Functional restoration of uterine architecture and fertility via a 3D-printed biomimetic scaffold: The role of macrophage-driven immunomodulation

Abstract

The structural complexity of the uterus makes identifying biomaterials and scaffold designs for supporting functional tissue regeneration challenging. Poly(L-lactide-co-caprolactone) (PLCL) has gained attention due to its biocompatibility and tunable degradation profile. In this study, a three-layered PLCL scaffold was fabricated via melt electrowriting (MEW) 3D printing to mimic the mechanical properties of the native uterine matrix. Using a mouse model of severe uterine injury, we demonstrated that the PLCL scaffold markedly promoted structural repair and functional recovery, as evidenced by improved pregnancy outcomes. Mechanistically, the scaffold enhanced macrophage recruitment and skewed their polarization toward an M2 phenotype, accompanied by upregulation of the hallmark genes SPP1, LGALS3, and TREM2. Furthermore, the scaffold stimulated parenchymal cell proliferation and migration and activated the JAK-STAT3 and YAP-Hippo signaling pathways, thereby establishing a pro-regenerative niche. Collectively, these findings highlight a scaffold-driven immunomodulatory mechanism that promotes uterine regeneration through macrophage polarization and activation of key signaling pathways.

Graphical abstract

A three-layered PLCL scaffold fabricated by melt electrowriting (MEW) facilitates uterine repair and regeneration. Implantation of the scaffold into a severely injured mouse uterus promotes functional and structural recovery. Mechanistically, regeneration is mediated by integrin binding, which induces actin cytoskeleton reorganization and activates the YAP pathway. Subsequent M2 macrophage polarization establishes a pro-regenerative microenvironment. This process culminates in myofibroblasts activation, upregulation of α-SMA, and tissue regeneration.

1. Introduction

The increasing incidence of uterine surgeries among women of reproductive age has raised concerns regarding subsequent uterine wall defects, which may impair healing and increase risks in future pregnancies [[1], [2], [3]]. Current therapeutic strategies, including hormonal therapies and surgical repair, often yield suboptimal outcomes, failing to fully restore uterine architecture and function [4,5]. In this context, tissue engineering and regenerative medicine provide innovative approaches for uterine reconstruction.

Melt electrowriting (MEW) is an emerging manufacturing technology that allows precise control over fiber diameter pore structure, and enables layer-by-layer construction of scaffold structures with high reproducibility. MEW achieves deterministic scaffold designs while maintaining micron-level resolution compared with those of traditional extrusion-based 3D printing or electrospinning, making it particularly suitable for constructing anisotropic and multilayered tissues. Scaffolds manufactured using MEW technology are being increasingly applied in various fields, such as skeletal muscle tissue and dental injury repair [6,7]. However, their use in repairing large-scale uterine damage remains to be explored.

The uterine wall is a highly organized, layered structure consisting of the endometrium and myometrium, which differ significantly in structural and mechanical properties. Therefore, accurate repair of full-thickness uterine defects requires a manufacturing strategy that can precisely control the scaffold geometry and mechanical properties. MEW-printed scaffolds can provide a stable three-dimensional structure and tunable mechanical properties, while maintaining good porosity that facilitates cell infiltration and tissue remodeling [[8], [9], [10]].

An ideal scaffold for uterine repair must fulfill several critical criteria: mechanical properties mimicking the native tissue [11,12], appropriate porosity to facilitate vascularization, a suitable degradation profile synchronized with tissue regeneration [13], biocompatibility to avert immune rejection [14], and customization for patient-specific applications [15,16]. Although decellularized scaffolds provide a natural extracellular matrix, they do not achieve efficient recellularization [[17], [18], [19]]. Despite their outstanding biocompatibility, collagen-based scaffolds often suffer from rapid degradation and poor shape fidelity [20]. Hyaluronic acid (HA) hydrogels support cell migration and proliferation but typically lack the mechanical stiffness and toughness required for uterine wall support [3,21,22]. Synthetic materials such as polycaprolactone (PCL) and polylactic acid (PLA) are used to fabricate scaffolds that offer structural strength; however, these scaffolds are limited by suboptimal cell-loading capacity. Consequently, scaffolds crafted from optimal materials that enable complex architectural designs are urgently needed.

Previous research has primarily focused on endometrial regeneration and adhesion prevention, with limited attention given to the repair of large, full-thickness uterine defects involving the endometrium and myometrium [17,18]. Although Park et al.'s study [23] made significant progress in this area, most research still does not adequately address the full-thickness uterine defect repair. To address this gap, we developed a biomimetic, three-layer scaffold using poly(L-lactide-co-caprolactone) (PLCL). The flexibility offered by PLCL is greater than that of more rigid polyester materials, such as PLA or PCL, making it more suitable for the dynamic mechanical environment of the uterus. Herein, through MEW 3D printing technology, we fabricated a PLCL scaffold designed to replicate the histological structure of the mouse uterus, comprising the endometrium and inner circular and outer longitudinal muscle layers. We systematically investigated its reparative efficacy and underlying mechanisms in a mouse model of uterine defects.

2. Materials and methods

2.1. Animal models

All animal procedures were approved by the Ethics Committee of Zhejiang Provincial People's Hospital (No. 20231213144812203080). Female C57/BL6 mice (6-8w) were obtained from Shanghai Slack Laboratory Animal Co., Ltd. (China) and maintained under standard laboratory conditions. Mice were randomly assigned to the excision-only (EX), PLCL scaffold implantation (PLCL), and sham-operated control groups. In the EX group, a segment encompassing the endometrium and myometrium (approximately 2/3 of the uterine horn circumference, 1 cm in length) was excised, preserving the mesometrial side. The PLCL group received the scaffold sutured at the injury site, while the control group only underwent laparotomy. One uterine horn was treated per mouse, with the contralateral horn serving as an internal control. Animals were euthanized at designated time points for sample collection.

