Reinforcement of Injectable Hydrogel for Meniscus Tissue Engineering by Using Cellulose Nanofiber from Cassava Pulp

Injectable hydrogels can be applied to treat damaged meniscus in minimally invasive conditions. Generally, injectable hydrogels can be prepared from various polymers such as polycaprolactone (PCL) and poly (N-isopropylacrylamide) (PNIPAAm). Poly (ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) block copolymer-diacrylate (PEO-PPO-PEO-DA) is an interesting polymer due to its biodegradability and can be prepared as water-insoluble injectable hydrogel after curing with UV light at low intensity. However, mechanical and cell adhesion properties are not optimal for these hydrogels. For the improved mechanical performance of the injectable hydrogel, cellulose nanofiber (CNF) extracted from cassava pulp was used as a reinforcing filler in this study. In addition, gelatin methacrylate (GelMA), the denatured form of collagen was used to enhance cell adhesion. PEO-PPO-PEO-DA/CNF/GelMA injectable hydrogels were prepared with 2-hydroxy-1-(4-(hydroxy ethoxy) phenyl)-2-methyl-1-propanone as a photoinitiator and then cured with UV light, 365 nm at 6 mW/cm2. Physicochemical characteristics of the hydrogels and hydrogels with CNF were studied in detail including morphology characterization, pore size diameter, porosity, mechanical properties, water uptake, and swelling. In addition, cell viability was also studied. CNF-reinforced injectable hydrogels were successfully prepared after curing with UV light within 10 min with a thickness of 2 mm. CNF significantly improved the mechanical characteristics of injectable hydrogels. The incorporation of GelMA into the injectable hydrogels improved the viability of human cartilage stem/progenitor cells. At optimum formulation, 12%PEO-PPO-PEO-DA/0.5%CNF/3%GelMA injectable hydrogels significantly promoted cell viability (>80%) and also showed good physicochemical properties, which met tissue engineering requirements. In summary, this work shows that these novel injectable hydrogels have the potential for meniscus tissue engineering.


Introduction
Scaffolds for tissue engineering act as structural support to repair damaged tissues by providing cell adhesion and proliferation, and subsequent development of new tissue. Currently, marketed scaffolds for tissue engineering are in various forms such as sponge, nanofibrous, 3D printing, and solid hydrogel scaffolds. They are typically produced in porous solid templates before implantation into the target tissue. The benefit of solid scaffold forming before implantation is that it is a simple and not complicated process. proline and hydroxyproline) sequences. The arginylglycylaspartic acid (RGD) motif, responsible for cell adhesion to ECM and cell proliferation is also present in the gelatin [12]. In addition, many studies reported that GelMA represents one of the most promising candidates for the development of scaffold material for tissue repair owing to its various advantages: water-insoluble, enhancement of cell adhesion and proliferation, biodegradability, non-toxicity, and promising in vitro and in vivo tissue regeneration [13]. Celikkin, N. et al. (2017) [14] studied the effect of GelMA concentration (5 and 10% w/v) on osteogenic differentiation in rat mesenchymal stem cells (MSCs). They discussed that MSCs in both 5% (w/v) GelMA and 10% (w/v) GelMA hydrogels showed good osteogenic differentiation. Especially, 5% (w/v) GelMA hydrogel encouraged higher MSC attachment. Nowadays, there are very few reports on the development of injectable hydrogels for meniscus tissue engineering [15][16][17], especially; there are no reports on CNF-reinforced PEO-PPO-PEO-DA/GelMA injectable hydrogel for meniscus tissue engineering. In this work, we have developed CNF-reinforced PEO-PPO-PEO-DA/GelMA injectable hydrogels for meniscus tissue engineering. The effect of CNF and GelMA on the physicochemical properties of a hydrogel, such as morphology, pore size diameter, porosity, mechanical properties, water uptake and swelling, gel fraction, and chemical interaction were investigated. In addition, cell cytotoxicity and proliferation were also studied. The overview of the study design is shown in Figure 1.

