Evaluation of Early Biomarkers of Atherosclerosis Associated with Polychlorinated Biphenyl Exposure: An in Vitro and in Vivo Study

Background: Miscellaneous cardiovascular risk factors have been defined, but the contribution of environmental pollutants exposure on cardiovascular disease (CVD) remains underappreciated. Objective: We investigated the potential impact of typical environmental pollutant exposure on atherogenesis and its underlying mechanisms. Methods: We used human umbilical vein endothelial cells (HUVECs) and apolipoprotein E knockout (ApoE−/−) mice to investigate how 2,3,5-trichloro-6-phenyl-[1,4]-benzoquinone (PCB29-pQ, a toxic polychlorinated biphenyl metabolite) affects atherogenesis and identified early biomarkers of CVD associated with PCB29-pQ exposures. Then, we used long noncoding RNAs (lncRNAs) HDAC7-AS1–overexpressing ApoE−/− mice and apolipoprotein E/caveolin 1 double-knockout (ApoE−/−/CAV1−/−) mice to address the role of these early biomarkers in PCB29-pQ–induced atherogenesis. Plasma samples from patients with coronary heart disease (CHD) were also used to confirm our findings. Results: Our data indicate that lncRNA HDAC7-AS1 bound to MIR-7-5p via argonaute 2 in PCB29-pQ–challenged HUVECs. Our mRNA sequencing assay identified transforming growth factor-β2 (TGF-β2) as a possible target gene of MIR-7-5p; HDAC7-AS1 sponged MIR-7-5p and inhibited the binding of TGF-β2 to MIR-7-5p. The effect of PCB29-pQ–induced endothelial injury, vascular inflammation, development of plaques, and atherogenesis in ApoE−/− mice was greater with MIR-7-5p–mediated TGF-β2 inhibition, whereas HDAC7-AS1–overexpressing ApoE−/− mice and ApoE−/−/CAV1−/− mice showed the opposite effect. Consistently, plasma levels of HDAC7-AS1 and MIR-7-5p were found to be significantly associated individuals diagnosed with CHD. Discussions: These findings demonstrated that a mechanism-based, integrated-omics approach enabled the identification of potentially clinically relevant diagnostic indicators and therapeutic targets of CHD mediated by environmental contaminants using in vitro and in vivo models of HUVECs and ApoE−/− and ApoE−/−/CAV1−/− mice. https://doi.org/10.1289/EHP9833


Table of Contents
Figure S1. Functional analyses of genes targeted by differentially expressed miRNAs. Highthroughput sequencing analysis of differentially expressed miRNAs after 5 μM PCB29-pQ treatment of HUVECs for 24 h (n = 3). Target gene of differentially expressed genes were screened and functional analyses performed. Gene ontology has three ontologies: molecular function, cellular component, and biological process. The exact data are presented in Excel Table  S3.

Figure S2. KEGG pathway analyses of genes targeted by differentially expressed miRNAs.
High-throughput sequencing analysis of differentially expressed miRNAs in HUVECs with 5 μM PCB29-pQ treatment for 24 h (n = 3). KEGG pathway enrichment in miRNA, the rich factor represents the degree of enrichment. The node size shows the number of selected genes, and the color scale represents -log (P value). The exact data are presented in Excel Table S4. Figure S3. The expression of HDAC7-AS1, MIR-7-5p and Ago2 in HUVECs exposed to PCB29-pQ. RT-qPCR was performed to validate (a) lncRNAs and (b) miRNAs identified as differentially expressed by RNA sequencing assay. β-Actin was used as housekeeping genes for lncRNAs quantification and U6 was used as housekeeping genes for miRNAs quantification, respectively. RP5-1057120.4 (termed as HDAC7-AS1) and MIR-7-5p were selected as target lncRNA and miRNA for further investigation (indicated with red boxes). Data are presented as mean ± SD (n = 3). (c) Anti-Ago2 RNA binding protein immunoprecipitation assay was used to pull down endogenous RNAs associated with Ago2; IgG served as the control. Ago2 in proteins from Ago2-RIP assay was measured by western blotting. Data are presented as mean ± SD (n = 3). P values were determined by two-way ANOVA, followed by Tukey's post hoc test. Data are graphed relative respective vehicle controls. The exact mean and SD value are presented in Table  S20. Figure S4. The expression of HDAC7 level in HUVECs transfected with HDAC7-AS1 siRNA or pEZ-M61-HDAC7-AS1. (a) Predicted relationship