The bio-based high-precision resin eResin-PLA Pro (PH100) was obtained from eSUN®. Primary antibodies including rabbit monoclonal anti-Vimentin (Vim, D21H3) XP antibody (#5741), horse chimeric monoclonal anti-11b/ITGAM antibody (#78923), rat monoclonal anti-CD44 (IM7) antibody (#39037), rabbit monoclonal anti-FAK (D5O7U) XP antibody (#71433) were obtained from Cell Signaling Technology, Inc. Primary antibodies including mouse monoclonal anti-Cytokeratin 20 antibody (CK20, GTX72046), goat polyclonal alpha Smooth Muscle Actin antibody (α-SMA, GTX89701), rat monoclonal anti-Mannose receptor antibody (CD206, GTX42265), mouse monoclonal anti-CD86 antibody (GTX34569), mouse monoclonal anti-PCNA antibody (GTX20029) were purchased from GeneTex Inc. RadioImmunoPrecipitation Assay buffer (20–188), dispase II (D4693), collagenase Type I (SCR103), trypsin-EDTA (T4049), dulbecco's modified eagle medium (D5030), fetal bovine serum (12103C) were obtained from Sigma-Aldrich. TMR (red) Tunel Cell Apoptosis Detection Kit (G1502), IF488-Phalloidin, a F-actin cytoskeleton staining reagent (G1248), 4',6-diamidino-2-phenylindole (DAPI) Staining Solution (ready-to-use, G1012), Hematoxylin and Eosin (H&E) Staining Kit (G1005), Masson’s Trichrome Staining Kit (MTC, G1006), Elastica van Gieson (EVG) Staining Kit (GP1035), Modified Picrosirius Red Staining Kit (RS, GP2088), Periodic Acid-Schiff (PAS) Staining Kit (G1008), Alcian Blue Staining Kit (AB, G1027), paraffin wax (WGHB-319213129), 20× Citrate Antigen Retrieval Solution (pH 6.0, G1202), Triton X-100 (GC204003), and Normal Donkey Serum for blocking (undiluted, G1217), Trizol (G3013), SweScript RT II Reagent Kit (G3333), were purchased from Wuhan Servicebio Technology Co., Ltd.. Meglumine diatrizoate (contrast agent, M5266) was purchased from Sigma-Aldrich (St. Louis, MO, USA), and the Calcium Colorimetric Assay Kit (S1063S) was obtained from Beyotime (Shanghai, China). Commercial sterile SIS (Cook Biotechnology, G34980) was used as the control scaffold. Dispase II (17105041), Collagenase Type I (17100017), and other cell culture reagents, including Trypsin-EDTA (0.25%, 15400054), Dulbecco's modified eagle medium (10566016), fetal bovine serum (A5670701) were purchased from Gibco (Thermo Fisher Scientific, Waltham, MA, USA). Rabbit inducible Nitric Oxide Synthase (iNOS, BY-ER772509) ELISA Kit and Rabbit Arginase 1 (Arg-1, BY-ER7101251) ELISA Kit were purchased from Boyan Biotechnology.

The fabrication of urethral defect scaffolds was based on the method described in a previously published study [37], with the same design parameters retained to suit the hollow anatomical characteristics of the urethra during urination. A cylindrical porous scaffold was designed using SolidWorks software (Dassault Systèmes, France) with an inner diameter of 2 mm, matching the typical clinical urethral lumen dimensions, and a wall thickness of 0.5 mm. The scaffold featured a grid-like porous architecture to facilitate tissue ingrowth and nutrient transport, while mimicking the three-dimensional structure of native urethral tissue and providing appropriate mechanical support. Upon completion of the design, the model was exported in standard tessellation language format for subsequent 3D printing.

A bio-based high-precision resin, eResin-PLA Pro, was selected as the printing material due to its excellent biocompatibility, favorable mechanical properties, and suitability for micro-scale 3D printing. According to the manufacturer’s documentation, this material has passed biocompatibility testing in accordance with ISO 10993 standards, making it suitable for in vivo implantation. Fabrication was carried out using an M-dental U60 DLP-based 3D printer (Ningbo Inteplast, China) equipped with a 405 nm light source. The STL file was imported into the slicing software, and the following printing parameters were set: layer thickness = 25 µm, exposure time per layer = 8–10 seconds, bottom exposure time = 60 seconds, with support structures automatically generated. Prior to printing, the build platform was calibrated to ensure geometric accuracy and surface quality, meeting the requirements for urethral repair applications.

Prior to printing, the build platform was calibrated to ensure geometric accuracy and surface quality, meeting the requirements for urethral repair applications. After printing, the scaffolds were subjected to ultraviolet irradiation for 30 minutes on each side for sterilization prior to biological use.

