Cell Viability of Porous Poly(d,l-lactic acid)/Vertically Aligned Carbon Nanotubes/Nanohydroxyapatite Scaffolds for Osteochondral Tissue Engineering

Treatment of articular cartilage lesions remains an important challenge. Frequently the bone located below the cartilage is also damaged, resulting in defects known as osteochondral lesions. Tissue engineering has emerged as a potential approach to treat cartilage and osteochondral defects. The principal challenge of osteochondral tissue engineering is to create a scaffold with potential to regenerate both cartilage and the subchondral bone involved, considering the intrinsic properties of each tissue. Recent nanocomposites based on the incorporation of nanoscale fillers into polymer matrix have shown promising results for the treatment of osteochondral defects. In this present study, it was performed using the recently developed methodologies (electrodeposition and immersion in simulated body fluid) to obtain porous superhydrophilic poly(d,l-lactic acid)/vertically aligned carbon nanotubes/nanohydroxyapatite (PDLLA/VACNT-O:nHAp) nanocomposite scaffolds, to analyze cell behavior and gene expression of chondrocytes, and then assess the applicability of this nanobiomaterial for osteochondral regenerative medicine. The results demonstrate that PDLLA/VACNT-O:nHAp nanocomposite supports chondrocytes adhesion and decreases type I Collagen mRNA expression. Therefore, these findings suggest the possibility of novel nanobiomaterial as a scaffold for osteochondral tissue engineering applications.


Introduction
The regeneration of articular cartilage injuries and defects remains as one of the most important challenges for orthopedic surgeons and researchers [1,2]. Due to its avascular nature, cartilage tissue Thus, herein was proposed the evaluation of porous PDLLA/CNT/nHAp scaffolds obtained from recently developed methodologies [41], in order to analyze their cell behavior and gene expression of chondrocytes. Therefore, it is aimed at to assess the applicability of this nanobiomaterial for osteochondral regenerative medicine.

Preparation of Superhydrophilic Vertically Aligned Multi-Walled Carbon Nanotubes Films (VAMWCNT-O)
Vertically aligned multi-walled carbon nanotubes (VAMWCNT) were produced using a microwave plasma chamber at 2.45 GHz (MWCVD), as described elsewhere [42]. Briefly, titanium (Ti) squares (10 mm), covered by a thin Fe layer (10 nm) deposited by an e-beam evaporator, were used as substrates. A pre-treatment was carried out for 5 min in N 2 /H 2 (10/90 cm 3 (STP) min −1 ) plasma, at 760 • C, in order to generate the catalyst for VAMWCNT growth. After this pre-treatment, CH 4 (14 cm 3 (STP) min −1 ) was inserted into the chamber for VAMWCNT nucleation. The pressure and temperature were kept around 800 • C and 30 Torr, respectively. Next, oxygen was incorporated by using a pulsed-direct current plasma reactor. Briefly, an oxygen flow rate of 1 cm 3 (STP) min −1 , at 85 mTorr, 700 V, frequency of 20 kHz and a time of treatment of 120 s was used. After oxygen plasma treatment, the samples were further referred as VAMWCNT-O.

Electrodeposition of nHA Crystals on the VAMWCNT-O (nHAp1)
nHAp crystals were electrodeposited onto the VAMWCNT-O films as previously reported [43], using Ca(NO 3 ) 2 ·4H 2 O and (NH 4 )·2HPO 4 electrolytes (pH = 4.8) at 0.042 mol·L −1 and 0.025 mol·L −1 , respectively. For the electrochemical measurements, a three-electrode cell apparatus coupled to an Autolab PGSTAT 128N equipment was used. VAMWCNT-O films were used as the working electrode. VAMWCNT-O films had a geometric area of 0.27 cm 2 (contact area with the electrolytic solution). A platinum coil wire was used as an auxiliary electrode, while an Ag/AgCl electrode was used as reference electrode. The production of the nHAp crystals occurred by simply applying a constant potential of −2.0 V for 30 min, and the solution temperature was kept at 70 • C in constant stirring.
The VAMWCNT-O samples were placed in plastic tubes and exposed to UV light in a biosafety chamber (BioProtector-12 Plus VECO, Campinas, Brazil) for 30 min. Next, 13 mL of SBF solution was added to the plastic tubes, which was stored in a refrigerated incubator (CT-712-R, Cientec, Porto Alegre, Brazil) under constant stirring (75 rpm at 36.5 • C) for 14 days. Finally, the samples were washed with deionized water (60 • C) and dried in an incubator (SP400, SPLABOR, Presidente Prudente Brazil) at 50 • C for 1 h [45,46].