between ENST00000080059 (HDAC7) and ENST00000599515 (HDAC7-AS1) via RNAplex (http://www.tbi.univie.ac.at/RNA/RNAplex.1.html) (b) After transfection with NC siRNA or HDAC7-AS1 siRNA (25, 50, or 100 nM) for 48 h, the silencing efficiency of HDAC7-AS1 siRNA in HUVECs was determined by RT-qPCR. Data are presented as mean ± SD (n = 3). (c) After transfection with 2.5 μg pEZ-M61-NC or pEZ-M61-HDAC7-AS1 (2.5, 5, or 7.5 μg) for 6 h, HUVECs were treated with 5 μM PCB29-pQ. HDAC7-AS1 level was determined by RT-qPCR. Data are presented as mean ± SD (n = 3). HDAC7 expression of PCB29-pQ-exposed cells or control after (d) transfection with 25 nM NC siRNA or 25 nM HDAC7-AS1 siRNA for 24 h or (e) transfection with 2.5 μg pEZ-M61-NC or 2.5 μg pEZ-M61-HDAC7-AS1 for 6 h. Data are presented as mean ± SD (n = 3). The housekeeping gene β-actin was used for quantification, and the primer information is shown Table S3. P values were determined by one-way ANOVA, followed by Tukey's post hoc test. For b and c, expression is shown relative to NC siRNA control and pEZ-M61-NC controls, respectively. The exact mean and SD values are presented in Table  S21. Figure S5. The localization of LncRNA HDAC7-AS1 in HUVECs. HUVECs were treated with 5 μM PCB29-pQ for 24 h. (n = 3). Fluorescent in situ hybridization (FISH) assay was used to investigate the localization of HDAC7-AS1 in in the cells. HDAC7-AS1 was stained red using an RNA probe, and the nucleus was stained blue with DAPI. Scale bar = 10 μm. Figure S6. Volcano plots analyzing differential expression with mRNA sequencing (mRNA-Seq) in the control group and PCB29-pQ group. The abscissa represents the logarithmic values of two different groups, and the ordinate represents mRNAs differences (fold change≥2 and P value＜0.05) between two groups. Red dots indicated PCB29-pQ group is higher relative to control group). Green dots indicated lower expression in PCB29-pQ group than control group (judgment standard is P value <0.05, and the difference multiple is more than 2). Black dots indicted no difference between two groups. Figure S7. Target genes TGF-β2 and PPME1 protein levels in HUVECs exposed to PCB29-pQ with MIR-7-5p inhibitor or TGF-β2/PPME1 siRNA. After co-transfection with 100 nM NC inhibitor or MIR-7-5p inhibitor and NC siRNA or TGF-β2/PPME1 siRNA (25 nM) for 24 h, HUVECs were treated with 5 μM PCB29-pQ for 24 h. (n = 3). Protein levels of (a) PPME1 and (b) TGF-β2 were detected by western blotting (upper panel). The relative protein expression levels (lower panel) were quantified by ImageJ software. Data are presented as mean ± SD (n = 3). β-Actin was used as an internal loading control. Data are graphed relative to the expression in cells exposed to the NC inhibitor and NC siRNA together. The exact mean and SD values are presented in Table S22. Figure S8. Luciferase analysis of the activity of MIR-7-5p bind to TGF-β2, PPME1, and HDAC7-AS1 in HUVECs transfected with NC mimic or MIR-7-5p mimic. (a) HUVECs were transfected with NC mimic or MIR-7-5p mimic for 48 h, together with NC-3'UTR, TGF-β2 (WT)-3'UTR, or TGF-β2 (MUT)-3'UTR. (b) HUVECs were transfected with 50 nM NC mimic or 50 nM MIR-7-5p mimic for 48 h, together with NC-3'UTR, PPME1 (WT)-3'UTR, or PPME1 (MUT)-3'UTR. (c) HUVECs were transfected with 50 nM NC mimic or 50 nM MIR-7-5p mimic for 48 h, together with Luc-NC, Luc-HDAC7-AS1-WT, or Luc-HDAC7-AS1-MUT. For all panels, Gaussia luciferase activity was analyzed, corresponding to TGF-β2, PPME1, and HDAC7-AS1 transcription. Data are presented as mean ± SD (n = 3). P values were determined by two-way ANOVA, followed by Tukey's post hoc test. Data are graphed relative to the NC-mimic treated, NC-3'UTR controls. The exact mean and SD values are presented in Table S23. Figure S9. The tube forming ability in HUVECs exposed to PCB29-pQ and transfected with a MIR-7-5p inhibitor. (Left panel) After transfection with 100 nM NC inhibitor or 100 nM MIR-7-5p inhibitor for 48 h, followed by exposure to 5 μM PCB29-pQ for 24 h, cells were plated on Matrigel to conduct tube formation assay. Scale bar = 200 μm. (Right panel) Quantification of tube formation through measurement of branch point number with ImageJ software. Data were presented as mean ± SD (n = 