An Aligned-to-Random PLGA/Col1-PLGA/nHA Bilayer Electrospun Nanofiber Membrane Enhances Tendon-to-Bone Healing in a Murine Model

Abstract

Background:

The challenge of achieving effective tendon-to-bone healing remains a significant concern in sports medicine, necessitating further exploration. Biomimetic electrospun nanomaterials present promising avenues for improving this critical healing process.

Purpose:

To investigate the biological efficacy of a novel aligned-to-random PLGA/Col1-PLGA/nHA bilayer electrospun nanofiber membrane in facilitating tendon-to-bone healing.

Study Design:

Controlled laboratory study.

Methods:

The bilayer membrane’s composition, combining PLGA/Col1 for tendon attachment and PLGA/nHA for bone integration, was examined using scanning electron microscopy, Fourier transform infrared spectroscopy, and mechanical testing. Positioned between the Achilles tendon and bone, its design aimed for harmonious integration with both types of tissue. In vitro, biocompatibility, cell adhesion, and proliferation of the biomaterial were evaluated using live/dead staining and the CCK-8 assay. Collagen secretion and mineralization were measured for 2 cell types. In vivo, tendon-to-bone insertion samples harvested from mice were analyzed: micro–computed tomography assessed bone formation; histological staining evaluated chondrogenesis, tendinogenesis, and the 4-layer structure of the insertion; and biomechanical testing measured insertion strength. Real-time polymerase chain reaction identified genes involved in tendon-to-bone healing, and transcriptome analysis elucidated the underlying cellular and molecular mechanisms.

Results:

The optimal composition was determined as 10% 3:1 for aligned PLGA/Col1 and 9% 5:1 for PLGA/nHA. Coculture showed minimal cell death, firm cell adherence, and steady proliferation, with PLGA/Col1 enhancing collagen secretion. In vivo, the material promoted bone and cartilage formation and improved tendon-to-bone interface strength. Transcriptome analysis indicated links to TNF and NF-κB pathways and to genes IL-1β, ADAM8, and EGR2.

Conclusion:

The novel aligned-to-random PLGA/Col1-PLGA/nHA bilayer nanofiber membrane outperformed other materials in both in vitro and in vivo evaluations, significantly enhancing tendon-to-bone healing. It notably improved cartilage and bone formation, tendon maturation, and biomechanical strength at the surgical interface. These effects may be associated with the TNF and NF-κB pathways and with the genes IL-1β, ADAM8, and EGR2.

Clinical Relevance:

This study introduces a biomimetic nanofiber membrane enhancing tendon-to-bone healing, which is crucial for sports medicine. Its efficacy in improving healing outcomes, including bone and cartilage formation and biomechanical strength, could significantly lower failure rates in surgical procedures such as rotator cuff repair and anterior cruciate ligament reconstruction. This advancement offers promising implications for patient recovery and the effectiveness of surgical interventions in tendon-to-bone injuries.

Keywords

aligned PLGA/Col1-PLGA/nHA PLGA collagen nHA tendon-to-bone healing

There is a special structure that lies between bone and the ligament or tendon known as the tendon-to-bone interface or enthesis. This structure is composed of 4 continuous layers: tendon, nonmineralized fibrocartilage, mineralized fibrocartilage, and bone.^1,2 These distinctions in composition smoothe the mechanical stress distribution curve and enhance the bonding strength between soft and hard tissue.^4,12 Typical anatomic structures featuring this interface include the rotator cuff, anterior cruciate ligament, patellar tendon, and Achilles tendon. Surgery is often necessary after unsatisfactory nonoperative treatment. However, the failure rate for small rotator cuff repair ranges from 20% to 40% and can even reach as high as 94% for massive rotator cuff repair.^5,6 The typical failure rate for anterior cruciate ligament reconstruction is 10% to 25%.¹⁹ Cadaveric studies have revealed that the original structure is replaced by scar tissue, and the strength of the tendon-to-bone interface is significantly lower than that of the natural tendon-to-bone interface.¹⁸

Strengthening the connection between the tendon and bone is of utmost importance. To address these challenges, both synthetic and natural biomaterials have been introduced and developed as patches and membranes to enhance tendon-to-bone healing. Among the various materials available, poly(lactic-co-glycolic) acid (PLGA) stands out as one of the most widely used biodegradable polymers, known for its exceptional biocompatibility, biodegradability, mechanical properties, and minimal inflammatory response within the body.¹³ Collagen type I (Col1), the primary component of the extracellular matrix in tendons, is a natural polymer with the ability to interact with numerous growth factors, thereby promoting cell growth, migration, and differentiation. Col1 also enhances the biocompatibility of polymers.²⁰ Nanohydroxyapatite (nHA) is a common choice in bone engineering because its physicochemical properties closely resemble those of natural nanocrystals in bone tissue, and numerous studies have confirmed its efficacy in bone regeneration.²⁴ A combination of these materials can mimic the natural extracellular matrix of the tendon-to-bone interface, fostering growth at both the tendon and bone ends.

Electrospinning techniques have been employed to fabricate scaffolds for artificial grafts. This method offers a viable approach to combining biomaterials and creating a nanofiber membrane that has been proven to exhibit excellent biocompatibility, high aspect ratios, and high porosity with small pore sizes.^9,23 These unique attributes enable nanofibers to mimic the hierarchical architecture of the extracellular matrix, which is crucial for cell adhesion and nutrient transport. Xie et al²¹ designed aligned nanofibers that run parallel to the long axis of tendon fibers. They observed that fibroblasts cultured on this scaffold displayed highly organized morphologies reminiscent of tendon fibers. Building on these findings, we devised an aligned PLGA/Col1-PLGA/nHA bilayer electrospun nanofiber membrane. In our design, the aligned portion is positioned at the tendon end, while the PLGA/nHA portion is placed at the bone end. This arrangement replicates the fiber structure from the tendon to bone, aiming to facilitate the transition of healing mechanisms from tendon-to-bone healing to both bone-bone and tendon-tendon healing. Our hypothesis was that this aligned PLGA/Col1-PLGA/nHA bilayer electrospun nanofiber membrane would enhance tendon-to-bone healing and ultimately improve the biomechanical properties of the tendon-to-bone interface after surgery.