All experiments were approved by the Medical Laboratory Animal Center of Dongguan University of Technology and conducted in strict accordance with the institutional guidelines for animal care and use. Thirty male New Zealand white rabbits (1.5 months old, weighing 1.5–2.0 kg) were randomly assigned as follows: 10 rabbits were subcutaneously implanted with 1 scaffold each, yielding 10 TEUGs for autologous in vivo urethral defect replacement surgery; 10 rabbits received commercial SIS and served as the in vivo control group; the remaining 10 rabbits were subcutaneously implanted with 2 scaffolds each, obtaining 20 TEUGs for in vitro biophysical, chemical, and other characterizations, as well as characterization experiments of interactions between in vitro urethral epithelial cells and TEUG.

Animals were anesthetized via intravenous injection of 2% pentobarbital sodium (45 mg/kg) through the marginal ear vein. Anesthesia depth was monitored by assessing reflexes, and body temperature was maintained with a heating pad. Postoperatively, meloxicam (0.2 mg/kg) was administered subcutaneously 10 minutes after surgery for pain relief, with daily supplementation as needed. Animals were closely observed during recovery and housed under standard conditions with ad libitum access to food and water.

TEUGs were prepared by implanting sterile PLA scaffolds (inner diameter: 2 mm, wall thickness: 0.5 mm, length: 2.2 cm) subcutaneously into the abdominal region of rabbits. Under anesthesia, a small skin incision was made, and the scaffold was inserted into a subcutaneous pocket. The wound was closed with intradermal sutures.

After 14 days, the implant was retrieved under anesthesia. The scaffold, together with the surrounding tissue capsule, was excised en bloc. A cylindrical segment (~ 0.2 mm in length) was cut from the middle of the construct and longitudinally sectioned to obtain inner and outer surface samples.

For scanning electron microscopy (SEM) analysis, samples were fixed in electron microscopy fixative for 2 h, dehydrated through a graded ethanol series, critically dried, and sputter-coated with gold (HITACHI MC1000). Surface morphology was examined using a HITACHI Regulus 8100 scanning electron microscope (Hitachi, Japan).

Body weight and body temperature were monitored before surgery and at regular intervals post-implantation to assess systemic responses. Unused samples were stored at − 20°C for future use.

Fourteen days after TEUG implantation, rabbits in both the TEUG and SIS groups underwent anterior (pendulous) urethral reconstruction surgery under intravenous anesthesia. A midline abdominal incision was made to expose the bulbar urethra extending to the suspensory ligament. A 2.2 cm full-thickness circumferential segment of the pendulous urethra was excised to create a long-segment urethral defect. The surgical site was continuously irrigated with sterile saline to prevent tissue desiccation.

The excised native urethral tissue was immediately placed in sterile physiological saline, maintained at 4°C during handling, and subsequently transferred to a − 20°C freezer for storage until further analysis. These samples were used for subsequent characterization by atomic force microscopy (AFM), SEM, and histological evaluation.

For reconstruction, the pre-formed TEUG was used directly for end-to-end anastomosis. In contrast, the SIS patch was first manually rolled into a tubular structure (inner diameter ~ 2.5 mm) and sutured with 6 − 0 absorbable sutures under sterile conditions before implantation. Anastomosis was performed in an end-to-end fashion using 8 − 0 polydioxanone (PDS-II) sutures in a continuous running pattern, ensuring a watertight, tension-free closure with maintained luminal patency.

An F8 silicone catheter was inserted intra-luminally and left in place for 10 days to provide structural support, prevent stricture formation, and facilitate urinary drainage. The catheter was exteriorized through the distal urethra and secured in place. The surrounding soft tissues were repositioned, and the abdominal wall and skin were closed in layers.

Postoperatively, all animals received subcutaneous meloxicam for analgesia and were monitored daily for urination patterns, wound healing, and potential complications such as infection or urinary leakage. Animals were housed under standard conditions with free access to food and water.

AFM was used to assess surface topography, roughness, and mechanical properties of TEUGs and native urethral tissues. Native tissue samples, previously stored at − 20°C, were thawed and equilibrated in physiological saline for 2 h before testing.

Measurements were performed in saline using an AFM (Bruker Dimension ICON, Bruker, Germany) with a silicon nitride cantilever (spring constant ~ 0.58 N/m, tip radius ~ 20 nm). Surface morphology and roughness (Ra) were analyzed in tapping mode over 20 × 20 µm² areas.

Nanomechanical properties were evaluated in Indentation Retrace (IdR) mode. Young’s modulus was calculated from force–distance curves using the Hertz model, and spatial maps of mechanical response (including elasticity and adhesion) were generated across the inner and outer surfaces of the samples.

Data were analyzed using standard software (e.g., NanoScope Analysis). At least five regions from three independent samples per group were measured.

To evaluate the surface chemistry, wettability, and viscoelastic properties of the TEUGs and native urethral tissues, the following analyses were performed.

X-ray Photoelectron Spectroscopy (XPS) was conducted using a Thermo Scientific K-Alpha 250Xi (Thermo Fisher Scientific, USA) equipped with a monochromatic Al Kα X-ray source (1486.6 eV). Survey and high-resolution scans (C 1s, O 1s, N 1s) were acquired at a pass energy of 100 eV and 30 eV, respectively. Data were analyzed using Avantage software, with charge correction referenced to the C–C/C–H peak at 284.8 eV.

