Effect And Mechanism Of The Nuclear Receptor LXR On Endothelial Repair After Vascular Injury

1. BackgroundVascular endothelial injury, which is the common pathophysiological basis ofhypertension, atherosclerosis and vascular disease in diabetes, is also the major cause ofpost-angioplasty restenosis and thrombosis. Endothelial regeneration and repair couldprevent thrombosis and suppress vascular smooth muscle cell proliferation after arterialinjury. Hence, accelerating re-endothelialization of injured arteries and restoring endothelialfuction are useful strategies for inhibiting post-angioplasty restenosis and thrombosis andpreventing vascular injured diseases.Liver X receptors (LXRs) are members of nuclear receptor superfamily. Onceactivated by endogenous ligands or synthetic agonists, LXR can regulate cholesterolmetabolism, glucose metabolism and inflammations by influencing the expression of anarray of LXR target genes. Studies performed recently have showed that LXR might playimportant roles in preventing cardiovascular diseases:(1) LXR agonists inhibited thedevelopment of atherosclerosis,(2) LXR agonists suppressed neointima formation inballoon-injured rat carotid arteries,(3) LXR activation promoted angiogenesis afterischemic stroke. Since endothelial regeneration and repair after vascular injury are thecommon steps in preventing above cardiovascular diseases, we hypothesize that activationof endogenous LXR might regulate all above pathophysiology processes by stimulatingendothelial regeneration.In our preliminary experiments, we have found that treatment of LXR agonistT0901317increased re-endothelialized area in a mouse model of carotid injury, suggestingthat activation of endogenous LXR could promote endothelial regeneration and repair of injured arteries. However, the mechanisms are not known. Endothelial regenerationdepends on:(1) neighbouring endothelial cells (ECs) proliferation and migration;(2)endothelial progenitor cells (EPCs) derived from bone marrow, spleen and etc.Accumulating evidence has demonstrated that EPCs, which mobilized to circulation byischemia, physical training, and the administration of statins, estrogen, and a variety ofcytokines, can be recruited to sites of endothelial injury. There, they promotere-endothelialization directly by incorporating into the recovering endothelium at the sitesof injury and indirectly by producing and releasing angiogenic growth factors.Based on these observations, we speculate that activation of LXR might promoteendothelial regeneration and repair after vascular injury via regulating ECs and EPCsbiological functions (such as proliferation, migration and etc.).2. Methods2.1. Effect of LXR activation on endothelial repair after vascular injuryTo explain the effect of LXR activation on endothelial repair after injury, weestablished an mouse model of carotid injury as previously described. Mice were orallyadministered of LXR agonist T0901317(30mg/kg per day) or vehicle starting2daysbefore the injury. To measure the re-endothelialized area, animals were perfused in vivowith Evans blue dye4,7, and14days after the injury.2.2. Effect and mechanism of LXR on ECs proliferation and migrationHuman umbilical vein endothelial cells (HUVECs) were obtained and cultured by ourestablished methods. Immunostaining for von Willebrand factor (vWF) confirmed that thecells were endothelial. Cells used for experiments were from passages3through5.HUVECs were treated with different concentrations of LXR agonist T0901317(0μmol/L,0.5μmol/L,2μmol/L,5μmol/L) for different time course. HUVECs proliferationwas evaluated using a colorimetric MTS assay kit and HUVECs migration assay wasperformed using a Transwell chamber (8μm pore size). In addition, western blot analysiswas used to determine the expression of phosphorylated Akt (Ser473), total Akt,phosphorylated eNOS (Ser1177) and total eNOS. LY294002(10μmol/L) and L-NAME (100μmol/L), when used, were added30minutes prior to the addition of LXR agonist.2.3. Effect and mechanism of LXR on EPCs biological functionsTwo distinct species and sources of EPCs, rat bone marrow-derived EPCs (BM-EPCs) and mouse spleen-derived EPCs (spleen-EPCs), were used for research. The EPCs wereisolated, cultured and characterized following our established protocol. After5~7days ofculture, the EPCs were used for experiments.To determine the expression of LXR in rat BM-EPCs and mouse spleen-EPCs, weused reverse transcriptase-PCR (RT-PCR), western blot, and confocal immunofluorescence.Rat