Gliquidone improves retinal injury to relieve diabetic retinopathy via regulation of SIRT1/Notch1 pathway

Diabetic retinopathy (DR) is a severe microvascular complication for diabetes mellitus (DM) patients, which is a leading cause of vision impairment for working-age adults worldwide [1, 2]. High level of blood glucose is the main contributing factor to affect DR progression, triggering pathological changes of the retinal cells [3]. In the pathogenesis of DR, endothelial dysfunction, inflammation, and oxidative stress are the three main pathological features [4‐6]. Currently, the prevalent treatments for retinal vascular dysfunction such as intraocular injection of anti-vascular endothelial growth factor (VEGF) drugs and steroids have developed rapidly [7‐10]. For example, intraocular glucocorticoids are approved for clinical use to treat diabetic macular edema, and topical ocular treatment has been explored with novel selective glucocorticoid receptor agonists [7]. But some adverse effects including retinal neuron destruction, irreversible retinal damage, and even vision loss are still occurred [8‐10]. Gliquidone (GLI) is an anti-diabetic medication approved for the clinical use and it is classified as a second-generation sulfonylurea [11‐13]. However, researches on the therapeutic efficacy of GLI on DR are relatively rare.

Silent information regulator 2 (SIR2), belonging to the histone deacetylases family, functions as anti-inflammation and oxidation resistance [14]. There are seven homologues of SIR2 named as SIRT1, SIRT2, SIRT3, SIRT4, SIRT5, SIRT6, and SIRT7, respectively. SIRT1 has been confirmed to be deeply involved in the progression of DR [15, 16]. For instance, Jiang et al. found that resveratrol, a specific activator of SIRT1, can remarkably promote the viability of human retinal endothelial cells (HRECs) and repress the secretion of inflammatory cytokines [15]. Ji et al. conducted a DR rat model and demonstrated a decreased expression of SIRT1 in high glucose-induced rat retinal endothelial cells (RRECs) and retinal tissues of DR rats [16]. Meanwhile, they further indicated that overexpressed SIRT1 can accelerate the proliferation of RRECs [16]. Additionally, Notch1 signaling is reported as a downstream effector pathway of SIRT1 to affect the development of human diseases [17‐19]. Notch signaling family is a well-known conserved pathway to regulate inflammatory reactions and oxidative stress, [20] and cellular processes such as proliferation, differentiation, and apoptosis [21]. Meanwhile, Dou et al. believed that Notch signaling plays vital roles in retinal angiogenic sprouting/vasculature maturation/vascular stability [22]. As one of the important receptors of Notch, the activation of Notch1 is confirmed to induce vascular permeability, which increases the risk of DR [23]. More importantly, a latest research revealed that SIRT1 can inactivate Notch1 signaling, eventually alleviating the development of sepsis [24]. Tian et al. constructed a diabetic nephropathy (DN) mouse model and showed that GLI may improve oxidative damage and renal interstitial fibrosis to delay the progression of DN through inhibiting Notch1 signaling [25]. However, whether GLI interacts with SIRT1/Notch1 pathway to affect DR progression is still unclear.

In this study, the function of GLI and its relationship with SIRT1/Notch1 pathway in the progression of DR were preliminarily investigated. These results may provide a clinical therapeutic method for DR.

More than 80% DM patients may suffer from DR at the first two decades of diabetes [34]. Although the management opinions for DR have greatly improved, some severe side effects are still confirmed to affect the therapeutic efficacy [8‐10]. Exploring effective drugs to furthest decrease the side effects in the treatment of DR is extremely urgent. GLI is a well-known hypoglycemic agent in clinic [11]. Previous study has revealed the therapeutic effect of GLI on diabetic nephropathy [25]. But there still no researches on the possible function of GLI on DR progression. This study for the first time uncovers that GLI can relieve DR through regulation of endothelial dysfunction, inflammation, and oxidative stress via SIRT1/Notch1 pathway, suggesting that GLI may be an effective agent for the treatment of DR in clinic.

