3D Printing of PVA/Hexagonal Boron Nitride/Bacterial Cellulose Composite Scaffolds for Bone Tissue Engineering

In this study, a novel Polyvinyl Alcohol (PVA)/Hexagonal Boron Nitride (hBN)/Bacterial Cellulose (BC) composite, bone tissue scaffolds were fabricated using 3D printing technology. The printed scaffolds were characterized by fourier transform infrared spectroscopy (FT-IR), scanning electron microscopy (SEM), tensile testing, swelling behavior, differential scanning calorimetry (DSC), and in vitro cell culture assay. Results demonstrated that bacterial cellulose addition affected the characteristic properties of the blends. Morphological studies revealed the homogenous dispersion of the bacterial cellulose within the 12 wt%PVA/0.25 wt%hBN matrix. Tensile strength of the scaffolds was decreased with the incorporation of BC and 12 wt%PVA/0.25 wt%hBN/0.5 wt%BC had the highest elongation at break value (93%). A significant increase in human osteoblast cell viability on 3D scaffolds was observed for 12 wt%PVA/0.25 wt%hBN/0.5 wt%BC. Cell morphology on composite scaffolds showed that bacterial cellulose doped scaffolds appeared to adhere to the cells. The present work deduced that bacterial cellulose doped 3D printed scaffolds with welldefined porous structures have considerable potential as a suitable tissue scaffold for bone tissue engineering (BTE).

occurring in human bone, regeneration or healing of bone tissue is carried out with threedimensional structures capable of mimicking the physical and chemical properties of the extracellular matrix (ECM) [2]. Bone grafting and transplantation is the current clinical assistance. There is an alternative to the conventional use of bone grafts which is engineered bone tissues. Tissue-engineered bone has a limitless supply, and it does not allow disease transmission. However, there are some limitations or difficulties in terms of clinical practice.
To overcome these challenges, bone tissue engineering (BTE) aims to regenerate bones via combination of cells, biomaterials and factor therapy [3]. BTE has become one of the favourite areas to facilitate bone regeneration and the treatment of diseased or damaged tissues. BTE leads regeneration of targeted tissues by combining osteoconductive scaffolds, osteogenic cells, and osteoinductive signals.
Tissue-engineered scaffolds enable cells attachment and effective tissue regeneration. Thus, tissue-engineered scaffolds should have necessary properties, which are biocompatibility, biodegradability, proper mechanical properties, and interconnected porous structures.
Fabrication technique also has a critical effect on the properties of the aimed artificial scaffolds. There are some traditional fabrication methods such as gas foaming, freeze-drying, fiber bonding, particulate/salt leaching, emulsification and phase separation/inversion [4].
These methods may have some problems with the control of the geometry, porosity and pore shape. 3D printing is a novel method to overcome these disadvantages of traditional methods.
This new technology has been used to build scaffolds with designed shapes and porosity Moreover, hBN addition also increased the thermal stability and swelling degree of the composite hydrogels [8]. Zaboroska et al. study investigated both nanoporous and microporous bacterial cellulose effect on NC3T3-E1 osteoprogenitor cells. This work, also reported that microporous BC is a promising biomaterial for bone tissue regeneration [9]. The effect of varying composition of PVA, hexagonal boron nitride and bacterial cellulose have been studied separately for bone tissue engineering. Still, no previous studies have been reported using these polymers together to obtain bone tissue scaffold.
In this study, 3D bioprinting technology was managed to produce 3D porous bone scaffolds that are compatible with bone physical and mechnanical structure, using polyvinyl alcohol (PVA)-hexagonal boron nitride (hBN)-bacterial cellulose (BC) polymeric blends to get an ideal scaffold for effective bone healing. The amount of bacterial cellulose in PVA/hBN matrix was investigated with different BC contents due to its superior biocompatible properties, crystallinity, non-toxicity, and hydrophilicity [10]. It has outstanding effects on cell adhesion with its interconnected porous structure [11,12]. PVA can provide desirable flexibility, hydrophilicity, outstanding chemical stability, and semi-permeability to the scaffolds as host material, which are crucial for transporting oxygen and nutrients for cell survival [13]. The hBN has been used as fillers due to its excellent thermal conductivity, thermal stability, and superior mechanical properties. Additionally, hBN has been used in biomedical applications as a promising material due to its excellent biocompatibility property.

Bacterial Cellulose Production by Gluconacetobacter Xylinus
Gluconacetobacter xylinus was used to produce bacterial cellulose membranes (Fig. S1a). The bacterium was cultivated on Hestrin and Schramm (HS) medium consisted of 2.0 wt% Dglucose, 0.5 wt% peptone, 0.5 wt% yeast extract, 0.27 wt% disodium hydrogen phosphate and 0.115 wt.% citric acid, and the pH at 5.0-6.0 with acetic acid. After incubating in a static incubator at 28 °C for 10 days, the BC membranes were purified by 0.1 M NaOH at 90 °C for 2 hours and soaked them in deionized water (DI) several times. The samples were sterilized at 121°C in an autoclave for 20 min, followed by oven-dried at 50 °C for 24 h. Oven-dried bacterial cellulose membranes were treated with 50% (v/v) sulphuric acid solution in a cellulose/acid ratio of approximately 20 g/L, at 50 °C for 2 h. The suspension was washed with deionized water and centrifuged and repeated until neutral pH (as shown in Fig. S1(b, c, d, e)). software, and this design was converted into the G-code via Simplify program, which provides to control the process parameters programmatically ( Fig. 1 (a, b, c)).   (Table   1). In this work, several concentrations of PVA solutions were used, but 12 wt%PVA was found best printable to form uniform pore distribution for determining optimum printable parameters with given parameters. After the PVA concentration was obtained, hBN particles J o u r n a l P r e -p r o o f were added into 12 wt%PVA matrix with a ratio of 0.25 wt% to increase the mechanical strength of the scaffolds. BC was added with three different rates to observe the BC effect on the properties of the scaffolds, especially cell viability and attachment. There was not used any chemical cross-linking agents to avoid their toxic effects. In the printing stage, 12 wt%PVA was fabricated at first, and the parameters during the printing process were 70% infill rate, 0.3 mL/hr flow rate at room temperature. 12 wt%PVA/0.25 wt%hBN was printed with same infill rate as 12 wt%PVA, but 0.5 mL/hr flow rate was reported for this stage. With the addition of 0.1, 0.25, 0.5 wt%BC into the 12 wt%PVA/0.25 wt%hBN, the value of flow rate was adjusted to 0.5, 0.8, and 1 mL/hr, respectively. After whole scaffolds were obtained and dried at room temperature for 24 hours, several characterization tests were performed to determine the properties of the printed scaffolds. Transform Infrared Spectroscopy (FT-IR) was used to analyze the presence of chemical groups between the components by using FT-IR-4700, Jasco which wavelength value ranges between 450 and 4000 cm -1 . Mechanical properties of the composite scaffolds were examined by using tensile test device (Shimadzu, Japan) with 5 kN load cell. Grup movement speed was adjusted to 5 mm/min, and two samples were selected for each concentration to increase the analysis, they were coated with gold for 90 seconds, 18 mA. The swelling behaviours of the composite scaffolds were tested with phosphate-buffered saline (PBS) for 7 days using thermo-shaker (BIOSAN) at 37 °C, 250 rpm. Before the swelling degree measurement, the dry weight of the scaffolds was measured and recorded. After that, they were placed in eppendorf tubes with PBS. Dry measurements were taken each day and after composite scaffolds were wiped with filter paper to remove the saline on the surface. The swelling degrees of the composite scaffolds were calculated with the formula used in the Li et al. study [8].

6. Statistical Analysis
All data were demonstrated as the mean ±standart deviation. All experiments were carried out in duplicates. Their averages were taken as the final value and differences among groups were considered significant at p < 0.05.
When compared to the peaks of the other spectrums, only one different peak (~1140.7 cm -1 ) was observed for 12 wt%PVA/0.25 wt%hBN/0.25 wt%BC (Fig. 2B(d)). There were observed small shifts compared to the other spectrums for 12 wt%PVA/0.25 wt%hBN/0.5 wt%BC composite scaffolds, and one peak was only seen in this spectrum, which was absorbed in ~1647.9 cm -1 (Fig. 2B(e)).

SEM Observations
SEM images of the composites were shown in Fig. 3 with labels. In Fig. 3a, 12 wt%PVA had a uniform structure and homogeneous pore distribution. The results indicated that the 3D printed composite scaffolds could be printed steadily over somewhat large areas and heights [17]. The average pore size value was about 291.71±19.94 µm. In Fig. 3b, 12 wt%PVA/0.25 wt%hBN composite scaffolds showed a relatively smooth surface, and its mean pore size value was 290.18±26.80 µm. In Fig. 3c, with 0.5 wt%BC addition, 12 wt%PVA/0.25 wt%hBN/0.5 wt%BC had smaller pores of 265.68±15.39 µm. In tissue engineering applications, pore size usually ranges from 150 to 500 µm [18][19][20]. It can be said that the pore size of synthesized scaffolds has a proper range of pore values to provide vascularization and nutrient transport for tissue engineering applications.

Uniaxial Tensile Testing
The typical mechanical properties of the printed scaffolds like tensile strength and elongation at break values were listed in Table 1 and showed in wt%PVA, but it has a huge standard deviation which may be due to different extrusion speeds of the prepared scaffolds (0.8 mL/h). Elongation at break values had generally increased with the incorporation of the additives as can be seen from Table 1. Bacterial cellulose and hBN increased the ductile nature of the scaffolds. Reason for this can be not only due to hydrogen bonds owing to intra-action of deposited gel but the interaction between deposited layers [8,9]. The swelling behaviour of the composite scaffolds was shown in Fig. 6. According to the results, the maximum swelling degree was observed for 12 wt%PVA. Other composites also exhibited an increased swelling degree during the time up to 5 th day. However, these values were lower than the matrix polymer (12 wt%PVA). It can be realized that swelling degree of BC free composites of 12 wt%PVA was higher than of the BC added composites. Also, hBN addition, as in the literature, increased the swelling properties when compared to BC added cases [8]. It was also observed that swelling degree increased with time for the first 120 h and then presented a decreasing tendency until the end of the experiment (168 h). In the bacterial cellulose additive groups, the swelling degree decreased with the increased amount of bacterial cellulose. The reason for this tendency could be the number of free spaces into 12 wt%PVA/0.25 wt%hBN composite scaffolds, because the free spaces within the matrix may be decreased with the incorporation of BC that were released during the uptake of the water inside the printed filaments. The high rigid structure in BC content groups not only decreases the water absorption and permeation but also limited the swelling behaviour of materials [27].
The reason for this decreasing behaviour might be due to strong intramolecular bonds occurred between functional groups of PVA and BC [28].

Conclusions
In this study, the effects of incorporation of bacterial cellulose into the host 12 wt%PVA/0.25 wt%hBN composite scaffolds in different ratios were examined. It was observed that the desired pore sizes and the homogeneous scaffold structures were obtained with threedimensional printing of the polymeric blends. According to DSC analysis, it was found that additive materials did not disrupt the crystal structure of PVA. The significant increase was