Methods

Materials and Agents

PLGA with a lactide:glycolide ratio of 85:15 and a molecular weight in the range of 190,000 to 240,000 g/mol was purchased from Aladdin. Hexafluoroisopropanol and nHA were also procured from Aladdin. Col1 was obtained from Macklin.

Fabrication and Characterization of Biomaterials

Aligned PLGA/Col1 and PLGA/nHA nanofiber membranes with different mass fractions and ratios were fabricated using an electrospinning machine. The characteristics of these biomaterials were analyzed using scanning electron microscopy (SEM), energy-dispersive spectroscopy (EDS), Fourier transform infrared spectroscopy (FTIR), and mechanical testing. Specific parameters for each device are provided in the Appendix (available in the online version of this article).

In Vitro Biocompatibility Experiments

Osteoblast and Fibroblast Culture

Murine NIH3T3 fibroblast cell lines and MC3T3 osteoblast cell lines, along with complete medium (DMEM + 10% calf serum + 1% penicillin/streptomycin), were obtained from Procell Life Science & Technology. A total of 4 nanofiber membranes—aligned PLGA/Col1, PLGA, random PLGA/Col1, and PLGA/nHA—were cut into 24-well plates with a circular pore size. Afterward, they were soaked in 75% ethanol at room temperature for 30 minutes. Subsequently, ethanol was discarded, and the membranes were left to dry at room temperature for 1.5 hours. To enhance hydrophilicity, a plasma cleaner (Zepto ONE; Diener) was utilized for 1 minute. After 40 minutes of ultraviolet sterilization, the material was fixed at the bottom of the 24-well plate using a sterile circular stainless-steel gasket of an appropriate size. The entire plate was then sterilized by ultraviolet light again for 40 minutes. Subsequently, 2 types of cell lines were seeded on the 24-well plate (10⁴ cells/well) with complete medium.

Cell Viability and Adhesion

NIH3T3 cells were seeded on PLGA, aligned PLGA/Col1, and random PLGA/Col1 nanofiber membranes, while MC3T3 cells were seeded on PLGA/nHA nanofiber membranes in 24-well plates. A blank group served as the control. Each group had 3 replicates. After culturing for 1 and 5 days, respectively, live and dead cells were examined by a live/dead cell double staining kit (Solarbio). In brief, each well was added with 400 uL of a prepared dye solution and incubated for 15 minutes. Then, the solution was discarded and washed twice with phosphate-buffered saline for 5 minutes each time. The material was placed in a confocal dish for live/dead staining and cell adhesion observation under a confocal microscope (LSM900; ZEISS).

Cell proliferation was assessed using a cell counting kit–8 (CCK-8) assay kit (Solarbio). In brief, NIH3T3 and MC3T3 cell suspensions were seeded on 96-well plates, and the extracts of aligned PLGA/Col1, PLGA, random PLGA/Col1, and PLGA/nHA were added, respectively. After 24, 48, and 72 hours of culture, 10 uL of CCK-8 solution was added to each well, and incubation occurred at 37°C with 5% CO₂ for 1.5 hours. Then, the absorbance was measured at 450 nm by a microplate reader (Victor X; PerkinElmer).

Collagen Secretion

To assess the potential of 3 materials (aligned PLGA/Col1, PLGA, random PLGA/Col1) in promoting collagen secretion by NIH3T3 cells, a hydroxyproline content assay kit (Solarbio) was employed to measure the collagen content in the cells. In summary, NIH3T3 cells were seeded on the 3 materials. After culturing for 5 days, cells were collected after trypsinization. Next, 1 mL of the extract was added, boiled for complete digestion, cooled with 10 mol/L sodium hydroxide to adjust the pH to a range of 6 to 8, and then diluted with distilled water to 2 mL. The mixture was centrifuged at 16,000 rpm and 25°C for 20 minutes, and the supernatant was collected. The spectrophotometer was used at 560 nm to quantify the optical density. Finally, the hydroxyproline content was calculated using the formula provided in the instructions.

Mineralization

To evaluate the potential of PLGA/nHA in promoting alkaline phosphatase (ALP) activity by MC3T3 cells, an ALP activity assay kit (Bestbio) was employed to determine ALP activity. In summary, MC3T3 cells were seeded on the PLGA/nHA nanofiber membrane, and a blank group was set as the control. After culturing for 5 days, cells were collected after trypsinization. The cells were then lysed using cell lysates, and the solution was added according to the instructions provided by the company. The spectrophotometer was set at 510 nm to measure the optical density. Finally, ALP activity was calculated according to the manufacturer’s specifications.

In Vivo Animal Experiments

Study Design

The animal experiment was approved by our university (No. AMUWEC20210782) and carried out in accordance with the ARRIVE guidelines. A total of 138 nine-week-old male C57BL/6 mice (Hunan SJA Laboratory) were obtained. For each time point (4, 8, and 12 weeks postoperatively), 43 mice were randomly allocated. They were further assigned to 5 groups: control, model, PLGA, aligned PLGA/Col1-PLGA/nHA, and random PLGA/Col1-PLGA/nHA. Specifically, 17 mice had their left hindfoot assigned to the control group and their right hindfoot assigned to the model group. Another 17 mice had their left hindfoot assigned to the PLGA group and their right hindfoot assigned to the aligned PLGA/Col1-PLGA/nHA group. Moreover, 9 mice had their hindfeet assigned to the random PLGA/Col1-PLGA/nHA group. An additional 9 mice from both the model group and the aligned PLGA/Col1-PLGA/nHA group were selected in the first month after surgery for transcriptome analysis. At 1, 2, and 3 months after surgery, mice were sacrificed for histological staining, micro–computed tomography (micro-CT), biomechanical testing, and real-time polymerase chain reaction (PCR) (Figure 1).

