Scaffolds drive meniscus tissue engineering

Abstract

Meniscus injury is a common sports injury. The removal of the injured meniscus predisposes the development of osteoarthritis. Tissue engineering provides a promising approach for the treatment of meniscus injury. Over the past few decades, researchers have put a lot of effort in developing various kinds of scaffolds for preventing osteoarthritis and relieving clinical symptoms. The scaffolds for meniscus tissue engineering (MTE) can be categorized into four classes: hydrogels, three-dimensional (3D) porous polymeric scaffolds, extracellular matrix (ECM) macromolecule-based 3D matrices, and tissue-derived platforms. Although the expected efficacies in tissue integration and chondro-protective function have not yet emerged, several scaffolds have been exploited to provide new substitutes for native meniscus, suggesting their great potential in promoting tissue engineered meniscus (TEM). In the current review, we provide and discuss evidences on different scaffolds in order to make recommendations for the further development of MTE.

---

Zheng-Zheng Zhang

Zheng-Zheng Zhang received his BS degree from Tianjin Medical University in 2009. He is now studying for his PhD degree under the supervision of Prof. Jia-Kuo Yu at the Institute of Sports Medicine of Peking University Third Hospital. He has published more than 10 academic papers. His research interests focus on tissue-engineered meniscus and related tissue regeneration in sports medicine.

Dong Jiang

Dong Jiang is currently an associate chief physician and assistant professor at the Institute of Sports Medicine of Peking University Third Hospital. He received his BS degree from Peking University in 2004 and obtained his MD degree from Peking University in 2009 under the supervision of Prof. Jia-Kuo Yu. He has published more than 20 academic papers and applied 2 Chinese patents. His research focuses on (1) tissue-engineered meniscus regeneration, (2) meniscus allograft transplantation in animal models and clinical outcomes, and (3) cartilage and ligamentous tissue regeneration using mesenchymal stem cells (MSCs) deriving from different sources.

Shao-Jie Wang

Shao-Jie Wang is now an associate chief physician in the Department of Joint Surgery, Zhongshan Hospital of Xiamen University. He is now studying for his PhD degree under the supervision of Prof. Jia-Kuo Yu at the Institute of Sports Medicine of Peking University Third Hospital. His research interests include cartilage repair and related tissue regeneration using MSCs of different sources.

Yan-Song Qi

Yan-Song Qi received his MS degree from Inner Mongolia Medical University in 2013. He is now studying for his PhD degree under the supervision of Prof. Jia-Kuo Yu at the Institute of Sports Medicine of Peking University Third Hospital. His research interests are conducting comparative studies about the clinical effects of different operative methods on anterior cruciate ligament and posterior cruciate ligament reconstructions, and related arthroscopic surgery clinical research.

Jian-Xun Ding

Jian-Xun Ding is currently an associate professor at Changchun Institute of Applied Chemistry (CIAC), Chinese Academy of Sciences (CAS). He received his BS degree from the University of Science and Technology of China (USTC) in 2007, and obtained his PhD degree at CIAC, CAS in 2013 under the supervision of Prof. Xuesi Chen. He has published more than 90 academic papers with an h-index of 20 and applied over 50 Chinese invention patents. His research focuses on (1) exploitations of stimuli-responsive polymers as the matrices of smart nanocarriers, (2) developments of cell-related nanoscale platforms for controlled drug delivery, and (3) regenerative biomaterials.

Jia-Kuo Yu

Jia-Kuo Yu is currently a professor and vice director at the Institute of Sports Medicine of Peking University Third Hospital. He is a member of the standing committee of the Chinese Association of Sports Medicine, member of the standing committee of the Chinese Society of Sports Medicine of the Chinese Medical Association, committee member of the Tissue Repair and Regeneration Society of the Chinese Medical Association. He has published more than 100 academic papers. His research focuses on (1) tissue-engineered meniscus and cartilage using MSCs derived from different sources, and (2) designing, proving, and manufacturing of personally compatible total knee prosthesis.

Xue-Si Chen

Xue-Si Chen received his PhD degree at Waseda University, Japan, in 1997, and completed his post-doctoral fellowship at the University of Pennsylvania, USA, in 1999. He has been a full professor at CIAC, CAS since 1999. He has published over 500 articles in academic journals, which have been cited more than 9000 times until now. In addition, he has applied over 250 Chinese patents and more than 120 have been authorized. His research interests focus on preparations and biomedical applications of biodegradable polymers, mainly focused on polyethers, polyesters, polypeptides, polycarbonates, and their copolymers.

---

1 Introduction

Meniscal tears are the most common knee injuries in all age groups.¹ The repair techniques, including suture anchor, synovial flap, and fibrin clot implantation, are clinically effective in the management of acute tears in the peripheral vascular part of the meniscus, but not the central avascular area.² Either partial or total meniscectomy is used to treat meniscus injury in the avascular region. Until 2014, the number of peoples requiring the above therapies was more than 1.5 million across the United States and Europe annually.³ However, the postoperative abnormal physiological stress on articular cartilage always leads to knee damage and osteoarthritis.

