Diabetic retinopathy (DR) is a highly hazardous and widespread complication associated with diabetes mellitus (DM). It affects approximately 30% of people with DM and is the primary cause of vision impairment and blindness in people over age of 40 [1,2]. 578 million people are predicted to live in DR by 2030, and 700 million by 2045 [3]. Currently, therapeutic strategies for DR, such as laser photocoagulation, anti-VEGF, glucocorticoid drugs, and vitreoretinal surgery, have demonstrated limited success and sometimes even adverse effects [4]. In this regard, it is imperative to enhance our knowledge of the mechanisms that contribute to DR and explore alternative therapeutic strategies.

DR's pathogenesis is complex and incompletely understood. However, it's recognized that oxidative stress and inflammation are important factors in DR development [5]. The primary source of inflammatory cytokines in DR are Müller cells, and their dysfunction is an important contributor to the vasculopathies of DR [6,7]. Chronic hyperglycemia causes metabolic abnormalities and the accumulation of intracellular reactive oxygen species (ROS), which in turn causes inflammation and oxidative stress in the retina [5,8]. Elevated inflammatory cytokines disrupt endothelial junctional proteins and damage blood-retinal barrier (BRB), which is a characteristic feature of DR [8]. Recent evidence suggests that neddylation is associated with oxidative stress and inflammation [9,10]. Neddylation refers to a posttranslational protein modification by which neuronal precursor cell-expressed developmentally down-regulated protein 8 (NEDD8) is conjugated to target proteins [11]. The most well-known substrates of NEDD8 are Cullins, which are the scaffold proteins of the Cullins-RING E3 ligases (CRLs). The regulation of nuclear factor erythroid 2-related factor 2 (Nrf2), which controls cellular ROS levels by modulating transcription of downstream antioxidant enzymes, is mediated by Cullin3-RING ligase (CRL3). However, the activity of Nrf2 signal pathway is impaired in DR [12]. Therefore, targeting NEDD8-mediated neddylation may be an effective method to attenuate ROS-induced retinal inflammation and relieve DR by activating Nrf2 signal pathway.

Numerous studies have revealed the beneficial impact of mesenchymal stem cells (MSCs) on DR animal models and patients [13,14]. Their positive effect is predominantly attributable to paracrine activity and the secretion of small extracellular vesicles (sEV) [15,16]. sEVs mediate cell-to-cell communication by transferring their cargoes (miRNAs, mRNAs, proteins, and lipids) into neighboring cells. MSC-derived sEVs (MSC-sEV) have advantages over MSCs, such as non-tumorigenic, non-immunogenic, and easy to be endocytosed by target cells to conduct specific functions [17]. The therapeutic mechanisms of MSC-sEV have been shown to be associated with their miRNA cargoes [[18], [19], [20]]. Recent investigations have demonstrated that intravitreal MSC-sEV injection was able to alleviate DR [[18], [19], [20]]. Nevertheless, the mechanisms have not yet been fully comprehended.

With aforementioned knowledge, our study aimed to examine the protective impact and mechanisms of MSC-sEV on ROS-induced retinal damage in DR, and also determine whether the protective impact is associated with the inhibition of neddylation. Additionally, based on miRNA sequencing of MSC-sEV and online bioinformatics software, we intended to identify the major miRNA in MSC-sEV that possesses the ability to inhibit NEDD8-mediated neddylation.