Assembly of Interfacial Polyelectrolyte Complexation Fiber Bundles with Mineralization Gradient for Physiologically-inspired Ligament Regeneration

doi:10.21203/rs.3.rs-3455055/v1

Download PDF

Article

Assembly of Interfacial Polyelectrolyte Complexation Fiber Bundles with Mineralization Gradient for Physiologically-inspired Ligament Regeneration

https://doi.org/10.21203/rs.3.rs-3455055/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Current synthetic grafts for ligament rupture repair often fail to integrate well with the surrounding biological tissue, leading to complications such as graft wear, fatigue, and subsequent re-rupture. To address this medical challenge, this study aims at advancing the development of a biological ligament through the integration of physiologically-inspired principles and tissue engineering strategies. In this study, we utilize interfacial polyelectrolyte complexation spinning technique, along with a custom-designed collection system, to fabricate a hierarchical scaffold mimicking native ligament structure. To emulate the bone-ligament interface and alleviate stress concentration, a hydroxyapatite mineral gradient is strategically introduced near both ends of the scaffold to enhance interface integration and diminish the risk of avulsion rupture. Biomimetic viscoelasticity is successfully displayed to provide similar mechanical support to native ligamentous tissue under physiological conditions. By introducing the connective tissue growth factor and conducting mesenchymal stem cells transplantation, we significantly amplify the regenerative potential of the synthetic ligament. This pioneering study offers a multifaceted solution combining biomimetic materials, regenerative therapies, and advanced techniques to potentially transform ligament rupture treatment.

Biological sciences/Biotechnology/Biomaterials/Tissues

Health sciences/Medical research/Translational research

ligament tissue engineering

interfacial polyelectrolyte complexation

hierarchical structure

mineralization gradient

bone-ligament interface

Ligament-like hierarchical architecture is developed to closely mimic the hierarchical fiber organization of ligaments and act as a structural cue for de novo tissue generation.
Enthesis-like hydroxyapatite gradient is incorporated to promote transition from soft to hard tissue and enhance ligament-to-bone interface healing.
Biomimetic viscoelasticity is created to provide similar mechanical support to native ligametous tissue under physiological conditions.
Tissue-engineered strategies including MSC inoculation and growth factor delivery are included to improve the regenerative outcome.

Ligamentous injuries, affecting around 17 million individuals annually, present a significant orthopedic challenge and impose a healthcare burden with costs exceeding $40 billion per year ¹. Additionally, they frequently give rise to secondary complications, including meniscal tears, chondral lesions, and post-traumatic osteoarthritis ². Biological grafts, commonly used for ligament reconstruction, are accompanied by notable constraints. These include donor site morbidity linked to autologous graft harvesting, and delayed healing or immunological complications following allograft implantation ³. Conversely, synthetic ligaments such as the PET-based ligament augmentation and reconstruction system (LARS) exhibit satisfactory short-term outcomes. However, they are associated with considerable long-term failure rates and complications due to their incompatible biological properties. Acknowledging these limitations and problems, this study aims at developing tissue-engineered ligament substitutes that leverage bio-inspired design strategies and multifunctional regenerative capabilities. The objective is to faithfully replicate native ligament performance and thereby facilitate effective ligament regeneration.

The interfacial polyelectrolyte complexation (IPC) spinning technique synthesizes biocompatible fibers exhibiting excellent properties for diverse biomedical applications. This emerging biofabrication method engenders fibers by harnessing electrostatic interaction between pairs of polycation and polyanion, thereby establishing a dynamic interface within an aqueous, ambient environment ⁴. The distinctive eco- and bio-friendly mechanism, devoid of organic solvents usage, distinguishes IPC fibers from other spinning techniques like electrospinning and 3D textiles commonly employed in ligamentous biomaterials. The IPC process demonstrates immense potential in efficiently encapsulating a wide range of bioactive molecules, from proteins, nucleic acids, and extracellular vesicles to cells ^5,6. Moreover, IPC fibers demonstrate exceptional mechanical strength, which can be further amplified through crosslinking process ⁷. The programmability of IPC reaction enables the fabrication of intricate structures, including multiphasic and core-sheath fibers, thereby broadening their bioengineering applicability ^8,9. Leveraging the advantage of IPC spinning technique, our study uses naturally-derived biopolymers to achieve a balance between bioactivity and mechanical stability through IPC process. The poly-D-lysine (PDL) synthetic polypeptide, known for its ability to promote cell adhesion and resistance to enzymatic degradation, is utilized to ensure long-term stability and integration of the artificial ligament, as well as to enhance ligament fibroblasts activity and matrix formation ^10,11. Furthermore, a natural polysaccharide, pectin, is incorporated not only to provide mechanical strength but also to possess significant potential in promoting osteogenesis at the graft site, thereby augmenting the healing at the juncture between the graft and the host bone and culminating in enhanced outcomes for ligament reconstruction ¹².

Ligament tissues demonstrate a restricted capacity of self-healing, impeded by inherent characteristics such as reduced cellularity, constrained vascularization, and diminished metabolic activity¹³. Therefore, our study tries to tackle this shortage by capitalizing on the capability of IPC scaffold to carry biomolecules, specifically to deliver connective tissue growth factor (CTGF) and exploit its potential in aiding the regeneration of ligament tissue. CTGF exhibits key properties beneficial to the healing process, including stimulating fibroblast proliferation, promoting extracellular matrix (ECM) secretion, enhancing collagen deposition, and upregulating the expression of ligament/tendon-related genes ¹⁴. Moreover, by harnessing unique affinity between CTGF and heparan sulfate proteoglycan (HSPG), it is possible to deliver CTGF effectively while being shielded within HSPG-based polyelectrolyte complex nanoparticles (PCNs) ¹⁵. This delivery system provides protection against enzymatic degradation, thereby ensuring the bioactivity of the growth factor. Besides, healing of the graft-to-bone interface also presents a significant clinical hurdle ¹⁶. One promising strategy to overcome this shortage is the use of hydroxyapatite (HAp). HAp, a bioactive material, mirrors the structure and chemistry of bone phosphates, thereby providing an environment conducive to bone integration ¹⁷. As an increasing research evidence, its significant role in accelerating graft-to-bone healing, stimulating bone formation, enhancing biomechanical stability, and potentially curbing the risk of postoperative osteoarthritis, underscores the substantial contribution of HAp in optimizing postoperative recovery ¹⁶.

Bone marrow-derived MSCs, in particular, offer distinct advantages due to their widespread availability, propensity for musculoskeletal tissues, and proven effectiveness in regenerative medicine ¹⁸. Particularly, studies have consistently associated these cells with positive outcomes, demonstrating their efficacy in improving ligamentization, promoting both intrinsic and extrinsic healing, and enhancing osseointegration in the graft-to-bone interface following ACL reconstruction ^19,20. Despite integrating numerous tissue engineering strategies, inherent MSCs in ligaments have been reported to hinder the repair process due to their limited regenerative capacity in tissue with limited vascularity²¹. This inherent limitation underscores the imperative for the supplement of therapeutic stem cells and bioactive factors.

Collectively, this study devises a biomimetic ligament graft with enhanced biological attributes (Fig. 1). With the assistance of the customized spinning instrumentation, our approach centers on the development of a hierarchically-structured IPC scaffold, composed of oppositely charged pectin and PDL polymers to closely mimic the structural complexity of the native ligament (Supplementary Video S1). This transplant is further fortified through targeted chemical and physical enhancements, ensuring its functional endurance mirrors that of natural ligaments. Furthermore, we strategically introduce HAp in a gradient configuration near both scaffold ends, fostering a seamless graft-to-bone integration and aligning with the inherent mineralization gradient of the enthesis. To further enrich the design, regenerative modalities are integrated: MSCs are seeded uniformly on the scaffold while CTGF is encapsulated within a collagen sheath, thus ensuring sustained therapeutic benefits. In essence, this intriguing approach is designed to significantly advance the field of ligament tissue engineering. Our methodology provides a pathway towards developing an artificial ligament substitute that possesses a physiological resemblance and offers biological cues to promote functional recovery.

