Characterization of osteogenic cells grown over modified graphene- oxide-biostable polymers (original) (raw)

Biomedical Materials

PAPER

Characterization of osteogenic cells grown over modified graphene-oxide-biostable polymers

Eraj Humayun Mirza 1,2{ }^{1,2} (D), Aftab Ahmed Khan 3{ }^{3} (D), Abdulaziz Abdullah Al-Khureif 2{ }^{2}, Selma Adnan Saadaldin 2{ }^{2}, Badreldin Abdelrhaman Mohamed 2{ }^{2}, Fatima Fareedi 2{ }^{2}, Muhammad Muzammil Khan 3{ }^{3}, Musaad Alfayez 4{ }^{4}, Randa Al-Fotawi 7{ }^{7}, Pekka K Vallittu 8{ }^{8} and Amer Mahmood 9{ }^{9}
1{ }^{1} Department of Biomedical Engineering, NED University of Engineering and Technology, L.E.J. Campus, 75270, Karachi, Pakistan
2{ }^{2} Dental Biomaterials Research Chair, College of Applied Medical Sciences, King Saud University, 11451, Riyadh, Saudi Arabia
3{ }^{3} Department of Clinical Dental Science, College of Dentistry, Princess Nourah Bint Abdulrahman University, 84428, Riyadh, Saudi Arabia
4{ }^{4} Department of Community Health, College of Applied Medical Sciences, King Saud University, 11451, Riyadh, Saudi Arabia
5{ }^{5} Department of Biomedical Engineering, Sir Syed University of Engineering and Technology, 75300, Karachi, Pakistan
6{ }^{6} Stem Cell Unit, Department of Anatomy, College of Medicine, King Saud University, 11451, Riyadh, Saudi Arabia
7{ }^{7} Department of Oral Surgery, College of Dentistry, King Saud University, 11451, Riyadh, Saudi Arabia
8{ }^{8} Department of Biomaterials Science and Turku Clinical Biomaterials Centre, Institute of Dentistry, University of Turku and City of Turku, Welfare Division, 20520, Turku, Finland
9{ }^{9} Author to whom any correspondence should be addressed.

E-mail: erajhumayun@hotmail.com, dreraj@neduet.edu.pk, afikhan@ksu.edu.sa, aalkhurail@ksu.edu.sa, sasaadaldin@pnu.edu.sa, bmohamed@ksu.edu.sa, fatima.fareedi1@gmail.com, mmuzamil@ssuet.edu.pk, alfaye@ksu.edu.sa, ralfotawei@ksu.edu.sa, pekval@ utu.fi and ammahmood@ksu.edu.sa

Keywords: bone tissue, gene expression, graphene oxide, osteoblasts, poly(methyl methacrylate) (PMMA)

Abstract

Graphene is an excellent filler for the development of reinforced composites. This study evaluated bone cement composites of graphene oxide (GO) and poly(methyl methacrylate) (PMMA) based on the proliferation of human bone marrow mesenchymal stem cells (hBMSCs), and the anabolic and catabolic effects of the incorporation of GO on osteoblast cells at a genetic level. Surface wettability and roughness were also evaluated at different GO concentrations (GO1: 0.024 wt %\% and GO2: 0.048 wt%\mathrm{wt} \% ) in the polymer matrix. Fabricated specimens were tested to (a) observe cell proliferation and (b) identify the effectiveness of GO on the expression of bone morphogenic proteins. Early osteogenesis was observed based on the activity of alkaline phosphatase and the genetic expression of the runrelated transcription factor 2. Moreover, bone strengthening was determined by examining the collagen type 1 alpha-1 gene. The surface roughness of the substrate material increased following the addition of GO fillers to the resin matrix. It was found that over a period of ten days, the proliferation of hBMSCs on GO2 was significantly higher compared to the control and GO1. Additionally, quantitative colorimetric mineralization of the extracellular matrix revealed greater calcium phosphate deposition by osteoblasts in GO2. Furthermore, alizarin red staining analysis at day 14 identified the presence of mineralization in the form of dark pigmentation in the central region of GO2. The modified GO-PMMA composite seems to be promising as a bone cement type for the enhancement of the biological activity of bone tissue.

