Abstract
Background. The impact of cysteine-rich angiogenic inducer 61 (Cyr61, also called CCN1) on endothelial progenitor cells (EPCs) from diabetic-rat-derived whole peripheral and bone marrow remains poorly understood. Therefore, the expression levels of CCN1, CCN1-induced C-X-C chemokine receptor type 4 (CXCR4), and stromal-cell-derived factor-1 (SDF-1) were explored under high glucose (HG) conditions.
Objectives. The aim of the study was to explore the effects of high CCN1 levels on EPC activity in diabetic rats through mitogen-activated protein kinase kinase (MEK)/extracellular signal-regulated kinase (ERK) pathway modulation.
Materials and methods. Primary EPCs were isolated from bone marrow and whole peripheral blood of streptozocin (STZ)-induced diabetic Sprague–Dawley rats and controls. Cell migration, tube formation ability and viability were determined using transwell, Cell Counting Kit-8 (CCK-8), and Matrigel®-based capillary-like tube formation assays. Protein and gene expression levels were measured by western blot and real-time quantitative polymerase chain reaction (RT-qPCR).
Results. The study findings showed that EPC migration, viability and tube formation ability were significantly lower under HG conditions. High CCN1 expression levels restored EPC function by inducing SDF-1 and CXCR4 in EPCs under HG conditions. Furthermore, HG suppressed MEK/ERK phosphorylation, while an ERK1/2 agonist rescued EPC CCN1-SDF-1/CXCR4 expression under HG conditions through the activation of the MEK/ERK pathway.
Conclusions. This study demonstrates that high CCN1 expression levels restored EPC functions, partly by modulating MEK/ERK signaling. These findings provide a basis for developing novel therapeutic methods for diabetic vascular neogenesis and vascular injury repair.
Key words: CXCR4, EPCs, CCN1, high-glucose MEK/ERK signaling pathway, SDF-1
Introduction
Cysteine-rich angiogenic inducer 61 (Cyr61, also called CCN family member 1 or CCN1) is a 40-kDa signaling protein localized in the extracellular matrix (ECM) and encoded by the CYR61 gene in humans.1 The CCN1 mediates inflammation during inflammatory responses, implying that its expression could be modulated to manage acute lung injury.2 In addition, CCN1 regulates several cellular processes, such as adhesion, apoptosis, proliferation, differentiation, and senescence, by interacting with heparan sulfate proteoglycans and cell surface integrin receptors. The CCN1 also plays a vital role in vascular integrity, blood vessel formation and cardiac septal morphogenesis in the placenta and during embryonic development.3, 4 In adults, CCN1 is involved in tissue repair and inflammation, and is associated with diseases involving persistent inflammation, such as retinopathy, diabetes-related nephropathy, atherosclerosis, rheumatoid arthritis, and various cancers.5
Neovascularization is impaired in diabetes mellitus, leading to peripheral artery disease attributed to endothelial progenitor cell (EPC) dysfunction.6 Endothelial progenitor cells infiltrate injured vessels from the bone marrow, participate in neovascularization, and promote endothelial regeneration, playing a significant role in angiogenesis.7 The levels of recruited EPCs and their function are significantly lowered under diabetic conditions,8, 9 and EPC dysfunction in diabetes might result in cardiovascular complications and defective angiogenesis. Low EPC levels in the blood and the EPC dysfunction are involved in diabetic vascular conditions. Notably, EPC transplantation restores their function and induces angiogenesis after hind limb ischemia in diabetic mice. Diabetic EPCs are characterized by impaired adhesion and proliferation and have a deformed morphology compared to nondiabetic EPCs,10 though studies have yet to explore the mechanisms behind diabetes-induced EPC impairment.
