A sequential and synergistic release sacrificial microgels scaffold for endometrial niche reprogramming and fertility restoration

Abstract

Intrauterine adhesions (IUA) remain a therapeutic challenge, necessitating advanced regenerative strategies. Herein, we propose a “programmed endometrial niche” strategy, a 3D-printed adipose-derived stem cells (ADSCs) hydrogel scaffold is constructed via microfluidic technology, incorporating platelet-rich plasma (PRP)-loaded thermosensitive gelatin microgels (PMS@AH) for sequential and synergistic delivery of PRP and ADSCs to restore the endometrial immune microenvironment and niche. In vitro assays demonstrate that PMS@AH possesses excellent biocompatibility and significantly promoted cell proliferation, migration, and angiogenesis, accompanied by potent anti-inflammatory activity. In both rat and cynomolgus macaque IUA models, it effectively restored endometrial structure, reduced fibrosis and inflammation, and successfully restored fertility. Mechanistically, RNA-sequencing analysis reveals that PMS@AH involves suppressing the IL-17/nuclear factor-κB (NF-κB) pathway to modulate the immune microenvironment and inhibiting the Wnt/β-catenin pathway to program the endometrial niche. This sequentially releasing scaffold presents a promising platform for treating IUA and broader applications in reproductive medicine.

Graphical abstract

Highlights

• •A thermosensitive sacrificial microgel scaffold for sequential and synergistic delivery was designed using 3D printing and electrospray technology, where early PRP release initiates repair and the ensuing porous network sustains ADSCs activity.
• •In both rat and cynomolgus monkey models of IUA, PMS@AH effectively reduced inflammation and fibrosis and restored fertility.
• •Mechanistic studies revealed that PMS@AH mitigated the IL-17/nuclear factor-κB (NF-κB) pathway to modulate the immune microenvironment and inhibiting the Wnt/β-catenin pathway to program the endometrial niche, thereby achieving a dual therapeutic mechanism for endometrial repair and regeneration.
• •This study not only offers a highly promising tissue engineering solution for intrauterine adhesions, but its sequential synergistic release strategy based on thermosensitive sacrificial microgels also holds broad application potential in other areas of female reproductive medicine.

1. Introduction

A healthy endometrium is essential for female reproductive function. Disruption of the endometrial environment can trigger and exacerbate various gynecological disorders, ultimately compromising fertility [1]. Intrauterine adhesion (IUA), also known as Asherman's syndrome, is characterized by symptoms such as irregular menstruation, pelvic pain, a thin endometrial layer, recurrent pregnancy loss, and infertility [2]. Niche formation arises from interactions between adjacent cells [3]. IUA develops due to iatrogenic disruption of the endometrial niche [4], typically occurring after intrauterine procedures or inflammation, and represents one of the most prevalent acquired endometrial disorders [5]. Its incidence ranges from 19% to 62.5% depending on the type of clinical procedure performed [6]. Currently, transcervical resection of adhesions (TCRA) remains the standard surgical intervention, yet recurrence occurs in up to 62.5% of severe cases [7]. Physical barriers such as intrauterine devices (IUDs) or Foley catheter balloons are also used to prevent re-adhesion; however, prolonged implantation often induces inflammation and necessitates removal, risking further injury [8]. In addition, intrauterine administration of hormones, pharmacological agents, or stem cells has shown partial efficacy, but clinical translation remains limited by poor absorption, low retention, and adverse side effects [9,10]. Therefore, developing effective strategies that effectively prevent re-adhesion, restore the endometrial niche, and recover uterine function remains an urgent clinical priority.

In recent years, regenerative medicine has introduced innovative strategies for managing IUA, emphasizing the reparative potential of platelet-rich plasma (PRP) and adipose-derived stem cells (ADSCs). PRP, a platelet-enriched plasma fraction obtained by centrifugation, is rich in growth factors that promote tissue regeneration [11,12]. As an emerging therapeutic modality, PRP has been applied to various clinical contexts [13], including endometrial repair. Numerous studies have demonstrated its immunomodulatory and pro-angiogenic properties, showing that PRP enhances endometrial regeneration, mitigates fibrosis, and improves fertility and pregnancy outcomes [14,15]. However, PRP is characterized by a rapid, burst-like release of bioactive factors in vivo, resulting in a short duration of activity and limited therapeutic efficacy [16]. ADSCs, endowed with multidirectional differentiation potential and robust paracrine activity, have also shown substantial promise in uterine repair and regeneration [17]. Nonetheless, early inflammatory and oxidative stress conditions within the injured endometrium markedly reduce ADSC survival and retention, diminishing their therapeutic benefit [18]. Therefore, overcoming the intrinsic limitations of PRP and ADSCs through rationally designed delivery systems is critical to enhancing their clinical efficacy and translational potential in endometrial regeneration.

In the field of tissue engineering, gelatin has been extensively employed owing to its excellent biocompatibility and thermosensitive gel–sol transition properties, which allow reversible phase changes in response to temperature variations [19]. This feature also makes gelatin an ideal carrier for controlled encapsulation and sustained release of bioactive molecules in drug delivery systems [20]. Furthermore, gelatin methacryloyl (GelMA), a photocrosslinkable derivative of gelatin, retains the bioactivity and cell-adhesive motifs of native gelatin while gaining superior mechanical tunability and structural stability [21]. Moreover, GelMA can serve as the primary structural matrix of the hydrogels, providing a biomimetic extracellular microenvironment that supports the encapsulation, survival, and function of stem cells. In Yang's research, porous GelMA hydrogels were utilized as delivery carriers for mesenchymal stem cells derived from gingival tissue [22]. However, traditional therapies in the field of tissue engineering often rely on single-component approaches, neglecting synergistic strategies and thus exhibiting limitations. Therefore, based on the properties of gelatin and GelMA, we propose the co-delivery of PRP and ADSCs to achieve sequential release and synergistic regeneration. This approach holds promise as a novel strategy for synergistically activating the restoration of the endometrial niche.

In this study, we developed a thermosensitive sacrificial microgels scaffold (PMS@AH) for the sequential and synergistic delivery of PRP and ADSCs (Fig. 1A). We first prepared PRP microgels (PMS) using microfluidic electrospray technology and cryopreserved them in liquid nitrogen. Subsequently, we mixed these PMS with pre-expanded ADSCs to formulate a bioink, which was then printed into three-dimensional PMS@AH scaffolds via microfluidic-assisted 3D bioprinting. The resulting composite hydrogel exhibited robust mechanical integrity and temperature-responsive degradability, with PMS dissolving at physiological temperature (37 °C) to enable the controlled early release of PRP. Moreover, the dissolution of PMS generated a microporous network within the hydrogel, thereby enhancing the viability, proliferation, and paracrine activity of ADSCs to achieve sequential release and synergistic therapeutic effects. In vitro assays demonstrated that PMS@AH mitigated early macrophage-mediated inflammation while enhancing the proliferation, migration, and tube formation of human umbilical vein endothelial cells (HUVECs). In a rat IUA model, PMS@AH markedly improved the endometrial architecture and stimulated cell proliferation and neovascularization through modulation of the inflammatory microenvironment and endometrial niche, ultimately promoting uterine repair and regeneration. Furthermore, in a non-human primate (cynomolgus macaque) IUA model, PMS@AH achieved comparable regenerative outcomes (Fig. 1B). Transcriptomic analysis revealed that PMS@AH attenuated inflammation and epithelial–mesenchymal transition (EMT) by suppressing the IL-17/NF-κB and Wnt/β-catenin signaling pathways, thereby improving the inflammatory microenvironment and endometrial niche of the damaged uterus (Fig. 1C). Collectively, this work introduces a novel thermosensitive sacrificial microgels scaffold capable of sequential and synergistic release of bioactive components, offering a promising therapeutic strategy for endometrial regeneration and broader applications in drug and stem cell delivery systems.

Fig. 1.