Introduction

Graphene contains carbon atoms arranged in the form of a honeycomb lattice [1]. Two-dimensional single atom thick graphene has distinct mechanical, electrical, thermal and optical properties, and has attracted widespread interest within the scientific community. Owing
to its commendable properties, graphene is among the most prominent contenders for the development of reinforced composites, and since its discovery in 2004 there has been ongoing interest in using graphene in medical and biomaterial applications [2-4].

Likewise, the derivative of graphene, graphene oxide (GO), has a similar structure and can be

chemically oxidized via the silanization method. As a result of oxidization, GO offers various functional groups on its basal plane such as those of carboxyl, hydroxyl and epoxy. Moreover, the biocompatibility and role of GO in enhanced cell proliferation for stem cells has paved the way for it use in the field of tissue engineering [5,6][5,6]. However, to utilize the potential of GO, it is critical to disperse it in a well-established matrix. Numerous polymers have been tested for tissue engineering applications with GO, as they easily disperse owing to the copious oxygen groups they contain [7]. The most common biostable polymer is poly (methyl methacrylate) (PMMA), which is extensively used as bone cement [8-11].

The advantages of PMMA include its low cost and ease of fabrication in conjunction with the use of poly-mer-powder-monomer liquid systems, and the possibility of exposing the fillers of the composite implant by dissolving the thermoplastic PMMA matrix [12-16]. In addition to these advantages, there are some disadvantages associated with PMMA, such as its insufficient hardness and strength, its brittleness and its fibrous capsule formation [17]. In a recent study, our lab has demonstrated incremental improvements in mechanical and physical properties when GO is incorporated with PMMA [18]. The overall hardness, surface roughness, elastic modulus and compressive strength were improved due to GO incorporation [18]. Fracture, delamination and the release of bone cement debris are the most frequently encountered problems associated with PMMA [19]; however, in this case, the expected biological effects of using reinforcement agents in the material do not take place because of the embedded reinforcement agents under the PMMA skin.

Porogen fillers and bioactive glasses (owing to the presence of hydroxyl groups) have recently been suggested for use with discontinuous, reinforcing fillers to enhance bone growth on the surface of PMMA and in PMMA bone cement [14, 20, 21]. The glass releases bioactive phosphate and calcium ions which enhance stem cell differentiation to osteogenic cell lines; however, these modified bone cements have not been clinically approved yet, and traditional powder-liquid PMMA bone cement is still in use [22].

Recently, our group has also published results pertaining to the incorporation of carbon nanotube fillers in PMMA-based bone implants that show considerable influence on biological properties [13]. Given that GO belongs to the carbon family, we envisage that the influence of GO would be the same and even be evident with low-filler loading in PMMA.

Therefore, this study aims to deepen our understanding of the roles of GO fillers in the polymer matrix in terms of surface properties, biological compatibility and cell proliferation. Additional tests were considered necessary to evaluate the effectiveness of GO on the expression of bone morphogenic proteins (BMPs). For this purpose, the activity of alkaline
phosphatase (ALP) and run-related transcription factor 2 (RUNX2), as well as the expressions of collagen type 1 alpha-1 (Col1A1) genes, were studied to determine the possible osteogenic effects.

Materials and methods

GO nanosheet particle preparation

To prepare the GO nanosheets, graphite flakes were purchased and ultrasonicated in a mixture of ethanol and water ( ∼40%−60%\sim 40 \%-60 \% ). The complete procedure is described by Chia et al [23]. Flakes of graphene were dispersed using deionized water ( 900 ml ), which was subsequently ultrasonicated. The resulting product was oxidized (with 10%HCl10 \% \mathrm{HCl} ), filtered and dehydrated at 85∘C85^{\circ} \mathrm{C} overnight. The dimensions of the resultant GO sheets ranged from 500 nm to 2μ m2 \mu \mathrm{~m} [24].