Surgical Procedure

Surgical procedures were performed according to the previous methods that we specially developed for investigating tendon-to-bone healing.²² In brief, the mice were given anesthesia with isoflurane and placed on their stomachs on a foam operating platform. After cleaning the hindfoot, a 2-mm incision was made about 2 mm above the calcaneus. First, tissue was carefully separated to expose the Achilles tendon and calcaneus. Second, a transverse bone tunnel was made in the calcaneus using a 30-gauge insulin needle. Then, a 6-0 needle with absorbable thread was passed through the bone tunnel and through the Achilles tendon, creating a stitch. The thread was looped around the tendon and stitched in place. Third, the Achilles tendon was cut near its attachment using a No. 11 blade, and cartilage in the tendon-to-bone area was gently removed. In the model group, nothing was inserted between the Achilles tendon and bone. In the nanofiber membrane groups (aligned PLGA/Col1-PLGA/nHA, PLGA, and random PLGA/Col1-PLGA/nHA), 3 × 1–mm nanofiber membranes were placed between the Achilles tendon and bone. Specifically, the end of PLGA/Col1 was attached to the Achilles tendon, with tendon fibers running parallel to the nanofibers. The end of PLGA/nHA was attached to the bone ends, and the orientation of PLGA had no specific requirements. In the control group, the tendon-to-bone interface retained its natural structure. Then, the thread ends were tied and tightened. Finally, the skin was closed with 1 to 2 stitches. After surgery, the mice were allowed to move freely in their cages, with access to water and food throughout the experiment (Figure 2).

Micro-CT Analysis

At 4, 8, and 12 weeks after surgery, mice were euthanized, and 15 hindfeet (3 from each group) were collected to observe bone formation in the tunnel. The Skyscan 1272 system (Bruker) with a voxel size of 10.0 µm was employed for micro-CT analysis. Imaging was performed in 4% paraformaldehyde with a tube potential of 60 kV, an X-ray intensity of 166 µA, and an exposure time of 1700 milliseconds. Reconstruction was performed using NRecon software (Version 1.6.10), and 3-dimensional images were generated from 2-dimensional images using CTVox software (Version 3.0.0) based on distance transformation of the grayscale originals. The bone-related parameters assessed in this experiment included bone mineral density (BMD; g/cm³), bone volume (BV; mm³), BV/tissue volume (BV/TV; %), bone surface (BS; mm²), BS/BV (1/mm), and BS/TV (1/mm). For consistency, a cylindrical region of interest was established, and the total volume of each sample was standardized to 0.0396 mm³.

Histomorphological Analysis

At 4, 8, and 12 weeks after surgery, mice were euthanized, and 15 hindfeet (3 from each group) were obtained. After removing the skin, the calcaneus and a portion of the Achilles tendon were harvested and preserved in 4% paraformaldehyde for 24 hours. After preservation, the specimens underwent decalcification using an EDTA decalcifying solution for 72 hours in a thermostatic shaker at 37°C. Subsequently, the specimens were embedded in paraffin and sliced sagittally. Various staining methods, including hematoxylin and eosin (H&E), safranin O/Fast green, and Alcian blue (Solarbio), were employed to examine the tendon-to-bone interface. H&E staining assessed the overall condition of the interface, including new tissue formation and cell morphology. Safranin O/Fast green and Alcian blue staining were applied to identify glycosaminoglycan and observe cartilage formation. Additionally, a modified tendon maturity assessment system,¹⁵ comprising 8 items with a total possible score of 32, was utilized to evaluate tendon maturity. This system included the following parameters: cellularity, proportion of fibers oriented parallel, proportion of fibers of a large diameter, vascularity, continuity, bone ingrowth, fibrocartilage cells, and tidemark. In this assessment, a higher score reflected greater maturity in tendon-to-bone healing.

Biomechanical Analysis

After euthanizing the mice, 30 hindfeet were obtained (6 from each group), and only the entire feet and tendons were collected at the 4-, 8-, and 12-week postoperative time points. ElectroForce mechanical test instruments (TA Instruments) were used to assess ultimate tensile strength (MPa), stiffness (N/mm), and the elastic modulus (MPa). The specimens were wrapped with gauze to enhance fixation friction and then secured on the tensile machine. After calibration, a 0.05-N preload was applied for 1 minute, followed by stretching the samples at a speed of 0.03 mm/s. Data related to terminal slip in tendons and fractures unrelated to tendon-to-bone insertion connections were excluded from analysis.

Real-Time PCR and Transcriptome Analysis

At each postoperative time point, 25 hindfeet were collected (5 from each group) to obtain tendon-to-bone interface samples, including small portions of both bone and the tendon. These specimens were sectioned, and total RNA was extracted using TRIzol reagent (Thermo Fisher Scientific). The RNA concentration was measured via spectrophotometry and normalized to the expression of the housekeeping gene GAPDH. Primer sequences for BMP2, Col1, SOX9, collagen type II, collagen type III, TGF-β1, TGF-β3, and GAPDH are listed in Table 1.

Table 1

Primer Sequences

Gene	Forward	Reverse
BMP2	tgc ttc tta gac gga ctg cg	ctg ggg aag cag caa cac ta
Col1a1	agc acg tct ggt ttg gag ag	gac att agg cgc agg aag gt
Col2a1	gcc agg atg ccc gaa aat ta	Tgg gtc ctt gtt cac ctg c
Col3a1	gcg agc ggc tga gtt tta tg	gca gct cag agt agc acc at
SOX9	gtg caa gct ggc aaa gtt ga	tgc tca gtt cac cga tgt cc
TGF-β1	gct gaa cca agg aga cgg aa	ggg ctg atc ccg ttg att tc
TGF-β3	gga aat ggg tcc acg aac ct	ccg tgc tat ggg ttg tgt ct
GAPDH	GGT TGT CTC CTG CGA CTT CA	TGG TCC AGG GTT TCT TAC TCC

For transcriptome analysis, RNA sequencing was conducted by LC-Bio Technologies. Libraries were prepared using the NEBNext Ultra RNA Library Prep Kit (E7490; New England Biolabs) on the NovaSeq 6000 PE150 system (Illumina) and purified with AMPure XP magnetic beads (Beckman Coulter). Cutadapt software (Version 1.9) was used with default parameters to remove adapters and low-quality reads. The resulting FASTQ files were aligned using the HISAT2 package (Version 2.2.1), and gene abundance was quantified with StringTie (Version 2.1.6) to determine transcript expression levels by calculating FPKM (fragments per kilobase of transcript per million mapped reads) values.

Differential gene expression analysis was performed using DESeq2, while comparisons between 2 samples were made utilizing edgeR. Genes with a false discovery rate <0.05 and an absolute fold change ≥2 were considered differentially expressed. Enrichment analysis for Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways was then conducted for the differentially expressed genes.