Preparation of CNF
Cassava pulp (CP) was ground and sieved with a mesh size of 150-250 µm. Then, an oven was used to dry CP at 120 • C for 24 h to remove excess moisture. After that, 200 g of CP was soaked in 4 L of 4 wt% sodium hydroxide (NaOH) solution at 80 • C for 2 h under mechanical stirring and then washed with deionized water several times until reaching a pH of 7. Finally, the CP after alkali treatment was dried for 1 day in an oven at 80 • C. Alkali-treated cassava pulp (TCP) was produced following alkali treatment. After alkali treatment, the TCP was mixed with a solution of acetate buffer (27 g NaOH and 75 mL glacial acetic acid, diluted to 1 L of distilled water) and aqueous chlorite (1.7% (w/v) NaClO 2 in water) in equal amounts to conduct the bleaching treatment. The ratio of TCP and bleaching solution was 1:20. Bleaching was performed at 80 • C for 6 h. Following bleaching, TCP was washed repeatedly in deionized water until the pH of the bleached TCP became neutral. The bleached TCP was put into an oven to dry at 80 • C for 24 h and then grinding and sieving processes were conducted with a mesh size of 38-63 µm. The sample resulting from the bleaching treatment was called bleached cassava pulp (BCP). CNF was prepared by an application of the high-pressure homogenizing process. Briefly, 0.5% (w/v) BCP aqueous dispersion was dispersed in phosphate-buffered saline (PBS) solution. The suspension was treated with a high-pressure homogenizer (Microfluidics M-110EH-30, Microfluidics International Corporation, Westwood, MA, USA) to separate the fibril bundles from BCP. The homogenization pressure was adjusted to 25,000 psi, and the homogenization number of cycles was 15 cycles.

Preparation of CNF-Reinforced PEO-PPO-PEO-DA/GelMA Injectable Hydrogels
To find the optimum CNF concentration for reinforcing in PEO-PPO-PEO-DA hydrogels, the physicochemical properties of injectable hydrogels were investigated using varying CNF concentrations. The hydrogels were prepared by solution mixing. In the first step, 0.3% (w/v) IR2959 as a photoinitiator was added into CNF suspension at 60 • C under stirring of 500 rpm, for 30 min. The CNF was added with the following concentrations: 0, 0.3, 0.5, and 0.7% (w/v). 12% (w/v) PEO-PPO-PEO-DA was dissolved in the CNF suspension at 4 • C for 24 h. After that, the liquid mixtures were stirred for 30 min, 500 rpm at room temperature. The liquid mixtures were injected into the mold and cured with UV light, 365 nm at 6 mW/cm 2 for 10 min, after curing the physicochemical properties of hydrogels were characterized.
After that, the optimum CNF concentrations were selected to prepare CNF/PEO-PPO-PEO-DA/GelMA injectable hydrogels. In the first step, two times of using IR2959 or 0.6% (w/v) IR2959 as a photo initiator was added into CNF suspension at 60 • C under stirring of 500 rpm for 30 min. Two times using PEO-PPO-PEO-DA or 24% (w/v) PEO-PPO-PEO-DA was dissolved in the CNF/IR2959 suspension at 4 • C for 24 h. GelMA with various as 2, 6, and 10% (w/v) was dissolved in the CNF/IR2959 suspension at 60 • C for 1 h. The prepared PEO-PPO-PEO-DA in CNF/IR2959 suspension and prepared GelMA in CNF/IR2959 suspension were mixed for 30 min, 500 rpm at room temperature. Finally, the mixtures gave IR2959 of 0.3% (w/v), PEO-PPO-PEO-DA concentration of 12% (w/v), and GelMA concentration of 1, 3, and 5% (w/v), respectively. The liquid mixtures were injected into the mold and cured with UV light, 365 nm at 6 mW/cm 2 for 10 min. After curing, the physicochemical properties of injectable hydrogels and cell viability assessment were characterized.

Preparation of Fibrin Glue
In this study, CNF-reinforced PEO-PPO-PEO-DA injectable hydrogels were compared with the injectable fibrin glue. The fibrin glue was prepared according to Numpaisal et al. (2016) [18]. Briefly, fibrinogen, 20 mg/mL was dissolved in PBS. Thrombin and CaCl 2 , with final concentration values of 5.3 U/mL and 4.12 mM, respectively, were added for initiating polymerization to an end fibrinogen concentration of 10 mg/mL. Both fibrinogen and thrombin/CaCl 2 solutions were injected into the mold, and the fibrin glue was allowed for polymerization at 37 • C.
2.5. Characterization of Cellulose Nanofiber 2.5.1. Morphology, Diameter, and Length A field emission scanning electron microscope (FE-SEM) (Carl Zeiss AURIGA ® , Thuringia, Germany) was employed for the morphology and diameter of CNF. Samples were sputter-coated with carbon on the surface. Images were acquired with a voltage of 5 kV. The CNF size was determined by ImageJ software (Wayne Rasband NIH, Washington, DC, USA). FE-TEM (Talos™ F200X, Thermo Fisher Scientific Inc., Waltham, MA, USA) was applied to determine the length of CNF. A drop of the diluted suspension was cast onto the surface of a clean copper grid and covered with a thin carbon film. TEM images of the CNF were negatively stained in Uranyless for 1 min, then washed in distilled water. Drying the sample at normal temperature was vital before FE-TEM analysis. The length of CNF was also measured by ImageJ software in 100 random fibers, and the diameter distribution curves of CNF were plotted by OriginLab software, (OriginPro ® 2021b, OriginLab Corporation, Northampton, MA, USA).

