DOCK2 Promotes Atherosclerosis by Mediating the Endothelial Cell Inflammatory Response

Abstract

The pathology of atherosclerosis, a leading cause of mortality in patients with cardiovascular disease, involves inflammatory phenotypic changes in vascular endothelial cells. This study explored the role of the dedicator of cytokinesis (DOCK)-2 protein in atherosclerosis. Mice with deficiencies in low-density lipoprotein receptor and Dock2 (Ldlr^−/−Dock2^−/−) and controls (Ldlr^−/−) were fed a high-fat diet (HFD) to induce atherosclerosis. In controls, Dock2 was increased in atherosclerotic lesions, with increased intercellular adhesion molecule (Icam)-1 and vascular cell adhesion molecule (Vcam)-1, after HFD for 4 weeks. Ldlr^−/−Dock2^−/− mice exhibited significantly decreased oil red O staining in both aortic roots and aortas compared to that in controls after HFD for 12 weeks. In control mice and in humans, Dock2 was highly expressed in the ECs of atherosclerotic lesions. Dock2 deficiency was associated with attenuation of Icam-1, Vcam-1, and monocyte chemoattractant protein (Mcp)-1 in the aortic roots of mice fed HFD. Findings in human vascular ECs in vitro suggested that DOCK2 was required in TNF-α–mediated expression of ICAM-1/VCAM-1/MCP-1. DOCK2 knockdown was associated with attenuated NF-κB phosphorylation with TNF-α, partially accounting for DOCK2-mediated vascular inflammation. With DOCK2 knockdown in human vascular ECs, TNF-α–mediated VCAM-1 promoter activity was inhibited. The findings from this study suggest the novel concept that DOCK2 promotes the pathogenesis of atherosclerosis by modulating inflammation in vascular ECs.

Graphical abstract

Atherosclerosis is the stiffening of arteries caused by injury to the endothelial cells (ECs), followed by chronic inflammation, dysregulated repair, and plaque development inside the vessel wall.1, 2, 3 It is the major common risk factor of coronary artery disease and stroke, and the primary cause of death in the developed world.4, 5, 6 Currently there is a lack of effective pharmaceutical options for the treatment of patients with atherosclerosis. Further understanding of the pathogenesis of this disease is therefore critical.

Insult to the local ECs represents the first step toward the formation of atherosclerotic lesions. In injured ECs without proper repair, proinflammatory responses occur and cytokines or chemokines are secreted, including tumor necrosis factor (TNF)-α, IL-1β, and monocyte chemoattractant protein (MCP)-1.7, 8, 9 The expression levels of intercellular adhesion molecule (ICAM)-1 and vascular cell adhesion molecule (VCAM)-1, proteins important for the adhesion and activation of leukocytes in inflammatory processes, are increased in ECs during the development of atherosclerosis. These mediators coordinate to induce the attraction and adhesion of monocytes, which differentiate into macrophages and further promote local inflammation. Subsequently, macrophages take up lipid particles and transform into foam cells that eventually appear as lipid plaque.10, 11, 12 Although considerable efforts have been committed to decipher the progression of atherosclerosis, effective therapeutic treatment approaches are limited, and cure remains elusive. Identifying novel mediators of the pathologic events crucial in atherosclerosis, particularly those involved in phenotypic changes in ECs, may provide insights into the pathogenesis and treatment of this disorder.

The dedicator of cytokinesis (DOCK)-2 protein is a member of the DOCK family of small GTPases, specifically the sub-A family.^13,14 It was originally reported to be expressed exclusively in hematopoietic cells. Increasing evidence supports the involvement of DOCK2 in inflammation-related diseases, including, but not limited to, cancer and lung fibrosis.13, 14, 15, 16, 17 It is important for lymphocyte chemotaxis and the differentiation of plasma cells.^14,18 Recent evidence also suggests a role of DOCK2 in immune surveillance.^19,20 DOCK2 modulates the phenotype of smooth muscle cells after vascular injury.^21,22 In addition, it may be involved in high-fat diet (HFD)-induced obesity, primarily by mediating adipose tissue inflammation.¹³ Plasma level of DOCK2 is associated with the extent of coronary artery stenosis.²³ DOCK2 is possibly involved in EC phenotypic modulation, which may be involved in the pathogenesis of atherosclerosis. Therefore, the study of the role of DOCK2 in atherosclerosis addresses a key knowledge gap and may provide novel insight into disease progression.

This study reports a newly defined function of DOCK2 in the progression of atherosclerosis, in particular, the modulation of inflammatory phenotype in ECs. Elevated levels of DOCK2 have been observed in atherosclerotic lesions in humans and in low-density lipoprotein receptor knockout (Ldlr^−/−) mice fed HFD. Interestingly, the up-regulated expression of DOCK2 was colocalized with EC marker CD31 in atherosclerotic lesions in humans and in Ldlr^−/− mice. Dock2 knockout (Dock2^−/−) effectively attenuated the development of atherosclerosis in both aortic roots and aortas in Ldlr^−/− mice fed HFD. DOCK2 deficiency was associated with significant inhibition of the expression of ICAM-1, VCAM-1, and MCP-1 in ECs both in vivo and in vitro. These findings support further investigation of DOCK2 as a novel candidate target for the treatment of patients with atherosclerosis.