2.2. Manufacturing and characterization of the PLCL scaffold

Scaffolds were designed using the EFL-PotatoE software and fabricated with an EFL-BP 6601 3D printer (Suzhou Intelligent Manufacturing Research Institute, China). The three-layer structure (total area: approximately 1200 mm²) comprises an ordered porous inner layer for cell loading, a triangular middle layer, and a square outer layer for mechanical support (Supplementary Fig. 2). Layers were sequentially printed, with printing times of approximately 31, 230, and 14 min, respectively.

The printing system was cleaned with hexafluoro isopropanol and 75% ethanol. PLCL copolymer (50:50 L-lactic and caprolactone, Jinan Daigang Bioengineering Co., Ltd., China), viscosity: 1.46 dL/g, Mw: approximately 180,000) was loaded into a preheated metal nozzle (170 °C) and cylinder (120 °C) and melt for 15 min to ensure its uniformity. Printing was performed at room temperature (20 °C) with the nozzle positioned 2 mm above a glass collector. Voltage and nozzle temperature were dynamically adjusted to ensure stable fiber deposition (Table 1). The nozzle temperature was gradually reduced from 170 to 130 °C across 1 h to maintain an optimal polymer flow.

Table 1.

Optimized melt electrowriting (MEW) parameters used for the fabrication of PLCL scaffolds.

Layer	Repeated Layers	Fiber diameter (μm)	Grid length of side (μm)	Nozzle Inner diameter(μm)	Nozzle Temperature (°C)	Collection Speed (mm/min)	Pressure (Kpa)	Voltage (KV)
Disordered (Inner)	2	3–5	1–5	150	170	500	20	4.5
Triangular (Middle)	20	15	160	150	170 → 130	2000	30	3
Square (Upper)	2	50	500	350	170	800	15	3

Morphology: Scanning electron microscopy (SEM) was performed to observe the morphology of the PLCL scaffold. Samples were air-dried for 24 h at room temperature, platinum-sputtered, and imaged under SEM (EVO 10, ZEISS, Germany) at an acceleration voltage of 15 kV [24].

Degradation in PBS buffer:Scaffolds were 3D printed in bulk using the same parameters and process to ensure consistency. Scaffold samples (20 × 15 mm) were placed in 60 mm Petri dishes containing 20 mL of PBS buffer and maintained in a 37 °C incubator. After the degradation was completed at designed time (0, 7, 15, 30, 60, 90, 120, and 150 days), the scaffolds were removed from the dish, cleaned with deionized water and 75% ethanol, air-dried, and weighed to calculate the mass retention rate.

Thermal degradation: For the granular PLCL raw material, preheating was performed at 170 °C for 15 min in a hot air drying oven, followed by continuous heating at 150 °C. Samples of PLCL granules were collected at 0, 15, 30, 45, 60, 90, and 120 min at 150 °C. The samples were frozen at −20 °C and stored for further molecular weight testing alongside the control samples (Supplementary Fig. 1b).

Molecular weight: The analysis was performed via size exclusion chromatography (SEC).Scaffold segments or particles from above degradation experiments were collected as samples, which were dissolved in tetrahydrofuran (THF) at 2 mg/mL, filtered through a 0.22 μm polytetrafluoroethylene (PTFE) membrane, and analyzed using an Agilent system (1260 Infinity II system, America) equipped with two PLgel 10 μm MIXED-B columns maintained at 40 °C, with a mobile phase flow rate of 1.0 mL/min.

Mechanical properties: The mechanical properties of scaffold samples were tested after degradation. Uniaxial tensile testing was conducted using a SUNS UTM 2203 testing frame equipped with a 20 N load cell. Specimens were clamped at both ends along the longitudinal direction, leaving a 10 mm gauge length exposed, and the speed of the crosshead was set to 5 mm/min consistently during the test.

Viscosity: PLCL circular films with a diameter of 20 mm and a thickness of 1 mm were prepared by preheating at 170 °C for 15 min. Rheological testing was conducted using a rotational rheometer (HAAKE MARS 60), with the temperature stabilized at 150 °C and a shear rate of 1 rad/s (Supplementary Fig. 1a).

SEM: The scaffold samples collected after degradation in PBS buffer were gently washed sequentially with purified water and 75% ethanol, followed by natural air-drying. The dried samples were then frozen at −20 °C for subsequent analysis. Prior to imaging, the scaffolds were thawed and dried again at room temperature. Subsequently, samples were sputter-coated with gold and imaged using a scanning electron microscope (HITACHI SU3500) operated at 5.00 kV.

Biocompatibility assessment of scaffolds: Primary uterine smooth muscle cells were isolated by enzymatic digestion of fresh mouse uterine myometrial tissue fragments (1–2 mm³) using 0.1% collagenase I (abx082404, YiSheng Biotechnology Co., Ltd, Shanghai, China) and 0.25% trypsin (E-EL-SR001-100 mL, YiSheng Biotechnology Co., Ltd, Shanghai, China) at 37 °C for 1–2 h. Cells were collected by centrifugation at 1000 rpm for 3 min, filtered through a 40 μm strainer, and cultured in DMEM (10566016, Gibco, US) supplemented with 10% fetal bovine serum. Primary uterine smooth muscle cells were resuspended and adjusted to a density of 1 × 10⁵ cells/mL. Subsequently, 100 μL of suspension (10,000 cells) was seeded into each well of a pre-coated 96-well plate, with five replicates per group. Following incubation at 37 °C and 5% CO₂ until complete cell adhesion, the culture medium in the experimental group was replaced with 100 μL of scaffold extract (prepared by incubating the scaffold in DMEM for 24 h), while the control group received fresh complete medium. Cell viability was evaluated at 0.5, 1, 2, and 3 days after treatment. At each time point, 10 μL of CCK-8 reagent was added to each well, followed by incubation for 2 h. Absorbance was then measured at 450 nm using a microplate reader, and cell viability was calculated based on the optical density values obtained.

