| BackgroundDiabetic vascular diseases, among chronic complications in diabetes mellitus, are related with atherosclerosis, diabetic retinopathy, nephropathy and diabetic foot. Ineffective vascular repair is likely to be an important contributor to the diabetic vascular disease, and this effect is linked with impaired neovascularization in diabetes. Efforts to investigate the mechanisms of impaired neovascularization in diabetes and find new ways to improve it are significant to therapies of diabetic vascular complications.Neovascularization is composed of angiogenesis and vasculogenesis, which involves distinct mechanisms. Vasculogenesis occurs in embryogenesis, and has been found to contribute to postnatal neovascularization due to the discovery of endothelial progenitor cells (EPCs). Asahara et al. reported for the first time the existence of EPCs in1997. Subsequent studies supported that EPCs played an crucial role in postnatal vasculogenesis. In the last decade, the important role played by EPCs in cardiovascular health is becoming increasing appreciated, especially for their crucial role in maintenance of endothelial integrity and function, as well as postnatal neovascularization.There is accumulating evidence showing reduced number and impaired EPC function in the presence of diabetes mellitus, and dissatisfactory effects of diabetic EPC therapy. which limits the use of EPCs in practice. However, the mechanisms of EPC dysfunction in diabetes are incompletely understood. Nitric oxide (NO) was reported to play an important role in EPC mediated vasculogenesis. However, eNOS-derived NO decreases in the setting of diabetes, which contributes to impaired EPC function. Evidence has showed that eNOS resides in caveolae and Cav-1could inhibit the activity of eNOS via direct interaction with eNOS. Though the expression of Cav-1was upregulated in several tissues in diabetes, the change of Cav-1in diabetic EPCs is not clear. In current study, we hypothesize that Cav-1may be also upregulated in diabetic EPCs and involves in dysregulated eNOS/NO pathway. We determined the expression of Cav-1in EPCs from type2diabetic db/db mice, and tried to reveal the relationship between Cav-1and metabolic abnormalities, like hyperglycemia and hyperleptinemia. Specifically, we generated lentivirus carrying Cav-1siRNA to elucidate the role of Cav-1in EPC function and the underlying mechanisms in the setting of diabetes.Objectives1. To measure the changes of expression of Cav-1in type2diabetic EPCs, and to investgate the underlying mechanisms.2. To investigate the role and mechanisms of Cav-1in regulating EPC functions in type2diabetes.Methods1. Animal modelMale db/db (BKS.Cg-m+/+Leprdbh/J) mice and their control mice(db/+), aged12weeks, were purchased from Model Animal Research Center of Nanjing University. In the current study, db/db mice with a blood glucose level>300mg/dl were considered diabetic and were used for EPC isolation.2. EPC isolation, culture and characterizationBone-marrow mononuclear cells were isolated from the femurs, tibias and pelvis of mice and then cultured with endothelial cell growth medium2in37℃5%CO2. After72hours, nonadherent cells were removed by changing medium and adherent cells were cultured continuously till day7for experimental use. The EPCs after7-day culture were characterized by Hoechst33258, DiI-acLDL and FITC-conjugated isolectin staining. To further characterize the phenotype of EPCs, the expressions of CD34, Sca-1, CD144, Flk-1and CD11b were analyzed by flow cytometry.3. EPC function assaysThe tube formation and migration assays were performed to determine the angiogenic function of EPCs. For tube formation assay,5×104EPCs were plated in a well of48-well plate pre-coated with growth factor-reduced Matrigel. After24-hour incubation at37℃5%CO2with EGM-2plus5%FBS, images were taken by an inverted microscope and tube lengths were measured. EPC migration was measured using Boyden chambers. Briefly,5×104EPCs were plated in the upper chamber of Transwell and allowed to migrate in a37℃5%CO2incubator. The upper cells were removed and the cells in the lower side were fixed and counted with a microscope.4. Lentivirus constructsWe constructed a siRNA for mouse caveolin-1and a control (scrambled) that does not affect any protein expression using oligo pairs. Lentiviruses encoding scrambled and caveolin-1siRNA were