Given the vital role of meniscus in knee stability and homeostasis, some advanced alternative approaches are required to facilitate meniscectomy or traditional repair. Tissue engineering, which aims to regenerate the damaged meniscus, offers a potential solution strategy. Tissue engineering includes three main variables: scaffolds, seed cells, and growth factors or exogenous stimuli, which can be implanted into the injured tissue individually or simultaneously.⁴ In meniscus tissue engineering (MTE), seed cells, including mesenchymal stem cells (MSCs),⁵ meniscal fibrochondrocytes,⁶ and chondrocytes,⁷ etc., play an important role in the regeneration of defect successfully. Furthermore, different grow factors⁵ and mechanical stimuli⁸ have also been demonstrated to promote the efficacy of meniscus repair. Above all, scaffolds, a three-dimensional (3D) porous platform, allow seed cells to attach and migrate to regenerate new tissue in vitro and/or in vivo.⁹ Numerous materials have been investigated to construct meniscal implants, suggesting their great prospect in upgrading tissue engineered meniscus (TEM) for relieving symptoms after partial meniscectomy or preventing the development of osteoarthritis. Grossly, the scaffolds for MTE can be categorized into four classes: hydrogels, 3D porous polymeric scaffolds, extracellular matrix (ECM) component-derived systems, and tissue-originated materials (Fig. 1).

Fig. 1 MTE strategies include 3 main variables: scaffolds, seed cells, and biomechanical or biochemical stimuli, which can be implanted into animal model for regeneration or replacement the meniscus alone or in combination. Particularly, scaffolds allow seed cells to attach and migrate to form new tissue in vitro or in vivo, which promote TEM.

The aim of this review is to summarize the development of different scaffolds for MTE. Within this topic, the emphases are placed on advances and inadequacies of scaffolds combined with their applications in preclinical models in order to promote the realization of meniscal regeneration in clinic.

2 Scaffold-promoted MTE

2.1 Hydrogels

Hydrogels, derived from natural or synthetic polymers, such as agarose, alginate, chitosan, and poly(N-isopropyl acrylamide) (PNIPAAm), are hydrophilic colloids capable of holding large amounts of water. The physical properties of hydrogels are largely influenced by their water contents.¹⁰ The most prominent feature of hydrogels is versatile. They can be fabricated by crosslinking various synthetic and natural materials through various methods,¹¹ and patterned with cells¹² and growth factors.¹³ However, hydrogels may hamper the phenotypes of meniscus cells,¹⁴ and the mechanical properties of hydrogels are less easily manipulated (Fig. 2A–C).

Fig. 2 Hydrogel scaffolds. (A–C) Gross appearances of constructs: (A) agarose in Week 1, (B) agarose static in Week 7, and (C) agarose rotating wall bioreactor in Week 7;¹⁴ (E–L) photographs of intact implants or cross-sectional views of alginate meniscal constructions: (E and I) Week 0; (F and J) Week 2; (G and K) Week 4; (H and L) Week 8. Scale bars represent (E–H) 10 mm and (I–L) 2 mm, respectively.²¹

Many researchers have focused on using the versatile hydrogels to create more native microenvironment for seed cells. As a typical example, the hybrid alginate–chitosan–hyaluronan hydrogels conjugated with arginine–glycine–aspartic acid (RGD) tripeptide and cultured with articular chondrocytes showed higher collagen (Col) and glycosaminoglycan (GAG) contents.¹⁵ However, the further works were required to investigate these scaffolds for their co-culture with meniscus cells, which might be created by the spatial patterns of various cell types.^16,17

Another property of hydrogels is their ability to reversibly gel in response to the environmental factors, such as temperature, compressive loading, electric field, pH, ultrasound, or salt concentration.¹⁸ For this characteristic, an injectable hydrogel that solidifies in the body can be produced.¹⁹ Though the implantation process is mini-invasive, the main disadvantage is lacking mechanical properties after solidification. Some researchers have improved them by crosslinking methods, but these approaches also influence cellular metabolism.²⁰ Ballyns et al. combined the imaging capabilities of magnetic resonance imaging (MRI) or computed tomography (CT) with robotic printing to create meniscus-shaped model (Fig. 2D–L). The alginate scaffolds seeded with meniscus cells in high density created a high geometrical fidelity to the targeting shape. However, the tensile properties have not been reported.²¹ Following loaded with loading platens via a custom bioreactor, the engineered constructs had a positive effect on ECM and mechanical properties, e.g., compressive modulus.²²

As mentioned above, hydrogels represent a versatile category of scaffolds, while their unfavorable bioactivities, especially in promoting meniscus cell phenotype and ECM synthesis, and mechanical properties, e.g., in tension, limit their applications in total meniscus replacement. Angele et al. repaired a critical-size defect of rabbit medial meniscus using a hyaluronan–gelatin composite scaffold seeded with MSCs. Moreover, this study demonstrated that the pre-cultured implants integrated with the host tissue and most of them contained meniscus-like fibrocartilage, while the mechanical properties of neo-meniscus have not been investigated,²³ which play a vital role in the protection of articular cartilage.

