All-natural medicine food homology herb-based self-gelling hemostatic powder for rapid hemostasis and diabetic wound management

Abstract

Diabetic wound management is characterized by a significantly delayed healing process and the induction of severe complications due to the significant alteration of the immune microenvironment. Additionally, uncontrolled bleeding after trauma further exacerbates the wound healing disorder. Medicine food homology herbs have a promising application prospect for diabetic wound management, however, their inherent properties lead to poor therapeutic effects. Self-gelling powder, a novel wound adjunct, retains the mechanical properties and adhesiveness, while also having the ability to adapt various wounds. Inspired by the concept of medicine food homology, this study designed a self-gelling powder (NR@Pue), composed of puerarin (Pue) and charred Nelumbinis Rhizomatis-derived carbon dots (NR-CDs), to collectively accelerate rapid hemostasis and diabetic wound repair. This powder rapidly self-gelled upon absorbing interfacial water, thereby forming a tight adhesion with the moist tissue surface. Consequently, in a diabetic mouse skin full-thickness defect model, NR@Pue powder accelerated hemostasis, promoted the M1 to M2 polarization of macrophages, enabling the wound to escape the inflammatory phase and enter the proliferative phase, thus enhancing collagen deposition, angiogenesis, and epithelial regeneration. Take together, NR@Pue demonstrated outstanding wound repair capabilities, providing a potential solution for the combined treatment of chronic wounds using self-gelling technology and medicine food homology concept.

Graphical abstract

Supplementary Information

The online version contains supplementary material available at 10.1186/s12951-026-04039-4.

Introduction

Diabetes, a widespread metabolic disorder, has become a major global public health issue [1]. For diabetic patients, skin ulcer complications are usually accompanied by chronic inflammation [2]. Abnormally elevated blood sugar levels lead to the accumulation of M1 type macrophages at the wound site, thus hindering the transformation of the healing process from the initial proliferation stage to the remodeling stage [3]. As reported, the physiological process of skin wound healing is extremely complex, mainly consisting of the consecutive stages of hemostasis, inflammation, proliferation, and remodeling [4]. Uncontrolled bleeding after severe trauma represents a key factor contributing to the high mortality rate during the healing process, making hemostasis as the crucial initial stage of healing particularly important [5]. Despite the wide application of nanofibers, nanoparticles, polymer bio-adhesives and zeolites on accelerating tissue regeneration, the inherent limitations still seriously affect their practical application [6]. Thereafter, in clinical practice, developing and selecting novel wound dressings with rapid hemostasis, good adhesion, qualified biocompatibility and promotion of diabetic wounds have become a widely concerned research direction in the field of medical materials [7].

Recently, self-gelling powder has increasingly gained attention in the field of hemostasis and wound management [8]. This material combines the mechanical support, tissue adhesion, and wound closure advantages of hydrogels, while retaining the high blood absorption capacity and the convenience of irregular wound filling of powder materials [6]. After spreading on the wound, self-gelling powder quickly transforms into an in-situ hydrogel on the moist tissue, effectively closing the bleeding wound and significantly improving the hemostatic performance. Additionally, the three-dimensional network structure in the powder regenerative gel provides a natural advantage for continuous drug delivery. Increasing evidence indicated that self-gelling powder shows much superiority to conventional therapies in treating diabetic wounds [9]. Nevertheless, current research on self-gelling powder constructed from polymer macromolecules mainly focuses on hemostatic performance, but neglects the subsequent wound repair [5]. More importantly, the chemical crosslinking agents or inducers used in the preparation of such hydrogels seriously damage the biocompatibility of self-gelling systems and are difficult to remove [10]. Besides, the drug loading performance of self-gelling is often overlooked, remarkably restricting its clinical application [11]. Therefore, developing self-gelling powder that does not require chemical modification or carriers and has significant pharmacological activity, provides an ideal and highly promising solution for diabetic wounds management.

Medicine food homology, refers to herbal materials that can be used both for medicine and as food, and highlights the close connection and mutual transformation between food and medicine [12]. Due to their strong biological activity and low toxicity, these plants have received extensive attention [13]. Pueraria lobata, as a medicine food homology herb, has a culinary history dating back to the “Shennong’s Herbal Classic”, which is known as “official medicine” [14]. Puerarin (Pue), an isoflavone compound derived from Pueraria lobata, is commonly used as a natural antioxidant and dietary supplement, and has a series of pharmacological activities, including pro-healing, cardiac protection and neuroprotection in diabetes treatment [15, 16]. Currently, the three main formulations of Pue, including injection, eye drops and freeze-dried powder, have been approved by the China National Food and Drug Administration for clinical treatment of inflammatory diseases, cardiovascular diseases and cerebrovascular diseases [17–19]. Increasing evidence indicates that Pue also improves chronic wound healing through multiple mechanisms, mainly involving alleviating cellular oxidative stress, accelerating fibroblast migration kinetics, enhancing collagen deposition, promoting angiogenesis and tissue regeneration [20]. However, Pue belongs to the biopharmaceutics classification system-IV (BCS-IV) class of drugs, with characteristics of low water solubility and poor permeability, resulting in limited absorption in the body and low bioavailability [21]. Increasing the dosage does not effectively enhance the bioavailability and may even lead to toxicity and side effects. In recent years, various nanotechnologies and preparation techniques have been progressed to improve the bioavailability of Pue, including microemulsions and dendrimer macromolecules, nanoparticles and nanocrystals [22]. However, these strategies have some problems such as low drug loading capacity, carrier toxicity, and poor biological safety. Notably, without additional modification, Pue with structure of hydrophilic pyran glucose and hydrophobic isoflavones, self-assembles into a hydrogel after heat-cooling process, thus solving application limitations caused by low hydrophilicity and low bioavailability [23]. Previously, SDF-1α/Pue hydrogel has been fabricated to promote cardiac repair, where Pue is not only used for delivering SDF-1α but also functions as a therapeutic agent for macrophage polarization [24]. Wan et al. constructed a hydrogel composed of hyaluronic acid, silanol and Pue, aiming to promote wound healing in diabetic patients [20]. Zhai et al. developed a multifunctional wound dressing by incorporating mesoporous polydopamine nanoparticles containing genistein into carboxymethyl chitosan/oxidized alginate-based hydrogels, which helps promote the healing of infected wounds [25]. Yang et al. successfully fabricated a fiber-water gel composite dressing based on methacrylic acid gelatin using electrospinning and 3D printing technology, which contained N-halogenated amine antibacterial agents 1 and Pue, could prevent infections and promote skin tissue regeneration [26]. Zeng et al. constructed injectable self-healing chitosan@puerarin hydrogel to inhibit the inflammatory reaction mediated by miR-29ab1 and promote the healing of diabetic wounds [27]. Despite the rapid progress of Pue hydrogel on wound management, the biological safety of these hybrid hydrogels still requires attention due to the addition of inactive components.