2.1 Morphological Observations of Ligament-like Scaffold

To create the hierarchical architecture of ligament tissue, a bottom-up approach was utilized via the IPC spinning technique. This apparatus was instrumental with sequential collection and assembly of the hierarchical architecture. In Fig. 2, the well-defined hierarchy of the ligament-like scaffold was clearly discernible under SEM analysis. It composed of easily identifiable fascicles and sub-fascicles corresponding to secondary and primary fiber bundles (Fig. 2A). Furthermore, the IPC fibers exhibited a porous microstructure like that distributed in a hydrogel (Fig. 2B). The porous nature of the fibers allowed cell infiltration and facilitated oxygen and nutrient exchange. Quantification using Image J revealed a pore diameter of 9.86 µm and a porosity of 50.05%. The fibrous morphology at each step of the biofabrication process was also examined and collected by critical point drying to remove water content. From the top view of the fibrils, fibers, to fiber bundles, the scaffold exhibited a remarkable resemblance to the microstructure found within ligaments (Fig. 2C, D). By comparing with the cross-section images (Fig. 2E, F, G), we determined that the ligament-like structure composed of five secondary fiber bundles, with each secondary fiber bundle consisting of several primary fiber bundles. The primary fiber bundle, in turn, comprised five IPC fibers. Higher levels of fiber bundles were formed by tightly packed and strictly compartmentalized fibers from lower levels, demonstrating a hierarchical structure throughout the scaffold.

An interesting discovery was noted regarding the distribution of polyionic molecules within the cross-section of a single IPC fiber (Fig. 2H). To visualize this, fluorescent dyes were conjugated to pectin and PDL respectively, and confocal microscopy was used to image the samples after frozen sectioning. It was found that a uniform distribution with an intra-fiber pattern was obtained. The PDL and pectin were closely bound together within each IPC fiber and did not mix or separate from each other. These findings demonstrated the five-level hierarchical architecture of our scaffold, ranging from the polyelectrolyte molecules to ligament-like tissue level. It also confirmed the success of this bottom-up approach in scaling up micro-strands to the macro-scale for practical applications.

2.2 Characterization of Enthesis-like HAp Gradient

To replicate the gradient structure found in the enthesis, a hydroxyapatite (HAp) coating was constructed using a wet synthesizing method (Fig. 3A) ²². The coating solution could spontaneously evaporate, resulting in spatially-varied deposition time across the scaffold, thereby creating a gradient in mineral content. To affirm the successful preparation of HAp on both ends of the scaffold, the crystal structure was confirmed using an X-ray diffraction (XRD) pattern (Fig. 3B). The analysis revealed noticeable characteristic diffraction peaks, corresponding to the HAp standard as indexed in the JCPDS data card. This discovery suggested successful in situ grain growth of HAp on the IPC scaffold. Further analysis of the elemental composition revealed the presence of both calcium and phosphorus on the scaffold surface with an atomic Ca/P ratio of approximately 1.67, which mirrors the hydroxyapatite formula (Fig. 3C). This quantitative analysis provided the evidence that HAp was successfully synthesized on the surface of the ligament-like scaffold. Moreover, the surface morphology revealed micrometer-sized particles, characterized by whisker and needle-like formation of HAp crystals under SEM observation (Fig. 3D). The crystals displayed a uniform size distribution, indicating consistent and controlled lattice growth during the synthesis process. Most importantly, the atomic Ca/(Ca + C) ratio that varied continuously along the scaffold was identified, ranging from 14–40% over 4 mm (Fig. 3E). The obtained data exhibited a high correlation coefficient (R² = 0.92), indicating a strong linear relationship between the deposition position and the mineral content.

2.3 Evaluation of the Strengthened Ligament-like Scaffold

Enhancement of the mechanical stability via the replacement of ionic bonds with covalent bonds was evident in Fig. 4A and confirmed by FT-IR spectrum (Supplementary Fig. S3). The spectrum demonstrated the EDC/NHS coupling-initiated formation of an amide bond between carboxyl groups from pectin and amino groups from PDL. This crosslinking resulted in significant improvements in the mechanical properties of scaffold under physiologically wet conditions, including increased ultimate tensile strength (UTS) from 1.24 MPa to 4.33 MPa, reduced deformation before breakage, and heightened stiffness, as indicated by the enhanced Young's modulus and summarized in Supplementary Table S2. These findings suggested that crosslinking effectively improved the mechanical stiffness of the scaffold. In addition to the crosslinking reinforcement, a custom-made twisting device was employed to enhance the structural integrity of the crosslinked scaffold. The tensile testing was conducted to evaluate the effect of different twisting angles on the mechanical performance of the fabricated scaffolds. As visualized in Fig. 4B, an increase in the twisting angle led to enhancements in ultimate tensile strength, strain at break, and stiffness. However, mechanical performance was significantly declined beyond 90°/cm, likely due to stress concentration and subsequent microfractures when over-twisting. The optimal performance was found with a twisting angle at 45°/cm, and scaffolds based on this parameter was chosen for further investigations.

2.4 Measurement of Tissue-like Viscoelasticity

To ensure that the ligament-like scaffold closely mimics the mechanical behavior of native tissue during daily activities, it is imperative to comprehensively evaluate its viscoelastic properties. Tests such as cyclic testing, creep, and stress relaxation serve as valuable tools for assessing the durability of the scaffold and the ability to withstand long-term mechanical stress. Figure 4C and Supplementary Table S3 depicted the creep results of ligament-like scaffolds without collagen coating. Under a constant stress of 6MPa for 5 minutes, a substantial creep deformation of 9.85% was observed together with persistent residual strain of 13.79% and the recovery ratio of 44.45%. This stress range of 6MPa faithfully emulates the peak stress encountered by the human ACL during typical walking, as deduced from the peak force data provided by Shelburne et al. and the average cross-sectional area data elucidated by Iriuchishima et al ²³. On the other hand, the collagen-coated scaffold presented a creep deformation of 4.22%, only slightly higher than the 1.57% deformation of the porcine ACL. Both the collagen-coated scaffold and the native porcine ACL exhibited a recovery ratio of approximately 90% after load release, reducing the strain to about 1%. The stress relaxation behavior, as shown in Fig. 4D, was significant when the uncoated scaffold was subjected to a constant 10% strain, a value derived from the peak stress in the force-deformation curve by Shelburne et al. and the average ACL length by Marieswaran et al. ²⁴. In the absence of collagen coating, the scaffold retained only 20% of the initial stress after 1000 seconds. However, when the scaffold with collagen coating, the stress relaxation phenomenon improved remarkably, preserving approximately 80% of the initial stress level.

For the cyclic test involving 100 loading-unloading cycles with a stress range from 0 MPa to 6 MPa as shown in Fig. 4E and 4F, the uncoated scaffold demonstrated an increase in peak strain from 22.27–54.08%, and residual strain from 10.06–41.21%, indicating significant permanent deformation and laxity. On the other hand, the collagen-coated scaffold only showed a slight increase in peak strain from 12.37–14.04% and residual strain from 2.97–5.06%. This magnitude was comparable to the increase seen in the porcine ACL, which exhibited an increase in peak strain from 13.94–18.13% and residual strain from 5.72–8.79%. It is suggested that the collagen-coated scaffold, like the porcine ACL, did not show any significant elongation or laxity. Additionally, the Young's modulus of these groups remained remarkably stable across the cycles, implying the absence of substantial stiffening or weakening in the examined region (Fig. 4G). From the above results, it is indicated that the incorporation of collagen coating layer enhances the fatigue resistance of the scaffold, thereby establishing collagen-coating as an optimal parameter for production of the proposed ligament substitute.

2.5 Quantification of Recombinant CTGF Release and Scaffold Degradation

Based on our initial consideration, the encapsulation of CTGF within IPC fibers was achieved by its integration into a polyelectrolyte solution, leading to successful protein entrapment. In Supplementary Fig. S4, release kinetics demonstrated that CTGF enclosed by an external sheath exhibited slower release profiles, with a release percentage of 5.78% at 14 days, as opposed to 11.86% from a uniformly distributed configuration. However, an alternative approach was adopted to strike a balance between release rate and therapeutic effectiveness, involving the incorporation of CTGF into a collagen sheath subjected to mild EDC crosslinking to prevent burst release. The previous complexation process of CTGF with PCN ensured the protection of CTGF from direct contact with EDC (Fig. 5A). As presented in Fig. 5B, collagen without EDC crosslinking displayed a burst release profile, while the introduction of EDC crosslinking resulted in a sustained release curve, with a 14-day release percentage of 26.57%. Protection through electrostatic binding with PCN successfully allowed CTGF to display controlled and prolonged release over a two-week period, achieving a cumulative release of 84.17% for two weeks.