Attenuated Total Reflectance–Fourier Transform Infrared (ATR-FTIR) spectroscopy was performed on a Bruker TENSOR27 spectrometer (Bruker, Germany) equipped with a diamond ATR crystal. Spectra were collected over the range of 4000–400 cm⁻¹ at a resolution of 4 cm⁻¹, with 32 scans averaged per sample. Background correction was performed prior to each measurement.

Static Water Contact Angle (WCA) was measured using a Dataphysics OCA Pro15 instrument (DataPhysics Instruments GmbH, Germany). A 5 µL droplet of deionized water was dispensed onto the flat surface of the sample (inner and outer layers separately), and images were captured immediately after droplet stabilization. Contact angles were determined by the sessile drop method using SCA20 software. At least five measurements were taken from different locations on each sample (n = 3 per group).

Cyclic Stress-Relaxation Testing was performed using a TA Instruments Q800 dynamic mechanical analyzer (TA, USA) in tension mode. Rectangular strips of TEUG and native urethral tissue (n = 5 per group) were preconditioned in physiological saline at 37°C. Tests were conducted under 25% strain amplitude for 10 consecutive cycles at a frequency of 1 Hz. After the 8th, 9th, and 10th cycles, the samples were held at zero stress for relaxation periods of 15, 20, and 30 minutes, respectively, while the decay in restoring force was recorded. The hysteresis loops and residual stress after each relaxation phase were analyzed to evaluate energy dissipation, elastic recovery, and viscoelastic stability.

All samples were equilibrated in physiological saline for 2 h before testing. At least three independent samples per group were analyzed for each assay.

Histological analysis was performed on paraffin-embedded tissue sections (5 µm) from TEUG obtained via subcutaneous self-assembly, and replacement samples harvested at 1-, 3-, and 6-months post-surgery. Sections were stained with H&E, MTC, EVG, PAS, AB (pH 2.5), and SR, following the manufacturers’ instructions.

For immunofluorescence staining, deparaffinized sections were subjected to heat-induced antigen retrieval using a citrate-based buffer (pH 6.0) in a pressure cooker, followed by blocking with 5% normal goat serum in PBS for 1 h at room temperature. Sections were then incubated overnight at 4°C with primary antibodies diluted in antibody diluent, washed, and incubated with fluorescently labeled secondary antibodies for 1 h at room temperature. Nuclei were counterstained with DAPI. All slides were imaged using a whole-slide scanner (3DHISTECH, Hungary), except for SR-stained sections, which were examined under both bright-field and polarized light microscopy (PM). Semi-quantitative analysis of positive staining areas was performed using CaseViewer v2.4 and Fiji ImageJ-win64 (NIH, USA) after thresholding and background subtraction.

Urethral epithelial cells (UECs) were isolated from rabbit urethral mucosa using a previously described enzymatic digestion and explant culture method [38]. In brief, the epithelial layer was separated and sequentially digested with dispase II at 4°C overnight and collagenase Type I at 37°C. Released cells were dissociated with 0.25% trypsin-EDTA, seeded in dulbecco's modified eagle medium supplemented with 15% fetal bovine serum, and cultured under standard conditions (37°C, 5% CO₂). Cells from passages 2–4 were used for experiments, with epithelial identity confirmed by immunofluorescence staining for CK20 in preliminary batches.

For interaction studies, sterile TEUGs and SIS scaffolds were cut into 8 mm diameter discs and pre-wetted in culture medium to ensure full hydration prior to cell seeding. UECs were seeded onto the luminal surface at a density of 5 × 10⁴ cells per scaffold. Cell adhesion, spreading, and monolayer formation were evaluated by SEM at 12, 48, 96, and 168 h post-seeding, with SIS serving as the control.

At 168 h, TEM was performed to examine focal adhesions and cytoskeletal (microfilament) organization of UECs on the luminal surface of TEUGs. Samples were fixed, embedded, and ultrathin-sectioned (~ 50 nm) using a Leica EM UC7 ultramicrotome. Sections were analyzed under a HT7700 TEM (Hitachi, Japan) at 100 kV.

Additionally, AFM was used in Dissipation Retrace (DsR) mode (Bruker Dimension ICON, Germany) to map the spatial distribution of energy dissipation during probe retraction, reflecting nanoscale variations in cell-surface interaction dynamics. Dissipation maps were acquired at high resolution, with color coding indicating local differences in energy loss across the luminal surfaces of TEUGs and SIS.

Immunofluorescence staining for FAK and F-actin/DAPI was performed to visualize the spatial organization of focal adhesions and the actin cytoskeleton. Fluorescence images were analyzed using Fiji ImageJ software to quantify FAK expression intensity on the inner surfaces of TEUGs versus SIS scaffolds.

To evaluate the structural and functional outcomes of urethral replacement, rabbits were subjected to multimodal follow-up assessments at predetermined time points after implantation. At 1 month postoperatively, cross-sectional computed tomography (CT) was performed using a TC-460 CT scanner (Shenzhen TUTOM Medical Technology Co., Ltd., Shenzhen, China) to visualize the morphology and luminal patency of the implanted TEUGs in vivo. Concurrently, calcium wave imaging was conducted to assess intercellular signaling and physiological responsiveness of epithelial cells at the graft site.