HG-induced HRECs are generally used as an in vitro model of DR due to the crucial role in angiogenesis [29, 35, 36]. In this study, HG treatment significantly increased the contents of glucose, triglyceride and 2DG6P in HRECs, which is consistent with the data of Long et al. [36]. The results suggested that DR cell model was established successfully. The pathological feature of DR is not only manifested as the increased level of blood glucose, but also the selective apoptosis of endothelial cells [37]. Similarly, we demonstrated that in HG-induced HRECs, the cell viability was decreased, while the apoptosis was promoted. Addition of GLI promoted cell viability and repressed apoptosis of HG-induced HRECs. The experiments indicated that GLI treatment can effectively inhibit the development of DR in a cellular level. Meanwhile, our in vivo experiments that GLI administration improved the decreased retinal thickness caused by DM. Interestingly, Zhu et al. investigated the effect of TGR5 receptor agonist (INT-777) on morphology of retinal tissues in a STZ-induced rat model and found a decrease in retinal thickness induced by diabetes, while intravitreal injection of INT-777 alleviated such morphological changes in the retinas, suggesting that TGR5 agonism is protective against diabetes-induced retinal vascular injury [38]. As a result, we also believed that GLI can alleviate retinal vascular injury in the progression of DR. As well accepted, excessive inflammation responses and oxidative damage are exited in the progression of DR [27, 28]. In the current study, we found enhanced inflammatory reactions and oxidative injuries in HG-induced HRECs and STZ-induced DM rats. In addition, we also demonstrated inflammatory reactions and oxidative stress were relieved by GLI administration both in vitro and in vivo. Similarly, Li et al. focused on the protective effects of sulforaphane on inflammation and peroxidation both in vitro and in vivo, and demonstrated that sulforaphane can attenuate the development of DR through inhibiting inflammation and oxidative stress in HG-induced rat Müller cells and STZ-induced DM rats [39]. Combined the previous results, we concluded that GLI may be also an effective agent to repress inflammation and oxidative stress in DR progression. Vascular hyperpermeability is the initial pathologic feature in DR capillaries, [40] and the increased level of VEGF is confirmed to be closely associated with retinal vascular permeability [32]. Previous researches have confirmed that some agents such as blueberry anthocyanins [41], resveratrol [42], and melatonin [43] can modulate vascular permeability via suppression of VEGF level in DR rats. In this investigation, as expected, DM rats showed relatively high level of VEGF, while administration of GLI repressed the release of VEGF. We speculated that GLI may be helpful for the inhibition of vascular hyperpermeability in the progression of DR. Based on the above data, we believed that GLI may be also an effective agent to ameliorate DR through suppressing a series of pathological processes of DR including endothelial cell apoptosis, inflammation, oxidative damage, and vascular hyperpermeability. In clinic, GLI may inhibit endothelial dysfunction, inflammation responses, oxidative injuries and vascular hyperpermeability, thereby contributing to the improvement of poor prognosis of DR patients.

Notch1 signaling and the corresponding effector genes act as important roles in prevention of inflammation and oxidative stress in numerous human diseases, such as arthritis, [44] enteritis, [45] myocardial injury, [46] and chronic obstructive pulmonary disease [47]. We speculated that there may be some relationships between GLI and Notch1 signaling. As expected, the in vitro and in vivo experiments showed that the protein levels of Notch1 and effector genes (Hes1 and Hey2) were elevated in HG-induced HRECs and STZ-induced DM rats, which suggested that the activation Notch1 signaling may accelerate the development of DR. Similarly, Miloudi et al. believed that the activation of Notch1 increases the risk of DR via inducing vascular permeability [23]. Additionally, we further indicated that GLI partly eliminated the promoting effects of HG on Notch1 signaling-related proteins. The results implied that GLI can repress the activation of Notch1 signaling. Notch1 signaling has been reported as a downstream effector pathway of SIRT1 to affect disorder progression, such as liver fibrosis, breast cancer, and lung cancer [17‐19]. We then further speculated GLI may also interact with SIRT1 in DR progression. Firstly, we found that SIRT1 protein level was downregulated in HG-induced HRECs and DM rats. Similarly, a decreased expression of SIRT1 is found in HG-induced RRECs and retinal tissues of DR rats [16]. However, GLI treatment significantly elevated the protein level of SIRT1, suggesting that GLI may induce the production of SIRT1. Based on the results that GLI can inactivate Notch1 signaling, we suggested that GLI may interact with SIRT1/Notch1 pathway to affect DR progression. As expected, we found that downregulation of SIRT1 partly attenuated the inhibiting effects of GLI on the levels of Notch1 signaling-related proteins. Therefore, the SIRT1/Notch1 signaling may be an important regulatory pathway in DR, providing a direction for the research of novel targeted drugs in clinic. Assuredly, except for SIRT1/Notch1 pathway, there are numerous signaling pathways involved in DR progression, such as PlGF/ERK [48], PLA2/COX-2/VEGF-A [49], TLR4/NF-kappaB [50], and PI3K/Akt/mTOR [51]. We speculated GLI may also ameliorate DR via these pathways. This is may be a limitation of this study and we will elucidate these issues in future studies. Additionally, some other factors have been linked to the progression of DR such as miRNAs [52, 53] and lncRNAs [54, 55]. The interaction of GLI with them should be further explored.

In summary, the present investigation indicates that GLI may be an effective agent to improve DR through inhibiting inflammation and oxidative stress via SIRT1/Notch1 pathway. Our findings clarify the action mechanism of GLI on the development of DR, and may provide an effective drug for DR therapy in clinic.