Degradation experiments were then conducted to evaluate material sustainability within physiological environments. As revealed in Fig. 5C, weight loss over time was used as an indicator of degradation progress. The IPC scaffold was stable when incubated in PBS, with initial degradation kinetics resulting in approximately 5% degradation by day three and then without further change subsequently. We also found the inherent resistance to enzymes of the IPC scaffold with an integrated collagen sheath. Even in the presence of collagenase, the scaffold still maintained stable at around 80% of its initial weight.

2.6 Scaffold Cytocompatibility and MSC Proliferation

The biocompatibility of the ligament-like scaffold was investigated with MSCs cultivation, employing three conditions: IPC scaffold, collagen-coated IPC scaffold, and HAp-deposited IPC scaffold. In cytotoxicity evaluation, scaffolds co-cultured with MSCs underwent assessment via the LDH assay (Fig. 5D). As shown in Fig. 5E, cell death percentages in all scaffold groups demonstrated no cellular damages during one, three, and five-day co-culture period, indicating the exceptional biocompatibility and potential of scaffold as an implant candidate.

To probe the efficacy of the ligament-like scaffolds as carriers for cell therapy, a refined cell seeding technique was adopted. V-shaped PDMS molds were employed to support cell attachment onto the fibrous scaffold within a confined tissue-culture plate area (Fig. 5F). After 24-hour incubation to encourage cell adhesion, the scaffold was transferred to a new culture well for continued culture. The CCK-8 assay quantitatively determined cellular mitochondrial activity on the scaffold, with an MSC-based standard curve established to ensure accurate cell count. Result findings indicated ongoing cellular growth on the coated scaffolds beyond one week (Fig. 5G). Notably, the collagen-coated scaffold surpassed the HAp-coated scaffold in proliferation rates and cell yields. To visualize cell attachment on the scaffold, a Live/Dead assay was also demonstrated, as detailed in Supplementary Fig. S5. Prior to this assay, Sudan Black B was employed to quench the autofluorescence generated by scaffold, and a CCK8 assay was conducted to confirm the non-influential impact of SBB on cell viability. The findings echoed the commendable cellular attachment properties found on the collagen-coating scaffold, as indicated by the discovery of unidirectional alignment. These outcomes confirmed the capability of the scaffold as a cell therapy carrier, promoting extended cellular growth to support further in vivo transplantation.

2.7 In vivo ACL Reconstruction in Rabbit Model

The plan of animal study incorporated multiple groups to evaluate ACL regeneration using various graft conditions. Experimental groups used IPC scaffold constructs with collagen coating and HAp mineralization. Additionally, the therapeutic potential of CTGF alone (Scaffold/CTGF) and combined with MSCs (Scaffold/CTGF/MSC) was investigated, respectively. Controls included commercially available LARS graft and Achilles tendon autograft. Post-implantation, samples were harvested at 4 and 10 weeks respectively, followed by histological staining, micro-CT, and biomechanical assessment to scrutinize graft performance.

Longitudinal sections of the harvested intra-articular graft were used to evaluate fiber alignment and tissue infiltration degree. Autograft samples exhibited consistently parallel fiber alignment at both 4 and 10 weeks. However, experimental groups at 4 weeks showed less cellularity, but they demonstrated extracellular matrix deposition and cellular penetration along the aligned IPC fibers. In the 10 weeks, the Scaffold, Scaffold/CTGF, and Scaffold/CTGF/MSC groups all exhibited enhanced cellularity within fiber interstices (Fig. 6A). Masson's Trichrome staining provided further insights, denoting cytoplasm as pink and collagen as blue (Fig. 6B). In the 4-week Scaffold and Scaffold/CTGF groups, minimal cytoplasmic and collagen deposition were observed. On the other hand, the Scaffold/CTGF/MSC group showed noticeable cytoplasmic and collagen expressions. At 10 weeks, progression in the Scaffold/CTGF group was apparent with increased collagen and cytoplasm depositions, while the Scaffold/CTGF/MSC group presented with more tissue infiltration.

To further understand the ligamentization process, the protein expressions of tenomodulin (TNMD) and tenascin C (TNC) were examined through immunohistochemistry. TNMD, a transmembrane glycoprotein predominantly located on the cell membrane, is recognized as a mature ligament cell marker. The influence of therapeutic factors, CTGF and MSC, was profoundly studied. As shown in Fig. 6C, groups supplemented with MSCs at 4 weeks demonstrated noticeable TNMD expression, thereby suggesting an accelerated regeneration. At 10 weeks after transplantation, an enhancement in TNMD expression within the Scaffold/CTGF group was then noted. This implied the high expression of TNMD was accompanied by a significant surge in cell quantity in the MSC supplemented group at 10 weeks. This finding not only re-affirms the regenerative capabilities but also suggests a potential ancillary role of MSCs in recruiting neighboring cells. TNC, a glycoprotein in the ligament extracellular matrix, also demonstrated a similar trend. At 4 weeks, minor TNC secretion was observed among interwoven fibers in the Scaffold/CTGF group (Fig. 6D). On the other hand, a substantial accumulation of TNC within the fibers was evident in the Scaffold/CTGF/MSC group. By 10 weeks, TNC expression in the Scaffold/CTGF group was more prominent, with the highest expression noted in the Scaffold/CTGF/MSC group. The IHC stain validated the efficacy of the synergistic effect involving CTGF and MSCs for fostering ligament regeneration.

The transverse sections of the bone tunnel stained with H&E and Masson's Trichrome staining were utilized to examine the tissue formation at the graft-bone interface, respectively (Fig. 7). The wide-field images with lower magnification of the scaffold interface and central region were presented in Supplementary Fig. S6. In the control, LARS and Autograft groups, fibrous tissue primarily constituted the graft-bone interface as identified in the H&E staining at both 4 and 10 weeks. For the Autograft group, the center of the graft displayed limited cellularity and loosely arranged collagen at the interface without significant progress over time. In the LARS group, a thickened fibrous tissue at the interface was formed. By the 10-week timepoint, few matured collagen fiber networks were observed within the LARS graft but still present with a gap between graft-bone interface. In contrast, histochemical staining unveiled remarkable tissue integration in the Scaffold, Scaffold/CTGF, and Scaffold/CTGF/MSC groups, and exhibited a narrower interface width, indicative of better graft-to-bone healing. As shown in Fig. 7A, it implied minimal tissue infiltration and preferential inter-fiber growth in the Scaffold and Scaffold/CTGF groups at 4 weeks by H&E staining. Until 10 weeks, these groups exhibited more tissue infiltration into fiber bundles, high cellularity, enhanced ECM deposition, and obscure graft-bone boundary (Fig. 7B). Masson's Trichrome staining highlighted a significant increase in collagen deposition in the Scaffold/CTGF group at both 4- and 10-week timepoints (Fig. 7C, D), supportive the role of CTGF in augmenting ECM deposition. Incorporation of MSCs distinctly expedited tissue regeneration as shown in the Scaffold/CTGF/MSC group. As early as 4 weeks, both H&E and Masson's Trichrome staining displayed matured ECM development and pervasive tissue infiltration within the scaffold. Intriguingly, by the 10 weeks, the emergence of trabecular bone-like dense tissue at the interface between bone and scaffold was found, suggesting the initiation of osteogenesis.

To further explore the osseointegration, immunohistochemical staining for osteocalcin (OCN), an osteoblast marker denoting mature osteogenesis, was performed. In 4 weeks, despite enhanced tissue infiltration was identified in some groups, there was no noticeable OCN expression in most scaffolds (Fig. 8A). Minor OCN expression was seen exclusively at the interface of the Scaffold/CTGF/MSC group, reflecting the prior observations of accelerated tissue repair. By 10 weeks, the Scaffold/CTGF/MSC group displayed pronounced OCN expression at the interface (Fig. 8B). This was absent in the LARS and Autograft groups. Additionally, trabecular bone features were identified in the Scaffold/CTGF/MSC scaffold, confirming its effectiveness in enhancing mineralization at the bone insertion site. These results corroborate the capacity of the scaffold for promoting bony ingrowth and highlight the synergistic therapeutic potential of stem cells for facilitating new bone formation at the graft site.

In addition, micro-CT imaging and quantitative analysis were employed 10 weeks after transplantation to evaluate the biomineralization within the bone tunnel. 3D reconstruction images demonstrated significant bone formation in the Scaffold/CTGF and Scaffold/CTGF/MSC groups (Fig. 8C). Both groups showed enhanced bone matrix along the bone tunnel wall and significant bony ingrowth in the central region, surpassing the Autograft and LARS groups, which displayed limited mineralization at the interface. Quantitative analysis of the ROI volume yielded prominent results (Fig. 8D). The Scaffold/CTGF and Scaffold/CTGF/MSC groups revealed a significant synergy in fostering osteogenesis and exhibiting a higher bone mineral density (BMD) and a slightly smaller bone tunnel area when compared with the Scaffold group. Furthermore, both groups showed an increased bone volume fraction (BV/TV) and trabecular number (Tb.N), supporting the promising osteogenic potential of CTGF and MSCs.

