Modulation Of Dickkopf1 Under Shear Stress And Roles Of Dickkopf1 In The Development Of Atherosclerosis

BackgroundAtherosclerosis (AS) is one of the most common cardiovascular diseases and its pathogenesis has been a research hotspot in the cardiovascular area. Atherosclerotic plaques occur primarily at the curves and bifurcations of arteries. The change of local shear stress is closely related to the specific distribution of AS plaques. In the straight parts of blood vessels, smooth blood flow produces a unidirectional physiological pulsatile shear stress(≥12dyne/cm2), which has a protective effect of inhibiting atherogenesis. Whereas, in the vascular curves and bifurcations, blood flow becomes turbulent and turns into atheroprone oscillatory shear stress with multidirection and decreased velocity (0±4 dyne/cm2). The vascular endothelium, located at the luminal surface of the vessel wall, is directly exposed to blood flow and plays an important role in vascular homeostasis in response to stress. Endothelial dysfunction is a key step in the early development of atherosclerosis.The DKK-related protein family is very conservative in the evolution. It consists of four members:DKK1, DKK2, DKK3 and DKK4, and DKK1 is the most widely studied. DKK1 is a secreted antagonist of the Wnt/β-catenin pathway. It could bind with Wnt membrane receptors, Krumen and LRP5/6, hinder the formation of normal Wnt receptor complexes, promote the phosphorylated degradation of cytoplasmic P-catenin and close the Wnt pathway, which in turn affect the corresponding biological functions.Previous studies found that DKK1 was closely related to the occurrence of cardiovascular disease. The increased expressions of DKK1 in the serum and AS plaques in the patients with coronary heart disease indicated that DKK1 was associated with atherosclerosis. However, the direct effects and underlying mechanisms of DKK1 in terms of atherogenesis are poorly understood. The changes of DKK1 expression under shear stress have also not been reported.Objectives1. To investigate whether endothelial DKK1 expression was modulated by shear stress both in vivo and in vitro;2. To investigate the role of DKK1 in regulating endothelial adhesion and cellular tight junction;3. To explore the role of DKK1 in atherogenesis and mechanisms involved.Methods1. Cell cultureHuman umbilical vein endothelial cells (HUVECs) were isolated from fresh human umbilical cords using trypsin digestion method and cultured in M199 base medium containing 10% fetal bovine serum (FBS),1% penicillin/streptomycin and 2 ng/ml of fibroblast growth factor-2 at 5% CO2 at 37℃. The medium was changed to remove the non-adherent cells next day. When the cells reached confluence, they were propagated to the next passage.2. In vitro shear stress interventionFor the shear stress stimulation, HUVECs from passages 4 to 7 were seeded on glass slides (75 by 25 mm) precoated with rat tail collagen I in a computer-controlled osci-flow apparatus. To investigate the effects of different shear stresses on the expression of DKK1, the cells were exposed to unidirectional pulsatile shear stress (12 dyne/cm2) or oscillatory shear stress (0±4 dyne/cm2). The cells of the control group were maintained in a static state.3. Partial carotid ligation mouse modelTo further confirm the effects of disturbed flow on the expression of DKK1, we used an animal model of partial ligation of the carotid artery to induce disturbed shear stress in arteries in C57BL/6 mice. The protocols were as follows. Anesthesia was induced with 2% isoflurane. Betadine was applied to the depilated area. A 4-6-mm vertical incision was made in the middle of the neck. The left common carotid artery (LCA) was bluntly dissected to expose the following four distal branches:the external carotid artery, the internal carotid artery, the occipital artery, and the superior thyroid artery. Three of the four branches (the left external carotid, the internal carotid, and the occipital artery) were ligated, leaving the superior thyroid artery intact. The incision was then sutured. The mice were placed in a warming chamber during recovery and underwent carotid ultrasonography to determine whether the partial ligation induced disturbed flow at 1 day after surgery.4. En face staining of mouse aortas and common carotid arteriesMice were anesthetized with 0.8%(wt/vol) pentobarbital sodium and then perfused with lukewarm saline and 4% paraformaldehyde transcardially. The aorta or common carotid artery was carefully dissected, removed and fixed in fixative solution, washed with phosphate-buffered saline (PBS) and dissected longitudinally before being permeabilized in 0.5% Triton X-100 PBS solution for 10 min, blocked with 5% bovine serum albumin, and incubated with primary antibodies at 4℃ overnight, then