipid lesions. J. Cell Sci. 113, 2355–2361 (2000).
Article CAS PubMed Google Scholar
Mallat, Z. et al. Inhibition of transforming growth factor-beta signaling accelerates atherosclerosis and induces an unstable plaque phenotype in mice. Circ. Res 89, 930–934 (2001).
Article CAS PubMed Google Scholar
Hartung, T. Anti-inflammatory effects of granulocyte colony-stimulating factor. Curr. Opin. Hematol. 5, 221–225 (1998).
Article CAS PubMed Google Scholar
Sinha, S. K., Mishra, V., Nagwani, S. & Rajavashisth, T. B. Effects of G-CSF on serum cholesterol and development of atherosclerotic plaque in apolipoprotein E-deficient mice. Int. J. Clin. Exp. Med. 7, 1979–1989 (2014).
PubMed PubMed Central Google Scholar
Budi, E. H., Schaub, J. R., Decaris, M., Turner, S. & Derynck, R. TGF-beta as a driver of fibrosis: physiological roles and therapeutic opportunities. J. Pathol. 254, 358–373 (2021).
Article CAS PubMed Google Scholar
Haley, K. J. et al. Overexpression of eotaxin and the CCR3 receptor in human atherosclerosis: using genomic technology to identify a potential novel pathway of vascular inflammation. Circulation 102, 2185–2189 (2000).
Article CAS PubMed Google Scholar
Freigang, S. et al. Fatty acid-induced mitochondrial uncoupling elicits inflammasome-independent IL-1alpha and sterile vascular inflammation in atherosclerosis. Nat. Immunol. 14, 1045–1053 (2013).
Article CAS PubMed Google Scholar
Elhage, R. et al. Involvement of interleukin-6 in atherosclerosis but not in the prevention of fatty streak formation by 17beta-estradiol in apolipoprotein E-deficient mice. Atherosclerosis 156, 315–320 (2001).
Article CAS PubMed Google Scholar
Veillard, N. R. et al. Antagonism of RANTES receptors reduces atherosclerotic plaque formation in mice. Circ. Res 94, 253–261 (2004).
Article CAS PubMed Google Scholar
Davenport, P. & Tipping, P. G. The role of interleukin-4 and interleukin-12 in the progression of atherosclerosis in apolipoprotein E-deficient mice. Am. J. Pathol. 163, 1117–1125 (2003).
Article CAS PubMed PubMed Central Google Scholar
Cardilo-Reis, L. et al. Interleukin-13 protects from atherosclerosis and modulates plaque composition by skewing the macrophage phenotype. EMBO Mol. Med. 4, 1072–1086 (2012).
Article CAS PubMed PubMed Central Google Scholar
Caligiuri, G. et al. Interleukin-10 deficiency increases atherosclerosis, thrombosis, and low-density lipoproteins in apolipoprotein E knockout mice. Mol. Med 9, 10–17 (2003).
Article CAS PubMed PubMed Central Google Scholar
Damond, N. et al. A map of human type 1 diabetes progression by imaging mass cytometry. Cell Metab. 29, 755–68 e5 (2019).
Article CAS PubMed PubMed Central Google Scholar
Giesen, C. et al. Highly multiplexed imaging of tumor tissues with subcellular resolution by mass cytometry. Nat. methods 11, 417–422 (2014).
Article CAS PubMed Google Scholar
Windhager, J. et al. An end-to-end workflow for multiplexed image processing and analysis. Nat. Protoc. 18, 3565–3613 (2023).
Article CAS PubMed Google Scholar
Becht, E. et al. Dimensionality reduction for visualizing single-cell data using UMAP. Nat. Biotechnol. 37, 38–44 (2018).
Article Google Scholar
Pan, H. et al. Single-cell genomics reveals a novel cell state during smooth muscle cell phenotypic switching and potential therapeutic targets for atherosclerosis in mouse and human. Circulation 142, 2060–2075 (2020).
Article CAS PubMed PubMed Central Google Scholar
Liu, L. & Shi, G. P. CD31: beyond a marker for endothelial cells. Cardiovasc Res 94, 3–5 (2012).
Article CAS PubMed Google Scholar
Ho, M. K. & Springer, T. A. Mac-2, a novel 32,000 Mr mouse macrophage subpopulation-specific antigen defined by monoclonal antibodies. J. Immunol. 128, 1221–1228 (1982).
Article CAS PubMed Google Scholar
Owens, G. K. Regulation of differentiation of vascular smooth muscle cells. Physiol. Rev. 75, 487–517 (1995).
Article CAS PubMed Google Scholar
Kimball, A. K. et al. A Beginner’s Guide to Analyzing and Visualizing Mass Cytometry Data. J. Immunol. 200, 3–22 (2018).
Article CAS PubMed Google Scholar