Nelumbinis Rhizomatis Nodus (NR), a common medicine food homology herb, is rich in multiple active components such as polyphenols, triterpenoids, flavonoids, and steroid compounds [28]. Nelumbinis Rhizomatis Nodus roots and its extracts display immunomodulatory, lipid-lowering, blood pressure-lowering, blood sugar-lowering, antioxidant, antibacterial capacities [29]. Charred Nelumbinis Rhizomatis Nodus, one of the representative charcoal drugs with astringent and hemostatic effects has been recorded in the Chinese Pharmacopoeia (2020 edition), showing a history of over 2000 years in Chinese medicine clinic [30]. Interestingly, the high-temperature carbonization process of the charcoal drugs and the high-temperature pyrolysis process of carbon dots (CDs) are exactly the same [31]. Previous study showed that CDs derived from charred Mume fructus presented a remarkable rapid hemostatic and promoting wound healing effects [32], which is attributed to the abundant active groups of CDs with super-small particle size. In addition, the addition of CDs increases the mechanical properties, adhesion, and stability of hydrogels [33, 34]. Through rational design, hydrogel nanoplatforms constructed from CDs can further promote the development of wound repair strategies, especially hydrogels derived from medicine food homology herbs.

This study introduced a medicine food homology strategy to combine Pueraria lobata with charred Nelumbinis Rhizomatis Nodus, and developed a self-gelling powder, mainly composed of Pue loaded with NR-CDs. For the first time, the self-gelling powder co-assembled from all-medicine food homology substances, which has the functions of rapid hemostasis and promoting the repair of diabetic wounds. Pue and NR-CDs are assembled in water solution through hydrogen bonds to form a water gel network. The NR@Pue powder was prepared from freeze-drying. After re-hydrating, the regenerated NR@Pue retained excellent self-healing, injectable and adhesive properties. Therefore, the NR@Pue powder could absorb blood and tissue fluid from the injured area and form a hydrogel including the liver, tail, and skin wounds. Specifically, the regenerated hydrogel has the functions of rapid hemostasis, regulating immune environment, promoting angiogenesis, collagen deposition and extracellular matrix remodeling during the critical stage of wound repair. In view of the high-safety as a medicine food homology substance, Pue and NR-CDs provide a promising auxiliary strategy for safer and more effective wound management therapies.

Results and discussion

Synthesis, characterization and free scavenging activity of NR-CDs

In this study, using charred Nelumbinis Rhizomatis Nodus (NR) as the carbon source, NR-CDs were synthesized through the hydrothermal and dialysis method, with a yield of approximately 6.52%. Firstly, transmission electron microscopy (TEM) was characterized to visualize the morphology of NR-CDs. As a result, the prepared NR-CDs showed monodisperse and uniform size distribution, with an average particle size of 2.67 ± 0.51 nm (Fig. 1A). Further, high-resolution TEM images revealed that NR-CDs had lattice stripes similar to those of graphene (100) (Fig. 1B), with a lattice spacing of 0.218 nm, indicating a graphite-like structure. The uniform spherical shapes with a narrow particle size distribution and clear lattice fringes suggested the high purity and carbonization efficiency of NR-CDs [35]. Additionally, X-ray diffraction spectra (XPS) showed a characteristic peak of NR-CDs appeared at 25.76°, corresponding to the (002) crystal plane of graphite carbon, indicating the successful preparation of NR-CDs with the complete crystal structure (Fig. 1C). The ζ-potential of NR-CDs was approximately − 6.35 mV (Figure S1). A clear peak at 279 nm in UV–vis spectra represented the π-π* transformation of aromatic C = C bonds (Figure S2A). Meanwhile, the optimal excitation wavelength and emission wavelength of NR-CDs were approximately 395 nm and 541 nm, respectively, and displayed excitation-dependent emission behavior (Figure S2B and S2C).

Fig. 1.

Preparation and characterization of NR-CDs. (A) TEM image and size distribution of NR-CDs, and (B) High resolution-TEM and lattice spacing of NR-CDs. (C) XRD spectrum and (D) Raman spectrum of NR-CDs. (E) FT-IR spectrum of NR and NR-CDs. (F) XPS survey spectra, and high-resolution XPS spectra of (G) C1s and (H) O1s for NR-CDs. (I) UV–vis absorption spectrum and (J) ·O₂⁻ scavenged by NR-CDs. (K) UV–vis absorption spectrum and (L) ·OH scavenged by NR-CDs. (M) UV–vis absorption spectrum and (N) ABTS⁺ scavenged by NR-CDs. (O) UV–vis absorption spectrum and (P) DPPH· scavenged by NR-CDs