Structure
The structural and chemical composition of CP, BCP, and CNF were characterized by a Tensor 27 FTIR spectrometer (Bruker Optics, Ettlingen, Germany). FTIR was performed in the spectra range of 4000-400 cm −1 .

Crystallinity
X-ray diffraction was used for crystallinity assessment of CP, BCP, and CNF's determination. Each sample in the form of milled powder was placed on the sample holder and flattened for uniform X-ray exposure. Sample analysis was carried out by an X-ray diffractometer (D8-Advance Bruker AXS GmbH, Ettlingen, Germany) at RT with a monochromatic CuK radiation source (λ = 0.1539 nm) in the step-scan mode with a 0.2 • angle ranging from 5 • to 40 • with a step of 0.02 and scanning time of 0.5 min. The Segal method was applied to obtain the crystallinity index (CrI) of the [19], as shown in Equation (1).
I 200 the maximum intensity of the (2 0 0) reflection at 2θ of 22 • -23 • , and I am is the lowest intensity of diffraction at 2θ of 18 • -19 • for the amorphous part. The morphology and pore size of the top surface and the cross-section of the hydrogels were assessed by FE-SEM, 5 kV voltage, (Carl Zeiss AURIGA ® , Thuringia, Germany), after gold sputtering. The pore size diameter of samples was determined by ImageJ software (Wayne Rasband NIH, Washington, DC, USA) in 100 random pores, and the pore size diameter distribution diagram of hydrogel samples was plotted by OriginLab software (OriginPro ® 2021b, OriginLab Corporation, Northampton, MA, USA).

Porosity
Hydrogel porosity was assessed by means of the liquid displacement method. Briefly, the lyophilized hydrogel was soaked in a defined volume (V 1 ) of ethanol in a 5 mL cylinder for 10 min. The total volume of ethanol and sample was recorded as V 2 . The ethanolimpregnated sample was removed from the cylinder, and the residual volume of ethanol was recorded as V 3 . The porosity was calculated according to Equation (2). The sample number was three per group (n = 3).

Gel Fraction
Hydrogels were dried in an oven at 40 • C for one day and then weighed (dry weight before immersion). After drying, the hydrogels were hydrated in distilled water at 37 • C for 24 h to remove the soluble fraction and dried at 40 • C for 24 h (dry weight after immersion), then the gel fraction of hydrogel samples was calculated as shown in Equation (3). Experiments were carried out in triplicate (n = 3).

Water Uptake and Swelling
The hydrogels were left soaking in distilled water at 37 • C for one day and then weighed (wet weight). The wet hydrogels were then dried in an oven at 40 • C for 24 h (dry weight after immersion). The water uptake and swelling of the hydrogels were obtained according to the following Equations (4) and (5), respectively. All experiments were performed in triplicate (n = 3).

Mechanical Properties
The compressive strength of the hydrogels (diameter = 5 mm and height = 4 mm) was evaluated with a texture analyzer (TA.XT-PLUS, London, UK) with a load cell of 5 kg, a compression speed of 3 mm/min and pressure distance 80% strain. The ultimate tensile strength and elongation at break of the hydrogel were also determined using a texture analyzer (TA.XT-PLUS, Texture Technologies Corp., London, UK) with a load cell of 5 kg. The samples (length = 30 mm, width = 5 mm, and thickness = 1.5 mm) were attached to the central part of the tensile grip jaw. The upper sample (10 mm) and the lower sample (10 mm) were gripped by the tensile grip jaw and finally, a gauge length for tensile testing was 10 mm. The tensile tests were performed at a speed rate of 10 mm/min. All the tests represented the average results from three test specimens (n = 3). The extract dilution method is more commonly adopted for the in vitro cytotoxicity evaluation of materials and devices used in the body since it can be applied to a wide variety of raw materials and finished products that may release toxins from exposed surfaces (un-crosslinked polymer solutions and residual crosslinker). The liquid samples were autoclaved at 121 • C, 15 psi pressure for 20 min. The liquid samples were pipetted into a 96-well plate and then induced into the hydrogels by UV irradiation, 365 nm at 6 mW/cm 2 for 10 min in a laminar flow cabinet. Cytotoxicity was determined by human cartilage stem/progenitor cell (CSPCs) viability after incubation in an extracted medium of the hydrogels for 24 h. The cell viability was assessed with an MTT (3-[4,5-dimethylthiazol-2-yl]-2,5 diphenyl tetrazolium bromide) assay. The extracted medium of the hydrogels was derived from hydrogel immersion in cell culture medium (Dulbecco's modified Eagle's medium (DMEM)-high glucose, 5% PLTGold ® Human platelet lysate, 1% L-glutamax and 1% antibiotic) for 24 h. The CSPCs were seeded onto 96-well plates at 2.5 × 10 4 cells per well in 200 µL of a culture medium and incubated at 37 • C, 5% CO 2 for 24 h. After 24 h of cell incubation, the old culture medium was removed and 200 µL of the extracted medium of the hydrogels was then added to the 96-well plates. The cells seeded 96-well plates were incubated in an extracted medium of the hydrogels for up to 24 h at 37 • C, 5% CO 2 . Then, the cells were cleaned with PBS two times, and 100 µL of MTT solution (5 mg MTT/mL of medium) was added. After a 2 h reaction time, MTT formazan was extracted with dimethyl sulfoxide (DMSO) for 10 min, followed by measuring the absorbance of the extracted medium at 570 nm with a microplate reader (Infinite M200Pro, Tecan Group Ltd., Männedorf, Switzerland). All the results were expressed as relative viability compared to cells grown in the culture medium without hydrogel immersion (control). The cell viability percentage was calculated by dividing MTT formazan absorbance for each sample by the absorbance formazan for control and multiplying this number per 100 as in Equation (6). Assays were carried out in triplicate for each sample (n = 3).