Biomechanical testing at 10 weeks provided critical insights into the mechanical performance of the femur-ligament-tibia complex (Supplementary Table S1 and Supplementary Fig. S7). The Autograft and LARS exhibited ultimate tensile strengths of approximately 12N and 10N, respectively, with failures occurring at the bone tunnel in the tibia, indicating concerns regarding bone-graft integration ability. On the other hand, the Scaffold group, without any therapeutic enhancements, recorded a modest ultimate tensile strength of 4.55N. Yet, the inclusion of CTGF and MSCs led to a significant improvement in tensile strength, nearly matching to the magnitude of the Autograft group. These results supported the beneficial impact of growth factor and stem cell therapy on the mechanical attributes of the scaffold.

In this study, we successfully reproduced the hierarchical structure of native ligament tissue, starting from a micro strand and progressing to a macroscale (Fig. 2). This was accomplished by developing a spinning instrument that seamlessly interfaces with IPC reactions. The collection procedure enabled the stable production of fiber-based scaffold. The resulting clinically applicable ligament-like scaffolds exhibited tightly bound yet strictly compartmentalized fiber bundles in their cross-sections, indicative of a well-defined hierarchical architecture. Notably, the fabrication process induced a step-by-step hierarchical organization. The fluorescence-labeled IPC fiber images revealed the spatial arrangement of pectin and PDL molecules within the fiber threads, demonstrating molecular-level hierarchy. In addition to that, a mineralization gradient at the bone-graft interface was also successfully created (Fig. 3). Our approach involved the development of a biofriendly environment that obviates the need for high temperature, specialized microchannel design, or fluid pump. Instead, by controlling at physiological temperature, we established the gradient driven by the evaporation of the liquid interface. This allowed us to achieve a linear distribution of the gradient within 4 mm. While it is acknowledged that the mineralization gradient of native enthesis is typically in hundreds µm ²⁵, our method offers a linear gradient capable of restoring the structure of the soft-to-hard tissue junction.

Mechanical properties play a pivotal role in the function of synthetic ligaments. Therefore, the chemical and physical enhancements to augment the material mechanical properties were adopted. Through chemical enhancement, we observed a substantial increase in mechanical strength. Importantly, all our tests were conducted under wet conditions, validating that IPC fibers after chemical crosslinking could maintain mechanical stabilities. Viscoelasticity and fatigue are critical characteristics of biological tissues. However, current polyester-based ligament scaffolds exhibit limited control in viscoelastic performance ²⁶. Theories underpinning ligament viscoelasticity involve fiber sliding between one another, which aligns with the characteristics of biopolymers and the mechanism of IPC through electrostatic attraction ²⁷. Our findings indicated that scaffolds lacking collagen coating prominently manifest stress relaxation, creep, and laxity (Fig. 4). The apparent fatigue may impose constraints on ligament tissue engineering with rapid relaxation phenomenon. The introduction of a collagen coating layer significantly augmented the fatigue resistance of the scaffold. The triple-helix structure of collagen allowed our ligament-like scaffold to approximate the viscoelastic behavior of porcine ACL under physiological conditions, encompassing peak stress and peak strain during human normal walking. Whether subjected to short-term cyclic loading-unloading or prolonged deformation or loading, the scaffold with collagen coating exhibited a physiology-relevant mechanical response with better durability.

The investigated system not only served as an effective ligament substitute but also demonstrated its potential as a tissue engineering platform, as supported by the in vitro analyses (Fig. 5). The IPC fibrous scaffold offered versatile drug-loading capability, making it feasible for drug encapsulation within the fibers via polyelectrolyte loading (Supplementary Fig. S4). Furthermore, the scaffold allowed for controlled release of bioactive cues through a core-sheath structure. The degradation of the collagen sheath in aqueous environments was extended using EDC crosslinking and twisting. The PCN complex successfully protected the bioactivity and prevented from direct interaction with the crosslinker and ensured the sustained release over time. These features collectively provided a robust and versatile drug-loading potential in this system. Furthermore, the scaffold supported cell adhesion and maintained cellular growth for weeks, presenting as an effective carrier for cell-based therapies. Considering the extended duration of ligament regeneration, which typically spans at least six months ³⁸, the scaffold employed in this study should exhibit excellent resistance to enzymatic degradation. As demonstrated in Fig. 5, the scaffold showed comparable degradation kinetics and stability in buffer and PBS containing collagenase. This resistance to enzymatic degradation was crucial for maintaining scaffold integrity throughout the regenerative process.

Animal study confirmed the compatibility and the effectiveness of the reported scaffold for ligament reconstruction (Fig. 6). Although autografts and LARS can provide adequate mechanical strength, bone-graft integration remains a challenging issue ^16,29. The autografts revealed the existence of fibrous scar tissue without prominent Sharpey's fibers and active bone protein expression at both the 4-week and 10-week intervals (Fig. 7). Sharpey's fibers are a matrix of connective tissue consisting of bundles of strong predominantly type I collagen fibers connecting periosteum to bone. This implies an immature remodeling process at the ligament-bone interface. Furthermore, micro-CT analysis revealed that limited new bone formed around the autograft, leading to inadequate fixation of the autograft (Fig. 8). Although fibrous scar tissue represents an initial step in bone remodeling, its limited integration compromised the graft fixation, resulting in subsequent graft pullout from the bone tunnel during mechanical stress³⁰.

In the case of LARS, previous studies have reported its ability to encourage fibroblast infiltration, which aligns with our findings. Our results demonstrated an increase in tissue thickness between PET fibers from 4 to 10 weeks. This confirmed the inherent limitations of PET in biological properties. Due to its non-degradable nature, newly formed matrix did not penetrate to the core area of the scaffold but instead resided between the fibers. Consequently, fibrous scar tissue remained the predominant feature, resulting in insufficient strength for tissue fixation and graft incorporation. In contrary, the scaffolds proposed in this study exhibit promising regenerative effects. While ACL has been reported to possess limited healing properties due to its inability to form a localized hematoma and poor to provide intrinsic cell source and nutrient ³¹, ligamentization was observed in the reported IPC grafts with synergistic therapy. On the other hand, significant tissue infiltration at the graft-to-bone interface was observed in our scaffolds (Fig. 7, 8). Even in the scaffold-only group, a higher number of cell nuclei was visible within the scaffold in 10 weeks, accompanied by enhanced tissue and cell penetration within the bone tunnel. At 10 weeks, integration between the scaffold and bone tissue was evident without distinct boundary between newly formed tissue and fiber bundles. Additionally, mineralization at the surroundings was observed in the micro-CT analysis, suggesting osteoconductive properties. Furthermore, the incorporation of CTGF into the scaffold significantly enhanced regeneration process characterized by heightened cell tissue infiltration, collagen production, and the display of ligament-specific ECM elements. We observed obvious expression of TNC, an ECM glycoprotein present in mature and growing ligament or tendon tissues, underlining the critical role of CTGF in initiating ligament healing and remodeling. As reported, TNC promotes cell-matrix interaction and assists in the precise positioning and alignment of collagen fibers ³². Additionally, a thin scar tissue layer surrounded by trabecular bone-like dense tissue was observed, indicating a more favorable and reliable graft-to-bone healing process. As confirmed by micro-CT analysis, there were notable improvements in BMD, BV/TV, Tb.N values, and a marked reduction in the bone tunnel area (Fig. 8). These multifaceted therapeutic effects of CTGF, which were also observed in previous studies ^30,33.

After incorporating rabbit bone marrow MSCs into the scaffold, we observed their therapeutic potential. Ligament regeneration is often hindered by a lack of cellular sources. With the intra-articular graft, notable regeneration, including cytoplasm infiltration and collagen deposition along with the expression of ligament-specific proteins like TNMD and TNC, was apparent as early as 4 weeks. Moreover, by 10 weeks post-surgery, a significant increase in TNMD positive cells in the ligament was observed, showcasing the effective recruitment of surrounding intrinsic cells by the integrated stem cell therapy. In the bone tunnel analysis, complete tissue infiltration in Scaffold/CTGF/MSC group was apparent after 4 weeks of transplantation. At 10 weeks, tissue resembling trabecular bone was observed with significant expression of OCN in the interface portion, indicating a more mature stage of bone formation. As shown in Supplementary Table S4, the biomechanical testing revealed improved stiffness in the groups treated with CTGF and MSCs, comparable to the magnitude of control groups, thereby demonstrating the reparative effects of MSCs in ACL reconstruction.