Schematic of 3D-printed PMS@AH scaffolds for ADSCs and PRP delivery with endometrial regeneration. (A) Preparation method of the PMS@AH composite scaffold. Using electrospray technology, a pre-mixed solution of PRP and gelatin was processed to generate PRP microgels (PMS). These PMS were then blended with ADSCs and a GelMA solution to formulate a bio-ink, which was subsequently utilized to fabricate a PMS@AH scaffold via microfluidic-assisted 3D bioprinting. Under physiological conditions (37 °C), the dissolution of PMS within the scaffold enables the early and controlled release of PRP, the PRP came into contact with calcium ions incorporated into the GelMA scaffold and was activated. Concurrently, this dissolution creates a microporous network inside the hydrogel, which enhances ADSCs viability. Together, these features achieve the sequential release of PRP and ADSCs, leading to a synergistic therapeutic effect (MS: Gelatin microgels; MSH: Gelatin microgels hydrogel; PMS: PRP gelatin microgels; PMSH: PRP gelatin microgels hydrogel; PMS@AH: Hydrogels loaded with PMS and ADSCs). (B) PMS@AH scaffolds were implanted into the damaged uteri of SD rats and cynomolgus macaque with IUAs to promote endometrial regeneration (n represents the total number of animal samples). (C) Molecular mechanisms of PMS@AH scaffold-mediated repair of damaged endometrium.

2. Results and discussion

2.1. Fabrication and Characterization of MSH scaffold

In this study, we designed and fabricated a novel hydrogel scaffold incorporating sacrificial microgels (MSH scaffold) through a synergistic combination of microfluidic electrospray and 3D printing technologies (Fig. 2A). GelMA, with its excellent photo-crosslinking properties and biocompatibility, was selected as the foundational material for the scaffold [23,24]. Through 3D printing technology, we first constructed a GelMA hydrogel scaffold with a stable and regular fiber network, providing a robust framework for subsequent microgel integration (Fig. 2B). Owing to its inherent thermosensitivity, gelatin, as a natural polymer material, was strategically chosen for constructing the thermo-responsive sacrificial microstructures [25]. Due to its sol-gel transition temperature being highly dependent on concentration, we first conducted rheological characterization on gelatin solutions of different concentrations (5%, 10%, 15%, 20% w/v). A temperature-sweep analysis revealed that the sol-gel transition point, defined by the crossover of the storage modulus (G′) and loss modulus (G″), shifted to higher temperatures with increasing gelatin concentration, occurring at 32 °C, 34.8 °C, 35.8 °C, and 37.3 °C, respectively (Fig. 2C). To ensure that the microgels could dissolve effectively at physiological temperature, while avoiding premature sacrifice due to body temperature fluctuations, we ultimately selected a 15% gelatin formulation because its phase transition temperature is 35.8 °C in this study, which is closest to body temperature without exceeding 37 °C. Subsequently, to ensure that the 15% gelatin electrospraying ink could be stably processed through the microfluidic electrospray system for microgels (MS) formation, we further evaluated its rheological properties (Fig. 2D). The flow step test confirmed the significant thermosensitivity of this electrospraying ink, it exhibited thixotropic gel behavior at 30 °C and transformed into a low-viscosity fluid at 37 °C. Furthermore, a flow sweep test demonstrated its excellent shear-thinning properties, a critical prerequisite for stable and continuous jetting through the microfluidic nozzle (Fig. 2E). To systematically investigate the influence of electrospray voltage on MS size distribution [26], we fabricated MS under a series of applied voltages (3.5, 4.0, 5.0, 5.5 kV). As shown in Fig. S1A and B (Supporting Information), the MS diameter exhibited an inverse correlation with the voltage, decreasing progressively until it stabilized at approximately 5 kV. When 5 kV was used as the voltage for the electrospray experiment, the MS exhibited smaller diameters and uniform morphology. As the voltage was further increased, there was no significant change in the diameter of the MS. Considering the potential risk of equipment damage at higher voltages, 5 kV was selected as the optimal and stable voltage for all subsequent MS preparation. Leveraging these optimized parameters, we successfully fabricated MS via microfluidic electrospray, which exhibited a uniform size with a mean diameter of 421.4 ± 24.8 μm (Fig. 2F).

Fig. 2.

Fabrication and Characterization of MSH Scaffold. (A) Schematic illustration for the fabrication of 3D-printed scaffold. (B) Digital photograph of during the bracket printing process (i) and pure scaffold (ii). Scale bars indicate 1 cm in (i) and 1 mm in (ii). (C) Temperature sensitivity curves of gelation electrospray inks in different concentrations (5%, 10%, 15% and 20%). (D) Viscosity stabilization duration of electrospraying ink. (E) Shear-thinning profile of the electrospraying ink. (F) Optical images of MS and particle size distribution curve of MS at a voltage of 5 kV. Scale bar, 500 μm. (G) EDS elemental mapping images of MS and PMS. Scale bar, 50 μm, Scale bars indicate 25 μm in (i, ii). (H) SEM images of MS (i, ii) and SEM images of MS surfaces (iii) and interiors (iv). Scale bars indicate 400 μm in (i), scale bars indicate 100 μm in (ii) and 25 μm in (iii, iv). (I) SEM images of pure scaffold (i, ii) and MSH scaffold (iii, iv). Scale bars indicate 500 μm in (i, iii) and 100 μm in (ii, iv). The red box indicates MS. (J) Digital photograph (i) and Confocal fluorescence microscope morphology (ii - iv) of the MSH scaffold. Scale bars indicate 1 mm in (i) and 200 μm in (ii - iv). MS: Gelatin microgels; PMS: PRP gelatin microgels; MSH: Gelatin microgels hydrogel.

To endow the sacrificial MS with early-stage regenerative capacity, we encapsulated PRP into the MS to prepare PMS. Since PRP activated in vitro undergoes a gel-like transformation and forms water-insoluble flocculent precipitates (Fig. S2C–D, Supporting Information), it cannot be mixed with a gelatin solution to form a homogeneous electro-ink (Fig. S2E, Supporting Information). Furthermore, PRP activated in vitro can no longer release growth factors and has virtually no reparative effects (Fig. S2F, Supporting Information). Therefore, in this study, we mixed fresh, unactivated PRP with a gelatin solution to form a homogeneous, uniformly dispersed electro-ink. PRP, as a concentrated source of endogenous growth factors, has demonstrated significant therapeutic potential in regenerative medicine fields such as endometrial repair [27]. We first successfully extracted PRP with a high platelet concentration of 2-3 × 10¹²/L (Fig. S2A–B, Supporting Information). The loading and distribution of PRP within the PMS were characterized by energy-dispersive X-ray spectroscopy (EDS). As shown in Fig. 2G, compared to the Control (MS), the PMS exhibited a distinct and homogeneous surface signal for sodium (Na), which served as a proxy for the presence of PRP. Scanning electron microscopy (SEM) further unveiled the porous internal structure of the MS, which facilitates sustained release of PRP (Fig. 2H). Subsequently, these MS were homogenously dispersed in the GelMA bio-ink to 3D print the MSH scaffold. Imaging via SEM validated the uniform distribution of the MS within the matrix and the preservation of the scaffold's structural integrity (Fig. 2I). To clearly observe the distribution of MS within the fibrous scaffold, we labeled MS and scaffold with FITC-NHS and Rhodamine dyes, respectively, as shown in Fig. 2J, the green-labeled MS were stably encapsulated within the red fibers. This elegant design leverages the sacrificial nature of the MS to create an interconnected porous network in situ, establishing an optimal microenvironment for key cellular activities such as infiltration and proliferation.