Statistical Analysis

The independent-samples t test and 1-way or 2-way analysis of variance were employed for comparisons between 2 groups and among multiple groups, respectively. Model assumptions were assessed using the Shapiro-Wilk test of normality and the Levene test for the homogeneity of variance, along with visual inspection of residual and fitted value plots. Post hoc comparisons after analysis of variance were conducted using the Tukey multiple comparison test. Statistical significance was defined as P < .05.

Results

Characteristics of Nanofiber Membranes

Figure 3A displays spectra at 1590 cm^–1 and 1638 cm^–1 for PLGA/Col1, revealing 2 characteristic stretching bands of Col1: namely, an amide I band and an amide II band. These findings support the incorporation of PLGA and Col1, as illustrated in Figure 3A(a). Additionally, evidence of nHA incorporation with PLGA is presented in Figure 3A(b). In the FTIR spectra at 866 cm^–1, 980 cm^–1, and 1019 cm^–1 for PLGA/nHA, stretching vibrations corresponding to carbonyl and phosphate groups were detected,³ validating the combination of PLGA and nHA.

Various mass fractions of aligned PLGA/Col1 and PLGA/nHA nanofiber membranes were examined with SEM. Aligned PLGA/Col1 nanofibers with 6% (432.5 ± 13.4 nm) (Figure 3B(a)), 8% (433.5 ± 55.9 nm) (Figure 3B(c)), and 10% (854.0 ± 248.9 nm) (Figure 3B(e)) mass fractions were observed, alongside PLGA/nHA nanofibers with 7% (383.5 ± 58.7 nm) (Figure 3B(g)), 9% (513.0 ± 55.2 nm) (Figure 3B(i)), and 11% (1017.0 ± 487.9 nm) (Figure 3B(k)). Aligned PLGA/Col1 nanofibers exhibited distinct directionality and alignment, while PLGA/nHA nanofibers displayed a random arrangement with an observed aggregation of nHA (Figure 3B). Additionally, chemical composition analysis using EDS revealed the presence of nitrogen, calcium, and phosphate elements in aligned PLGA/Col1 and PLGA/nHA nanofibers (Figure 3C).

The tensile properties for each nanofiber membrane are illustrated in Figure 3D. In the aligned PLGA/Col1 group, 3 mass fractions exhibited no significant differences in the elastic modulus. However, aligned PLGA/Col1 at 8% and 10% demonstrated a significantly higher maximum force and ultimate tensile strength compared with aligned PLGA/Col1 at 6%. While no significant difference was observed between 8% and 10%, the biomechanical performance of 10% was superior to 8% (Figure 3D(b-d)). Therefore, the mass fraction of 10% was selected for aligned PLGA/Col1. Within the PLGA/nHA group, 3 mass fractions showed no significant differences in the elastic modulus, maximum force, and ultimate tensile strength. However, PLGA/nHA at 9% exhibited greater ductility (Figure 3D(e)). Consequently, the mass fraction of 9% was chosen for PLGA/nHA.

After establishing the mass fractions for aligned PLGA/Col1 and PLGA/nHA, we explored different ratios of PLGA and Col1 as well as PLGA and nHA. SEM results in Appendix Figure A1A and A1B (available online) depict aligned PLGA/Col1 10% and PLGA/nHA 9% under various ratios. Notably, different ratios of aligned PLGA/Col1 10% consistently exhibited clear directionality and alignment, while the nanofibers of PLGA/nHA 9% remained randomly arranged, with evident nHA deposition. These findings, consistent with the earlier EDS results, further validate the successful combination of PLGA with Col1 and PLGA with nHA, as illustrated in Appendix Figure A1C (available online). In tensile performance testing (Appendix Figure A1D [available online]), varying ratios of aligned PLGA/Col1 and PLGA/nHA revealed no significant differences in the elastic modulus, maximum force, and ultimate tensile strength. However, the specific ratio of aligned PLGA/Col1 10% at 3:1 displayed superior results compared with other ratios (Appendix Figure A1D(b-d) [available online]). Similarly, PLGA/nHA 9% at 5:1 exhibited greater ductility (Appendix Figure A1D(e) [available online]). Consequently, these optimal configurations—aligned PLGA/Col1 10% at 3:1 and PLGA/nHA 9% at 5:1—were chosen as the final mass fractions and ratios for the subsequent experiments.

Figure 3E presents co-electrospinning images of aligned PLGA/Col1 10% 3:1 and PLGA/nHA 9% 5:1 using SEM. The cross-sectional SEM images revealed a 2-layered structure (Figure 3E(a, b)). The lower layer exhibited distinct directionality and alignment (Figure 3E(d)), while the upper layer contained nHA particles (Figure 3E(c)). EDS results further confirm the successful incorporation of PLGA and Col1 (Figure 3E(f)) in the lower layer and PLGA and nHA (Figure 3E(e)) in the upper layer.

Biocompatibility of Fabricated Nanofiber Membranes

Fluorescence-based live/dead staining conducted with confocal microscopy at 1 day of culture revealed a high proportion of cells that died on the PLGA nanofiber membranes, while others showed minimal cell death (Figure 4A). By the fifth day of culture, all materials exhibited few to no dead cells. Additionally, the alignment of aligned PLGA/Col1 nanofibers became apparent (Figure 4B(a, b)), and the NIH3T3 cells started aligning themselves with the fiber direction (Figure 4B(b), white arrow). Overall, 3-dimensional images were captured to illustrate the relationship between materials and cells. In the depth coding mode of confocal microscopy, the hierarchical structure of cells and materials, represented by different colors, was observed at both 1 and 5 days. The colors of cells corresponded with the materials, indicating robust cell growth on the materials over time.

CCK-8 assay demonstrated that in the NIH3T3 cell group, all types of materials inhibited cell proliferation speed to some extent compared with the blank group, but cell numbers increased over time. Aligned PLGA/Col1 was significantly superior to PLGA and random PLGA/Col1 nanofibers after 72 hours of culture (Figure 4C(a)). In the MC3T3 cell group, PLGA/nHA nanofibers significantly promoted cell growth compared with the blank group after 48 and 72 hours of culture (Figure 4C(b)). In summary, these materials provided a suitable environment for cell proliferation and adhesion over time, with aligned PLGA/Col1 nanofiber membranes showing significantly better results than the other 2 materials.