Cell Proliferation
Cell proliferation in hydrogels was determined by CSPCs viability at days 1, 7, and 14. The MTT assay was used for cell viability studies. Samples were autoclaved at 121 • C, 15 psi pressure for 20 min, and pipetted into a 96-well plate and then induced into the hydrogels by UV irradiation, 365 nm at 6 mW/cm 2 for 10 min in a laminar flow cabinet. The CSPCs were seeded onto the hydrogels in 96-well plates at 2.5 × 10 4 cells per well in 200 µL of cell culture medium. Culture of cell seeded hydrogels was performed in DMEM-high glucose, 5% PLTGold ® human platelet lysate, 1% L-glutamax, and 1% antibiotic, put in 37 • C with a 5% CO 2 humidified incubator. After the cell culture in the hydrogels for 1, 7, and 14 days, the cells were washed with PBS twice, and 100 µL of MTT solution (5 mg MTT/mL of medium) was added. MTT formazan was extracted 2 h later with dimethyl sulfoxide (DMSO) for 10 min, followed by measuring absorbance of the extracted solution at 570 nm with a microplate reader (Infinite M200Pro, Tecan Group Ltd., Männedorf, Switzerland). All the results were expressed as relative viability compared to cells grown in the culture medium without hydrogel immersion (control). The cell viability percentage was calculated by dividing MTT formazan absorbance for each sample by the absorbance formazan for control and multiplying this number per 100 as in Equation (6). Assays were performed in triplicate for each sample (n = 3).

Statistical Analysis
The results were expressed as mean ± standard deviation. All quantitative data were analyzed with one-way ANOVA and paired t-test followed by Tukey's post hoc comparison test. To identify statistical differences between the comparison groups, a p-value < 0.05 was considered statistically significant. the culture medium without hydrogel immersion (control). The cell viability percentage was calculated by dividing MTT formazan absorbance for each sample by the absorbance formazan for control and multiplying this number per 100 as in Equation (6). Assays were performed in triplicate for each sample (n = 3).

Statistical Analysis
The results were expressed as mean ± standard deviation. All quantitative data were analyzed with one-way ANOVA and paired t-test followed by Tukey's post hoc comparison test. To identify statistical differences between the comparison groups, a p-value < 0.05 was considered statistically significant.   Figure 3a shows the FTIR spectra of CP, BCP, and CNF with homogenizing technique. The peaks between 3600-3000 cm −1 of all samples corresponded to -OH stretching. This band was related to intermolecular hydrogen bonds of cellulose [15,16]. The peaks at 1318 cm −1 , 1204 cm −1 , and 898 cm −1 of BCP and CNF were attributed to CH2 wag, OH-deformation, and beta glycosidic linkages of the glucose ring; these bands were referred to the structure of the cellulose components [20,21]. The peaks at 1732 cm −1 and 1505 cm −1 of raw cassava pulp corresponded to C=O stretching and C=C aromatic ring; these referred to lignin and hemicellulose [20,21]. Apparently, these peaks disappeared in the pretreated cassava pulp and CNF, indicating partial removal of hemicelluloses and lignin from the structure of fibers leading to higher purity of cellulose. Furthermore, no significant differences were observed among the spectra of BCP and CNF. The results proved that the cellulose molecular structure remained unchanged with the homogenizing method.