At 3 months, urethral pressure profilometry was performed to evaluate urinary continence function and lumen patency. A pediatric urodynamic catheter was inserted retrogradely into the urethra under anesthesia, and continuous saline perfusion was maintained while slowly withdrawing the catheter. The BL420 Biological Signal Acquisition System (Chengdu Taimeng Science and Technology Co., Ltd., Chengdu, China) was used to record the maximum urethral closure pressure and pressure profiles along the urethral axis, allowing assessment of urethral closure capacity, smooth muscle contractility, and overall patency.

At 6 months, retrograde urethrography was carried out using the Elite Urethrogram System (Shenzhen TUTOM Medical Technology Co., Ltd., Shenzhen, China) to assess the anatomical structure and morphological integration of the regenerated urethra. After catheterization via the external urethral orifice, iopamidol contrast agent was injected retrogradely, and radiographic images were acquired in anteroposterior and oblique views to evaluate lumen continuity, diameter uniformity, and the presence of strictures or diverticula.

All procedures were performed under intravenous anesthesia, and animals were monitored for general health and voiding behavior throughout the 6-month follow-up period.

TEUG and SIS grafts (mung bean-sized each) were harvested from the urethral defect sites of New Zealand rabbits at 6 months post-operation, along with healthy posterior urethral tissue from rabbits without urethral replacement surgery (served as positive control). Tissues for qPCR were immediately frozen in liquid nitrogen and stored at − 80°C, while grafts for ELISA were homogenized on ice with radioimmunoprecipitation assay lysis buffer, centrifuged at 12,000 rpm for 20 minutes at 4°C, and the supernatant was collected to determine protein concentration.

For qPCR, total RNA was extracted using Trizol reagent. After verifying RNA purity (A260/A280 ratio: 1.8–2.0) with a NanoDrop spectrophotometer (UH5210, Hitachi, Japan), 1 µg of RNA was reverse-transcribed into cDNA using the SweScript RT II Reagent Kit. Amplification was performed on a Bio-Rad CFX96 Touch™ System (Bio-Rad, USA) with TB Green™ Premix Ex Taq™ II (Takara, Japan) in a 20 µL reaction system. The protocol included initial denaturation at 95°C for 30 s, 40 cycles of 95°C for 5 s and 60°C for 30 s, followed by melting curve analysis (65–95°C). Relative mRNA expression levels of CD206, CD86, CD27, CD28, and CD4 were calculated via the 2 − ΔΔCt method, with GAPDH as the endogenous reference; results were presented as fold changes relative to the positive control.

The rabbit-specific primer sequences were as follows:

GAPDH: Forward: 5′-TGAAGGTCGGAGTGAACGGAT-3′ Reverse: 5′-CGTTCTCAGCCTTGACCGTG-3′; CD206: Forward: 5′-GCTAAATGGGAGAATCTGGAATGT-3′ Reverse: 5′-GACTAGGACAGTTGGTGGGCA-3′; CD86: Forward: 5′-GGACTGAGTGTCACGGTCTTTG-3′, Reverse: 5′-GATACACGCCCTTGTCCTTGA-3′; CD27: Forward: 5′-GTCTCGGAGACAGAAACAAGCAA-3′, Reverse: 5′-TTTCTTGTAGCAGCAACCTCCAC-3′; CD28: Forward: 5′-ACGATGATTCTCAGGCTGCTC-3′, Reverse: 5′-GCTCAGGTTGACCTCGTTGT-3′; CD4: Forward: 5′-AACATGCGAACCAGGTCAAGA-3′, Reverse: 5′-CCGAGTCGTCCATCCTGAGAT-3′.

ELISA was conducted following the instructions of Rabbit iNOS and Arg-1 ELISA Kits. 100 µL of standards or samples was added to each well and incubated at 37°C for 1 hour. After discarding the liquid, detection antibody was added for another 30-minute incubation at 37°C. The plate was washed 5 times, then HRP-conjugated secondary antibody was added and incubated for 30 minutes at 37°C, followed by 5 more washes. Substrate chromogenic solution was added for 15-minute light-protected incubation at 37°C. Stop solution was then added, and absorbance at 450 nm was measured using a microplate reader to calculate the target protein expression levels.

Data are presented as mean ± standard deviation (SD). Comparisons between two groups were performed using Student’s t-test. For three or more groups, one-way ANOVA was used followed by Tukey’s post hoc test for multiple comparisons. All statistical analyses were conducted using GraphPad Prism 9.0 (GraphPad Software, USA). A p value < 0.05 was considered statistically significant.

