Chronic wounds, particularly those associated with diabetes, represent a major clinical challenge due to their prolonged healing times and susceptibility to infection, often resulting from excessive inflammation and oxidative stress (Falanga, 2005; Armstrong et al., 2017; Ellis et al., 2018; Powers et al., 2016). The diabetic wound microenvironment is characterized by impaired angiogenesis, persistent inflammation, and elevated levels of reactive oxygen species (ROS) and reactive nitrogen species (RNS), which collectively exacerbate tissue damage (Schäfer and Werner, 2008). These reactive species trigger the release of pro-inflammatory cytokines, including interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), and matrix metalloproteinases (MMPs), which degrade the extracellular matrix (ECM), hinder epithelialization, and promote apoptosis of critical cell types such as keratinocytes and fibroblasts (Brem and Tomic-Canic, 2007; Trüeb, 2002; Ju et al., 2016). This vicious cycle not only delays wound closure but also increases the risk of severe complications, such as chronic ulcers and amputations, imposing a significant burden on healthcare systems globally (Sen et al., 2009). Current therapeutic strategies, such as non-steroidal anti-inflammatory drugs (NSAIDs) (Andrade Del Olmo et al., 2022; Nochos et al., 2013), topical antibiotics (Xu et al., 2016; Tiwari and Pathak, 2023; Heal et al., 2016), and advanced wound dressings (Boateng et al., 2008; Jones et al., 2006) (e.g., hydrocolloids and hydrogels), often provide symptomatic relief but fail to address the underlying oxidative and inflammatory imbalances effectively. This limitation underscores the urgent need for innovative, targeted therapies that can modulate these pathological processes and restore the natural healing cascade.

Curcumin, a polyphenolic compound extracted from the rhizome of Curcuma longa (turmeric), has emerged as a promising natural agent due to its well-documented antioxidant, anti-inflammatory, and antimicrobial properties (Aggarwal and Harikumar 2009). Its ability to neutralize ROS, inhibit pro-inflammatory signaling pathways (e.g., NF-κB), and promote collagen synthesis positions it as a potential candidate for wound healing applications (Ak and Gülçin, 2008; Liu et al., 2018; Wang et al., 2024). However, the clinical translation of curcumin is hindered by its poor water solubility, rapid metabolism, and low bioavailability, which result in suboptimal therapeutic concentrations at the wound site (Anand et al., 2007). To overcome these challenges, nanotechnology has offered a transformative approach by encapsulating or coordinating natural compounds with metal ions to enhance their stability, solubility, and targeted delivery (Sun et al., 2020). Among these, iron-based coordination nanoparticles have garnered attention for their ability to improve the physicochemical properties of bioactive molecules while leveraging the catalytic potential of iron to mitigate oxidative stress (Khodaei et al., 2022). For example, iron-curcumin nanoparticles have been successfully employed in inflammatory diseases such as osteoarthritis (OA), where they enhance ROS/RNS scavenging and modulate key pathways, including the nuclear factor-erythroid 2 related factor-2 (Nrf2) and NLRP3 inflammasome pathways, to reduce cartilage degradation and inflammation (Patra et al., 2015).

Building on these advancements, we hypothesize that iron-curcumin-coordinated nanoparticles can be engineered as a dual-action therapeutic platform to regulate inflammation and promote diabetic wound healing. In this study, we synthesized Fe-Cur-P nanoparticles by coordinating curcumin with iron ions and loading carboxy-PTIO, a nitric oxide scavenger, to specifically target elevated NO levels in diabetic wounds (Scheme 1). This design aims to mitigate oxidative stress by neutralizing ROS and RNS, suppress excessive inflammation by downregulating pro-inflammatory cytokines, and enhance tissue regeneration through sustained drug release and improved biocompatibility. The therapeutic potential of Fe-Cur-P nanoparticles is evaluated through a comprehensive approach, including in vitro studies using RAW 264.7 macrophage cell lines to assess anti-inflammatory and anti-apoptotic effects, and in vivo experiments in a diabetic mouse model to monitor wound closure, histological regeneration, and immunomodulatory responses. Our investigation focuses on key molecular markers, such as TGF-β for tissue repair and myeloperoxidase (MPO) for neutrophil infiltration, alongside histological assessments of vascularization and collagen deposition. This multifaceted analysis aims to establish Fe-Cur-P nanoparticles as a novel, effective platform for managing chronic diabetic wounds, with implications for broader applications in regenerative medicine.