3). P values were determined by two-way ANOVA, followed by Tukey's post hoc test. The exact mean and SD values are presented in Table S24. Figure S11. HDAC7-AS1, MIR-7-5p, TGF-β2, PPME1 mRNA levels, TGF-β2 and PPME1 protein levels, and apoptosis or proliferation rates in HUVECs exposed to PCB29-pQ and transfected with a MIR-7-5p inhibitor or HDAC7-AS1 siRNA. After co-transfection with 100 nM NC inhibitor or 100 nM MIR-7-5p inhibitor and NC siRNA or HDAC7-AS1 siRNA (25 nM) for 24 h, HUVECs were treated with 5 μM PCB29-pQ for 24 h. RNAs expression of (a) MIR-7-5p, (b) HDAC7-AS1, (c) TGF-β2 and (d) PPME1 were detected by RT-qPCR. β-actin was used as a housekeeping gene, except U6 was used for MIR-7-5p quantification. The primer information is shown in Table S3-4. (e) Protein levels of TGF-β2 and PPME1 were detected by western blotting. β-Actin was used as an internal loading control. (f) TGF-β2, (g) PPME1 protein expression levels were quantified by ImageJ software. Data were presented as mean ± Standard Deviation (SD) (n = 3). (h) Cell viability was measured by CCK-8 kit. Data were presented as mean ± SD (n = 3). (i) Cell apoptosis was performed by Annexin V-FITC/PI assay with a flow cytometer. (j) Cell proliferation was performed by BrdU/PI staining. P values were determined by one-way ANOVA, followed by Tukey's post hoc test. Data are graphed relative to the cells exposed to NC inhibitor, NC siRNA, and vehicle control. The exact mean and SD values are presented in Table S25. Male ApoE -/mice were i.v. injected with AAV-HDAC7-AS1 via tail vein (4×10 10 particles/mouse) to create an HDAC7-AS1 overexpressed mice model. Male ApoE -/mice that received the AAV vector were considered as AAV control mice. Control and HDAC7-AS1 overexpressed ApoE -/mice were fed a Western high-fat diet for 12 weeks and tap water ad libitum. Mice were injected with 5 mg/kg body weight of PCB29-pQ or equal volumes of corn oil by i.p. injection once a week for 12 continuous weeks with the first two injections during the 1 st week (3 days apart). (Left panel) The presence of p-CAV1 in aortic root cross-sections was detected by double immunostaining with the use of antibodies against p-CAV1 (red), endothelial cell marker CD31 (green), and nucleus marker DAPI (blue), respectively. Scale bar = 100 μm. White arrows represent the colocalization of p-CAV1 (red) and CD31 (green). (Right panel) Colocalization of p-CAV1 and CD31 was analyzed by Pearson's correlation coefficient. Data were presented as mean ± SD (n = 3). The exact mean and SD values are presented in Table S28.  Table S2.

Figure S16. Immunohistochemistry staining of CD68 and TC and LDL-C levels in ApoE -/-
and ApoE -/-/CAV1 -/mice. ApoE -/mice were crossed with CAV1 -/mice to generate ApoE -/-/CAV1 -/mice. ApoE -/or ApoE -/-/CAV1 -/mice were fed a Western high-fat diet for 12 weeks and tap water ad libitum. Mice were treated with 5 mg/kg body weight of PCB29-pQ or equal volumes of corn oil by i.p. injection. (a) (Left panel) Immunohistochemistry staining of CD68 detected macrophage infiltration located at the aortic wall. Scale bar = 200 μm. (Right panel) Quantification proportion of positive CD68 expressions area to total aortic section using was performed by ImageJ software. Data are presented as mean ± SD (n = 3). Serum (b) TC level and (c) LDL-C levels. Data are presented as mean ±SD (n = 5). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. The exact mean and SD values are presented in Table S29. Figure S17. Adhesion molecules, pro-inflammatory cytokines and p65 protein expression levels in HUVEC exposed to PCB29-pQ. HUVECs were treated with 5 μM PCB29-pQ for 1, 3, or 6 h. (a) IL-1β, IL-6, TNF-α, ICAM-1, and VCAM-1 expressions in cell lysates were analyzed by western blotting. (b-f) IL-1β, IL-6, TNF-α, ICAM-1, and VCAM-1 expressions were quantified by ImageJ software. Data are presented as mean ± Standard Deviation (SD) (n = 3). (g) Calcein-AM-loaded THP-1 cells (10 6 cells/ml) were added to HUVECs and then incubated for 3 h. The unbound cells were washed off, and attached fluorescent monocytes were visualized using an optical microscope. (Left panel) The fluorescence intensity of Calcein-AM was quantified using ImageJ software. Data are