2.3. Histological analysis

Tissue sections underwent standard processing, including antigen retrieval and blockade of endogenous peroxidase activity, followed by incubation with specific primary antibodies at 4 °C overnight. Primary antibodies employed for immunohistochemical staining were purchased from Abcam (Cambridge, UK), including anti-epithelial cadherin (E-CAD, ab231303), anti-cluster of differentiation 31 (CD31, ab182981), anti-α smooth muscle actin (α-SMA, ab 124964), anti-cluster of differentiation 86 (CD86, ab239075), and anti-cluster of differentiation (CD163, ab182422). Sections were then incubated with secondary antibody (ab205718) at room temperature for 25 min and counterstained with hematoxylin for 30 s. For negative controls, isotype-specific immunoglobulin (ab172730) was used instead of primary antibodies. The number of E-CAD-positive tubular glands, composed of a single layer of cuboidal epithelial cells, as well as CD31-positive blood vessels, were quantified using Image-J software (US National Institutes of Health, USA).

2.4. Immunofluorescence staining

After fixation, permeabilization, and blocking, tissue sections were incubated overnight at 4 °C with primary antibodies against CD86 and CD163 (ab220188, ab322551, Abcam, Cambridge, UK,1:200). After washing, sections were incubated with fluorophore-conjugated secondary antibodies (ab150077, ab150078, Abcam, Cambridge, UK,1:200), and nuclei were counterstained with DAPI. Images were acquired using a Leica Microscope (SP8 Confocal, Germany), and the number of CD86 and CD163-positive cells was quantified using ImageJ software.

2.5. Fertility test

To evaluate the functional competence of the reconstructed uterus in supporting embryo implantation and development, female mice from both the EX and PLCL groups (n = 10 per group) were mated with fertile males. Successful mating was confirmed by the presence of vaginal plugs. Cesarean sections were performed between post-coitum days 16–19 to assess embryonic development.

2.6. Macrophage depletion by clodronate liposomes injection

A macrophage depletion model was established via intraperitoneal administration of clodronate liposomes (5 mg/mL) (40337ES10, YiSheng Biotechnology Co., Ltd., Shanghai, China). An initial dose of 200 μl per mouse was administered one day prior to uterine defect induction, followed by daily maintenance injections of 50 μl. On day 15 post-surgery, macrophage depletion was verified in spleen tissues by flow cytometric analysis of F4/80 (NBP2-81029MFV610, Youningwei Biotechnology Co., Ltd., Shanghai, China) and CD11b (561114, Youningwei Biotechnology Co., Ltd., Shanghai, China) expression, and uterine repair was evaluated histologically.

2.7. Co-culture of PLCL loaded with RAW.264.7 macrophages and primary uterine smooth muscle cells

Primary uterine smooth muscle cells were seeded into 6-well plates at a density of 2 × 10⁵ cells per well and cultured for 3 days in a humidified incubator at 37 °C with 95% air and 5% CO₂. Concurrently, mouse monocyte/macrophage leukemia cells (Raw264.7, CL-0190, Wuhan Procell Life Technology Co., Ltd., Wuhan, China) were seeded onto PLCL scaffolds and maintained in DMEM medium for 3 days. When smooth muscle cells reached approximately 90% confluence, a uniform scratch (approximately 0.5 mm in width) was created in the cell monolayer using a sterile 1 mL pipette tip. After gently washing to remove detached cells, fresh medium was added. Subsequently, the PLCL scaffolds loaded with Raw264.7 cells were transferred into the 6-well plates for co-culture to evaluate the cell migration ability, and the PLCL scaffolds without cells were used as blank scaffold group. Images of the wound area were captured at 0, 24, and 36 h post-scratch [25]. The migration distance was quantified by calculating the average of ten randomly selected fields of view.

2.8. ELISA detection of cytokine levels

Uterine tissues were harvested at postoperative days 3, 7, and 14 for homogenate preparation. Total protein concentration was determined using a BCA assay (23225, Thermo Fisher Scientific, US). Levels of interleukin (IL)-6 (EM0004), tumor necrosis factor (TNF)-α (EM0010), IL-10 (EM0005), and TGF-β (EH0012) were quantified using commercial ELISA kits (Hua'an Biotechnology Co., Ltd, China), with results normalized to total protein content (pg/mg).

2.9. RNA sequencing

Total RNA was extracted from tissues collected 7 days post-modeling using Trizol reagent (15596026CN,Thermo Fisher Scientific, US). Poly-A-selected mRNA was reverse-transcribed into cDNA, and sequencing libraries were constructed and sequenced on the Illumina platform. Downstream analyses were conducted to explore changes in gene expression and potential molecular mechanisms. For single-cell RNA sequencing, freshly harvested tissues were dissociated under low-temperature conditions using a combination of mechanical disruption and enzymatic digestion with collagenase and DNase I (abx082404,10325ES80, YiSheng Biotechnology Co. Ltd., Shanghai, China). Following filtration and red blood cell lysis, single-cell suspensions, were obtained (cell viability >85% as confirmed by trypan blue exclusion). Single-cell libraries were constructed using the 10x Genomics Chromium Single Cell 3′ v3 platform and sequenced on the Illumina NovaSeq 6000 platform. Raw sequencing data were processed with Cell Ranger for alignment to the reference genome and generation of gene expression matrices. Downstream analyses included differential expression analysis and cell type annotation based on established marker genes.