2.2 3D porous scaffolds from synthetic polymers

In contrast to the hydrophilic hydrogels, the hydrophobic synthetic polymers have also been investigated for preparing meniscus scaffolds. Synthetic polymers are macromolecules that do not exist in the body, such as poly(L-lactide) (PLLA), polyglycolide (PGA), poly(lactide-co-glycolide) (PLGA), and poly(ε-caprolactone) (PCL). The 3D scaffolds have obvious advantage and disadvantage of superior mechanical properties and inferior bioactivities, respectively.²⁴

Polyurea–poly(L-lactide) (PU–PLLA) is widely applied as vascular prosthesis in cardiovascular surgery, which has demonstrated its characteristics of biocompatibility and biodegradability.²⁵ In MTE, the optimal implant was constructed with PU–PLLA at a weight ratio of 95 : 5 and reinforced with carbon fiber. Although the elastic modulus at 2% compression was 11 kPa and the Young's modulus was 42.5 MPa, its carbon component induced synovitis.²⁶ Another study have also indicated that a PU scaffold (i.e., Estane 5701 F) with improved properties comprising compression modulus at 20%, strain of 300 kPa, and porosity at 78% could not be used as a scaffold for articular cartilage, which was damaged especially in the tibial plateau (Fig. 3A).²⁷

Fig. 3 3D synthetic polymer scaffolds. (A) Scanning electron micrograph (SEM) micrograph of Estane implant.²⁷ (B) 3D PCL–PU scaffold.³⁰ (C) SEM microimage of aligned PCL scaffold produced via electrospinning planting MSCs on it after one day of culture. Scale bar: 50 mm.³⁴ (D) PCL and HYAFF® meniscus implant augmented with circumferential PLA fiber.³⁷ (E and F) Gross assessment of operated joints in sheep model after 12 months showing lower joint degeneration: (E) femoral condyle and (F) tibial plateau.³⁹ (G) Meniscus-shaped PGA scaffold and native rabbit meniscus.⁶ (H) Gross view of neo-meniscus retrieved at 36 weeks after transplantation with cell-seeded scaffold.⁶

Further refinements of PU follow because Estane is thought to potentially release carcinogenic compounds upon degradation.²⁸ A new scaffold of PCL–PU showed superior tissue in growth compared to Estane scaffold. However, it could not prevent the degradation of cartilage, which might be explained by its inferior compressive characteristics.²⁹ A 24 month follow-up study demonstrated that the PCL–PU implant was disintegrated by fragmentation with the lacking Col fiber orientation. Although its compression modulus was improved after tissue ingrowth, the damage to articular cartilage was also observed (Fig. 3B).³⁰

Since the load and compression of meniscus in vivo are high,³¹ the property of mechanical anisotropy is important for scaffolds. The application of a rotating collection platform during electrospinning facilitated the alignment of fibers in PCL/PLA scaffolds,^32,33 which exhibited a better improvement in tensile modulus. In addition, the characteristic of alignment was beneficial to the orientations of cells and ECM (Fig. 3C),^34,35 which demonstrated that the shear properties of fibrocartilage tissues could match or exceed native tissue benchmarks.³⁶ However, the scaffold fabricated using electrospinning was dense and difficult for cell and tissue ingrowth. Kon, Chiari, and colleagues developed a scaffold consisting of PCL and hyaluronan (HA) derivative (i.e., HYAFF®) to serve either as a partial or a total meniscus substitute in sheep model (Fig. 3D).^37–39 HYAFF® (Fidia Advanced Biololymerse, AbanoTerme, Italy) is a commercialized natural polymer obtained by the etherification of HA with different alcohols.⁴⁰ The PCL–HYAFF® scaffold was reinforced with a polyethylene terephthalate (PET) net for the partial meniscal substitute and with the circumferential PLA fibers in the total meniscus substitute. After 6 weeks, cartilage degeneration was more severe in the operated joints, though the implant showed excellent tissue ingrowth to the capsule, and the integration issue was also observed between the implant and the original meniscus.³⁷ In addition, after seeding autologous chondrocytes, the scaffolds were sutured to the capsule and the meniscal ligament, or fixed using a transtibial method for 4 months. The better implant appearance and integrity were observed in the group using former fixation, and lower joint degeneration was observed in the cell-seeded group with respect to the cell-free implant.³⁸ At 12 months of follow-up, the implant dislocation and wrinkling of scaffold in the posterior horn were also observed. Unfortunately, the damage to articular cartilage was not prevented, although less pronounced compared to meniscectomized control (Fig. 3E and F).³⁹ Recently, Mao et al. reported a 3D-printed PCL scaffold with spatially released human connective tissue growth factor and transforming growth factor-β3 (TGF-β3), which could regenerate into knee meniscus in sheep model. PCL was molten at 120 °C and dispensed following the 3D design of the internal microstructures of native meniscus, which could be optimal to approximate mechanical properties. Although there was no significant in tensile modulus between the native meniscus and regenerated construct, the long-term chondro-protective effect should also been investigated.⁵

PLLA, PGA, and their copolymer of PLGA are widely used in tissue engineering, and have gained the approval of US Food and Drug Administration (FDA) for several applications. Although PGA scaffold with a porosity of 95% demonstrated successful fibrochondrocytes seeding, it lacked sufficient mechanical properties, especially after culture for 7 weeks.¹⁴ The mechanical characteristics could be improved by physically bonding of PLGA onto both sides of PGA fiber mesh, but the scaffold without seeding cells demonstrated early partial degradation and inferior signs of meniscal tissue regeneration in a rabbit model, indicating the scaffold could not prevent articular cartilage damage (Fig. 3G and H).⁶ Another copolymer of poly(L-lactide-co-D-lactide) (PLDLA) had similar mechanical properties to PLLA but a superior degradation profile. After the addition of poly(caprolactone-triol) (PCL-T) to enhance compressive modulus, the scaffold was implanted into the medial knee meniscus of rabbit. However, the degradation of articular cartilage was investigated only by analysis of the amount of chondrocytes in the hyaline cartilage of femoral condyles, which demonstrated a higher amount in the implanted group compared to meniscectomized knee.⁴¹