Subsequently, the structure and surface groups of NR-CDs were characterized by Raman spectroscopy, Fourier transform infrared (FT-IR) spectroscopy, and X-ray photoelectron spectroscopy (XPS). In the Raman spectrum, the D-band and G-band of NR-CDs corresponded to two characteristic peaks at 1385 cm⁻¹ and 1580 cm⁻¹, respectively, with a 0.877 of ID/IG ratio, suggesting that NR-CDs had obvious surface defects and a degree of graphitization (Fig. 1D). FT-IR spectrum of NR-CDs displayed characteristic peaks at 1045, 1452, 1633, 2925, and 3417 cm^− 1, respectively, corresponding to the stretching vibrations of C-O, C = O, C = C, C-H and O-H groups (Fig. 1E) Notably, NR-CDs retained the surface groups of the NR carbon source, with characteristic peaks at 1010, 1440, 1620, 2920, and 3290 cm⁻¹, corresponding to C-O, C = O, C = C, C-H and O-H groups, respectively. Further, X-ray photoelectron spectroscopy (XPS) was used to analyze the chemical composition and functional groups of NR-CDs. The XPS full spectrum in Fig. 1F showed that NR-CDs were mainly composed of C and O, accounting for 60.28% and 38.75%, respectively. Specifically, the high-resolution XPS spectrum of C1s could be decomposed into three binding energy peaks at 287.9 eV, 286.7 eV, and 284.8 eV, corresponding to C = O, C-O, and C-C/C = C bonds, respectively (Fig. 1G). Meanwhile, the O1s peaks at 533.6 eV, 532.9 eV, and 531.8 eV corresponded to C = O-OH, C = O, and C-O bonds, respectively (Fig. 1H). Therefore, above results confirmed the successful synthesis of ultrasmall-sized NR-CDs with rich surface functional groups, providing crucial structural support for the generation of pharmacological activity from CDs. Besides, the significant water solubility of NR-CDs might be attributed to the abundant hydrophilic groups on its surface, including carboxyl and hydroxyl groups.

Chronic diabetic wound sites generate excessive reactive oxygen species (ROS), ultimately impeding wound repair [36]. Thereafter, eliminating ROS at the wound site is an effective strategy to alleviate inflammation and promote wound repair [37–39]. A system containing riboflavin and methionine was constructed to evaluate the ·O₂⁻ clearance ability of NR-CDs. Specifically, ·O₂⁻ generated from the oxidation process of methionine, thus oxidizing nitroblue tetrazolium chloride (NBT) into a blue-purple NBT triphenylamine with strong absorption at 560 nm. After adding NR-CDs, the UV–vis absorption of NBT significantly decreased, indicating that NR-CDs can effectively remove ·O₂⁻ (Fig. 1I). Additionally, as the concentration of NR-CDs reached to 125 µg/mL, the scavenging rate of ·O₂⁻ exceeds 70% (Fig. 1J). Subsequently, we constructed the Fenton reaction system consisted of 3,3’,5,5’-tetramethylbenzidine (TMB), Fe²⁺ and H₂O₂ to detect the ·OH scavenging ability of NR-CDs. Under the action of the oxidant, TMB is oxidized into a green solution, with distinct absorption peak at 652 nm. The addition of NR-CDs caused a decrease in the UV–vis of TMB in the system (Fig. 1K). Quantitative data indicated that 250 µg/mL of NR-CDs could achieve a nearly 60% ·OH scavenging rate, confirming that NR-CDs have a significant ability to remove ·OH free radicals (Fig. 1L). The ·O₂⁻ and ·OH free radicals have extremely short half-lives and high reactivity, which makes their detection quite challenging. After captured by DMPO, the short-lived free radicals are converted into stable free radicals, followed by identified using electron spin resonance (ESR) spectroscopy [40]. According to the ESR spectra, the signals of the characteristic peaks of ·O₂⁻ and ·OH were significantly decreased after adding NR-CDs and showed a dose-dependent effect (Figures S3A and S3B), suggesting that CDs could effectively remove ·O₂⁻ and ·OH. Further, 2, 2’-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS) would generate green ABTS⁺ under the action of the oxidant, with the significant absorbance at 734 nm [41]. As shown in Fig. 1M and N, the ABTS⁺ scavenging ability of NR-CDs showed a dose-dependent behavior. Over 80% of ABTS⁺ free radicals were scavenged by 500 µg/mL. Besides, the scavenging ability of 1,1-diphenyl-2-nitrobenzhydrazide radical (DPPH·) was proportional to the antioxidant capacity of the substance [42]. As shown in Fig. 1O and P, the clearance rate of DPPH· by NR-CDs increased with the increased concentration, reaching 55% at 7.8 µg/mL and over 80% at 125 µg/mL, demonstrating a concentration-dependent antioxidant potential. In summary, these results clearly suggested that NR-CDs showed remarkable dose-dependent free radical clearance characteristics, however, a quantitative comparison with native enzymes or alternative nanozymes need to be further evaluated. Given that the outbreak of ROS leads to the problem of diabetic wound repair, the ROS clearance activity of NR-CDs is expected to have potential application value in the prevention and treatment of chronic wounds.

In recent years, nanomaterials derived from herbal sources have shown great potential in promoting various types of wound healing [3]. Our previous study suggested that CDs prepared from carbonized Platycladus orientalis could promote endogenous hemostasis by triggering the exogenous coagulation pathway and platelet activation [43]. Zhu et al. synthesized CDs using Ligusticum wallichii as the precursor, and these carbon dots exhibited strong antibacterial activity and showed effective wound healing in infected skin wound model with Staphylococcus aureus [44]. In this study, we innovatively propose the NR as carbon source, a medicine food homology charcoal drug to synthesize CDs. By integrating the overall regulatory concept of Chinese medicine’s theory of stopping bleeding through charcoal treatment with modern formulation technology, this study provides a model for the modernization of Chinese medicine and the development of original drugs. Compared with other CDs originated from herbal medicine, NR-CDs had the advantages of safer sources, stronger sustainability, and broader prospects for clinical transplantation.