In this study, we successfully developed a physiologically-inspired ligament substitute with several key features. First, an innovative interfacial polyelectrolyte complexation (IPC) fiber harvesting system was employed for the fabrication of hierarchical structure to biomimetic the organization of native ligaments. Additionally, a mineralization gradient technique was introduced, replicating the complex anatomical ligament-bone junctions, and enhancing the integration with surrounding soft tissues. To ensure its clinical viability, viscoelastic analysis confirmed that the transplant possessed mechanical behaviors comparable to native ligaments, guaranteeing its durability and capacity to withstand biomechanical demands post-implantation. Its capability to load with growth factors and stem cells further enhanced its regenerative potential, creating opportunities for synergistic therapeutics in ligament tissue regeneration. The in vivo study showcased the superior tissue integration capabilities of this graft compared to autograft and commercial artificial ligament. Remarkably, when incorporated with growth factors and stem cell therapy, the substitute not only demonstrated ligamentization within the intra-articular graft but also induced osteogenesis at the bone insertion site, ultimately leading to promising mechanical performance. After 10-week transplantation, its biomechanical performance approached to that of the autograph group, demonstrating its potential for successful ligament reconstruction. Overall, the synergistic combinations of the biomimetic structure, gradient mineralization, and favorable biomechanical behavior positioned this bioinspired ligamentous substitute as a highly attractive candidate for further clinical considerations. The contributions arising from this study hold the potential to advance the domain of ligament tissue engineering forward and establish efficacious strategies for creating functional ligament tissue.

5.1 Set-up of IPC spinning instrument

A customized instrument was designed for fiber collection in accordance with the methods described in the previous study (Supplementary Fig. S1A) ³⁴. The primary component was a 75 cm x 33 cm x 32 cm modified metal cabinet, offering ample sealing to minimize contamination, while allowing real-time observation of the spinning process through a transparent front panel. The device was equipped with a motor featuring a speed control mechanism, meticulously configured to facilitate a consistent and uninterrupted spinning process (Oriental Motor, Japan). The speed control unit managed all speed control gear motors of the apparatus to regulate the respective movements for each step, including fiber production and spinning(3-45mm/s), fusion (2.5-40r/min), and collection (0.25-40r/min). The fiber bundle collector was assembled with a circular plate and stainless-steel rods. To maintain an appropriate humidity environment conducive for the IPC reaction, we housed a humidifier and hygrometer inside the metal chamber. In addition, a twisting device was self-designed to facilitate the fiber twisting process (Supplementary Fig. S1B). It included an acrylic plate, a metal shaft, and copper plates to secure the fiber bundles. The acrylic backboard could be adjusted to accommodate varying fiber lengths during weaving, ranging from 20 to 100 mm in length.

5.2 Fabrication and characterization of ligament-like scaffolds

The IPC spinning process was conducted according to the methodology outlined in the previous study ³⁵. In brief, 2% (w/v) Pectin (Taiwan Gum, Arabic) and 2% (w/v) Poly-D-lysine hydrobromide (Mw = 84kDa, Alamanda-Polymers, USA) were separately dissolved in deionized (DI) water to prepare the polyanion and polycation solutions. Five pairs of polyanion-polycation solutions were placed onto the loading plate to form a core-sheath structure. Fibers were produced through a controlled fiber formation platform, allowing the polyanion and polycation solutions to generate an interface in between. Once fibers were formed and reached 5 cm in length, the loading plate was rotated to create an intersection point for fiber fusion into a primary fiber bundle. Pectin solution was used again as fusion solution and applied to ensure the merging fibers properly successfully fused. Secondary fiber bundles were gathered using a specifically-designed fiber bundle collector. The final diameter of the secondary fiber bundle was determined by the number of collection repetitions. To create the ligament-like scaffold, five of these secondary fiber bundles were stacked together and then arranged on a fiber-twisting device. The dimensions of the resulting scaffold were adjusted according to the requirement. For material characterization and in vitro testing, the scaffold was prepared with a length of 20 mm and a diameter of 0.5 mm.

After configuring the ligament-like scaffold, a crosslinking solution containing 5% (w/v) N-(3-Dimethylaminopropyl)-N’-ethylcarbodiimide (EDC; Sigma-Aldrich, USA) and 2% (w/v) N-Hydroxysuccinimide (NHS; Fluka, Switzerland) in a 0.1M MES buffer (pH = 6.5, 0.3 M NaCl) was used for a preliminary crosslinking step. This was to ensure initial mechanical stability of the scaffold. The crosslinked scaffold was further immersed in a fresh crosslinking solution for 4 hours, followed by rinsing with DI water to remove residual crosslinking agents. The modified chemical bonds were analyzed using Fourier-transform infrared spectroscopy (Bruker, Vertex 80v).

Scanning electron microscopy (SEM) was utilized to observe the morphology and structure of the ligament-like scaffold. High-resolution micrographs were taken using a thermal field emission scanning electron microscope (JEOL, JSM-7610F) operating at 15keV and under a vacuum of 9.6×10^-5 Pa. For overall structure observation, the ligament-like scaffold underwent lyophilization, while for morphology analysis during the IPC spinning process, sample were entailed sequential dehydration procedures and critical point drying (CPD). Prior to SEM imaging, specimens were coated with gold using an auto fine coater (JOEL, JEC-3000FC). Image J analysis was then used to quantify the porosity of the samples from the captured micrographs.

To visualize the distribution of pectin and PDL in individual IPC fibers, pectin was labeled with Fluorescent Dye 645-A Amine (Abnova, Taiwan) via an EDC/NHS crosslinking process. Concurrently, PDL was labeled with Fluorescein Isothiocyanate (FITC, Fluka, Switzerland). The fluorescently-labeled IPC fiber was then created, embedded in an optimum cutting temperature compound (OCT; Sigma), and sectioned into 15 µm-thin slices using a Cryostat Microtome (Leica, CM1950). The OCT was subsequently removed with methanol. Finally, the fiber cross-sections were examined using confocal microscopy (ZEISS LSM 800 & Axio Observer Z1 TIRF).

5.3 In situ deposition and characterization of HAp gradient

HAp gradient was synthesized using a wet chemical synthesis method. The HAp coating solution was prepared by dissolving 0.5M Ca(OH)₂ (Fluka, Switzerland) and 0.3M (NH₄)₂HPO₄ (Showa Chemical, Japan) separately in DI water at 37°C. Equal volume of (NH₄)₂HPO₄ solution was slowly added to the Ca(OH)₂ solution at 37°C under stirring at 300 rpm. The edge of scaffold was immediately immersed in the HAp coating solution. For material characterization and in vitro study, each end was immersed to a depth of 0.5 cm. The scaffold was maintained in the HAp solution at 37°C for 12 hours to allow HAp deposition. When the solution was evaporated, it resulted in the formation of graded mineralization. The chemical equation is as follows:

$$10Ca{\left(OH\right)}_{2}+{\left({NH}_{4}\right)}_{2}\bullet HP{O}_{4}\to {Ca}_{10}{\left({PO}_{4}\right)}_{6}{\left(OH\right)}_{2}+18{H}_{2}O+12{NH}_{3}$$

The crystalline structure of HAp was identified using X-ray diffraction (Bruker, D2 Phaser), and its characteristic diffraction pattern was cross-referenced with the Joint Committee on Powder Diffraction Standards (JCPDS) No. 09-0432 (HAp) data card. The morphology of HAp nanocrystalline on the scaffold surface was visualized with a thermal field emission scanning electron microscope (JEOL, JSM-7610F), and the elemental composition, including the crucial calcium and phosphorus contents, was analyzed using energy-dispersive X-ray spectroscopy (EDS). The mineralization gradient on the scaffold was plotted using X-ray Photoelectron Spectroscopy (ULVAC, PHI PHI 5000 Versaprobe II), which provided insights into the calcium to carbon ratio.