presented as mean ± SD (n = 3). P value was determined by unpaired Student's t-test. (h) HUVECs were treated with 5 μM PCB29-pQ for 1, 3, or 6 h. IκBα and p-p65 levels were analyzed by western blotting analysis. β-Actin was used as an internal loading control. (i-j) IκBα and p-p65 levels were quantified by ImageJ software. Data are presented as mean ± SD (n = 3). (k) HUVECs were pretreated with 5 μM p65 inhibitor PDTC for 1 h, followed with 5 μM PCB29-pQ exposure for 6 h. IL-1β, IL-6, and TNF-α in cell lysates were analyzed by western blotting. β-Actin was used as an internal loading control. (l-n) IL-1β, IL-6, and TNF-α protein expression levels were quantified by ImageJ software. Data were presented as mean ± SD (n = 3). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. The exact mean and SD values are presented in Table S30. Figure S18. Adhesion molecules, pro-inflammatory cytokines and p65 protein expression levels in HUVEC exposed to PCB29-pQ with CAV1 siRNA. HUVECs were transfected with NC siRNA or 25 nM CAV1 siRNA for 48 h and then treated with 5 μM PCB29-pQ for 6 h. (a) IL-1β, IL-6, TNF-α, ICAM-1 and VCAM-1 expressions were analyzed by western blotting. β-Actin was used as an internal loading control. (b-f) IL-1β, IL-6, TNF-α, ICAM-1 and VCAM-1 expression levels were quantified by ImageJ software. Data were presented as mean ± Standard Deviation (SD) (n = 3). (g) IκBα, p-p65, and p65 expressions were analyzed by western blotting. β-Actin was used as an internal loading control. (h-i) IκBα, p-p65, and p65 expression levels were quantified by ImageJ software. Data were presented as mean ± SD (n = 3). (j) (Left panel) Immunofluorescence analysis of p65. Green staining represents the location of p65. Nuclei shown in blue were stained with DAPI. Scale bar = 10 μm. (Right panel) Co-localization of p-CAV1 and DAPI was analyzed by Pearson's correlation coefficient. Data were presented as mean ± SD (n = 3). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. The exact mean and SD values are presented in Table S31. Figure S19. Inflammatory factors and ROS levels in HUVEC exposed to PCB29-pQ. HUVECs were pretreated with 40 μM VC, 20 μM VE or 5 mM NAC for 1 h, followed with 5 μM PCB29-pQ exposure for 6 h. (n = 3). (a) p-p65, p65, IL-1β, IL-6, TNFα, ICAM-1 and VCAM-1 in cell lysates were analyzed by western blotting. (b-g) p-p65, p65, IL-1β, IL-6, TNFα, ICAM-1 and VCAM-1 expression levels were quantified by ImageJ software. Data were presented as mean ± Standard Deviation (SD) (n = 3). (h) ROS levels were detected by DCFH-DA (10 μM) probe. Data were presented as mean ± SD (n = 3). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. In all graphs, data is normalized to the control cell groups. The exact mean and SD values are presented in Table S32. Figure S20. p-CAV1 and CAV1 levels in HUVEC exposed to PCB29-pQ and antioxidants. HUVECs were pretreated with 40 μM VC, 20 μM VE or 5 mM NAC for 1 h, followed by 5 μM PCB29-pQ exposure for 6 h. HUVECs were pretreated with (a) 40 μM VC, 20 μM VE or 5 mM NAC, (b) 200 U/ml PEG-SOD, 500 U/ml PEG-CAT and 5 mM GSH-MEE for 1 h, followed with 5 μM PCB29-pQ exposure for 6 h. (Left panel) p-CAV1 and CAV1 in cell lysates were analyzed by western blotting. β-Actin was used as an internal loading control. (Right panel) p-CAV1 and CAV1 expression levels were quantified by ImageJ software. Data were presented as mean ± Standard Deviation (SD) (n = 3). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. All data is graphed relative to the control cell groups. The exact mean and SD values are presented in Table S33. Figure S21. Heatmap showing correlation scores between factors. Plasma RNA was extracted by TRNzol universal reagent. CHD group (n = 77) contains patients with > 50% coronary artery stenosis, and the control group (n = 50) contains subjects with < 50% coronary artery stenosis. RT-qPCR analysis of HDAC7-AS1, PPME1,IL6 and TNFα expressions. Spearman's rank correlation coefficient was used to assess the correlation between the two indicated factors. P value of < 0.05 was considered significant. The size and color of circle represents the correlation between two factors. X's mean no significance. Summary data can be found in Table S11.