Fabrication and characterization of NR@Pue powder

Pue, one of the representative herbal natural products, can self-assemble into thermosensitive hydrogels without any structural modification or additives via heating-cooling process [23]. In this study, the aqueous solution containing Pue and NR-CDs displayed a brown transparent color, and a light brown NR@Pue hydrogel was obtained through the heating-cooling process (Fig. 2A). Subsequently, hydrogel was freeze-dried into powder, aiming to adapt the various shapes of the wound. As shown in Fig. 2B, SEM image of Pue presented as slender fibrous structure. After the addition of NR-CDs, NR@Pue showed a punctate protrusion structure, showing completely different microscopic morphology with Pue. This phenomenon suggested that there might be significant interactions between Pue and NR-CDs, resulting in a huge change in the microstructure of the hydrogel. In addition, ζ-potential is closely associated with the stability of the hydrogel system. A relatively high charge value will cause a strong electrostatic repulsion between the particles, effectively preventing the particles from approaching and aggregating. As a result, the hydrogel remains in a uniformly dispersed state, reducing the possibility of precipitation, aggregation or phase separation, thereby enhancing the stability of the system [45]. In this study, compared with Pue (−7.45 mV), the ζ-potential of NR@Pue was approximately − 19.7 mV, showing higher stability (Fig. 2C). According to the results of UV-vis spectrum, Pue had two characteristic absorption peaks at 256 nm and 306 nm. The addition of NR-CDs led to a slight redshift of the characteristic absorption peaks of NR@Pue, indicating a significant interaction between NR-CDs and Pue (Fig. 2D). XRD test was performed to further analyze the crystal structure changes of the hydrogel. The typical diffraction peaks of Pue in the XRD pattern were located at 6.62°, 11.68°, 13.88°, 15.98°, 18.92° and 23.48°, respectively, suggesting the crystal structure of Pue powder. Nevertheless, NR@Pue presented a wide diffraction peak in the 20–25° range, indicating that the hydrogel formed by Pue and NR-CDs was amorphous (Fig. 2E). This might be due to the interference of the amorphous structure of NR-CDs on the crystal structure of Pue. FTIR spectra showed the characteristic peaks of O − H, C-H, C = O, and C − O groups in Pue were located at 3210, 2900, 1630, and 1050 cm^− 1, respectively (Fig. 2F). After adding NR-CDs, the peak of O − H and C-H groups red-shifted to 3290 and 2930 cm^− 1, while the peak positions of C = O and C − O groups blue-shifted to 1620 and 1010 cm^− 1, respectively. This phenomenon could be attributed to the hydrogen bond interaction between NR-CDs and Pue during the co-assembly process. Despite the surface groups confirmed by the FT-IR data, the peak integration and deconvolution to measure the NR-CDs and Pue conjugation still need to be further investigated. According to the circular dichroism (CD) analysis results, both Pue and NR@Pue had a characteristic peak at 251 nm (Fig. 2G). Subsequently, rheological experiments were performed to further determine the mechanical properties of the hydrogel. As shown in Figs. 2H-2J, within the frequency range of 0.1–100 radians/second (or 0.01–10.01 Hz), the storage modulus (G’) was higher than the loss modulus (G’’), indicating the significant elastic properties of NR@Pue. However, NR@Pue showed better elastic properties and shear viscosity compared to that of Pue, suggesting that NR-CDs could significantly enhance mechanical and adhesive properties. Furthermore, the release characteristics of Pue from NR@Pue hydrogel were characterized by incubating in pH 7.4 PBS medium at 37 ℃. The NR@Pue hydrogel showed the sustained release of Pue, with the cumulative release amount exceeding 80% after 72 h (Figure S4). Thus, the sustained release characteristics of Pue in the hydrogel will ensure its therapeutic effect.

Fig. 2.

Fabrication and characterization of NR@Pue hydrogel and powder. (A) Schematic diagram of preparing process of NR@Pue hydrogel and powder. (B) SEM images of Pue and NR@Pue powder. (C) ζ-potential, (D) UV–vis absorption spectrum, (E) XRD spectrum, (F) FT-IR spectrum and (G) circular dichroism spectrum of Pue and NR@Pue. The storage modulus (G’) and the loss modulus (G’’) within the and (H) 0.01–10 Hz, (I) frequency range of 0.1–100 rad/s and (J) Shear viscosity of NR@Pue and Pue hydrogel, measured at 0.1% strain. (K) Self-gelling, (L) injectability (M) adhesion property of NR@Pue hydrogel. P value: “**” p < 0.01

The hemostatic powder material is highly regarded due to its ability to adapt to the irregular shape of the wound and its effective absorption of blood. To enhance the hemostasis and wound closure capabilities, NR@Pue was freeze-dried to obtain a powder form. As shown in Fig. 2K, the NR@Pue powder could rapidly revert to a gel state upon contact with water. The hemostatic and cyclic regeneration capabilities of self-gelling biomaterials are closely related to their hydrophilicity [46]. Meanwhile, NR@Pue hydrogel displayed good injectability (Fig. 2L). During the wound healing process, adhesion is important for preventing the hydrogel from falling off and maintaining the tissue structure [47]. Thereafter, we analyzed the adhesion performance of NR@Pue hydrogel. As shown in the Fig. 2M, NR@Pue firmly adhered to the major organs of mice. This indicated that NR@Pue has broad application prospects in the field of tissue adhesion and hemostasis, and can be effectively applied in seamless tissue repair and wound healing. Additionally, placing the hydrogel at the finger joint and bending it at different angles could quickly recover, which to some extent could prove the ability of the hydrogel to withstand pressure and restore its shape (Fig. 2N). Nevertheless, to develop into medical hydrogel applications, further evaluations on burst pressure, tensile strength, and degradation rate of NR@Pue hydrogel should be performed [8, 48]. Moreover, it is essential to investigate the crosslinking densities or rheological qualities of NR-CDs since these elements may affect injectability and stability. Together, through hydrogen bond interactions as the driving force, this hydrogel with excellent injectability, tissue adhesiveness, adaptability and thermal stability was successfully prepared by co-assembling Pue and NR-CDs. Particularly, it was demonstrated for the first time that the NR@Pue powder has self-gelation ability, which can rapidly form a hydrogel after being moistened, which gives it great potential advantages in hemostasis and wound repair intervention.