5.4 PCN synthesis and CTGF loading

A 0.6 mg/mL chitosan solution (Sigma-Aldrich, USA) and a 1.2 mg/mL heparan sulfate solution (YickVic, Hong Kong) were prepared in a 0.1 M acetic acid solution at pH 5.0. Both solutions were purified using 0.22 µm MCE syringe filters. The purified chitosan solution was mixed with the heparan sulfate solution at a 1:6 volume ratio. The reaction proceeded for 3 hours with stirring at 800 rpm in the dark. Afterward, the resulting polyelectrolyte complex nanoparticles (PCNs) were dialyzed against DI water using an 8000 Da MWCO membrane for 24 hours. The dialyzed PCN was then lyophilized and stored for subsequent CTGF conjugation. CTGF and PCNs were mixed in DI water at a mass ratio of 100 ng/mg (CTGF/PCNs). The mixture was vortexed to ensure thorough mixing and allowed to react for 30 minutes. Subsequently, the product was subjected to lyophilization to remove water and preserve it for future use.

5.5 Coating of collagen sheath

SurgiAid® collagen (Maxigen Biotech Inc., Taiwan) was dissolved in a 1.5% (v/v) acetic acid solution and stirred overnight to prepare a 2% (w/v) collagen solution. A V-shaped PDMS mold was created using SYLGARD™ 184, and the scaffold was coated with 40ul of collagen solution containing 2% (w/v) EDC and 200ng CTGF. The coated scaffold was air-dried in a laminar flow hood to stabilize the collagen coating.

5.6 Mechanical properties analyses

The mechanical properties of the ligament-like scaffold were extensively characterized. Prior to each test, the samples were kept hydrated by storing them in PBS and the diameter was ascertained using a Vertical Fluorescence Microscope (Zeiss, Axiostar scope. A1) and analyzed using Image J. For the preparation of native tissue, porcine ACL tissues were prepared from fresh porcine knees. Femur and tibia regions were sliced into 1cm-thin rectangular planks for secure fastening during testing. Mechanical testing was conducted using a Universal Tester (Shimadzu, Autograph AGS-X) along with the software TRAPEZIUM LITE X for parameter calculations. The tensile tests assessed the strength, strain at break, and Young's modulus at a constant strain rate of 10%/min. For creep tests, samples were initially subjected to stress of 6 MPa at a stress rate of 30 MPa/min, which was held constant for 300 seconds. Then, the stress was slowly decreased to 0 MPa at a stress rate of 1.2 MPa/min. The initial strain (𝜺₀, %), peak strain (𝜺_Max, %), permanent strain (𝜺_min, %) were subsequently determined. Following these measurements, the recovery ratio and creep deformation were calculated using the following formulas:

$$Recovery ratio=\frac{{\epsilon }_{Max}-{\epsilon }_{Min}}{{\epsilon }_{Max}}\times 100\%$$

$$Creep deformation={\epsilon }_{0}-{\epsilon }_{Max}$$

For stress relaxation, samples were subjected to a deformation of 10% strain at a strain rate of 10%/min, with this deformation kept constant for 1000 seconds. The stress values were normalized by the initial stress when the strain reached 10%. For cyclic loading-unloading tests, the process involved a loading rate of 6MPa and 100 cycles of unloading at a constant stress rate of 30MPa/min. Parameters including maximum strain, minimum strain, loading modulus, and unloading modulus were calculated for each cycle.

5.7 Degradation kinetics and CTGF release behavior

Lyophilized ligament-like scaffolds and collagen-coated ligament-like scaffolds were weighed and divided into 30 mg portions per 2 mL tube. These samples were then incubated in either 1.5 mL PBS or 1.5 mL PBS containing 50 IU/mL type I collagenase (Gibco, USA). The degradation buffers were changed every other day to maintain the enzymatic activity. At different time point, the samples were collected and washed with DI to remove residual buffer, following by lyophilization to measure the dry weight of materials.

To evaluate the CTGF release kinetics, two different methodologies were tested with each sample consisting of 200 ng of CTGF in a 500 µl release buffer containing 0.1% (w/v) BSA in PBS. The first test entailed the integration of CTGF by premixing it with pectin solution prior to scaffold fabrication. This approach targeted the achievement of a core-sheath pattern and uniform encapsulation of CTGF within the scaffold. The second test involved the loading of CTGF post-scaffold synthesis as kind of collagen sheath coating. Specifically, this approach was divided into three conditions: CTGF mixed with uncrosslinked collagen, CTGF mixed with 2% EDC-crosslinked collagen, and CTGF bound to PCNs and then mixed with 2% EDC crosslinked collagen. Subsequently, CTGF release was monitored and samples were collected at predetermined time intervals. Quantification was performed using a CTGF ELISA kit (BioVendor, Czech Republic).

5.8 In vitro cell culture

The human mesenchymal stem cells (hMSCs) utilized in the in vitro study were acquired from the Bioresource Collection and Research Center, Taiwan. The culture medium for MSCs consisted of Gibco™ MEM non-essential amino acids (MEM-NEAA; Gibco, USA) supplemented with 1.5 g/L sodium bicarbonate (Sigma-Aldrich, USA), 1.0 mM sodium pyruvate (Gibco, USA), 10% fetal bovine serum (Gibco, USA), 4 ng/mL Human-bFGF (Gibco, USA), and 100 U/mL Penicillin/Streptomycin (Gibco, USA). Cell cultures were maintained in a controlled environment at 37°C with 5% CO₂. The culture medium was refreshed every 3 days to ensure optimal cell growth and viability. For the subsequent experiments, MSCs at passages 3–5 were utilized.

5.9 Biocompatibility of ligament-like scaffold

Biocompatibility of the scaffolds was tested using an LDH Assay. 2×10⁴ MSCs were co-cultured with sterilized scaffolds in 48-well culture plate and cytotoxicity was evaluated at 1, 3, and 5 days. The LDH assay was conducted according to the manufacture’s guidelines. Absorbance at 490 nm was measured using a microplate reader (Molecular Devices, SpectraMaxPlus384). Cytotoxicity was determined based on a formula by comparing the absorbance values from the sample and negative control against the positive control, Triton X-100-induced cell lysis. The cytotoxicity percentage was calculated as follows:

$$Cytotoxicity \left(\%\right)=\frac{{OD}_{Sample}-{OD}_{Negative}}{{OD}_{Positive}-{OD}_{Negative}}\times 100\%$$

where OD represents the absorbance obtained from the samples, negative control, and positive control, respectively.

5.10 Cell adhesion and proliferation on ligament-like scaffold

To validate the suitability of the ligament-like scaffold as a cell carrier for stem cell therapy, evaluation of cell adhesion and proliferation on the ligament-like scaffold was performed using a Cell Counting Kit 8 (WST-8/CCK8 Assay; Abcam, USA). In brief, scaffolds were positioned in V-shaped PDMS molds within a 6-well culture plate and seeded with 2x10⁵ MSCs in 0.2 mL of medium. Due to the inherent hydrophobicity of PDMS and the effects of gravity, MSCs were able to adhere and accumulate on the scaffold. After a 24-hour period, cell-laden scaffolds were transferred to a fresh 12-well plate to facilitate cell growth. Cell count quantification was achieved by measuring absorbance at 450 nm after scaffold incubation with CCK8 solution at designated time points. The number of cells in the samples was calculated using interpolation from a standard curve, generated under identical conditions with the same cell passages.

5.11 Live/Dead assay

To corroborate the quantified results and observe cell morphology, a LIVE/DEAD assay was employed after an auto-fluorescence quenching process via Sudan Black-B (SBB, Abcam, USA). The samples were initially immersed in a quenching buffer of 0.1% SBB in 70% ethanol overnight at 37°C followed by washing thrice with 70% ethanol and PBS. MSCs were then seeded onto these scaffolds and then placed on a PDMS mold for 24 hours for cell attachment. Next, scaffolds were transferred to a new 12-well culture plate and rinsed thrice with PBS to discard unattached cells. Subsequently, Live/Dead staining (Invitrogen, USA) was performed 7 day after cell seeding, with live cells marked by 2 µM calcein AM (Green) and dead cells with 4 µM Ethidium Homodimer-1 (Red) in DPBS for 30 minutes at room temperature. These stained cells were visualized using fluorescence inverted microscopy (Axiovert 40 CFL, Carl Zeiss).