Matos, T. R., Liu, H. & Ritz, J. Research techniques made simple: Mass cytometry analysis tools for decrypting the complexity of biological systems. J. Invest Dermatol 137, e43–e51 (2017).
Article CAS PubMed Google Scholar
Ran, F. A. et al. Genome engineering using the CRISPR-Cas9 system. Nat. Protoc. 8, 2281–2308 (2013).
Article CAS PubMed PubMed Central Google Scholar
Cory, A. H., Owen, T. C., Barltrop, J. A. & Cory, J. G. Use of an aqueous soluble tetrazolium/formazan assay for cell growth assays in culture. Cancer Commun. 3, 207–212 (1991).
Article CAS PubMed Google Scholar
Endemann, G. et al. CD36 is a receptor for oxidized low density lipoprotein. J. Biol. Chem. 268, 11811–11816 (1993).
Article CAS PubMed Google Scholar
Salina, A. C. G., Fortes-Rocha, M. & Cunha, L. D. In vitro assessment of efferocytic capacity of human macrophages using flow cytometry. Bio Protoc. 13, e4903 (2023).
CAS PubMed PubMed Central Google Scholar
De Meyer, G. R. Y., Zurek, M., Puylaert, P. & Martinet, W. Programmed death of macrophages in atherosclerosis: mechanisms and therapeutic targets. Nat. Rev. Cardiol. 21, 312–325 (2024).
Article PubMed Google Scholar
Kojima, Y., Weissman, I. L. & Leeper, N. J. The role of efferocytosis in atherosclerosis. Circulation 135, 476–489 (2017).
Article CAS PubMed PubMed Central Google Scholar
Bennett, M. R., Sinha, S. & Owens, G. K. Vascular smooth muscle cells in atherosclerosis. Circ. Res. 118, 692–702 (2016).
Article CAS PubMed PubMed Central Google Scholar
Meng, X. M. et al. Inflammatory macrophages can transdifferentiate into myofibroblasts during renal fibrosis. Cell Death Dis. 7, e2495 (2016).
Article CAS PubMed PubMed Central Google Scholar
Alexander, M. R. & Owens, G. K. Epigenetic control of smooth muscle cell differentiation and phenotypic switching in vascular development and disease. Annu. Rev. Physiol. 74, 13–40 (2012).
Article CAS PubMed Google Scholar
Papaspyridonos, M. et al. Galectin-3 is an amplifier of inflammation in atherosclerotic plaque progression through macrophage activation and monocyte chemoattraction. Arterioscler Thromb. Vasc. Biol. 28, 433–440 (2008).
Article CAS PubMed Google Scholar
Holness, C. L. & Simmons, D. L. Molecular cloning of CD68, a human macrophage marker related to lysosomal glycoproteins. Blood 81, 1607–1613 (1993).
Article CAS PubMed Google Scholar
Feil, S. et al. Transdifferentiation of vascular smooth muscle cells to macrophage-like cells during atherogenesis. Circ. Res. 115, 662–667 (2014).
Article CAS PubMed Google Scholar
Amere Subbarao, S. Cancer vs. SARS-CoV-2 induced inflammation, overlapping functions, and pharmacological targeting. Inflammopharmacology 29, 343–366 (2021).
Article CAS PubMed PubMed Central Google Scholar
Binder, C. J. et al. IL-5 links adaptive and natural immunity specific for epitopes of oxidized LDL and protects from atherosclerosis. J. Clin. Invest 114, 427–437 (2004).
Article CAS PubMed PubMed Central Google Scholar
Tyrrell, D. J. & Goldstein, D. R. Ageing and atherosclerosis: vascular intrinsic and extrinsic factors and potential role of IL-6. Nat. Rev. Cardiol. 18, 58–68 (2021).
Article CAS PubMed Google Scholar
Dong, K. et al. Mesenchyme homeobox 1 mediates transforming growth factor-beta (TGF-beta)-induced smooth muscle cell differentiation from mouse mesenchymal progenitors. J. Biol. Chem. 293, 8712–8719 (2018).
Article CAS PubMed PubMed Central Google Scholar
Kim, Y. M. et al. Proteomic identification of ADAM12 as a regulator for TGF-beta1-induced differentiation of human mesenchymal stem cells to smooth muscle cells. PLoS One 7, e40820 (2012).
Article CAS PubMed PubMed Central Google Scholar
Kinner, B., Zaleskas, J. M. & Spector, M. Regulation of smooth muscle actin expression and contraction in adult human mesenchymal stem cells. Exp. Cell Res. 278, 72–83 (2002).
Article CAS PubMed Google Scholar
Wang, D. et al. Proteomic profiling of bone marrow mesenchymal stem cells upon transforming growth factor beta1 stimulation. J. Biol. Chem. 279, 43725–43734 (2004).
Article CAS PubMed Google Scholar
Dominici, M. et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 8, 315–317 (2006).
Article CAS PubMed Google Scholar
Gorelik, L. & Flavell, R. A. Transforming growth factor-beta in T-cell biology. Nat. Rev. Immunol. 2, 46–53 (2002).