In vitro anti-oxidant, pro-healing and hemostatic capability of NR@Pue powder

The healing process of diabetic wounds is remarkably hindered due to its high susceptibility to bacterial growth and infection, contributing to chronic inflammation, excessive production of ROS, and severe oxidative stress [2]. Thereafter, to evaluate the ROS scavenging ability of NR@Pue, we conducted in vitro experiments using L929 and RAW264.7 cells as the model system. Specifically, lipopolysaccharide (LPS) was used to induce an in vitro inflammatory model in L929 and RAW264.7 cells, and then Pue, NR-CDs, and NR@Pue were intervened, respectively (Fig. 3A). Through the detection with the dichlorodihydrofluorescein diacetate (DCFH-DA) probe, after 1 µg/mL LPS stimulation, a strong green fluorescence was observed due to the production of a large amount of ROS, indicating the successful establishment of the inflammatory model (Fig. 3B). After treatment with Pue or NR-CDs, the intracellular ROS significantly decreased, especially for NR@Pue the weakest green fluorescence of ROS. Previously, Pue-based hydrogels have been proved to promote tissue repair by reducing oxidative damage and enhancing the viability of cells under stress conditions [20]. In addition, Pue inhibits the NLRP3 activation of inflammatory corpuscles of endothelial cells through ROS-dependent oxidation pathway [49]. According to the above data, NR-CDs showed remarkable ROS scavenging capacity. Therefore, the significantly enhanced ROS scavenging and antioxidant effects of NR@Pue might be attributed to the collective action of Pue and NR-CDs. Further, using LPS-induced L929 and RAW264.7 cells as cell model, we simulated the effect of NR@Pue on the migration of skin cells under inflammatory conditions. As shown in Fig. 3C, compared with the LPS group, Pue, NR-CDs, and NR@Pue treatment significantly enhanced the migration rate of L929 cells. Especially for NR@Pue group, the migration rate reached approximately 42% after 24 h, while in LPS group was approximately 27% (Fig. 3C and D). Therefore, the above results confirmed that NR@Pue could effectively reduce ROS production, alleviate oxidative stress, protect cells from ROS-induced damage, and promote wound migration.

Fig. 3.

Assessment of intracellular oxidative eliminating, in vitro pro-healing, and hemostatic ability of NR@Pue powder. (A) Illustration of cell experiments process. (B) Fluorescence images of LPS-induced L929 cells or RAW 264.7 cells using DCFH-DA probe (scale bar: 100 μm). (C) Cell scratching healing and (D) quantitative analysis of LPS-induced L929 cells after treated with Pue, NR-CDs and NR@Pue at 0 and 24 h (scale bar: 200 μm). (E) Representative images, (G) hemostatic time, and (H) blood loss of rat liver hemorrhage model. (F) Representative images, (I) hemostatic time, and (J) blood loss of rat tail amputation model with the treatment of Pue, NR-CDs and NR@Pue powder. P value: “*” p < 0.05, “**” p < 0.01 and “***” p < 0.001

Wound repair is a complex process consisting of four consecutive overlapping stages, and the most successful outcome depends on the treatment of all stages [50]. Hemostasis is the earliest stage of wound repair. Therefore, hydrogels with rapid hemostatic function can accelerate wound repair. In vivo hemostatic potential of NR@Pue was evaluated using rat liver trauma and tail amputation models. In the rat liver trauma hemorrhage model, a 3-mm deep and 10-mm long incision was made in the liver. The control group recorded the highest blood loss of 2.98 g, while the blood loss of Pue and NR-CDs was reduced to 2.31 and 1.13 g, respectively. Notably, the blood loss of the NR@Pue group was only 0.46 g, approximately one-sixth of the control group (Fig. 3G). Additionally, the hemostatic time of the control group was 275 s. Pue and NR-CDs achieved closure by contacting with blood, further shortening the hemostasis time to 128 s and 121 s, respectively. Nevertheless, NR@Pue treatment significantly reduced the time to 77 s, approximately one-fourth of the control group (Fig. 3H). The bloodstain images taken after hemostasis were consistent with the quantitative results (Fig. 3E). In the rat tail amputation model, an amputation incision of 7-cm from the rat tail was made (Fig. 3F). Compared with the control group (blood loss: 2.82 g, hemostatic time: 213 s), NR@Pue treatment effectively decreased the blood loss and hemostasis time to 0.52 g and 100 s, demonstrating a strong hemostatic effect (Fig. 3I and J). Above results indicated that NR@Pue powder encountered blood, quickly adhered to the tissue and achieve hemostasis in the liver and tail, with very little blood loss. Therefore, the self-gelling NR@Pue powder exhibited excellent hemostatic properties. Upon contacting with blood, the powder rapidly formed a hydrogel and tightly adhered to the surface of the surrounding tissues at the bleeding site, providing tissue sealing for hemostasis. In this study, the hemostatic effects of Pue, NR-CDs and NR@Pue were preliminarily investigated at a fixed dose, however, the hemostatic performance at different doses still requires further experimental design for further study.

In our previous study, carbonized Platycladus orientalis derived CDs promoted endogenous and exogenous hemostasis through activation of platelet and coagulation pathways [43]. In addition, CDs prepared from charred Mume Fructus with good biosafety and remarkable ROS scavenging ability, has been confirmed to enhance hemorrhagic wound healing [32]. Zhang et al. reported that Scutellariae Radix Carbonisata-derived CDs showed the hemostatic effects, which could inhibit the release of inflammatory cytokines and activate fibrin system and endogenous coagulation by regulating myD88/NF-κB signaling pathway [51]. In this study, NR@Pue exhibited a stronger hemostatic effect than Pue, however, its deep hemostatic effect remains unclear. Moreover, carefully designed experiments are needed to deeply explore the hemostatic mechanism of NR@Pue and to determine the role of CDs in the hemostasis of NR@Pue.