The CCN1 is a regulator of angiogenesis, involved in reducing EPC levels and functions under high glucose (HG) conditions.11 Furthermore, atherosclerotic plaques have high CNN1 expression levels that initiate cerebrovascular, cardiovascular and peripheral arterial diseases.12 The CCN1 is a significant genetic regulator in coronary artery disease (CAD) and plays a role in protecting the murine heart against ischemia reperfusion injury.13 The C-X-C chemokine receptor type 4 (CXCR4) and stromal cell-derived factor 1 (SDF-1 or CXCL12) are essential CCN1 targets during the modulation of angiogenesis, cell proliferation and energy metabolism.14 The SDF-1 is overexpressed in injured tissues, and the SDF-1/CXCR4 pathway modulates hematopoiesis, wound healing, angiogenesis, and progenitor homing.15 Moreover, the SDF-1/CXCR4 axis is involved in EPC recruitment into ischemic tissue during angiogenesis.16 Kawakami et al.17 showed the involvement of the SDF-1/CXCR4 pathway in EPC mobilization and incorporation in fracture healing.
The EPCs isolated from diabetic animals exhibit altered migration in response to CCN1, and the migration ability of EPCs to SDF-1 binding is defective in patients with type 1 or type 2 diabetes.18 Recently, EPC transplantation has been found to improve limb function in diabetic mice with unilateral hind limb ischemia by restoring local blood flow through the overexpression of CCN1.11 Yet, it is unknown whether the CCN1 pathway plays a role in the mechanisms of EPC therapy in diabetes.
The mitogen-activated protein kinase kinase (MEK)/extracellular signal-regulated kinase (ERK) pathway (Ras-Raf-MEK-ERK pathway) mediates communications from the cell surface to the nucleus. The EPC-induced activities, such as capillary sprouting, are significantly reduced by the inhibition of MEK/ERK signaling.19 The CCN1 inhibition using an anti-CCN1 antibody suppressed MEK and ERK phosphorylation in acute myeloid leukemia (AML) cells.20 A study on type 2 diabetic patients with aberrant EPC number, migration and nitrogen oxide synthase (NOS) activity showed that these phenotypes were associated with low SDF-1/CXCR4 expression levels and downregulation of the MEK/ERK signaling pathway.21 These findings provide a basis for the potential mechanisms underlying hyperglycemia-impaired EPC migration in type 2 diabetes mellitus.
Herein, the impact of HG on EPC function was explored in diabetic rats. In addition, the role of CCN1 in MEK/ERK signaling-dependent modulation of EPC activity was evaluated.
Objectives
The study aimed to explore the impact of CCN1 on EPCs using whole peripheral blood and bone marrow samples derived from diabetic rats. The expression levels of CCN1, CXCR4 and SDF-1 were evaluated under HG conditions, and the effects of high CCN1 levels on the MEK/ERK pathway-dependent modulation of EPC activity were assessed in diabetic rats.
Materials and methods
Animals
Our animal experiments were in accordance with the guidelines of the Animal Care Committee of Sir Run Run Shaw Hospital (approval No. SYXK 2017-0006). Sprague–Dawley rats (7–8-week old males) were obtained from Shanghai S&P (Shall Kay Laboratory Animal Co., Ltd., Shanghai, China). The rats were housed in a humidity- and temperature-controlled environment, with free access to water and food, and were separated into 2 groups of 3 animals each (Figure 1). Group 1 was a control group and was administered a standard laboratory diet (SLD). Group 2 received an intraperitoneal injection of 45 mg/kg streptozotocin (STZ) in 50 mM of sodium citrate buffer, pH 4.5, and was fed a SLD. Blood glucose levels were determined 3 weeks after STZ administration using a glucometer. The rats exhibiting non-fasting blood glucose (non-FBG) levels ≥11.1 mmol/L were considered diabetic.