with TRITC-or FITC-conjugated secondary antibody at 37℃ for 1 h. Nuclei were counterstained with 4’,6-diamidino-2-phenylindole (DAPI) for 10 min. Samples were photographed under a laser-scanning confocal microscope.5. Transient Transfection5.1 DKK1 siRNA transfectionDKK1 siRNA and negative control siRNA were all synthesized by Shanghai GenePharma Biotechnology Company. Transfection was in accordance with the instructions for the lip2000. Cell culture medium was changed 6 hours after transfection. Different stimulations were given and corresponding detections were taken 24-48 hours after transfection.5.2 Construction and transfection of DKK1 overexpression lentivirus vectorDKK1 gene overexpression lentivirus vector and empty vector carrying green fluorescent protein GFP were synthesized by Shanghai Genechem Company. Transfection was in accordance with lentivirus operation manual.48 hours after transfection, transfection efficiency was observed under fluorescence microscopy.6. Quantitative real-time PCRTotal RNA was collected from the common carotid artery or HUVECs and reverse transcribed into cDNA by using a Prime Script RT reagent kit (TaKaRa Bio, Japan). qPCR analysis of gene expression involved the use of an SYRB Premix Ex Taq kit (TaKaRa Bio, Japan). The primer sequences for the target genes were synthesized by Shanghai GenePharma Biotechnology Company. Quantification involved the 2-ΔΔCt method.7. Western blotting analysisEqual amounts of protein extracted from tissue and HUVECs were separated on SDS-PAGE, transferred to PVDF membranes, and incubated with primary antibodies overnight at 4℃. The bands were recorded by using the LAS-4000 luminescent image analyzer and analyzed with use of Adobe Photoshop CS5.8. ImmunofluorescenceHUVECs were fixed, permeabilized, blocked, and incubated with primary antibodies at 4℃ overnight, rinsed with PBS, and then incubated with FITC-or TRITC-conjugated antibodies. Nuclei were stained with DAPI. Cells were observed under a laser-scanning confocal microscope.9. In vitro monocyte adhesion assayHuman peripheral blood mononuclear leukocytes (THP-1 cells) were obtained from the American Type Culture Collection and cultured in RPMI 1640 medium with 10% FBS and 1% penicillin/streptomycin and labeled in PBS with 25μM 5(6)-carboxyfluorescein diacetate N-succinimidyl ester (CFSE) for 15 min.Following shearing, the slides seeded with HUVECs were removed from the flow chamber and incubated with a labeled THP1 cell suspension (5×105 cells/ml) for 1 h. Non-adherent cells were washed away. The bound THP-1 cells on the HUVECs were counted in eight randomly selected microscopy fields from each slide under a×20 objective; quantification involved the use of Image-Pro Plus.10. Effect of DKK1 silence on the development of AS10.1 Construction of DKK1 silent lentivirus vectorA recombinant lentivirus vector pGLV3/Hl/GFPtPuro (pGLV3) harboring a short-hairpin RNA sequence targeting DKK1 (pGLV3-shRNA-DKKl) was produced by Shanghai Genechem Company. Empty vectors carrying green fluorescent protein GFP were transfected as negative control.10.2 Partial carotid ligation AS model and histopathologyNinety eight-week-old ApoE-/-mice were fed with high fat diet after two weeks’ accommodation and divided into three groups:(1) Con group (n= 30):intravenous injection of 100ul physiological saline; (2) Scramble group (n=30):intravenous injection of 100ul,5×107ifu lentivirus vector only with GFP; (3) shRNA-DKK1 group (n=30):intravenous injection of 100ul,5×107ifu shRNA-DKK1. Two weeks after transfection, lentivirus transfection efficiency was evaluated by observation of GFP fluorescence intensity and detection of the expression of DKK1 using real-time quantitative PCR and Western Blotting. The remaining mice underwent left common carotid artery ligation to induce pathological shear stress. After 1 week 16 mice of each group were randomly selected. En face staining was used to examine monocyte-endothelial adhesion and endothelial tight junction proteins in vivo. Western Blotting was used to detect the expression of related proteins. Three weeks after ligation, the rest of mice were sacrificed. Sections of the left carotid artery underwent HE and oil red O staining. Plaque area was calculated by use of Image-Pro Plus.10.3 Long-term high fat feeding AS model and histopathologyThirty eight-week-old male ApoE-/-mice were fed with high fat diet during the whole experiment after 2 weeks accommodation and divided into three groups which were similar to the acute AS model. Each group included ten mice. Lentivirus vectors harboring an RNA sequence targeting DKK1 (shRNA-DKK1) were delivered into mice each one month. All the mice were sacrificed after three months. Sections of aortic arches underwent oil red O staining. Plaque area was calculated