Besides, nanomaterials with negative charges attract positively charged platelets and red blood cells in the blood through electrostatic interaction, causing these cells to gather on the surface of the material and accelerating the formation of blood clots, thereby achieving hemostasis [52]. Additionally, positively charged materials can adsorb and activate coagulation factors, accelerating the coagulation process. For instance, positively charged silk protein-based hydrogels promote platelet activation through electrostatic interaction, enhance the formation of the fibrin network, and shorten the clotting time [53]. As previously reported, an appropriate ζ-potential helps maintain the stability of the material in the blood and avoids material failure due to excessive adsorption or aggregation of blood components [54]. By adjusting the ζ-potential to achieve a balanced surface charge of the material, it is ensured that the material can continue to function during the hemostasis process. Moreover, identifying the density of functional groups, such as carboxyl or hydroxyl, might improve the mechanistic connection between biological activity and charge. Therefore, it is necessary to conduct a further assessment of the relationship between the surface charge density and functional groups of NR@Pue with its hemostatic effect.

NR@Pue powder facilitates rapid healing of diabetic wounds on mice

To investigate the in vivo effects of NR@Pue powder in chronic wound repair, we firstly established a diabetic C57BL/6J mouse model, followed by created a 10 mm circular full-thickness wound on the back. The wounds were treated with PBS, Pue powder, NR-CDs powder, and NR@Pue powder, respectively, and the changes in wound morphology were recorded on days 0, 3, 6, 9, and 12 (Fig. 4A). Meanwhile, wounds created from healthy mice served as the NC group. Figure 4B showed the changes in wound morphology during 12-day’s treatment. As shown in Figure S5A, during the whole treatment, the body weight of all groups of mice did not change significantly, suggesting that the good in vivo biosafety of NR@Pue hydrogel. Furthermore, the blood glucose levels of the NC group mice were less than 10 mmol/L, while those of the model group and the treatment group remained at around 20 mmol/L (Figure S5B), indicating that the entire treatment process was maintained in a hyperglycemic state and did not reach the normal level. Therefore, the local treatment with NR@Pue hydrogel could not affect blood glucose. Future research efforts will require the control of systemic blood glucose or the combination treatment with insulin or other metabolic regulators. Figure 4C simulated the process of wound repair with different interventions. Further, the heatmap of wound area results showed that significant differences were observed among the Pue, NR-CDs, NR@Pue treatment groups and the Mod group on day 3 (Fig. 4D). Compared with the Mod group, the treatment groups of Pue, NR-CDs, and NR@Pue all showed a faster healing rate, with wound closure rates reaching 80.00 ± 7.79%, 81.28 ± 6.19%, and 90.26 ± 3.32% on day 6, respectively (Fig. 4E). These effects could be ascribed to the free radical scavenging property of NR-CDs and the therapeutic advantages of the self-gelling powder. Notably, on day 12, the wounds treated with NR@Pue powder were almost completely healed, with a healing closure rate of 98.27 ± 0.73%. This remarkable therapeutic effect is likely to be caused by the collective action of NR-CDs and Pue. The residue of hydrogel on the wound surface could not be ignored, as it may stimulate the wound tissue, trigger inflammatory or immune responses, affect the healing process of the wound, and even lead to scar formation or an increased risk of infection [55]. In this study, after 12 days’ treatment of NR@Pue, almost no residue was observed on the wounds and was suitable for clinical transplantation.

Fig. 4.

In vivo evaluation of NR@Pue powder on mouse diabetic wounds. (A) Establishment of the full-thickness diabetic skin wound and NR@Pue treatment process. (B) Representative images of the wounds area changing from day 0 to 12. (C) Illustration showing the wound repair process at 0, 3, 6, 9 and 12 d. (D) Wound area (cm²) visualized by heatmap within 12 days. (E) Wound closure (%) treated during the 14 day-treatment. (F) HE staining and (G) Masson staining images of the wounds at day 12 (scale bar: 1 mm, 200 μm). P value: “*” p < 0.05, “**” p < 0.01 and “***” p < 0.001

HE staining was further performed to assess the ability of NR@GA powder to accelerate wound healing. After 12 days of treatment, all treatment groups exhibited signs of granulation tissue formation and epidermal regeneration (Fig. 4F). Compared with the Mod group, the Pue, NR-CDs, and NR@Pue treatments showed narrower granulation gaps. Specifically, the wounds treated with NR@Pue formed a large number of hair follicles, presenting the most regular epidermal structure. Further, the collagen deposition at the wound site was visualized through Masson staining was used to evaluate. By the 12th day, the NR@Pue group displayed the highest collagen content, with more uniform collagen fibers arrangement. In contrast, other treatment groups had less collagen deposition and more disordered collagen fiber arrangement (Fig. 4G). Together, NR@Pue, with effective free radical elimination and hemostatic effect at the wound site, collectively promoted epithelial regeneration, dermal reconstruction, and the regeneration of skin appendages.

In this study, NR@Pue was locally administrated on the skin wounds, however, its pharmacokinetic characteristics remain unclear. According to previous report, the pharmacokinetic study of Pue reveals its species specificity, with two-compartment open model observed in rat and canine while three-compartment open mode in rabbits [56]. After intravenous injection, Pue can be detected in most organs, including the heart, lungs, stomach, liver, breast, kidneys, spleen, and even can cross the blood-brain barrier [57]. As substrates for the activity of P-glycoprotein, multidrug resistance-associated proteins and various metabolic enzymes, the pharmacokinetics of Pue can be affected by different pathological states and drug interactions [17]. Prescriptions containing Pueraria lobata such as Gegen Qinlian Decoction and Compound Longnao Pill, represent compatibility “herb pair” in Chinese medicine [58]. The compatible herbs in the formula may affect the pharmacokinetic characteristics of Pue. Studies have shown that compared with the Pue extract, the Pue in the Gegen Qinlian Decoction has more effective absorption and a slower elimination in the plasma of rats [56]. As previously reported, glycyrrhizic acid significantly inhibits the intestinal absorption of Pue, however, high concentrations of baicalin and berberine can promote its absorption [56]. Moreover, the intestinal absorption of Pue can be improved by Schisandra chinensis extract, Angelica sinensis extract, Angelica root extract, and Ligusticum chuanxiong root extract. Ligusticum enhances the absorption and pharmacokinetics of Pue through increasing the solubility of Pue, regulating the efflux of P-glycoprotein, and reducing the expression of claudin-5 [56]. However, due to the complexity of the ingredients in traditional Chinese medicine combinations, the interaction mechanisms of most herb pairs are still unclear. In this study, the influence of NR-CDs on the pharmacokinetic properties of Pue in NR@Pue hydrogel is worthy of further exploration.