Endothelial progenitor cell isolation, culture and identification
All healthy rats were euthanized by cervical dislocation, and the femur and tibia were harvested aseptically for EPC isolation. A 20-mL peripheral blood and bone marrow cell suspension was prepared, and mononuclear cells were isolated using a separation medium (Lonza, Basel, Switzerland). Samples were washed twice with phosphate-buffered saline (PBS) (Josonbio, Shanghai, China), and the cells were resuspended in a complete medium supplemented with 10% fetal bovine serum (FBS) (Gibco, Baltimore, USA). Then, 10 mg/L of basic fibroblast growth factor (FGF) and 10 mg/L of vascular endothelial growth factor (VEGF) (both from TBD Science, Tianjin, China) were added to the cells. The cells were seeded on 25 cm2 flasks at 5×105 cells/mL and cultured in a humidified incubator at 37°C with 5% carbon dioxide (CO2). The medium was replaced every 4 days to remove non-adherent cells and maintained for 2 weeks. The EPC phenotype was explored using fluorescence microscopy by double-positive staining for fluorescein isothiocyanate-Ulex Europaeus Agglutinin-1 (FITC-UEA-1) binding (green) and Dil acetylated low-density lipoprotein (Dil-ac-LDL) (red). Glucose was added to the EPC cultures at 5 or 22 mM to mimic normal glucose (NG) and HG, respectively.
Recombinant adenovirus
A first-generation adenovirus was regulated by the cytomegalovirus promoter using an open reading frame (ORF) shuttling system (Vigene Biosciences, Inc., Rockville, USA). Rat CCN1 complementary deoxyribonucleic acid (cDNA) (NM_024359) (Vigene Biosciences, Inc.) was harvested from the pENTR vector and transferred to a pAD-ORF transfer vector. Recombinant adenoviral constructs were transfected into 293 cells to obtain a recombinant adenovirus with a high Ad-CCN1 titer (1×1010 plaque-forming unit (pfu)/mL). The Ad-CCN1 was obtained using a customized adenovirus (Vigene Biosciences, Inc.). An empty adenovirus was used as the Ad-control. Infection efficiency was determined by labeling all adenovirus vectors with a flag. The effects of adenovirus infection were explored by western blot analysis to verify whether Ad-CCN1 was highly expressed.
Cell viability assay
Cell viability was explored using the Cell Counting Kit-8 (CCK-8) assay. The EPCs were seeded to flat-bottomed 96-well microplates at a density of 1×104 cells/well. Then, the cells were incubated for 24 h in an endothelial basal medium-2 (EBM-2; Lonza, Walkersville, USA) containing 2% FBS (6 wells per group). The culture medium was replaced with EBM-2 medium containing 10% FBS and cultured for 6 h. The cells were then cultured in HG media (22 mM of glucose for 24 h). Control cells were not treated. The CCK-8 (10 μL/well) was added to the wells at the end of the experiment. The absorbance was measured on a microplate reader at 450 nm after incubation at 37°C for 48 h, and the proliferation of the treated EPCs was determined relative to the control EPCs.
Cell migration assay
The EPC migration rate was evaluated using transwell assays with 8-mm pore filters (Corning Inc., Corning, USA). After treating EPCs with HG or overexpressing CCN1, 5×104 cells in 100 µL of non-FBS EBM-2 medium were seeded into the upper chamber, and 500 µL of culture medium with 20% FBS was added to the lower chamber. The EPCs were cultured for 24 h, then the transwell membranes were stained with 0.1% crystal violet for 30 min. The number of migrated cells was determined in 3 random fields of view using an IX51 inverted fluorescence microscope (Olympus, Tokyo, Japan). All experiments were performed in triplicate.
Matrigel-based capillary-like tube formation assay
The impact of HG on the capillary-like tube formation capacity of EPCs was determined using the capillary tube formation angiogenesis assay kit (ECM625; Merck Millipore, Burlington, USA). A total of 1×104 cells/well were seeded onto Matrigel-precoated 96-well plates (Corning Inc.) after treatment of EPCs with HG for 24 h. An IX51 inverted light microscope (Olympus) was used to observe tube formation. Three independent fields were measured per well, and the average number of tubes was determined using WimTube quantitative image analysis (Wimasis, Cordoba, Spain).