5.12 Animal and surgical procedures of ACL reconstruction model

The experimental protocols were approved by the Institutional Animal Care and Use Committee (IACUC) at Laboratory Animals Center in National Tsing Hua University, Taiwan (IACUC Protocol No. 111049). The in vivo model utilized 3kg female New Zealand White Rabbits, sourced from Livestock Research Institute, Taiwan. Prior to surgery, morphine (2-5mg/kg) was administered for analgesia followed by 5mg/kg xylazine and 28.4mg/kg pentobarbital for anesthesia. The surgical site on the right knee was shaved and disinfected. After skin incision, a lateral parapatellar arthrotomy was utilized to expose the ACL, which was subsequently segmentally excised using a surgical blade. Bone tunnels in the tibia and femur were established using a 3.0-mm drill. The size of the proposed graft is 70 mm in length and 3 mm in diameter. The graft was passed proximally through the femoral tunnel and distally through the tibial tunnel, followed by fixation to the femoral and tibial periosteum using 3 − 0 PDS sutures in figure-of-eight fashion. The joint capsule and skin were then repaired using 4 − 0 PDS sutures and 4 − 0 nylon sutures respectively. Postoperatively, cefazolin (0.5 mg/kg) was administered intramuscularly, and the surgical site was managed with antibiotic ointment and sterile dressings. A plaster cast was applied to immobilize the hindlimb.

Rabbit bone marrow derived MSCs, procured from Cell Biologics (USA), were cultivated in Mesenchymal Stem Cell Medium (ScienCell, USA) at 37°C with 5% CO2. The culture medium was replaced every other day to maintain cell growth and vitality. MSCs from passages 3–5 were employed in the ensuing experiments.

The control group consisted of autologous grafts and commercial grafts (LARS), denoted as Autograft and LARS, respectively. The autologous graft (70 mm in length and 3 mm in diameter) was harvested from the Achilles tendon of the same hindlimb, and was prepared and rinsed with normal saline before ACL reconstruction. The LARS graft was reformed into grafts measuring 70 mm in length and 3 mm in diameter. The experimental groups included: a ligament-like scaffold with collagen coating and HAp deposition (Scaffold), the same scaffold incorporated with 200ng CTGF (Scaffold/CTGF), and the scaffold incorporated with 200ng CTGF and 2×10⁶ rabbit MSCs, (Scaffold/CTGF/MSCs). For in vivo studies, the grafts in all groups had a length of 70 mm and a diameter of 3 mm. A description of the experimental setup in more details was provided in Supplementary Table S1. The in vivo assessments comprised of histological and immunohistochemical analyses at 4- and 10-weeks after implantation, with the addition of micro-CT evaluations and biomechanical testing performed at the 10-week point post-surgery (Supplementary Fig. S2).

5.13 Osseointegration analysis by micro-CT

The knee joint was subjected to a 3-day formalin fixation prior to scanning using a 3D micro-CT system (SkyScan 1276, Bruker). The scanning process employed an isotropic resolution of 36 um in thickness. CT Analyzer software (v 1.20.8, Bruker) was utilized for subsequent analyses, specifically for evaluating the characteristics of the bone tunnel. For this purpose, a cylindrical volume of interest (VOI) with a width of 4.0 mm was selected to assess the 3.0-mm drilled bone tunnel. Within the VOI, various parameters such as bone mineral density (BMD), bone volume fraction (BV/TV, %), and trabecular number (Tb.N, 1/mm) were quantified. Additionally, the area of the bone tunnel (measured in mm²) was determined at a depth of 5 mm from the joint surface.

5.14 Histological and immunohistochemical analyses

In the histological analysis, H&E and Masson’s Trichrome staining techniques were employed to visualize tissue repair in both the ligament tissue and bone insertion regions. After micro-CT analysis, decalcification of samples was carried out using Leica Rapid Decalcifier II (Leica, Germany) for a week, with replacement of the solution every two days. After decalcification, the surrounding soft tissue was excised, and the femur-ligament-tibia complex was segmented into small portions. This included longitudinal sections for observing the intra-articular graft and transverse sections for bone insertion site examination.

Post decalcification, the tissue sections underwent a manual dehydration process comprising serial immersions in graded ethanol and xylene solutions, then embedded in paraffin using an Embedding Workstation (Thermal Fisher Scientific, HistoStar). The process encompassed stages in 70%, 80%, and 95% ethanol for specified durations followed by two rounds of each sample in 100% ethanol and xylene, and an overnight incubation in 65℃ paraffin. Thin sections of 7 µm thickness were obtained using a rotary microtome (Thermal Fisher Scientific, HM 315) after paraffin embedding. Hematoxylin and Eosin (H&E; Abcam, USA) staining was employed as per Abcam's protocol to elucidate cell nuclei (hematoxylin) and cytoplasm (eosin), enabling an evaluation of tissue morphology and cell organization. Masson's Trichrome staining (Abcam, USA), also guided by Abcam's instructions, was used to stain cell nuclei with Weigert's iron hematoxylin, muscle fibers and collagen with Biebrich Scarlet-Acid Fuchsin, differentiating with a phosphomolybdic/phosphotungstic acid solution, and counterstaining collagen fibers with Aniline Blue. This staining technique rendered details about tissue fibrosis and collagen deposition.

For immunohistochemistry (IHC), sections were rehydrated and treated with proteinase K for antigen retrieval at 37°C for 20 min followed by a 10-min hydrogen peroxide block and a 10-min protein block using an Abcam ABC IHC kit. When using primary antibodies sourced from rabbits, a rabbit-to-rabbit blocking reagent (ScyTek, USA) was applied for 2 hours at room temperature to prevent non-specific binding. Histological sections were then incubated with the primary antibody overnight at 4°C followed by a secondary antibody for 1 hour. For rabbit primary antibodies, biotinylated goat anti-polyvalent (Abcam, USA) was added for 10 min. The chromogenic reaction involved a 10-min streptavidin peroxidase and a 2-min DAB solution. Counterstaining was done using hematoxylin. The antibody dilution factors used were as follows: primary rabbit anti- Tenomodulin antibody, 1:100 (Bioss, USA); primary mouse anti- Tenascin C antibody, 1:70 (Cloud-clone, USA); primary guinea pig anti-osteocalcin antibody, 1:40 (Cloud-clone, USA); secondary biotinylated anti-mouse antibody, 1:100 (Arigo, Taiwan); and secondary biotinylated anti-guinea pig antibody, 1:1000 (Genetex, USA).

5.15 Biomechanical analysis

Biomechanical investigations were conducted using a Universal Tester (Shimadzu, AGS-X). Knee specimens were harvested and prepared as femur-graft-tibia complex immediately after sacrifice, with removal of other connective tissues except for the ACL replacement. The mechanical failure test was performed at an elongation rate of 1 mm/min to determine the ultimate tensile strength and generate a load-elongation curve. The breaking condition was recorded to evaluate the integration between the replacement and the surrounding tissue.

5.16 Statistical analysis

The data are presented as mean ± standard deviation (S.D.) of the mean. Statistical significance was determined using one-way analysis of variance (ANOVA) and T-test. The following notations were used to indicate the significance level: N.S. for no significant difference, * for p ≤ 0.05, ** for p ≤ 0.01, *** for p ≤ 0.001, and **** for p ≤ 0.0001.

Acknowledgement

We thank Dr. Andrew C.A. Wan. at Institute of Bioengineering and Nanotechnology, Singapore for technical guidance in IPC instrumentation. We also extend our appreciation to the Taiwan Mouse Clinic, Academia Sinica, and Taiwan Animal Consortium for their assistance in micro-CT scanning. We thank the Instrument Center at NTHU for their support with SEM, EDS, ESCA, XRD, and confocal microscopy. We also acknowledge Po-Yu Chen, Shang-Hsiu Hu, and Chieh-Cheng Huang from NTHU for their respective support with mechanical analysis, confocal microscopy and cryostat microtome. This research was funded by the Taiwan Ministry of Science and Technology under grants MOST 111-2628-B-007 -005 -MY3 and NSTC 112-2221-E-007 -037 -MY3.

Author contribution

Y.C.L., S.H.C, C.H.K, and T.W.W conceptualized the project and drafted the manuscript. Y.C.L. executed the majority of experiments, analyzed data, and assisted in animal studies. S.H.C devised the animal model, conducted surgeries and managed animal care. C.H.K contributed to manuscript writing. Discussions on vital project facets were enriched by C.H.K, W.Y.H, and H.X.C. T.W.W provided overall supervision, guiding experimental design, data interpretation, and manuscript refinement.

Competing Interest

There is no conflict of interests.

