Diabetic HDL is dysfunctional in stimulating endothelial cell migration and proliferation due to down regulation of SR-BI expression

Abstract

Background: Diabetic HDL had diminished capacity to stimulate endothelial cell (EC) proliferation, migration, and adhesion to extracellular matrix. The mechanism of such dysfunction is poorly understood and we therefore sought to determine the mechanistic features of diabetic HDL dysfunction.

Methodology/principal findings: We found that the dysfunction of diabetic HDL on human umbilical vein endothelial cells (HUVECs) was associated with the down regulation of the HDL receptor protein, SR-BI. Akt-phosphorylation in HUVECs was induced in a biphasic manner by normal HDL. While diabetic HDL induced Akt phosphorylation normally after 20 minutes, the phosphorylation observed 24 hours after diabetic HDL treatment was reduced. To determine the role of SR-BI down regulation on diminished EC responses of diabetic HDL, Mouse aortic endothelial cells (MAECs) were isolated from wild type and SR-BI (-/-) mice, and treated with normal and diabetic HDL. The proliferative and migratory effects of normal HDL on wild type MAECs were greatly diminished in SR-BI (-/-) cells. In contrast, response to diabetic HDL was impaired in both types suggesting diminished effectiveness of diabetic HDL on EC proliferation and migration might be due to the down regulation of SR-BI. Additionally, SR-BI down regulation diminishes diabetic HDL's capacity to activate Akt chronically.

Conclusions/significance: Diabetic HDL was dysfunctional in promoting EC proliferation, migration, and adhesion to matrix which was associated with the down-regulation of SR-BI. Additionally, SR-BI down regulation diminishes diabetic HDL's capacity to activate Akt chronically.

Figure 1. Diabetic HDL is less efficient in stimulating EC proliferation. A)

HUVECs were treated with N-HDL and D-HDL for 24 hours, and cell proliferation was measured using MTT assay. Mean ± SEM, n = 6, ***, p<0.001 by student’s t-test. B) HUVECs were treated with N-HDL or G-HDL for 24 hours at apoA-I concentrations of 50 and 100 µg/ml respectively, and cell proliferation was measured using MTT assay. Although N-HDL promoted cell proliferation at 24 hours, G-HDL at 100 g/ml apoA-I decreased cell proliferation (mean ± SEM; **, p<0.01 by student’s t-test). C) HUVECs were treated with N-HDL, D-HDL, G-HDL or Ox-HDL for 12, 24, 36, 48, 60 or 72 hours at apoA-I concentrations of 100 µg/ml and cell proliferation using BrdU proliferation assay was shown (mean ± SEM; **, p<0.01 by ANOVA and Bonferroni’s Multiple Comparison Test).

Figure 2. Diabetic HDL is much less efficient in promoting EC migration. A)

Scratched HUVEC monolayers were treated with media alone (C, control), N-HDL (N), or D-HDL (D) for 24 hours, and migration into the wound was photographed (100× objective lens). B) N-HDL or D-HDL (n = 10 each) was added to each well and transwell migration was evaluated after 8 hours. C) Quantification migration in the wound healing assay (n = 3, mean ± SEM; ***, p<0.001 by ANOVA and Bonferroni’s Multiple Comparison Test). D) Cell migration based upon an 8 hour incubation in the transwell migration assay (n = 10, mean ± SEM, ns, p>0.05 and ***, p<0.001 by ANOVA and Bonferroni’s Multiple Comparison Test) was shown. E) Scratched HUVEC monolayers were treated with N-HDL (N), or D-HDL (D) for 3, 6, 12 and 24 hours, and the migration into the wound was photographed (100× objective lens). F) The migration in wound healing assay was quantified (n = 3, mean ± SEM; ***, p<0.001 by ANOVA and Bonferroni’s Multiple Comparison Test). G) Transwell migration assay was applied to HUVECs treated with media alone (C, control), N-HDL, G-HDL, or Ox-HDL at an apoA-I concentration of 100 µg/ml for 8 hours (mean ± SEM, **, p<0.01 and ***, p<0.001 by ANOVA test and Bonferroni’s Multiple Comparison Test).