Table S1. Sources of antibodies using in this study.  Table S10. Negatively correlated miRNA-gene pairs: MIR-7-5p and its target genes. Table S11. Results of correlation analysis between two indicated factors. Table S12. Univariable and multivariable logistic regression analysis for determining the independent risk factors of CHD. Table S13. High-throughput lncRNA and miRNA sequence analysis and relevant qPCR results from HUVECs treated with 5 μM PCB29-pQ. Table S14. The expression of HDAC7-AS1, MIR-7-5p and TGF-β2/PPME1 in HUVECs exposed to 5 μM PCB29-pQ. Table S15. Endothelial injury and atherogenesis in ApoE -/mice exposed to PCB29-pQ, and apoptotic rate in HUVECs exposed to PCB29-pQ. Table S16. Measures of endothelial injury in cells and protein and mRNA expression of TGF-β2 and PPME1 in HUVECs transfected with a HDAC7-AS1 overexpression vector or a MIR-7-5p mimic exposed to 5 μM PCB29-pQ. Table S17. Atherosclerosis and inflammation in ApoE-/-mice exposed to PCB29-pQ accompanied by AAV-HDAC7-AS1 treatment. Table S18. Inflammation and atherogenesis in ApoE -/and ApoE -/-CAV1 -/mice exposed to PCB29-pQ. Table S19. Plasma HDAC7-AS1 (a), MIR-7-5p (b), TGF-β2 (c), PPME1 (d), IL-1β (e), IL-6 (f), and TNFα (g) levels in patients with CHD. Table S20. The expression of select lncRNAs in HUVECs exposed to PCB29-pQ.  Table S22. Protein TGF-β2 and PPME1 levels in HUVECs exposed to PCB29-pQ with MIR-7-5p inhibitor or TGF-β2/PPME1 siRNA. Table S23. Luciferase analysis of the activity of MIR-7-5p bind to TGF-β2, PPME1, and HDAC7-AS1 in HUVECs transfected with NC mimic or MIR-7-5p mimic. Table S24. The tube forming ability in HUVECs exposed to PCB29-pQ and transfected with a MIR-7-5p inhibitor. Table S25. HDAC7-AS1, MIR-7-5p, TGF-β2, PPME1 levels and apoptosis or proliferation rates in HUVECs exposed to PCB29-pQ and transfected with a MIR-7-5p inhibitor or HDAC7-AS1 siRNA. Table S26. HE staining of aortic root cross-sections and total cholesterol (TC) and lowdensity lipoprotein (LDL) levels in ApoE -/mice were i.v. injected with AAV-HDAC7-AS1. Table S27. TGF-β2, PPME1, and inflammatory factors level in HUVECs exposed to PCB29-pQ with or without CAV1 siRNA. Table S28. Pearson's correlation coefficient for expression of p-CAV1 and CD31 in ApoE -/mice that were i.v. injected with AAV-HDAC7-AS1. Table S29. Immunohistochemistry staining of CD68, total cholesterol (TC), and low-density lipoprotein level (LDL) in ApoE -/and ApoE -/-/CAV1 -/mice. Table S30. Adhesion molecules, pro-inflammatory cytokines and p65 protein expression level in HUVEC exposed to PCB29-pQ. Table S31. Adhesion molecules, pro-inflammatory cytokines and p65 protein expression levels in HUVEC exposed to PCB29-pQ with CAV1 siRNA. Table S32. Protein expression of inflammatory factors and ROS levels in HUVEC exposed to PCB29-pQ. Table S33. p-CAV1 protein expression relative to CAV1 total protein in HUVEC exposed to PCB29-pQ and antioxidants.   Excel Table S4. Figure S3. The expression of HDAC7-AS1, MIR-7-5p and Ago2 in HUVECs exposed to PCB29-pQ. RT-qPCR was performed to validate (a) lncRNAs and (b) miRNAs identified as differentially expressed by RNA sequencing assay. β-Actin was used as housekeeping genes for lncRNAs quantification and U6 was used as housekeeping genes for miRNAs quantification, respectively. RP5-1057120.4 (termed as HDAC7-AS1) and MIR-7-5p were selected as target lncRNA and miRNA for further investigation (indicated with red boxes). Data are presented as mean ± SD (n = 3). (c) Anti-Ago2 RNA binding protein immunoprecipitation assay was used to pull down endogenous RNAs associated with Ago2; IgG served as the control. Ago2 in proteins from Ago2-RIP assay was measured by western blotting. Data are presented as mean ± SD (n = 3). P values were determined by two-way ANOVA, followed by Tukey's post hoc test. Data are graphed relative respective vehicle controls. The exact mean and SD value are presented in Table S20. presented as mean ± SD (n = 3). (c) After transfection with 