As a natural antioxidant, curcumin has garnered significant attention; however, its biomedical applications are often limited by poor biocompatibility and low water solubility. (Jia et al., 2021; Hosseinzadeh et al., 2023). In chronic diabetic wounds, sustained oxidative stress and unresolved inflammation form a self-amplifying loop that impairs angiogenesis, re-epithelialization and ECM remodeling; therefore, biomaterials capable of reshaping the redox/inflammatory microenvironment are increasingly considered a rational strategy to restore healing trajectories (Worsley et al., 2023). In this context, curcumin-based dressings have been widely explored due to their intrinsic antioxidant and immunomodulatory potential, yet their translational utility is frequently constrained by poor aqueous solubility, chemical instability and limited local bioavailability—motivating chemical/coordination modification and carrier-assisted delivery designs. In this study, iron ions were introduced into a methanol solution of curcumin (Figure 1a). We utilized transition metal iron to coordinate with curcumin, yielding fully water-soluble curcumin-coordinated nanoparticles (Fe-Cur). Notably, metal–polyphenol coordination has been reported as an effective route to simultaneously improve the dispersibility of hydrophobic polyphenols and introduce catalytic (“nanozyme-like”) ROS-regulating functions. For example, Fe–curcumin nanoparticles have been shown to exhibit ROS-scavenging and anti-inflammatory activities via modulation of cytokine production and redox-linked inflammatory signaling, supporting the design rationale that Fe coordination can convert curcumin from a poorly soluble antioxidant into a water-dispersible, bioactive nanomedicine (Fan et al., 2024). Transmission electron microscopy (TEM) revealed that these nanoparticles have a diameter of approximately 9.8 ± 0.6 nm (Figures 1b,c). Furthermore, Fe-Cur nanoparticles were employed as carriers to load carboxy-PTIO (a nitric oxide scavenger) via π-π interactions, forming Fe-Cur-P nanoparticles (Figure 1b). Beyond ROS, diabetic wounds also feature dysregulated reactive nitrogen species (RNS) chemistry. While physiological NO can be pro-healing, excessive nitrosative stress in inflamed tissues may aggravate damage, and NO can rapidly react with superoxide to form peroxynitrite, a highly reactive oxidant implicated in protein/lipid oxidation and inflammatory amplification. Thus, “context-dependent” local modulation of NO/RNS (rather than indiscriminate NO elevation) is increasingly recognized as a complementary axis to ROS control in chronic wounds. In this regard, carboxy-PTIO is a commonly used NO scavenger in biochemical and biological studies, providing a practical handle to attenuate NO-driven nitrosative stress in situ (Pfeiffer et al., 1997). The size and zeta potential of Fe-Cur-P nanoparticles changed after loading carboxy-PTIO, indicating successful structural modification and good stability in aqueous solutions (Figures 1d,e). Additionally, the characteristic peaks of carboxy-PTIO were observed in Fe-Cur-P nanoparticles, confirming the successful loading of carboxy-PTIO onto the nanoparticle surface (Figure 1e). As the yellow curcumin solution diminished, the pale-yellow iron solution transformed into a dark black product, indicating a coordination reaction between curcumin and iron ions (Figure 1f). Based on the carboxy-PTIO peak in Fe-Cur-P and the carboxy-PTIO standard curve from high-performance liquid chromatography (HPLC), the drug loading efficiency of carboxy-PTIO in Fe-Cur-P nanoparticles was determined to be 21.31% (Supplementary Figure S1). The structure of Fe-Cur-P nanoparticles was evaluated using Fourier-transform infrared spectroscopy (FT-IR), which showed spectra consistent with those of Fe-Cur nanoparticles. Characteristic peaks at approximately 3400 cm⁻¹ and 1200 cm⁻¹ corresponded to O-H and pyrrole groups, respectively, as shown in Figure 1f. Moreover, a characteristic PTIO peak at approximately 1550 cm⁻¹, attributed to imidazole ring vibrations, was observed in the spectra of Fe-Cur-P nanoparticles (Supplementary Figure S2).