Semi-quantitative real-time polymerase chain reaction
Total ribonucleic acid (RNA) was extracted using TRIzol (Invitrogen, Carlsbad, USA), and 0.5 μg of RNA was used for reverse transcription using the PrimerScript™ RT reagent kit (Takara, Shiga, Japan). Samples were incubated for 1 h at 42°C, and the reaction was terminated at 70°C for 15 min. The CCN1 forward primer (5’-TCA CCC TTC TCC ACT TGA CC-3’) and reverse primer (5’-AAT TGC ATT CCA GCC CCT TG-3’), and the β-actin forward primer (5’-CAC GAT GGA GGG GCC GGA CTC ATC-3’) and reverse primer (5’-TAA AGA CCT CTA TGC CAA CAC AGT-3’) were used for polymerase chain reaction (PCR) amplification. The PCR products were separated and analyzed using agarose gels, and the CCN1 bands were confirmed through sequencing.
Real-time polymerase chain reaction
Total RNA was extracted using TRIzol (Invitrogen) and retrotranscribed with the PrimeScript 1st strand cDNA synthesis kit (D6110A; Takara Bio USA, Inc., San Jose, USA). The real-time (RT) PCR assay used SYBR Green Real-time PCR Master Mix (QPK-201, QPK-201T; Toyobo, Osaka, Japan) under the following conditions: 95°C for 30 s, followed by 40 cycles at 95°C for 5 s and 60°C for 30 s. The analysis was performed in triplicate, and the gene expression levels were determined using the 2−ΔΔCT method. The CCN1 forward primer was 5’-TTG TAG GCA CGG CTG CTA TGC T-3’, the CCN1 reverse primer was 5’-GGT GCT CCA TTC TCA GAA CCT G-3’, the SDF-1 forward primer was 5’-GGA GGA TAG ATG TGC TCT GGA AC-3’, the SDF-1 reverse primer was 5’-AGT GAG GAT GGA GAC CGT GGT G-3’, the CXCR4 forward primer was 5’-GAC TGG CAT AGT CGG CAA TGG A-3’, the CXCR4 reverse primer was 5’-CAA AGA GGA GGT CAG CCA CTG A-3’, the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) forward primer was 5’-AGA CAG CCG CAT CTT CTT GT-3’, and the GAPDH reverse primer was 5’-CTT GCC GTG GGT AGA GTC AT-3’.
Western blotting
The EPCs were lysed with a lysis buffer containing 100 mM of phenylmethanesulfonyl fluoride, and the proteins were resolved using 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). The proteins were transferred onto polyvinylidene difluoride (PVDF) membranes and incubated with primary antibodies against CCN1 (1:1000; Life Technologies, Carlsbad, USA), SDF-1 (1:1000; Life Technologies), CXCR4 (1:1000; Life Technologies), ERK1/2 (1:1000; Santa Cruz Biotechnology Inc., Dallas, USA), p-ERK1/2 (1:1000; Santa Cruz Biotechnology Inc.), ERK (1:1000; Santa Cruz Biotechnology Inc.), p-ERK (1:1000; Santa Cruz Biotechnology Inc.), and GAPDH (1:3000; Santa Cruz Biotechnology Inc.). The western blot signal was developed using an enhanced chemiluminescence detection kit (Merck Millipore).
The first experiment was considered a pre-experiment. Based on the results and their reliability, the second experiment was conducted with the same design and sample selection criteria to ensure stronger credibility of the experimental results. Using successfully isolated, cultured, and identified EPCs in the first experiments, all experiments were repeated twice, including cell migration, tube formation ability and viability. Protein and gene expression levels were measured using western blot and real-time quantitative polymerase chain reaction (RT-qPCR) tests, respectively (detailed experimental procedures were described previously). Finally, data from the 2 experiments were combined for statistical analyses.
