Type 1 diabetes is associated with increased cardiovascular disease (CVD). Decreased endothelial progenitor cells (EPCs) number plays a pivotal role in reduced endothelial repair and development of CVD. We aimed to determine if cardioprotective effect of metformin is mediated by increasing circulating endothelial progenitor cells (cEPCs), pro-angiogenic cells (PACs) and decreasing circulating endothelial cells (cECs) count whilst maintaining unchanged glycemic control. This study was an open label and parallel standard treatment study. Twenty-three type 1 diabetes patients without overt CVD were treated with metformin for 8 weeks (treatment group-TG). They were matched with nine type 1 diabetes patients on standard treatment (SG) and 23 age- and sex-matched healthy volunteers (HC). Insulin dose was adjusted to keep unchanged glycaemic control. cEPCs and cECs counts were determined by flow cytometry using surface markers CD45dimCD34+VEGFR-2+ and CD45dimCD133−CD34+CD144+ respectively. Peripheral blood mononuclear cells were cultured to assess changes in PACs number, function and colony forming units (CFU-Hill’s colonies). At baseline TG had lower cEPCs, PACs, CFU-Hills’ colonies and PACs adhesion versus HC (p < 0.001-all variables) and higher cECs versus HC (p = 0.03). Metformin improved cEPCs, PACs, CFU-Hill’s colonies number, cECs and PACs adhesion (p < 0.05-all variables) to levels seen in HC whilst HbA1c (one-way ANOVA p = 0.78) and glucose variability (average glucose, blood glucose standard deviation, mean amplitude of glycaemic excursion, continuous overall net glycaemic action and area under curve) remained unchanged. No changes were seen in any variables in SG. There was an inverse correlation between CFU-Hill’s colonies with cECs. Metformin has potential cardio-protective effect through improving cEPCs, CFU-Hill’s colonies, cECs, PACs count and function independently of hypoglycaemic effect. This finding needs to be confirmed by long term cardiovascular outcome studies in type 1 diabetes. Trial registration ISRCTN26092132

Cardiovascular disease is the leading cause of morbidity/mortality worldwide. Metformin is the first therapy offering cardioprotection in type 2 diabetes and non-diabetic animals with unknown mechanism. We have shown that metformin improves angiogenesis via affecting expression of growth factors/angiogenic inhibitors in CD34⁺ cells under hyperglycemia-hypoxia. Now we studied the direct effect of physiological dose of metformin on human umbilical vein endothelial cells (HUVEC) under conditions mimicking hypoxia-hyperglycemia. HUVEC migration and apoptosis were studied after induction with euglycemia or hyperglycemia and/or CoCl₂ induced hypoxia in the presence or absence of metformin. HUVEC mRNA was assayed by whole transcript microarrays. Genes were confirmed by qRT-PCR, proteins by western blot, ELISA or flow cytometry. Metformin promoted HUVEC migration and inhibited apoptosis via upregulation of vascular endothelial growth factor (VEGF) receptors (VEGFR1/R2), fatty acid binding protein 4 (FABP4), ERK/mitogen-activated protein kinase signaling, chemokine ligand 8, lymphocyte antigen 96, Rho kinase 1 (ROCK1), matrix metalloproteinase 16 (MMP16) and tissue factor inhibitor-2 under hyperglycemia-chemical hypoxia. Therefore, metformin’s dual effect in hyperglycemia-chemical hypoxia is mediated by direct effect on VEGFR1/R2 leading to activation of cell migration through MMP16 and ROCK1 upregulation, and inhibition of apoptosis by increase in phospho-ERK1/2 and FABP4, components of VEGF signaling cascades.

Cardiovascular disease (CVD) is the leading cause of morbidity and mortality in patients with diabetes mellitus (DM). To identify the most effective treatment for CVD, it is paramount to understand the mechanism behind cardioprotective therapies. Although metformin has been shown to reduce CVD in Type-2 DM clinical trials, the underlying mechanism remains unexplored. CD34+ cell-based therapies offer a new treatment approach to CVD. The aim of this study was to investigate the effect of metformin on the angiogenic properties of CD34+ cells under conditions mimicking acute myocardial infarction in diabetes. CD34+ cells were cultured in 5.5 or 16.5 mmol/L glucose ± 0.01 mmol/L metformin and then additionally ± 4 % hypoxia. The paracrine function of CD34+ cell-derived conditioned medium was assessed by measuring pro-inflammatory cytokines, vascular endothelial growth factor A (VEGFA), and using an in vitro tube formation assay for angiogenesis. Also, mRNA of CD34+ cells was assayed by microarray and genes of interest were validated by qRT-PCR. Metformin increased in vitro angiogenesis under hyperglycemia–hypoxia and augmented the expression of VEGFA. It also reduced the angiogenic-inhibitors, chemokine (C–X–C motif) ligand 10 (CXCL10) and tissue inhibitor of metalloproteinase 1 (TIMP1) mRNAs, which were upregulated under hyperglycemia–hypoxia. In addition metformin, increased expression of STEAP family member 4 (STEAP4) under euglycemia, indicating an anti-inflammatory effect. Metformin has a dual effect by simultaneously increasing VEGFA and reducing CXCL10 and TIMP1 in CD34+ cells in a model of the diabetic state combined with hypoxia. Therefore, these angiogenic inhibitors are promising therapeutic targets for CVD in diabetic patients. Moreover, our data are commensurate with a vascular protective effect of metformin and add to the understanding of underlying mechanisms.