The stimuli-responsive release behavior was evaluated under simulated diabetic wound conditions (pH = 6.0) and physiological conditions (pH = 7.4). Fe-Cur-P nanoparticles exhibited ideal acid-responsive release of carboxy-PTIO (a nitric oxide scavenger) under oxidative conditions mimicking a diabetic wound environment (Figure 1g), suggesting that Fe-Cur-P may hold therapeutic potential by reducing nitric oxide levels at diabetic wound sites. Importantly, diabetic wound microenvironments are frequently characterized by mild acidity and elevated oxidative burden, and stimulus-responsive release systems have been repeatedly leveraged to enhance on-site efficacy while minimizing off-target exposure. From a positioning standpoint, our pH/oxidation-associated release behavior aligns with the broader design logic of “microenvironment-adaptive” wound therapeutics reported in recent diabetic wound biomaterials literature, while introducing an added RNS-regulating component through carboxy-PTIO loading. Furthermore, the sustained release of Fe-Cur-P monomers from the nanoparticles persisted for over 28 days, indicating a significantly prolonged half-life (Supplementary Figure S3). In conclusion, Fe-Cur-P nanoparticles with controlled release properties have been successfully developed.

Curcumin, a natural reducing agent derived from plants, suggests that the synthesized Fe-Cur-P nanoparticles may exhibit potent scavenging capabilities for reactive nitrogen species (RNS) and reactive oxygen species (ROS). To evaluate the antioxidant capacity of Fe-Cur-P nanoparticles, we employed multiple probes, including 1,1-diphenyl-2-picrylhydrazyl (DPPH·), 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) radical (ABTS+·), and methylene blue (MB, an indicator of hydroxyl radicals (·OH)). Initially, DPPH· and ABTS+· radicals were used as model RNS to assess the antioxidant activity of Fe-Cur-P nanoparticles. The characteristic absorption peak of DPPH· at 517 nm gradually diminished upon incubation with antioxidants. Notably, the absorbance of DPPH· decreased significantly after mixing with Fe-Cur-P nanoparticles, indicating robust free radical scavenging activity (Figure 1h). This finding was corroborated by the ABTS+· assay, which yielded similar results (Figure 1i). Evidently, Fe-Cur-P nanoparticles at appropriate concentrations effectively scavenged the majority of DPPH· and ABTS+· radicals. Hydroxyl radicals (·OH), known for their potent reactivity and ability to induce cellular damage, were also investigated. The scavenging capacity of Fe-Cur-P nanoparticles against ·OH, generated via the Fenton reaction, was evaluated. As anticipated, the presence of Fe-Cur-P nanoparticles significantly attenuated the degradation of MB by the Fenton reaction, demonstrating strong ROS/RNS scavenging activity (Figure 1j). Collectively, these results unequivocally confirm that Fe-Cur-P nanoparticles, enriched with phenolic groups, serve as a potent scavenger of ROS and RNS, offering significant potential for anti-inflammatory treatment in diabetic wound healing. When benchmarked against prior ROS-regulating diabetic wound dressings, nanozyme-reinforced hydrogels (e.g., ceria- or manganese oxide–containing systems) have been reported to accelerate closure by reducing oxidative stress, improving angiogenesis and suppressing pro-inflammatory signaling. Our multi-probe radical assays (DPPH·, ABTS+· and Fenton-associated MB decolorization) are consistent with this “redox normalization” paradigm, and the inclusion of an NO-scavenging cargo further extends the activity spectrum from ROS toward ROS/RNS co-regulation, which may be advantageous in chronic inflammatory wound settings where nitrosative stress can co-exist with oxidative stress.

To further evaluate the intracellular reactive oxygen species (ROS) scavenging behavior of Fe-Cur-P nanoparticles (NPs), we investigated their effects in a cellular model. Chronic diabetic wounds are characterized by elevated levels of pro-inflammatory mediators, where oxidative stress induced by ROS and nitric oxide (NO) plays a critical role in exacerbating inflammation. This leads to the overactivation of inflammatory responses, resulting in the production of pro-inflammatory cytokines and chemokines, which contribute to the apoptosis of epithelial and vascular endothelial cells, thereby worsening diabetic wound conditions. Studies have shown that suppressing the expression of ROS-related genes, such as interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), matrix metalloproteinases (MMP-1, -3, and -13), and cyclooxygenase-2 (COX-2), or enhancing their expression, exerts anti-inflammatory effects in epithelial cells, vascular endothelial cells, and animal models of diabetic wounds. Additionally, current therapies for managing inflammation in diabetic wounds include non-steroidal anti-inflammatory drugs, underscoring the importance of inflammation control in diabetic wound treatment. Thus, controlling inflammation in diabetic wounds is critical for achieving therapeutic efficacy. Mechanistically, a key hallmark of diabetic wounds is the persistence of pro-inflammatory macrophage phenotypes and a delayed/insufficient transition toward pro-reparative states, which sustains cytokine production (e.g., IL-1β, IL-6, TNF-α) and perpetuates oxidative injury. Multiple reviews highlight macrophages as actionable therapeutic targets in diabetic wound healing and emphasize that successful interventions often converge on dampening inflammatory cytokines while restoring a pro-healing microenvironment. In this context, our in vitro findings—showing suppression of canonical LPS-induced inflammatory mediators together with ROS reduction—are in line with the mechanistic expectations of macrophage-centered diabetic wound therapies reported previously. Given the remarkable structural properties of Fe-Cur-P nanoparticles, we applied them to diabetic wound models to assess their anti-inflammatory potential, as the accumulation of pro-inflammatory factors at the wound site is a primary cause of delayed healing in diabetic wounds. Different macrophage subsets are closely associated with immune cell infiltration, disease progression, and pain levels in diabetic wound tissues. Both innate and adaptive immune systems play key roles in the low-grade inflammation associated with diabetic wounds. Therefore, RAW 264.7 cells, a representative murine macrophage cell line, were used as a model for immune cells to simulate the immune response in osteoarthritis (OA). Specifically, RAW 264.7 cells were treated with lipopolysaccharide (LPS, 100 ng/mL) for 24 h, followed by incubation with Fe-Cur-P nanoparticles for an additional 24 h (Figure 2a). At concentrations ranging from 10 to 200 μg/mL, both Cur and Fe-Cur nanoparticles exhibited good biocompatibility, with Fe-Cur-P nanoparticles maintaining excellent biocompatibility at concentrations up to 100 μg/mL (Figure 2a). Therefore, in both in vivo and in vitro experiments, a concentration of 100 μg/mL was used for the nanoparticles.

Calcein AM/PI live/dead cell staining images demonstrated that Fe-Cur-P nanoparticles effectively inhibited LPS-induced apoptosis in RAW 264.7 cells (Figures 2b,c). Flow cytometry analysis was employed to assess intracellular ROS levels in epithelial cells. Compared to the control group, ROS levels in LPS-stimulated epithelial cells increased from approximately 0.3%–26.9%. While Cur and Fe-Cur partially mitigated epithelial cell apoptosis (approximately 22.8% and 20.3%, respectively), Fe-Cur-P nanoparticles significantly reduced the apoptosis rate to approximately 5.83% (Figures 2d,e). Activated caspase 3/7 cleaves poly (ADP-ribose) polymerase (PARP), a nuclear protein critical for cell differentiation, proliferation, and apoptosis, producing an 89 kDa fragment. Accordingly, we measured the levels of cleaved PARP-1 in RAW 264.7 cells. As expected, Fe-Cur-P nanoparticles (at a concentration of 100 μg/mL) eliminated the production of LPS-induced cleaved PARP-1 in cells (Supplementary Figure S4). The anti-inflammatory potential of Fe-Cur-P nanoparticles was further elucidated through the quantitative analysis of pro-inflammatory mediators in RAW 264.7 cells, as depicted in Figures 2f–m. The expression levels of interleukin-1β (IL-1β), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α), and cyclooxygenase-2 (COX-2) were significantly elevated following LPS stimulation, with IL-1β mRNA levels increasing approximately 4-fold (Figure 2f), IL-6 mRNA levels rising by about 15-fold (Figure 2g), TNF-α mRNA levels increasing by approximately 3-fold (Figure 2h), and COX-2 mRNA levels showing a 10-fold increase (Figure 2i) compared to the control group. Treatment with Fe-Cur-P nanoparticles markedly attenuated these elevations, reducing IL-1β mRNA to near baseline levels, IL-6 mRNA by approximately 80%, TNF-α mRNA by about 60%, and COX-2 mRNA by roughly 50% relative to the LPS-only group. Similarly, at the protein level, LPS stimulation resulted in substantial increases, with IL-1β protein levels rising to approximately 500 pg/mL (Figure 2J), IL-6–80 pg/mL (Figure 2K), TNF-α to 1000 pg/mL (Figure 2L), and COX-2–8000 pg/mL (Figure 2M). In contrast, Fe-Cur-P treatment significantly suppressed these protein levels, reducing IL-1β to approximately 200 pg/mL, IL-6–20 pg/mL, TNF-α to 500 pg/mL, and COX-2–4000 pg/mL, demonstrating a robust inhibitory effect. These findings indicate that Fe-Cur-P nanoparticles effectively downregulate both the transcriptional and translational expression of key pro-inflammatory cytokines and enzymes, likely due to their potent ROS and RNS scavenging capabilities, thereby offering a promising therapeutic strategy for mitigating inflammation in diabetic wound healing.

The selection of animal sex remains a contentious issue in experimental research, with some investigators advocating for balanced gender representation and the systematic consideration of sex as a biological variable. Drawing on prior studies of wound healing and to mitigate potential influences of estrogen on wound repair, we exclusively utilized male animals in this study. While this approach may limit the clinical applicability of our findings to some extent, future investigations will aim to address this limitation. In our experimental design, we employed a diabetic mouse model, inducing full-thickness skin defects measuring 1.5 cm on their backs to evaluate the therapeutic efficacy of nanoparticles in wound healing.

Infected diabetic mice were randomly assigned to four groups: a control group, a Cur group, an Fe-Cur group, and an Fe-Cur-P group. The diabetic wounds were treated with different nanoparticle formulations, and healing progress was monitored over 2 weeks (Figure 3a). Photographic evidence revealed that the Fe-Cur-P group exhibited significantly faster wound closure compared to the Cur, Fe-Cur, and control groups (Figure 3b). Notably, the Fe-Cur-P group demonstrated statistically significant differences in wound closure rates on days 3, 7, 10, and 14. By day 14, the wound closure rate in the Fe-Cur-P group reached 93.5% ± 2.4%, nearly achieving complete healing, whereas the Cur and Fe-Cur groups achieved closure rates of 67.2% ± 3.1% and 76.2% ± 4.5%, respectively (Figure 3c). The final healing time points for the mice were also recorded (Figure 3d), with the control, Cur, Fe-Cur, and Fe-Cur-P groups achieving complete healing at 20.5 days, 18.2 days, 18.4 days, and 13.6 days, respectively.

We conducted histological and immunomodulatory analyses to further elucidate the impact of Fe-Cur-P on wound healing. Masson’s trichrome and hematoxylin-eosin (H&E) staining were performed to assess the regeneration of blood vessels, hair follicles, and collagen at the wound site on day 14, while evaluating the reconstruction of skin architecture across different treatment groups. By day 14, wounds in the Fe-Cur-P group exhibited robust recovery, featuring a continuous epithelial layer, substantial collagen deposition, and newly formed blood vessels and hair follicles (Figures 4a,b). These features were notably absent in the control, Cur, and Fe-Cur groups. Wound margins treated with Fe-Cur-P displayed extensive regeneration of blood vessels and hair follicles, alongside enhanced epithelialization and collagen deposition. Quantitative analysis revealed that Fe-Cur-P-treated wounds significantly outperformed other groups in terms of vascular and follicular regeneration (Supplementary Figure S5) and showed superior epithelialization and collagen deposition. High-magnification H&E staining images indicated minimal fibroblast proliferation in the control, Cur, and Fe-Cur groups, accompanied by substantial infiltration of inflammatory cells. In contrast, Fe-Cur-P-treated wounds showed negligible inflammatory cell infiltration but abundant fibroblast participation in tissue reconstruction. These findings suggest that Fe-Cur-P markedly enhances epithelialization, reduces tissue inflammation, and improves wound healing quality by promoting the regeneration of blood vessels, hair follicles, and collagen.

To further explore the mechanisms underlying tissue repair, wound samples were collected on day 14 for immunohistochemical staining of transforming growth factor-β (TGF-β) (Figures 4c,d). Fe-Cur-P significantly elevated TGF-β levels at the wound site, aligning them with those of the normal wound group, indicating its exceptional capacity to regulate TGF-β and promote tissue repair. TGF-β is a pivotal cytokine in wound healing, orchestrating fibroblast activation, collagen synthesis, and extracellular matrix (ECM) remodeling, which are essential for restoring tissue integrity (Abdoli et al., 2017). The observed increase in TGF-β expression with Fe-Cur-P treatment suggests a robust stimulation of these reparative processes, likely mediated by the nanoparticle’s ability to mitigate oxidative stress and create a favorable microenvironment for regeneration. This finding is particularly significant in the context of diabetic wounds, where impaired TGF-β signaling is a known contributor to delayed healing (Mokoena et al., 2018). Consistent with prior reports, reduced or dysregulated TGF-β signaling in diabetic wounds is closely linked to impaired granulation tissue formation and collagen remodeling; therefore, the restored TGF-β staining observed here supports a reparative shift at the tissue level rather than a purely anti-inflammatory effect. Comparative analysis with the control, Cur, and Fe-Cur groups revealed that Fe-Cur-P’s enhanced TGF-β modulation may be attributed to its sustained release of carboxy-PTIO, which reduces nitric oxide levels, thereby alleviating the oxidative inhibition of TGF-β pathways. These results highlight Fe-Cur-P’s potential to bridge the gap between inflammation resolution and tissue regeneration, offering a dual therapeutic benefit.

Inflammation-related cytokines (IL-6, IL-1β, and TNF-α) were also stained on day 14 (Figures 4c,d). These cytokines were highly expressed at the wound site in the control group, reflecting the chronic inflammatory state typical of diabetic wounds, which perpetuates tissue damage and hinders healing (Yamauchi et al., 2016). However, Fe-Cur-P outperformed the control, Cur, and Fe-Cur groups by effectively suppressing IL-6, IL-1β, and TNF-α expression, reducing their levels to those comparable to or lower than the normal wound group. This suppression is likely driven by the nanoparticle’s potent ROS and RNS scavenging capabilities, which disrupt the upstream signaling cascades (e.g., NF-κB and MAPK pathways) that induce cytokine production. The superior performance of Fe-Cur-P over Cur and Fe-Cur suggests that the incorporation of carboxy-PTIO enhances its anti-inflammatory efficacy, possibly by targeting nitric oxide-mediated amplification of inflammation. Quantitative densitometry of the immunohistochemical staining revealed a reduction in cytokine expression by approximately 70%–85% in the Fe-Cur-P group compared to the control, underscoring a dose-dependent and sustained anti-inflammatory effect.

This study demonstrates the successful development of iron-curcumin-coordinated nanoparticles (Fe-Cur-P) as a novel therapeutic platform for enhancing diabetic wound healing. By addressing the inherent limitations of curcumin’s poor water solubility and bioavailability through coordination with iron ions, Fe-Cur-P nanoparticles achieved a uniform size of 9.8 ± 0.6 nm and exhibited controlled release of carboxy-PTIO, with a drug loading efficiency of 21.31%. The nanoparticles effectively scavenged reactive oxygen species (ROS) and reactive nitrogen species (RNS), as evidenced by significant reductions in DPPH·, ABTS+·, and hydroxyl radical levels. In vitro experiments with RAW 264.7 macrophages revealed substantial suppression of LPS-induced apoptosis and downregulation of pro-inflammatory cytokines (IL-1β, IL-6, TNF-α, and COX-2), highlighting their potent anti-inflammatory properties. In vivo studies using a diabetic mouse model further confirmed the therapeutic efficacy of Fe-Cur-P, with wound closure rates reaching 93.5% ± 2.4% by day 14, alongside enhanced tissue regeneration marked by increased vascularization, collagen deposition, and hair follicle regeneration. Histological and immunomodulatory analyses underscored reduced inflammation, elevated TGF-β expression, and diminished myeloperoxidase (MPO) activity, suggesting a comprehensive mechanism for promoting wound healing. These findings establish Fe-Cur-P nanoparticles as a promising candidate for clinical translation, offering a targeted approach to modulate inflammation and oxidative stress in chronic diabetic wounds. Future research will focus on optimizing scalability, evaluating long-term safety, and exploring gender-balanced models to broaden clinical applicability.