2.2. Biocompatibility and Physicochemical Properties of the MSH scaffold

Excellent biocompatibility is a prerequisite for biomaterials intended for tissue regeneration [28]. Given that the injured endometrial surface may present with exposed microvasculature, the hemocompatibility of the scaffold is particularly critical [29]. A hemolysis assay revealed that the MSH scaffold induced a hemolysis ratio of less than 5%, which is within the safe threshold, indicating its excellent blood compatibility and a low risk of causing additional hemolytic damage upon uterine implantation (Fig. 3A). Furthermore, the cytocompatibility of the MSH scaffold was evaluated through co-culture with HUVECs and ADSCs. Calcein AM staining assays performed over 3 days demonstrated that cells co-cultured with the MSH scaffold exhibited high viability and maintained a healthy morphology, with proliferation rates comparable to those of the Control group (Fig. 3B; Fig. S3, Supporting Information). These results collectively confirmed the outstanding cytocompatibility of the MSH scaffold. To ensure the structural stability of the MSH scaffold in the dynamic environment of the uterine cavity, we conducted in-depth characterization of its physicochemical and mechanical properties. Rheological analysis revealed that the storage modulus of the MSH scaffold based on 10% GelMA matrix was approximately 100-1000 Pa [30], which closely matched the modulus range of natural human endometrium (Fig. 3C). This biomimetic mechanical property is crucial for promoting tissue integration. Stepwise cyclic rheological tests also revealed the excellent self-healing ability of the scaffold, which helps it quickly recover and maintain structural integrity (Fig. 3D). In addition, the introduction of MS enhanced the overall mechanical properties of the scaffold without affecting its shear-thinning characteristics as a bioink (Fig. 3E). The thermosensitive sacrificial property of the MS within the MSH scaffold was functionally verified through compression experiments. As shown in Fig. 3F and Fig. S4 (Supporting Information), the compressive strength of the MSH scaffold at 37 °C physiological temperature was significantly lower than that at 25 °C. This mechanical property change was attributed to the phase transition of MS from solid to sol state, confirming its on-demand sacrificial design function. To further evaluate the retention of the scaffold in the uterus, we conducted in vitro swelling and degradation experiments for further simulation. Compared with pure GelMA scaffolds, MSH scaffolds exhibited higher swelling rates and faster degradation rates (Fig. 3G and H). This may be attributed to the formation of a porous structure within the MSH scaffold after the sacrifice of MS at physiological temperature, increasing the specific surface area and thereby accelerating its swelling and degradation [31]. On this basis, to verify the early and effective release of PRP in the MSH scaffold (PMSH scaffold), we evaluated the release behavior of a similar biomacromolecular substance, BSA. From the in vitro simulation release results, it can be seen that the simulated drug showed a slow release in the first 24 h of the experiment. As the MS within the MSH scaffold were sacrificed, the release rate gradually increased and reached a plateau at the 48th hour of the experiment (Fig. 3I; Fig. S5, Supporting Information). This result indicated that due to the presence of sacrificial MS, PRP could be released in an early and timed manner through the sacrifice of MS in the early stage of uterine injury repair. Additionally, it is well known that active PRP can release a large number of growth factors, thereby promoting the repair of damaged tissues [32]. To evaluate the active loading of PRP within the PMSH scaffold, we further assessed the release of its growth factors. As shown in Fig. 3J and K, over a period of more than 48 h, PMSH scaffold maintained good activity and was capable of supporting the sustained release of vascular endothelial growth factor (VEGF) and epidermal growth factor (EGF). Additionally, due to the presence of the hydrogel scaffold, the release of growth factors achieved a certain degree of sustained release effect compared to the rapid release of pure PRP, which could more effectively promote early repair. In summary, these results fully demonstrate that the MSH scaffold we prepared has excellent biocompatibility and biomimetic physicochemical properties, and can achieve active loading and early release of PRP, laying a good foundation for subsequent experiments.

Fig. 3.

Biocompatibility and Physicochemical Properties of the MSH Scaffold. (A) Hemolysis test results. One-way ANOVA analysis was utilized for the statistical Analysis, n = 3, ∗∗∗p < 0.001 compared with CON group. (B) Fluorescence images of HUVECs co-cultured with the MSH scaffold for 1, 2, and 3 days by Calcein AM staining. Scale bar, 100 μm. (C) Amplitude sweep curves of GelMA and MSH materials. (D) Alternate step strain curves of GelMA and MSH materials. (E) Shear-thinning profile of the GelMA and MSH precursor solution. (F) compression curves of MSH scaffold under 25 °Cand under 37 °C. (G, H) Swelling and loss curves of GelMA and MSH materials. (I) Protein release curve of the MSH materials. n = 3. (J, K) Release kinetic curves of VEGF and EGF in the MSH materials. n = 3. GelMA: GelMA hydrogel; MSH: Gelatin microgels hydrogel; PRP: Pure PRP; PMSH: PRP gelatin microgels hydrogel.

2.3. In vitro anti-inflammatory, pro-angiogenic, cell migration and proliferation evaluation

Adipose tissue harvested from the abdominal region of adult rats was utilized to isolate ADSCs, which demonstrated the characteristic fibroblast-like morphology (Fig. S6A, Supporting Information). Flow cytometry analysis revealed that these ADSCs exhibited high expression levels of CD29 and CD90, while CD106 and CD45 were not expressed (Fig. S6B, Supporting Information). Furthermore, the differentiation potential of ADSCs into adipogenic, osteogenic, and chondrogenic lineages was evaluated using Oil Red O, Alizarin Red, and Alcian Blue staining assays, respectively (Fig. S6C–E, Supporting Information), corroborating previous findings on typical ADSCs. ADSCs were subsequently encapsulated into PMSH scaffold (PMS@AH scaffold). Consequently, the behavior of ADSCs cultured within the scaffold was directly visualized using a 3D model (Fig. S7, Supporting Information). Calcein AM staining revealed that ADSCs maintained high viability within the hydrogel scaffold after 24 h of culture.

Following an injury, the inflammatory response is essential for maintaining tissue homeostasis [33,34]. However, excessive and prolonged inflammation can impede or delay wound healing. Macrophages play a pivotal role in modulating inflammation levels in the wound area and have the potential to facilitate tissue regeneration. To explore this, we utilized lipopolysaccharide (LPS) to activate mouse peritoneal primary macrophages (MPMs) and establish an in vitro model of inflammation. The in vitro anti-inflammatory effects of PMS@AH materials were evaluated using immunofluorescence (IF) staining and enzyme-linked immunosorbent assay (ELISA). According to the IF staining results (Fig. 4A and B), the PMS@AH scaffold group demonstrated a significant reduction in the M1 macrophage marker (CD86) and a marked increase in the expression of the M2 macrophage marker (CD206) when compared to both the LPS group and the MSH scaffold group which lacks ADSCs and PRP. Furthermore, ELISA assays indicated that the PMS@AH scaffold effectively decreased the levels of secreted inflammatory cytokines IL-6 and TNF-α (Fig. 4C). These findings suggest that the PMS@AH scaffold contributes to the attenuation of inflammation in vitro.

Fig. 4.

In vitro anti-inflammatory, pro-angiogenic, cell migration and proliferation evaluation of PMS@AH scaffold. (A) Representative immunofluorescence staining images for CD86 and CD206 of MPMs from different groups and (B) statistical data. Scale bars indicate 50 μm in A. (C) The concentrations of IL-6 and TNF-α in supernatants from MPMs measured by ELISA assay. (D) Representative images and (G) statistical data of migratory HUVECs cultured with different scaffolds. (E) Representative images and statistical data (H) of HUVECs tube formation in different groups. (F) Immunofluorescence images and quantification (I) of EdU staining in HUVECs. (J) CCK8 assays of the HUVECs after different treatments for 1, 2, and 3 days. Scale bars indicate 100 μm in D-F. One-way ANOVA analysis was utilized for the statistical Analysis, n = 3, ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 compared with Model or Control group. Control group: Blank control group; Model group: LPS group; MSH group: Gelatin microgels hydrogel without ADSCs and PRP; ADSCs group: Pure ADSCs; PRP group: Pure PRP; PMS@AH: Hydrogels loaded with PRP microgels (PMS) and ADSCs.