Materials and Methods

Animals

Wild-type C57BL/6 mice were purchased from The Jackson Laboratory (Bar Harbor, ME). Dock2^−/− mice were generated on the C57BL/6 background as reported previously.15, 16, 17^,21,24 The Ldlr^−/− mice on C57BL/6 background (B6.129S7-Ldlrtm1Her/J) were obtained from The Jackson Laboratory. The Dock2^−/− mice were bred with Ldlr^−/− mice to generate double-heterozygous mice. The F₁ generation was further crossed to generate homozygous mice deficient in Ldlr and Dock2 (Ldlr^−/−Dock2^−/−). Genotype was verified using PCR with specific primers for Dock2 and Ldlr. Primers for Dock2 genotyping were as follows: DOCK2SC1 (reverse), 5′-ATCTGTCTGCATGATGGATGCTT-3′; DOCK2SC2 (forward), 5′-AATGCCTGCTCTTTACTGAAGG-3′; and DOCK2SCB2 (forward), 5′-AAGTGACCTTACCTGTGACAG-3′. Primers for Ldlr genotyping were as follows: OIMR0092 (forward), 5′-AATCCATCTTGTTCAATGGCCGATC-3′; OIMR3349 (forward), 5′-CCATATGCATCCCCAGTCTT-3′; and OIMR3350 (reverse), 5′-GCGATGGATACACTCACTGC-3′. Male Ldlr^−/−Dock2^−/− and Ldlr^−/− (control) mice aged 8 to 10 weeks were used for the experiments and fed either chow diet or HFD (catalog number d12108C; Research Diets Inc., New Brunswick, NJ) for 4 or 12 weeks to induce atherosclerosis.^25,26 All mice were housed in conventional conditions and received humane care in compliance with the Principles of Laboratory Animal Care formulated by the National Society for Medical Research and the NIH’s Guide for the Care and Use of Laboratory Animals.²⁷ All of the experimental procedures used in this study were approved by the Institutional Animal Care and Use Committee of The University of Georgia (Athens, GA) and The University of Texas Health Science Center at Tyler (Tyler, TX).

Atherosclerosis Specimen Collection from Human Donors

Normal and atherosclerotic human coronary arteries were obtained from autopsy materials, as described previously.²⁸ The protocol for the use of human tissue samples was approved by the Institutional Review Board at the University of Missouri (Columbia, MO; project 2026026). All of the experiments using human specimens were conducted in accordance with the relevant regulations and guidelines.

Atherosclerotic Lesion Analysis

With mice under anesthesia with 2% isoflurane inhalation, mouse aortic roots and whole aortas were collected as previously described.²⁹ To detect en face plaque(s), the aortas were cut open longitudinally, and then fixed in 4% paraformaldehyde overnight. Oil red O (ORO) staining was performed as previously reported.^30,31 Briefly, aortic roots were embedded in optimal cutting temperature compound, and mouse aortic root sections (10 μm) were used for ORO staining. In addition, the aortic roots were fixed in buffered neutral formalin overnight, followed by dehydration in a series of ethanol and xylene solutions and embedding in paraffin. Sections (5 μm) were then deparaffinized and rehydrated. Staining with hematoxylin and eosin (catalog number KTHNEPT) and Verhoeff-Van Gieson elastic (catalog number KTMTR2) were performed as described^21,32,33 using commercially available kits purchased from American MasterTech Scientific (Lodi, CA). Images were taken using a Nikon microscope (Nikon Instruments, Melville, NY). The atherosclerotic lesion areas were quantified using NIH ImageJ software version 1.52a (NIH, Bethesda, MD; http://imagej.nih.gov.ij).

Cell Culture

Human umbilical vein (HUV) ECs (catalog number C2517A) and human coronary artery (HCA) ECs (catalog number CC-2585) were purchased from Lonza (Walkersville, MD) and cultured in EGM^TM-2 medium (catalog number CC-3202; Lonza) in 37°C, 5% CO₂ conditions. Cells at less than eight passages were used. TNF-α (catalog number 210-TA; R&D Systems, Minneapolis, MN) and oxidized low-density lipoprotein (ox-LDL; catalog number J65591; Alfa Aesar, Haverhill, MA) were used to treat the cells at indicated doses and durations. The NF-κB inhibitor ammonium pyrrolidinedithiocarbamate (catalog number P8765) was purchased from Sigma-Aldrich (St. Louis, MO).

Mouse aortic artery ECs from WT and Dock2^−/− mice were isolated and cultured as reported previously.^34,35 Briefly, the aortas from the aortic arch to the abdominal aorta were dissected out and briefly washed with serum-free Dulbecco's modified Eagle's medium (DMEM). The aortas were filled with a solution containing 1 mg/mL collagenase type II (catalog number LS004176; Worthington Biochemical Corporation, Lakewood, NJ) in serum-free DMEM and incubated at 37°C for 45 minutes. ECs were removed from the aorta by flushing with 5 mL DMEM supplied with 20% fetal bovine serum. The ECs were centrifuged, resuspended in DMEM with 20% fetal bovine serum, and seeded in dishes precoated with type I collagen. After incubation for 2 hours at 37°C, the medium was aspirated, and the cells were washed with phosphate-buffered saline (PBS). The attached ECs were then cultured with medium G^34,35 until confluence.

Western Blot Analysis

Cell lysate preparation and Western blot analysis were performed as previously reported.^15,17,36 Human vascular ECs were washed with cold PBS followed by protein extraction using radioimmunoprecipitation assay buffer (catalog number AAJ63306AK; Thermo Fisher Scientific, Waltham, MA) supplied with protease inhibitors (catalog number 87786; Thermo Fisher Scientific) and a phosphatase inhibitor (catalog number 78428; Thermo Fisher Scientific). Total protein concentration was determined using the bicinchoninic acid protein assay reagent (catalog number 23225; Thermo Fisher Scientific) with the Infinite 200 Pro plate reader (VWR International, Radnor, PA). Lysates were denatured by boiling in Laemmli SDS buffer. SDS-PAGE electrophoresis was run and polyvinylidene difluoride membrane was used for transfer, followed by blocking with 5% nonfat milk and incubation with primary antibodies overnight at 4°C. Horseradish peroxidase–conjugated secondary antibodies were used for incubation for 1 hour at room temperature with gentle shaking. The primary antibodies used included DOCK2 (1:500; catalog number 09-454, Sigma-Aldrich), VCAM-1 (1:1000; catalog number ab134047; Abcam, Cambridge, MA), ICAM-1 (1:1000; catalog number sc-8439; Santa Cruz Biotechnology, Dallas, TX), MCP-1 (1:1000; catalog number AF279NA; R&D Systems), phospho (p)-NF-κB p65 (1:500; catalog number 3033L; Cell Signaling Technology, Danvers, MA), NF-κB p65 (1:1000; catalog number 8242S; Cell Signaling Technology), lamin B1 (1:1000; catalog number SC-374015; Santa Cruz Biotechnology), glyceraldehyde-3-phosphate dehydrogenase (GAPDH; 1:5000; catalog number 60004-1-Ig; Proteintech, Rosemont, IL), and α-tubulin (1:3000; catalog number T5168; Sigma-Aldrich).