2.10. RT-PCR validation

Total RNA was reverse-transcribed into cDNA using the PrimeScript RT reagent kit (RR037A, Takara Biotechnology, Japan). Quantitative real-time PCR was performed using SYBR Green reagent (639676, Takara Biotechnology, Japan) on an Applied Biosystems 7500 Real-Time PCR System (Applied Biosystems, US). Gene expression levels were normalized to β-actin.

2.11. Protein isolation and western blot analysis

Tissues lysates were resolved by SDS-PAGE and transferred to PVDF membranes (FFP39, Beyotime Biotechnology, Shanghai, China). Membranes were probed with primary antibodies, followed by incubation with horseradish peroxidase HRP-conjugate secondary antibodies (ab6721, Abcam, Cambridge, UK). Protein bands were visualized using chemiluminescence and expression levels were quantified densitometrically using ImageJ software, with GAPDH (ab181602, Abcam, Cambridge, UK) as the loading control.

2.12. Statistical analysis

Data were presented as mean ± standard deviation. Comparisons between two groups were performed using independent-sample t-tests, while multiple-group comparisons were conducted by one-way analysis of variance (ANOVA). Statistical analyses were conducted using Prism 8 (GraphPad, USA), with a p-value <0.05 considered statistically significant.

3. Results

3.1. Design, fabrication, and characterization of the PLCL scaffold

the PLCL Scaffold comprised a porous, disordered inner layer simulating the endometrium; a middle layer composed of triangular grids mimicking the inner circular muscle; and a square-grid outer layer representing the outer longitudinal muscle (Fig. 1, Supplementary Fig. 3). The total scaffold thickness was approximately 300 μm, closely matching the native murine uterine myometrial dimensions, based on our own measured data. Specifically, the outer structural framework, with a thickness of approximately 100 μm, was constructed from two superimposed layers of 50 μm diameter fibers organized in a 500 μm square grid pattern, providing essential mechanical integrity and shape stability. The intermediate cell-guiding layer, of approximately 200 μm thick, consisted of 20 interwoven triangular grids formed by 15 μm diameter fibers with 160 μm spacing, engineered to facilitate cell infiltration and migration while maintaining structural patency. The innermost layer, resembling a non-woven mat with fiber diameters ranging from 1 to 5 μm and a thickness of 10 μm, was designed for rapid in vivo degradation, functioning as a transient barrier to prevent endometrial overgrowth into the myometrial compartment.

Fig. 1.

Structural and physicochemical characterization of the melt electrowritten PLCL scaffold. (a) Representative image of a molten PLCL filament during extrusion from the MEW printer nozzle. (b) In vitro degradation profile. (c-i) Key in vitro properties of the scaffold, including (c) biocompatibility, (d) elastic modulus, (e) weight-average molecular weight (Mw), (f) retention rate, (g) toughness, (h) elongation at break, and (i) ultimate tensile strength. Abbreviations: DMEM, Dulbecco's Modified Eagle Medium; F12, Ham's F12 Medium; FBS, Fetal Bovine Serum; PLCL, Poly(L-lactide-co-ε-caprolactone); PCL, Polycaprolactone; ECM, Extracellular Matrix.

This design was achieved using the MEW system, which employs metal cylinders to maintain the polymer in a molten state under increased temperatures, with material extrusion achieved through precise air pressure control during extended printing cycles. MEW involves a low flow rate and the material is stored in the metal container, where it is continuously heated. This process keeps the material in a molten, heated state throughout. Given that PLCL has a high melting point, FDM printing requires temperatures above 140 °C [26]. In our study, MEW processing required temperatures ranging from 130 to 170 °C Rheological and thermal degradation experiments simulating this process (Supplementary Fig. 1a and b) demonstrated significant thermal degradation of PLCL, leading to a gradual decline in both viscosity and molecular weight over time. Previous studies have shown that extended MEW processing (e.g., 12 h) of PLCL with an initial molecular weight of 60 kDa can reduce molecular weight by up to 25%. To mitigate these effects, we implemented a processing strategy involving intermittent supplementation with small material aliquots to minimize cumulative thermal exposure, as detailed in Section 2.2. Although the newly introduced material exhibited minor rheological variations compared to those of the initial charge, the established printing parameters remained sufficiently robust to complete the scaffold fabrication (Supplementary Fig. 5).

The in vivo degradation study revealed rapid resorption of the inner layer within 14days post-implantation. The triangular middle layer underwent substantial degradation by one month and was completely resorbed within two months. Conversely, the structurally supportive outer framework demonstrated the slowest degradation rat.(Supplementary Fig. 4).

The cell viability assay results indicate that the scaffold had no significant toxic effect on cell growth at 0.5, 1, 2, and 3 days compared with the normal complete culture medium, suggesting the good biocompatibility of the scaffold (Fig. 1c).

The PLCL scaffolds fabricated using the MEW process exhibited excellent flexibility, with mechanical properties closely resembling those of the native uterine wall (Fig. 1d). The outstanding extensibility of PLCL makes it particularly suitable for soft, expansile tissues such as the uterus, which are subjected to significant tensile stress. Through mechanical testing, we compared the tensile properties of PLCL with those of common scaffold materials—PCL and decellularized extracellular matrix (DECM). The PLCL-based scaffolds had lower stiffness and a softer modulus compared with those of traditional PCL scaffolds, which is more akin to the mechanical behavior of natural extracellular matrix. Despite having lower stiffness, PLCL scaffolds exhibited higher strength and a greater elongation at break compared to DECM scaffolds. This facilitates suturing during surgical procedures and prevents rupture or damage during the early stages of tissue regeneration.