In the context of wound repair, the migration of endothelial cells is crucial for the angiogenesis process [35]. The ability of PMS@AH scaffold to enhance cell migration was assessed using HUVECs, and results from crystal violet staining demonstrated that PMS@AH significantly enhanced endothelial cell migration compared to other groups (Fig. 4D and G). Also the formation of new blood vessels is essential in tissue engineering as it provides nutrients that facilitate tissue recovery [36]. The angiogenic potential of PMS@AH scaffold was evaluated using primary HUVECs, with observations indicating that most tubules began forming within 2-4 h and reached their maximum number between 3 and 12 h [37], an 8-h period was selected as the optimal time for observation. The data demonstrated that the PMS@AH group enhanced tube formation with a greater number of branches compared to the other groups, indicating that PMS@AH may potentiate angiogenesis in vitro (Fig. 4Eand H). To evaluate the cell proliferation effect of PMS@AH, the 5-ethynyl-2-deoxyuridine (EdU) staining and CCK-8 assay were used to analyze cell survival following coculturing hydrogels with HUVECs, the PMS@AH group had more proliferative cells (red) in EdU-stained HUVECs compared to the other groups (Fig. 4Fand I). At designated time intervals, the CCK8 assay results revealed that PMS@AH significantly promoted the proliferation of HUVECs on days 1, 2 and 3 of incubation relative to the control and MSH groups (Fig. 4J). Previous studies have confirmed that PRP is rich in growth factors associated with angiogenesis, including PDGF, IGF-1, bFGF, and VEGF [38]. Furthermore, ADSCs facilitate the migration, proliferation, and differentiation of HUVECs by secreting various biologically active substances, such as exosomes, growth factors, and anti-apoptotic factors [39]. Our findings suggested that although both the ADSCs and PRP alone groups supported endothelial cell proliferation, migration, and angiogenesis, their effects were not as pronounced as those observed in the PMS@AH group. These results suggest that PMS@AH scaffold enables sequential delivery and synergistic therapy of PRP and ADSCs, collectively providing beneficial stimulation to HUVECs. In summary, the in vitro findings demonstrate that PMS@AH can reduce macrophage inflammation and promote cell migration, proliferation, and angiogenesis.

2.4. Examination of endometrial restoration in rats post various treatments

To evaluate the endometrial regeneration potential of PMS@AH in vivo (Fig. 5A; Figure S8, Supporting Information), we developed a standard rat model of IUA using a mechanical scraping technique as outlined in previous studies [40]. Following various treatments, the rats were either euthanized on the 14th day to collect uterine tissue or housed for mating with males at a 1:1 ratio. Selecting day 14 not only allows for the assessment of early tissue reconstruction but also enables the observation of mid-stage repair stability, including re-epithelialization, matrix remodeling, angiogenesis, and the expression of functional biomarkers [41]. Furthermore, based on the results of material degradation and sustained release, this timeframe allows sufficient time to observe the subsequent regenerative effects mediated by ADSCs, as well as the initial microenvironmental regulation mediated by PRP. To assess the safety of the material treatment, histological examination using hematoxylin and eosin (H&E) staining was conducted on the major organs of the rats (heart, liver, spleen, lung, and kidney), revealing no significant pathological abnormalities in any group, thereby confirming the safety of all implanted scaffolds (Figure S9, Supporting Information).

Fig. 5.

Enhanced endometrial repair process in vivo by PMS@AH. (A) Procedure for inducing injury and administering the treatment. (B) Representative H&E (Scale bar, 300 μm) and (C) Masson's trichrome staining images of the uterus after different treatments. Scale bars indicate 300 μm in (i) and 25 μm in (ii). (D, E) Immunohistochemical staining for Ki67 and IL-6. Scale bar, 200 μm. (F-J) Quantitative analysis of endometrial thickness (F), endometrial glands (G), collagen deposition (H), Ki67 (I) and IL-6 (J) in different groups. One-way ANOVA analysis was utilized for the statistical Analysis, n = 5, ∗P < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 compared with IUA group. Sham group: Sham surgery group; IUA group: Model group; MSH group: Gelatin microgels hydrogel without ADSCs and PRP; ADSCs group: Pure ADSCs; PRP group: Pure PRP; PMS@AH: Hydrogels loaded with PRP microgels (PMS) and ADSCs.

After 14 days of treatment for each group (Figure S10, Supporting Information), no significant cross-interference was observed between the bilateral horns. Subsequently, H&E staining was employed to evaluate the endometrial architecture, with measurements taken for endometrial thickness and gland count. The IUA group displayed a significantly reduced endometrial thickness (approximately 450 μm) compared to the sham group (approximately 1100 μm). In contrast, the PMS@AH treatment group achieved an endometrial thickness of 900 μm, which, although less than that of the sham group, was substantially greater than that of the IUA group (Fig. 5B and F). Furthermore, while the number of glands was stable and structurally intact in the sham group, it was significantly diminished in the IUA group. Compared to the IUA group and other treatment groups, the PMS@AH treatment group exhibited a higher number of glands and superior structural integrity (Fig. 5G). To conduct a more detailed analysis of endometrial fibrosis in rats, we utilized Masson trichrome staining to assess collagen deposition within the endometrium across different experimental groups. The blue-stained regions served as indicators of uterine fibrosis levels. Compared to the sham group, all treatment groups demonstrated an overall increase in collagen deposition, consistent with the initial phase of extracellular matrix remodeling [42]. Notably, collagen deposition was significantly reduced in the PMS@AH group compared to the IUA group and other treatment groups, underscoring the efficacy of PMS@AH in mitigating fibrosis during endometrial repair (Fig. 5C and H).

To evaluate the impact of PMS@AH treatment on endometrial regeneration, we assessed the expression of Ki67, a nuclear antigen present during all phases of cell proliferation. As depicted in Fig. 5D and I, the IUA group exhibited a limited number of proliferating cells, whereas the PMS@AH group showed a significant increase in Ki67 expression within the endothelial fraction. These findings suggest that PMS@AH material promotes endometrial regeneration in IUA rats. The immunohistochemical (IHC) analysis of IL-6 staining indicated that treatment with PMS@AH material effectively reduced inflammation in endometrial wounds, corroborating the in vitro findings (Fig. 5E and J). Endometrial angiogenesis was evaluated using double immunofluorescence staining with the vascular markers alpha-smooth muscle actin (α-SMA) and CD31 [43]. After 14 days of treatment, the PMS@AH group showed significantly increased expression levels of these vascular markers compared to the other groups. This was evidenced by a greater number and density of blood vessels, suggesting that the PMS@AH material facilitated neovascularization in the injured endometrium (Fig. 6A and C).

Fig. 6.

Examination of endometrial restoration and Fertility in rats post various treatments. (A) Typical immunofluorescence staining images of the uterine tissues from different groups: (i) CD31 and α-SMA double staining, (ii) LIF and (iii) ESRα staining (CD31: red, α-SMA, LIF and ESR-α: green, DAPI: blue). Scale bar, 200 μm. (B-D) Statistical data of the staining images. One-way ANOVA analysis was utilized for the statistical Analysis, n = 5, ∗P < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 compared with IUA group. (E) Pregnancies in different groups. (F) Comparison of the pregnancy rate, (G) the number of embryos and (H) embryonic developmental state in different groups. One-way ANOVA analysis was utilized for the statistical Analysis, n = 3, ∗∗∗p < 0.001 compared with IUA group. Sham group: Sham surgery group; IUA group: Model group; MSH group: Gelatin microgels hydrogel without ADSCs and PRP; ADSCs group: Pure ADSCs; PRP group: Pure PRP; PMS@AH: Hydrogels loaded with PRP microgels (PMS) and ADSCs.

Additionally, IF staining for estrogen receptor α (ESRα) and leukemia inhibitory factor (LIF), both critical for regulating endometrial receptivity, was performed on postoperative day 14 to evaluate the restoration of endometrial function and tolerance following surgery [44,45]. Among all treatment groups, the PMS@AH group exhibited the highest levels of endometrial receptivity (Fig. 6B–D and E). These results imply that PMS@AH promotes the long-term functional regeneration of the endometrium and supports fertility recovery.

2.5. Evaluation of intrauterine restoration of fertility

Since the main goal of endometrial repair and regeneration is to achieve fertility restoration, we further explored uterine recovery in IUA rats following PMS@AH material treatment by evaluating the rates of pregnancy and the development of embryos. The rats were randomly assigned to three groups: Group 1 (sham), consisting of normal rats not subjected to uterine injury surgery; Group 2 (IUA), comprising rats that underwent uterine injury surgery without subsequent treatment; and Group 3 (PMS@AH), consisting of rats that received PMS@AH material post-uterine injury surgery. Successful fertilization was confirmed through vaginal smears and examination of vaginal plugs (Fig. S11, Supporting Information), after which the rats were euthanized on day 14 to evaluate reproductive outcomes post-fertilization. Notably, the number of uterine horns was significantly reduced in the IUA group compared to the sham group. However, treatment with PMS@AH material resulted in enhanced uterine fertility (Fig. 6F). Additionally, embryos in the PMS@AH group exhibited superior development, with significantly greater numbers and sizes compared to those in the IUA group (Fig. 6G). Furthermore, the uterine pregnancy rate was 60% in the IUA group, increasing to 80% in the PMS@AH group (Fig. 6H). Simultaneously, embryos from the PMS@AH group exhibited normal development, comparable to those in the sham group (Fig. 6I). These observations indicate that treatment with PMS@AH material in rats with endometrial injury resulted in fertility restoration, highlighting its potential to enhance pregnancy outcomes.