BM-EPCs and mouse spleen-EPCs were treated with different dosages of LXRagonists T0901317(0μmol/L,0.5μmol/L,2μmol/L,5μmol/L) or GW3965(0μmol/L,0.5μmol/L,2μmol/L,5μmol/L) for different time course. EPCs proliferation was evaluatedusing a colorimetric MTS assay kit and EPCs migration assay was performed using aTranswell chamber (8μm pore size). In addition, western blot analysis was used todetermine the expression of phosphorylated Akt (Ser473), total Akt, phosphorylated eNOS(Ser1177) and total eNOS. LY294002(10μmol/L) and L-NAME (100μmol/L), when used,were added30minutes prior to the addition of LXR agonists.EPCs were treated with LXR agonist T0901317(2μmol/L) for24hours, and then,semi-quantitative RT-PCR was used to determine the mRNA expressions of angiogenicgrowth factors (VEGF, SDF-1, HGF, IGF-1and G-CSF).EPCs were treated with different concentrations of LXR agonists T0901317(0μmol/L,0.5μmol/L,2μmol/L,5μmol/L) or GW3965(0μmol/L,0.5μmol/L,2μmol/L,5μmol/L)for different time course. In addition, to detect the VEGF mRNA, protein expression andsecretion, we used Real Time-PCR, Western blot, and ELISA.2.4. Statistical analysisAll values are expressed as the means±S.D. of at least three experiments.Comparisons between the groups were analyzed using one-way ANOVA followed by theappropriate post-hoc test. A value of P <0.05was considered to be statistically significant.3. Results3.1. Effect of LXR activation on endothelial repair after vascular injuryTreatment of LXR agonist T0901317increased re-endothelialized area as comparedwith control at all time points (4d,7d,14d) after arterial injury (P＜0.05,n=8), suggestingthat activation of endogenous LXR could promote endothelial regeneration and repair ofinjured vessels.3.2. Effect and mechanism of LXR on ECs proliferation and migration 3.2.1. Characterization of HUVECsHUVECs were grown as confluent monolayers with a "cobblestone" morphology. Thecells were homogenous, closely apposed, large, flat, and polygonal. Cells showing positivestaining for von Willebrand factor were HUVECs (＞95%, n=6).3.2.2. LXR agonist enhances HUVECs proliferation and migrationTreatment with LXR agonist T0901317(0.5μmol/L,2μmol/L,5μmol/L) profoundlyincreased the HUVEC proliferation and migration in dose-dependent manners (P＜0.05,P＜0.01,n=4-6). These findings suggested that LXR agonist could augment HUVECsproliferation and migration.3.2.3. Mechanism of LXR activation on ECs proliferation and migration3.2.3.1. LXR agonist influences PI3K/Akt/eNOS signaling in HUVECsLXR agonist T0901317induced significant increases in the Ser473Akt phosphorylationand Ser1177eNOS phosphorylation in time-and dose-dependent manners in HUVECs (P＜0.05,P＜0.01,n=3-5), which were abolished by the PI3K inhibitor LY294002. Theseobservations suggested that LXR activation could influence PI3K/Akt/eNOS signaling inHUVECs. However, T0901317did not change the levels of phosphorylated Akt andphosphorylated eNOS within1hour, suggesting that LXR activation might activate thePI3K/Akt/eNOS pathway via an indirect mechanism.3.2.3.2. Activation of LXR promote HUVECs proliferation and migration via thePI3K/Akt/eNOS signaling pathwayIt was showed that both the PI3K inhibitor LY294002(10μmol/L) and the eNOSinhibitor L-NAME (100μmol/L) significantly attenuated T0901317(5μmol/L)-stimulatedHUVECs proliferation and migration (P＜0.05, P＜0.01, n=4-6), whereas neitherLY294002nor L-NAME treatment influenced basal HUVECs proliferation and migration inthe absence of LXR agonist (P=N.S., n=4-6). These data suggested that thePI3K/Akt/eNOS signaling pathway was involved in LXR activation-induced HUVECsproliferation and migration.3.3. Effect and mechanism of LXR on EPCs biological functions3.3.1. Characterization of rat bone marrow-derived EPCs (BM-EPCs)The adherent rat bone marrow-derived mononuclear cells (BM-MNCs) wereoval-shaped, spindle-like, and polygonal. Cell clusters and colonies appeared in early phase (4d~14d). In late phase (>14d), cord-like structures and network formation were observedwhen cell density was low, where as the cells exhibited typical "cobblestone" morphologyat high cell density conditions. The majority of the adherent BM-MNCs (>90%) werepositive for both ac-LDL (red) uptake and UEA-I lectin binding (green), which indicatedendothelial cell characteristics. These adherent BM-MNCs were further characterized bydemonstrating the expression of the stem cell marker CD133(89.2%),CD34(75.4%), theendothelial cell lineage antigen CD31(91.9%) and VEGFR-2(92.9%) by