Collagen Secretion and Mineralization

On the fifth day of NIH3T3 cell culture, the aligned PLGA/Col1 group demonstrated significantly higher secretion of Col1 compared with the PLGA group. Moreover, there was no significant difference between the blank group and the aligned PLGA/Col1 group (Figure 4D(a)). In the MC3T3 cell group, the PLGA/nHA group exhibited a significantly higher ALP activity level compared with the blank group (Figure 4D(b)). These results suggest the superiority of aligned PLGA/Col1 in promoting fibroblast function compared with the other 2 materials, and the PLGA/nHA group demonstrated better performance than the blank group in promoting osteoblast function.

Histological Analysis

After harvesting at various time points, the Achilles tendons exhibited different degrees of thickening, with a rough and lusterless surface. At 1 month postoperatively, H&E staining revealed a disordered structure at the tendon-to-bone interface, marked loss of the normal physiological structure, and the infiltration of inflammatory cells. Safranin O/Fast green staining indicated a reduced tendon volume with a notable accumulation of cartilage near the bone end, suggesting potential endochondral ossification. Alcian blue staining showed similar results to Safranin O/Fast green staining.

By 2 months postoperatively, H&E staining showed improvement in all groups. The tendon-to-bone interface became clearer, and tendon quality and orientation started to emerge. The aligned PLGA/Col1-PLGA/nHA group displayed performance closest to that of the control group. Safranin O/Fast green staining revealed the best tendon quality and orientation in the aligned PLGA/Col1-PLGA/nHA and PLGA groups. Although tendon fibers increased in the model and random PLGA/Col1-PLGA/nHA groups, the structure remained disordered. Regarding cartilage formation, both safranin O/Fast green and Alcian blue staining indicated obvious cartilage formation at the tendon-to-bone interface in the PLGA and aligned PLGA/Col1-PLGA/nHA groups, with the latter closer to the control group. The model group and random PLGA/Col1-PLGA/nHA group lacked clear cartilage formation at the interface, and the latter showed abnormal cartilage cell aggregation.

At 3 months postoperatively, H&E staining demonstrated further improvement in all groups, with a significant increase in tendon tissue. The tendon-to-bone interface structure in the aligned PLGA/Col1-PLGA/nHA, random PLGA/Col1-PLGA/nHA, and model groups was more complete. Safranin O/Fast green staining indicated maturity in the aligned PLGA/Col1-PLGA/nHA group, closely resembling the normal physiological structure, while the other groups showed enhanced tendon quality and quantity. Regarding cartilage formation, safranin O/Fast green and Alcian blue staining revealed obvious cartilage formation in the aligned PLGA/Col1-PLGA/nHA group, similar to the natural structure, and the random PLGA/Col1-PLGA/nHA group also exhibited noticeable cartilage formation.

Furthermore, we used the modified tendon maturity score⁸ to evaluate maturity. In the first 2 months after surgery, all groups showed comparable performance with a disordered tendon-to-bone interface. By 3 months after surgery, the aligned PLGA/Col1-PLGA/nHA group demonstrated significant superiority over the other groups. In summary, in the first 2 months after surgery, each group exhibited comparable performance with a disordered tendon-to-bone interface. However, in the third month after surgery, the aligned PLGA/Col1-PLGA/nHA group exhibited better tendon maturity and cartilage-forming ability (Figure 5).

Micro-CT Analysis

Figure 6 shows the 3-dimensional reconstruction of the calcaneus across different groups and time points. Complete healing of the bone tunnel in the calcaneus was observed in the aligned PLGA/Col1-PLGA/nHA group by 1 month, in both the PLGA group and the random PLGA/Col1-PLGA/nHA group by 2 months, and in the model group by 3 months. All groups exhibited heterotopic ossification formation around the surgical sites, with the aligned PLGA/Col1-PLGA/nHA group showing less heterotopic ossification formation compared with the other groups. In summary, the aligned PLGA/Col1-PLGA/nHA group demonstrated stronger bone-forming ability with reduced heterotopic ossification formation. Table 2 presents quantitative data of bone formation, except for the control group representing the natural structure of the enthesis. In the first or second month after surgery, the aligned PLGA/Col1-PLGA/nHA group demonstrated significant superiority over the other groups in terms of BMD (compared with the model and PLGA groups), BV (in comparison to the other groups), BV/TV (compared with the other groups), BS (in comparison to the PLGA and random PLGA/Col1-PLGA/nHA groups), BS/BV (compared with the PLGA group), and BS/TV (compared with the PLGA and random PLGA/Col1-PLGA/nHA groups). However, in the third month after surgery, except for the control group, the groups exhibited no statistical significance in BS/TV, BS/BV, BS, and BMD. Only in BV and BV/TV did the aligned PLGA/Col1-PLGA/nHA group show a significant difference from the random PLGA/Col1-PLGA/nHA group. These results indicated that apart from the control group representing the natural enthesis structure, the aligned PLGA/Col1-PLGA/nHA group exhibited superior bone-forming ability compared with the other 3 groups in the first 2 months, while at the third time point, each group demonstrated comparable bone-forming ability.