2.6. Evaluation of endometrial regeneration and remodeling effects in cynomolgus macaques

In this research, invasive surgery was used to successfully establish an IUA model in a nonhuman primate, the cynomolgus macaque (Fig. 7A), which share a genetic background, endocrine system, menstrual cycle, and anatomical structure with humans [46]. This large animal experimental model facilitates further investigation into the tissue-repair-promoting effects of PMS@AH materials, particularly focusing on thin endometrial adhesions resulting from endometrial injury in non-human primate. Prior to inducing mechanical injury, macaques with regular menstrual cycles were monitored and recorded over a continuous two-month period. The IUA model was established in cynomolgus macaques through open abdominal surgery (Figure S12, Supporting Information), uterine tissue was collected on the 14th day post-surgery, a stage corresponding to the late proliferative phase of the macaque menstrual cycle, during which epithelial regeneration, stromal cell proliferation, angiogenesis, and stromal remodeling were all underway. Therefore, this time point is suitable for evaluating the early-to-mid-term reparative efficacy of the scaffold, while avoiding the secretory phase, which is significantly influenced by hormonal fluctuations. In addition, we considered that using the same time point (14 days) as in the rat experiments would facilitate cross-species comparative analysis and avoid introducing time-point differences as an additional variable. Hysteroscopic imaging confirmed the successful establishment of the IUA model in macaques, revealing a pale, irregular endometrial surface with areas of adhesions, the application of PMS@AH treatment effectively reversed these conditions and facilitated the repair of damaged uterine tissue (Fig. 7B and C; Videos, Supporting Information). Furthermore, the safety of the PMS@AH material was assessed by histological examination of the primary organs (liver, lung, and kidney) of the cynomolgus macaques, with H&E staining indicating no significant pathological abnormalities (Figure S13, Supporting Information). H&E staining was used to assess changes in the structure of endometrial tissue, revealing that the endometrium of the IUA group was disordered and had incomplete glands, whereas the PMS@AH group demonstrated a more intact glandular architecture and a smoother endometrial surface compared to the IUA group (Fig. 7D). Similarly, Masson staining was utilized to assess the degree of tissue fibrosis, corroborating the therapeutic efficacy of PMS@AH (Fig. 7E). To investigate the role of PMS@AH in modulating local inflammation and cell proliferation during endometrial regeneration in cynomolgus macaques, IHC staining for CD86, CD206, and Ki67 was conducted. The outcomes were consistent with those observed in rat models, suggesting that PMS@AH may exert a beneficial effect on these processes (Fig. 7F–H). Additionally, the IF staining of CD31/α-SMA demonstrated that there was more neovascularization in the PMS@AH group (Fig. 7I). Furthermore, endometrial receptivity following PMS@AH treatment was evaluated through IF staining of ERα and LIF in the uteri of cynomolgus macaques. The results supported the initial findings, indicating that the PMS@AH group exhibited enhanced endometrial receptivity compared to the IUA group (Fig. 7J and K). Ultimately, the positive effects of PMS@AH material after uterine adhesion injury were further substantiated in non-human primate experiments.

Fig. 7.

Evaluation of endometrial regeneration and remodeling effects in non-human primate experiments from the IUA group and PMS@AH treatment group. (A) Flow chart showing the establishment of IUA model and treatment of PMS@AH scaffold in cynomolgus macaques. (B) Hysteroscopic photographs and (C)digital photographs of cynomolgus macaque uterine cavities. (D) Representative H&E and (E) Masson's trichrome staining images of the Endometrial. (F-H) Representative images of immunohistochemical staining for endometrial CD86, CD206 and Ki67. Typical immunofluorescence staining images of the endometrium: (I) CD31 and α-SMA double staining (CD31/α-SMA is used to label blood vessels), (J) LIF and (K) ESRα staining. Scale bar, 200 μm. (L) Statistical data of the staining images. Student's t-test was utilized for the statistical Analysis, “n = 3” indicates that three independent random multi-point sampling analyses were conducted per specimen, and the statistical results were derived from these repeated measurements. ∗P < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 compared with IUA group. IUA group: Model group; PMS@AH: Hydrogels loaded with PRP microgels (PMS) and ADSCs.

2.7. Regulatory mechanism of PMS@AH scaffold via RNA sequencing

To elucidate the mechanisms behind the observed endometrial repair, an IUA model was developed in rats through mechanical injury (IUA group) and treated with PMS@AH (PMS@AH group), tissue samples from the wound location were collected for RNA sequencing to investigate differentially expressed genes (DEGs) between the IUA and PMS@AH groups. Fig. 8A and B illustrated that Principal Component Analysis (PCA) and volcano plots revealed notable differences in gene expression between the IUA and PMS@AH groups. According to the heatmap of DEGs (Fig. 8D), the upregulated DEGs were mostly related to inhibiting inflammatory responses, reducing oxidative stress in cells, promoting cell growth, and modulating the immune system compared to the IUA group, in opposition, the downregulated genes were chiefly linked to the enhancement of inflammatory responses and the initiation of oxidative stress. By applying biological processes, cellular components, and molecular functions as criteria, we carried out gene ontology (GO) enrichment analyses of these DEGs, revealing that the PMS@AH group negatively influenced the inflammatory response, furthermore, it affected the proliferation of stem cells, which suggested a wide range of modulatory impacts on immune function and cell proliferation (Fig. 8F). Additionally, the KEGG enrichment analysis of DEGs demonstrated connections to the immune system, including pathways like IL-17, cytokine-cytokine receptor interaction, and MAPK signaling (Fig. 8G). Concurrently, the results from gene heatmaps and Co-expression Network maps in these pathways suggested that PMS@AH may have a pivotal role in immunomodulation by recruiting immune cells with chemokines and thus blocking the IL-17 signaling pathway (Fig. 8C–E). Building on the analysis above, we investigated how PMS@AH treatment affects immune responses by using IF staining for M1 (CD86) and M2 (CD206) macrophage markers, along with RT-qPCR, the results indicated that PMS@AH might enhance the polarization of macrophages from M1 to M2, decrease pro-inflammatory cytokine secretion, and improve the inflammatory microenvironment via the IL-17 signaling pathway (Fig. 8H–J). It has been reported in some studies that IL-17 is upstream of NF-κB signaling, and its activation leads to the intensification of the inflammatory cascade [47,48], so we detected the proteins associated with NF-κB signaling pathway by western blot analysis, the findings were consistent with our predictions, showing that PMS@AH was able to markedly inhibit the phosphorylation of p65 and IKB-α (Fig. 8K and L).

Fig. 8.

RNA sequencing of uterine tissues of rats in IUA and PMS@AH treatment groups to analyze their repair mechanisms. (A) PCA analysis of the IUA group and PMS@AH treatment group. (B) Volcano plot of differentially expressed genes (DEGs) (|log2FoldChange| > 1, padj <0.05). (C) Cluster heatmap of DEGs of interest which involve in IL-17 signaling pathway, cytokine-cytokine receptor interaction, and cell adhesion molecules. (D) Heat map of up-regulated and down-regulated DEGs between IUA group and PMS@AH treatment group. (E) Co-expression Network maps of DEGs in immune-related pathways. (F) GO and (G) KEGG enrichment analysis of DEGs. (H) Typical immunofluorescence staining images of CD86/F480 double staining and CD206/F480 double staining. Scale bar, 400 μm. (I) Corresponding semi-quantitative analysis of F480, CD86 and CD206 fluorescence intensity. (J) Relative mRNA expression of pro-inflammatory (IL-17a, IL-17f, IL-6, TNF-α and IL-1β) and anti-inflammatory factors (IL-10, IL-4 and Arg-1). (K, L) Western blot of P-ikbα, ikbα, P-P65, P65, β-catenin and C-myc. (M) Relative mRNA expression levels of mesenchymal markers (CDH2/N-cadherin, ZEB1 and FN1/Fibronection) and epithelial markers (CDH1/E-cadherin, TJP1/ZO-1 and OCLN/Occludin). Student's t-test was utilized for the statistical Analysis, n = 3, ∗P < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001 compared with IUA group. IUA group: Model group; PMS@AH: Hydrogels loaded with PRP microgels (PMS) and ADSCs.