2.5 μg pEZ-M61-NC or pEZ-M61-HDAC7-AS1 (2.5, 5, or 7.5 μg) for 6 h, HUVECs were treated with 5 μM PCB29-pQ. HDAC7-AS1 level was determined by RT-qPCR. Data are presented as mean ± SD (n = 3). HDAC7 expression of PCB29-pQexposed cells or control after (d) transfection with 25 nM NC siRNA or 25 nM HDAC7-AS1 siRNA for 24 h or (e) transfection with 2.5 μg pEZ-M61-NC or 2.5 μg pEZ-M61-HDAC7-AS1 for 6 h. Data are presented as mean ± SD (n = 3). The housekeeping gene β-actin was used for quantification, and the primer information is shown Table S3. P values were determined by one-way ANOVA, followed by Tukey's post hoc test. For b and c, expression is shown relative to NC siRNA control and pEZ-M61-NC controls, respectively. The exact mean and SD values are presented in Table S21.  in the control group and PCB29-pQ group. The abscissa represents the logarithmic values of two different groups, and the ordinate represents mRNAs differences (fold change≥2 and P value＜0.05) between two groups. Red dots indicated PCB29-pQ group is higher relative to control group). Green dots indicated lower expression in PCB29-pQ group than control group (judgment standard is P value <0.05, and the difference multiple is more than 2). Black dots indicted no difference between two groups. Figure S7. Target genes TGF-β2 and PPME1 protein levels in HUVECs exposed to PCB29-pQ with MIR-7-5p inhibitor or TGF-β2/PPME1 siRNA. After co-transfection with 100 nM NC inhibitor or MIR-7-5p inhibitor and NC siRNA or TGF-β2/PPME1 siRNA (25 nM) for 24 h, HUVECs were treated with 5 μM PCB29-pQ for 24 h. (n = 3). Protein levels of (a) PPME1 and (b) TGF-β2 were detected by western blotting (upper panel). The relative protein expression levels (lower panel) were quantified by ImageJ software. Data are presented as mean ± SD (n = 3). β-Actin was used as an internal loading control. Data are graphed relative to the expression in cells exposed to the NC inhibitor and NC siRNA together. The exact mean and SD values are presented in Table S22.  Table S23.  Table S24. Figure S11. HDAC7-AS1, MIR-7-5p, TGF-β2, PPME1 mRNA levels, TGF-β2 and PPME1 protein levels, and apoptosis or proliferation rates in HUVECs exposed to PCB29-pQ and transfected with a MIR-7-5p inhibitor or HDAC7-AS1 siRNA. After co-transfection with 100 nM NC inhibitor or 100 nM MIR-7-5p inhibitor and NC siRNA or HDAC7-AS1 siRNA (25 nM) for 24 h, HUVECs were treated with 5 μM PCB29-pQ for 24 h. RNAs expression of (a) MIR-7-5p, (b) HDAC7-AS1, (c) TGF-β2 and (d) PPME1 were detected by RT-qPCR. β-actin was used as a housekeeping gene, except U6 was used for MIR-7-5p quantification. The primer information is shown in Table S3-4. (e) Protein levels of TGF-β2 and PPME1 were detected by western blotting. β-Actin was used as an internal loading control. (f) TGF-β2, (g) PPME1 protein expression levels were quantified by ImageJ software.

Additional File-Excel Document
Data were presented as mean ± Standard Deviation (SD) (n = 3). (h) Cell viability was measured by CCK-8 kit. Data were presented as mean ± SD (n = 3). (i) Cell apoptosis was performed by Annexin V-FITC/PI assay with a flow cytometer. (j) Cell proliferation was performed by BrdU/PI staining. P values were determined by one-way ANOVA, followed by Tukey's post hoc test. Data are graphed relative to the cells exposed to NC inhibitor, NC siRNA, and vehicle control. The exact mean and SD values are presented in Table S25.  Table S26. Figure S13. CAV1 phosphorylation and TGF-β, PPME1, and inflammatory factor levels in HUVECs exposed PCB29-pQ with CAV1 siRNA. (a) HUVECs were treated with 5 μM PCB29-pQ for 1, 3, and 6 h. (upper panel) Protein levels of p-CAV1 and CAV1 were detected by western blotting.
β-Actin was used as an internal loading control. The relative protein expression levels (lower panel) were quantified by ImageJ software. Data are presented as mean ± SD (n = 3). After transfection with 25 nM NC siRNA or CAV1 siRNA for 48 h, HUVECs were treated with 5 μM PCB29-pQ for 24 h.