Quantitative Real-Time RT-PCR

The quantitative real-time (RT-PCR (RT-qPCR) method has been described previously.^16,17,37 Briefly, culture media were removed, and cells were collected into 1 mL of TRIzol reagent (catalog number 15596018; Thermo Fisher Scientific) for the extraction of total RNA. After the determination of RNA concentration using the plater reader, 1 μg of total RNA was used for reverse transcription to generate cDNA template using the iScript cDNA Synthesis kit (catalog number 1708891; Bio-Rad Laboratories, Hercules, CA). The cDNA templates were used for qPCR analysis of target gene expression using iTaq SYBR Green supermix (catalog number 1725121; Bio-Rad). The cyclophilin gene was used as an internal control. All samples were run in triplicate. The primers for the genes were as follows: DOCK2: forward, 5′-TGCTGAAGTGGCGTATGAAG-3′; reverse, 5′-AATTTCCGGTCTGCAATGAG-3′; VCAM1: forward, 5′-TTGGGAGCCTCAACGGTACT-3′; reverse, 5′-GCAATCGTTTTGTATTCAGGGGA-3′; ICAM1: forward, 5′-GTGATGCTCAGGTATCCATCCA-3′; reverse, 5′-CACAGTTCTCAAAGCACAGCG-3′; CCL2: forward, 5′-CAGCCAGATGCAATCAATGCC-3′; reverse, 5′-TGGAATCCTGAACCCACTTCT-3′; and CYP (cyclophilin gene): forward, 5′-GTGGTCTTTGGGAAGGTGAA-3′; reverse, 5′-TTACAGGACATTGCGAGCAG-3′.

IHC and immunofluorescence Staining

Staining for immunohistochemistry (IHC) analysis was performed as previously reported, with some modification.^13,15,21 Paraffin-embedded tissue sections were rehydrated, incubated with 3% H₂O₂ in ethanol for 15 minutes, and washed with PBS. Sections were then blocked with 10% goat serum in PBS for 30 minutes at room temperature. The sections were probed with primary antibodies (1:100) against DOCK2, ICAM-1, VCAM-1, p–NF-κB p65, and F4/80 (catalog number ab6640; Abcam). The DAB kit (catalog number SK4100; Vector Laboratories, Newark, CA) was used to visualize the staining. The sections were counterstained with hematoxylin and sealed with paramount for image capture using a NiU microscope (Nikon). Quantification was conducted using ImageJ. Immunofluorescence (IF) staining of sections was performed in a similar manner. The primary antibodies were DOCK2 (1:100; catalog number UM800039; Thermo Fisher Scientific) and CD31 (1:100; catalog number ab28364; Abcam), and the secondary antibodies were goat anti-mouse Alexa Fluor Plus 488 (1:100; catalog number A32723) or goat anti-rabbit Texas Red (1:100; catalog number T-2767) (InvitroGen, Waltham, MA). Nuclei were stained with DAPI (catalog number H-1500DAPI; Vector Laboratories). Fluorescence images were captured using the NiU microscope.

Cell IF staining was performed as reported previously.36, 37, 38 Briefly, HCAECs on glass coverslips were fixed with 4% paraformaldehyde for 10 minutes, permeabilized with 0.1% Triton X-100 in PBS for 10 minutes, and blocked with 10% normal goat serum for 30 minutes at room temperature. The cells were then incubated with anti-DOCK2 antibody overnight at 4°C. The cells were washed with PBS with Tween 20 and incubated with secondary antibodies for 1 hour at room temperature. Nuclei were stained with DAPI. Fluorescence images were captured using a fluorescence microscope (Nikon).

Monocyte–Endothelial Cell Adhesion Assay

THP-1 (catalog number TIB-202; ATCC, Manassas, VA), the human leukemia monocytic cell line, was used for this assay.³⁴ The cells were labeled using incubation with 2.5 μmol/L calcein AM (catalog number C1359; Sigma-Aldrich) for 30 minutes in a 37°C 5% CO₂ incubator. The labeled cells were washed with 10 mL of PBS three times and then resuspended at 1 × 10⁶ cells/mL in culture medium. The labeled cells (1 × 10⁶) were then added to confluent primarily cultured ECs from wild-type and Dock2^−/− mice, which were pretreated with vehicle or TNF-α (20 ng/mL) for 24 hours. After incubation for 1 hour in an incubator at 37°C with 5% CO₂, the cells were washed with PBS, and the adherent labeled THP-1 cell images were captured using a fluorescence microscope (Nikon). Labeled THP-1 cells from six random fields were counted.