FCM.3.3.2. Characterization of mouse spleen-derived EPCs (spleen-EPCs)The adherent mouse spleen-derived mononuclear cells (spleen-MNCs) wereoval-shaped, spindle-like, and polygonal. Cell clusters and colonies could be found in earlyphase (4d~14d). In late phase (>14d), cord-like structures and network formation wereobserved when cell density was low, where as the cells exhibited typical "cobblestone"morphology at high cell density conditions. The majority of the adherent spleen-MNCs(>90%) were positive for both ac-LDL (red) uptake and UEA-I lectin binding (green),consistent with endothelial lineage cells. These adherent spleen-MNCs were furthercharacterized by demonstrating the expression of the stem cell marker Sca-1(75.4%),CD34(82.3%) and c-kit (64.9%), the endothelial cell lineage antigen CD31(84.6%) andVEGFR-2(86.2%) by FCM.3.3.3. LXRα and LXRβ are expressed in EPCsLXRα and LXRβ mRNA and protein were present in the rat BM-EPCs and mousespleen-EPCs as determined by RT-PCR and western blot. Additionally, LXRα and LXRβimmunofluorescence was observed primarily in the nucleus of the EPCs, whereas onlyweak staining was observed throughout the cytoplasm. Treatment of the EPCs with theLXR agonists T0901317or GW3965led to a significant up-regulation of the LXR targetgene ABCA1, which indicated that the LXRs expressed in EPCs are biologically active.3.3.3. LXR agonists enhance EPCs proliferation and migrationAfter treatment of synthetic LXR agonists T0901317(0.5μmol/L,2μmol/L,5μmol/L)or GW3965(0.5μmol/L,2μmol/L,5μmol/L), the proliferation and migration of EPCs (ratBM-EPCs and mouse spleen-EPCs) increased profoundly in dose-dependent manners (P＜0.05,P＜0.01,n=4-6). These findings suggested that LXR activation could enhance EPCsproliferation and migration. 3.3.4. Mechanism of LXR activation on EPCs proliferation and migration3.3.4.1. LXR activation influences PI3K/Akt/eNOS signaling in EPCsBoth synthetic LXR agonists T0901317and GW3965induced significant increases inthe Ser473Akt phosphorylation and Ser1177eNOS phosphorylation in time-anddose-dependent manners in EPCs (rat BM-EPCs and mouse spleen-EPCs)(P＜0.05,P＜0.01,n=3-5), which were diminished by the PI3K inhibitor LY294002. These observationssuggested LXR activation influences PI3K/Akt/eNOS signaling in EPCs. Nevertheless,treatment with T0901317did not change the levels of phosphorylated Akt andphosphorylated eNOS within1hour, showing that LXR activation might activate thePI3K/Akt/eNOS pathway through an indirect mechanism.3.3.4.2. LXR agonists promote EPCs proliferation and migration via thePI3K/Akt/eNOS signaling pathwayWe found that both the PI3K inhibitor LY294002(10μmol/L) and the eNOS inhibitorL-NAME (100μmol/L) significantly attenuated T0901317(2μmol/L)-or GW3965(5μmol/L)-enhanced EPCs (rat BM-EPCs and mouse spleen-EPCs) proliferation andmigration (P＜0.05,P＜0.01,n=4-6), whereas neither LY294002nor L-NAME treatmentinfluenced basal EPC proliferation and migration in the absence of LXR agonists (P=N.S.,n=4-6). These data demonstrated suggested that the PI3K/Akt/eNOS signaling pathway wasinvolved in LXR activation-promoted EPCs proliferation and migration.3.3.5. Effect of LXR on the production of angiogenic growth factors by EPCs3.3.5.1. Effect of LXR activation on the mRNA expressions of angiogenic growthfactorsLXR agonist T0901317(2μmol/L) treatment didn't change the mRNA levels ofSDF-1, HGF, IGF-1and G-CSF, but increased the VEGF mRNA expression (P＜0.05,n=4),suggesting that LXR activation might up-regulate VEGF mRNA expression in EPCs.3.3.5.2. LXR agonists up-regulate VEGF expression and induces VEGF secretion inEPCsTreatment of EPCs with either T0901317or GW3965led to an increased level ofVEGF mRNA expression in time-and dose-dependent manners (P＜0.05, P＜0.01, n=4-6).In addition, T0901317or GW3965treatment similarly increased the VEGF protein level (P＜0.05, n=6) and stimulated VEGF secretion (P＜0.01, n=5). These observations demonstrated that LXR agonists could up-regulate VEGF expression and induce VEGFsecretion in EPCs.4. Conclusion4.1. LXR activation increases ECs proliferation and migration throughPI3K/Akt/eNOS signaling pathway.4.2. LXRα and LXRβ are expressed in EPCs. LXR agonists enhance EPCsproliferation and migration via PI3K/Akt/eNOS pathway. In addition, LXR agonistsup-regulate VEGF expression and induce VEGF secretion by EPCs.4.3. Activation of LXR could promote endothelial regeneration and repair aftervascular injury through regulating the functions of ECs and EPCs.