In vivo immune regulatory, angiogenesis and collagen deposition efficacy of NR@Pue powder

Chronic inflammation is a hallmark feature of diabetic wounds, characterized by the continuous polarization of local macrophages to the pro-inflammatory M1 phenotype, causing to the continuous release of pro-inflammatory cytokines (such as TNF-α and IL-6) [59]. Effective tissue repair requires that macrophages at the wound site promptly transform from the M1 to M2 phenotype. To investigate the effect of NR@Pue powder on macrophage polarization, we evaluated the expression of phenotypic markers and quantified the secreted cytokines. Figure 5A showed representative images of M1-type (CD86, green) and M2-type (CD206, red) macrophages stained in wound tissue on the 12th day. Compared with the Mod group, the expression levels of CD206 red fluorescence in the wound tissue after NR-CDs, Pue, and NR@Pue treatments was significantly increased, while the green fluorescence of CD86 was significantly decreased, indicating that NR@Pue could promote the polarization of M1-type macrophages at the wound site to the M2-type. Especially after NR@Pue treatment, the expression of CD206 in the wound tissue was the highest and that of CD86 was the lowest (Figure S6). Subsequently, we detected the release of pro-inflammatory factors TNF-α and IL-6 in the wound tissue. Compared with the Mod group, the expression levels of TNF-α and IL-6 in the NR-CDs, Pue, and NR@Pue groups were significantly reduced (Fig. 5B and D). The expression levels of TNF-α were reduced by 73.13%, 70.72%, and 95.04%, respectively (Fig. 5C). The expression levels of IL-6 were decreased by 44.43%, 66.48%, and 87.73% respectively (Fig. 5E). These results further confirmed that NR@Pue promotes the polarization of macrophages, thus repairing the immune microenvironment. Moreover, the sustained release characteristic of the gel system prolongs the free radical scavenging and anti-inflammatory duration of NR-CDs, reduces the interference of oxidative stress on repair, and accelerates the transition from the inflammatory phase to the proliferative phase. To gain more comprehensive understanding of the dynamic changes in macrophage polarization and the resolution of inflammation, it is necessary to conduct pathological staining and immunofluorescence analysis at more time points during the wound healing process, such as on day 3, 6 and 9.

Fig. 5.

Immunomodulatory, angiogenesis, collagen deposition capacity of the NR@Pue powder. (A) Representative images of immunofluorescence staining for CD86 (Green), CD206 (Red) and DAPI (Blue) in diabetic wound tissues on day 12 (scale bar: 50 μm). Representative images of immunofluorescence staining of (B) TNF-α, (D) IL-6, (F) α-SMA, (H) CD31, (J) Collagen I and (L) Collagen III of diabetic tissues on day 12 (scale bar: 50 μm). Quantification of (C) TNF-α, (E) IL-6, (G) α-SMA, (I) CD31, (K) Collagen I and (M) Collagen III of wound tissues on day 12. All data were shown as mean ± SD (n = 3). P value: “*” p < 0.05, “**” p < 0.01 and “***” p < 0.001

The proliferative phase is a key stage during wound repair, involving fibroblast proliferation, neovascularization, and extracellular matrix deposition [60]. Myofibroblasts are particularly sensitive to mechanical stress at the wound site, and their expression of α-smooth muscle actin (α-SMA) enables them to generate strong contractile force [61]. This characteristic allows myofibroblasts to actively participate in wound contraction, thereby accelerating healing. As shown in Fig. 5F, after NR@Pue powder treatment, the number of α-SMA positive cells significantly increased, indicating that NR@Pue can promote the proliferation of myofibroblasts in vivo. Quantitative analysis showed that compared with the Mod group, the expression level of α-SMA was increased by 87.44% (Fig. 5G). Further, through CD31 immunofluorescence staining to visualize the formation of new blood vessels at the 12th day of the wound site. Compared with the Mod group, the wound tissue treated with NR@Pue powder showed a greater number and larger size of blood vessels (Fig. 5H). Quantitative analysis showed that the expression level of CD31 in the tissue of the NR@Pue powder treatment group was increased by 71.76% compared with the Mod group (Fig. 5I). These results are attributed to NR@Pue’s effective regulation of the immune microenvironment, promoting the increase of fibroblasts and the formation of new blood vessels. Additionally, fibroblasts gradually synthesize collagen and convert it into different crystalline forms of fibrillar morphology. Type III collagen predominates in the early stage of wound healing, and with cell proliferation and tissue remodeling, type I collagen gradually deposits. As shown in Fig. 5J and K, the expression level of collagen I in the NR@Pue group was significantly higher than that of the Mod group, Pue group, and NR-CDs group, respectively, by 7.7, 1.2, and 1.5 folds. Meanwhile, the expression level of collagen III was significantly higher in the NR-CDs group than in the Mod group, Pue group, and the control group, being 7.6, 1.9, and 3.0 folds higher, respectively (Fig. 5L and M). In summary, the above results indicated that the self-gelling NR@Pue powder could effectively regulate the inflammatory and proliferation stages during the chronic wound healing process, and accelerate vascular proliferation and collagen deposition during the remodeling stage, thereby significantly improving diabetic wound repair.