The primary pathological alterations observed in IUA include inflammation and the accumulation of fibrinogen within the extracellular matrix (ECM), ultimately leading to endometrial fibrosis and disruption of the endometrial niche [[49], [50], [51]]. Epithelial-mesenchymal transition (EMT) is a process whereby epithelial cells acquire mesenchymal properties, playing a crucial role in the pathogenesis of both inflammation and fibrosis. As illustrated in Fig. 8M, RT-qPCR results demonstrated a significant upregulation of mesenchymal markers (CDH2/N-cadherin, ZEB1, FN1/Fibronectin, and VIM/Vimentin) and a concomitant downregulation of epithelial markers (CDH1/E-cadherin, TJP1/ZO-1, and OCLN/Occludin) in the IUA model. Conversely, treatment with PMS@AH effectively reversed EMT, and thus restore the endometrial niche. The Wnt/β-catenin signaling pathway serves as a pivotal mechanism for initiating EMT, with canonical Wnt signaling being integral to development, differentiation, and proliferation. Activation of Wnt/β-catenin signaling facilitates the translocation of β-catenin into the nucleus, where it binds to downstream transcription factors, thereby directly activating EMT-related target genes [[52], [53], [54]]. The western blot analysis indicated that PMS@AH suppresses the aberrant activation of the Wnt/β-catenin signaling pathway, subsequently inhibiting EMT to restore the endometrial niche (Fig. 8K and L). Furthermore, existing literature has documented that the Wnt/β-catenin pathway can activate the NF-κB pathway, with both pathways synergistically contributing to inflammation and EMT [[55], [56], [57]]. Moreover, we obtained similar experimental results and conclusions through validation in non-primate cynomolgus macaques (Fig. S14, Supporting Information). Our findings propose that PMS@AH mitigates the inflammatory response and EMT by obstructing the IL-17/NF-κB and Wnt/β-catenin pathways respectively (Fig. 1C), thereby improving the inflammatory microenvironment and endometrial niche of the damaged uterus, achieving endometrial repair and regeneration.

3. Conclusion

In conclusion, leveraging the reversible thermosensitive cross-linking behavior of unmodified gelatin, we fabricated temperature-responsive PMS via electrohydrodynamic technology. Building upon this design, a thermosensitive sacrificial microgel system was established using microfluidic 3D printing, enabling the sequential and synergistic release of PRP and ADSCs. The resulting PMS@AH scaffold exhibited high printing precision, robust mechanical integrity, and excellent biocompatibility. These features not only supported efficient encapsulation and retention of ADSCs but also facilitated scaffold implantation within sites of endometrial injury. Besides, in vitro studies demonstrated that PMS@AH significantly promoted the migration, proliferation, and tube formation of HUVECs while attenuating LPS–induced inflammation. In both rat and cynomolgus macaque models of IUA, implantation of PMS@AH markedly enhanced endometrial niche restoration and functional endometrial regeneration by suppressing inflammation, promoting cellular proliferation, inhibiting fibrosis, and stimulating angiogenesis. RNA sequencing further revealed that PMS@AH modulated the immune microenvironment and reversed EMT by downregulating the IL-17/NF-κB and Wnt/β-catenin signaling pathways, respectively. Collectively, this 3D-printed GelMA–ADSCs–PRP sacrificial microgels scaffold demonstrates powerful immunomodulatory effects and reshaping of the endometrial niche properties, offering a robust platform for endometrial repair and a promising translational strategy for treating IUA and related reproductive disorders.

4. Materials and methods

Materials: Porcine Gelatin, Methacrylic acid anhydride and sodium carbonate were purchased from Sigma (USA). Lithium phenyl (2,4,6-trimethyl benzoyl) phosphinate (LAP) was obtained from Aladdin (Shanghai, China). Rat adipose MSC medium, osteogenic, adipogenic and osteogenic differentiation medium were obtained from OriCell (China). Calcein-AM, CCK-8, and the BCA protein assay kit were bought from Beyotime (Beijing, China). Matrigel matrix (Growth Factor Reduced) was purchased from Corning (USA). Elisa assay kit was obtained from LiankeBio (China). Rhodamine B, H&E and Masson's Trichrome staining kits were provided by Solarbio (Beijing, China). CD86, CD206 and F480 antibody were purchased from Santa Cruz (USA). The Ki67, ESRα and IL-6 antibody were purchased from Abcam (USA). The LIF, α-SMA and CD31antibody was obtained from Proteintech (Chicago, USA). The P-ikbα,ikbα, P-P65, P65, β-catenin and C-myc antibody were bought from CST (USA). The PrimeScript™RT Master Mix and TB Green® Permix Ex Tap™ Kit were purchased from TaKaRa (Janpan). Detailed information on the reagents used in this study is provided in Table S2 of the Supplementary information.

Preparation of PRP: For the preparation of PRP, blood samples were collected from female SD rats aged 9–10 weeks. We added 1 mL of PRP extraction solution (mlBio, China) to 9 mL of fresh whole blood and mixed the contents by inverting the centrifuge tube several times. The mixture was then centrifuged at 4 °C at 1000 rpm for 10 min to yield 2-3 mL of supernatant, finally the supernatant was enriched by secondary centrifugation at 1000 rpm for 10 min to obtain 0.5 mL PRP. Platelet content in PRP was measured using a fully automated platelet counter (XN-1000, SYSMEX, Japan). In this study, the platelet concentration (Approximately 2.5 × 10¹²/L, Fig. S2B) in PRP was three to five times that of baseline whole blood.

Electrospraying of PMS: To make electrospray ink, 15% w/v unmodified gelatin was dissolved completely in PBS at 37 °C, followed by the addition of PRP. The PMS prepared with nozzle-ring electric field. The nozzle was selected as 30 G (outer diameter 0.3 mm and inner diameter 0.15 mm), and the flow rate of the ink is adjusted to 100 μL/min. By adjusting the voltage to achieve PMS of the desired diameter, the PMS were ultimately collected using liquid nitrogen, moved to PBS, and stored in a refrigerator at 4 °C for a thermo-crosslinking duration of 400 s. And the PMS and MS were vacuum freeze-dried and subsequently treated with a sputter coater, finally the morphologies and EDS elemental mapping images of the samples were observed and captured with SEM (FE-SEM, SU8010, Hitachi, Japan).

GelMA synthesis: GelMA was synthesized according to a protocol that was previously developed [58]. Under constant stirring, 10 g of gelatin was dissolved in 100 mL of carbonate-bicarbonate buffer at a temperature of 50 °C. 2 mL of methacrylic anhydride was gradually added, and the pH was adjusted to 8 by hydrochloric acid or sodium hydroxide. After reacting for 3 h, GelMA was produced following the dialysis, filtration, and freeze-drying of the resulting solution.

Isolation and Culture of Rat ADSCs: The ADSCs were meticulously extracted from the inguinal fat pads of 3-week-old male Sprague-Dawley rats, dissected and chopped in a sterile environment. Subsequently, adipose tissue homogenates were digested with 0.1% type I collagenase at 37 °C for 1-1.5 h. The samples were then filtered and centrifuged at 1200 rpm for 10 min. The supernatant was discarded, and then the cells were resuspended in rat ADSCs complete medium and cultured at 37 °C in 5% CO₂ for 48 h. The medium was replaced every 2 days to remove non-adherent cells until the cells were 80% confluent.

Characterization of ADSCs: Characterization of ADSCs by Flow Cytometry: In the third passage (P3), ADSCs were digested using trypsin to create a single-cell suspension and subsequently incubated with antibodies, including anti-CD90-APC, anti-CD45-FITC, anti-CD106-PE and anti-CD29-FITC (BD Biosciences, USA).The cells underwent antibody incubation at 37 °C in the dark for 30 min and were then analyzed using flow cytometry (Cytoflex, Beckman Coulter, Germany).