Statistical analyses
Statistical analyses employed SPSS v. 21.0 (IBM Corp., Armonk, USA) software. One-way analysis of variance (ANOVA) and multiple comparison post-hoc tests (Bonferroni) compared means between experimental groups. Due to the small sample size, we used a bootstrap ANOVA. The differences between the 2 groups were determined using Student’s t-test, with p-values ≤0.05 considered statistically significant (* p < 0.05, statistically significant difference; ** p < 0.01, very significant difference; *** p < 0.001, extremely significant difference). In all figures, the long horizontal line in the middle indicates the sample mean of the group, the range between the short horizontal lines at both ends indicates the 95% confidence intervals, and the points indicate the within-sample distribution.
Results
Endothelial progenitor cell isolation, establishment and characterization
Bone marrow EPCs (BM-EPCs) and whole peripheral blood EPCs (PB-EPCs) were isolated from normal and STZ-treated diabetic rats. The cells were passaged and characterized, as described by Brandl et al.22 Then, they were centrifuged using Percoll density gradient centrifugation and cultured for 2 weeks. The EPC phenotype was confirmed by staining with Dil-ac-LDL (red) and FITC-UEA-1 (green) and observed under fluorescence microscopy (model BX53; Olympus Corp., Tokyo, Japan) (Figure 2A). After less than 24 h in culture, cells appeared small and round. The EPCs exhibited a spindle-shaped morphology after 4–7 days and developed cord-like structures after 10 days in culture (Figure 2B). Additionally, after 10 days, significantly more control PB-EPCs were positive for FITC-UEA-1 (72.80 ±1.92%) compared to diabetic PB-EPCs (56.80.3 ±4.55%, p ≤ 0.05). Consistently, significantly more control BM-EPCs (70.00 ±1.58%) were positive for Dil-ac-LDL and FITC-UEA-1 compared to diabetic BM-EPCs (57.00 ±2.92%, p ≤ 0.05). Overall, diabetic rats had significantly fewer EPCs (Figure 2C and Supplementary Table 1).
High glucose affected endothelial progenitor cell migration, viability and capillary-like tube formation
Angiogenesis involves EPC activation, proliferation, migration, and capillary-like tube formation. The impact of glucose on EPC activity, viability, tube formation, and migration was explored under NG (5 mM of glucose) and HG (22 mM of glucose) conditions. The CCK-8 assay determined EPC viability, which was significantly lower in diabetic rats compared to normal rats (Figure 3). In addition, EPC levels were lower in the bone marrow and peripheral blood samples cultured in the HG medium than in the NG medium (all p < 0.05, Figure 3A and Supplementary Tables 2,3).
Angiogenesis is influenced by effective migration. Therefore, the impact of HG on EPC migration was explored using transwell assays. High glucose significantly decreased EPC migration in diabetic rats compared to the control rats. The migration rate was markedly lower in the HG medium compared to the NG medium in the bone marrow and peripheral blood samples (Figure 3B,C and Supplementary Tables 4,5).
The inhibitory effect of HG on EPC tube formation was explored. The HG medium suppressed EPC tube formation, as shown by a marked decrease in segment length, mean tube length and junction numbers (Figure 3D). These findings indicate that HG inhibits EPC tube formation, implying that it can play a suppressive role in EPC neo-angiogenesis.