Production of Porous PDLLA/VAMWCNT-O:nHAp Scaffolds
First, the nanoparticles of nHAp1 and nHAp2 from their Ti substrates were removed and dispersed in chloroform (0.3 wt%) under sonication (3 min, 1200 J mL −1 ). Thereafter, PDLLA (copolymer of L-lactide and D-lactide, 96/04, PLD-9655, Purasorb ® ) was dissolved in the VAMWCNT-O:nHAp/chloroform solution (10% w/v) (nHA1 and nHA2) under mechanical stirring for 120 min [47]. Subsequently, the solutions were placed into molds (0.5 mm in diameter) and kept under the following conditions: 80% controlled humidity (~80%) and room temperature. Next, a simple and fast functionalization to obtain superhydrophilicity in the porous polymer membranes was performed using a pulsed-direct current plasma reactor with an oxygen flow rate of 1 cm 3 (STP) min −1 , at a pressure of 85 mTorr, 700 V, at a repetition rate of 20 kHz [41]. A PDLLA membrane without any treatment or nanoparticles was prepared as a control.

Characterization of Porous PDLLA/VAMWCNT-O:nHAp Membranes
Scanning electron microscopes (SEM) were used to evaluate the composition and morphology of the porous membranes.

Cell Culture Experiments and Analysis
Healthy human chondrocyte cultures were isolated from normal articular cartilage of three young adults (age range: 26-36) undergoing ACL (anterior cruciate ligament) surgery. Small slices of cartilage were minced and digested overnight in 0.25% type I collagenase (Sigma-Aldrich, St. Louis, MO, USA), at 37 • C. The cells were then seeded onto tissue culture flasks, for expansion in monolayer, and kept as sub-confluent monolayers in growth medium, Dulbecco's modified Eagle's medium (DMEM) supplemented with 1.5 mL glutamine, 10% fetal bovine serum, and 100 units/mL penicillin-streptomycin (Gibco-Invitrogen, MA, USA). The incubation occurred in a humidifier at 37 • C and 5% CO 2 . The medium was replaced 3 times a week until cells reached 80% of confluence. Then, the cells were washed with phosphate buffer saline (PBS), removed with trypsin/ ethylenediaminetetra-acetic acid solution, and replaced at the same density until the third passage. The study was conducted in full accordance with local ethics guidelines, approved by the Ethics Committee (CEP/Hospital Israelita Albert Einstein nº10/1268; CAAE 0006.0.028.000-10), and cartilage samples were collected after obtaining written informed consent of the donors. Then, all samples were sterilized by UV light exposition for 2 h. Human chondrocytes were seeded in these membranes at a concentration of 5 × 10 5 cells/mL (50 µL of cell suspension) and incubated for five days. Chondrocytes cultivated on plastic and PDLLA were used as the control group.
To determine the cell adhesion and viability, the chondrocyte density in each membrane was observed. After five days in culture, attached cells on membranes were fixed with 4% paraformaldehyde solution for 15 min, washed three times in PBS buffer and stained with a 0.05% crystal violet solution for 5 min, washed with PBS twice and then analyzed by optical microscope (FSX 100, Olympus, Tokyo, Japan). Cells were counted using ImageJ software (NIH, Bethesda, MD, USA).
The morphology of chondrocytes on the different membranes was evaluated by SEM (Zeiss EVO MA 10 microscope, Oberkochen, German) after five days in the culture. Cells were fixed with 3% glutaraldehyde/0.1 M sodium cacodylate buffer for 1 h, followed by a dehydration step with 10 min of incubation in a graded ethanol solution series (30%, 50%, 70%, 95%, and 100%). Next, a drying stage with a 1:1 solution of ethanol/hexamethyldisilazane at room temperature was performed. Samples were coated with a thin gold film to improve image acquisition.

Cytotoxicity Assay
PDLLA, PDLLA/VAMWCNT-O:nHAp1 and PDLLA/VAMWCNT-O:nHAp2 scaffold-seeded cell were reacted with a lactate dehydrogenase (LDH) assay for the cytotoxicity test. After 24 h of culture, each group was analyzed by the LDH colorimetric assay kit (ab102526, Abcam) following the manufacturer's instructions. In brief, 50 µL of supernatant of each group and LDH positive control was mixed with the reaction mix (substrate), incubated at 37 • C for 30 min. The result was determined by the spectrophotometer at 450 nm. Total LDH was expressed as [OD sample −OD blank ]. Cells cultivated on plastic were used as the control. Results shown are representative of at least two independent experiments performed in triplicate and are expressed as mean ± SD (error bars) of three replicates.

Gene Expression Analysis
Relative quantification of mRNA expression of Sox-9, Aggrecan, MMP3, Type I and II Collagen was performed using Polymerase chain reaction quantitative real time (qRT-PCR). Total RNA from chondrocytes cultured in PDLLA, PDLLA/VAMWCNT-O:nHAp1, and PDLLA/VAMWCNT-O:nHAp2 membranes, for five days, was extracted by RNeasy kit (QIAGEN, Hilden, Germany) and reverse transcriptase reaction (QuantiTect Reverse Transcription kit, QIAGEN) was performed. qRT-PCR was carried out using an ABI7500 thermocycler (Applied Biosystems, Foster City, CA, USA) and the Maxima SYBR Green qPCR Master Mix (Thermo Fisher Scientific, Waltham, MA, USA), according to manufacturer's recommendations. Primer sequences and PCR parameters were provided upon request. Sequences were detailed on Table 1. Expression of target genes was normalized by β-actin mRNA levels measured concurrently. The level of expression was then calculated as 2-∆∆Ct and expressed as the mean. All quantitative RT-qPCR results were representative of at least two independent experiments, each with three technical replicates and are expressed as mean ± SD (error bars).

Statistics Analysis
Data were expressed as mean ± SD. Statistical significance was determined by a two-tailed unpaired t test, one-way ANOVA in Graph Pad Prism 6 ® software (GraphPad, CA, USA). Significance was determined at p < 0.05. Figure 1 shows micrographs of the produced porous PDLLA (Figure 1a), PDLLA/VAMWCNT-O:nHAp1 (Figure 1b), and PDLLA/VAMWCNT-O:nHAp2 (Figure 1c) membranes. The differences can be clearly observed as regards to the surface texture/morphology after the nanoparticles were incorporated. Figure 1a illustrates the smaller pores (~11-18 µm) on the surface of PDLLA. Conversely, the loading of both nHAp1 and nHAp2 led to larger pores on the surface (~27-39 µm, Figure 1b,c, respectively).