by use of Image-Pro Plus.11. Statistical analysisData are presented as mean±SEM. SPSS 16.0 was used for statistical analysis. Unpaired t test and one-way ANOVA were used for the analysis. P<0.05 was considered statistically significant.Results1. The expression of endothelial DKK1 in vivo in areas of disturbed flow was higher than that in the physiological shear stress regionsThe vascular endothelium at the lesser curvature of aortic arches in mice mainly underwent atheroprone pathological oscillatory shear stress and low shear stress, whereas endothelium at the straight segments of thoracic aorta was exposed to the atheroprotective physiological pulsatile shear stress. The expression of DKK1 was higher in the inner curvature of the aortic arch than in the thoracic aorta (p<0.05). Partial ligation of the carotid artery model was used to induce disturbed shear stress in arteries in C57BL/6 mice. DKK1 expression was significantly greater in the endothelium of the left common carotid artery of mice with partial ligation than sham-operated mice. Therefore, disturbed flow upregulated the expression of DKK1 in endothelial cells in vivo.2. Pulsatile shear stress decreased DKK1 expression, whereas oscillatory shear stress upregulated DKK1 levels in HUVECsCompared with static controls, DKK1 mRNA and and protein levels in HUVECs were both decreased and peaked at 6 h in the setting of pulsatile shear stress (p<0.05). In response to oscillatory shear stress, DKK1 expression was increased and peaked at 6 h compared with static controls (p<0.05). A unidirectional pulsatile shear stress or oscillatory shear stress was applied to HUVECs for 6 h. Immunofluorescence assay revealed greater DKK1 protein level with oscillatory shear stress and lesser DKK1 protein level with pulsatile shear stress than static conditions.3. DKK1 participated in the modulation of shear stress on the endothelial adhesion with monocytesCompared with static controls, pulsatile shear stress inhibited the adhesion of endothelial cells with monocytes (P<0.05) and decreased the expression of endothelial adhesion molecules VCAM-1, ICAM-1 and E-selectin (P< 0.05). DKK1 overexpression blocked the inhibition and upregulated VCAM-1, ICAM-1 and E-selectin levels (P< 0.05).Adhesion of THP1 cells to endothelial cells was greater under oscillatory shear stress than static conditions (P< 0.05) and the expression of VCAM-1, ICAM-1 and E-selectin in HUVECs were also upregulated by oscillatory shear stress (P< 0.05). siRNA knockdown of DKK1 attenuated the increased adhesion (P< 0.05) and reduced the expression of VCAM-1, ICAM-1 and E-selectin under oscillatory shear stress (P< 0.05).4. DKKl was involved in the regulation of shear stress on the endothelial tight junction proteins expressionCompared with static controls, pulsatile shear stress upregulated endothelial tight junction proteins ZO1, ZO2, occludin and Claudin 5 levels (P< 0.05); overexpression of DKK1, their expressions were decreased (P< 0.05).Oscillatory shear stress suppressed the expression of ZO1, ZO2, occludin and Claudin 5 (P< 0.05); DKK1 knockdown significantly increased their protein levels (P <0.05).5. In vivo gene silencing of DKKl limited atheroma formation in the areas of pathological shear stressTwo weeks after transfection, the expression of GFP was obvious in the common carotid artery of ApoE/-mice. Compared with Con group, DKKl mRNA and protein levels in the carotid artery in shRNA-DKK1 group were both significantly reduced (P<0.05). By three weeks after ligation, atherosclerosis had developed in the left common carotid artery and was inhibited by DKK1 silencing (P<0.05). In the chronic high fat diet model, atherosclerosis development in both the aortic arch and the arterial branches was also significantly inhibited by DKK1 silencing (P<0.05).6. DKK1 silencing attenuated the adhesion of endothelium with monocytes in areas of pathological shear stress in vivo in ApoE-/-miceDKK1 silencing attenuated monocyte cell adhesion to the vascular endothelium and decreased the expression of endothelial adhesion molecules VCAM-1, ICAM-1 and E-selectin in the left common carotid artery which was exposed to pathological shear stress at one week post-ligation in the carotid ligation model (P<0.05).7. DKK1 silencing upregulated the expression of endothelial tight junction proteins in areas of pathological shear stress in vivo in ApoE-/-miceDKK1 silencing alleviated the endothelial tight junction disruption induced by pathological shear stress in the left common carotid artery at one week post-ligation in the partial carotid ligation model (P<0.05). The expression of ZO1, ZO2, occludin and Claudin 5 were also increased after DKK1 silencing (P<0.05).Conclusions1. Pulsatile shear stress decreased whereas oscillatory shear stress increased