www.sciencesignaling.org/cgi/content/full/8/389/ra79/dc1 Supplementary Materials for HDL-bound sphingosine 1-phosphate acts as a biased agonist for the endothelial cell receptor S1P 1 to limit vascular inflammation Sylvain Galvani, Marie Sanson, Victoria A. Blaho, Steven L. Swendeman, Hideru Obinata, Heather Conger, Björn Dahlbäck, Mari Kono, Richard L. Proia, Jonathan D. Smith, Timothy Hla* The PDF file includes: *Corresponding author. E-mail: tih2002@med.cornell.edu Published 11 August 2015, Sci. Signal. 8, ra79 (2015) DOI: 10.1126/scisignal.aaa2581 Fig. S1. S1P 1 localization in en face preparations of the mouse aorta. Fig. S2. ICAM-1 and VCAM-1 abundance in aorta upon S1P 1 deletion. Fig. S3. Effect of albumin and albumin-s1p on TNF -induced increases in ICAM-1 and VCAM-1 abundance. Fig. S4. Characterization of the sphingolipid composition of human HDL. Fig. S5. Effect of S1P carrier on S1P 1 retention at the cell surface. Fig. S6. Effect of S1P 1 inhibition in S1P-induced activation of MAPK. Fig. S7. Subcellular localization of -arrestin 2 depends on the S1P carrier. Fig. S8. Genetic modulation of S1P 1 expression in endothelial cells and modification of ICAM-1 abundance. Fig. S9. Analysis of weight, plasma cholesterol, and cell populations in Apoe / S1pr1 wild-type and ECKO animals after 16 weeks on a high-fat diet. Fig. S10. Plasma sphingolipid characterization. Fig. S11. Analysis of plaque composition. Fig. S12. Atherosclerotic root lesion quantification.
Supplemental Figure 1. S1P1 localization in en face preparations of the mouse aorta. Aortae were dissected and intima was stained in en face preparations. (A) S1P1, VE-Cadherin, (B) ICAM-1 and (C) VCAM-1 abundance were analyzed by immunofluorescence in descending aorta and lesser curvature. Images are representative of 6-8 mice obtained from 4 independent experiments.
Supplemental Figure 2. ICAM-1 and VCAM-1 abundance in aorta upon S1P1 deletion. (A) Aortic lysates from aortic arch and descending aortae of WT and S1pr1ECKO animals were analyzed for VCAM-1 and ICAM-1 abundance by immunoblot analysis. (B) Data represent combined analysis from 2 representative experiments (one that contained 3 aortae for each condition pooled together and another that contained 2 aortae for each condition pooled together). (Data are mean ± SD. *p< 0.05 by Mann Whitney test). (C) In vivo cell surface biotinylation followed by streptavidin precipitation on total aortic lysates. ICAM-1 abundance was determined by immunoblot analysis. Results are from an experiment that contained two mice in each group.
Supplemental Figure 3. Effect of albumin and albumin-s1p on TNFα-induced increases in ICAM-1 and VCAM-1 abundance. (A) Serum starved HUVECs were stimulated with TNFα and with increasing doses of albumin or albumin-s1p as indicated. ICAM-1 abundance was assessed by immunoblot. Immunoblots are representative of an experiment that was repeated 3 times. (B) Data represent combined analysis from 3 independent experiments (mean± SD). (C) Serum starved HUVECs were stimulated by TNFα for 6h and with recombinant ApoM (ApoM), recombinant ApoM loaded with S1P (ApoM-S1P, [S1P]=100nM),or albumin-s1p ([S1P]=100nM) as indicated. VCAM-1 abundance was assessed by flow cytometry. (Data are mean ± SD. n = 4 independent experiments analyzed separately *p< 0.05 by unpaired Student s t-test).
Supplemental Figure 4. Characterization of the sphingolipid composition of human HDL HDL particles were isolated from human plasma and their sphingolipid composition was analyzed by liquid chromatography tandem mass spectrometry. N=2 different pools of human HDL isolated from two different plasma samples.