High glucose downregulated CCN1, SDF-1 and CXCR4 mRNA expression by EPCs
The CCN1 messenger RNA (mRNA) and protein expression levels were determined after culturing EPCs in HG medium using western blotting, RT-qPCR and semi-quantitative RT-PCR (sqRT-PCR). Furthermore, the protein levels of EPC targets (SDF-1 and CXCR4) were determined after culture in the HG medium. The sqRT-PCR and RT-qPCR analyses showed that HG significantly suppressed the expression of CCN1 (Figure 4A,B and Supplementary Table 6). The CCN1 protein levels in whole peripheral blood and bone marrow were significantly downregulated in EPCs from diabetic rats and HG medium (Figure 4C,D and Supplementary Table 7). Western blot showed significant decreases in the levels of CCN1 targets (SDF-1 and CXCR4) (Figure 4E–G and Supplementary Tables 8,9). These findings imply that CCN1 inhibition and its target genes, SDF-1/CXCR4, are involved in HG-induced EPC dysfunction.
CCN1 overexpression ameliorated high-glucose impairment of EPC viability, migration and capillary-like tube formation
The findings outlined above showed that HG correlated with low CCN1, SDF-1 and CXCR4 expression levels, which are known neovascularization inducers in EPCs. Recombinant adenoviral particles containing the CCN1 ORF controlled by a constitutive CMV promoter explored whether CCN1 could alleviate the negative impact of HG on EPC functions. Cell migration, viability and tube formation were assessed, as detailed above, for BM-EPCs under NG or HG conditions after adenovirus infection for 48 h. Notably, the inhibition of cell viability and migration in BM-EPCS under HG (Figure 5A–C and Supplementary Tables 10–13) was reversed by the overespression of CCN1. In addition, BM-EPC tube formation impairment under HG was improved by the upregulation of CCN1 (Figure 5D). The RT-qPCR and western blot analysis revealed that the upregulation of CCN1 rescued the reduced SDF-1 and CXCR4 mRNA and protein levels triggered by HG (Figure 5E–G and Supplementary Tables 14,15). The CCN1 mRNA (Figure 5E) and protein (Figure 5F,G) levels were upregulated after adenoviral infection and normalized by HG medium. These findings show that the overexpression of CCN1 improved EPC tube formation ability, migration and viability after HG treatment. The results imply that HG-induced CCN1 downregulation, leading to CCN1-mediated SDF-1/CXCR4 suppression, could be involved in neovascularization in EPCs of diabetic rats.
High-glucose exposure decreased CCN1 expression levels in EPCs from diabetic rats by suppressing the MEK/ERK pathway
We explored whether CCN1 plays a role in alleviating the negative effects of HG on EPCs through MEK/ERK signaling. The MEK/ERK protein levels and their phosphorylated forms (p-MEK/p-ERK) under HG and NG conditions were evaluated by western blotting. The phosphorylation of MEK and ERK was inhibited by HG (Figure 6A,B). To evaluate if the rescue effect of CCN1 on EPC dysfunction is mediated through the MEK/ERK signaling pathway, the CCN1 protein levels were measured in the absence or presence of wortmannin, a MEK agonist, with or without HG. Wortmannin markedly restored the HG-reduced p-MEK and p-ERK protein levels. Moreover, HG suppression of CCN1 was restored by the MEK agonist (Figure 6C,D and Supplementary Tables 16–18). These findings indicate that CCN1-mediated attenuation of EPC dysfunction caused by HG in diabetic rats requires MEK/ERK signaling activation, at least partially.
Discussion
Vascular complications are significantly responsible for the morbidity and mortality in diabetes mellitus. Indeed, hyperglycemia is a key factor leading to vascular disorders and results in neovascularization and endothelial dysfunction.23 Endothelial progenitor cells circulating in the blood modulate neovascularization and vascular repair.18 The findings of the present study show that the negative impact of HG on EPCs in diabetic rats is mediated through the suppression of CCN1 expression and its target genes, SDF-1 and CXCR4.14, 24 Furthermore, the overexpression of CCN1 alleviated EPC dysfunction by activating MEK/ERK signaling,25 implying that defective EPCs can be modulated by pharmacological interventions using MEK agonists to restore cell function. In addition, this study showed that CCN1 modulates MEK/ERK signaling in EPCs, potentially by regulating SDF-1 and CXCR4 activity.26 These findings indicate that CCN1 is highly expressed in EPCs and performs its function through the MEK/ERK pathway by modulating SDF-1 and CXCR4 activity.