To calculate the mass retention rate in vitro (Fig. 1f), the scaffolds were removed from the dish, cleaned with deionized water and 75% ethanol, air-dried, and weighed. The tensile test was conducted only up to day 30, as the scaffolds had visibly broken by day 60 and fractured before being clamped for testing (Fig. 1g). We observed a gradual decrease in the polymer's molecular weight consistent with the trend in the mechanical properties over time (Fig. 1e). Electron microscopy images (Fig. 1b) revealed that fibers degraded in vitro retained a relatively intact structure, which may be due to the degradation predominantly occurring through hydrolysis, lacking in vivo enzymatic degradation and active cellular phagocytosis [27,28].

Regarding the mechanical properties (Fig. 1i), there was an increase in strength during the early stages of degradation, which may be attributed to polymer chain rearrangement and crystallization [24]. This enhancement maintained the scaffold's structure in the early stages of regeneration, preventing collapse. As degradation progresses, the strength and elongation at break gradually decreased (Fig. 1h and i). This indicates that the scaffold can reduce the restriction on the uterus during the gradual process of uterine repair. Fig. 1e shows the changes in molecular weight (MW) of PLCL at different degradation time points. The "raw" data represents the initial material of PLCL, with a molecular weight of approximately 160 kDa. A significant decrease in molecular weight was observed compared to that of raw material at day 0, primarily due to thermal degradation occurring during the printing process. From days 0 to 90, the scaffold gradually degraded, with a continuous reduction in molecular weight. Notably, mechanical properties were significantly compromised during degradation, with elongation at break demonstrating a pronounced decrease (Fig. 1h) and tensile strength declining after a transient initial increase (Fig. 1i). SEM analysis revealed characteristic surface erosion features, including micro-pits and striations, with localized fracture emergence in advanced degradation stages (Fig. 1b, Supplementary Fig. 2)—observations consistent with mechanical testing data. We posit that the synergistic effects of macrophage encapsulation, phagocytic activity, and enzymatic/radical-mediated degradation significantly accelerate scaffold breakdown in vivo.

3.2. PLCL scaffold-mediated uterine repair and functional recovery in vivo

To evaluate the effect of the PLCL scaffold in promoting regeneration after extensive uterine defect, mice were randomly assigned to the EX, PLCL, and sham-operated control groups.

Postoperative assessment revealed profoundly distinct tissue regeneration outcomes between the EX and PLCL scaffold implantation groups (Fig. 2a–d, Fig. 3a and b). At the 30-day time point, uteri in the EX group displayed pronounced luminal stenosis and hydrometra, accompanied by substantial myometrial discontinuity and extensive endometrial fibrosis, collectively indicating compromised regenerative capacity. In contrast, the PLCL group exhibited well-preserved luminal architecture without evident stenosis, concurrent with partial scaffold degradation and significantly attenuated fibrotic deposition (Fig. 3a and b). Immunohistochemical analysis further demonstrated markedly increased expression levels of key functional markers in the PLCL group, including smooth muscle cells marker α-SMA, endothelial cell marker CD31, and epithelial cell-cell junction marker E-CAD, as well as significantly reduced fibrotic area compared to the EX group.

Fig. 2.

In vivo implantation and evaluation of PLCL scaffolds in a uterine defect model. (a) Schematic diagram of the experimental timeline and key pregnancy-related outcome measures. (b) Establishment of the injury-only control (EX) and scaffold-implanted (PLCL) models, accompanied by representative gross views of uterine tissues harvested at 30- and 60-days post-operation. (c) Histological analysis via Hematoxylin and Eosin (H&E) and immunohistochemical staining for α-smooth muscle actin (α-SMA, red) at the scaffold implantation site, with nuclei counterstained in blue. (d) Fertility outcomes. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3.

PLCL scaffold implantation promotes functional uterine repair over time. Histological and immunohistochemical analysis of tissue regeneration at 30 (a) and 60 (b) days post-treatment. Representative histological sections from the Excision (EX), PLCL, and Control (Ctrl) groups, stained with hematoxylin and eosin (H&E) (tissue architecture), CD31 (vascularization), E-cadherin (E-CAD, epithelialization), Masson's trichrome (fibrosis), and α-SMA (smooth muscle regeneration). Quantification of key parameters at different time points. All quantifications are shown as mean ± standard deviation (SD). ∗p < 0.05, ∗∗p < 0.01; ns, not significant.

By postoperative day 60, while the EX group showed limited evidence of endogenous regeneration, their uteri maintained notable stenosis and hydrometra. Conversely, the PLCL group achieved comprehensive structural restoration, characterized by near-complete myometrial reconstruction with organized smooth muscle bundle alignment, diminished collagen accumulation, and sustained increase in α-SMA expression (Fig. 3a and b). Furthermore, the PLCL group demonstrated significantly enhanced microvascular density, as quantified by CD31-positive staining, as well as expanded endometrial regeneration evidenced by E-CAD immunoreactivity—collectively indicating robust neovascularization and successful tissue reconstitution.

Notably, functional competence of the regenerated uterus was confirmed through mating trials. PLCL scaffold implantation resulted in significantly improved pregnancy outcomes, with embryos successfully implanting and developing within the scaffold-regenerated uterine horns. Quantitative analysis revealed that the PLCL group supported a significantly greater number of developing embryos compared to those of the EX group, unequivocally demonstrating the functional efficacy, superior regenerative performance, and excellent biocompatibility of the PLCL scaffold in restoring uterine reproductive capacity (Fig. 2b,c,d).

3.3. Single-cell transcriptomic profiling reveals scaffold-mediated immune reprogramming

Uterine tissue samples collected 7 days after implantation were subjected to single-cell RNA sequencing to characterize the mechanism underlying scaffold-mediated uterine tissue regeneration. Comparative analysis between the PLCL and EX groups demonstrated that the implantation of scaffold induced substantial recruitment of immunoregulatory cells—particularly neutrophils and macrophages—to the injury site. Subcluster analysis of immune cells showed that PLCL scaffolds specifically promoted the polarization of macrophages to the M2 phenotype. The proportion of M2 in the PLCL group increased compared with that of the EX group, whereas the proportion of M1 decreased. This scaffold-induced immunomodulation effectively established a pro-regenerative microenvironment conducive to tissue repair and structural reconstruction (Fig. 4).