Luciferase Activity Analysis

The human VCAM-1 promoter region (−504 to +45 bp relative to the mapped transcription start site) was amplified by PCR using human genomic DNA. The products were cloned into the pGL4 vector as previously reported.³⁴ Adenovirus-expressing control (Ad-shCtrl) or DOCK2 shRNA (Ad-shDOCK2) was used to infect HUVECs.^16,21 The next day, VCAM-1 promoter fragment containing vector was transfected into the cells using Lipofectamine LTX reagent (catalog number 15338100; Invitrogen) for 24 hours. Transfected cells were then treated with vehicle or 20 ng/mL TNF-α for an additional 24 hours. Luciferase activity was measured according to the manufacturer’s manual for the Dual-Luciferase Reporter assay system (catalog number E1960; Promega, Madison, WI).^36,39,40

Statistical Analysis

All data are expressed as the means ± SD. In two-group comparisons, the two-tailed t-test was used. In three or four groups of data, one-way analysis of variance followed by the Bonferroni post hoc test for multiple comparisons was used to determine the significance of differences using Prism software version 9.0 (GraphPad Software, Boston, MA). P < 0.05 was considered statistically significant.

Results

DOCK2 Expression Is Increased in Atherosclerotic Lesions

The expression of Dock2 in a preclinical model of atherosclerosis was first determined. Ldlr^−/− mice fed HFD for 2 weeks showed lesions of atherosclerosis in aortic roots, as indicated by hematoxylin and eosin staining (Figure 1A). Lesions were increased after 4 weeks of HFD feeding (Figure 1A). Interestingly, Dock2 expression was observed with the progression of atherosclerosis. It was locally increased in the lesions of aortic roots in Ldlr^−/− mice after 2 weeks of HFD feeding, as indicated by IHC staining. It encompassed the whole lesion area after 4 weeks of HFD feeding (Figure 1, B and C). Similar results were observed with IF staining (Supplemental Figure S1). The adhesion markers Icam-1 and Vcam-1 were also increased in the aortic roots of Ldlr^−/− mice fed with HFD for 4 weeks but not with chow diet (Supplemental Figure S2). These findings suggest a potential role of DOCK2 in the pathogenesis of atherosclerosis. To test for potential clinical relevance, DOCK2 expression in humans with atherosclerosis was further determined. DOCK2 expression was significantly elevated in the coronary arteries of atherosclerotic patients compared to control arteries from healthy donors (Figure 1, D and E), as indicated by IHC staining. This suggests that DOCK2 may also contribute to human atherosclerosis.

Figure 1.

DOCK2 expression is enhanced in atherosclerotic lesions. A: Representative images of aortic roots of Ldlr^−/− mice fed chow diet or HFD for 2 and 4 weeks, by hematoxylin and eosin staining. Arrows indicate lesion formation. Images on the right represent the boxed regions (left) at increased magnification; dashed lines indicate the borders of atherosclerotic lesions. B and C: Representative immunohistochemistry (IHC) staining (B) and quantification (C) of Dock2 expression in aortic roots in Ldlr^−/− mice fed chow diet or HFD for 2 and 4 weeks. In B, images on the right represent the boxed regions (left) at increased magnification; dashed lines indicate the borders of atherosclerotic lesions. D and E: IHC staining (D) and quantification (E) of DOCK2 expression in human normal vessel and lesions in atherosclerotic (AS) patients. Data are expressed as means ± SD. n = 4 per group (E); n = 5 mice per group (C). ∗P < 0.05 versus chow diet (one-way analysis of variance followed by Bonferroni post hoc test); ^†P < 0.05 versus normal vessel (two-tailed t-test). Scale bars: 200 μm (A and B); 100 μm (D).

Dock2 Deficiency Attenuates HFD-Induced Atherosclerosis in Ldlr^−/− Mice

To further determine whether DOCK2 plays a role in atherosclerosis, Dock2-deficient (Dock2^−/−) Ldlr^−/− mice were generated. En face ORO showed dramatic staining of lipid accumulation (fatty streak, red) in the aortas of Ldlr^−/− mice fed HFD for 12 weeks, with scattered staining along the abdominal aorta. In contrast, Ldlr^−/−Dock2^−/− mice that received HFD for 12 weeks had much less ORO staining in the aorta (Figure 2A). Quantification of positive areas of ORO staining further demonstrated a significant decrease of lipid deposition in the Ldlr^−/−Dock2^−/− group compared to their Ldlr^−/− counterparts (Figure 2B). On staining with hematoxylin and eosin (Figure 2C) and elastic fiber (Verhoeff-Van Gieson elastic) (Figure 2D) of the aortic roots, the area of lesions in the Ldlr^−/− group was notably larger than that in the Ldlr^−/−Dock2^−/− group. On cross-sectional views of ORO staining of the aortic roots from those mice, results were consistent with those from en face staining. No lipid deposition was found in the aortic roots of mice in the Ldlr^−/− and Ldlr^−/−Dock2^−/− groups with the chow diet. Conversely, with the HFD for 12 weeks, lipid accumulation was dramatically increased in the inner layer of the aortic roots in Ldlr^−/− mice compared with that in the Ldlr^−/− mice fed chow diet (Figure 2, E and F). With Dock2 deficiency, HFD-induced lipid deposition was significantly decreased in the aortic roots (Figure 2, E and F). These findings implicate DOCK2 as a protein factor contributing to the development of atherosclerosis.

Figure 2.

Dock2^−/− attenuates atherosclerosis formation in Ldlr^−/− mice. A: Representative images of plaques in aortas in Ldlr^−/− and Ldlr^−/−Dock2^−/− mice fed HFD for 12 weeks, by en face oil red O (ORO) staining. B: Quantification of plaque areas as a percentage of total aortic areas in A. C and D: Representative images of aortic root sections in Ldlr^−/− and Ldlr^−/−Dock2^−/− mice fed HFD for 12 weeks, by staining with hematoxylin and eosin (C) or Verhoeff-Van Gieson elastic (D). E and F: Representative cross-sectional images (E) and quantification (F) of lesions in aortic roots in Ldlr^−/− and Ldlr^−/−Dock2^−/− mice fed chow diet or HFD for 12 weeks, by ORO staining. Data are expressed as means ± SD. n = 5. ∗P < 0.05 versus Ldlr^−/− mice fed HFD (two-tailed t-test); ^†P < 0.05 versus Ldlr^−/− mice fed chow diet (analysis of variance followed by Bonferroni post hoc test); and ^‡P < 0.05 versus Ldlr^−/− mice fed HFD (analysis of variance followed by Bonferroni post hoc test). Scale bars = 200 μm (C–E).