Characterization of Cellulose Nanofiber
From the XRD curve shown in Figure 3b, the crystalline peaks of all samples showed 2θ at 16° and 22°. The crystalline peak of the CP had the lowest intensity. From the XRD pattern, the crystallinity index (CI) of all samples could be determined. The Segal method was used to calculate the CI with the principal diffraction peak of cellulose I around 22° and the lowest intensity at around 18°. The CP showed the lowest CI of 19%, since it contains a great amount of amorphous area. On the other hand, the BCP and CNF give high  Figure 3a shows the FTIR spectra of CP, BCP, and CNF with homogenizing technique. The peaks between 3600-3000 cm −1 of all samples corresponded to -OH stretching. This band was related to intermolecular hydrogen bonds of cellulose [15,16]. The peaks at 1318 cm −1 , 1204 cm −1 , and 898 cm −1 of BCP and CNF were attributed to CH 2 wag, OHdeformation, and beta glycosidic linkages of the glucose ring; these bands were referred to the structure of the cellulose components [20,21]. The peaks at 1732 cm −1 and 1505 cm −1 of raw cassava pulp corresponded to C=O stretching and C=C aromatic ring; these referred to lignin and hemicellulose [20,21]. Apparently, these peaks disappeared in the pretreated cassava pulp and CNF, indicating partial removal of hemicelluloses and lignin from the structure of fibers leading to higher purity of cellulose. Furthermore, no significant differences were observed among the spectra of BCP and CNF. The results proved that the cellulose molecular structure remained unchanged with the homogenizing method.
From the XRD curve shown in Figure 3b, the crystalline peaks of all samples showed 2θ at 16 • and 22 • . The crystalline peak of the CP had the lowest intensity. From the XRD pattern, the crystallinity index (CI) of all samples could be determined. The Segal method was used to calculate the CI with the principal diffraction peak of cellulose I around 22 • and the lowest intensity at around 18 • . The CP showed the lowest CI of 19%, since it contains a great amount of amorphous area. On the other hand, the BCP and CNF give high crystallinity of 43-46% (Table 1). This occurred due to the deletion of hemicelluloses and lignin bound to the fibers of cellulose.
Polymers 2023, 15, 2092 9 of 23 crystallinity of 43-46% (Table 1). This occurred due to the deletion of hemicelluloses and lignin bound to the fibers of cellulose.

Effect of CNF Concentration on Physicochemical Properties of PEO-PPO-PEO-DA Injectable Hydrogels
PEO-PPO-PEO-DA injectable hydrogels with several CNF concentrations at 0, 0.3, 0.5, and 0.7% (w/v) were successfully prepared after curing with UV light within 10 min. All the injectable hydrogels were uniformly translucent, non-brittle, and strong enough to handle without deformation ( Figure 4). The morphologies of hydrogels with different CNF concentrations were characterized by FE-SEM. The FE-SEM images shown in Figure  5 reveal the porous structures of the hydrogels, which play a fundamental role in cell proliferation. The FE-SEM image of the surface and cross-section of all the hydrogels show highly microporous structures with an interconnecting network, pore size diameter of a surface range of 11-128 μm and a cross-section range of 9-168 μm ( Figure 6 and Table 2). All the hydrogels show a porosity of 65-93% ( Table 2). As a result of augmenting the CNF concentration, porosity in the hydrogels diminishes, and the hydrogel structure becomes tighter, suggesting that the CNF may be dispersed in the polymeric structure. Lately, scaffolds displaying pores from 5 to 350 μm in diameter and porosity of ≥ 55% have been used for tissue regeneration [22]. This suggests that all the hydrogels show an adequate porosity and pore siz e suitable for chondrocyte growth and proliferation.

Effect of CNF Concentration on Physicochemical Properties of PEO-PPO-PEO-DA Injectable Hydrogels
PEO-PPO-PEO-DA injectable hydrogels with several CNF concentrations at 0, 0.3, 0.5, and 0.7% (w/v) were successfully prepared after curing with UV light within 10 min. All the injectable hydrogels were uniformly translucent, non-brittle, and strong enough to handle without deformation (Figure 4). The morphologies of hydrogels with different CNF concentrations were characterized by FE-SEM. The FE-SEM images shown in Figure 5 reveal the porous structures of the hydrogels, which play a fundamental role in cell proliferation. The FE-SEM image of the surface and cross-section of all the hydrogels show highly microporous structures with an interconnecting network, pore size diameter of a surface range of 11-128 µm and a cross-section range of 9-168 µm ( Figure 6 and Table 2). All the hydrogels show a porosity of 65-93% ( Table 2). As a result of augmenting the CNF concentration, porosity in the hydrogels diminishes, and the hydrogel structure becomes tighter, suggesting that the CNF may be dispersed in the polymeric structure. Lately, scaffolds displaying pores from 5 to 350 µm in diameter and porosity of ≥ 55% have been used for tissue regeneration [22]. This suggests that all the hydrogels show an adequate porosity and pore siz e suitable for chondrocyte growth and proliferation.  (Table 1). This occurred due to the deletion of hemicelluloses a lignin bound to the fibers of cellulose.