Some other studies focus on the bioactivities of synthetic polymers. A matrix metalloproteinase-2 (MMP-2)-degradable peptide scaffold showed more biomimetic, but it has not yet been applied to MTE.⁴² Another study by comparing PLLA nanofiber and microfiber scaffolds reported that the former scaffolds increased the production of Col II, GAG, cartilage link protein, and aggrecan for the native components of ECM are nanoscale molecules.⁴³ Although synthetic polymers have been improved, the main disadvantage of the scaffolds is the inefficient construction of a functional matrix before scaffold degradation in vivo.

2.3 3D matrices based on ECM macromolecules

Integration with neighboring host tissue remains an issue to be tackled for 3D porous polymeric scaffolds. As far as biocompatibility, ECM scaffolds would logically constitute a native microenvironment for seed cells or host tissue.

ECM component scaffolds are platforms formed primarily from the component macromolecules of natural matrices, which include Col meniscus implants (CMI) (Fig. 4A), HA scaffolds, and so on. Col scaffolds are fabricated by nanofiber electrospinning, anisotropic deposition, and so on, and the mechanical properties are comparable to the scaffolds originated from synthetic polymers.^24,44 In combination with Col or gelatin, HA scaffolds were also widely used in MTE, which have positive effect on the healing of meniscal defect for their excellent biocompatibility and biodegradability.^45,46 Therefore, ECM scaffolds can provide a native microenvironment for seed cells, which suggest they are more biomimetic than hydrogels and 3D scaffolds from synthetic polymers.

Fig. 4 ECM component scaffolds. (A) Medial and lateral CMI®. (B) Arthroscopic second look at 18 months after grafting CMI implant (Menaflex®, ReGen Biologics, Inc., Hackensack, NJ, USA).⁵² (C) Histological evaluation of biopsy at 5 years after CMI implantation: (1) compact connective tissues below, and (2) looser connective tissues above vessels (arrows).⁵² (D) Arthroscopic look of CMI implant.⁵⁵ (E) Arthroscopic relook in 6.3 years after placement of Col meniscus implant. The defect is completely filled with new tissue (N), moreover the new and the native tissue is barely distinguishable (arrow).⁵⁵ (F) Masson's trichrome staining of biopsy at 6.3 years after CMI implant: fibrocartilaginous tissue that is meniscus-like in appearance. The original magnification is ×25.⁵⁵

Since meniscus cells rest in a network of Col and GAG molecules, ECM scaffolds compose of the above components would provide a native microenvironment for TEM. An early study showed that Col II–GAG matrix provided a better microenvironment for meniscus cell proliferation and GAG deposition, meanwhile less contraction compared to Col I–GAG matrix.⁴⁷ Another research has shown that aggrecan surface promoted more effective deposition of meniscus cell ECM than Col I surface.⁴⁸ In general, since not all ECM molecules are equally effective, different ECM component scaffolds show various bioactivities.^49,50 Further studies need to be investigated in them and their combination, i.e., a hybrid scaffold consisting of chitosan, HA, chondroitin-6-sulfate, Col I, and/or Col II macromolecules may even induce the redifferentiation of previously dedifferentiated meniscus cells.⁵¹

In clinic, although a multi-center clinical trial of CMI showed positive results in activity levels (Fig. 4B and C),^52,53 several drawbacks still exist, such as scaffold degradation, shape incongruence, and technique of suturing (Fig. 4D–F).^54,55 Moreover, these acellular scaffold have not completely migrated host cells to produce meniscus matrix.⁵⁴ Nowadays, CMI has been forbid in clinic by FDA.

2.4 Tissue-derived scaffolds

Similar to the favorable bioactivities of ECM components, the applications of tissue-derived scaffolds also exhibit advantages: providing a native microenvironment for cell seeding, proliferation, differentiation, migration, and ECM deposition. Tissue-derived 3D materials include decellularized ECM (dECM)⁵⁶ and other byproducts of living tissue, such as small intestinal submucosa (SIS).^57–59 Although they have high bioactivities, there are still some disadvantages: first, for the scaffolds derived from natural tissues, the supply is limited; moreover, the processing protocols, such as decellularization, affect the mechanical integrity of tissue-derived materials.

An early work found that the unseeded porcine SIS transplanted in canine meniscus defects prevented the development of erosion in articular cartilage.⁵⁷ Similar results have also been observed in a long-term study of SIS implantation, but the study reported that the SIS implantation showed inferior biomechanics property (Fig. 5A–C).⁵⁸ Moreover, the study demonstrated that the knee joints of goats that were implanted with unseeded porcine SIS developed cartilage degeneration combining with the regeneration of meniscus.⁶⁰ The above described issues highlight that this regenerated tissue is mechanically insufficient.