Colocalization of DOCK2 with the EC Marker CD31

EC dysfunction plays an important role in the development of atherosclerosis. Co-staining of DOCK2 with the endothelial marker CD31⁴¹ has been detected in the atherosclerotic lesions. As indicated by IF staining, Dock2 (green) was expressed in the atherosclerotic lesions of the aortic roots (Figure 3A) in Ldlr^−/− mice fed HFD. Interestingly, the expression of Dock2 was partially colocalized with that of Cd31 (red, Figure 3A). Approximately 51% of CD31⁺ cells were positively stained with DOCK2 (Figure 3C). The results suggest that ECs contribute to increased DOCK2 expression during the development of atherosclerosis. Consistent results were observed in humans with atherosclerosis. DOCK2 (green) expression in the coronary arteries of atherosclerotic patients was partially colocalized with CD31 (red) expression (Figure 3B), with about 80% of CD31⁺ cells positively stained with DOCK2 (Figure 3C). Together, these findings suggest that DOCK2 is expressed in ECs in atherosclerotic lesions.

Figure 3.

DOCK2 is induced and colocalized with endothelial cell marker CD31 in atherosclerotic lesions. A and B: Representative images of DOCK2 (green) and CD31 (red; ECs) in aortic roots in Ldlr^−/− mice fed HFD for 12 weeks (A) and in human vessels with atherosclerotic (AS) lesions (B), by IF staining. DAPI (blue) stains the nuclei. Arrows indicate a representative area of colocalization. C: Percentages of DOCK2⁺CD31⁺ cells relative to the total CD31⁺ cells quantified from tissue sections in A and B. n = 3 per group. Scale bars = 50 μm (A and B).

DOCK2 Is Increased in Human Vascular ECs along with Phenotypic Changes

Proinflammatory cytokines and mediators play indispensable roles in atherogenesis. Given that DOCK2 is expressed in the ECs of atherosclerotic lesions, it is interesting to know whether TNF-α and ox-LDL mediate DOCK2 expression in human vascular ECs. In HUVECs treated with various concentrations of TNF-α for 24 hours, DOCK2 expression significantly increased with TNF-α at 20 ng/mL and above (Figure 4, A and B). Significant increases in adhesion-related molecules VCAM-1 and ICAM-1 were also observed with TNF-α treatment (Figure 4, A and B). Similarly, the expression levels of DOCK2, VCAM-1 and ICAM-1 were increased in HUVECs treated with ox-LDL (Figure 4C). To explore the potential relationship between DOCK2 expression and human coronary artery disease, the expression of DOCK2 in HCAECs in response to TNF-α and ox-LDL treatment was determined. Consistently, DOCK2 as well as VCAM-1 and ICAM-1 were increased in HCAECs with both TNF-α and ox-LDL (Figure 4, D and E). A time-dependent increase in DOCK2, along with ICAM-1 and VCAM-1, was observed with TNF-α in HUVECs (Supplemental Figure S3A) and HCAECs (Supplemental Figure S3B). DOCK2 mRNA, along with VCAM1 and ICAM1, was significantly increased in HUVECs after TNF-α treatment (Supplemental Figure S4, A–C). These findings suggest that DOCK2 is induced by TNF-α and ox-LDL, along with transformation of human vascular ECs to an inflammatory phenotype.

Figure 4.

DOCK2 expression is induced in human endothelial cells with phenotypic alteration. A and B: Expression (A) and quantification (B) of DOCK2, VCAM-1, and ICAM-1 in human umbilical vein (HUV) ECs treated with TNF-α (0, 20, 40, and 60 ng/mL) for 24 hours. C: DOCK2, VCAM-1, and ICAM-1 expression in HUVECs treated with oxidized low-density lipoprotein (ox-LDL; 0, 50, and 100 ng/mL) for 24 hours. D and E: DOCK2, VCAM-1, and ICAM-1 expression in human coronary artery ECs treated with TNF-α (D) or ox-LDL (E). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is the loading control. n = 4. ∗P < 0.05 versus vehicle-treated control (0 ng/mL; analysis of variance followed by Bonferroni post hoc test).

DOCK2 Contributes to the Acquisition of EC Proinflammatory Phenotype

To further explore whether DOCK2 contributes to EC phenotypic changes, HUVECs were transfected with scramble and DOCK2 siRNA (siDOCK2) for 24 hours, followed by TNF-α (20 ng/mL) treatment for an additional 24 hours. Western blot analysis indicated a significant attenuation of the expression of both VCAM-1 and ICAM-1 with TNF-α in HUVECs with DOCK2 knockdown (Figure 5, A and B), suggesting an essential role of DOCK2 in TNF-α-induced inflammatory EC phenotype. In addition, with DOCK2 knockdown by adenovirus DOCK2 shRNA, TNF-α–induced VCAM-1 and ICAM-1 expression in HCAECs was dramatically inhibited (Figure 5C), suggesting a potential role of DOCK2 in the development of coronary artery disease in humans. Importantly, in HFD-induced atherosclerosis, the expression of Vcam-1 and Icam-1 in aortic root was significantly lower in Ldlr^−/−Dock2^−/− mice compared with that in Ldlr^−/− mice, as determined by IHC staining (Figure 5, D and E). In addition to VCAM-1 and ICAM-1, ECs might express MCP-1 to facilitate monocyte attachment and transmembrane migration to promote lesion formation.42, 43, 44 Given that DOCK2 has been associated with increased MCP-1 expression in human lung fibroblast cells to induce lung injury,¹⁵ whether DOCK2 induces MCP-1 expression in ECs was tested. MCP-1 expression time- and dose-dependently increased with TNF-α in HUVECs (Supplemental Figures S4D and S5, A and B). Importantly, with knockdown of DOCK2 using siRNA, the induction of MCP-1 was attenuated (Supplemental Figure S5C). With Dock2 deficiency, the expression of Mcp-1 was attenuated in the aortic roots of Ldlr^−/− mice fed HFD for 12 weeks (Supplemental Figure S5D). The role of DOCK2 in the monocyte–EC interaction was evaluated to assess the functional significance of DOCK2 on EC phenotypic changes. Monocyte adhesion to ECs increased with TNF-α treatment (20 ng/mL for 24 hours). However, with DOCK2 knockout, the number of monocytes attached to ECs ex vivo significantly decreased (Figure 5, F and G). Together, these findings suggest an essential role of DOCK2 in mediating the formation of a proinflammatory EC phenotype.