produced by transfection of HEK293T which is performed commercially by GENECHEM.5. Western Blot Analysis20-50μg protein was subjected to SDS-polyacrylamide gel electrophoresis. The PVDF membranes were probed with primary antibodies against Cav-1, p-eNOS1177, eNOS and Actin. The blots were scanned with an ImageQuant4000system and band intensity was quantified with Quantity One System.6. Real-time PCRFor each sample,1μg RNA was reverse transcribed using a reagent kit. Real-time PCR was performed using SYBR Green PCR Master Mix on a iCycler iQ Real-time PCR System. Variation in transcription levels was calculated using2△CT.7. Measurement of NO releaseAfter being serum-starved for24hours, EPCs were treated with VEGF (50ng/ml) for30minutes. Nitrite concentration in the culture medium was measured using the Griess reaction, and sodium nitrite served as a standard. The optical density of samples was measured in a reader system.8. Statistical AnalysisData were presented as means±SEM. Differences between two groups were compared by Student’s t-test, and one-way ANOVA with Newman-Keuls multiple comparison tests was used for multiple groups. In all the tests, a value of P<0.05was taken as statistically significant.Results1. Leptin regulated Cav-1mRNA and protein expressionIt was reported that leptin could upregulate the expression of Cav-1in HCAEC and HUVEC. To test whether leptin could regulate expression of Cav-1in EPCs, db/+EPCs were incubated with increasing concentration of leptin (0.20,50,100ng/ml) for24h, then cells were harvested for mRNA and protein analysis. Our data showed the mRNA and protein of Cav-1was upregulated by leptin at the concentration of20ng/ml. When the concentration of leptin was increased, Cav-1expression rose accordingly.2. Cav-1protein expression was upregulated in db/db EPCs independent of high leptin levelCompared with db/+EPCs, db/db EPCs showed higher expression of Cav-1. We treated db/db EPCs with100ng/ml leptin to measure the Cav-1expression. The result showed a high level of leptin could not upregulate the expression of Cav-1in db/db EPCs.3. High glucose upregulated Cav-1protein expression in db/db and db/+EPCsAs compared with that in control medium (5mmol/l glucose), the expression of Cav-I in db/+EPCs assessed by Western blot was significantly increased, by94.77%in high-glucose medium. High glucose also upregulated Cav-1expression in db/db EPCs by39.36%, though db/db EPCs had a higher expression of Cav-1originally. By contrast, the osmotic control with mannitol did not change the expression of Cav-1.4. Cav-1knockdown improved functions of db/db EPCsOur result showed db/db EPCs formed significantly fewer networks than db/+EPCs. In contrast, when Cav-1expression was knocked down by lentivirus-mediated siRNA, Matrigel tube formation was increased by69.32%in db/db EPCs. Cav-1knockdown also significantly increased migration of db/db EPCs.5. Cav-1knockdown increased phosphorylation of eNOS and NO productionEPCs were incubated in EGM-2supplemented with VEGF (50ng/ml) for30minutes, then the medium was collected for NO detection and cells were harvested for protein analysis. The eNOS phosphorylation at Ser1177shown by immunoblotting was significantly decreased in db/db EPCs compared with that in db/+EPCs. This reduction in eNOS phosphorylation was associated with a decrease in EPC-derived NO production. Nevertheless, Cav-1knockdown significantly increased eNOS phosphorylation by88.72%in db/db EPCs coupled with significantly augmented NO production.Conclusions1. High glucose upregulated Cav-1protein expression independent of high leptin level.2. Cav-1knockdown in db/db EPCs improved the angiogenic function.3. Cav-1knockdown increased phosphorylation of eNOS and NO production in db/db EPCs BackgroundCaveolae are50-100nm flask-shaped invaginations of the plasma membrane which are implicated in cholesterol transport, signal transduction. endocytosis. Caveolin, an important protein component of caveolae membrane coats, is known to have three isoforms which are caveolin-1(Cav-1), Cav-2and Cav-3. Cav-1has been widely and intensively studied, especially after the generation of Cav-1knockout mice. The putative importance of Cav-1in cardiovascular system have been validated in several physiological and pathological models, in which Cav-1influences vascular permeability, angiogenesis and progress of atherosclerosis. Notably, endothelial