Multilineage Potential Assay of ADSCs: Third passage rat ADSCs were cultured in differentiation conditions to assess their ability for multilineage differentiation. For osteogenic differentiation of ADSCs, ADSCs were incubated with osteogenic differentiation medium for 3 weeks, and osteogenesis was monitored by Alizarin Red S staining. To assess adipocyte differentiation, ADSCs were cultured in an adipogenic medium and stained with Oil Red O after 14 days. For chondrogenic differentiation, ADSCs were incubated with osteogenic differentiation medium for 3 weeks and stained with Alcian Blue.

Microfluidic Printing of Microsphere Hydrogel scaffold (MSH scaffold): To form a gel precursor, GelMA was dissolved in PBS at 37 °C and subsequently mixed with 0.5% (w/v) of the photoinitiator (LAP). After the solution reached room temperature, a 1:1 ratio of MS solution was incorporated. Using a microfluidic 3D printing approach, the MSH scaffold was fabricated and its structure was confirmed, with stability maintained by ongoing UV light exposure. Finally the morphologies images of the scaffolds were observed and captured with SEM and laser confocal scanning microscope (Leica Microsystems, Wetzlar, Germany).

Fabrication of PMS@AH scaffold: The medium that included ADSCs was combined with PMS to produce precursor solution A. Afterward, GelMA was dissolved in a medium containing calcium at 37 °C and mixed with LAP to form precursor solution B. To create the bioprinting ink, precursor solutions A and B were combined in a 1:1 vol ratio at room temperature. The PMS@AH scaffold was subsequently fabricated with microfluidic 3D printing.

The Biocompatibility of MSH Scaffold: To assess the biocompatibility of the scaffold, CCK8 assay was performed in which the ADSCs (3 × 10⁴ cells/well) were co-cultured with scaffolds in 24-well plates for 1, 3, and 5 days, after discarding the medium, a 10% CCK8 working solution was added to each well and incubated at 37 °C and 5% CO₂ for 2 h. Finally, 100 μL of the sample solution was transferred to a 96-well plate for absorbance measurement at 450 nm. To further assess the cytotoxicity of MSH scaffolds, they were co-cultured with ADSCs and HUVECs for 1, 3, or 5 days, and live cells were stained using Calcein-AM, and the staining of cell samples was observed using a laser confocal scanning microscope (Leica Microsystems, Wetzlar, Germany).

Hemolysis Experiments: Sodium citrate-anticoagulated whole blood was collected and centrifuged at 3000 rpm for 15 min to isolate red blood cells (RBCs), and the purified RBCs were washed three times, and then 20 μL of RBCs were separately aspirated, mixed with 1 mL of the prepared components, and incubated at 37 °C for 1 h before being centrifuged at 3000 rpm for 15 min. And absorbance of the sample supernatant (As) was measured at 542 nm, PBS was used as a negative control (Ap) and ultrapure water (ddH₂O) was used as a positive control (A_h). The hemolysis percentage was calculated using the following formula:

$$\text{Hemolysis}\text{rate}(\%)=({\mathrm{A}}_{\mathrm{s}}-{\mathrm{A}}_{\mathrm{p}})/({\mathrm{A}}_{\mathrm{h}}-{\mathrm{A}}_{\mathrm{p}})\ast 100\%$$

Rheological test: A rheometer (DHR-2, TA, USA) was used to assess the rheological properties of the bioinks. A parallel plate rotor with a diameter of 40 mm was employed for every test. For electrospray ink, to acquire the temperature sensitivity curve of the electrospray ink, the strain was fixed at 1 %, the angular frequency was maintained at 5 rad/s, and the modulus was recorded from 20 °C to 40 °C at a rate of 5 °C min⁻¹. During the flow step measurement of gelatin electrojet ink, the shear rate was set at 10 s⁻¹, with a sampling period of 6 s and a total test duration of 1200 s. The rheometer was adjusted to temperatures of 30 °C and 37 °C to assess viscosity. During shear thinning tests, the rheometer was kept at a temperature of 37 °C, with shear rates ranging from 1 s⁻¹ to 100 s⁻¹.

For GelMA precursor solution, during the amplitude sweep test of 10% GelMA hydrogel and MSH, the temperature was maintained at 25 °C, with an angular frequency of 1 Hz, and the strain varied from 0.1% to 1000% to assess the storage (G’) and loss modulus (G”). The GelMA hydrogel and MSH were scanned at 25 °C with a frequency of 1 Hz during alternate step strain scanning, the sample underwent multiple strain steps, including 1% strain for 60 s and 500% strain for 60 s, repeated three times. In flow sweep, the temperature was fixed at 25 °C, and the shear rate was adjusted from 0.1 s⁻¹ to 100 s⁻¹ to evaluate viscosity.

Compression tests and visco-elastic properties: The compressive deformation potential of the bracket was measured on a universal compression tester (5944, Instron, USA), with a strain rate set at 5 mm per minute.

Swelling Ratio and Degradation of MSH materials: For swelling test, the swelling characteristic of the GelMA hydrogel and MSH materials were assessed using the weighing method. For each group, three samples were prepared and their dry weights (W0) were measured by freeze-drying right after the gels formed. Then the freeze-dried samples were placed in a 37 °C water bath and weighed at specific intervals to measure their wet weights (W1). The swelling ratio is determined using the following equation:

$$\text{swelling}\text{ratio}=\left(\mathrm{W}1-\mathrm{W}0\right)/\mathrm{W}0\times 100\%$$

For vitro degradation of MSH materials, the initial dry weight of the MSH materials (W0) was recorded at the beginning of the test. Afterward, the samples were placed in PBS at 37 °C with shaking. They were then washed, dried, and weighed at the specified times (W1). The degradation ratio is determined using the following equation:

$$\text{degradation}\text{ratio}=\left(\mathrm{W}0-\mathrm{W}1\right)/\mathrm{W}0\times 100\%$$

Drug Release of MSH materials: BSA powder was enclosed within MS to simulate the drug release pattern of the scaffold, and the scaffolds were then incubated in PBS at 37 °C with shaking. 150 μL of supernatant was collected and replaced with fresh PBS at every predetermined time interval. In the end, BCA kits were used to measure cumulative release concentrations.

In vitro growth factor release kinetics: Enzyme-linked immunosorbent assay (ELISA) with VEGF and EGF kits (MULTISCIENCES, Hangzhou, China) was used to quantitatively measure the release profiles of growth factors in both the PRP and PMS hydrogel groups. All samples were tested according to the manufacturer's instructions. In this experiment, 1 mL of PMSH sample was placed in 1 mL of PBS solution in a 24-well plate, stirred at 80 rpm, and incubated at 37 °C. At specific time intervals, the supernatant was gathered and kept at −80 °C for later examination. The pure PRP group was prepared following the same method.

Extraction of mouse peritoneal primary macrophages (MPMs): Several 6-8 weeks male C57BL/6J mice were taken and injected intraperitoneally with 3 mL of 6% thioglycolate solution (0.3 g beef extract, 1.0 g tryptone, and 0.5 g NaCl dissolved in 100 mL water and filtrated with 0.22 μm filter). After two days, the mice were euthanized, and 5 mL of 1640 medium was used to flush the peritoneal cavity. The fluid was then collected and centrifuged to isolate the MPMs. Ultimately, MPMs were counted and plates were prepared for further experiments.

Immunofluorescence (IF) staining: The MPMs were fixed with 4% paraformaldehyde, permeabilized with 0.5% tritonX-100, then blocked with 5% BSA and incubated overnight at 4 °C with CD86 and CD206 antibody. In the end, the appropriate secondary antibodies were applied for staining and visualized with a laser confocal scanning microscope. (Nikon, C2si, Tokyo, Japan).

ELISA assay: The MPMs were inoculated into 24-well plates (2 × 10⁵ cells/well in the lower chamber). After 12 h, the MSH scaffolds, ADSCs, PRP or PMS@AH scaffolds were placed into the upper chamber of the transwell inserts (0.4 μm pore-sized filters) respectively. Then MPMs were incubated with LPS (0.5 μg/mL) for 24 h. According to the manufacturer's guidelines, cell supernatants were gathered and tested for IL-6, TNF-α, and IL-1β expression levels using ELISA.