Article CAS PubMed Google Scholar
Poirier, P. et al. Obesity and cardiovascular disease: pathophysiology, evaluation, and effect of weight loss: an update of the 1997 American Heart Association Scientific Statement on Obesity and Heart Disease from the Obesity Committee of the Council on Nutrition, Physical Activity, and Metabolism. Circulation 113, 898–918 (2006).
Article PubMed Google Scholar
Björkegren, J. L. M. & Lusis, A. J. Atherosclerosis: recent developments. Cell 185, 1630–1645 (2022).
Article PubMed PubMed Central Google Scholar
Zhuang, T. et al. ALKBH5-mediated m6A modification of IL-11 drives macrophage-to-myofibroblast transition and pathological cardiac fibrosis in mice. Nat. Commun. 15, 1995 (2024).
Article CAS PubMed PubMed Central Google Scholar
Libby, P., Ridker, P. M. & Hansson, G. K. Progress and challenges in translating the biology of atherosclerosis. Nature 473, 317–325 (2011).
Article CAS PubMed Google Scholar
Marx, S. O., Totary-Jain, H. & Marks, A. R. Vascular smooth muscle cell proliferation in restenosis. Circul. Cardiovascular Interv. 4, 104–111 (2011).
Article CAS Google Scholar
Deglise, S., Bechelli, C. & Allagnat, F. Vascular smooth muscle cells in intimal hyperplasia, an update. Front. Physiol. 13, 1081881 (2022).
Article PubMed Google Scholar
Alsaigh, T., Evans, D., Frankel, D. & Torkamani, A. Decoding the transcriptome of calcified atherosclerotic plaque at single-cell resolution. Commun. Biol. 5, 1084 (2022).
Article CAS PubMed PubMed Central Google Scholar
de Winther, M. P. J. et al. Translational opportunities of single-cell biology in atherosclerosis. Eur. Heart J. 44, 1216–1230 (2023).
Article PubMed Google Scholar
Williams, J. W. et al. Single cell RNA sequencing in atherosclerosis research. Circ. Res. 126, 1112–1126 (2020).
Article CAS PubMed PubMed Central Google Scholar
Grainger, D. J., Reckless, J. & McKilligin, E. Apolipoprotein E modulates clearance of apoptotic bodies in vitro and in vivo, resulting in a systemic proinflammatory state in apolipoprotein E-deficient mice. J. Immunol. 173, 6366–6375 (2004).
Article CAS PubMed Google Scholar
Gistera, A., Ketelhuth, D. F. J., Malin, S. G. & Hansson, G. K. Animal models of atherosclerosis-supportive notes and tricks of the trade. Circ. Res. 130, 1869–1887 (2022).
Article CAS PubMed Google Scholar
Christ, A. et al. Western diet triggers NLRP3-dependent innate immune reprogramming. Cell 172, 162–75 e14 (2018).
Article CAS PubMed PubMed Central Google Scholar
Grosse, L. et al. Defined p16(high) senescent cell types are indispensable for mouse Healthspan. Cell Metab. 32, 87–99 e6 (2020).
Article CAS PubMed Google Scholar
Childs, B. G. et al. Senescent intimal foam cells are deleterious at all stages of atherosclerosis. Science 354, 472–477 (2016).
Article CAS PubMed PubMed Central Google Scholar
Wang, J. et al. Vascular smooth muscle cell senescence promotes atherosclerosis and features of plaque vulnerability. Circulation 132, 1909–1919 (2015).
Article CAS PubMed Google Scholar
Bai, B. et al. Cyclin-dependent kinase 5-mediated hyperphosphorylation of sirtuin-1 contributes to the development of endothelial senescence and atherosclerosis. Circulation 126, 729–740 (2012).
Article CAS PubMed Google Scholar
Covarrubias, A. J. et al. Senescent cells promote tissue NAD(+) decline during ageing via the activation of CD38(+) macrophages. Nat. Metab. 2, 1265–1283 (2020).
Article CAS PubMed PubMed Central Google Scholar
Amici, S. A. et al. CD38 is robustly induced in human macrophages and monocytes in inflammatory conditions. Front. Immunol. 9, 1593 (2018).
Article PubMed PubMed Central Google Scholar
Jablonski, K. A. et al. Novel markers to delineate murine M1 and M2 macrophages. PLoS One 10, e0145342 (2015).
Article PubMed PubMed Central Google Scholar
Terao, R. et al. LXR/CD38 activation drives cholesterol-induced macrophage senescence and neurodegeneration via NAD(+) depletion. Cell Rep. 43, 114102 (2024).
Article CAS PubMed PubMed Central Google Scholar
Koike, T. et al. IDOL deficiency inhibits cholesterol-rich diet-induced atherosclerosis in rabbits. Arterioscler Thromb. Vasc. Biol. 45, 669–671 (2025).