Table 2

Micro–Computed Tomography Findings^a

	Control	Model	Aligned PLGA/ Col1-PLGA/nHA	Random PLGA/ Col1-PLGA/nHA	PLGA
BMD, g/cm³
1 mo	1.294 ± 0.113^{b , c}	0.809 ± 0.062	0.612 ± 0.241	0.493 ± 0.116^d	0.800 ± 0.182
2 mo	1.129 ± 0.048^c	0.551 ± 0.041^e	0.920 ± 0.023^f	0.788 ± 0.248	0.481 ± 0.104^g
3 mo	1.313 ± 0.017^{h , i}	0.849 ± 0.035	0.977 ± 0.052	0.868 ± 0.057	0.814 ± 0.016
BV, mm³
1 mo	0.032 ± 0.001	0.022 ± 0.004	0.027 ± 0.010	0.014 ± 0.004^h	0.019 ± 0.004
2 mo	0.036 ± 0.002ⁱ	0.017 ± 0.001^e	0.032 ± 0.004^f	0.021 ± 0.008^h	0.012 ± 0.005^g
3 mo	0.037 ± 0.000	0.029 ± 0.005	0.033 ± 0.005	0.019 ± 0.007^h	0.034 ± 0.002
BV/TV, %
1 mo	79.883 ± 2.460	54.254 ± 9.485	69.143 ± 25.852	34.975 ± 9.822^h	46.666 ± 11.091
2 mo	91.702 ± 4.870ⁱ	43.580 ± 1.863^e	81.985 ± 10.342^f	52.282 ± 19.204^h	29.555 ± 12.012^e
3 mo	93.683 ± 0.709	73.536 ± 12.243	82.576 ± 13.006	48.369 ± 18.565^h	86.552 ± 5.647
BS, mm²
1 mo	0.870 ± 0.082^g	2.015 ± 1.218	2.926 ± 0.608	0.938 ± 0.113^g	0.855 ± 0.196^{d , g}
2 mo	0.987 ± 0.038^h	1.136 ± 0.147	2.287 ± 0.743	1.462 ± 0.386	1.248 ± 0.626
3 mo	1.169 ± 0.095	1.524 ± 0.324	1.830 ± 0.604	1.252 ± 0.207	1.384 ± 0.213
BS/BV, 1/mm
1 mo	27.531 ± 3.478^e	89.529 ± 40.395	128.747 ± 88.015	69.900 ± 12.243	47.354 ± 13.796^h
2 mo	27.180 ± 1.024	65.905 ± 9.460	69.818 ± 16.498	75.842 ± 31.614	103.284 ± 17.607
3 mo	31.481 ± 2.763	54.390 ± 19.176	58.735 ± 29.179	72.572 ± 29.935	40.709 ± 8.916
BS/TV, 1/mm
1 mo	21.936 ± 2.061^g	50.848 ± 30.750	73.897 ± 15.432	23.648 ± 2.841^g	21.564 ± 4.941^{d , g}
2 mo	24.902 ± 0.955^h	28.658 ± 3.698	57.728 ± 18.778	36.884 ± 9.738	31.481 ± 15.792
3 mo	29.482 ± 2.402	38.441 ± 8.162	46.175 ± 15.236	31.588 ± 5.220	34.899 ± 5.368

Data are shown as mean ± SD. BMD, bone mineral density; BS, bone surface; BV, bone volume; Col1, collagen type I; nHA, nanohydroxyapatite; PLGA, poly(lactic-co-glycolic) acid; TV, tissue volume.

P < .0001 compared to the aligned PLGA/Col1-PLGA/nHA group.

P < .0001 compared with the model group.

P < .05 compared with the model group.

P < .01 compared to the aligned PLGA/Col1-PLGA/nHA group.

P < .01 compared with the model group.

P < .001 compared to the aligned PLGA/Col1-PLGA/nHA group.

P < .05 compared to the aligned PLGA/Col1-PLGA/nHA group.

P < .001 compared with the model group.

Biomechanical Analysis

Regarding ultimate tensile strength (Figure 7A(a)), the aligned PLGA/Col1-PLGA/nHA group showed significant superiority over the other groups in the second month, with the exception of the control group. The PLGA group showed a statistically significant difference in stiffness compared with the model and random PLGA/Col1-PLGA/nHA groups at 3 months after surgery. No statistical difference in stiffness was noted for the other groups at the same time point (Figure 7A(b)). Concerning the elastic modulus (Figure 7A(c)), with the exception of the control group, no significant differences were observed between the groups in the first 2 months after surgery. Notably, the aligned PLGA/Col1-PLGA/nHA group demonstrated superiority over the model and random PLGA/Col1-PLGA/nHA groups.

In summary, apart from the control group, the aligned PLGA/Col1-PLGA/nHA group demonstrated superior ultimate tensile strength compared with the other groups. Additionally, it exhibited superiority over the model and random PLGA/Col1-PLGA/nHA groups in terms of the elastic modulus. These findings support the conclusion that placing aligned PLGA/Col1-PLGA/nHA between the tendon and bone enhanced biomechanical properties in the tendon-to-bone healing process.

Real-Time PCR

Figure 7B illustrates the PCR results of all groups, with the PLGA group and the random PLGA/Col1-PLGA/nHA group showing abnormally high expression across all indicators. We speculate that this could be attributed to the abnormal aggregation of chondral cells, heterotopic ossification, and a significant presence of tendon fibers without disorder, as shown by the histological and micro-CT results. Notably, the aligned PLGA/Col1-PLGA/nHA group demonstrated a significant difference from the other groups in BMP2 and collagen type II, excluding the PLGA and random PLGA/Col1-PLGA/nHA groups. These findings are in line with our histological and micro-CT results for the aligned PLGA/Col1-PLGA/nHA group.

Transcriptome Analysis

Pearson correlation analysis demonstrated a significant correlation among the samples (Appendix Figure A2A [available online]). Principal component analysis further revealed distinct gene expression profiles between the model and aligned PLGA/Col1-PLGA/nHA groups (Appendix Figure A2B [available online]). RNA sequencing analysis identified 62 differentially expressed genes, with 15 downregulated and 47 upregulated genes in the aligned PLGA/Col1-PLGA/nHA group compared with the model group. Heatmaps illustrated an opposing trend between the 2 groups (Figure 8A).

Moreover, GO and KEGG enrichment analyses identified over 50 GO terms and 20 KEGG pathways. Specifically, 3 GO terms (GO:0045670, GO:0030278, GO:0071260) were identified, each associated with a specific differentially expressed gene (ADAM8, EGR2, and IL-1β) (Figure 8B). Additionally, KEGG enrichment analysis suggested that tumor necrosis factor (TNF) and nuclear factor–kappa B (NF-κB) signaling pathways (Figure 8A(b)) might be the most relevant ones.

Discussion

The challenging issue of poor healing at the tendon-to-bone interface, leading to a high recurrence rate postoperatively, remains a crucial problem awaiting resolution in the field of sports medicine. Drawing inspiration from the natural structure of the tendon-to-bone interface, we devised a bilayer electrospun nanofiber membrane, seamlessly transitioning from aligned PLGA/Col1 to random PLGA/nHA. The nanofibers in the aligned PLGA/Col1 layer were meticulously aligned with the direction of tendon fibers, while those in the PLGA/nHA layer were randomly arranged to mimic the fiber direction at the bone end.⁴ Rigorous characterization by FTIR, SEM, EDS, and mechanical testing informed the determination of mass fractions and ratios for aligned PLGA/Col1 and PLGA/nHA. The final nanofiber membrane, combining aligned PLGA/Col1 10% 3:1 with PLGA/nHA 9% 5:1, was successfully fabricated.