Fig. 5 Tissue-derived scaffolds. (A) SIS repair technique.⁵⁸ (B) Gross appearance of operated meniscus from dogs using SIS repair in 12 months after surgery.⁵⁸ (C) Histological appearance of operated meniscus from dogs in 12 months after surgery. (D) Gross inspection of acellular meniscus scaffold.⁵⁸ (E) Hematoxylin and eosin (H&E), (F) 4′,6-diamidino-2-phenylindole dihydrochloride (DAPI), and (G) Masson's trichrome staining: there is a visible abundance of Col remaining in the meniscus scaffold without cellular and nuclear contents. The original magnification is ×100.⁶¹

Another researches center on the creating appropriate pore sizes for scaffolds. Since cell infiltration is various through the depths of whole processed tissues (i.e., SIS)⁵⁹ and decellularized meniscus,⁵⁶ recent works have increased decellularized ovine meniscus porosity to a value of 80% in the outer meniscus as well as connected pore volume. Unfortunately, the process harmed their compressive properties (Fig. 5D–G).⁶¹

The decellularization process also induces the losses in GAG content. Recent study demonstrated that meniscus cells were successfully cultured within these scaffolds after the treatment of ovine meniscus with an enzymatic solution. In addition, the cultivation showed no reaction to the major histocompatibility complex class-1 (MHC-1) and major histocompatibility complex class-2 (MHC-2), and preservation of compressive stiffness. But more GAG loss and non-uniform cell distribution were observed in the re-seeded scaffold.⁶² Another study revealed that the treatment of human meniscus with 2% sodium dodecyl sulfate (SDS) finally maintained Col structure and mechanical properties after decellularization.⁶³ However, the use of SDS is associated with the loss of cellular DNA and GAG.⁶⁴

Another tissue-derived materials applied in MTE is Silk. Mandal et al. used silkfibroin obtained from Bombyxmori silkworm cocoons for MTE.^65,66 They designed a multilayered silk scaffold composing of three individual layers with different pore sizes (i.e., 500–600, 350–400, and 60–80 μm) and orientations. Although seed cells could be migrated though the large pores, and smaller pores improved the deposition of ECM, the mechanical properties of whole construct would have to be investigated to meet values comparable to the native human meniscus. Yan et al. demonstrated that the structural and mechanical properties of scaffold could be modified by altering initial silkfibroin concentrations. However, the porosity and interconnectivity of scaffold decreased.⁶⁷

2.5 Scaffoldless self-assembly of tissue

Although the support scaffolds described above display promising in mechanics, bioactivities, and logistics, another issues, e.g., potential toxicity, derived from degradation products and stress-shielding effects exerted by scaffolds, should also been considered.²⁴ Aside from using traditional scaffolds for MTE, cell self-assembly has begun to gain recognition in the generation of functional cartilage and fibrocartilage in recent years.^4,68 In this strategy, no scaffolds make the seed cells in a high density and promote cell to cell adhesion and cells to matrix adhesion. Meanwhile, the approach encourages cells to integrate into a matrix with quantifiable mechanical properties.^69,70

The self-assembled constructions have also responded well to the exogenous agents, such as growth factors^70,71 and mechanical stimulation.⁷² After treatment with transforming growth factor-β1 (TGF-β1) and chondroitinase ABC, the self-assembled scaffoldless construct resulted in maturational growth as evidenced by synergistic enhancement of the compressive relaxation modulus by 68% and the radial tensile modulus by 5-fold.⁷¹ Under hydrostatic pressures, the constructs were enhanced by increasing Young's modulus (92%), aggregate modulus (96%), GAG per wet weight (52%), and Col per wet weight (51%).⁷³ However, the self-assembled systems have not been implanted in vivo.

3 Conclusion and perspectives

Although different categories of scaffolds offer various advantages, none of them can meet all the three fundamental requirements for MTE, that is, mechanics, bioactivities, and logistics.²⁴ First, hydrogels represent a versatile category of scaffolds. The most prominent feature is their ease surgical implantation, but their unsatisfactory mechanical properties, especially in tension, limit their applications. Second, the 3D synthetic polymer scaffolds have obvious advantage and inadequacy: superior mechanical properties and inferior bioactivities. Meanwhile, the promotion of joint lubrication to prevent tissue tear at bone–cartilage interface needs further researches. Third, ECM scaffolds represent promising properties in mechanics and logistics. However, the development of suitable alternative tissues within ECM scaffolds in vivo remains a key issue in multi-center clinical trials. Forth, the bottlenecks of tissue-derived scaffolds are their supply and process standardization, though they own the most favorable bioactivities. Furthermore, none have demonstrated their chondro-protective function in humans. It is to be expected that the properties of scaffolds will have to be optimized.