Article CAS PubMed Google Scholar
Fan, J. et al. Rabbit models for the study of human atherosclerosis: from pathophysiological mechanisms to translational medicine. Pharm. Ther. 146, 104–119 (2015).
Article CAS Google Scholar
Baigent, C. et al. Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 366, 1267–1278 (2005).
Sabatine, M. S. et al. Evolocumab and clinical outcomes in patients with cardiovascular disease. N. Engl. J. Med. 376, 1713–1722 (2017).
Article CAS PubMed Google Scholar
Robinson, J. G. et al. Efficacy and safety of alirocumab in reducing lipids and cardiovascular events. N. Engl. J. Med. 372, 1489–1499 (2015).
Article CAS PubMed Google Scholar
Nissen, S. E. et al. Bempedoic acid and cardiovascular outcomes in statin-intolerant patients. N. Engl. J. Med 388, 1353–1364 (2023).
Article PubMed Google Scholar
Cannon, C. P., Blazing M. A. & Braunwald E. Ezetimibe plus a statin after acute coronary syndromes. N. Engl. J. Med. 373, 1476–1477 (2015).
Sachdeva, A. et al. Lipid levels in patients hospitalized with coronary artery disease: an analysis of 136,905 hospitalizations in Get With The Guidelines. Am. Heart J. 157, 111–7 e2 (2009).
Article CAS PubMed Google Scholar
Ridker, P. M. et al. Antiinflammatory therapy with canakinumab for atherosclerotic disease. N. Engl. J. Med. 377, 1119–1131 (2017).
Article CAS PubMed Google Scholar
Ridker, P. M. et al. Low-dose methotrexate for the prevention of atherosclerotic events. N. Engl. J. Med. 380, 752–762 (2019).
Article CAS PubMed Google Scholar
Reindl, M., Metzler, B. & Reinstadler, S. J. Low-dose colchicine after myocardial infarction. N. Engl. J. Med. 382, 1667 (2020).
PubMed Google Scholar
Hu, G. et al. Nanoparticles targeting macrophages as potential clinical therapeutic agents against cancer and inflammation. Front. Immunol. 10, 1998 (2019).
Article CAS PubMed PubMed Central Google Scholar
Aouadi, M. et al. Orally delivered siRNA targeting macrophage Map4k4 suppresses systemic inflammation. Nature 458, 1180–1184 (2009).
Article CAS PubMed PubMed Central Google Scholar
Merry, T. L. et al. The CSF1 receptor inhibitor pexidartinib (PLX3397) reduces tissue macrophage levels without affecting glucose homeostasis in mice. Int. J. Obes. (Lond.) 44, 245–253 (2020).
Article PubMed Google Scholar
Fujiwara, T. et al. CSF1/CSF1R signaling inhibitor pexidartinib (PLX3397) reprograms tumor-associated macrophages and stimulates T-cell infiltration in the sarcoma microenvironment. Mol. Cancer Ther. 20, 1388–1399 (2021).
Article CAS PubMed PubMed Central Google Scholar
Grabert, K. et al. A transgenic line that reports CSF1R protein expression provides a definitive marker for the mouse mononuclear phagocyte system. J. Immunol. 205, 3154–3166 (2020).
Article CAS PubMed Google Scholar
Clausen, B. E., Burkhardt, C., Reith, W., Renkawitz, R. & Forster, I. Conditional gene targeting in macrophages and granulocytes using LysMcre mice. Transgenic Res 8, 265–277 (1999).
Article CAS PubMed Google Scholar
Norton, J. M. & Munck, A. Glucose transport in murine macrophages: in vitro characterization of the monosaccharide transport system of the thioglycollate-elicited mouse peritoneal macrophage. J. Immunol. 125, 252–258 (1980).
Article CAS PubMed Google Scholar
Graidist, P. et al. Fortilin binds Ca2+ and blocks Ca2+-dependent apoptosis in vivo. Biochem J. 408, 181–191 (2007).
Article CAS PubMed PubMed Central Google Scholar
Fujita, T. et al. Human fortilin is a molecular target of dihydroartemisinin. FEBS Lett. 582, 1055–1060 (2008).
Article CAS PubMed PubMed Central Google Scholar
Pinkaew, D. et al. Fortilin interacts with TGF-beta1 and prevents TGF-beta receptor activation. Commun. Biol. 5, 157 (2022).
Article CAS PubMed PubMed Central Google Scholar
Pinkaew, D., Hutadilok-Towatana, N., Teng, B. B., Mahabusarakam, W. & Fujise, K. Morelloflavone, a biflavonoid inhibitor of migration-related kinases, ameliorates atherosclerosis in mice. Am. J. Physiol. Heart Circ. Physiol. 302, H451–H458 (2012).
Article CAS PubMed Google Scholar