Currently, the effects of EPCs on vascular dysfunction in diabetes are not well understood. Indeed, there are significant differences in endothelial cell behaviors, vascularization and impaired microvascular processes in the eye,27 which can be partly attributed to the variability of circulating EPC subpopulations involved in the regenerative processes of impaired vascular beds. Available data from type 1 and type 2 diabetic rodent models suggests that diabetic EPCs are not involved in vascular injury repair.28 Notably, endothelial dysfunction precedes atherosclerosis and its clinical manifestations. Thus, approaches aimed at restoring the endothelial cell layer and endothelial functions play crucial roles in maintaining healthy vessels.29 Neovascularization is driven by the accumulation of EPCs at the endothelial injury site, the migration and proliferation of differentiated endothelial cells, and the incorporation of EPCs into the nascent endothelium.30 In animal models, EPCs constitute 25% of all endothelial cells in newly formed vessels.29
The CNN1 has low expression levels under normal conditions and high expression levels in pathological states such as atherosclerosis, colitis, diabetic retinopathy, rheumatoid arthritis, and Graves’ orbitopathy,13, 31 and has been reported to have a causative role in atherosclerosis,31 while CNN1 polymorphisms are associated with the risk of acute coronary syndrome (ACS) in humans.32 Low CCN1 expression levels in rats with carotid balloon injury restore vascular smooth muscle cell (VSMC) proliferation, alleviating vascular intimal hyperplasia.33 In addition, the suppression of CCN1 signaling results in reduced VSMC senescence in the smooth muscle cell layer of the human coronary artery.34 Moreover, CCN1 was reported to affect 30-day mortality in CAD and acute heart failure (AHF) patients, and it could identify myocardial ischemic injury and the clinical progression of ACS.35 Hence, even though the present study reports that the high expression of CCN1 is beneficial for EPC functions, it also appears to be involved in CAD. As such, it could be hypothesized that CCN1 is indeed causative in CAD, but EPCs evolve to respond to increased CCN1 expression. However, this hypothesis needs to be examined in future studies. Furthermore, the relationship between CAD and circulating CCN1 in diabetic and non-diabetic patients has not been fully explored.
Endothelial progenitor cells have a direct role in vasculogenesis and pro-angiogenic factor secretion. The SDF-1, and its receptor CXCR4, are involved in the retention and quiescence of EPCs within their niche in the bone marrow.36 The SDF-1 desensitizes insulin in adipocytes, and its expression is induced in obesity and during fasting. Moreover, SDF-1 plays a role in the chemotactic recruitment of several types of cells, e.g., hematopoietic stem cells and mesenchymal progenitor cells.37 Studies report that the SDF-1/CXCR4 axis modulates fracture healing by fine-tuning of the recruitment and differentiation of progenitor and stem cells at fracture sites.15, 38 The SDF-1 is also involved in the pathogenesis, progression and diverse pathological effects of type 2 diabetes, including adipose tissue inflammation, nephropathy and insulitis.39 Arakura et al.40 reported that SDF-1 and CXCR4 expression and localization at fracture sites showed changes in the course of fracture healing in a diabetic model, probably associated with defective fracture healing and the inhibition of angiogenesis. The dysregulation of the interaction between SDF-1 and CXCR4 is a potential approach for EPC mobilization.
Previous studies report that low SDF-1 expression levels in diabetic wound tissues are involved in impaired diabetic wound healing. The activation of MEK/ERK signaling induces the phosphorylation of the insulin receptor substrate 1 (IRS-1) protein at a serine residue, degradation of IRS-1, glucose uptake, and attenuation of insulin-facilitated protein kinase B (AKT) phosphorylation.21, 41 Furthermore, plasma SDF-1 levels correlate with type 2 diabetes.24 The present study showed that HG reduced CCN1 expression, and the overexpression of CCN1 restored EPC function. The CCN1 overexpression also rescued HG-induced suppression of SDF-1 and CXCR4 protein levels, enhancing EPC viability, migration and tube formation.