Effect of CNF Concentration on Physicochemical Properties of PEO-PPO-PEO-DA Injectable Hydrogels
PEO-PPO-PEO-DA injectable hydrogels with several CNF concentrations at 0, 0.5, and 0.7% (w/v) were successfully prepared after curing with UV light within 10 m All the injectable hydrogels were uniformly translucent, non-brittle, and strong enough handle without deformation (Figure 4). The morphologies of hydrogels with differ CNF concentrations were characterized by FE-SEM. The FE-SEM images shown in Fig  5 reveal the porous structures of the hydrogels, which play a fundamental role in cell p liferation. The FE-SEM image of the surface and cross-section of all the hydrogels sh highly microporous structures with an interconnecting network, pore size diameter o surface range of 11-128 μm and a cross-section range of 9-168 μm ( Figure 6 and Table All the hydrogels show a porosity of 65-93% ( Table 2). As a result of augmenting the C concentration, porosity in the hydrogels diminishes, and the hydrogel structure becom tighter, suggesting that the CNF may be dispersed in the polymeric structure. Lately, sc folds displaying pores from 5 to 350 μm in diameter and porosity of ≥ 55% have been u for tissue regeneration [22]. This suggests that all the hydrogels show an adequate por ity and pore siz e suitable for chondrocyte growth and proliferation.       The mechanical properties of injectable hydrogels must be taken into consideration in relation to resistance to damage when scaffolds are applied for tissue engineering. In this study, the CNF-reinforced PEO-PPO-PEO-DA injectable hydrogels were compared with the conventional injectable hydrogel, fibrin glue. Figure 7 shows the poor mechanical properties of the fibrin glue: compression strength of 47 kPa, compression modulus of 12 kPa, ultimate tensile strength of 18 kPa, tensile modulus of 8 kPa, and elongation at break of 130%. Especially, all the CNF-reinforced PEO-PPO-PEO-DA injectable hydrogels show remarkably higher mechanical properties than those of the fibrin glue. Interestingly, the mechanical properties of the injectable hydrogels significantly improved with the increase of CNF from 0 to 0.5% (w/v). However, the mechanical properties of the hydrogels were slightly reduced with the increase of CNF from 0.5 to 0.7% (w/v). All the CNFreinforced PEO-PPO-PEO-DA injectable hydrogels show good mechanical properties: compression strength (429-583 kPa), compression modulus (143-160 kPa), ultimate tensile strength (70-110 kPa), tensile modulus (139-154 kPa) and elongation at break (499-566%) which met the hydrogel scaffold requirements. In addition, the CNF-reinforced PEO-PPO-PEO-DA injectable hydrogels exhibit remarkably better mechanical properties than other hydrogel materials (summarized in Table 3), which confirms the improvements brought by incorporating CNF in the semi-interpenetrating network. Subsequently, the CNF-reinforced PEO-PPO-PEO-DA injectable hydrogels were selected for subsequent experiments. Hydrogel scaffold requirement ≥100 ≥100 ≥50 ≥100 ≥100 [28] Note: / means not mentioned. Water uptake and swelling are fundamental for cell growth on hydrogels aimed at biological applications. Typically, water uptake and swelling of hydrogels for use as tissue engineering should be more than 50% and 100%, respectively [29]. As shown in Table 4, all hydrogels show high water uptake of 86-87%. However, after augmenting CNF concentration, the water uptake capacity of the hydrogels slightly decreased because of additional physical interaction due to hydrogen bonds and the resulting decrease in porosity [29]. Table 4 shows the water swelling of PEO-PPO-PEO-DA and CNF-reinforced PEO-PPO-PEO-DA injectable hydrogels. All of the hydrogels show high water swelling of 618-643% hinting high porosity for the hydrogels. However, the swelling characteristics of PEO-PPO-PEO-DA hydrogel were higher than for other hydrogels. Again, the reason may be found in the CNF creating a tighter, denser network and diminishing porous content [29]. Gel fraction is the degree of crosslinking of hydrogels produced by UV irradiation. Typically, gel fraction of hydrogels for use as tissue engineering should be more than 80% [30]. All the hydrogels show high gel fraction of >80% (Table 4), which meet the requirement, indicating that the hydrogels are cross-linked three-dimensional hydrophilic polymer networks capable of swelling without losing integrity when in an aqueous environment.
As evidenced by all the results, this indicates that 0.5% (w/v) CNF-reinforced polymer matrix gives the highest mechanical properties without deterioration of other properties that meet the physicochemical properties of tissue engineering requirements. Therefore, CNF at 0.5% (w/v) was selected for reinforcing in PEO-PPO-PEO-DA/GelMA injectable hydrogels.

Characterization of CNF-Reinforced PEO-PPO-PEO-DA/GelMA Injectable Hydrogels
The purpose of incorporating GelMA into PEO-PPO-PEO-DA/CNF injectable hydrogel was to promote cell adhesion and proliferation. In this study, the effects of GelMA concentration on physicochemical properties were investigated. In addition, cell cytotoxicity, and cell proliferation were also studied.