Results and Discussion
Different composites using PDLLA have been described as biocompatible support materials for cell culture, focusing especially on their applicability in bone tissue engineering [41,47,48]. Moreover, many studies using PDLLA/HAp composites have been investigated employing different processes [49][50][51][52][53]. The incorporation of HAp and/or CNT in PDLLA polymer matrices is an important strategy to induce cell differentiation and ossification. Recent studies in the research group showed the biocompatibility of nHAp obtained by the method of immersion in SBF and also by electrodeposition [46,54]. Cell viability is one of the preliminary tests conducted to demonstrate the possible cytotoxicity effects of a biomaterial candidate to be used as scaffold. The LDH test can show death and cell lysis after 24 h of culture and in a single incubation time (5 days); it is a biocompatibility test widely used for basic research on biomaterials, mainly involving osteochondral applications. Accordingly, the analysis of lactate dehydrogenase (LDH) showed that all membranes were non-toxic to human chondrocytes ( Figure 2). This result illustrated that the different combinations of PDLLA and hydroxyapatite can alter the viability of human chondrocytes similar to that of the control (cells cultivated at plastic).  Cell viability is one of the preliminary tests conducted to demonstrate the possible cytotoxicity effects of a biomaterial candidate to be used as scaffold. The LDH test can show death and cell lysis after 24 h of culture and in a single incubation time (5 days); it is a biocompatibility test widely used for basic research on biomaterials, mainly involving osteochondral applications. Accordingly, the analysis of lactate dehydrogenase (LDH) showed that all membranes were non-toxic to human chondrocytes ( Figure 2). This result illustrated that the different combinations of PDLLA and hydroxyapatite can alter the viability of human chondrocytes similar to that of the control (cells cultivated at plastic). Cell viability is one of the preliminary tests conducted to demonstrate the possible cytotoxicity effects of a biomaterial candidate to be used as scaffold. The LDH test can show death and cell lysis after 24 h of culture and in a single incubation time (5 days); it is a biocompatibility test widely used for basic research on biomaterials, mainly involving osteochondral applications. Accordingly, the analysis of lactate dehydrogenase (LDH) showed that all membranes were non-toxic to human chondrocytes ( Figure 2). This result illustrated that the different combinations of PDLLA and hydroxyapatite can alter the viability of human chondrocytes similar to that of the control (cells cultivated at plastic).    (Figure 4b,c). The chondrocytes protruded into the interior of membrane pores, occupying the full diameter and forming a cellular layer.  The expression of typical chondrogenic markers was evaluated by PCR analysis. Figure 5       Conversely, the different PDLLA preparation did not alter the expression levels of the other genes analyzed. The absence of change in all sets of genes can be elucidated by a short period of time and a low number of samples analyzed. These results corroborate with other researches that reported that the detection of the type II Collagen and Aggrecan mRNA expression only occurs after 7-14 days of the cell culture on biomaterial [55,56]. Stenhamre et al. (2013) reported that the chondrocytes re-differentiation maintained fibroblastoid format after 7 days of culture when cultivated on 3D nanofibers mats [57]. Similar results were attained after 5 days of cultivation on porous membranes with incorporated nanoparticles present. The lower type I Collagen mRNA expression observed in chondrocytes cultivated on porous PDLLA/VAMWCNT-O:nHAp1 and PDLLA/VAMWCNT-O:nHAp2 membranes suggest that, that is the first step of re-differentiation state, shifting the microenvironment similar to health cartilage.
In monolayer culture on a substrate, chondrocytes lose their spherical morphology and acquire a fibroblast-like format. This expansion causes a process of dedifferentiation characterized by decreased synthesis of type II collagen and increased synthesis of type I collagen [58][59][60]. The re-differentiation process, which promotes cartilage regeneration, depends on several factors such as time of culture, physical and chemical characteristics of the scaffold [57]. However, the controlled polymer degradation and the porosity to allow free exchange of body fluids, can promote the rescue to the spherical format and phenotype characteristic of chondrocytes [61].
The results showed that cells adhered on the porous membranes (Figures 3 and 4), and do not present cytotoxic effects ( Figure 2). However, the cells behave differently in porous PDLLA/VAMWCNT-O:nHAp membranes, the results showed an increase in adhesion and formation of monolayer cells in the PDLLA/VAMWCNT-O:nHAp membranes. It is probably due to the increase in porosity of surfaces, similar behavior has been previously described by Siqueira et al. (2015) in bone regeneration. Besides that, others researchers demonstrated the influence of surface wettability and increased porosity on cell adherence behavior and re-differentiation of chondrocytes because these surface properties improve the cell adhesion capacity [61][62][63][64][65][66][67].
This paper has shown that both the porous PDLLA/VAMWCNT-O:nHAp1 and PDLLA/VAMWCNT-O:nHAp2 membranes promote chondrocytes adhesion (Figures 3 and 4), non-cytotoxic effects ( Figure 2) and decrease the type I Collagen mRNA expression ( Figure 5). Therefore, the clinical implication of this study was that since previously these nanobiomaterial had already demonstrated biocompatibility with osteoblasts [41], the findings are very encouraging for its use as scaffolds for the treatment of osteochondral defects. Moreover, this study contributes to the knowledge of the chondrocytes behavior on different biomaterials.
In the future, complementary studies, including increases in sample number, longer periods of time culture and an in vivo evaluation, will be performed before translation to clinical use.

Conclusions
The spreading, adhesion, and gene expression of human chondrocytes cultivated on two different compositions of membranes was performed. The porosity and roughness in PDLLA/VAMWCNT-O:nHAp were associated to partial hydrophobicity control after a simple and fast oxygen plasma etching due to oxygen groups attached on the surface. All these surface controls promote human chondrocytes adhesion, decrease type I Collagen mRNA expression and non-cytotoxic effects. However, although these results suggest porous PDLLA/VAMWCNT-O:nHAp1 and PDLLA/VAMWCNT-O:nHAp2 membranes as potential alternatives in osteochondral repair, additional in vivo assays will be necessary to fully elucidate the clinical implication of our observations.