In general, the use of scaffolds for MTE is at an early stage, and the following requirements should be considered. First, the initial mechanical properties of scaffolds should be equivalent to those of native meniscus, that is, compressive modulus from 75 to 150 kPa; tensile modulus of 75–150 MPa,^74,75 although the mechanical strength of scaffolds would attenuate with their degradation. Unfortunately, only few reported scaffolds are comparable to the tensile modulus of meniscus up to now.⁵ Second, the degradation profiles should extend to 9–12 months as our advised. In such way, the adequate ECM and tissue ingrowth can maintain mechanical support. Moreover, surface or bulk degradation profiles should also be considered. Based on the above issue, PCL would be a promising material for MTE, which loses its molecular weight in vivo considerably slower than other aliphatic polyester.^76,77 Third, to promote cell proliferation and matrix deposition, large mean pore sizes (±300 μm), high interconnectivity by smaller micropores (±10–50 μm), and moderate percentage of porosity (±70%) should be taken into account on material selection and fabrication. However, the balance between biological proliferation and mechanical property needs further research.⁷⁸ Forth, the implantation of TEM should be as minimal as possible, perfectly via arthroscopy. Surgical fixation techniques should be rigid enough to withstand extrusive forces and the hoop stresses during loading. Fifth, seed cells combining with growth factors and/or mechanical stimulation offer a promising protocol for the regeneration of TEM before and after implantation.⁷⁹ Therefore, a bioreactor designed for seeding cells and providing biomechanical stimuli will be applicable.^80,81 Although biomechanical stimulation is perhaps an opportunity for improving the regeneration of TEM, the variable parameters, that is, method, time of application, frequency of stimulation and magnitude, used for mechanical stimuli remain controversial and undiscovered. The further advanced MTE and the final clinical implementation need to progress with the multidisciplinary of chemistry, materials science, biology, medicine, and even industrial.

Acknowledgements

This study was supported by the National Natural Scientific Foundation of China (Grant No. 51273004, 31200725, and 51303174), the National High Technology Research and Development Program of China (863 Program) (Grant No. 2012AA020502), and the Science and Technology Planning Project of Changchun City (Grant No. 14KG045).

Notes and references

1. M. Englund, E. M. Roos, H. Roos and L. Lohmander, Rheumatology, 2001, 40, 631–639.
2. C. Weinand, G. M. Peretti, S. B. Adams, Jr., M. A. Randolph, E. Savvidis and T. J. Gill, Arch. Orthop., Mechanother. Unfallchir., 2006, 126, 599–605.
3. C. J. Moran, A. Busilacchi, C. A. Lee, K. A. Athanasiou and P. C. Verdonk, Arthroscopy, 2015, 31, 944–955.
4. D. J. Huey, J. C. Hu and K. A. Athanasiou, Science, 2012, 338, 917–921.
5. C. H. Lee, S. A. Rodeo, L. A. Fortier, C. Lu, C. Erisken and J. J. Mao, Sci. Transl. Med., 2014, 6, 266ra171.
6. S. W. Kang, S. M. Son, J. S. Lee, E. S. Lee, K. Y. Lee, S. G. Park, J. H. Park and B. S. Kim, J. Biomed. Mater. Res., Part A, 2006, 77, 659–671.
7. G. M. Peretti, T. J. Gill, J. W. Xu, M. A. Randolph, K. R. Morse and D. J. Zaleske, Am. J. Sports Med., 2004, 32, 146–158.
8. N. J. Gunja and K. A. Athanasiou, J. Biomed. Mater. Res., Part A, 2010, 92, 896–905.
9. R. K. Das and O. F. Zouani, Biomaterials, 2014, 35, 5278–5293.
10. Y. Jiang, J. Chen, C. Deng, E. J. Suuronen and Z. Zhong, Biomaterials, 2014, 35, 4969–4985.
11. L. S. Teixeira, J. Feijen, C. A. van Blitterswijk, P. J. Dijkstra and M. Karperien, Biomaterials, 2012, 33, 1281–1290.
12. A. Kang, J. Park, J. Ju, G. S. Jeong and S. H. Lee, Biomaterials, 2014, 35, 2651–2663.
13. C. N. Jones, N. Tuleuova, J. Y. Lee, E. Ramanculov, A. H. Reddi, M. A. Zern and A. Revzin, Biomaterials, 2009, 30, 3733–3741.
14. A. C. Aufderheide and K. A. Athanasiou, Tissue Eng., 2005, 11, 1095–1104.
15. S. H. Hsu, S. W. Whu, S. C. Hsieh, C. L. Tsai, D. C. Chen and T. S. Tan, Artif. Organs, 2004, 28, 693–703.
16. T. M. Hammoudi, H. Lu and J. S. Temenoff, Tissue Eng., Part C, 2010, 16, 1621–1628.
17. N. Tuleuova, J. Y. Lee, J. Lee, E. Ramanculov, M. A. Zern and A. Revzin, Biomaterials, 2010, 31, 9221–9231.
18. S. N. Pawar and K. J. Edgar, Biomaterials, 2012, 33, 3279–3305.
19. J. Wu, Q. Ding, A. Dutta, Y. Wang, Y. H. Huang, H. Weng, L. Tang and Y. Hong, Acta Biomater., 2015, 16, 49–59.
20. M. B. Browning, V. Guiza, B. Russell, J. Rivera, S. Cereceres, M. Hook, M. S. Hahn and E. M. Cosgriff-Hernandez, Tissue Eng., Part A, 2014, 20, 3130–3141.
21. J. J. Ballyns, J. P. Gleghorn, V. Niebrzydowski, J. J. Rawlinson, H. G. Potter, S. A. Maher, T. M. Wright and L. J. Bonassar, Tissue Eng., Part A, 2008, 14, 1195–1202.
22. J. J. Ballyns and L. J. Bonassar, J. Biomech., 2011, 44, 509–516.
23. P. Angele, B. Johnstone, R. Kujat, J. Zellner, M. Nerlich, V. Goldberg and J. Yoo, J. Biomed. Mater. Res., Part A, 2008, 85, 445–455.
24. E. A. Makris, P. Hadidi and K. A. Athanasiou, Biomaterials, 2011, 32, 7411–7431.
25. Y. Kawamoto, A. Nakao, Y. Ito, N. Wada and M. Kaibara, J. Mater. Sci.: Mater. Med., 1997, 8, 551–557.
26. H. Elema, J. H. de Groot, A. J. Nijenhuis, A. J. Pennings, R. P. H. Veth, J. Klompmaker and H. W. B. Jansen, Colloid Polym. Sci., 1990, 268, 1082–1088.
27. T. G. Tienen, R. G. Heijkants, J. H. de Groot, A. J. Pennings, A. J. Schouten, R. P. Veth and P. Buma, Am. J. Sports Med., 2006, 34, 64–71.
28. M. Szycher, J. Biomater. Appl., 1988, 3, 297–402.
29. T. G. Tienen, R. G. Heijkants, J. H. de Groot, A. J. Schouten, A. J. Pennings, R. P. Veth and P. Buma, J. Biomed. Mater. Res., Part B, 2006, 76, 389–396.
30. R. T. Welsing, T. G. van Tienen, N. Ramrattan, R. Heijkants, A. J. Schouten, R. P. Veth and P. Buma, Am. J. Sports Med., 2008, 36, 1978–1989.
31. A. J. Fox, F. Wanivenhaus, A. J. Burge, R. F. Warren and S. A. Rodeo, Clin. Anat., 2015, 28, 269–287.
32. W. J. Li, R. L. Mauck, J. A. Cooper, X. Yuan and R. S. Tuan, J. Biomech., 2007, 40, 1686–1693.
33. J. Baek, X. Chen, S. Sovani, S. Jin, S. P. Grogan and D. D. D'Lima, J. Orthop. Res., 2015, 33, 572–583.
34. B. M. Baker and R. L. Mauck, Biomaterials, 2007, 28, 1967–1977.
35. M. B. Fisher, E. A. Henning, N. Soegaard, J. L. Esterhai and R. L. Mauck, Acta Biomater., 2013, 9, 4496–4504.
36. T. P. Driscoll, N. L. Nerurkar, N. T. Jacobs, D. M. Elliott and R. L. Mauck, J. Mech. Behav. Biomed. Mater., 2011, 4, 1627–1636.
37. C. Chiari, U. Koller, R. Dorotka, C. Eder, R. Plasenzotti, S. Lang, L. Ambrosio, E. Tognana, E. Kon, D. Salter and S. Nehrer, Osteoarthritis Cartilage, 2006, 14, 1056–1065.
38. E. Kon, C. Chiari, M. Marcacci, M. Delcogliano, D. M. Salter, I. Martin, L. Ambrosio, M. Fini, M. Tschon, E. Tognana, R. Plasenzotti and S. Nehrer, Tissue Eng., Part A, 2008, 14, 1067–1080.
39. E. Kon, G. Filardo, M. Tschon, M. Fini, G. Giavaresi, L. Marchesini Reggiani, C. Chiari, S. Nehrer, I. Martin, D. M. Salter, L. Ambrosio and M. Marcacci, Tissue Eng., Part A, 2012, 18, 1573–1582.
40. B. Grigolo, L. Roseti, M. Fiorini, M. Fini, G. Giavaresi, N. N. Aldini, R. Giardino and A. Facchini, Biomaterials, 2001, 22, 2417–2424.
41. A. R. Esposito, M. Moda, S. M. Cattani, G. M. de Santana, J. A. Barbieri, M. M. Munhoz, T. P. Cardoso, M. L. Barbo, T. Russo, U. D'Amora, A. Gloria, L. Ambrosio and E. A. Duek, BioRes. Open Access, 2013, 2, 138–147.
42. Y. Chau, Y. Luo, A. C. Cheung, Y. Nagai, S. Zhang, J. B. Kobler, S. M. Zeitels and R. Langer, Biomaterials, 2008, 29, 1713–1719.
43. W. J. Li, Y. J. Jiang and R. S. Tuan, Tissue Eng., 2006, 12, 1775–1785.
44. K. R. Stone, W. G. Rodkey, R. Webber, L. McKinney and J. R. Steadman, Am. J. Sports Med., 1992, 20, 104–111.
45. M. Sonoda, F. L. Harwood, M. E. Amiel, H. Moriya, M. Temple, D. G. Chang, L. M. Lottman, R. L. Sah and D. Amiel, Am. J. Sports Med., 2000, 28, 90–97.
46. U. Freymann, M. Endres, U. Goldmann, M. Sittinger and C. Kaps, Osteoarthritis Cartilage, 2013, 21, 773–781.
47. S. M. Mueller, S. Shortkroff, T. O. Schneider, H. A. Breinan, I. V. Yannas and M. Spector, Biomaterials, 1999, 20, 701–709.
48. N. J. Gunja and K. A. Athanasiou, Arthritis Res. Ther., 2007, 9, R93.
49. C. Chiari, U. Koller, B. Kapeller, R. Dorotka, U. Bindreiter and S. Nehrer, Tissue Eng., Part A, 2008, 14, 1295–1304.
50. G. K. Tan, D. L. Dinnes, P. T. Myers and J. J. Cooper-White, Biomaterials, 2011, 32, 5600–5614.
51. G. K. Tan, D. L. Dinnes, L. N. Butler and J. J. Cooper-White, Biomaterials, 2010, 31, 6104–6118.
52. P. Bulgheroni, L. Murena, C. Ratti, E. Bulgheroni, M. Ronga and P. Cherubino, Knee, 2010, 17, 224–229.
53. S. Zaffagnini, G. M. Marcheggiani Muccioli, N. Lopomo, D. Bruni, G. Giordano, G. Ravazzolo, M. Molinari and M. Marcacci, Am. J. Sports Med., 2011, 39, 977–985.
54. A. J. Schoenfeld, W. J. Landis and D. B. Kay, Am. J. Orthopsychiatry, 2007, 36, 614–620.
55. J. R. Steadman and W. G. Rodkey, Arthroscopy, 2005, 21, 515–525.
56. T. Yamasaki, M. Deie, R. Shinomiya, Y. Izuta, Y. Yasunaga, S. Yanada, P. Sharman and M. Ochi, J. Biomed. Mater. Res., Part A, 2005, 75, 23–30.
57. J. L. Cook, J. L. Tomlinson, J. M. Kreeger and C. R. Cook, Am. J. Sports Med., 1999, 27, 658–665.
58. J. L. Cook, D. B. Fox, P. Malaviya, J. L. Tomlinson, K. Kuroki, C. R. Cook and S. Kladakis, Am. J. Sports Med., 2006, 34, 32–42.
59. J. L. Cook, D. B. Fox, K. Kuroki, M. Jayo and P. G. de Deyne, Am. J. Vet. Res., 2008, 69, 148–156.
60. M. P. Bradley, P. D. Fadale, M. J. Hulstyn, W. R. Muirhead and J. T. Lifrak, Orthopedics, 2007, 30, 650–656.
61. K. J. Stabile, D. Odom, T. L. Smith, C. Northam, P. W. Whitlock, B. P. Smith, M. E. van Dyke and C. M. Ferguson, Arthroscopy, 2010, 26, 936–948.
62. D. Maier, K. Braeun, E. Steinhauser, P. Ueblacker, M. Oberst, P. C. Kreuz, N. Roos, V. Martinek and A. B. Imhoff, J. Orthop. Res., 2007, 25, 1598–1608.
63. G. H. Sandmann, S. Eichhorn, S. Vogt, C. Adamczyk, S. Aryee, M. Hoberg, S. Milz, A. B. Imhoff and T. Tischer, J. Biomed. Mater. Res., Part A, 2009, 91, 567–574.
64. T. W. Stapleton, J. Ingram, J. Katta, R. Knight, S. Korossis, J. Fisher and E. Ingham, Tissue Eng., Part A, 2008, 14, 505–518.
65. B. B. Mandal, S. H. Park, E. S. Gil and D. L. Kaplan, Tissue Eng., Part A, 2011, 17, 2749–2761.
66. B. B. Mandal, S. H. Park, E. S. Gil and D. L. Kaplan, Biomaterials, 2011, 32, 639–651.
67. L. P. Yan, J. M. Oliveira, A. L. Oliveira, S. G. Caridade, J. F. Mano and R. L. Reis, Acta Biomater., 2012, 8, 289–301.
68. J. C. Hu and K. A. Athanasiou, Tissue Eng., 2006, 12, 969–979.
69. G. Ofek, C. M. Revell, J. C. Hu, D. D. Allison, K. J. Grande-Allen and K. A. Athanasiou, PLoS One, 2008, 3, e2795.
70. E. A. Makris, R. F. MacBarb, N. K. Paschos, J. C. Hu and K. A. Athanasiou, Biomaterials, 2014, 35, 6787–6796.
71. D. J. Huey and K. A. Athanasiou, Biomaterials, 2011, 32, 2052–2058.
72. B. D. Elder and K. A. Athanasiou, Tissue Eng., Part A, 2009, 15, 1151–1158.
73. B. D. Elder and K. A. Athanasiou, PLoS One, 2008, 3, e2341.
74. M. A. Sweigart and K. A. Athanasiou, Proc. Inst. Mech. Eng., Part H, 2005, 219, 337–347.
75. S. D. Masouros, I. D. McDermott, A. A. Amis and A. M. Bull, Knee Surg. Sports Traumatol. Arthrosc., 2008, 16, 1121–1132.
76. H. Kweon, M. K. Yoo, I. K. Park, T. H. Kim, H. C. Lee, H. S. Lee, J. S. Oh, T. Akaike and C. S. Cho, Biomaterials, 2003, 24, 801–808.
77. C. G. Pitt, M. M. Gratzl, G. L. Kimmel, J. Surles and A. Schindler, Biomaterials, 1981, 2, 215–220.
78. I. Zein, D. W. Hutmacher, K. C. Tan and S. H. Teoh, Biomaterials, 2002, 23, 1169–1185.
79. Z. Z. Zhang, D. Jiang, S. J. Wang, Y. S. Qi, J. Y. Zhang and J. K. Yu, ACS. Appl. Mater. Interfaces, 2015, 7, 15294–15302.
80. N. L. Nerurkar, S. Sen, B. M. Baker, D. M. Elliott and R. L. Mauck, Acta Biomater., 2011, 7, 485–491.
81. C. Liu, R. Abedian, R. Meister, C. Haasper, C. Hurschler, C. Krettek, G. von Lewinski and M. Jagodzinski, Biomaterials, 2012, 33, 1052–1064.

---