Role of lipotoxicity in vascular and β-cell damage in type 2 diabetes

Abstract

Background. Diabetes is a complex disease which currently affects 425 million people worldwide and type 2 diabetes (T2D) accounts for 80-90% of all cases. Development and progression of T2D is caused by insulin resistance (IR) and pancreatic β-cell failure, the latter due to dysfunction or destruction of β-cells. Nevertheless, the most prevalent cause of mortality in T2D patients is represented by cardiovascular (CV) disease which occurs with two-three times higher rate in these subjects than in adults without diabetes. Sodium-glucose cotransporter 2 inhibitors (SGLT2-I) – in particular empagliflozin, dapagliflozin and canagliflozin – belong to a recently introduced class of anti-diabetic drugs capable to improve both hyperglycemia and CV risk. The mechanisms underlying SGLT2-I-mediated CV protection are still unclear but, among other hypotheses, it has been proposed the inhibition of the sodium/H+ exchanger (NHE) as putative non-canonical mediator of SGLT2-I-mediated effects. A key phenomenon in the pathogenesis of both IR, β-cell dysfunction and CV risk, is represented by lipotoxicity, which consists in the deleterious effects caused by elevated free fatty acid (FFA) levels. The present study focuses on the possible effects of lipotoxicity on two cell types directly involved in (1) vascular complication, with specific focus on impaired reparatory mechanisms, and (2) progression of β-cell dysfunction. As a cell population with a pivotal role in endothelial repair, lipotoxicity has been studied on myeloid angiogenic cells (MACs), which are a subset of cells with pro-angiogenic function. Importantly, a reduction in MAC number and function is associated with both T2D and increased CV morbidity and mortality. Aims. (1) to investigate the effects of lipotoxicity – specifically mediated by physiological concentrations of the saturated stearic acid (SA) – on MAC viability and function and the capacity of novel anti-diabetic drugs to curb SA-induced lipotoxicity in MACs and (2) to improve the in vitro protocol of differentiation of iPSCs into functional β-cells to create novel cell models to study β-cell lipotoxicity. Results. (1) SA induces lipo-apoptosis in MACs in a dose- and time-dependent manner. Moreover, SA triggers pro-inflammatory cytokines and chemokines and endoplasmic reticulum (ER) and oxidative stress marker gene expression at physiological concentration (100 μM) in MACs. Interestingly, JNK activation has been found to mediate pro-inflammatory response, whilst pro-apoptotic response seems to be mediated by PERK signaling in ER stress response. Of note, SA exposure affects angiogenic function in MACs. At the maximum concentration tested (100 μM), both empagliflozin and dapagliflozin curb SA-induced expression of pro-inflammatory and oxidative stress markers, restoring baseline values. NHE isoform 1-6 and 9, but not SGLT2, have been detected in MACs. Amiloride (an aspecific NHE inhibitor), and only partially cariporide (NHE1 specific inhibitor), mimic SGLT2-I mediated anti-inflammatory effects on SA-treated MAC, supporting the hypothesis of an involvement of NHE blocking, independent of SGLT2 inhibition, in SGLT2-I-mediated anti-lipotoxic action. (2) To date, findings on lipotoxicity in β-cells have been hampered by the difficulty to study the disease tissue, i.e. limited availability and high variability of islet preparations. An alternative solution to human pancreatic islets is represented by the human β-cell line EndoC-βH1. In our experience, pilot experiments show that EndoC-βH1 cells are only moderately susceptible to lipoapoptosis induced by physiological concentration of SA and palmitic acid (PA), although higher concentration (500 μM) succeeds to be pro-apoptotic. However, in other experiments - led by our collaborator Prof Scharfamann (Paris) - EndoC-βH1 cells are rather resistant to similar concentration of SA and PA and such protection seems to be mediated by the high expression of stearoyl-CoA desaturase (SCD), a key enzyme involved in the synthesis of unsaturated from saturated FFAs. Given our contrasting data and considering that EndoC-βH1 are transformed pseudodiploid cells in continuous expansion and, as a consequence, they should not be considered as a direct equivalent of primary β-cells, we can conclude that EndoC-βH1 cells might not represent the best model to study β-cell lipotoxicity. In the present project, β-cells differentiated from human induced pluripotent stem cells (iPSCs) have been identified as the optimal tool to study β-cell lipotoxicity. Using a 7-stage protocol which mimics the embryonic development of the endocrine pancreas, iPSCs have been differentiated into β-cells. At the end of the process, the yield of insulin-positive β-cells is comparable to that of human islets. In iPSC-derived β-cells at stage 7, high glucose stimulation slight increases insulin release, which is further augmented in response to high glucose plus forskolin (that acts by raising intracellular cAMP levels). Next, iPSC-derived β-cell transplantation, under the kidney capsule of NOD-SCID mice, further boosts their functional maturation. Mice transplanted with iPSC-derived β-cells show an increase in human C-peptide in response to glucose injection and succeed in maintain normoglycaemia after streptozotocin injection, compensating the rodent β-cells ablation. Their dynamic responsiveness has also been confirmed by in situ kidney perfusion assay. Among the advantages of iPSC-derived β-cells, it should be recognized that iPSCs are a patient-relevant cell model, as they are directly reprogrammed from patient’s cells. In this regard, iPSCs derived from fibroblasts of a patient affected by Friedreich ataxia – which is an autosomal recessive neurodegenerative disease caused by an intronic repeat expansions in FXN gene, characterized by reduced expression of the frataxin protein and with a high prevalence of diabetes – have been differentiated in β-cell by using the protocol described above. The differentiation has been highly efficient, resulting in 58% insulin-positive β-cells and slight glucose-responsive cells. Frataxin levels in these iPSC-derived β-cells result lower than those of healthy control cell lines. Then, this patient-relevant model of β-cells has been exploited to evaluate the effect of an incretin analog, which was found to mildly increase frataxin expression. This study demonstrates the feasibility and the advantages of differentiate iPSCs into functional iPSCs and suggests their exploitation for investigating β-cell dysfunction. Conclusion. This project points to a critical role of lipotoxicity, specifically induced by SA, in driving the development of T2D and of its major complication, i.e. CVD.