Human umbilical vein endothelial cell (HUVEC)-based gene expression studies carried out under hypoxia and/ or hyperglycemia bear huge potential in modelling endothelial cell response in cardiovascular disease and diabetes. However, such studies require reference genes that are stable across the whole range of experimental conditions. These reference genes have not been comprehensively defined to date. We applied human genome-wide microarrays and quantitative real-time PCR (qRT-PCR) on RNA obtained from primary HUVEC cultures that were incubated for 24 h either in euglycemic or hyperglycemic conditions and then subjected to short-term CoCl₂-induced hypoxia of either 1, 3 or 12 h. Using whole-transcript arrays, we selected ten commonly used reference genes with no significant expression variation across 8 different conditions. These genes were ranked using NormFinder software according to their stability values. Consequently, five genes were selected for validation by quantitative real-time PCR (qRT-PCR). These were: ribosomal protein large P0 (RPLP0), transferrin receptor (TFRC), glyceraldehyde-3-phosphate dehydrogenase (GAPDH), β-glucuronidase (GUSB), and β-actin (ACTB). All five genes displayed stable expression under hyperglycemia. However, only RPLP0 and TFRC genes were stable under hypoxia up to 12 h. Under hyperglycemia combined with hypoxia up to 12 hours the expression of RPLP0, TFRC, GUSB and ACTB genes remained unchanged. Our findings strongly confirm that RPLP0 and TFRC are the most suitable reference genes for HUVEC gene expression experiments subjected to hypoxia and/or hyperglycemia for the given experimental conditions. We provide further evidence that even commonly known references genes require experimental validation for all conditions involved.

Type 1 diabetes is associated with raised inflammation, impaired endothelial progenitor cell mobilisation and increased markers of vascular injury. Both acute and chronic exercise is known to influence these markers in non-diabetic controls, but limited data exists in Type 1 diabetes. We assessed inflammation, vascular repair and injury at rest and after exercise in physically-fit males with and without Type 1 diabetes. Ten well-controlled type 1 diabetes (27 ± 2 years; BMI 24 ± 0.7 kg.m2; HbA1c 53.3 ± 2.4 mmol/mol) and nine non-diabetic control males (27 ± 1 years; BMI 23 ± 0.8 kg.m2) matched for age, BMI and fitness completed 45-min of running. Venous blood samples were collected 60-min before and 60-min after exercise, and again on the following morning. Blood samples were processed for TNF-α using ELISA, and circulating endothelial progenitor cells (cEPCs; CD45dimCD34+VEGFR2+) and endothelial cells (cECs; CD45dimCD133-CD34+CD144+) counts using flow-cytometry. TNF-α concentrations were 4-fold higher at all-time points in Type 1 diabetes, when compared with control (P < 0.001). Resting cEPCs were similar between groups; after exercise there was a significant increase in controls (P = 0.016), but not in Type 1 diabetes (P = 0.202). CEPCs peaked the morning after exercise, with a greater change in controls vs. Type 1 diabetes (+139 % vs. 27 %; P = 0.01). CECs did not change with exercise and were similar between groups at all points (P > 0.05). Within the Type 1 diabetes group, the delta change in cEPCS from rest to the following morning was related to HbA1c (r = -0.65, P = 0.021) and TNF-α (r = -0.766, P = 0.005). Resting cEPCs and cECs in Type 1 diabetes patients with excellent HbA1c and high physical-fitness are comparable to healthy controls, despite eliciting 4-fold greater TNF-α. Furthermore, Type 1 diabetes patients appear to have a blunted post-exercise cEPCs response (vascular repair), whilst a biomarker of vascular injury (cECs) remained comparable to healthy controls.

Type 1 diabetes (T1DM) is associated with increased cardiovascular disease (CVD) and reduced life expectancy. We thus hypothesized that anti-angiogenic miRs are increased in T1DM, and the cardioprotective effect of metformin is mediated via reducing those miRs. In an open label, case-controlled study, 23 T1DM patients without CVD were treated with metformin for eight weeks (TG), matched with nine T1DM patients on standard treatment (SG) and 23 controls (CG). Plasma miR-222, miR-195, miR-21a and miR-126 were assayed by real-time RT-qPCR. The results were correlated with: endothelial function (RHI), circulating endothelial progenitor cells (cEPCs) (vascular repair marker, CD45^dimCD34⁺VEGFR2⁺ cells) and circulating endothelial cells (cECs) (vascular injury marker, CD45^dimCD34⁺CD133^-CD144⁺ cells). miR-222, miR-195 and miR-21a were higher in T1DM than CG; p = 0.009, p < 0.0001, p = 0.0001, respectively. There was an inverse correlation between logmiR-222 and logRHI (p < 0.05) and a direct correlation between logmiR-222 and logCD34⁺ (p < 0.05) in TG. Metformin reduced miR-222, miR-195 and miR-21a levels in TG; p = 0.007, p = 0.002 p = 0.0012, respectively. miRs remained unchanged in SG. miR-126 was similar in all groups. There was a positive association between changes in logmiR-222 and logcECs after metformin in TG (p < 0.05). Anti-angiogenic miRs are increased in T1DM. Metformin has cardioprotective effects through downregulating miR-222, miR-195 and miR-21a, beyond improving glycemic control.