Fig. 4.

Single-cell RNA sequencing (scRNA-seq) reveals the PLCL scaffold-mediated remodeling of the uterine cellular landscape at 7 days post-implantation. (Upper panel) UMAP visualization and quantitative analysis of major parenchymal cell populations in the EX and PLCL groups. Cell types annotated include smooth muscle (SMCs), stromal (SCs), endothelial (ECs), immune (ICs), and epithelial (EPs) cells. (Lower panel) UMAP visualization and proportional analysis of immune cell subsets, including B cells, dendritic cells (DCs), M1 and M2 macrophage subsets, neutrophils (Ne), natural killer (NK) cells, and T cells (TCs), between the EX and PLCL groups.

3.4. Temporal dynamics of macrophage recruitment and phenotypic polarization

Complementing the transcriptomic findings, we performed detailed immunofluorescence analysis of macrophage infiltration and phenotypic polarization at days 3 and 7 post-implantation. Immunostaining for M1 (CD86⁺) and M2 (CD163+) macrophage subsets demonstrated that the PLCL scaffold significantly enhanced overall macrophage recruitment compared to that of the EX group. Temporal analysis revealed a dynamic polarization pattern: while M1 macrophages exhibited an initial increase followed by subsequent resolution, M2 macrophages demonstrated sustained activation throughout the observation period. This temporal shift in the M1/M2 ratio indicated a progressive polarization toward the M2 phenotype, suggesting that the PLCL scaffold populations during the healing process (Fig. 5a)

Fig. 5.

The PLCL scaffold modulates the immune microenvironment by promoting macrophage recruitment and M2 polarization to drive regeneration. (a) Quantitative immunofluorescence staining of macrophage subtypes (M1, red; M2, green) at the implantation site on days 3 and 7, demonstrating a temporal shift from a pro-inflammatory (M1) to a pro-regenerative (M2) phenotype and M1/M2 ratio. (b) Gating strategy for flow cytometric analysis of macrophage depletion and Schematic and results of the scratch wound co-culture assay, where RAW.264.7 macrophages seeded on PLCL scaffolds promoted the migration of primary uterine smooth muscle cells (SMCs), as well as quantitative analysis of SMC migration distance below.(c)Concentrations of cytokines (interleukin [IL]-6, tumor necrosis factor [TNF]-α, transforming growth factor [TGF]-β, and IL-10) in uterine homogenates on days 7 and 14 post-implantation measured by ELISA. (d)Hematoxylin and eosin (H&E) staining of uterine tissues 15 days after clodronate liposome-mediated macrophage depletion. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

Immunofluorescence staining was employed to quantitatively assess temporal macrophage infiltration dynamics at days 1, 3, and 7 post-implantation. During the initial phase (day 1), no statistically significant differences in total macrophage presence were observed between experimental groups. However, by day 3, the PLCL group demonstrated a substantial increase in overall macrophage recruitment, characterized by a significant predominance of the M2 phenotype (CD163⁺, p < 0.01). In contrast, the EX group exhibited only marginal, non-significant increases in both M1 (CD86⁺) and M2 subsets. This polarization trend became more pronounced by day 7, with the PLCL group showing significantly enhanced total macrophage infiltration and a remarkable predominance of M2 macrophages (p < 0.001). Collectively, these findings demonstrate that the PLCL scaffold implantation induces robust macrophage recruitment accompanied by a distinct temporal polarization pattern: an early transient M1 response followed by sustained M2 dominance, suggesting active immunomodulation throughout the repair process (Fig. 5a).

To investigate the effect of macrophages on the migration and proliferation of smooth muscle cells (SMCs), RAW 264.7 macrophages were seeded onto PLCL scaffolds and co-cultured with SMCs. Cell dynamics were observed at 0, 24, and 36 h; the intercellular distance in the co-culture group progressively decreased over time compared with that of the SMCs cultured alone, with the most pronounced difference observed at 36 h (p < 0.05; Fig. 5c). These findings suggest that the presence of macrophages significantly enhances the viability and migratory capacity of SMCs, indicating a potential regulatory role in tissue regeneration.

To functionally validate macrophage involvement in the regenerative process, we established a macrophage depletion model using clodronate liposomes. Flow cytometric analysis confirmed the significant reduction of splenic macrophage populations (F4/80⁺ cells) following 15 days of clodronate administration, while the concurrent increase of CD11b⁺ levels suggested potential compensatory recruitment of other myeloid lineages. Histological evaluation via H&E staining demonstrated that macrophage depletion substantially compromised uterine repair outcomes. Although the PLCL group maintained superior regeneration compared to that of the EX group under depletion conditions, the characteristic well-organized tubular architecture was notably absent (Fig. 5b). These results provide direct functional evidence that macrophages are essential scaffold-mediated uterine regeneration, with their absence resulting in markedly impaired tissue restoration despite scaffold presence, thereby confirming their crucial contribution to the reparative microenvironment.

We compared the complementary ELISA quantification of uterine lysates of EX, PLCL, and sham control groups, and revealed their corresponding cytokine expression profiles. Analysis at day 7 post-implantation showed significantly increased levels of macrophage-derived cytokines in the PLCL group compared to those of both the EX and control groups, which was consistent with the observed cellular recruitment (Fig. 5d). By day 14, while levels of pro-inflammatory mediators (TNF-α and IL-6) remained increased, and a significant increase in the anti-inflammatory cytokine IL-10 was detected specifically in the PLCL group. This cytokine shift indicates a progressive transition from inflammatory dominance toward an immunoregulatory state, coinciding with the observed M2 polarization and suggesting mechanistic involvement in tissue repair resolution.