The high sugar environment of diabetic wounds greatly accelerates the proliferation of bacteria, thereby prolonging the inflammatory period of the wounds and extending the healing process of diabetic wounds. Eliminating bacteria plays a crucial role in promoting the healing of diabetic wounds [62]. At present, Pue-based hydrogel dressings have made a series of advancements in the healing of infected wounds. For example, Wei et al. developed a multifunctional hydrogel self-assembled by Pue, copper ion cross-linked adipic dihydrazide grafted xanthan gum and poly dopamine nanoparticles to treat irregular wound defects, which has appropriate mechanical properties, immunomodulation, antibacterial, antioxidant and anti-inflammatory activities [63]. In addition, a ROS-responsive hydrogel loading silk fibroin methacrylate, modified type III collagen, antibacterial peptides and Pue, demonstrates remarkable antibacterial efficacy, the ability to regulate inflammatory responses, and the capacity to control vascular functions, which provides a simple and effective approach for treating chronic diabetic infectious wounds [64]. In another study, an innovative multifunctional hydrogel system containing carboxymethyl chitosan, microalgae, hyaluronic acid and Pue, has been reported to possess remarkable properties such as exudate absorption, mechanical flexibility, hemostasis and antibacterial efficacy [65]. Therefore, in future research, we should conduct supplementary evaluations on antibacterial detection for common wound pathogens such as Staphylococcus aureus and Escherichia coli to further verify the potential efficacy of NR@Pue hydrogel.

In vitro and in vivo biocompatibility evaluation of NR@Pue powder

High biocompatibility of wound dressings is an essential prerequisite for their application. We first verified the cytocompatibility of NR@Pue powder through the CCK-8 assay. Specifically, three different cell lines (HUVEC, L929, and RAW264.7) were co-cultured with different concentration gradients of Pue, NR-CDs, and NR@Pue for 24 h, respectively. As shown in Fig. 6A and C, NR@Pue powder hardly caused any harm to the cells, with the cell viability rate over 80% at the highest concentration of 2 mg/mL. Additionally, live/dead staining was performed to verify the cytocompatibility of NR@Pue powder. The vast majority of living cells exhibit green fluorescence (Fig. 6E and F). In this study, we only evaluated cell compatibility under non oxidative stress conditions, however, exploring the cell protective activity induced by lipopolysaccharide is of great significance for further evaluating cell compatibility of NR@Pue. Subsequently, the hemocompatibility of NR@Pue powder was determined through a hemolysis test. During this process, Triton X-100, a non-ionic surfactant, served as the positive control group, causing hemolysis with red blood cells, with a hemolysis rate close to 100% (Fig. 6D). After co-incubation with NR@Pue powder, the supernatant was clear and transparent after centrifugation, with a hemolysis rate of less than 2%, indicating excellent hemocompatibility. To study in vivo biocompatibility, C57BL/6J mice were administered with NR@Pue powder for 14 days. On day 14, the heart, liver, spleen, lung, kidney, and blood samples were collected (Fig. 6G). As shown in Fig. 6H, after NR@Pue powder administration, the blood routine, and liver and kidney functions of the mice were all at normal levels. Further, the HE staining results indicated that there were no obvious inflammatory responses or tissue damage in the heart, liver, spleen, lung, and kidney of the NR@Pue-treated mice (Fig. 6I). Therefore, the prepared NR@Pue powder had excellent in vitro and in vivo biocompatibility, which might be attributed to the hydrogel constructed from all-natural medicine food homology herbs.

Fig. 6.

Biocompatibility evaluation of NR@Pue powder in vitro and in vivo. CCK-8 analysis showing the cytotoxicity of Pue, NR-CDs and NR@Pue powder in (A) HUVEC, (B) L929 and (C) RAW 264.7 cell lines for 24 h. (D) Hemolysis rate with different treatments. Live/dead staining images of (E) L929 and (F) RAW 264.7 cell lines after treated with Pue, NR-CDs and NR@Pue powder, respectively (scale bar: 100 μm). (G) Schematic diagram of the blood and major organs of the mice after 14-day administration. (H) Heatmap showing the key indicators of blood routine tests, and liver and kidney functions among different groups on day 14. (I) HE staining of major organs after 14-days’ administration (scale bar: 100 μm). All data were shown as mean ± SD. P value: “***” p < 0.001

Conclusion

Overall, we developed a novel self-gelling powder based on the concept of medicine food homology, containing Pue and NR-CDs. Upon contacting with water, this self-gelling powder quickly formed a gel with high adhesion, biocompatibility, and mechanical strength, thereby rapidly stopped bleeding and effectively promoted the diabetic wound repair. Benefiting from the free radical scavenging activity of NR-CDs, NR@pue could effectively eliminate ROS production induced by LPS and promote cell migration on in vitro cell model. In liver injury and tail amputation hemorrhage models, the performance of NR@Pue was better than that of Pue alone or NR-CDs, highlighting the combined hemostatic ability of Pue and NR-CDs. More importantly, NR@Pue remarkably regulated the immune microenvironment by promoting the transformation of macrophages from M1 to M2, decreasing the secretion of inflammatory factors such as TNF-α and IL-6, thus accelerating the transformation from the inflammatory phase to the proliferative phase during the wound healing process. In summary, we concluded that the NR@Pue system achieved combined treatment through a cascade pattern of “immune microenvironment regulation-collagen deposition-vascular network remodeling”, expanded the application of self-gelling technology in the delivery of medicine food homology herbs, and provided a novel paradigm for wound treatment strategies with both theoretical innovation and translational potential.

Author contributions

Mijia Zhang: Data curation, Formal analysis, Methodology and Writing-original draft. Jianhua Peng: Conceptualization, Investigation, Funding acquisition and Writing-review & editing. Xin Peng, Qiwen Qin: Data curation, Formal analysis, Methodology. Tao Ye, Li Deng: Formal analysis, Methodology. Yongzhou Wang: Conceptualization, Investigation. Yong Jiang: Conceptualization, Investigation and Writing-review & editing. Pan Liang: Conceptualization, Funding acquisition, Investigation, Project administration and Writing-original draft.

Funding

The authors acknowledge support from the National Natural Science Foundation of China (82405190, U24A20689), Southwest Medical University (2023ZYYQ05).

Data availability

Data will be made available on request.

Declarations

Ethics approval and consent to participate

All animal experiments were conducted in strict accordance with Chinese national standards, and all followed the experimental protocols approved by the Animal Ethics Committee of Southwest Medical University.

Consent for publication

All authors have approved the manuscript and agree for the submission.

Competing interests

The authors declare no competing interests.