The siRNA information is shown in Table S6. (b) Protein levels of p-CAV1, CAV1, IL-1β, IL-6, and TNFα were detected by western blotting. β-Actin was used as an internal loading control. (c-h) p-CAV1, CAV1, IL-1β, IL-6, and TNFα were quantified by ImageJ software. Data are presented as mean ± SD (n = 3). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. Data are graphed relative to the NC siRNA group. The exact mean and SD values are presented in Table   S27.  Table S28.  Table S2.
Mice were treated with 5 mg/kg body weight of PCB29-pQ or equal volumes of corn oil by i.p.  presented as mean ± SD (n = 3). P value was determined by unpaired Student's t-test. (h) HUVECs were treated with 5 μM PCB29-pQ for 1, 3, or 6 h. IκBα and p-p65 levels were analyzed by western blotting analysis. β-Actin was used as an internal loading control. (i-j) IκBα and p-p65 levels were quantified by ImageJ software. Data are presented as mean ± SD (n = 3). (k) HUVECs were pretreated with 5 μM p65 inhibitor PDTC for 1 h, followed with 5 μM PCB29-pQ exposure for 6 h. IL-1β, IL-6, and TNF-α in cell lysates were analyzed by western blotting. β-Actin was used as an internal loading control. (l-n) IL-1β, IL-6, and TNF-α protein expression levels were quantified by ImageJ software.
Data were presented as mean ± SD (n = 3). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. The exact mean and SD values are presented in Table S30. Figure S18. Adhesion molecules, pro-inflammatory cytokines and p65 protein expression levels in HUVEC exposed to PCB29-pQ with CAV1 siRNA. HUVECs were transfected with NC siRNA or 25 nM CAV1 siRNA for 48 h and then treated with 5 μM PCB29-pQ for 6 h. (a) IL-1β, IL-6, TNFα, ICAM-1 and VCAM-1 expressions were analyzed by western blotting. β-Actin was used as an internal loading control. (b-f) IL-1β, IL-6, TNF-α, ICAM-1 and VCAM-1 expression levels were quantified by ImageJ software. Data were presented as mean ± Standard Deviation (SD) (n = 3). (g) IκBα, p-p65, and p65 expressions were analyzed by western blotting. β-Actin was used as an internal loading control. (h-i) IκBα, p-p65, and p65 expression levels were quantified by ImageJ software. Data were presented as mean ± SD (n = 3). (j) (Left panel) Immunofluorescence analysis of p65. Green staining represents the location of p65. Nuclei shown in blue were stained with DAPI. Scale bar = 10 μm. (Right panel) Co-localization of p-CAV1 and DAPI was analyzed by Pearson's correlation coefficient. Data were presented as mean ± SD (n = 3). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. The exact mean and SD values are presented in Table S31. Figure S19. Inflammatory factors and ROS levels in HUVEC exposed to PCB29-pQ. HUVECs were pretreated with 40 μM VC, 20 μM VE or 5 mM NAC for 1 h, followed with 5 μM PCB29-pQ exposure for 6 h. (n = 3). (a) p-p65, p65, IL-1β, IL-6, TNFα, ICAM-1 and VCAM-1 in cell lysates were analyzed by western blotting. (b-g) p-p65, p65, IL-1β, IL-6, TNFα, ICAM-1 and VCAM-1 expression levels were quantified by ImageJ software. Data were presented as mean ± Standard Deviation (SD) (n = 3). (h) ROS levels were detected by DCFH-DA (10 μM) probe. Data were presented as mean ± SD (n = 3). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. In all graphs, data is normalized to the control cell groups. The exact mean and SD values are presented in Table S32. Figure S20. p-CAV1 and CAV1 levels in HUVEC exposed to PCB29-pQ and antioxidants.
HUVECs were pretreated with 40 μM VC, 20 μM VE or 5 mM NAC for 1 h, followed by 5 μM PCB29-pQ exposure for 6 h. HUVECs were pretreated with (a) 40 μM VC, 20 μM VE or 5 mM NAC, (b) 200 U/ml PEG-SOD, 500 U/ml PEG-CAT and 5 mM GSH-MEE for 1 h, followed with 5 μM PCB29-pQ exposure for 6 h. (Left panel) p-CAV1 and CAV1 in cell lysates were analyzed by western blotting. β-Actin was used as an internal loading control. (Right panel) p-CAV1 and CAV1 expression levels were quantified by ImageJ software. Data were presented as mean ± Standard Deviation (SD) (n = 3). P values were determined by one-way ANOVA, followed by Tukey's post hoc test. All data is graphed relative to the control cell groups. The exact mean and SD values are presented in Table S33.