3.5. PLCL scaffolds drive M2 polarization through coordinated activation of Hippo-YAP and JAK-STAT3 signaling pathways

Transcriptomic profiling of uterine tissues harvested 7 days post-implantation revealed substantial alterations in gene expression patterns between experimental groups. Comparative RNA sequencing analysis identified 148 and 231 significantly upregulated and downregulated genes in PLCL-treated tissues, respectively, compared to those of EX controls. Functional annotation of these differentially expressed genes demonstrated predominant enrichment in biological processes related to immune system activation and positive regulation of host defense mechanisms.

KEGG pathway enrichment analysis further clarified the specific signaling networks modulated by scaffold implantation, identifying significant upregulation in several immunologically relevant pathways; this included phagosome formation and cytokine-cytokine receptor interactions. Protein-protein interaction network construction from the differentially expressed gene set identified several hub proteins, with CD68, Tyrobp, and Fcer1g emerging as central nodes within the PLCL-specific interaction network.

Independent validation by qRT-PCR confirmed significant transcriptional upregulation of multiple macrophage-associated genes, including those encoding chemotactic factors (CCL9, C5AR1, and CCR1), adhesion molecules (ADAM8), and established markers of M2 polarization (LGALS3, APOE, SPP1, HMOX-1, and TREM-2). These molecular findings provide compelling evidence that the PLCL scaffold favors a microenvironment conducive to macrophage recruitment and subsequent polarization toward the M2 phenotype (Fig. 6c).

Fig. 6.

PLCL scaffold promotes uterine regeneration by activating YAP/STAT3 signaling and modulating macrophage-related genes. (a) Transcriptome profiling via RNA-seq reveals global gene expression changes in the regenerating uterine tissue at the scaffold-implanted site relative to those of the EX control site. (b) Protein-level validation by Western blot demonstrates the activation (e.g., increased phosphorylation) of the YAP and STAT3 pathways. (c) qPCR analysis shows a significant increase in the expression of specific genes driving macrophage polarization following scaffold implantation.

To elucidate the downstream signaling mechanisms underlying enhanced myometrial regeneration, we performed Western blot analysis of key regenerative pathways. The results confirmed that PLCL scaffold implantation significantly activated both the Hippo-YAP and JAK-STAT3 signaling axes, as demonstrated by increased phosphorylation of YAP and STAT3, accompanied by increased expression of their downstream target proteins (Fig. 6b). The coordinated activation of these evolutionarily conserved pathways that regulate cell proliferation, survival, and immune modulation, provides a mechanistic foundation for the observed functional recovery in scaffold-implanted uteri.

4. Discussion

In this study, we engineered a biomimetic three-layer PLCL for replicating the structural complexity of the native uterus. Beyond its structural role, the scaffold actively enabled an immunomodulatory response characterized by significant M2 macrophage polarization, favoring a pro-regenerative microenvironment that ultimately led to functional uterine restoration and successful pregnancy outcomes in a murine model. Furthermore, the scaffold effectively reduced scar tissue formation and promoted the remodeling of the smooth muscle layer. Histological evaluation revealed a significant reduction in fibrosis and a more organized regeneration of the myometrial layer due to the PLCL scaffold. A complete and well-structured smooth muscle layer is crucial for maintaining uterine contractile function and structural integrity, which are key factors for achieving live birth.

Current paradigms in uterine tissue engineering have predominantly emphasized endometrial regeneration employing decellularized uterine matrices, collagen-based constructs, PCL composites, and various hydrogel systems. While these approaches have demonstrated efficacy in repairing focal endometrial injuries and supporting subsequent pregnancy [29,30], they inefficiently address extensive, full-thickness defects involving both endometrial and myometrial layers. The inherent regenerative capacity of the uterus proves insufficient to reconstruct native architecture and reproductive functionality in such severe cases [31,32], with myometrial repair being a particularly underexplored domain [33]. Our results address this clinical challenge, revealing that while rodents display spontaneous endometrial regeneration, myometrial repair proceeds with considerably slower kinetics and inferior quality. The differential regenerative rates between these tissue layers—with endometrial proliferation outpacing myometrial restoration—creates a fundamental biological discrepancy. Unchecked endometrial expansion may potentially impose contact inhibition upon the underlying myometrium, resulting in a compromised, disorganized muscular architecture incapable of supporting normal gestation to term, even when embryo implantation initially occurs.

Our molecular analyses provide compelling evidence that the PLCL scaffold functions as a potent immunomodulatory platform. Transcriptomic profiling revealed significant upregulation of macrophage polarization-associated genes, consistent with observed enhancement in macrophage recruitment and skewing toward the M2 phenotype. This immunomodulatory capacity was further validated through in vitro co-culture systems, wherein macrophage-laden scaffolds elicited pro-regenerative responses in primary smooth muscle cells. The functional necessity of macrophages was unequivocally demonstrated through depletion experiments using clodronate liposomes, which substantially attenuated the regenerative process. The concomitant increase of CD11b + cells following depletion suggests compensatory involvement of alternative myeloid lineages, highlighting the complexity of immune network interactions during scaffold-mediated repair. Collectively, these findings establish macrophage-dependent immunomodulation as essential in the scaffold-enhanced regeneration mechanism.