Physicochemical Characterization Appearance
PEO-PPO-PEO-DA/CNF injectable hydrogels with various GelMA concentrations at 0, 1, 3, and 5% (w/v) were successfully prepared after curing with UV light within 10 min. All the injectable hydrogels were uniformly translucent, non-brittle, and with enough strength to be handled without showing deformation (Figure 8).  (Figure 9) with a pore size diameter of surface range of 11-128 μm and a cross-section range of 8-140 μm ( Figure 10 and Table 5). All the

Morphology, Pore Size Diameter, and Porosity
The FE-SEM micrographs of the topical surface and cross-section of the PEO-PPO-PEO-DA/CNF and all the PEO-PPO-PEO-DA/CNF/GelMA show highly microporous structures with an interconnecting network ( Figure 9) with a pore size diameter of surface range of 11-128 µm and a cross-section range of 8-140 µm ( Figure 10 and Table 5). All the hydrogels show high porosity of 75-99%. With the increasing GelMA concentration in the hydrogels, porosity increases and the pore size diameter range of the surface and crosssection of the hydrogels show a narrower range (Table 5). Especially, all the hydrogels show suitable pore sizes and porosity that meet the scaffold requirements (pore sizes ranging from 7-350 µm and porosity of ≥55%).

Morphology, Pore Size Diameter, and Porosity
The FE-SEM micrographs of the topical surface and cross-section of the PEO-PPO-PEO-DA/CNF and all the PEO-PPO-PEO-DA/CNF/GelMA show highly microporous structures with an interconnecting network ( Figure 9) with a pore size diameter of surface range of 11-128 μm and a cross-section range of 8-140 μm ( Figure 10 and Table 5). All the hydrogels show high porosity of 75-99%. With the increasing GelMA concentration in the hydrogels, porosity increases and the pore size diameter range of the surface and crosssection of the hydrogels show a narrower range (Table 5). Especially, all the hydrogels show suitable pore sizes and porosity that meet the scaffold requirements (pore sizes ranging from 7-350 μm and porosity of ≥55%).    Gel Fraction, Water Uptake, and Swelling The PEO-PPO-PEO-DA/CNF and PEO-PPO-PEO-DA/CNF/GelMA injectable hydrogels meet the property requirements with high gel fraction (82-83%), water uptake (86-88%), water swelling (622-730%), and porosity (74-98%) ( Table 6). Interestingly, water uptake and swelling of the hydrogels are augmented with GelMA concentration. Table 6. Gel fraction, water uptake, and water swelling of the injectable hydrogels.

Mechanical Properties
The mechanical properties of the injectable hydrogels and fibrin glue are shown in Figure 11 and Table 7. All the injectable hydrogels show remarkably higher mechanical properties than those of the fibrin glue ( Figure 10). Unfortunately, the mechanical properties of the injectable hydrogels decreased with the increasing GelMA from 1 to 5% (w/v) resulting in PPO-PEO-DA/0.5%CNF/5%GelMA showing lower mechanical properties than hydrogel scaffold requirements (Table 7). However, the CNF-reinforced PEO-PPO-PEO-DA with the addition of GelMA from 1 to 3% show good mechanical properties, such as compression strength (417-451 kPa), compression modulus (134-143 kPa), ultimate tensile strength (77-100 kPa), tensile modulus (103-109 kPa), and elongation at break (231-487%), which meet the hydrogel scaffold requirements (Table 7).   Chemical Interaction FTIR spectra of CNF, GelMA, and hydrogels are shown in Figure 12. CNF shows a characteristic absorption band at 3600-3000 cm −1 (OH stretching) and 1606 cm −1 (O-H bending vibrations owing to absorbed water with some contributions from carboxylate groups). PEO-PPO-PEO-DA exhibits a typical absorption band at 3600-3400 cm −1 (OH stretching),

In Vitro Cell Culture Studies
Cell Cytotoxicity The cytotoxicity test aims at assessing the ability of certain chemicals to destroy living cells. Cytotoxicity measurements are important indicators for evaluating toxicological endpoints of biomedical materials because of their ease of performance, high speed, and high sensitivity, and they can reduce animal experimentation.