In subsequent in vitro experiments, we assessed the biocompatibility, collagen secretion, and mineralization of various materials. Aligned PLGA/Col1 and PLGA/nHA nanofiber membranes demonstrated significantly superior results in terms of cell proliferation, cell adhesion, and collagen secretion compared with the other experimental groups. PLGA/nHA exhibited the ability to mineralize comparable with the blank group. Moving to in vivo experiments, mice implanted with aligned PLGA/Col1-PLGA/nHA exhibited significantly superior outcomes in bone formation, BMP2 and collagen type II expression, cartilage formation, and biomechanical properties after surgery. To delve into the underlying cellular mechanisms, we conducted transcriptome analysis, screening 3 genes from GO enrichment and 2 pathways from KEGG enrichment. The potential mechanism underlying these outcomes may involve the TNF and NF-κB signaling pathways along with the genes IL-1β, ADAM8, and EGR2.

Recent endeavors in the realm of sports medicine have seen extensive exploration of electrospun biomaterials to improve tendon-to-bone healing. However, in the methods sections of these studies, authors have often provided specific mass fractions and ratios of materials without delving into the rationale behind their choices. Moreover, some studies^11,15-17 have employed identical materials without varying mass fractions and ratios. In light of this, our approach involved designing a gradient in mass fractions and carefully determining the ratios between materials to identify the optimal composite materials. In their study, Chandrasekar et al³ delved into the synthesis and characterization of nHA, presenting specific wavenumbers and corresponding stretching modes associated with nHA. Interestingly, we observed similarities when combining PLGA with nHA, mirroring their findings. This parallel extended to combining PLGA with Col1, as demonstrated by Sun et al.¹⁷ Furthermore, our results are in line with this combination, as verified by EDS, confirming the chemical composition of both PLGA and Col1 as well as PLGA and nHA.

SEM was employed to observe the topography of composite materials. Aligned PLGA/Col1 displayed clear directionality and alignment, while PLGA/nHA exhibited random fibers and the aggregation of nHA particles. In a study by Moffat et al,¹⁴ PLGA was used to design a scaffold for rotator cuff repair and augmentation. They assessed the mechanical performance and fibroblast response of aligned and unaligned PLGA scaffolds, noting that fibroblasts cultured on the aligned scaffold attached along the nanofiber’s long axis, with cells on the unaligned scaffold appearing polygonal and randomly oriented. Additionally, cells on the aligned scaffold produced more integrin. Moreover, aligned nanofibers demonstrated significantly superior mechanical properties compared with unaligned nanofibers. Our SEM and mechanical testing results for aligned PLGA/Col1 and PLGA/nHA were consistent with Moffat et al’s¹⁴ findings. These structural and mechanical characteristics suggest that our materials could provide a conducive environment for cells and surgeons, making them promising for clinical applications.

In our CCK-8 assay, all materials inhibited the proliferation speed of fibroblasts compared with the blank group; however, fibroblast numbers increased steadily over time. Notably, fibroblasts on aligned PLGA/Col1 nanofibers exhibited the fastest increase, potentially attributed to the addition of Col1 and its alignment. Surprisingly, the proliferation of MC3T3 cells on PLGA/nHA nanofibers was significantly better than the blank group. Alignment and biocompatibility were confirmed using confocal microscopy, and as mentioned earlier, fibroblast morphology was influenced by the aligned fibers. Additionally, fibroblasts on PLGA/Col1 nanofibers demonstrated the ability to produce more Col1, consistent with Moffat et al’s¹⁴ findings.

In histological staining, aligned PLGA/Col1-PLGA/nHA exhibited superior tendon maturation, tendon-to-bone interface, and cartilage formation compared with the other experimental groups. The random PLGA/Col1-PLGA/nHA group and the PLGA group displayed more tendon fibers and chondral cell aggregation, but the structure of the tendon, cartilage, and tendon-to-bone interface was disordered. These findings provide insight into the abnormal higher results in PCR for these 2 groups, including indicators of bone, tendon, and cartilage formation. Aligned PLGA/Col1-PLGA/nHA demonstrated enhanced bone-forming ability compared with the other experimental groups. However, in the third month after surgery, each experimental group exhibited comparable bone formation results. When considering biomechanical results, the critical phase for aligned PLGA/Col1-PLGA/nHA in enhancing tendon-to-bone healing appeared to be within the first 2 months after surgery.

Many studies^7,10,11,24 have traditionally examined their biomaterials in larger animals, such as rabbits, because of the ease of placing materials at specific sites. However, in this study, we overcame the challenge of a limited operating space in mice by strategically placing our membrane between the Achilles tendon and bone. This approach allowed us to delve deeper into the molecular mechanisms underlying the effects of this material. Transcriptome analysis showed that the potential mechanism underlying these improving outcomes might involve the TNF and NF-κB signaling pathways, along with the genes IL-1β, ADAM8, and EGR2. This incremental understanding of the molecular mechanisms of biomaterials represents a significant stride toward advancements in future clinical practice.

The design of this novel aligned-to-random PLGA/Col1-PLGA/nHA bilayer electrospun nanofiber membrane was inspired by the 4 layers of the natural tendon-to-bone interface. The aligned layer aimed to promote tendon maturation, while the nHA layer was intended to facilitate bone formation. This innovative approach sought to transform tendon-to-bone healing into a dual process involving both bone-bone and tendon-tendon healing. However, certain limitations should be acknowledged. First, the anatomic, genetic, and postoperative rehabilitation differences in small animals may affect the generalizability of the findings to humans. Second, the 3 genes and 2 signaling pathways were not investigated, leaving room for potential avenues of exploration. Lastly, a more in-depth exploration of the molecular and cellular mechanisms is warranted.

Conclusion

The performance of the novel aligned-to-random PLGA/Col1-PLGA/nHA bilayer electrospun nanofiber membrane was demonstrated to be superior to other experimental materials both in vivo and in vitro. Notably, it excelled in enhancing tendon-to-bone healing, specifically in terms of cartilage and bone formation, tendon maturation, and the biomechanical strength of the tendon-to-bone interface after surgery. The potential molecular mechanisms underlying these effects may be correlated with the TNF and NF-κB signaling pathways as well as the genes IL-1β, ADAM8, and EGR2.