The crucial role of macrophage phenotypic plasticity in tissue repair is increasingly recognized across diverse regeneration models, including cutaneous wound healing [34], myocardial infarction [35] and arthritis conditions [36]. Macrophages demonstrate remarkable functional adaptability in response to microenvironmental cues, traditionally categorized into pro-inflammatory (M1) and anti-inflammatory, pro-regenerative (M2) phenotypes. The transition to M2 polarization is characterized by the secretion of anti-inflammatory cytokines (e.g., IL-10 and TGF-β) and an array of growth factors that collectively coordinate extracellular matrix reorganization, angiogenesis, and parenchymal cell proliferation. Thus, the timely transition from inflammatory to regenerative macrophage populations represents a critical determinant of tissue homeostasis and repair efficacy [[37], [38], [39]].

Understanding the mechanisms behind macrophage polarization is crucial for developing scaffolds that not only support tissue regeneration but also modulate the immune response. In this context, the properties of biomaterials, particularly their architecture and biophysical cues, are crucial for guiding macrophage behavior.

Biomaterial properties profoundly influence this phenotypic switching. While material composition is important, emerging evidence underscores the critical role of scaffold geometry and biophysical cues in immunomodulation. Studies have demonstrated that specific architectural features, such as the size of pore structures from 200 to 500 μm in polyetheretherketone scaffolds, preferentially promote M2 polarization compared to their non-porous counterparts [40,41]. Crucially, recent studies utilizing MEW have elucidated the link between fiber architecture, cell morphology, and immune response. Tylek et al. demonstrated that MEW scaffolds with precise micro-porosity (down to 40 μm) induce significant macrophage elongation. This morphological restriction serves as a potent biophysical signal, driving spontaneous polarization toward a pro-healing M2-like phenotype even in the absence of exogenous biochemical stimuli [42]. Similarly, Mondadori et al. reported that specific fibrous geometries, such as rhomboidal grids, can actively modulate the cytokine secretion profile of macrophages, enhancing the release of anti-inflammatory factors (e.g., IL-10 and IL-13) compared to that of flat films [43]. Agreeing with these findings, our PLCL scaffold's fibrous architecture likely provides similar geometric constraints that induce cell alignment and elongation, thereby mechanically priming the observed M2 dominance.

The propensity of PLCL to facilitate M2 polarization has been similarly documented in vascular regeneration contexts [44]. The unique anatomical configuration of the uterus—featuring distinct tissue layers, asynchronous regenerative requirements, and necessity for profound volumetric expansion during pregnancy—imposes particular design constraints for biomaterial development. Consequently, the creation of customized scaffolds capable of directing macrophage phenotypic conversion represents an emerging frontier in uterine regenerative medicine. Our strategic approach, involving precise calibration of material parameters such as degradation kinetics and molecular weight across different scaffold regions, is aimed to spatiotemporally coordinate with the heterogeneous regenerative programs of distinct uterine tissues, with the ultimate objective of restoring functional competence.

Beyond immunomodulation, our findings highlight the involvement of mechanotransductive signaling pathways in the regenerative outcomes. Yes-associated protein (YAP) activation demonstrates exquisite sensitivity to mechanical inputs, including substrate stiffness [[45], [46], [47]]. Our results indicate that the PLCL scaffold, with its multilayered architecture and bulk stiffness approximating 100 kPa, provides sufficient mechanical stimulation to activate YAP phosphorylation. This finding is crucial, as YAP serves as a well-established master regulator of cellular proliferation and fate determination. Ectopic YAP activation drives hyperplastic growth in multiple organ systems and promotes expansion of progenitor cell populations in diverse tissues, including the skin and intestinal epithelium [[48], [49], [50]]. Particularly relevant is the demonstrated role of YAP/TAZ signaling in the regenerative capacity of cardiac injury models [[51], [52], [53]], suggesting the conservation of this mechanosensitive pathway across multiple soft tissue regeneration contexts, including uterine repair. The coordinated activation of both YAP and STAT3 pathways by our scaffold indicates a sophisticated interplay between mechanosensitive and cytokine-mediated signaling networks in driving functional uterine regeneration.

5. Conclusion

This study clarifies a sophisticated mechanism through which a biomimetically engineered PLCL scaffold facilitates the functional regeneration of complex uterine defects encompassing both endometrial and myometrial components. We demonstrated that the scaffold functions as a structural support and active biological platform that favors a complex regenerative response, which includes the targeted recruitment and M2 polarization of macrophages, enhanced proliferation and migration of uterine parenchymal cells, and concurrent activation of the mechanosensitive Hippo-YAP and cytokine-responsive JAK-STAT3 signaling pathways. The synergy of these immunomodulatory, cellular, and molecular responses effectively establishes a pro-regenerative niche that supports coordinated tissue restoration, including neovascularization and the reconstruction of histologically distinct uterine layers. Consequently, this scaffold-mediated approach achieves comprehensive repair of extensive uterine defects and culminates in the restoration of reproductive competence. This study advances our fundamental understanding of uterine regeneration and establishes a foundational framework for developing immunomodulatory biomaterials in regenerative medicine. Future studies should focus on elucidating the temporal coordination between the identified pathways and evaluating the translational potential of this scaffold-based strategy for clinical applications in human uterine repair.

Ethics approval

All animal experimentation was endorsed by the Animal Ethics Committee of Zhejiang Provincial People's Hospital (No. 20231213144812203080).

CRediT authorship contribution statement

Qianqian Wei: Investigation, Writing – original draft. Jing Zhang: Data curation, Funding acquisition, Writing – review & editing. Jiahao Weng: Methodology, Validation. Jiawen Lan: Formal analysis, Investigation. Jing He: Software, Visualization. Yuewei Chen: Funding acquisition, Methodology, Validation. Zhongfei Zou: Funding acquisition, Supervision. Yong He: Conceptualization, Funding acquisition, Resources. Yanpeng Wang: Conceptualization, Funding acquisition, Investigation. Jing Shu: Conceptualization, Funding acquisition, Methodology, Project administration.

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 ASupplementary data

The following is the supplementary data to this article:

Data availability

Data will be made available on request.