expression of Cav-1is closely involved in these conditions, and both lower and higher levels of Cav-1expression may impair vascular functions.Mature endothelial cells possess limited regenerative capacity, therefore, there is growing interest into circulating endothelial progenitor cells (EPCs). Not only have EPCs been used to diagnose and predict cardiovascular diseases, EPCs have also served as a useful tool for cell and gene therapy in ischemic diseases. Though Cav-1has been well investigated in endothelial cells, there is few reports concerning its role in EPCs. Sbaa et al. reported that Cav-1knockout mice presented impaired EPC mobilization from bone marrow, which resulted in defective neovascularization in a hindlimb ischemia model. This study suggested that Cav-1might play an important role in EPC functions. In current study, we would investigate the impact of Cav-1expression on EPC functions. Our previous study showed EPCs isolated form type2diabetic db/db mice presented impaired angiogenic functions, which caused deficient new vessel formation in skin repair, and subsequently delayed wound healing. This suggests EPC is an essential opponent participating in wound healing. Herein, we mainly examined the effects of Cav-1expression in EPCs on neovascularization and cutaneous wound healing. To address this question, Cav-1KO mice and Cav-1KO mice reconstituted with a transgene expressing Cav-1specifically in endothelial cells (Cav-1RC mice) were used. We assessed neovascularization and wound healing in WT, Cav-1KO and RC mice, and examined whether reconstitution of Cav-1regulates wound healing via EPC mediated vasculogenesis.It is well known that stromal cell-derived factor-1alpha (SDF-1α) is a predominant chemokine which is upregulated in response to ischemic stimuli and acts as a homing signal for EPCs. SDF-1α gene transfer has become a novel chemokine therapy for ischemic disease and tissue repair via recruiting EPCs to ischemic sites. It was proved that SDF-1α treatment accelerated wound closure in diabetes, while inhibition of SDF-1α further impaired diabetic wound healing, and the main mechanism was closely related with EPC mediated vasculogenesis. It was found in Cav-1KO mice plasma level of SDF-1α was2-fold higher than WT mice, which suggested Cav-1might regulate SDF-1α. Based on the evidence that SDF-la therapy effectively improves the injury repair in several models, our study also determined whether SDF-la therapy achieved by SDF-1α/AAV treatment enhances wound healing in Cav-1KO and RC mice, and whether the effects of SDF-la is Cav-1dependent.Objectives 1. To determine the role of Cav-1in EPC function;2. To investigate the role of Cav-1in cutaneous would healing3. To investigate the role and mechanisms of EPCs’ Cav-1in SDF-1α enhanced neovascularization and would healing;Methods1. AnimalsAll animal studies were performed with approval of the University of Pittsburgh Institutional Animal Care and Use Committee. The generation of Cav-1KO mice and Cav-1RC mice was previously described. All KO and RC mice were supplied by Dr. Michael Bauer. The WT mice (B6129SF2/J) were purchased from The Jackson Laboratory. Male mice with8-12weeks old were used in this study.2. Full-thickness excisional woundsMice were anesthetized using isoflurane and then depilated with an electric shaver and depilatory creamin dorsum. Mouse skin was cleaned by Betadine and70%alcohol. Then full-thickness skins, containing epidermis and dermis, were removed by a6-mm punch. Wounds were covered with a bioclusive transparent oxygen permeable wound dressing. Wound closure rate was measured every other day till day16on which WT mice were almost cured completely.4. EPC function assaysThe tube formation assay and in vitro wound healing assay were performed to determine the angiogenic function of EPCs. For tube formation assay,5×104EPCs were plated in a well of48-well plate pre-coated with growth factor-reduced Matrigel. After24-hour incubation, images were taken by an inverted microscope and tube lengths were measured. For in vitro wound healing assay, EPCs ware seeded at1.5×105cells/well into a24-well plate. After24-hour incubation, EPCs were almost confluent, and scratched using a10μl pipette tip. After being washed3times using PBS, EPCs were kept on culturing. Images were taken at the time of wounding0h,12h and24h. Migration