Endothelial cell-derived exosomes: A natural shield against osteoblast ferroptosis in diabetes

Abstract

In this editorial, we discuss a significant recent study published in World Journal of Diabetes by Shao et al. Diabetic osteoporosis (DOP) is a severe skeletal complication of diabetes. Emerging evidence indicates that ferroptosis, an iron-dependent form of regulated cell death, plays a pivotal role in osteoblast dysfunction and bone loss under hyperglycemic conditions. However, the precise mechanisms through which osteoblast ferroptosis can be effectively targeted for the treatment of diabetic osteopenia remain elusive. The study by Shao et al advances our understanding of DOP by demonstrating that endothelial cell-derived exosomes function as a natural delivery system that protects osteoblasts from high glucose-induced ferroptosis. The authors identify microRNA-335-3p (miR-335-3p) as the key molecular mediator within this protective mechanism, acting by directly targeting and suppressing prostaglandin endoperoxide synthase 2, a recognized marker of ferroptosis. Clinically, an inverse correlation was observed between serum miR-335-3p levels and disease severity, highlighting that miR-335-3p not only represents a potential therapeutic target but also a promising non-invasive biomarker for early detection and monitoring of DOP, thereby underscoring its translational relevance.

Key Words: Diabetic osteoporosis; Endothelial cell-derived exosomes; Ferroptosis Prostaglandin endoperoxide synthase 2; microRNA-335-3p; Bone mineral density

Core Tip: Diabetic osteoporosis (DOP), a common diabetic complication lacking targeted treatments, stems from chronic hyperglycemia disrupting the bone microenvironment. Key pathophysiological mechanisms include the accumulation of advanced glycation end products, oxidative stress, inflammation, and cell death. Currently, the field of DOP is moving beyond bone mineral density to understand the molecular and cellular dysregulation at the bone-vascular interface, aiming for precise diagnostics and therapies. This editorial highlight endothelial cell-derived exosomes and their cargo microRNA-335-3p as emerging and critical players in the pathophysiology of DOP. It discusses their role in inhibiting osteoclast ferroptosis and alleviating bone loss, while also identifying serum microRNA-335-3p as a promising non-invasive biomarker for early detection and monitoring, bridging molecular mechanisms with clinical translation.

This editorial refers to "Endothelial cell-derived exosomes inhibit high glucose-induced osteoblast ferroptosis by activating microRNA-335-3p/prostaglandin endoperoxide synthase 2" by Shao et al, 2026; https://doi.org/10.4239/wjd.v17.i1.111165.

---

INTRODUCTION

Diabetic osteoporosis (DOP) is a chronic metabolic bone disease characterized by abnormal bone tissue structure and reduced bone strength. It is pathologically defined by diabetes-induced reductions in bone mass and microstructural damage, leading to increased bone fragility and an elevated risk of fractures. According to related data, approximately 50%-66% of patients with diabetes mellitus (DM) have decreased bone mineral density (BMD), and about 33% of them have been diagnosed with DOP[1]. Studies have demonstrated that both type 1 DM and type 2 DM (T2DM) are significantly associated with an increased risk of fractures[2]. Critically, there are currently no approved therapies specifically for DOP, which highlights a major unmet medical need and drives research to identify disease-specific targets.

THE PATHOGENESIS OF DOP

DOP results from a complex and dysregulated “bone microenvironment” triggered by chronic hyperglycemia and associated metabolic disturbances. Key interconnected pathophysiological mechanisms currently under investigation include the accumulation of advanced glycation end products (AGEs), oxidative stress, inflammation, and deficiencies in insulin and insulin-like growth factor 1.

Chronic hyperglycemia induces non-enzymatic glycation of collagen and other bone matrix proteins, leading to the accumulation of AGEs. These cross-linked AGEs compromise the material properties of bone, increasing brittleness and reducing resistance to crack propagation, even when BMD appears preserved. Furthermore, osteoblasts under high-glucose conditions exhibit reduced proliferative capacity, delayed extracellular matrix synthesis, and impaired maturation and mineralization. Together, these alterations underscore the detrimental impact of a hyperglycemic milieu on bone formation and the maintenance of bone mass[3].

Hyperglycemia also drives excessive production of reactive oxygen species (ROS), establishing a state of chronic oxidative stress. The accompanying chronic inflammation contributes to micro- and macrovascular complications in T2DM while adversely affecting bone remodeling[4,5]. Additionally, a high-glucose environment promotes the lipogenic differentiation of bone marrow mesenchymal stem cells and enhances bone marrow fat deposition, accompanied by the release of free fatty acids and pro-inflammatory cytokines[6].