Figure S21. Heatmap showing correlation scores between factors. Plasma RNA was extracted by
TRNzol universal reagent. CHD group (n = 77) contains patients with > 50% coronary artery stenosis, and the control group (n = 50) contains subjects with < 50% coronary artery stenosis. RT-qPCR analysis of HDAC7-AS1, MIR-7-5p, TGF-β2, PPME1, IL-1β, IL6 and TNFα expressions. Spearman's rank correlation coefficient was used to assess the correlation between the two indicated factors. P value of < 0.05 was considered significant. The size and color of circle represents the correlation between two factors. X's mean no significance. Summary data can be found in Table S11.   Table S4. RT-qPCR primer sequences for lncRNAs and mRNAs. Gene Sequences are human unless otherwise specified.
Sequences are human unless otherwise specified.     Spearman's rank correlation coefficient was used to assess the correlation between two indicated mRNAs. Mean and SD values are the ones presented in Figure 1c-d. Data reported relative to control or Bio-NC. Mean and SD values are the ones presented in Figure 2a, b, c, d and f. Expression is relative to control or NC mimic for data from 2a, b, c, and d. For Figure 2F, data is expressed as luciferase activity (a.u.) Table S15. Endothelial injury and atherogenesis in ApoE -/mice exposed to PCB29-pQ, and apoptotic rate in HUVECs exposed to PCB29-pQ. Mean and SD values are the ones presented in Figure 3a, b and d.

Figure 6a (Right panel)
Data in figure 6c normalized to control ApoE -/mice.          Cq values are the ones presented in Figure 7a-g.

Figure S3a
Control Mean and SD values are the ones presented in Figure S3a-b. All data reported relative to control treated cells. Mean and SD values are the ones presented in Figure S7a-b (Lower panel). All data is presented as fold difference relative to NC inhibitor + NC siRNA. Mean and SD values are the ones presented in Figure S8a-c. All data presented as fold difference relative to NC mimic. All data is expressed as luciferase activity (a.u.). Mean and SD values are the ones presented in Figure S9 (Right panel). Number of branch points is presented. Mean and SD values are the ones presented in Figure S11a (mRNA expression of MIR-7-5p), b (mRNA expression of HDAC7-AS1), c (mRNA expression of TGFβ-2), and d (mRNA expression of PPME1), and f (protein expression of TGF-β2), g (protein expression of PPME1), and h (% cell viability). All data presented relative to NC inhibitor + NC siRNA.  Individual values  Individual values  Individual values  Individual values  267  310  211  187  210  330  103  198  201  352  190  208  340  401  165  288  215  290  157  276 Mean and SD values are the ones presented in Figure S12a (Right panel) and Figure S12a (plaque area relative to aortic section), b (TC; mmol/l), and c (LDL; mmol/l). Table S27. TGF-β2, PPME1, and inflammatory factors level in HUVECs exposed to PCB29-pQ with or without CAV1 siRNA. Mean and SD values are the ones presented in Figure S13a (Lower panel; p-CAV1/CAV1 ratio after treatment with PCB29-pQ for 0, 1, 3, or 6h, no CAV1 siRNA) and Figure c (p-CAV/CAV ratio), d (TGF-β2), e (PPME1), f (IL-1β), g (IL-6), and h (TNF-α). Data for c-h were normalized to β-actin expression and presented as fold difference compared to NC siRNA. Mean and SD values are the ones presented in Figure S14 (Right panel). Mean and SD values are the ones presented in Figure S16a (Right panel;), b (TC, mmol/l), and c (LDL, mmol/l). Mean and SD values are the ones presented in Figure S17b (IL-1β), c (IL-6), d (TNFα), e (ICAM-1), f (VCAM-1), g (The fluorescence intensity of Calcein-AM) (a.u.), i (p-p65/p65 ratio), j (IκBα), l (IL-1β), m (IL-6), and n (TNF-α). Data in b-e normalized to β-actin and presented as fold difference compared to 0 h control. Data in i presented as fold difference compared to 0 h. Data in j normalized to β-actin and presented as fold difference compared to 0 h. Data in l-n normalized to β-actin and presented as fold difference compared to control. Mean and SD values are the ones presented in Figure S18b (IL-1β), c (IL-6), d (TNF-α), e (ICAM-1), f (VCAM-1), h (p-p65/p65 ratio), I (IκBα), and j (Pearson's coefficient). Data in c-f and i normalized to β-actin and presented as fold difference compared to NC siRNA. Data in h presented as fold difference compared to NC siRNA. Mean and SD values are the ones presented in Figure S19b (p-65/p65 ratio), c (IL-1β), d (IL-6), e (TNF-α), f (ICAM-1), g (VCAM-1), and h (ROS level). Data in c-g are normalized to β-actin. All data presented as fold difference relative to control. Table S33. p-CAV1 protein expression relative to CAV1 total protein in HUVEC exposed to PCB29-pQ and antioxidants.