Various studies indicate that HG reduces EPC function by impairing MEK/ERK signaling. Indeed, the impairment of early EPCs induced by HG can be restored by modulating the mitogen-activated protein kinase (MAPK) pathway.42 Consistent with these findings, the current study showed that HG downregulated MEK/ERK protein levels and phosphorylation. An ERK1/2 agonist rescued EPC dysfunction caused by HG and restored CCN1-SDF-1/CXCR4 protein levels.43 Yan et al.36 reported that the interaction between SDF-1 and CXCR4 upregulates matrix metalloproteinase 9 (MMP-9) activity and VEGF expression, inducing angiogenesis and tissue regeneration. Furthermore, the binding of SDF-1 to CXCR4 induces the activation of ERK, AKT and mammalian target of rapamycin (mTOR) signaling and promotes cell proliferation and differentiation, resulting in angiogenesis and tissue regeneration and ultimately ordered remodeling. The findings indicate that the CCN1-mediated attenuation of EPC dysfunction due to HG partly requires the activation of MEK/ERK signaling.
Impaired EPCs were observed in type 1 and 2 diabetes, prediabetes, metabolic syndrome, and insulin resistance,44, 45 and were also identified as a cause of vascular complications of diabetes.46 The exact mechanisms of EPC impairment in diabetes remain poorly understood. The reduced numbers of EPCs are thought to be related to decreased mobilization from the bone marrow, decreased proliferation and shorter survival, as observed in the present study. The EPC mobilization is controlled by chemokines such as SDF-1, VEGF, granulocyte-colony stimulating factor (G-CSF), and CXCR4.47, 48 Still, the SDF-1/CXCR4 axis is activated in diabetes,49, 50 indicating that other mechanisms are at play. Insulin resistance is a possible culprit, since the development of insulin resistance and diabetes has been shown to lead to weaker EPCs.9, 51, 52 Knocking down nuclear factor kappa-B (NF-κB), a key player in insulin resistance and diabetes, improved EPC functions.52 Unfortunately, the biomarkers of insulin resistance and genes/proteins involved in insulin resistance were not assessed in the present study, though they will be explored in future research.
Limitations
Endothelial progenitor cells were isolated without differentiating between early and late EPCs,7, 8, 10, 53 which reflects the biological reality more accurately, since both types of cells can be found in animals and humans. Still, future studies could explore possible differences between EPC subtypes. Only a few genes and proteins were examined, preventing the determination of the exact mechanisms involved in EPC functional impairment in diabetes.
Conclusions
High glucose significantly affected the viability, migration and tube formation ability of EPCs. The study shows that the mechanism underlying impaired EPC function is associated with the CCN1-SDF-1/CXCR4-MEK/ERK signaling. These findings indicate that MEK agonists could be used in patients with diabetes mellitus-associated vascular disorders. However, further studies should explore the effects of CCN1 overexpression and ERK agonists in vivo.
Supplementary data
The supplementary materials are available at https://doi.org/10.5281/zenodo.8219770. The package contains the following files:
Supplementary Table 1. Number of EPCs in each group of rats.
Supplementary Tables 2–5. High glucose affects EPC migration, viability and capillary-like tube formation.
Suplementary Tables 6–9. High glucose downregulates CCN1, SDF-1 and CXCR4 mRNA expression by EPCs.
Supplementary Tables 10–15. CCN1 overexpression ameliorates EPCs viability, migration and capillary-like tube formation impairment by high glucose.
Supplementary Tables 16–18. High-glucose exposure decreases expression levels of CCN1 in EPCs from rats with diabetes by suppressing MEK/ERK pathway.