Proliferation, Migration, and Angiopoiesis of HUVECs: For proliferation assay, the HUVECs were inoculated into 24-well plates (1 × 10⁵ cells/well in the lower chamber). The upper chamber of the transwell inserts (0.4 μm pore-sized filters) were managed as described before. And then the CCk8 kit was used to assess cell proliferation at 24, 48, and 72 h. For EdU detection, HUVECs were seeded at a density of 5 × 10⁴ cells per well in a 24-well plate. After cell adhesion, the hydrogels were co-cultured with cells for 24 h. Subsequently, the EdU solution was added, and incubation continued for 4 h. Finally, imaging was performed according to the kit instructions.

For migration assay, the HUVECs (1 × 10⁴ cells/well) were seeded to the upper chamber (8 μm pore-sized filters), ADSCs (2 × 10⁵), PRP (10 μL), 20 μL MSH hydrogel scaffold or 20 μL PMS@AH scaffold (approximately containing 10 μL of PRP and 2 × 10⁵ ADSCs) were arranged in the lower chamber. The HUVECs were stained with 0.5% crystal violet after 24 h of incubation, and images were captured using a microscope ((Nikon, Tokyo, Japan).

For angiopoiesis assay, the HUVECs were placed in 24-well plates that were coated with growth factor reduced Matrigel matrix, the management of the upper chamber of the transwell inserts (0.4 μm pore-sized filters) followed the previously described method. After 8h, the formation of the endothelial network was observed under the microscope.

Preparation of IUA model in vivo and treatment: The famale Sprague-Dawley rats (8-10 weeks) were sourced from Beijing Vital River Laboratory Animal Technology Co., Ltd. All protocols for animal experiments were approved by the Animal Policy and Welfare Committee of Wenzhou Medical University after review(WYYY-AEC-2025-104). The rats were randomly divided into the following four cages (Fig. S10, n = 6 per cage): sham group (left and right uterus); L (left uterus): R (right uterus) = IUA group: MSH scaffold group; L: R = PRP group: ADSCs group; PMS@AH scaffold group (left and right uterus), No obvious cross-interference was observed between bilateral horns. The IUA model was formed by employing the mechanical damage method. As shown in Fig. S8 (Supporting Information), a 3-mm incision was created on each side of the uterus, and the uterus was scraped with an endometrial curette until it showed signs of congestion and the uterine wall became rough. The IUA group received no treatment, while the MSH scaffold group, PRP group, ADSCs group, and PMS@AH scaffold group were each implanted with 50 μL of MSH hydrogel, 50 μL of diluted PRP (containing 25 μL of PRP), 50 μL of ADSCs (5 × 10⁵ ADSCs), and 50 μL of PMS@AH scaffold (approximately containing 25 μL of PRP and 5 × 10⁵ ADSCs), respectively [6]. Each rat was administered 100,000 IU of penicillin daily for three days following the surgery.

Two female cynomolgus macaques were purchased from Hainan Xinzhengyuan Biotechnology Co. Ltd. The two female cynomolgus macaques were selected for IUA model after mechanical injury (Fig. S12, Supporting Information), and one macaque was received PMS@AH scaffold (200 μL of hydrogel approximately containing 100 μL of PRP and 2 × 10⁶ ADSCs) treatment for its uterine wounds [46]. Two weeks later, endometrial recovery was checked with an endoscope, and the uterus was collected for further analysis. To enhance data reliability, random multi-point sampling was performed on the uterine specimens from each macaque in this study. The macaques were fed at the Hainan Provincial Drug Safety Evaluation Research Center [SYXK(Qiong)2017-0013]. All macaque experimental procedures were reviewed and approved by the Animal Care and Use Committee of Hainan Provincial Drug Safety Evaluation Research Center [approval nos. 20240711-NHP(03)] and conformed to the guide for the Association for Assessment and Accreditation of Laboratory Animal Care.

Histological analysis: Two weeks post-surgery, the animals were euthanized, and the surgical sites on each uterine horn were processed and stained using H&E and Masson trichrome. ImageJ software was used to measure endometrial thickness and collagen fiber area. The levels of inflammation and proliferation were determined through IHC for IL-6 and Ki67. For endometrial recovery, it was assessed by IF staining for ESR-α and LIF. Furthermore, serial sections were immunostained with CD31 and α-SMA to evaluate vascularization. In order to examine the improvement of the immune microenvironment in vivo, CD206, CD86, and F480 were identified using immunofluorescence staining. And all images were quantitatively analyzed with ImageJ Fiji software (version 1.53c, NIH, USA).

Assessment of endometrial function: After two weeks from the operation, male and female rats from each group (Sham group, IUA group, PMS@AH group. n = 5 per group) were placed together in a 1:1 ratio in a single cage (one male and one female per cage), and vaginal smears were taken daily at 9:00 a.m. to check for spermatozoa. Fertilization was confirmed by the presence of sperm in vaginal smears from female rats (Day 0 of gestation). The pregnant rats were then isolated and euthanized on Day 14 of gestation. A total of five female mice from each group were included in the pregnancy assessment, and the pregnancy rate, number of embryos, and embryo size were recorded. The pregnancy rate was calculated using the following formula: (number of pregnant females/total number of females mated) × 100%.

RNA sequencing (RNA-seq) and bioinformatic analysis: Following the creation of an IUA model with mechanically injured rats (IUA group) and the administration of PMS@AH treatment (PMS@AH group), total RNA was isolated from trauma tissue samples using TRIzol reagent (Sigma, USA). The RNA libraries were sequenced on illumine Novaseq Xplus (illumine,USA) platform by Personalbio Co., Ltd (Shanghai, China). To explore the potential biological mechanisms, a functional enrichment analysis was conducted, incorporating Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. In this research, a significant difference in DEGs, GO, and KEGG enrichment analysis was marked by a corrected p value (padj) of under 0.05.

Quantitative real-time PCR: The extraction of total RNA from uterine tissue was done using TRIzol reagent, followed by reverse transcription with the PrimeScript™ RT kit. Following this, the expression levels of the specified genes were measured using real-time PCR with TB Green Master. The internal controls were the PCR results of Gapdh, and the 2-ΔΔCT method was applied for calculations. The primers used are listed in Table S1 (Supporting Information).

Western blot: The extraction of proteins from uterine tissue was performed with lysate, followed by concentration measurement using a BCA kit. An equal quantity (30 μg) of protein was used for immunoblotting, following the previously described method. The detection of protein levels was carried out using electrochemiluminescence assay.

Statistical analysis: All data were obtained from at least three independent replicates and calculated as the means ± standard deviation (mean ± SD). Statistical analyses were performed using GraphPad Prism 9.0 software. One-way analysis of variance (ANOVA) was used for comparisons involving three or more groups. When ANOVA indicated statistically significant differences (p < 0.05), Tukey's HSD post hoc test was used for pairwise multiple comparisons. For comparisons involving only two groups, an independent two-tailed Student's t-test was used. A p-value of <0.05 was considered statistically significant (∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001).

Availability of data and materials

The data that were utilized to support the findings of this study can be obtained from the author who corresponds to this work.

Ethical statement

All protocols for animal experiments were approved by the Animal Policy and Welfare Committee of Wenzhou Medical University after review(approval nos. WYYY-AEC-2025-104). All macaque experimental procedures were reviewed and approved by the Animal Care and Use Committee of Hainan Provincial Drug Safety Evaluation Research Center [approval nos. 20240711-NHP(03)] and conformed to the guide for the Association for Assessment and Accreditation of Laboratory Animal Care. All the authors listed have approved the manuscript that is enclosed. We have prepared the manuscript in compliance with the Ethics in Publishing Policy as described in the Guide for Authors.

CRediT authorship contribution statement

Qian Liu: Investigation, Methodology, Writing – original draft. Shun Lu: Funding acquisition, Writing – original draft. Junhua Zhou: Methodology, Writing – original draft. Jie Xu: Methodology. Jiaqin Shao: Methodology. Jianqiong Zheng: Methodology. Yinhe Zhang: Investigation. Anqi Ye: Investigation. Shuhong Tian: Investigation. Chuanqi Qiu: Investigation. Mengqi Jin: Methodology. Maolin Qin: Methodology. Zichang Wu: Methodology. Somy Yoon: Investigation. Ling Xie: Conceptualization, Funding acquisition, Writing – review & editing. Lei Chen: Conceptualization, Funding acquisition, Writing – review & editing. Jian Xiao: Conceptualization, Funding acquisition, Supervision, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Data availability

Data will be made available on request.