Robertson, A. K. et al. Disruption of TGF-beta signaling in T cells accelerates atherosclerosis. J. Clin. Invest 112, 1342–1350 (2003).
Article CAS PubMed PubMed Central Google Scholar
Halim, H. et al. Ticagrelor induces paraoxonase-1 (PON1) and better protects hypercholesterolemic mice against atherosclerosis compared to clopidogrel. PLoS One 14, e0218934 (2019).
Article CAS PubMed PubMed Central Google Scholar
Karimi, E. et al. Single-cell spatial immune landscapes of primary and metastatic brain tumours. Nature 614, 555–563 (2023).
Article CAS PubMed PubMed Central Google Scholar
Bertocchi, A. et al. Gut vascular barrier impairment leads to intestinal bacteria dissemination and colorectal cancer metastasis to liver. Cancer Cell 39, 708–24 e11 (2021).
Article CAS PubMed Google Scholar
Lamprecht, M. R., Sabatini, D. M. & Carpenter, A. E. CellProfiler: free, versatile software for automated biological image analysis. BioTechniques 42, 71–75 (2007).
Article CAS PubMed Google Scholar
Schneider, C. A., Rasband, W. S. & Eliceiri, K. W. NIH Image to ImageJ: 25 years of image analysis. Nat. methods 9, 671–675 (2012).
Article CAS PubMed PubMed Central Google Scholar
Celada, A., Gray, P. W., Rinderknecht, E. & Schreiber, R. D. Evidence for a gamma-interferon receptor that regulates macrophage tumoricidal activity. J. Exp. Med 160, 55–74 (1984).
Article CAS PubMed Google Scholar
Giuliano, C. J., Lin, A., Girish, V. & Sheltzer, J. M. Generating Single Cell-Derived Knockout Clones in Mammalian Cells with CRISPR/Cas9. Curr. Protoc. Mol. Biol. 128, e100 (2019).
Allen, R. M. et al. LDL delivery of microbial small RNAs drives atherosclerosis through macrophage TLR8. Nat. Cell Biol. 24, 1701–1713 (2022).
Article CAS PubMed PubMed Central Google Scholar
Gu, W. et al. Smooth muscle cells differentiated from mesenchymal stem cells are regulated by microRNAs and suitable for vascular tissue grafts. J. Biol. Chem. 293, 8089–8102 (2018).
Article CAS PubMed PubMed Central Google Scholar
Luo, Y. et al. Macrophagic CD146 promotes foam cell formation and retention during atherosclerosis. Cell Res. 27, 352–372 (2017).
Article CAS PubMed PubMed Central Google Scholar
Howard, O. M. et al. Autoantigens signal through chemokine receptors: uveitis antigens induce CXCR3- and CXCR5-expressing lymphocytes and immature dendritic cells to migrate. Blood 105, 4207–4214 (2005).
Article CAS PubMed PubMed Central Google Scholar
Drevets, D. A. & Campbell, P. A. Macrophage phagocytosis: use of fluorescence microscopy to distinguish between extracellular and intracellular bacteria. J. Immunol. Methods 142, 31–38 (1991).
Article CAS PubMed Google Scholar
Chunhacha, P., Pinkaew, D., Sinthujaroen, P., Bowles, D. E. & Fujise, K. Fortilin inhibits p53, halts cardiomyocyte apoptosis, and protects the heart against heart failure. Cell Death Discov. 7, 310 (2021).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We thank all members of the Fujise Laboratory for their collaborative work, and Sophie Wolfson for her assistance with laboratory management. We thank the McGill University Goodman Cancer Institute Single Cell Imaging and Mass Cytometry Analysis Platform Core Facility for imaging mass cytometry staining and data acquisition. This multi-year project was supported in part by grants from the National Heart, Blood, and Lung Institute within the National Institutes of Health (HL138992, HL117247, HL152723, and HL152683 to K.F.); by the Locke Research Awards and the Harold T. Dodge-John L. Locke Chair in Cardiovascular Medicine at the Division of Cardiology, Department of Internal Medicine, the University of Washington; by an American Heart Association Postdoctoral Fellowship Award (25POST1374001 to M.N.); and by the John Sealy Centennial Chair at the Division of Cardiology, Department of Internal Medicine, the University of Texas Medical Branch at Galveston (to K.F.). Research in the Ihrie Lab was supported by a pilot award from the Vanderbilt Brain Institute; by the National Institutes of Health (R01NS118580 to R.A.I.); by the Ben & Catherine Ivy Foundation (to R.A.I.); and by a gift from the Michael David Greene Brain Cancer Fund at the Vanderbilt–Ingram Cancer Center (to R.A.I. and A.A.B.).