Insulin promotes the expression of runt-related transcription factor 2, a key transcription factor involved in bone metabolism, thereby stimulating osteoblast differentiation and the maturation of the bone matrix[7]. In type 1 DM, insulin deficiency due to autoimmune destruction of pancreatic β-cells from the onset of the disease leads to impaired bone mineralization during adolescence. In T2DM, systemic insulin resistance and compensatory hyperinsulinemia in the early stages may initially enhance bone mineralization[8]. However, as islet function declines in advanced disease, insulin secretion decreases, leading to a significant reduction in BMD[9,10]. Owing to its structural similarity to insulin, insulin-like growth factor 1 also exerts anabolic effects on osteoblasts, promoting osteogenesis by accelerating collagen synthesis and bone matrix mineralization. Furthermore, a pathological interplay exists among chronic hyperglycemia, sustained inflammation, and cellular ferroptosis. Hyperglycemia fuels a pro-inflammatory state, which can, in turn, upregulate ferroptosis-promoting pathways (e.g., lipid peroxidation), creating a vicious cycle that exacerbates cellular dysfunction and tissue damage in this context.

FERROPTOSIS: THE EMERGING EXECUTIONER IN DOP

Ferroptosis is a new form of iron-dependent cell death characterized by an imbalance of iron homeostasis and lipid peroxidation[11]. It has rapidly gained recognition as a key pathological process in diabetic complications, including nephropathy, retinopathy, and cardiomyopathy. Its core biochemical hallmarks, including iron overload, glutathione (GSH) depletion, inactivation of GSH peroxidase 4 (GPX4), and the accumulation of toxic lipid peroxides, are exacerbated by the hyperglycemia-driven overproduction of ROS[1]. The system Xc^-/GSH/GPX4 axis is the primary intracellular defense against ferroptosis. The cystine/glutamate antiporter, composed of a catalytic subunit [solute carrier family 7 member 11 (SLC7A11)] and a chaperone subunit [solute carrier family 3 member 2 (SLC3A2)], imports cystine for GSH synthesis. GPX4 then utilizes GSH to detoxify lipid hydroperoxides. In the case of DOP, this protective axis is crippled[12]. The inhibition of GPX4 activity leads to the accumulation of lipid peroxides, which induces ferroptosis.

The study meticulously recapitulates this ferroptosis cascade in osteoblasts exposed to high glucose. The authors show a marked increase in ROS and malondialdehyde (MDA; a lipid peroxidation product), coupled with a decrease in GSH. At the molecular level, the results confirm the downregulation of the ferroptosis suppressors GPX4, SLC7A11, and SLC3A2, and the upregulation of prostaglandin endoperoxide synthase 2 (PTGS2), a critical marker of ferroptosis[13]. PTGS2 is an inducible enzyme that catalyzes the rate-limiting step in the biosynthesis of prostaglandins, such as prostaglandin E2. It is a key mediator of inflammation, pain, and fever, and is rapidly upregulated in response to cellular stress, cytokines, and growth factors[14]. Beyond its classic role in inflammation and prostaglandin synthesis, it is transcriptionally upregulated during ferroptosis and may further fuel lipid peroxidation cycles. Its elevated expression has been linked to poor bone outcomes in glucocorticoid-induced osteoporosis[15].

THE RESCUE MISSION: EC-EXOS DELIVER A MOLECULAR ANTIDOTE

Against this backdrop, the study by Shao et al[13], in the recent issue of the World Journal of Diabetes, offers a timely and significant contribution. The authors elegantly bridge two rapidly advancing frontiers in bone biology: The role of non-apoptotic programmed cell death (specifically ferroptosis) in bone cell dysfunction, and the therapeutic potential of extracellular vesicles, particularly endothelial cell-derived exosomes (EC-Exos), as natural, targeted delivery systems.

The central finding of the study is that EC-Exos act as a powerful exogenous protective signal against high glucose-induced ferroptosis. Co-culture with EC-Exos reversed the biochemical and molecular signatures: It lowered ROS and MDA levels, restored GSH levels, and normalized the expression of GPX4, SLC7A11, SLC3A2, and PTGS2. This finding confirms the therapeutic potential of the osteovascular axis.

The authors then embarked on a rigorous mechanistic quest to identify the “active ingredient” within EC-Exos responsible for this effect. Using high-throughput sequencing, they compared the miRNA profiles of osteoblasts under high glucose conditions, both with and without EC-Exos treatment. From a panel of differentially expressed miRNAs, microRNA-335-3p (miR-335-3p) emerged as a lead candidate, showing a dramatic 3.7-fold upregulation upon EC-Exos treatment. Bioinformatics analysis predicted PTGS2 as a direct target of miR-335-3p, a prediction decisively validated by a dual-luciferase reporter assay, showing that miR-335-3p binds specifically to the 3’ untranslated region of PTGS2 mRNA to suppress its expression.

The functional centrality of this axis was established through elegant loss-of-function experiments. When miR-335-3p was inhibited in osteoblasts, the protective effects of EC-Exos were completely abolished. The cells reverted to a ferroptosis state, exhibiting high PTGS2, low GPX4/SLC7A11/SLC3A2, elevated levels of ROS and MDA, and depleted GSH. These compelling data delineate a clear pathway: EC-Exos protect osteoblasts from high glucose-induced ferroptosis by regulating miR-335-3p/PTGS2.

In this finding, the authors found that miR-335-3p levels were low in the EC-Exos themselves but high in recipient osteoblasts after treatment, which indicates that EC-Exos do not simply deliver miR-335-3p as cargo. Instead, they likely deliver a signal, which could be a protein, another RNA species, or a lipid, that activates the endogenous expression of the miR-335-3p gene within the osteoblast. While the precise mechanism requires further investigation, this upregulation in recipient osteoblasts could be mediated by EC-Exo delivery of specific transcription factors or through the activation of cell surface receptor signaling pathways (e.g., via exosomal ligands), ultimately enhancing miR-335-3p transcription. This represents a more sophisticated, amplifier-like mode of action and highlights an exciting area for future investigation.

FROM BENCH TO BEDSIDE: CLINICAL TRANSLATION AND DIAGNOSTIC POTENTIAL

The most translational aspect of this work lies in its clinical correlation. The researchers analyzed serum samples from well-phenotype female patients with T2DM alone and those with T2DM and osteoporosis. The results were striking and consistent with the cellular model: DOP patients had significantly lower serum levels of miR-335-3p and significantly higher levels of PTGS2 compared to the T2DM controls. A strong inverse correlation was observed between these two molecules. In multivariate logistic regression, serum miR-335-3p emerged as an independent factor associated with DOP, alongside established risk factors such as lower body mass index and fasting C-peptide.

Most notably, receiver operating characteristic curve analysis suggested that serum miR-335-3p has promising diagnostic utility for distinguishing DOP from uncomplicated T2DM, with an area under the curve of 0.875. This finding elevates miR-335-3p from being merely a mechanistic player to a potential circulating biomarker. While BMD is the diagnostic gold standard for osteoporosis, it reflects bone quantity rather than the dynamic pathological activity driving drug-induced bone loss. The strong inverse correlation of miR-335-3p with disease severity, coupled with its area under the curve, positions it as a promising pathology-specific biomarker. It could potentially serve a complementary role in detecting early, active bone loss before significant changes in BMD occur or in monitoring therapeutic responses at the molecular level.

Although the receiver operating characteristic analysis is promising and indicates strong diagnostic potential, it is crucial to note that the clinical correlations were established within a cohort restricted to female patients. Future validation in broader, mixed-sex populations is essential to confirm the biomarker’s generalizability. In a field hampered by the inadequacy of BMD, a readily measurable serum marker that reflects the underlying pathological activity of the ferroptosis/PTGS2 axis could revolutionize risk stratification, enable earlier diagnosis of bone fragility in diabetic patients, and potentially monitor response to therapy.

BROADER IMPLICATIONS AND FUTURE HORIZONS

This study by Shao et al[13] opens several important avenues for both basic science and clinical medicine.

Reinforcing the osteovascular paradigm

It provides robust experimental evidence that vascular health is integral to bone health, not merely through perfusion but also via active exosome-mediated signaling. This underscores the need for a holistic approach to DOP that takes endothelial function into account.

Identifying a novel theragnostic axis

The miR-335-3p/PTGS2 pathway is positioned as a key regulatory node in the diabetic osteoblast fate. This makes it an attractive dual-purpose target: For therapy (via miR-335-3p mimics or PTGS2 inhibitors) and for diagnosis (through serum miR-335-3p levels).

Validating exosomes as a therapeutic platform

The work strengthens the rationale for developing EC-Exos, or biomimetic synthetic nanoparticles engineered with EC-Exos surface markers and miR-335-3p-enhancing cargo, as a targeted, cell-free therapy for DOP and potentially other forms of metabolic bone disease.

Connecting to a larger narrative

miR-335-3p has been implicated in improving insulin sensitivity and protecting islet function. It is downregulated in other diabetic complications, such as retinopathy, and in inflammatory bone diseases, including osteoarthritis and disc degeneration[16,17]. This study adds DOP to that list, suggesting that miR-335-3p may serve as a common regulator of metabolic and inflammatory stress across various tissues.

UNANSWERED QUESTIONS AND PATHS FORWARD

As with any pioneering study, new questions arise to guide future research.

The upstream trigger

What is the precise molecular component(s) within EC-Exos that triggers the upregulation of miR-335-3p in osteoblasts? The upstream likely involves one or more of the following candidate mechanisms: (1) An exosomal transcription factor that directly activates miR-335-3p gene expression; (2) A regulatory RNA that sequesters a repressor of miR-335-3p; and (3) A membrane-bound ligand that activates a pro-survival receptor signaling cascade in osteoblasts. Unraveling this will be key to engineering optimized therapeutics.

In vivo validation

Building upon the promising in vitro findings, the critical next step is to validate the therapeutic potential in relevant animal models of DOP, such as db/db mice or streptozotocin-induced diabetic rodents with bone loss. The central questions to address include: (1) Does systemic administration (e.g., intravenous or intraperitoneal) of EC-Exos or synthetic miR-335-3p mimics effectively target bone tissue in a diabetic milieu; (2) Can this intervention measurably improve key pathological outcomes, including bone microarchitecture, biomechanical strength, and ultimately reduce fracture susceptibility in vivo; and (3) What are the systemic safety and pharmacokinetic profiles of such treatments? Addressing these questions will bridge the mechanistic insights to preclinical therapeutic efficacy.

CELLULAR SPECIFICITY AND INTERACTIONS

The identified pathway likely initiates in osteoblasts, but its ultimate therapeutic efficacy depends on its effects throughout the bone niche. Do osteocytes, upon receiving EC-Exos signals, amplify the anti-ferroptosis response through their canalicular network? Could modulating osteoclast precursors alter the resorptive environment and indirectly support osteoblast survival? Investigating these interactions will reveal whether the therapy establishes a positive feedback loop that stabilizes the bone microenvironment or acts primarily as a one-time cellular rescue. Understanding this dynamic regulation is essential for designing rational treatment schedules and combination therapies.

BIOMARKER TRANSLATION AND CLINICAL VALIDATION

To translate miR-335-3p into a clinically useful biomarker, its diagnostic and prognostic potential must be rigorously validated. This requires prospective, multi-center longitudinal cohorts encompassing both men and women with T2DM, stratified by bone health status. Analyses must rigorously control for key confounders, including age, sex, body mass index, diabetes duration, glycemic control (glycated hemoglobin), renal function (estimated glomerular filtration rate), and concurrent medications. The validation should progress through three stages: (1) Establishing its correlation with BMD and microarchitectural parameters; (2) Evaluating its diagnostic accuracy for DOP against the gold standard; and (3) Assessing its ability to predict future fracture risk, independently of established clinical risk factors.

More importantly, the translational path for EC-Exo-based diagnostics or therapeutics will need to address significant practical challenges. These include the development of standardized, scalable methods for EC-Exo isolation and characterization, as well as establishing protocols for their long-term storage, stability, and effective in vivo delivery, and dosing.

CONCLUSION

In summary, Shao et al[13] have presented a compelling and multifaceted study that illuminates a novel protective communication from the vasculature to bone. They identify the EC-Exos/miR-335-3p/PTGS2 axis as a crucial defender against the ferroptosis death of osteoblasts in the diabetic milieu. By seamlessly integrating mechanistic cell biology with insightful clinical correlation, their work provides a cohesive narrative that advances our understanding of DOP pathogenesis while simultaneously proposing tangible tools for future diagnosis and therapy. This research exemplifies the power of interdisciplinary investigation at the intersection of endocrinology, bone biology, and vascular science. It moves us closer to the ultimate goal: Developing precise, effective, and targeted strategies to protect the fragile bones of the ever-growing diabetic population.