In Vitro Cell Culture Studies Cell Cytotoxicity
The cytotoxicity test aims at assessing the ability of certain chemicals to destroy living cells. Cytotoxicity measurements are important indicators for evaluating toxicological endpoints of biomedical materials because of their ease of performance, high speed, and high sensitivity, and they can reduce animal experimentation.  On Day 14, cell viability of 12%PEO-PPO-PEO-DA/0.5%CNF and 12%PEO-PPO-PEO-DA/0.5%CNF/1%GelMA hydrogels slightly decreased from 61 to 46 and 53 to 42%, respectively. Although these hydrogels are non-cytotoxic, they gave low cell viability. This suggests that these hydrogels have low cell adhesion and proliferation in the long term. On the other hand, PEO-PPO-PEO-DA/CNF hydrogel incorporating 3% and 5% (w/v) of GelMA, showed high cell viability of more than 80% at every time point implying good cell adhesion and proliferation and non-toxicity. Especially, 12%PEO-PPO-PEO-DA/0.5%CNF/3%GelMA and 12%PEO-PPO-PEO-DA/0.5%CNF/5%GelMA hydrogels showed significantly higher cell viability (p < 0.05) than 12%PEO-PPO-PEO-DA/0.5%CNF and 12%PEO-PPO-PEO-DA/0.5%CNF/1%GelMA hydrogels at every time point. The results confirm our hypothesis that the incorporation of GelMA at optimum concentration into an injectable hydrogel can promote cell adhesion and proliferation.
All of the experiment results verify that CNF-reinforced PEO-PPO-PEO-DA/GelMA injectable hydrogels represent a system alternative to existing therapeutics. This system is important to develop a novel scaffold for reducing the surgical wound area and healing time by minimally invasive surgery instead of open incisions. These hydrogels enhance the physicochemical properties which ensure superior performance compared to conventional hydrogels, fibrin glues, and other developed hydrogel materials (summarized in Table 3). Especially, this injectable hydrogel system significantly promotes cell proliferation and also shows good physicochemical properties which meet meniscus tissue engineering requirements. The results indicate that the mechanical characteristics of the injectable hydrogels and cell viability are significantly improved via incorporating CNF and GelMA, respectively, into the injectable hydrogels which proves our hypothesis correct.  All of the experiment results verify that CNF-reinforced PEO-PPO-PEO-DA/GelMA injectable hydrogels represent a system alternative to existing therapeutics. This system is important to develop a novel scaffold for reducing the surgical wound area and healing time by minimally invasive surgery instead of open incisions. These hydrogels enhance the physicochemical properties which ensure superior performance compared to conventional hydrogels, fibrin glues, and other developed hydrogel materials (summarized in Table 3). Especially, this injectable hydrogel system significantly promotes cell proliferation and also shows good physicochemical properties which meet meniscus tissue engineering requirements. The results indicate that the mechanical characteristics of the injectable hydrogels and cell viability are significantly improved via incorporating CNF and GelMA, respectively, into the injectable hydrogels which proves our hypothesis correct.

Conclusions
We have shown here CNF-PEO-PPO-PEO-DA/GelMA injectable hydrogel after curing with UV light within 10 min with a thickness of 2 mm, CNF significantly improves its mechanical properties. The incorporation of GelMA into the injectable hydrogel has led to promote cell proliferation of human cartilage stem/progenitor cells. At optimum formulation, 12%PEO-PPO-PEO-DA/0.5%CNF/3%GelMA injectable hydrogels significantly   All of the experiment results verify that CNF-reinforced PEO-PPO-PEO-DA/GelMA injectable hydrogels represent a system alternative to existing therapeutics. This system is important to develop a novel scaffold for reducing the surgical wound area and healing time by minimally invasive surgery instead of open incisions. These hydrogels enhance the physicochemical properties which ensure superior performance compared to conventional hydrogels, fibrin glues, and other developed hydrogel materials (summarized in Table 3). Especially, this injectable hydrogel system significantly promotes cell proliferation and also shows good physicochemical properties which meet meniscus tissue engineering requirements. The results indicate that the mechanical characteristics of the injectable hydrogels and cell viability are significantly improved via incorporating CNF and GelMA, respectively, into the injectable hydrogels which proves our hypothesis correct.

Conclusions
We have shown here CNF-PEO-PPO-PEO-DA/GelMA injectable hydrogel after curing with UV light within 10 min with a thickness of 2 mm, CNF significantly improves its mechanical properties. The incorporation of GelMA into the injectable hydrogel has led to promote cell proliferation of human cartilage stem/progenitor cells. At optimum formulation, 12%PEO-PPO-PEO-DA/0.5%CNF/3%GelMA injectable hydrogels significantly

Conclusions
We have shown here CNF-PEO-PPO-PEO-DA/GelMA injectable hydrogel after curing with UV light within 10 min with a thickness of 2 mm, CNF significantly improves its mechanical properties. The incorporation of GelMA into the injectable hydrogel has led to promote cell proliferation of human cartilage stem/progenitor cells. At optimum formulation, 12%PEO-PPO-PEO-DA/0.5%CNF/3%GelMA injectable hydrogels significantly promote cell proliferation (cell viability > 80%) and also show good physicochemical properties which meet tissue engineering requirements. To sum up, our data show that the novel injectable hydrogel is a good candidate for applications in meniscus tissue engineering.