Author information

Decha Pinkaew
Present address: Division of Pulmonary, Critical Care and Sleep Medicine, Department of Medicine, Houston Methodist Hospital, Houston, TX, USA
Preedakorn Chunhacha
Present address: Department of Biochemistry and Microbiology, Faculty of Pharmaceutical Sciences, Chulalongkorn University, Bangkok, Thailand
Preedakorn Chunhacha
Present address: Center of Excellence in Cancer Cell and Molecular Biology, Faculty of Pharmaceutical Sciences, Chulalongkorn University, Bangkok, Thailand
Asa A. Brockman & Rebecca A. Ihrie
Present address: Pediatrics - Neurology, University of Colorado Anschutz Medical Campus, Aurora, CO, USA
Hasseri B. Halim
Present address: Integrative Pharmacogenomics Institute (iPROMISE), Universiti Teknologi MARA Selangor, Selangor, Malaysia
These authors contributed equally: Nattaporn Wanachottrakul, Decha Pinkaew, Sandipan Mukherjee, Preedakorn Chunhacha.

Authors and Affiliations

Division of Cardiology, Department of Internal Medicine, University of Texas Medical Branch at Galveston, Galveston, TX, USA
Nattaporn Wanachottrakul, Decha Pinkaew, Preedakorn Chunhacha, Hasseri B. Halim & Ken Fujise
Division of Cardiology, Department of Internal Medicine, University of Washington, Seattle, WA, USA
Decha Pinkaew, Sandipan Mukherjee, Mari Nakashima, Uttariya Pal, Lena Tanaka, Hanna Huynh & Ken Fujise
Departments of Cell & Developmental Biology and Neurological Surgery, Vanderbilt University School of Medicine, Nashville, TN, USA
Asa A. Brockman & Rebecca A. Ihrie
Rosalind and Morris Goodman Cancer Institute, McGill University, Montreal, Quebec, Canada
Yuhong Wei
Department of Biochemistry and Molecular Biology, University of Texas Medical Branch at Galveston, Galveston, TX, USA
Kota Ramana
Department of Biochemistry, Noorda College of Osteopathic Medicine, Provo, UT, USA
Kota Ramana
Department of Surgery, University of Missouri, Columbia, MO, USA
Shiyou Chen
The Institute of Translational Sciences, University of Texas Medical Branch at Galveston, Galveston, TX, USA
Ken Fujise
Department of Laboratory Medicine and Pathology, University of Washington, Seattle, WA, USA
Ken Fujise