In this study, a novel strategy combining "3D-printed biodegradable scaffold subcutaneous pre-implantation" with "host cell self-assembly and ECM secretion" was employed to engineer TEUGs. A 3D-printed biodegradable tubular scaffold (inner diameter: 2 mm, wall thickness: 500 µm, length: 2.2 cm) was subcutaneously implanted into New Zealand rabbits for 2 weeks (Figure S1). During this period, host-recruited cells, including fibroblasts, monocytes, and adipose-derived stem cells, migrated into the scaffold and actively secreted and organized ECM through autocrine signaling, with fibroblasts and stem cells specifically contributing to the synthesis of PG matrix (a core functional component of ECM), ultimately forming a bioactive tubular graft rich in integrated PG matrix. The experimental workflow is illustrated schematically in Fig. 1A. Body temperature and body weight of animals with and without subcutaneous TEUG implantation were comparable at all postoperative time points (Figure S2), with no statistically significant difference (p > 0.05). The resulting TEUG (Fig. 1B) exhibited a well-defined tubular morphology, meeting the macroscopic requirement for maintaining luminal patency during urination.

To characterize the cellular composition of the TEUG, triple-label four-color immunofluorescence staining was performed (Fig. 1C). Co-staining for Vim (a fibroblast marker), CD11b (a macrophage marker), CD44 (a cell surface molecule), and DAPI (nuclear staining) revealed abundant host-derived cells within the TEUG. Quantitative analysis (Fig. 1D) showed the following cell proportions: fibroblasts (Vim⁺ cells: 48.55% ± 7.55%), macrophages (CD11b⁺ cells: 12.75% ± 3.12%), and CD44⁺ putative stem/progenitor cells (3.62% ± 0.76%). These cell populations collectively constitute the primary cellular basis for ECM production.

Histological and quantitative analyses demonstrated that the TEUG closely recapitulates the biochemical characteristics of native urethral ECM, with PG matrix serving as a key bridge between structural components (collagen, elastin) and functional properties. EVG staining and quantitative analysis revealed no significant difference in elastin content between TEUG and native urethra (Fig. 1E). MTC and SR staining, combined with PM (Fig. 1F), showed that the collagen architecture in TEUG closely resembles that of native urethra: under PM, type I collagen exhibited bright yellow birefringence with an orderly arrangement (predominant component), while type III collagen displayed green birefringence with a fine mesh-like distribution (supporting component). Quantitative analysis of collagen I/III gray values showed no significant difference between TEUG and native urethra (p > 0.05, Fig. 1E).

AB staining (targeting aGAGs) and PAS staining (targeting nGAGs) (Fig. 1H) confirmed the presence of both aGAGs and nGAGs in both TEUG and native urethra, these GAGs are the major carbohydrate moieties of PG matrix, indicating consistency in the fundamental types of polysaccharides within PG matrix and associated functional properties such as water retention and structural support. However, quantitative analysis using the rose plot in Fig. 1I revealed notable differences in polysaccharide proportions: in TEUG, aGAGs accounted for 23%–36% and nGAGs for 50%–67%, whereas in native urethra, aGAGs constituted 50%–59% and nGAGs 31%–45%. This indicates significant differences in the relative abundance and total amount of GAGs (and thus PG matrix subtypes) between the two tissues. Collectively, these results demonstrate that the TEUG accurately reproduces the biochemical composition, spatial organization, and relative abundance of major ECM components, such as collagen, elastin, and PG matrix (via its GAGs components), present in native urethral tissue.

To evaluate the structural and mechanical biomimetic properties of TEUG, we employed AFM imaging and force mapping techniques to characterize the ultrastructure and nanomechanical properties of its ECM. Commercial SIS was not used as a control in the experiment; instead, native rabbit urethral tissue was selected for comparison. The reasons are as follows: SIS is a decellularized dense laminate that lacks the critical PG matrix of native urethral tissue, retaining only hydrophobic collagen and elastin fibers. Additionally, its dense and heterogeneous structure easily causes AFM probe slippage and unreliable measurement results, failing to meet the requirement for data accuracy. In contrast, native rabbit urethral tissue contains a complete PG-collagen-elastin network, and its structural and mechanical properties can serve as a physiological benchmark, directly verifying the ability of TEUG to replicate native tissue and providing reliable data for subsequent studies. As shown in Fig. 2A, low-magnification and high-magnification SEM morphological images of cross-sectional samples revealed that TEUG exhibited a continuous and dense tubular structure with uniform ECM distribution, which was highly similar to the overall structural organization of native urethra. Figure 2B further displayed high-magnification views of the inner and outer surfaces of TEUG, showing a porous network composed of interwoven nanofibers—this structure was highly biomimetic to the topological characteristics of native ECM. Quantitative analysis of the diameter of cross-sectional nanofibers (Fig. 2C) showed that the average diameter of TEUG was (73.21 ± 21.4) nm, while that of native urethra was (81.26 ± 18.9) nm, with no statistically significant difference between the two groups (p > 0.05). This indicated that TEUG successfully replicated the nanofiber scale characteristics of native urethral ECM.

AFM topographic imaging (Fig. 2D), presented in both 2D and 3D formats, combined with quantitative analysis (Fig. 2E), reveals a surface roughness Ra of (65.33 ± 24.3) nm for TEUG and (54.33 ± 26.1) nm for native urethra, with no significant difference, suggesting excellent topographical biomimicry of TEUG. Further AFM-based Young’s modulus mapping (Fig. 2F) shows that TEUG exhibits a modulus range of (10.4–27.2) GPa, with a predominantly yellow-green color distribution, indicating relatively uniform stiffness. In contrast, the native urethra displays a modulus range of (12.6–25.4) GPa, with localized red regions (> 25 GPa) likely corresponding to dense collagen bundles or nuclear structures. Analysis of 3D mechanical topography indicates that while the native urethra exhibits higher local rigidity, its structure is more orderly, whereas TEUG, though slightly stiffer on average, shows greater surface irregularity and some mechanical heterogeneity.

Force–displacement curve analysis (Fig. 2G) further reveals that the modulus distribution in native urethra is more concentrated, while that of TEUG is broader, suggesting microstructural inhomogeneity due to incomplete ECM maturation. Nevertheless, elastic modulus mapping (Fig. 2H) demonstrates that the spatial distribution of mechanical properties in TEUG is highly similar to that of native urethra, exhibiting comparable elastic heterogeneity, highlighting its favorable structural biomimicry and mechanical compatibility. These data demonstrated that the TEUG closely recapitulates the ECM ultrastructure, surface morphology, and nanoscale mechanical behavior of native urethral tissue.

XPS analysis (Fig. 3A) revealed that both the TEUG and native rabbit urethral tissue exhibit dominant C1s, N1s, and O1s peaks on their surfaces, these elements are the primary components of PG matrix (via its protein core and GAG sugar chains), indicating that both materials are primarily composed of organic components rich in carbon, nitrogen, and oxygen. High-resolution deconvolution of the C1s spectrum identified characteristic functional groups associated with proteins and polysaccharides, including aliphatic carbon chains, carbonyl (C = O), and hydroxyl (C–OH) moieties. The relative abundances of these components are highly similar between TEUG and native tissue, suggesting good consistency in surface chemical composition.

In the ATR-FTIR spectra (Fig. 3B), both TEUG and native urethral tissue display a broad and intense absorption band around 3300 cm⁻¹, attributed to overlapping O–H and N–H stretching vibrations from hydrogen-bonded water, proteins, and GAGs (the carbohydrate component of PG matrix). Both samples show characteristic ECM signatures: amide I and II bands at 1654 cm⁻¹ and 1750 cm⁻¹ (indicative of peptide backbone –CONH–), alkyl chain vibrations at 1381 cm⁻¹ and 1358 cm⁻¹ (characteristic of elastin-like hydrocarbon structures), C–O–C stretching vibrations at 1181 cm⁻¹ (assigned to GAG sugar ring motifs), and = C–H bending vibrations at 870 cm⁻¹ (suggestive of aromatic glycoproteins). These shared molecular features indicate that TEUG effectively recapitulates the biochemical molecular structure of the native urethral ECM.

Static WCA measurements (Fig. 3C) show that the contact angle of TEUG is 57.6° ± 2.0°, which is not significantly different from that of native tissue (55.6° ± 2.1°), indicating comparable surface wettability and similar hydrophilic character. The moderate contact angles suggest favorable cell adhesion potential, likely due to the presence of hydrophilic polysaccharide components within the ECM-mimetic architecture, reflecting a good reproduction of the native tissue’s surface physicochemical properties by TEUG.

Cyclic tensile-relaxation testing (Fig. 3D) further compares the dynamic mechanical behavior of TEUG and native urethral tissue. Under 10 consecutive cycles at 25% strain, with stabilization over the first seven cycles and relaxation periods of 15 min (cycle 8), 20 min (cycle 9), and 30 min (cycle 10) applied at 37°C, the stress-strain hysteresis loops of TEUG gradually narrow and stabilize, with minimal residual strain, demonstrating good elastic recovery and mechanical integrity. Notably, TEUG maintains a certain level of stress responsiveness and exhibits a relatively rapid strain recovery trend after relaxation, this elastic performance is partially supported by the cross-linking between PG matrix and collagen/elastin networks, whereas the native urethral tissue shows moderate stress softening and a reduced hysteresis loop, suggesting viscoelastic relaxation and mild energy dissipation under repeated loading. Combined findings indicate TEUG matches native tissue in static biochemical/surface characteristics and dynamic viscoelastic properties.

To assess TEUG’s ability to regulate UEC behavior, New Zealand rabbit UECs were seeded on TEUG and commercial SIS scaffolds and monitored at 12, 48, 96, and 168 hours (Fig. 4A). SEM showed that on TEUG, cells adhered by 12 hours with early filopodial extensions, spread progressively between 48 and 96 hours, and formed a continuous, dense urothelium-like monolayer by 168 hours. In contrast, cells on SIS remained sparsely distributed throughout the culture period and failed to reach confluence even after 168 hours, indicating poor support for sustained proliferation and spreading. TEM analysis at 168 hours (Fig. 4B) revealed that UECs on TEUG exhibited well-defined nuclei, abundant mitochondria, microfilaments aligned along matrix fibers, and mature focal adhesions, consistent with stable integrin-mediated anchoring. On SIS, organelles were sparse, actin filaments lacked directional organization, and focal adhesions were weak or absent, reflecting inadequate cytoskeletal assembly.

AFM dissipation mapping at 168 hours (Fig. 4C) further confirmed these differences. UECs on TEUG displayed a broader dissipation signal range (0.90–4.60 mV), plump morphology, and continuous high-dissipation domains, suggesting robust membrane dynamics and effective mechanical coupling with the substrate. In contrast, cells on SIS showed weaker signals (1.41–3.81 mV) with fragmented distribution, indicative of unstable adhesion and limited interfacial activity. Line-scan profiles over a 3.0 µm distance at 168 hours (Fig. 4D) revealed higher peak amplitudes and smoother trends on TEUG, whereas SIS exhibited sharp fluctuations and frequent local spikes, consistent with heterogeneous cell-substrate contact and potential attachment instability.

Immunofluorescence staining at 168 hours (Fig. 4E and Supplementary Figures S3–S4) showed well-organized F-actin bundles and strong FAK enrichment at cell peripheries on TEUG, while FAK on SIS was weak and diffusely distributed. Quantitative analysis at 168 hours confirmed significantly higher FAK fluorescence intensity on TEUG (Fig. 4F), demonstrating its superior capacity to promote molecular assembly of functional adhesion complexes. Together, these results demonstrate that TEUG provides a biomimetic, bioactive interface that effectively supports UEC adhesion, spreading, cytoskeletal organization, and functional epithelial barrier formation, substantially outperforming commercial SIS scaffolds.

Figure 5A shows that the suture retention strength of TEUG was (2.0 ± 0.7) N, which is not significantly different from the tensile strength of natural urethral tissue ((1.5 ± 0.4) N; ns), indicating that its mechanical properties are within the normal physiological range of the urethra. In contrast, the suture strength of commercial SIS was (2.7 ± 1.2) N, significantly higher than both TEUG and natural urethral tissue (*p < 0.05), suggesting a potential risk of postoperative stress concentration. The intraoperative suture images shown in Fig. 5B indicate that TEUG exhibits good flexibility and can be uniformly sutured without significant tissue deformation. Conversely, due to its greater rigidity, SIS tends to cause local wrinkling and pulling deformations during suturing.

Postoperative one-month CT imaging revealed that, under the filled state, the TEUG bridging segment could maintain a stable hollow tubular structure. When emptied, it transformed into a semi-closed elliptical shape, demonstrating excellent compliance. Calcium imaging detected regular calcium signal fluctuations in the TEUG-internalized urethral smooth muscle cells during the filling-emptying transition, indicating the regeneration of functional smooth muscle. Electrical stimulation experiments further confirmed the excitability of the TEUG region, achieving successful capture and pacing responses at 2 Hz (Fig. 5D). In contrast, the SIS bridging segment failed to form a regular hollow tube when filled and presented multiple folds and a semi-closed shape when emptied. Its calcium signals were weak and disorganized, with significantly lower electrical stimulation response amplitudes compared to TEUG. Three months postoperatively, urethral pressure measurements showed that the TEUG group produced continuous and stable rhythmic contractions under 50 Hz and 5000 ms electrical stimulation, maintaining intraluminal pressures between 30–60 mmHg (Fig. 5E). The SIS group, however, exhibited nearly zero intraluminal pressure with no significant pressure fluctuations. Six months postoperatively, urethrography showed that the TEUG bridging segment remained fully patent, allowing contrast agent to pass through smoothly without retention. However, the SIS bridging segment developed stenosis near the proximal end, leading to contrast agent retention and difficulty in drainage. Quantitative analysis indicated that the average diameter of the repaired urethra in the TEUG group was (2.5 ± 0.7) mm, significantly larger than the (1.7 ± 0.8) mm observed in the SIS group (*p < 0.05). In summary, TEUG demonstrates superior biocompatibility, tissue integration capability, and physiological function reconstruction potential in urethral defect repair compared to commercial SIS.

Histological evaluation of urethral regeneration bridged by TEUG and SIS at 1-, 3-, and 6-months post-implantation showed that TEUG was significantly superior to SIS in inflammation control, tissue structure reconstruction, and functional component regeneration (Fig. 6A). H&E staining results revealed that the TEUG group exhibited a mild inflammatory response throughout the observation period, with no obvious foreign body giant cell formation and FCD, and maintained a patent lumen without stenosis or obstruction. In contrast, the SIS group showed persistent inflammatory infiltration and interstitial fibrosis over time, which progressed to significant luminal stenosis in the later stage. MTC staining demonstrated that the TEUG group had orderly and uniformly distributed collagen fiber deposition, and smooth muscle tissue gradually regenerated over time; a continuous circular muscle layer structure was formed by 6 months post-operation. By comparison, the SIS group had sparse and disorderly arranged smooth muscle cells, with dense and disorganized collagen fiber distribution, and lacked the formation of a mature muscle layer. EVG staining further indicated that elastic fibers in the TEUG group began to regenerate in the early stage, and a continuous and regular elastic fiber network was formed by 6 months. However, almost no elastic fiber regeneration was observed in the SIS group, with a lack of elastic structure.

