Mechanical Activation of Hypoxia-Inducible Factor 1α Drives Endothelial Dysfunction at Atheroprone Sites

Supplemental Digital Content is available in the text.


HIF1a is expressed by endothelial cells overlying plaques at low shear stress regions.
ApoE -/mice aged 8 weeks were exposed to a high fat diet for 6 weeks (n=4). C57BL/6 (wild-type) mice aged 14 weeks were also studied (n=4). The expression of HIF1a was quantified at a low wall shear stress region (inner curvature) by en face staining. EC were identified by co-staining with anti-CD31 antibodies conjugated to FITC (green). Cell nuclei were identified using TOPRO3 (blue). A representative image of HIF1a expression in EC overlying a plaque in the ApoE -/model is shown (left). Mean levels of HIF1a expression (+/-SEM) were calculated for ApoE -/and wild-type mice (right). Differences between means were analysed using a paired ttest. n.s., not significant. Validation of HIF1a siRNA. HUVEC were transfected with siRNA targeting HIF1a or with scrambled sequences. After 24 h, cells were exposed to DMOG for 4 h. The expression levels of HIF1a and HIF2a were assessed by Western blotting using specific antibodies, and anti-Calnexin antibodies were used to control for total protein levels. Representative blots are shown. Figure'1' 100"kD" "100"kD"

HIF1α
HIF1α'siRNA' Endothelial cells exposed to low shear stress in atmospheric oxygen were not hypoxic. HUVEC were exposed to orbital flow to generate low (5 dyn/cm 2 ) or high (11 dyn/cm 2 ) wall shear stress (WSS) either within a hypoxic chamber (1% O 2 ) or under atmospheric levels O 2 . After 72 h, the cells were treated with Pimonidazole (60 ng/ml) for 90 min before fixation with 4% PFA and staining using rabbit polyclonal antipimonidazole antibodies. Hypoxic cells were assessed by Pimonidazole staining (red). Cell nuclei were identified using DAPI (blue). Figure' Low shear stress induced HIF1a mRNA in coronary artery EC. Human coronary artery EC were exposed to low (5 dyn/cm 2 ) or high (11 dyn/cm 2 ) wall shear stress (WSS) for 72 h using an orbital system or were exposed to static conditions. HIF1a mRNA was quantified by qRT-PCR. Data were pooled from three different donors and mean HIF1a expression levels +/-SEM are shown. Differences between means were analysed using a 1-way ANOVA with the Bonferroni correction for multiple pairwise comparisons.

Validation of enforced IkBa expression.
HUVEC were transfected with pCMV-IkBa to inhibit NF-kB or with an empty plasmid as a control. After 24 h, the expression levels of IkBa and VCAM-1 were assessed by Western blotting using specific antibodies, and anti-Calnexin antibodies were used to control for total protein levels. Representative blots are shown.

Supplementary Figure VII.
Validation of antibodies used to detect Cezanne by Western blotting. HUVEC were transfected with siRNA targeting Cezanne or with scrambled sequences. Cells were then exposed to orbital flow to generate low WSS (5 dyn/cm 2 ). After 72 h, the expression levels of Cezanne were assessed by Western blotting using specific antibodies, and anti-PDHX antibodies were used to control for total protein levels. Representative blots are shown.
Supplementary Figure VIII. Cezanne positively regulates the expression of glycolysis genes and inflammatory molecules in EC exposed to low shear stress. HUVEC were transfected with siRNA targeting Cezanne or with scrambled sequences. Cells were subsequently exposed to orbital flow to generate low WSS (5 dyn/cm 2 ) for 72 h. The expression levels of Cezanne, HIF1a (A) or HK2, ENO2, PFKFB3, VCAM-1, ICAM-1, E-selectin, IL-8 and MCP1 (B) mRNA were assessed by qRT-PCR. Data were generated from three independent experiments and differences between means were analysed using a paired t-test. Low shear stress primed EC for hypoxic signaling. HUVEC were exposed to orbital flow to generate low (5 dyn/cm 2 ) or high (11 dyn/cm 2 ) wall shear stress (WSS). After 72h, cells were treated with DMOG for 4 h. The expression levels of HK2, ENO2 and PFKFB3 were assessed by Western blotting using specific antibodies, and anti-Calnexin antibodies were used to control for total protein levels. (A) Representative blots are shown. (B) Bands were quantified by densitometry. Data were pooled from three independent experiments and mean expression +/-SEM is shown. Differences between means were analysed using a 2way ANOVA with the Bonferroni correction for multiple pairwise comparisons. HUVEC were transfected with siRNA targeting HIF1a or with scrambled sequences. After 24 h, cells were exposed to orbital flow to generate low (5 dyn/cm 2 ) or high (5 dyn/cm 2 ) shear stress. After 72 h, EC proliferation was quantified by immunofluorescent staining using anti-Ki67 antibodies and co-staining using DAPI. Representative images are shown. The % Ki67-positive cells were calculated for multiple fields of view. Data were pooled from three independent experiments and mean expression +/-SEM is shown. Differences between means were analysed using a 2-way ANOVA with the Bonferroni correction for multiple pairwise comparisons. Low shear stress enhanced endothelial proliferation (Ki67 staining) via glycolysis. HUVEC were exposed to orbital flow to generate low shear stress (5 dyn/cm 2 ) in the presence of 2DG (5 mM) or 3PO (10 µM) or DMSO vehicle as a control. After 72 h, EC proliferation was quantified by immunofluorescent staining using anti-Ki67 antibodies and co-staining using DAPI. Representative images are shown. The % Ki67-positive cells were calculated for multiple fields of view. Data were pooled from three independent experiments and mean expression +/-SEM is shown. Differences between means were analysed using a 1-way ANOVA with the Bonferroni correction for multiple pairwise comparisons. FDG treatment did not alter the expression of Cezanne or HIF1a in EC. HUVEC were exposed to FDG for 24 h or 72 h or remained untreated as a control. The expression levels of HIF1a and Cezanne were assessed by Western blotting using specific antibodies, and anti-PDHX antibodies were used to control for total protein levels. Representative blots are shown. Bands were quantified by densitometry. Data were pooled from three independent experiments and mean levels +/-SEM are shown. Differences between means were analysed using 1-way ANOVA with the Bonferroni correction for multiple pairwise comparisons. n.s., not significant.