Supplemental Figure 5. Effect of S1P carrier on S1P1 retention at the cell surface. (A) Serum starved HUVECs were stimulated as indicated. Cell surface proteins were biotinylated, enriched by streptavidin immunoprecipitation and analyzed by Western blot. (B) Data represent combined densitometric analysis from 3 separate experiments (data are expressed as mean±sd. *p<0.05, **p<0.01 compared to unstimulated by one way ANOVA followed by Dunnett s multiple comparison.
Supplemental Figure 6. Effect of S1P1 inhibition in S1P-induced activation of MAPK. Serum starved HUVECs were preincubated or not with phosphorylated FTY720 (100nM) for 30 minutes followed by stimulation with either albumin-s1p or HDL-S1P (100nM) in presence or in absence of P-FTY720. ERK (phospho-form and total ERK) were analyzed by Western blot. Immunoblots are representative of 3 independent experiments.
Supplemental Figure 7. Subcellular localization of arrestin 2 depends on the S1P carrier. HUVECs were transduced with lentivirus vector for the expression of DS-Red-tagged arrestin-2. Serum starved cells were stimulated with huhdl ([S1P]=100nM), or albumin bound S1P ([S1P]=100nM) for the indicated time. Confocal microscopy was performed to assess the localization of DS-Red tagged arrestin-2. Images are representative of 3 independent experiments.
Supplemental Figure 8. Genetic modulation of S1P1 expression in endothelial cells and modification of ICAM-1 abundance. Aortae from Apoe -/- S1pr1 f/f WT and Apoe -/- S1pr1ECKO mice were dissected and intima was stained in en face preparations. (A) Immunostaining of S1P1 and VE-Cadherin. (B) Immunostaining for ICAM-1 in presence of absence of endothelial S1P1. Images are representative of 6 mice per genotype obtained from 3 independent experiments.
Supplemental Figure 9. Analysis of weight, plasma cholesterol, and cell populations in Apoe -/- S1pr1 wild-type and ECKO animals after 16 weeks on a high-fat diet. Apoe -/- S1pr1 WT and Apoe -/- S1pr1ECKO littermates were placed on high fat diet for 16 weeks. (A) Body weight was monitored weekly Data are expressed as mean SD. Statistical analysis was done using 2 way ANOVA followed by Dunnett s multiple comparison. After 16 weeks, mice were euthanized and plasma and blood were collected. (B) Plasma cholesterol concentrations were assessed using Cholesterol E CHOD-DAOS Method kit. Blood cell populations were analyzed by flow cytometry. (C) Analysis of circulating innate immune cell population and (D) circulating lymphocyte populations in Apoe -/- S1pr1 WT and Apoe -/- S1pr1 ECKO animals. Data are expressed as mean SD (6 Apoe -/- WT mice, 8 Apoe -/- S1pr1ECKO mice). Statistical analysis was done using the Mann-Whitney test.
Supplemental Figure 10. Plasma sphingolipid characterization. Plasma Apoe -/- S1pr1 WT and Apoe -/- S1pr1 ECKO animals from animals euthanized in Fig S8 were isolated and sphingolipid composition was analyzed by LC/MS. Data are represented as mean value SD (4 Apoe -/- S1pr1 WT mice and 9 Apoe -/- S1pr1 ECKO mice). The Mann- Whitney test was used to analyze statistical significance.
Supplemental Figure 11. Analysis of plaque composition. Lesions from descending aortic were isolated, fixed for 20 min in PFA 4%, then placed in 10% sucrose for 24 hours prior to embedding in OCT. (A) Histologic staining was done. (Hematoxylin/Eosin, Masson Trichrome, Verhoff s Stain with Van Gieson counter stain). (B) Immunofluorescent staining for CD31, smooth muscle actin (SM-actin) and ICAM-1 were done as described. Nuclei were counterstained using TOPRO-3 (n=2 mice for each genotype).
Supplemental Figure 12. Atherosclerotic root lesion quantification. (A) Aortic roots from mice euthanized in Fig S8 were stained using Oil Red O. Shown here is a representative image of aortic root section from each group. (B) Quantification of atherosclerotic plaques in the aortic root was assessed (6 Apoe -/- WT mice, 8 Apoe -/- S1pr1ECKO mice). The Mann-Whitney test was used to analyze statistical significance.