Immunofluorescence staining for CK20 (epithelial marker) and α-SMA (smooth muscle marker) was performed to further evaluate epithelialization and smooth muscle regeneration. The results showed that in the TEUG group, CK20-positive epithelial cells were observed covering the surface at 1-month post-implantation; a complete and continuous urethral epithelial layer was formed at 3 months. Meanwhile, the number of α-SMA-positive smooth muscle cells increased significantly and showed a tendency of circular arrangement (Fig. 6B and S5), indicating the ordered reconstruction of a functional muscle layer. On the contrary, the SIS group showed discontinuous and incomplete CK20-positive epithelial layers at all time points, weak α-SMA expression and scattered distribution of smooth muscle cells, failing to form mature muscle bundles (Fig. 6B and S6). Quantitative analysis results in Figs. 6C and 6D further confirmed that the thickness of the CK20-positive epithelial layer and α-SMA-positive smooth muscle layer in the TEUG group was significantly higher than that in the SIS group at 1, 3, and 6 months (*p < 0.05, ** p < 0.01, *** p < 0.001). Both indicators showed a consistent upward trend over time, and the TEUG group maintained good regenerative capacity even at 6 months. In contrast, the SIS group had a thin epithelial layer and underdeveloped smooth muscle layer, with a slow regeneration process, which was significantly lower than the TEUG group at all time points. These findings indicate that compared with SIS, TEUG can more effectively maintain the urethral structure and expandability during urethral defect repair, resist FBR, promote tissue integration and functional reconstruction, and ultimately achieve unobstructed urination.

At 6 months post-implantation, immunofluorescence staining for CD206 (a marker of M2 macrophages) and CD86 (a marker of M1 macrophages) was performed to evaluate the local immune response. In the TEUG group, CD206⁺ macrophages were circumferentially distributed around the epithelial and smooth muscle layers, with minimal CD86⁺ cells, indicating an anti-inflammatory and pro-repair immune microenvironment. In contrast, the SIS group exhibited extensive CD86⁺ macrophage infiltration, weak CD206⁺ signals, and cellular aggregation within the narrowed lumen, suggesting persistent inflammation and foreign body reaction. Quantitative analysis (Fig. 7B) showed that the CD206⁺/CD86⁺ ratio in the TEUG group (2.0 ± 0.8) was significantly higher than that in the SIS group (1.0 ± 0.5), further confirming a pro-regenerative immune bias.

To elucidate the underlying immunomodulatory mechanisms, gene expression analysis was performed on CD45⁺ immune cells isolated from the anastomotic sites (Fig. 7C). In the TEUG group, expression of pro-regenerative markers CD206 and CD4 (T helper cell marker) was upregulated, while pro-inflammatory and antigen-presenting molecules CD86 and CD27 were downregulated. The SIS group displayed the opposite trend, indicating activation of inflammatory and immune-rejection pathways. Consistently, ELISA detection of Arg-1 (a key functional marker of M2 macrophages) and iNOS (a signature pro-inflammatory marker of M1 macrophages) in the harvested grafts further confirmed these findings, with significantly higher Arg-1 levels and lower iNOS expression in the TEUG group compared to the SIS group.

Ultrastructural analysis (Fig. 7E) revealed that the luminal surface of the TEUG-reconstructed urethra was lined with intact umbrella cells exhibiting a flat or polygonal morphology, with cell bodies protruding into the lumen to form a "barrier layer." The cytoplasm was rich and contained numerous pinocytic vesicles. Short, densely packed microvilli (diameter: 0.1–0.2 µm, length: 0.5–1 µm) were observed on the apical surface, and nuclei displayed clear nuclear membranes and prominent nucleoli—features characteristic of mature urothelium with effective urine barrier function and mechanical adaptability. In contrast, SIS-reconstructed urethras exhibited cell membrane rupture, cytoplasmic vacuolization, and inflammasome formation, indicating severe structural damage.

PCNA immunofluorescence and TUNEL assays revealed that proliferating cells (PCNA⁺) in the TEUG group were evenly distributed across tissue layers, while apoptotic cells (TUNEL⁺) were sporadically present, indicating active yet balanced tissue remodeling. In the SIS group, PCNA⁺ signals were weak and disorganized, whereas TUNEL⁺ signals were intense and concentrated along fibrotic regions (Fig. 7F). Suggesting impaired cellular turnover and tissue degeneration. Quantitative analysis (Fig. 7G) showed a PCNA⁺/TUNEL⁺ ratio of 4.3 ± 3.8 in the TEUG group, significantly higher than the 0.4 ± 0.3 observed in the SIS group, highlighting its superior regenerative capacity and ability to maintain tissue homeostasis. These results confirm that TEUG has better bio-integration and regenerative efficacy than commercial SIS-based grafts.