Figure 5.

DOCK2 is essential for TNF-α–induced VCAM-1 and ICAM-1 expression and monocyte–EC interaction. A and B: Western blot analysis (A) and quantification (B) of DOCK2, VCAM-1, and ICAM-1 expression in human umbilical vein ECs transfected with scramble or DOCK2 siRNA for 24 hours and treated with vehicle (−) or TNF-α (20 ng/mL) (+) for 24 hours. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is the loading control. C: DOCK2, VCAM-1, and ICAM-1 expression in human coronary artery ECs infected with adenovirus expressing control (Ad-shCtrl) or DOCK2 shRNA (Ad-shDOCK2) for 24 hours and treated with vehicle (−) or TNF-α (20 ng/mL) (+) for 24 hours, by Western blot analysis. D: Vcam-1 and Icam-1 expression in lesions of aortic roots in Ldlr^−/− and Ldlr^−/−Dock2^−/− mice fed HFD for 12 weeks, by immunohistochemistry (IHC) analysis. E: Quantitative analysis of Vcam-1 and Icam-1 expression, by measurement of the IHC staining as the percentage of total lesion areas shown in D. F: Adhesion of calcein-AM–labeled human monocytes to quiescent or TNF-α–activated (20 ng/mL, 24 hours) wild-type (WT) or Dock2 knockout (Dock2^−/−) ECs. G: Quantification of adherent monocytes shown in F. Data are expressed as means ± SD. n = 4 (B and E); n = 6 (G). ∗P < 0.05 versus vehicle-treated (−) with scramble siRNA; ^†P < 0.05 versus TNF-α–treated scramble siRNA; ^‡P < 0.05 versus Ldlr^−/− mice fed HFD for VCAM-1; ^§P < 0.05 versus Ldlr^−/− mice fed HFD for ICAM-1; ^¶P < 0.05 versus vehicle-treated WT; and ^‖P < 0.05 versus TNF-α–treated WT (all, analysis of variance followed by Bonferroni post hoc test). Scale bars: 50 μm (D); 100 μm (F).

DOCK2 Promotes Proinflammatory EC Phenotype by Activating NF-κB Signaling

NF-κB is a major pathway involved in EC inflammation.^45,46 DOCK2 has been associated with NF-κB activation in macrophages and lung fibroblasts after lung injury.^15,47 Whether DOCK2 regulates VCAM-1 and ICAM-1 expression through NF-κB signaling in ECs was investigated next. With the pharmaceutical inhibitor of NF-κB, ammonium pyrrolidinedithiocarbamate, TNF-α–induced expression of VCAM-1 and ICAM-1 in HUVECs was significantly attenuated (Figure 6, A and B). Importantly, with knockdown of DOCK2 using siRNA, TNF-α–induced phosphorylation of NF-κB was significantly attenuated in HUVECs (Figure 6, C and D). Consistently, with DOCK2 knockout, p–NF-κB expression in atherosclerotic lesions was suppressed (Supplemental Figure S6). These findings suggest that DOCK promotes EC proinflammatory phenotype, at least partially, through the regulation of NF-κB signaling.

Figure 6.

DOCK2 is required for NF-κB activation in ECs. A and B: Western blot analysis (A) and quantification (B) of VCAM-1 and ICAM-1 expression in human umbilical vein (HUV) ECs treated with ammonium pyrrolidinedithiocarbamate (PDTC) (NF-κB inhibitor; 10 μmol/L) for 30 minutes followed by TNF-α (20 ng/mL) (+) or vehicle (−) for 24 hours. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) is the loading control. C and D: Western blot analysis (C) and quantification (D) of p–NF-κB and NF-κB expression in HUVECs transfected with scramble or DOCK2 siRNA (siDOCK2) for 24 hours and treated with vehicle (−) or TNF-α (20 ng/mL) (+) for 2 hours. Data are expressed as means ± SD. n = 4. ∗P < 0.05 versus vehicle (−); ^†P < 0.05 versus TNF-α; ^‡P < 0.05 versus vehicle-treated scramble (−); and ^§P < 0.05 versus TNF-α–treated scramble (all, analysis of variance followed by Bonferroni post hoc test).

DOCK2 Enhances VCAM-1 Transcription in ECs

To further dissect the role of DOCK2 in mediating proinflammatory mediators in ECs, whether DOCK2 can act as a transcriptional (co)factor was determined. TNF-α–induced DOCK2 expression was localized in both the nuclei and cytoplasm in HCAECs (Figure 7A), as indicated by IF staining. Subcellular fractionation Western blot analysis showed that DOCK2 was localized in both the cytoplasmic and nuclear compartments with TNF-α treatment (Figure 7B). Given that DOCK2 nuclear portion was dramatically elevated with TNF-α treatment, its transcriptional activity was tested. A luciferase assay was used for determining whether DOCK2 regulates VCAM-1 promoter activity. HCAECs were transduced with Ad-shCtrl or Ad-shDOCK2, followed by transfection with VCAM-1 promoters and subsequent TNF-α induction. With TNF-α, VCAM-1 promoter activity was significantly enhanced, as previously reported.^34,48 With DOCK2 knockdown, however, TNF-α–mediated VCAM-1 promoter activity was significantly inhibited (Figure 7C). These findings suggest that DOCK2 may serve as a transcription (co)factor to increase adhesion molecule expression in ECs.