Authors

Nattaporn Wanachottrakul
View author publications
Search author on:PubMed Google Scholar
Decha Pinkaew
View author publications
Search author on:PubMed Google Scholar
Sandipan Mukherjee
View author publications
Search author on:PubMed Google Scholar
Preedakorn Chunhacha
View author publications
Search author on:PubMed Google Scholar
Mari Nakashima
View author publications
Search author on:PubMed Google Scholar
Asa A. Brockman
View author publications
Search author on:PubMed Google Scholar
Uttariya Pal
View author publications
Search author on:PubMed Google Scholar
Hasseri B. Halim
View author publications
Search author on:PubMed Google Scholar
Yuhong Wei
View author publications
Search author on:PubMed Google Scholar
Lena Tanaka
View author publications
Search author on:PubMed Google Scholar
Hanna Huynh
View author publications
Search author on:PubMed Google Scholar
Kota Ramana
View author publications
Search author on:PubMed Google Scholar
Shiyou Chen
View author publications
Search author on:PubMed Google Scholar
Rebecca A. Ihrie
View author publications
Search author on:PubMed Google Scholar
Ken Fujise
View author publications
Search author on:PubMed Google Scholar

Contributions

K.F. conceived the general idea and framework of the project, designed the majority of the experiments, and oversaw the project to its completion. D.P., N.W., P.C., and S.M. contributed equally to the experimental design and its performance, including in vitro, cellular, and animal experiments, in this multi-year project. M.N. performed THP1 cell proliferation and apoptosis assays as well as immunocytochemistry and immunofluorescence staining. L.T. generated hypercholesterolemic mice with fortilin wild-type and knockout genotypes and prepared atherosclerotic arteries for immunofluorescence staining. H.B.H. performed the phagocytosis assay. K.R. and S.M. performed multiplex cytokine assays of mouse and human cytokines, respectively, and interpreted the results. N.W. and S.M. performed in vivo and in vitro foam cell formation assays, respectively. U.P. performed organ fibrosis assays. S.C. helped interpret TGF-β1-related data. S.M. and H.H. performed MΦ transdifferentiation and MSC differentiation assays. S.M. performed efferocytosis assays. K.F. directed the generation and characterization of fortilin^flox/flox mice. D.P. and K.F. generated both fortilin^KO-MΦ and fortilin^KO-MΦ-HC mice. Y.W. performed staining and data acquisition in the IMC experiments. A.A.B. and R.A.I. compiled and analyzed the IMC data in consultation with K.F. D.P., S.M., and K.F. analyzed data, except for imaging mass cytometry (IMC) data. K.F. wrote the manuscript and produced figures. K.F., D.P., S.M., and M.N. proofread the manuscript.

Corresponding author

Correspondence to Ken Fujise.

Ethics declarations

Competing interests

There is no competing financial or non-financial interest present for any authors of the current manuscript. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Peer review

Peer review information

Communications Biology thanks Devendra K. Agrawal, Sivareddy Kotla and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Primary Handling Editors: Jesmond Dalli and Dario Ummarino. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Transparent Peer Review file

Supplementary Information

Description of Additional Supplementary files

Supplementary Data 1

Reporting Summary

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Wanachottrakul, N., Pinkaew, D., Mukherjee, S. et al. Fortilin deficiency induces anti-atherosclerotic phenotypes in macrophages and protects hypercholesterolemic mice against atherosclerosis. Commun Biol 8, 1040 (2025). https://doi.org/10.1038/s42003-025-08425-w

Download citation

Received: 07 June 2024
Accepted: 20 June 2025
Published: 11 July 2025
Version of record: 11 July 2025
DOI: https://doi.org/10.1038/s42003-025-08425-w

Search

Advanced search

Quick links

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing