Cartilage Reconstruction Using Self-anchoring Implant with Functional Gradient

aDepartament of Mechanical Engineering, University of São Paulo – USP, Av. Trabalhador São Carlense, 400, CEP 13566-590, São Carlos, SP, Brazil bCenter for Natural Sciences and Humanities, University of ABC, Av. dos Estados, 5001, CEP 09210-971, Santo André, SP, Brazil cBioengineering Interunits, University of São Paulo – USP, Av. Trabalhador São Carlense, 400, CEP 13566-590, São Carlos, SP, Brazil dFederal University of São Carlos – UFSCar, Rod. Washington Luís, Km 235, CEP 13565-905, São Carlos, SP, Brazil eLaboratory of Morfology, Universidade Estadual Paulista Júlio de Mesquita – UNESP, Rua Humaitá, 1680, CEP 14801-903, Araraquara, São Paulo, SP, Brazil


Introduction
Articular cartilage is a highly resilient connective tissue covering the bone joint surfaces.In typical synovial joints, articular cartilage is a tribological and lubricated surface that promotes movement without considerable friction.This type of cartilage is hyaline and is mainly composed of chondrocytes surrounded by extracellular matrix 1,2 .The biomechanical properties of articular cartilage depend largely on the composition and integrity of the extracellular matrix.Because it is an avascular tissue, articular cartilage has a low regeneration capacity.Consequently, the cartilage lesions are healed by the formation of scar tissue composed mostly of fibrocartilage, which has weaker mechanical and biological properties than the original hyaline cartilage and which gradually degenerates over time, resulting in permanent loss of structure and function and leading to severe pain 1,3,4 .
The degenerated cartilage can lead to either temporary or permanent physical disability and can be a result of a number of conditions, such as microorganisms (tuberculosis, syphilis), injuries secondary to inflammatory rheumatic diseases, traumatic injuries, and metabolic diseases 5 .
The phenomenon of tissue healing can be divided into four natural phases: hemostasis, inflammation, proliferation, and remodeling.An injury restricted to the superficial layers of cartilage (chondral defect) does not entirely satisfy these three phases due to the avascular nature of the hyaline cartilage.On the other hand, when there is an osteochondral defect with exposure of the subchondral bone, all three stages occur naturally 6 .
In most clinical situations, the joint itself can repair damage that does not disturb the integrity of the articular surface.A mechanical disruption of the surface cartilage stimulates chondral synthesis, but rarely results in injury repair.In turn, although bone and chondral repair can be induced by osteochondral lesions, the repaired tissue does not have the same mechanical and biological properties as the original cartilage 7 .Treatments to reconstitute damaged articular cartilage such as drilling, subchondral Cartilage Reconstruction Using Self-anchoring Implant with Functional Gradient bone microfracture, and abrasion arthroplasty promote the formation of fibrous cartilage with different properties than those of the original hyaline tissue and therefore produce repairs of limited duration 8 .
Focal lesions of the articular cartilage require surgical treatment ranging from classical methods for bone marrow stimulation such as debridement, multiple perforations, abrasions, and microfractures, to modern biological methods like pericondral and periosteal transplantation, autologous implantation of cultured chondrocyte, osteochondral autologous grafts 6,9 , and more recently, to the use of biomaterials.
Among all the biomaterials used for tissue engineering, polyurethane has a structure that allows products with a wide range of physical and mechanical properties to be produced, making it suitable for use in a variety of applications, such as soft tissue and cartilage reconstruction and bone regeneration 10 .In technical applications, the polyurethane is applied as a coating following abrasion in water environment.Recent applications of biomaterials for tissue engineering have focused on the interaction with surrounding tissues 11 , improving bioactivity by incorporating fillings or coatings of porous bioceramics on polymeric matrices 12 .Bioactive glass is an example of a biomaterial that causes a specific biological response in vivo, resulting in the formation of a strong bond between the living tissue (e.g.bone) and the biomaterial 13 .45S5 bioglass is a important bioactive glass composed of SiO 2 , Na 2 O, CaO, and P 2 O 5 13,14 . The main advantage of bioglass is the ability to unite both hard and soft tissues, since its bioactivity index is greater than 8.Moreover, it has been shown that the dissolution product of bioglass stimulates gene expression in osteoblasts 11 as well as angiogenesis 15 ; thus, bioglass has been considered a material of choice for the development of bioactive composites for bone engineering 14 .
For cartilage implants to be successful, it will be necessary to solve challenges associated with mechanical loading properties, friction/lubrication, mechanical anchorage, durability, and interaction with surrounding cartilage.To that end, this study describes a conceptually new implant and application to repair osteochondral defects in knees of rabbits.The sample was a cylindrical model composed of polyurethane, bioglass microfiber, and variations in porosity, designed as a biological structural and functional gradient material (FGM).

Implant concept
An implant with functionally graded characteristics, once deployed, must establish a compromise between mechanical properties and biocompatibility.In order to achieve this goal, implants composed of bioglass microfiber and porous polyurethane were produced using available and reproducible manufacturing techniques.The upper side of the implant is responsible for the mechanical characteristics, gradually shifting to the lower side, which permits mechanical fixation by osteointegration in the osteochondral region, as shown in Figure 1.Therefore, this design of composite material has the characteristics of a possible replacement for articular cartilage.The processing methods and materials used were selected in order to meet the desired characteristics for the implant components, i.e., dimensional accuracy, mechanical reliability, service performance, repeatability, and acceptable cost.As a result, a structural material was obtained in functional gradient.The bioglass microfiber confers mechanical strength, rigidity, and adherence to biological tissue, while the polyurethane is responsible for the elastic-plastic behavior.This implant, when applied in vivo, is expected to maintain both the necessary mechanical properties and tribological characteristics (Figure 2).The overall dimensions of the samples for implantation were 3.0 mm diameter by 4.0 mm height.The aim is to design a material with new properties and functions that are impossible to find in a conventional homogeneous material.

Materials
The new material was developed using medical polyurethane (Tecothane® TPU TT-1074A, manufactured by Lubrizol® Advanced Materials Inc.) as a matrix and bioglass microfiber (manufactured by Mo-Sci Corporation) as a reinforcing and bone-integrating agent.TT-1074A has aromatic properties, a hardness of 75A-77D, a specific gravity of 1.1 g/cm 3 , a tensile strength of 41 MPa, radiopacity, a normal color, and a melt processing temperature ranging from 190 °C to 220 °C.Mo-Sci bioglass microfiber has a specific gravity of 2.7 g/cm 3 , an elastic modulus of 30 to 35 GPa, a tensile strength of 40 to 60 MPa, a refractive index of 1.55, a softening temperature of 550 °C, a dissolution rate of ~ 150 mg/cm 2 /day, and thermal expansion of 16×10 -6 cm/cm/°C.The following are typical chemical compositions: 45% SiO 2 , 24.5% Na 2 O, 6% P 2 O 5 .300 µm diameter NaCl particles were used as a porogenic agent.After the implants were formed, they were leached with distilled water to remove NaCl, and were dried in vacuum desiccators over a 24-hour period.

Manufacturing the implants
The implants were manufactured in a metal mold with a 3.0 mm inner diameter and two punches (superior and inferior), warmed and kept at a stable temperature of 180 °C.The mold created by the inferior punch had its cavity filled in three stratified phases.The first layer was composed of a mixture of polyurethane micro-pellets (40% by volume), microfiber bioglass particles (20 vol%), and NaCl (40 vol%); the second, a mixture of polyurethane micro-pellets (85 vol%) and microfiber bioglass (15 vol%); and the third layer was composed entirely of polyurethane Figure 2. Conceptualization of the specimen in functional gradient.Cartilage Reconstruction Using Self-anchoring Implant with Functional Gradient micro-pellets.After being filled, the mold was closed with a punch and heated for thirty seconds.Subsequently, an uniaxial pressure of 350 MPa was applied for one minute, followed cooling under pressure with circulating water to achieve temperatures of between 40 °C and 50 °C.The samples were radiosterilized with a CO-60 Panoramic dose of 25 kGy and sealed until the time of surgery.

Determination of kinetic friction coefficient
The kinetic friction coefficients of cartilage surfaces against polyurethane, silicone, and glass were experimentally determined.5 mm-diameter cartilage samples from cow hip joints, a normal force of 10 N, and a displacement velocity of 5 mm/s were employed over a test distance of 10 mm in a saline bath environment

Scanning Electron Microscopy (SEM)
Three samples were selected for SEM analyses.Cuts of thickness ranging from 20 µm to 22 µm underwent SEM analysis after fixation, demineralization, dehydration, diaphanization, and insertion into paraffin.SEM photomicrographs were taken using the Oxford LEO (Model 440) detector operating with an electron beam of 20 kV.The samples were coated with 10 nm of gold in a BAL-TEC MED 020 and kept in desiccators until the time of analysis using standard for co-calibration beam of 20 kV, focal length of 25 mm, 30% dead time, current of 2.82 A and I probe of 950 pA.Mapping was obtained with 20 frames and line-scan with 3000 frames.

Atomic Force Microscopy (AFM)
The microscope was operated in intermittent contact mode using a silicon needle with a spring constant of 70 n/ma and a frequency of 330,000 Hz.This test is justified by the analysis of micro-tribological surface of the polyurethane (merger, union, pores, roughness, etc.) and the osteointegration surface (porosity).This test was performed in two samples (n = 2).

Evaluation of the materials' biocompatibility and cytotoxicity
Vero cells were used from an established fibroblast cell line originating from the kidney of the African green monkey (Cercopithecus aethiops) and obtained by the Adolfo Lutz Institute (Sao Paulo, SP, Brazil).The cells were maintained in Ham's F-10 medium supplemented with 10% fetal bovine serum.In vitro tests were performed at the Laboratory of Biomaterials in Orthopedics (Labimo), State University of Campinas -Unicamp.
The biomaterials were tested in accordance with current national standards 16 using Vero cells, as recommended by ASTM F813-83 standard 17 to verify cytotoxicity.
Indirect Cytotoxicity -Extracts of 45S5 blown fiber materials and polyurethane were obtained by incubation in Ham F-12 medium with 10% FCS at 37 °C for 48 hours without agitation, at a rate of 01 g/mL 14 .The medium was then collected, allowing the evaluation of the possible effects of the substances released by the different materials that could be present.As a positive control, a solution of phenol of 1% was used in the culture medium.After obtaining the extracts, they were used for the cultivation of Vero cells.The cells were inoculated at a concentration of 1.0×10 5 cells/mL in 24-well plates in Ham F-12 medium with 10% FCS.After 2 hours of incubation, and enough time for cell adhesion, the culture medium was replaced by the extracts or respective control and cultured for 24 hours.Five fields were randomly chosen for each well and the cells were counted using a direct counting method.Images of these fields were also obtained for photographic documentation.
Direct Cytotoxicity -The evaluation of the in vitro cytotoxicity of different materials was performed using the cell viability test and the direct cell counting method.
In vivo studies -The surgical protocol was approved by the Ethics Committee of the Sao Carlos Federal University (UFSCar) (approval number: 031/2010).In this study, we used eight adult male New Zealand rabbits with an average weight of 2.5 kg maintained in the bioterium of the Interunits Post-graduation Program in Bioengineering, School of Engineering of São Carlos, São Carlos, Brazil.
Rabbits in the experimental group (n = 6) received a polyurethane and bioglass microfiber implant with a diameter of 3 mm and a depth of 4 mm in the center of the intercondylar region (trochlear groove).The control group (n = 2) received only the bone defect with a diameter of 3 mm and a depth of 4 mm.The animals were kept in individual 70×70×70 cm cages with a ceramic container for industrial water and food pellets ad libitum.The vivarium lighting followed a cycle of 12/12 h light/dark.The experimental period was 15, 30 and 90 days.
The knee joint was chosen since it is exposed to high mechanical stress over time and because it presents anatomical features that facilitate the surgical approach and the evaluation of the implants' performance.We selected the region of the femoral groove, and not the region of the femoral condyles, because its easier approach makes it a safe and reproducible surgical procedure.Furthermore, the trochlear region of rabbits is exposed to repeated mechanical stresses during walking in the same manner as the region of the condyles 18 .

Surgical procedure
Anesthesia was induced by an intramuscular injection of a 5% ketamine hydrochloride solution at a ratio of 50 mg/ kg, combined with a 2% xylazine hydrochloride solution at a ratio of 3 mg/kg.Shaving and assepsy were done at the surgical site with a 10% povidone-iodine solution.The surgical technique used was described by Reiff 18 .The surgical access on the knee is the medial longitudinal, extending from the lower region of the patella towards the tibial tuberosity to expose the patellar tendon.
The incision was made in the central portion of the patellar tendon in the longitudinal direction, extending over its entire length and removing the infrapatellar fat in the caudal direction, with subsequent visualization of the trochlear groove of the knee.The osteochondral defect was made in the central region of the femoral trochlea, using a low speed drill of 3 mm of diameter (Figure 3a).The implant was carefully inserted into the osteochondral defect in such way that the tribological surface was at the same level as bone cartilage (Figure 3b).
For three consecutive days immediately following the surgery, each animal received an intramuscular injection of 1% ketoprofen and a topical application of povidone-iodine on the suture.
After recovery from anesthesia, a normal load on the operated limb was immediately permitted, without any restriction or detention.
The animals were sacrificed 15, 30 or 90 days after surgery.Anesthesia similar to the preoperative anesthesia was used, followed by intravenous injection of 60 mg/kg of potassium chloride.

Macroscopic and microscopic analysis
Macroscopic analysis was used to inspect the external features of the repair of defect.The main characteristics to be analyzed, according to Ribeiro 6 , are the surface brightness and consistency of the tissue formed.
The histological samples were collected and fixed in 10% formalin (in 0.1 M phosphate buffer at pH 7.2) for 24 hours.The bones were kept in a decalcification solution until histological sectioning was possible.Samples were then washed in water, dehydrated using increasing concentrations of ethanol, and embedded in paraffin for routine histological techniques.5mm thick sections were made for different samples.The sections were then stained with hematoxylin and eosin.

Cytotoxicity tests
In the studied material, for both direct and indirect cytotoxicity, the quantity of cells is statistically similar to the negative control, and statistically different from the positive control (Figure 4).This indicates no direct or indirect toxicity caused by the sample.
Qualitative analysis showed that the appearance of cells growing under conditions of direct or indirect cytotoxicity is similar to the negative control and different from the positive control (Figure 5).

AFM
The measure of roughness of the tribological side of the specimen, taken using AFM, was 15.74 nm. Figure 6 shows the surface of the melted and subsequently cooled polymer, presenting rounded reliefs and minimal porosity.

Determination of kinetic friction coefficient
A suitable friction between material and the cartilage tissue requires low friction coefficient.In this study, glass exhibited the smallest friction coefficient (see Table 1), and it can be used for calibration.Silicone is a commonly proposed cartilage substitute material.However, polyurethane was  considered more suitable than silicone for tribologic performance.

SEM of the Samples
The lateral surface of the specimen can be seen in Figure 7a, which shows the differentiation of the proximal, middle and distal tribological surface generated by functional gradient manufacturing.Figure 7c shows the   substrates used to compose the novel material: polyurethane and microfiber bioglass 45S5.When viewing the sample with the backscattered electron function, the presence of NaCl (evidenced in white) is notable in the porosity remaining after leaching (Figure 7b).

Macro and microscopic evaluation -In Vivo Protocol
The rabbits sacrificed 15 days after implantation showed macroscopic features such as plane surface in the osteochondral defects, growth of fibrocartilaginous tissue, translucent coloration, and a depression-free surface, as shown in Figure 8.Although the rabbits sacrificed after 30 days also showed plane surface, translucent coloration, and a depression-free surface, there was no growth of fibrocartilaginous tissue (Figure 8).Those sacrificed after 90 days presented the following characteristics: presence of plane surface, growth of fibrocartilaginous tissue, transparent color, and no depression on the surface, as shown in Figure 8.
Immediately after the surgical procedure, all rabbits had normal ambulation.The animals showed no signs of infection and the joints remained stable after 15, 30 and 90 days, demonstrating the validity of the experimental method used.
The histological analysis of the control group sacrificed after 90 days showed the presence of bone tissue as well as fibrocartilaginous tissue with chondrocytes and blood vessels (Figure 9).
The SEM photomicrographs of the control group sacrificed after 30 days showed local reabsorption with remodeling of tissues.Figure 10 shows the porous bone formation filling the defect space in the first 15 days of the control group.This structure is similar to the bone trabeculae and was considered biomimetic.Figure 11 shows the occupation of almost the entire porous region by the bone formation after 30 days of implantation, filling the trabeculae with osteoid bone.Figure 12 shows the consolidation of bone formation with dense filling of the porous region and osteon formation 90 days postexperimentation.

Discussion
Surgical treatment of chondral and osteochondral injuries that affect load-bearing joints, especially the knee, still represents a challenge for orthopedics due to the characteristics of hyaline articular cartilage, which is devoid of vascularization and has limited potential of cicatrization 6 .
Didactically, it is possible to subdivide the tissue regeneration process in three phases: necrosis, inflammation, and repair 19 .Frenkel et al. 20 state that superficial lesions of hyaline cartilage that do not reach the subchondral bone cannot heal normally.Over the time, the progression of the damage produces an irregular and dull macroscopic appearance.However, in injuries extending into the wellvascularized subchondral bone (osteochondral defects), all three stages naturally occur 21 .In the current study, all three phases of regeneration were shown.
At the base of the defect, in the region contacting the subchondral bone, bone formation occurs extending toward the joint.For reasons that are not yet clear, bone tissue formation is often interrupted in the transition zone, allowing the remaining defect to be filled with fibrocartilaginous tissue 7,21 .Histological analysis 15 days after surgery indicated the presence of bone tissue, fibrous tissue with collagen fibers, fibroblasts, blood vessels, and the absence of inflammation.Histological analysis 90 after surgery showed the presence of chondrocytes near the region of the implant, as well as bone tissue and fibrocartilaginous tissue rich in blood vessels.
In the other hand, Huntley et al. 22 state that this new formed tissue that fills the osteochondral defect significantly differs from normal cartilage tissue in its composition and in the arrangement of interior elements.Although chondrocytes synthesize new proteoglycan molecules, they have a lower molecular weight than those found naturally.Similarly, type II collagen fibers have a smaller diameter and more irregular distribution.This arrangement, combined with the fact that there is imperfect integration between the new tissue and the cartilage, favors water permeability.Furthermore, the neoformed fibrocartilaginous tissue has a lower modulus of elasticity than cartilaginous tissue 23,24 .
According to Hasegawa et al. 25 , the fact that the implants have securely attached since the date of surgery favored    the formation of bone tissue adjacent to the subchondral bone.Furthermore, they reported the formation of tissue macroscopically similar to cartilage on a part of the implant surface, probably due to the friction between the implant and the patellar surface.Histological examination revealed the newly formed tissue in different period groups.
Biomaterials have contributed significantly to the advance of modern medicine.Driven by collaboration between biomedical engineers, materials scientists, chemists, physicists, biologists and physicians, the science of biomaterials has matured.It has become an area of multidisciplinary research, involving the use of fundamental concepts in the development of materials for clinical practice application.Rapid progress has been achieved in this area, resulting in the synthesis of new ceramic and polymeric biomaterials, the acquisition of knowledge about interactions between biomaterials and biological tissues, the development of artificial organs using cell culture, the advancement of systems for controlled release of drugs, improvement in cardiovascular grafts and devices, miniaturization of prostheses, bone grafts and orthopedic devices.
The characterization of a biomaterial is crucial for determining its performance and consequent definition of the minimum requirements to be met by the industry.In this study, was investigated an essential aspect in developing a biomaterial and its interaction with tissues.Was developed an implant that presents a concept to be applied in vivo.The implant maintains the desired mechanical properties and tribological characteristics, and simultaneously ensures good cell adhesion to the bone surface in the intercondylar region of the knee.Once implemented, it is expected to pose a compromise between mechanical properties and biocompatibility.Was described the development of a combination of materials (polyurethane biomaterial compound, microfiber 45S5 bioglass and porous) using manufacturing techniques that are available, reproducible, and technically functional.
With the methodology used in this study, the macroscopic findings in both groups were considered biologically acceptable.The term "biologically acceptable", as defined by Amiel et al. 26 , is used to describe repairs with the macroscopic appearance of newly formed tissue similar to fibrocartilage and that are smooth, shiny, firm, and in continuity with the adjacent cartilage.This is a macroscopic criterion of evaluation, and an exact correlation with the histological, biochemical and biomechanical features of the repair cannot be established 6 .Analysis performed 15, 30, and 90 days after surgery demonstrated the characteristics mentioned above in our sample.A SEM photomicrograph taken after 90 days of trial showed the consolidation of bone formation with densified filling of the porous region and osteon formation.
This study is likely to contribute to future research.The biomaterial composed of polyurethane and microfiber bioglass 45S5 described above behaved promisingly due to its biocompatibility, stability, and the absence of inflammatory reaction after 90 days of trial.