Bone, U. & Initiative, J. The burden of musculoskeletal diseases in the United States (BMUS). Rosemont, IL: OrgtoPaedic Research Society (2014).
Simon, D. et al. The relationship between anterior cruciate ligament injury and osteoarthritis of the knee. Adv. Orthop. 2015, 928301 (2015).
Lim, W. L., Liau, L. L., Ng, M. H., Chowdhury, S. R. & Law, J. X. Current progress in tendon and ligament tissue engineering. Tissue Eng. Regen. Med. 16, 549-571 (2019).
Wan, A. C. et al. Fibers by interfacial polyelectrolyte complexation–processes, materials and applications. Mater. Today 19, 437-450 (2016).
Liao, I.-C., Wan, A. C., Yim, E. K. & Leong, K. W. Controlled release from fibers of polyelectrolyte complexes. J. Control. Release 104, 347-358 (2005).
Cutiongco, M. F. A., Tan, M. H., Ng, M. Y. K., Le Visage, C. & Yim, E. K. F. Composite pullulan–dextran polysaccharide scaffold with interfacial polyelectrolyte complexation fibers: A platform with enhanced cell interaction and spatial distribution. Acta Biomater. 10, 4410-4418 (2014).
Tonegawa, H. et al. Synthesis of enzymatically crosslinkable peptide‐poly (l‐lysine) conjugate and creation of bio‐inspired hybrid fibers. Macromol. Biosci. 4, 503-511 (2004).
Wan, A. C., Leong, M. F., Toh, J. K., Zheng, Y. & Ying, J. Y. Multicomponent fibers by multi‐interfacial polyelectrolyte complexation. Adv. Healthc. Mater. 1, 101-105 (2012).
Du, C., Narayanan, K., Leong, M. F. & Wan, A. C. Induced pluripotent stem cell-derived hepatocytes and endothelial cells in multi-component hydrogel fibers for liver tissue engineering. Biomaterials 35, 6006-6014 (2014).
Lacombe, J.-G. et al. Primary human ligament fibroblast adhesion and growth on 3d-printed scaffolds for tissue engineering applications. Surgeries 4, 196-211 (2023).
Hernandez, D. S., Schunk, H. C., Shankar, K. M., Rosales, A. M. & Suggs, L. J. Poly-d-lysine coated nanoparticles to identify pro-inflammatory macrophages. Nanoscale Adv. 2, 3849-3857 (2020).
Hu, Z. et al. PVA/pectin composite hydrogels inducing osteogenesis for bone regeneration. Mater. Today Bio. 16, 100431 (2022).
Docheva, D., Müller, S. A., Majewski, M. & Evans, C. H. Biologics for tendon repair. Adv. Drug Deliv. Rev. 84, 222-239 (2015).
Shen, H. et al. The effect of adipose-derived stem cell sheets and CTGF on early flexor tendon healing in a canine model. Sci. Rep. 8, 1-10 (2018).
Holbourn, K. P., Acharya, K. R. & Perbal, B. The CCN family of proteins: structure–function relationships. Trends Biochem. Sci. 33, 461-473 (2008).
Jiang, Q. et al. Canine ACL reconstruction with an injectable hydroxyapatite/collagen paste for accelerated healing of tendon-bone interface. Bioact. Mater. 20, 1-15 (2023).
Thirumalai, J. Introductory chapter: The testament of hydroxyapatite: New prospects in regenerative medicinal treatments. Hydroxyapatite: Advances in composite nanomaterials, biomedical applications and its technological facets, 3-14 (2018).
Wang, C. et al. Application of stem cell therapy for ACL graft regeneration. Stem Cells Int. 2021 (2021).
Oe, K. et al. New strategies for anterior cruciate ligament partial rupture using bone marrow transplantation in rats. Stem Cells Dev. 20, 671-679 (2011).
Soon, M. Y., Hassan, A., Hui, J. H., Goh, J. C. & Lee, E. An analysis of soft tissue allograft anterior cruciate ligament reconstruction in a rabbit model: a short-term study of the use of mesenchymal stem cells to enhance tendon osteointegration. Am. J. Sports Med. 35, 962-971 (2007).
Nau, T. & Teuschl, A. Regeneration of the anterior cruciate ligament: Current strategies in tissue engineering. World J. Orthop. 6, 127-136 (2015).
Bernard, L., Freche, M., Lacout, J. & Biscans, B. Preparation of hydroxyapatite by neutralization at low temperature—influence of purity of the raw material. Powder Technol. 103, 19-25 (1999).
Iriuchishima, T. et al. Evaluation of ACL mid-substance cross-sectional area for reconstructed autograft selection. Knee Surg. Sport Tr. A. 22, 207-213 (2014).
Marieswaran, M., Jain, I., Garg, B., Sharma, V. & Kalyanasundaram, D. A review on biomechanics of anterior cruciate ligament and materials for reconstruction. Appl. Bionics Biomech. 2018 (2018).
Genin, G. M. et al. Functional grading of mineral and collagen in the attachment of tendon to bone. Biophys. J 97, 976-985 (2009).
Hahn, J. et al. Viscoelastic behavior of embroidered scaffolds for ACL tissue engineering made of PLA and P(LA-CL) after in vitro degradation. Int. J. Mol. Sci. 20 (2019).
Wang, F. et al. Multi‐Colored Fibers by Self‐Assembly of DNA, Histone Proteins, and Cationic Conjugated Polymers. Angewandte Chemie 126, 434-438 (2014).
Silva, M., Ferreira, F. N., Alves, N. M. & Paiva, M. C. Biodegradable polymer nanocomposites for ligament/tendon tissue engineering. J. Nanobiotechnology 18, 1-33 (2020).
Tulloch, S. J. et al. Primary ACL reconstruction using the LARS device is associated with a high failure rate at minimum of 6-year follow-up. Knee Surg. Sport Tr. A. 27, 3626-3632 (2019).
Zhu, G. et al. Bone physiological microenvironment and healing mechanism: Basis for future bone-tissue engineering scaffolds. Bioact. Mater. 6, 4110-4140 (2021).
Petrigliano, F. A., McAllister, D. R. & Wu, B. M. Tissue engineering for anterior cruciate ligament reconstruction: a review of current strategies. Arthroscopy 22, 441-451 (2006).
Li, Y., Wu, T. & Liu, S. Identification and distinction of tenocytes and tendon-derived stem cells. Front. Cell Dev. Biol. 9, 629515 (2021).
Arnott, J. A. et al. The role of connective tissue growth factor (CTGF/CCN2) in skeletogenesis. Crit. Rev. Eukaryot. 21 (2011).
Leong, M. F. et al. Patterned prevascularised tissue constructs by assembly of polyelectrolyte hydrogel fibres. Nat. Commun. 4, 2353 (2013).
Wan, A. C., Liao, I.-C., Yim, E. K. & Leong, K. W. Mechanism of fiber formation by interfacial polyelectrolyte complexation. Macromolecules 37, 7019-7025 (2004).

There is NO Competing Interest.

IPCSpinningProcessJoshuapaper.mp4
Supplementary Video S1. IPC spinning process.
Supplementaryinformation.docx
Supplementary information

Download PDF

Version 1

posted

You are reading this latest preprint version

Assembly of Interfacial Polyelectrolyte Complexation Fiber Bundles with Mineralization Gradient for Physiologically-inspired Ligament Regeneration

Status:

Version 1

Abstract

Figures

The highlights of this research are

1. Introduction

2. Results

2.1 Morphological Observations of Ligament-like Scaffold

2.2 Characterization of Enthesis-like HAp Gradient

2.3 Evaluation of the Strengthened Ligament-like Scaffold

2.4 Measurement of Tissue-like Viscoelasticity

2.5 Quantification of Recombinant CTGF Release and Scaffold Degradation

2.6 Scaffold Cytocompatibility and MSC Proliferation

2.7 In vivo ACL Reconstruction in Rabbit Model

3. Discussion

4. Conclusion

5. Methods

5.1 Set-up of IPC spinning instrument

5.2 Fabrication and characterization of ligament-like scaffolds

5.3 In situ deposition and characterization of HAp gradient

5.4 PCN synthesis and CTGF loading

5.5 Coating of collagen sheath

5.6 Mechanical properties analyses

5.7 Degradation kinetics and CTGF release behavior

5.8 In vitro cell culture

5.9 Biocompatibility of ligament-like scaffold

5.10 Cell adhesion and proliferation on ligament-like scaffold

5.11 Live/Dead assay

5.12 Animal and surgical procedures of ACL reconstruction model

5.13 Osseointegration analysis by micro-CT

5.14 Histological and immunohistochemical analyses

5.15 Biomechanical analysis

5.16 Statistical analysis

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1