Figure 7.

DOCK2 localizes in nuclei and regulates VCAM-1 promoter activity in ECs. A: DOCK2 expression and cellular location in human coronary artery (HCA) ECs were treated with vehicle or TNF-α (20 ng/mL) for 12 hours, on immunofluorescence staining performed using DOCK2 antibody. DAPI stains the nuclei. TNF-α increases DOCK2 nuclear and cytoplasmic expression. B: DOCK2 expression in HCAECs treated with TNF-α (20 ng/mL) (+) or vehicle (−) for 24 hours, on Western blot analysis. Nuclear and cytoplasmic proteins were isolated using nuclear extraction kit. Nuclear protein internal control, lamin B1; cytoplasmic protein internal control, α-tubulin. C: DOCK2 regulates VCAM-1 promoter activity in HCAECs transduced with Ad-shCtrl or Ad-shDOCK2 and transfected with VCAM-1-promoter-luc construct followed by TNF-α treatment for 12 hours, on luciferase assay. n = 4. ∗P < 0.05 versus vehicle-treated Ad-shCtrl; ^†P < 0.05 versus TNF-α–treated Ad-shCtrl (both, analysis of variance followed by Bonferroni post hoc test). Original magnification, ×400 (A).

Discussion

Atherosclerosis may be fatal when the buildup of plaque breaks and clogs the vessel to the heart or brain. The development of atherosclerosis is a chronic process that entails phenotypic alteration and inflammatory infiltration in ECs.^1,10 DOCK2 is important for the inflammatory response in several cell types, including adipocytes and fibroblasts.^13,15 The findings from the present study suggest a role of DOCK2 in the regulation of vascular EC phenotype during the development of atherosclerotic lesions. In mice with Dock2 deficiency, HFD-induced lesion formation was reduced and proinflammatory cytokine expression in the lesions was inhibited. Findings in human vascular ECs in vitro suggested that DOCK2 is crucial for TNF-α–mediated expression of VCAM-1, ICAM-1, and MCP-1 in ECs, and that DOCK2 is a participant in the pathogenesis of atherosclerosis.

In Ldlr^−/− mice fed HFD, an often-used model of atherosclerosis,^25,26 Dock2 expression was increased in the lesions of aortic roots where the worst lesion occurred (Figure 1 and Supplemental Figure S1) at the early stages of atherosclerosis (ie, 2 or 4 weeks after HFD feeding). Interestingly, Dock2 expression remained after longer exposure to HFD (12 weeks) (Figure 3A), suggesting a potential role of DOCK2 in disease progression. The enhanced levels of DOCK2 in the lesions of humans with atherosclerosis (Figures 1 and 3B) strongly imply that DOCK2 could likewise contribute to coronary artery atherosclerosis in humans. DOCK2 is involved in chronic inflammation contributing to pulmonary fibrosis^16,17 and inflammatory bowel disease.⁴⁹ The present study identified a function of DOCK2 in regulating EC inflammation and contributing to the development of atherosclerosis. Dock2 deficiency in mice significantly decreased atherosclerotic lesion size, attenuated lipid deposition in aortic roots and aortas (Figure 2), and blunted the HFD-induced expression of Icam-1, Vcam-1, and Mcp-1 in aortic roots in Ldlr^−/− mice (Figure 5D and Supplemental Figure S5).

Phenotypic alteration in ECs is a crucial pathologic event in the development of atherosclerosis. Injury to the ECs, regardless of origin, causes the release of ICAM-1, VCAM-1, and MCP-1, which are crucial for the subsequent attraction and attachment of macrophages to the site of vessel injury.^7,50,51 DOCK2 is a mediator of inflammatory EC phenotype that is induced by TNF-α and ox-LDL. DOCK2 expression was increased and colocalized with CD31 (an EC marker) in the aortic roots of mice fed HFD for 12 weeks (Figure 3A) and in the lesions of humans with atherosclerosis (Figure 3B). DOCK2 expression was increased in both HUVECs and HCAECs (Figure 4 and Supplemental Figure S4), and EC phenotype was altered as indicated by ICAM-1 and VCAM-1. DOCK2 expression was increased with TNF-α and ox-LDL with the cell phenotypic changes (Figure 4, D and E). Knockdown of DOCK2 significantly attenuated the induction of ICAM-1 and VCAM-1 (Figure 5). DOCK2 knockdown or knockout dramatically inhibited MPC-1 expression both in vitro and in vivo (Supplemental Figure S5). Importantly, monocyte adhesion to vascular ECs was impaired in Dock2-deficient mouse ECs (Figure 5, F and G). Macrophage infiltration in the atherosclerotic lesions was reduced in Dock2-deficient mice fed HFD for 12 weeks (Supplemental Figure S7). Collectively, these results support the involvement of DOCK2 in the development of atherosclerosis, at least in part, by promoting EC inflammation and monocyte adhesion to ECs.

The NF-κB signaling pathway contributes to the progression of atherosclerosis,⁵² and in mice, the inhibition of NF-κB in ECs protects against atherosclerosis.⁵³ DOCK2 is involved in the activation of NF-κB signaling in cell types including macrophages and fibroblasts.^15,21 In the present study, DOCK2 was found to be in involved in the activation of NF-κB p65 in ECs. siRNA knockdown of DOCK2 attenuated TNF-α–induced phosphorylation of NF-κB p65 (Figure 6). Moreover, a large portion of DOCK2 was located in the nuclei of HCAECs, as shown by IF staining and subcellular fractionation Western blot analysis (Figure 7, A and B), implying that DOCK2 may act as a transcription (co)factor in the mediation of gene transcription in ECs. DOCK2 knockdown also attenuated TNF-α activation of VCAM-1 promotor (Figure 7C). These data support the inference that DOCK2 induces EC inflammatory phenotype through the activation of NF-κB signaling and further induction of the transcription of proinflammatory genes. An open question is whether DOCK2 mediates NF-κB activation in macrophages to contribute to atherosclerosis. This specific hypothesis was not addressed in the present study but will be tested as a future extension of the current work.