the expression of DKK1 in endothelial cells.2. DKK1 was involved in the regulation of shear stress on the endothelial adhesion and tight junction proteins expression.3. DKK1 accelerated atheroma formation in the areas of pathological shear stress via promoting endothelial adhesion with monocytes and decreasing tight junction proteins expression.BackgroundAtherosclerotic plaque occurs primarily at the curves, bifurcations and stenosis of arteries, which are subjected to pathological oscillatory shear stress and low shear stress. The vascular endothelium, located at the luminal surface of the vessel wall, is directly exposed to blood flow. It could perceive the change of local microenvironment and transmit chemical signals into cell, which then influences gene expression. Endothelial dysfunction is an essential process in the early occurrence of atherosclerosis. The surface of endothelium, including lumen surface, intercellular junction and basal surface, contains numerous mechanical receptors, which can activate several intracellular signal transduction pathways after activation, affect the combination of various nuclear transcription factors with shear stress responsive elements (SSREs), and thus influence corresponding cell functions. At the regions where oscillatory shear stress and low shear stress occur, such as bifurcations of arteries, the expression of anti-atherosclerotic genes are suppressed and atheroprone genes are upregulated, which promotes the development and progression of atherosclerosis.Clinical researches have reported that DKK1 is tightly connected with atherosclerosis. Our advanced study demonstrated that pathological oscillatory shear stress upregulated the expression of DKK1 in endothelium, which then accelerated endothelial adhesion with monocytes and decreased the expression of intercellular tight junction molecules, thus promoting atherogenesis in disturbed flow areas. However the mechanism involved in the regulation of DKK1 under oscillatory shear stress still remains unknown.Except as an inhibitor of canonical Wnt/β-catenin, DKK1 is also the target gene of p-catenin. β-catenin could combine with TCF/LEF nuclear transcription factor and then regulate the expression of DKK1. Protease activated receptor 1 (PAR1), a seven-transmembrane G-protein-coupled receptor, could participate in the regulation of endothelial inflammation and permeability. It was reported that PAR1 activated CREB in melanoma cells, upregulated the expression of adhesion molecular MCAM/MUC18 and in further promoted the metastasis of melanoma. However, whether PAR 1/CREB pathway and β-catenin mediated the upregulation of DKK1 by oscillatory shear stress in endothelial cells has not been reported.Recently, the study of long non-coding RNA (LncRNA) has become a novel research hotspot. LncRNA plays important roles in the regulation of diverse cell functions, including cell differentiation, epigenetic modification and cell apoptosis. A recent study reported that LncRNA NBATI affected the metastasis of breast cancer by regulating DKK1 expression in breast cancer cells. Whether NBAT1 was involved in the regulation of endothelial DKK1 expression under oscillatory shear stress still need further exploration.Objectives1. To investigate the role β-catenin played in the regulation of oscillatory shear stress-induced DKK1 expression in endothelial cells;2. To investigate the role of PAR1/CREB pathway in the upregulation of DKK1 by oscillatory shear stress in endothelial cells;3. To explore the role of LncRNA NBAT1 played in the upregulation of DKK1 by oscillatory shear stress in endothelial cells and its relationship with β-catenin and PAR1/CREB pathway.Methods1. Cell cultureHuman umbilical vein endothelial cells (HUVECs) were isolated from fresh human umbilical cords using trypsin digestion method and cultured in M199 base medium containing 10% fetal bovine serum (FBS),1% penicillin/streptomycin, and 2 ng/ml of fibroblast growth factor-2 at 5% CO2 at 37℃. The medium was changed to remove the non-adherent cells next day. When the cells reached confluence, they were propagated to the next passage.2. In vitro shear stress interventionFor the shear stress stimulation, HUVECs from passages 4 to 7 were seeded on glass slides (75 by 25 mm) precoated with rat tail collagen I in a computer-controlled osci-flow apparatus. To investigate the effects of oscillatory shear stress on the expression of β-catenin, PAR1, CREB and NBAT1, the cells were exposed to oscillatory shear stress (0±4 dyne/cm2) with different time. The cells of the control group were maintained in a static state.3. DKK1 siRNA transfectionsiRNA of β-catenin, PAR1, CREB, NBAT1 and negative control were all synthesized by Shanghai GenePharma Biotechnology Company. Transfection was in accordance with the