Conclusion
This study reported on the development and successful validation of a biomaterial composed of polyurethane and bioglass microfiber 45S5 applied as a substitute for articular cartilage, as evidenced by the results of the  in vitro and in vivo assays, histological evaluation, SEM and AFM.The biomaterial composed of polyurethane and bioglass microfiber 45S5 showed neither direct nor indirect toxicity in samples and promoted cell growth and spreading, allowing further study using in vivo experiments in rabbits.An implant made of polyurethane and bioglass microfiber 45S5 promoted bone formation and complete filling of pores, as well as the presence of chondrocytes.Furthermore, independent of the follow-up time, the implant of polyurethane and bioglass microfiber 45S5 is a promising material because of the following characteristics: biocompatibility after 90 days of experimentation; absence of inflammatory reaction; stability; and the presence of fibroblasts, chondrocytes and osteoid tissue, satisfying the need for cartilage reparation.
Finally, the main advantage is that, due to stability and auto-anchorage to bone and cartilaginous tissues, the implant does not require immobilization, allowing normal weight bearing immediately after surgery.

Figure 4 .
Figure 4. Toxicity of the samples.The positive control of toxicity was significantly different (p = 0.01) from the other samples.

Figure 5 .
Figure 5. Qualitative morphological assessment of the toxicity of the samples.a) Negative control (non-toxic), b) positive control (toxic); c) direct cytotoxicity, d) indirect cytotoxicity.

Figure 6 .
Figure 6.Three-dimensional projection of the tribological surface of the test body.

Figure 7 .
Figure 7. Implant of bioglass microfiber and polyurethane images performed with SEM: a) surface; b) the same image but with backscattered electrons showing the presence of NaCl in the porosities; c) Section view showing polyurethane matrix (A) and bioglass microfiber 45S5 (B).

Figure 10 .
Figure 10.Images of bone formation and space occupation in the porosity with space occupation after the first 15 days under different levels of magnification.The structure is similar to that of the trabeculae bone, which was considered biomimetic.

Figure 11 .
Figure 11.Bone formation occupying almost all the porosity 30 days after implantation under different levels of magnification.

Figure 12 .
Figure 12.Consolidation of bone formation at 90 days with densified filling of the porous region and osteon formation.

Table 1 .
Kinetic friction coefficient of cartilage versus materials.