Supplemental Material

sj-pdf-1-ajs-10.1177_03635465241310530 – Supplemental material for An Aligned-to-Random PLGA/Col1-PLGA/nHA Bilayer Electrospun Nanofiber Membrane Enhances Tendon-to-Bone Healing in a Murine Model

Supplemental material, sj-pdf-1-ajs-10.1177_03635465241310530 for An Aligned-to-Random PLGA/Col1-PLGA/nHA Bilayer Electrospun Nanofiber Membrane Enhances Tendon-to-Bone Healing in a Murine Model by Baoyun Xu, Yunjiao Wang, Gang He, Xu Tao, Shang Gao, Mei Zhou, Yuzhen Tang, Kang-lai Tang, Lin Guo and Wan Chen in The American Journal of Sports Medicine

Footnotes

Submitted April 26, 2024; accepted November 7, 2024.

One or more of the authors has declared the following potential conflict of interest or source of funding: This study was funded by the National Natural Science Foundation of China (No. 82130071 and No. 82102633) and the Natural Science Foundation of Chongqing (cstc2021jcyj-msxmX0137). AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.

References

Angeline

Rodeo

SA.

Biologics in the management of rotator cuff surgery. Clin Sports Med. 2012;31(4):645-663.

Atesok

Wolf

, et al. Augmentation of tendon-to-bone healing. J Bone Joint Surg Am. 2014;96(6):513-521.

Chandrasekar

Sagadevan

Dakshnamoorthy

Synthesis and characterization of nano-hydroxyapatite (n-HAP) using the wet chemical technique. Int J Phys Sci. 2013;8:1639-1645.

Font Tellado

Balmayor

Van Griensven

Strategies to engineer tendon/ligament-to-bone interface: biomaterials, cells and growth factors. Adv Drug Deliv Rev. 2015;94:126-140.

Galatz

Ball

Teefey

Middleton

Yamaguchi

The outcome and repair integrity of completely arthroscopically repaired large and massive rotator cuff tears. J Bone Joint Surg Am. 2004;86(2):219-224.

Harryman

Mack

Wang

Jackins

Richardson

Matsen

FA.

Repairs of the rotator cuff: correlation of functional results with integrity of the cuff. J Bone Joint Surg Am. 1991;73(7):982-989.

Honda

Gotoh

Kanazawa

, et al. Hyaluronic acid accelerates tendon-to-bone healing after rotator cuff repair. Am J Sports Med. 2017;45(14):3322-3330.

Ide

Kikukawa

Hirose

, et al. The effect of a local application of fibroblast growth factor-2 on tendon-to-bone remodeling in rats with acute injury and repair of the supraspinatus tendon. J Shoulder Elbow Surg. 2009;18(3):391-398.

Inui

Kokubu

Makino

, et al. Potency of double-layered poly L-lactic acid scaffold in tissue engineering of tendon tissue. Int Orthop. 2010;34(8):1327-1332.

10.

Kovacevic

Fox

Bedi

, et al. Calcium-phosphate matrix with or without TGF-β3 improves tendon-bone healing after rotator cuff repair. Am J Sports Med. 2011;39(4):811-819.

11.

Fan

Sun

Liu

Cheng

Fan

Functional regeneration of ligament-bone interface using a triphasic silk-based graft. Biomaterials. 2016;106:180-192.

12.

Lui

PPY

Zhang

Chan

Qin

. Biology and augmentation of tendon-bone insertion repair. J Orthop Surg Res. 2010;5(1):59.

13.

Makadia

Siegel

SJ.

Poly lactic-co-glycolic acid (PLGA) as biodegradable controlled drug delivery carrier. Polymers (Basel). 2011;3(3):1377-1397.

14.

Moffat

Kwei

ASP

Spalazzi

Doty

Levine

HH.

Novel nanofiber-based scaffold for rotator cuff repair and augmentation. Tissue Eng Part A. 2009;15(1):115-126.

15.

Ngiam

Liao

Patil

Cheng

Chan

Ramakrishna

The fabrication of nano-hydroxyapatite on PLGA and PLGA/collagen nanofibrous composite scaffolds and their effects in osteoblastic behavior for bone tissue engineering. Bone. 2009;45(1):4-16.

16.

Sadeghi-Avalshahr

Khorsand-Ghayeni

Nokhasteh

Molavi

Naderi-Meshkin

Synthesis and characterization of PLGA/collagen composite scaffolds as skin substitute produced by electrospinning through two different approaches. J Mater Sci Mater Med. 2017;28(1):14.

17.

Sun

Han

Zhang

, et al. A synthetic bridging patch of modified co-electrospun dual nano-scaffolds for massive rotator cuff tear. J Mater Chem B. 2016;4(45):7259-7269.

18.

Thomopoulos

Genin

Galatz

LM.

The development and morphogenesis of the tendon-to-bone insertion: what development can teach us about healing. J Musculoskelet Neuronal Interact. 2010;10(1):35-45.

19.

Vergis

Gillquist

Graft failure in intra-articular anterior cruciate ligament reconstructions: a review of the literature. Arthroscopy. 1995;11(3):312-321.

20.

Wang

Yang

Cheng

, et al. Collagen/silk fibroin composite scaffold incorporated with PLGA microsphere for cartilage repair. Mater Sci Eng C Mater Biol Appl. 2016;61:705-711.

21.

Xie

Lipner

, et al. “Aligned-to-random” nanofiber scaffolds for mimicking the structure of the tendon-to-bone insertion site. Nanoscale. 2010;2(6):923.

22.

Wang

Tang

Guo

Chen

A novel and efficient murine model for investigating tendon-to-bone healing. J Orthop Surg Res. 2024;19(1):90.

23.

Yang

ZZ.

Polylactic acid nanofiber scaffold decorated with chitosan islandlike topography for bone tissue engineering. ACS Appl Mater Interfaces. 2017;9(25):21094-21104.

24.

Zhao

Han

Zhang

Sun

Lin

Chen

Hydroxyapatite-doped polycaprolactone nanofiber membrane improves tendon-bone interface healing for anterior cruciate ligament reconstruction. Int J Nanomedicine. 2015;10:7333-7343.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.38 MB

0.00 MB