was estimated by counting cell numbers within wounded region.5. Enzyme-linked immunosorbent assay (ELISA) SDF-1α protein levels in serum and bone marrow were determined by using the mouse SDF-1α Quantikine ELISA Kit based on the manufacturer’s instructions. Peripheral blood was obtained by cardiac aspiration and then centrifuged to collect serum. BM was obtained from femurs, tibias and pelvis. The bones were flushed in500μl PBS. After centrifugation, the supernatant was collected for SDF-la ELISA and total protein quantification.6. SDF-la AAV topical treatmentAAV vectors carrying SDF-1α gene has been used to treat wounds in mice. In this study, human SDF-la/AAV and control vector, green fluorescent protein (GFP)/AAV, which were gifts from Dr. Yanfang Chen (Wright State University, Ohio), were constructed by inserting the human SDF-la or GFP gene into AAV2/9vector. After wounds were created,100μl AAV at1012particle/ml in PBS was injected into the wound beds. Mice were maintained under anesthesia for30minutes, and then a bioclusive dressing was placed over the wound. Wounds were measured every other day till day16.7. Statistical AnalysisData were presented as means±SEM. Differences between two groups were compared by Student’s t-test, and one-way ANOVA with Newman-Keuls multiple comparison tests was used for multiple groups. The rate of wound healing among groups was analyzed by2-way ANOVA, followed by Bonferroni’s correction to control type I error. In all the tests, a value of P<0.05was taken as statistically significant.Results1. Reconstitution of Cav-1restores tube formation of Cav-1KO EPCsCav-1KO EPCs formed significantly fewer networks than WT EPCs, and relative tube length decreased by nearly50%. And this impairment was completely restored in Cav-1RC group by reconstitution of Cav-1in EPCs. In vitro wound healing assay showed these three groups presented similar migrating pattern which was analyzed at the time point of12hours and24hours after creating wound.2. Reconstitution of Cav-1in endothelium is insufficient to restore impaired wound healing in global Cav-1knockout miceThese three strains of mice underwent a single6-mm dorsal punch biopsy to create a full-thickness excisional wound, and then the rate of wound closure was measured every other day till day16. Cav-1KO mice exhibited slower healing rate than WT mice, which was significant from day6to day16. Nevertheless, reexprcssion of Cav-1in endothelium failed to rescue the impaired wound healing in global Cav-1knockout background, and exhibited similar recovery pattern with Cav-1KO mice. At day16, wounds in WT mice were almost entirely closed, whereas unhealed in Cav-1KO and Cav-1RC mice (89.02%and90.25%respectively).3. Wounded skins in Cav-1KO and RC mice showed decreased new vessel formationWe wondered whether deficient neovascularization led to impaired wound healing in Cav-1KO and RC mice, so we measured the vessel number in skins16days after wound creation. As showed by immunofluorescence staining, the percentage of vessel density in Cav-1KO and RC mice wounded skin were32.64%and38.03%respectively compared with that of WT skin. The impaired neovascularization in wounded skin might lead to the delayed wound healing in Cav-1KO and RC mice.4. Topical treatment of SDF-1α AAV accelerated wound healing in Cav-1RC mice coupled with increased neovascularization in skinFollowing wounding,1011viral particles of SDF-1α/AAV in100μl PBS was injected onto wound base for30minutes in WT, Cav-1KO, and Cav-1RC mice, while same amount of GFP/AAV was administered to control groups. During16-day observation, Cav-1RC mice showed enhanced wound healing rate and were restored to the WT level, however. Cav-1KO mice didn’t respond to the treatment. We next measured the neovascularization of wounded skins treated with SDF-1α/AAV or GFP/AAV. Results showed Cav-1RC mice treated with SDF-1α/AAV had more new vessel formation than those treated with GFP/AAV, while Cav-1KO mice showed no response to SDF-la treatment.Conclusions1. Expression of Cav-1is essential to the angiogenic function of FPCs. 2. Cav-1is required for cutaneous wound healing, and reconstitution only in endothelium is insufficient to restore the impaired wound healing in Cav-1KO mice.3. SDF-1α improved neovascularization and wound healing, in which expression of Cav-1in EPCs played an essential role. |
PDF Full Text Request