Figure 3. Diabetic HDL decreases cell adhesion to ECM. A)

HUVECs were pretreated with N-HDL or D-HDL for 24 hours, and relative cell adhesion to ECM was determined after a 1 hour incubation, n = 6, mean ± SEM, *, p<0.05 by student’s t-test. B) HUVECs were pretreated with N-HDL or D-HDL for 6 hours. Integrin αv expression was determined by flow cytometry, n = 3, mean ± SEM, *, p<0.05 by student’s t-test. C) Flow cytometry overlay histogram shows integrin αv expression of HUVECs treated with N-HDL or D-HDL compared to control.

Figure 4. EC SR-BI expression is reduced by diabetic HDL. A)

HUVECs were treated with PBS, N-HDL or D-HDL for 12 hours. Real-time PCR experiment shows that the gene expression of SR-BI, and the ratio of SR-BI/GAPDH (mean ± SEM) is shown, n = 6 each, ***, p<0.001 by one-way ANOVA. B) HUVECs were treated with PBS, N-HDL or D-HDL for 24 hours. Representative western blot shows that D-HDL down-regulated the expression of SR-BI. C) The density of the SR-BI bands was normalized to the density of the β-actin band. The ratio of SR-BI/β-actin (mean ± SEM) is shown, p<0.01 by student’s t-test. D) HUVECs were treated with N-HDL or D-HDL for 5, 10, 20, 30, 60 minutes, 3, 6, 8, 12, or 24 hours, and SR-BI expression was shown. E) HUVECs were treated with N-HDL or D-HDL for 24 hours, and ABCG1 expression was shown. F) HUVECs were treated with N-HDL or D-HDL for 8 or 24 hours. SR-BI levels on the cell surface were shown as percentage of control, mean ± SEM, *, P<0.05 by one-way ANOVA. G) HUVECs were treated with with media alone (C, control), N-HDL, G-HDL, and Ox-HDL at an apoA-I concentration of 100 µg/ml for 24 hours, and western blotting assay was performed. H) The density of the SR-BI bands was normalized to the β-actin band (***, p<0.001 by a Student’s t test).

Figure 5. SR-BI mediates the effects of N-HDL on MAEC proliferation and migration. A)

Representative agarose gel of SR-BI genotype assay by PCR. Lanes 1 and 2 are derived from 2 littermates from the offspring of SR-BI +/− parents. Lane 3 indicates molecular weight markers. Lanes 4 and 5 are derived from a wild type (SR-BI +/+) and a heterozygous (SR-BI +/−) individual. Lane 6 is negative control with water. B) Protein expression of SR-BI in liver from SR-BI (+/+) and SR-BI (−/−) mice was measured using western blot. C) MAECs from SR-BI (+/+) and SR-BI (−/−) mice were treated with N-HDL or D-HDL for 24 hours. Relative cell number was compared to untreated control cells using the MTT assay. Box-whisker plots of cell number are shown (**, p<0.01 by non parametric ANOVA test, n = 6). D) MAECs from SR-BI (+/+) or SR-BI (−/−) mice were treated with N-HDL or D-HDL for 8 hours and transwell migration assays were performed (***, p<0.001 by non parametric ANOVA test, n = 6).

Figure 6. Akt phosphorylation induced by HDL. A)

HUVECs were treated without (control) or with the addition of N-HDL or D-HDL for 5, 10, 20, 30, 60 minutes, 3, 6, 12, or 24 hours. Expression levels of phospho-Akt (Ser473) and Akt1/2 were analyzed by western blotting. B) HUVECs were treated with N-HDL or D-HDL (n = 3 each) for 20 minutes or 24 hours. Expression levels of phospho-Akt (Ser473), Akt1/2, and β-actin were analyzed by western blotting. C) The density of the phospho-Akt bands at 20 minutes was normalized to the β-actin band (ns, p>0.05). D) The density of the phospho-Akt bands at 24 hours was normalized to the β-actin band (**, p<0.01 by a student’s t test). E) The density of the phospho-Akt bands at 20 minutes was normalized to the Akt1/2 band (ns, p>0.05). F) The density of the phospho-Akt bands at 24 hours was normalized to the Akt1/2 band (*, p<0.05 by a student’s t test). G) MAECs from SR-BI (+/+) or SR-BI (−/−) mice were treated without (control) or with N-HDL or D-HDL for 24 hours, and expression levels of phospho-Akt (Ser473), Akt1/2, and β-actin were analyzed by western blotting. H) The density of the phospho-Akt bands of MAECs from SR-BI (+/+) or SR-BI (−/−) mice was normalized to the Akt1/2 band (*, p<0.05 and **, p<0.01 by a student’s t test).