instructions for the lip2000. Cell culture medium was changed 6 hours after transfection. Different stimulations were given and corresponding detections were taken 24-48 hours after transfection.4. Quantitative real-time PCRTotal RNA collected from HUVECs was reverse transcribed into cDNA by using a Prime Script RT reagent kit (TaKaRa Bio, Japan). qPCR analysis of gene expression involved the use of an SYRB Premix Ex Taq kit (TaKaRa Bio, Japan). The primer sequences for the target genes were synthesized by Shanghai GenePharma Biotechnology Company.5. Western blotting analysisEqual amounts of protein extracted from HUVECs were separated on SDS-PAGE, transferred to PVDF membranes, and incubated with primary antibodies overnight at 4℃. The bands were recorded by using the LAS-4000 luminescent image analyzer and analyzed with use of Adobe Photoshop CS5.6. Fluorescence in situ hybridization (FISH)Experiments were operated according to the protocol provided in the kit. The FISH probe of LncRNA NBAT1 and corresponding kit were all purchased from RIOBIO Biocompany (Guangzhou). After fixed with 4% paraformaldehyde and penetrated with 0.5% Triton X-100 PBS, HUVECs were incubated with NBAT1 FISH probe overnight. Scour off uncombined probes, stain nucleus and seal slide for fluorescent detection the next day.7. Statistical analysisData are presented as mean±SEM. SPSS 16.0 was used for statistical analysis. Unpaired t test and one-way ANOVA were used for the analysis. P<0.05 was considered statistically significant.Results1. β-catenin participated in the upregulation of DKK1 in HUVECs by oscillatory shear stressHUVECs were stimulated by oscillatory shear stress for different time. Western Blotting showed that oscillatory shear stress upregulated P-catenin expression in HUVECs, compared with static control group (P<0.05). siRNA transfection downregulated the expression of β-catenin, the upregulation of DKK1 by oscillatory shear stress was inhibited as well (P<0.05). The results indicated that β-catenin was involved in the regulation of DKK1 expression under oscillatory shear stress in HUVECs.2. PAR1/CREB pathway was involved in the upregulation of DKK1 by oscillatory shear stress in HUVECsAfter exposed to the stimulation of oscillatory shear stress for different time, compared with static control, the expression of PAR1,β-CREB (Ser 133) and CREB in HUVECs were significantly increased (P<0.05). Knockdown of PAR1 or CREB with corresponding siRNAs, the upregulation of DKK1 by oscillatory shear stress was all inhibited (P<0.05). Meanwhile, the expression of p-CREB and CREB were also decreased after the application of PAR1 siRNA or inhibitor SCH79797, compared with negative control (P<0.05). Therefore, in the setting of oscillatory shear stress, PAR1 modulated DKK1 expression via regulating the expression and phosphorylation of CREB.3. NBAT1 participated in the upregulation of DKK1 under oscillatory shear stress in HUVECsTo examine the regulation of oscillatory shear stress on NBAT1 expression, qPCR and FISH were used. Results showed that NBAT1 mainly expressed in the nucleus of HUVECs and compared with static control, oscillatory shear stress increased NBAT1 expression significantly (P<0.05). siRNA knockdown of NBAT1, the expression of DKK1 decreased as well, compared with negative control (P<0.05), indicating the participation of NBAT1 in the oscillatory shear stress-induced DKK1 expression.4. Regulation of NBAT1 on DKK1 expression under oscillatory shear stress in HUVECs was not via the affection of p-catenin expressionTo explore whether NBAT1 modulated DKK1 expression via affecting the expression of β-catenin under oscillatory shear stress, Western Blotting examination of β-catenin expression was done after NBAT1 knockdown. Results showed that NBAT1 interference had no significant effects on the expression of β-catenin under oscillatory shear stress, indicating that the regulation of NBAT1 on DKK1 level was not via the affection of β-catenin expression.5. PAR1/CREB pathway was involved in the NBAT1 modulated DKK1 upregulation under oscillatory shear stress in HUVECsTo further explore the role of PAR1/CREB pathway in the NBAT1 modulated DKK1 upregulation under oscillatory shear stress, we examined the expression of PAR1, p-CREB and CREB after NBAT1 siRNA transfection. Results showed that the upregulation of PAR1,β-CREB and CREB were all significantly inhibited after NBAT1 siRNA transfection (P<0.05), which indicated that NBAT1 could participate in the regulation of DKK1 in HUVECs under oscillatory shear stress via activating PAR1/CREB pathway.Conclusions1.P-catenin was involved in the upregulation of DKK1 by oscillatory shear stress in HUVECs.2. NBAT1 mediated the upregulation of DKK1 in HUVECs by oscillatory shear stress via activating PAR1/CREB pathway.