The present study had a few limitations. EC-specific Dock2 knockout mice were not used. Given that diverse cell types are involved in the development of atherosclerosis, it is important to test the cell-specific role of DOCK2 in the development of atherosclerosis, particularly using EC-specific Dock2 knockout mice. The VEcadherin-Cre mouse is a widely used model for generating endothelial-specific knockout mice.⁵⁴ An ongoing priority is the generation of EC-specific Dock2 knockout mice by crossing Dock2^fl/fl mice with VEcadherin-Cre mice and identifying the role of endothelial DOCK2 in the development of atherosclerosis. Another limitation was that the potential interaction of DOCK2 with NF-κB was not studied. DOCK2 is expressed in the nuclei and mediates VCAM-1 promoter activity in ECs; it is likely that DOCK2 interacts with NF-κB to activate the transcription of proinflammatory genes. Further understanding of the mechanisms of DOCK2 in mediating EC phenotypic alteration is an interesting subject for future study. Only male mice were used in this study, representing a limitation in experimental design. Given that the outcome of atherosclerosis can be different in female and male mice,^55,56 whether there is a sex-dependent effect of DOCK2 in the development of atherosclerosis warrants an independent study in the future. Another limitation was that pharmacologic or other therapeutic interventions were not used for assessing the role of DOCK2 in atherogenesis, but such studies may likewise be conducted in the future.

In summary, the findings from this study support the novel role of DOCK2 in the regulation of EC phenotypic alteration and that DOCK2 is essential for the progression of atherosclerosis. DOCK2, as a transcription (co)factor, increased activation of NF-κB signaling as well as ICAM-1 and VCAM-1 expression levels. Further studies targeting DOCK2 as a novel strategy for the treatment or prevention of atherosclerosis is warranted.

Author Contributions

X.G. and S.-Y.C. designed the research; X.G., G.Q., O.A., and D.C. performed experiments; X.G. and G.Q. analyzed the data and wrote the manuscript; and T.A.T., S.I., and S.Y.-C. edited and revised the manuscript. All of the authors approved the final version of the manuscript.

Disclosure Statement

None declared.

Supplemental Data

Supplemental Figure S1

Dock2 expression is induced in lesions of aortic roots from Ldlr^−/− mice given chow or HFD for 4 weeks. A and B: Representative images of immunofluorescence (IF) staining (A) and quantification (B) of Dock2 (red) and DAPI (blue, nuclei) in aortic roots in Ldlr^−/− mice fed with chow or HFD for 4 weeks. Arrows indicate Dock2 expression. Data are expressed as means ± SD. n = 4 mice per group. ∗P < 0.05 versus Ldlr^−/− mice given chow. Scale bars = 100 μm.

Supplemental Figure S2

The expression levels of adhesion molecules ICAM-1 and VCAM-1 are increased in lesions of aortic roots in Ldlr^−/− mice given chow or HFD for 4 weeks. Representative images of IHC staining for ICAM-1 and VCAM-1 in aortic roots in Ldlr^−/− mice given chow or HFD for 4 weeks. Scale bars = 50 μm.

Supplemental Figure S3

DOCK2 expression is time-dependently increased with EC phenotypic change. A and B: DOCK2, VCAM-1, and ICAM-1 expression in human umbilical vein ECs (A) or human coronary artery ECs (B) treated with TNF-α (20 ng/mL) for various durations, via Western blot analysis. Glyceraldehyde phosphate dehydrogenase (GAPDH) was used as an internal control.

Supplemental Figure S4

DOCK2 mRNA expression is increased with TNF-α in human umbilical vein (HUV) ECs. A–D:DOCK2 (A), VCAM1 (B), ICAM1 (C), and MCP1 (D) mRNA expression in HUVECs treated with TNF-α (20 ng/mL) for various durations. The cyclophilin gene was used as an internal control. n = 3. ∗P < 0.05 versus vehicle-treated control (0 ng/mL).

Supplemental Figure S5

DOCK2 regulates MCP-1 expression in ECs during atherosclerosis formation. A and B: Inflammation cytokine MCP-1 expression in human umbilical vein (HUV) ECs treated with TNF-α at different doses (A) or various durations (B) as indicated, by Western blot analysis. C: MCP-1 expression in HUVECs transfected with scramble or DOCK2 siRNA for 24 hours and treated with vehicle (−) or TNF-α (20 ng/mL) (+) for 16 hours, by Western blot analysis. D: MCP-1 expression in lesions of aortic roots in Ldlr^−/− and Ldlr^−/−Dock2^−/− mice given HFD for 12 weeks, by IHC analysis. Scale bars = 50 μm.

Supplemental Figure S6

Dock2 knockout suppresses p–NF-κB expression in atherosclerotic lesions in mice. Representative images of p–NF-κB expression in lesions of aortic roots in Ldlr^−/− and Ldlr^−/−Dock2^−/− mice given HFD for 12 weeks, by IHC analysis. n = 3 mice per group. Scale bars = 50 μm. L, lumen.

Supplemental Figure S7

Dock2 knockout decreases macrophage infiltration in atherosclerosis lesions in mice. Representative images of F4/80 in lesions of aortic roots in Ldlr^−/− and Ldlr^−/−Dock2^−/− mice given HFD for 12 weeks, by IHC analysis. n = 3 mice per group. Scale bars = 50 μm. L, lumen.