Fabrication of GO composite

GO-PMMA composites were fabricated with the use of autopolymerizing (via a peroxide-amine initiator system) PMMA powder and a monomer of methyl methacrylate (MMA) (Eco Cryl Cold, Protechno, Vilamalla Girona, Spain). Experimentally, the GO1 and GO2 groups respectively contained 0.025 g and 0.05 g of GO sheets, which were embedded in the MMA monomers and then sonicated. The GO1 group was a composite of GO ( 0.024wt/wt%0.024 \mathrm{wt} / \mathrm{wt} \% ), and the GO2 group was also a GO composite ( 0.048wt/wt−%0.048 \mathrm{wt} / \mathrm{wt}-\% ) which had been fabricated in a polymer matrix. The neat polymer (i.e., the polymer without GO) was considered as the control group © throughout the study. The powder/monomer ratio was 1.3:1.0. Three blocks from each group were fabricated in a silicon mold (dimensions of 62×62×3.5 mm362 \times 62 \times 3.5 \mathrm{~mm}^{3} ), and the mixing of PMMA powder and MMA liquid was performed manually until the mixture solution acquired a paste-like viscosity. An aluminum foil was used to cover the specimens for 20 min . A load of 100 N was then administered after which the remaining polymerization occurred at room temperature and pressure. After curing, composite cements were removed and stored in a desiccator.

A total of 30 specimens (dimensions of 14×14×2.5 mm314 \times 14 \times 2.5 \mathrm{~mm}^{3} ) were cut ( n=10n=10 for each of the three groups) by sectioning larger pieces from each group for different testing parameters. The cutting surface exposed the GO sheets and the surface was used for additional experiments.

Raman spectroscopy

The raman spectra of pure GO were obtained by exciting a He-Ne green laser source using a Reimshaw inVia Raman microscope.

Surface topography analysis

The average surface roughness (Sa)\left(S_{\mathrm{a}}\right) was determined with a three-dimensional (3D) optical noncontact

surface profiler (Bruker, Campbell, CA, USA). The mean SaS_{\mathrm{a}} value of the five specimens was measured in the central (4.0 mm×6.0 mm)(4.0 \mathrm{~mm} \times 6.0 \mathrm{~mm}) region of the specimens.

Contact angle analysis

Dynamic water contact angle analyses of the GOPMMA and control materials were performed using a digital tensiometer (Theta Lite, Dyne Technology, Staffordshire, UK). On each specimen, a 4.0μl4.0 \mu \mathrm{l} droplet of distilled H2O\mathrm{H}_{2} \mathrm{O} was dispensed and imaged after 30 s . The contact angle was automatically calculated with commercially available software.

Cell culture methods

The human telomerase reverse transcriptase gene in hBMSC was overexpressed to generate clones of human bone marrow stromal cells (TERT-hBMSCCL1). A subclone was derived from TERT-hBMSCs (referred to as CL1) owing to its excellent osteogenic, adipogenic and chondrogenic differential potential. The hBMSC-CL1 cells were grown as described earlier [25].

Cell viability assay with Alamar blue

To evaluate viability, CL1 cells were seeded on specimens. A cellular viability assay was performed using the alamarBlue ®{ }^{\circledR} cell viability assay according to the manufacturer’s directives (AbD Serotec, Raleigh, NC, USA). Briefly, 10μl10 \mu \mathrm{l} of alamarBlue ®{ }^{\circledR} substrate was added to a 96 - well plate and the cells were directly grown and nurtured for 1 h at a temperature of 37∘C37^{\circ} \mathrm{C} in a dark room. Fluorescence at an excitation wavelength of 530 nm and at an emission wavelength of 590 nm was measured with a microplate reader (BioTek Inc., Winooski, VT, USA).

In vitro osteoblast differentiation

For growing and differentiating hBMSC-CL1 cells, DMEM was used as a growth media in six-well plates. Once a cell confluency of 70%−80%70 \%-80 \% was achieved, the test cells were supplemented with an osteoblast (OB) induction mixture, which contained 50μ gml−150 \mu \mathrm{~g} \mathrm{ml}^{-1} L-ascorbic acid (Wako Chemicals, Neuss, Germany), 1%1 \% penicillin streptomycin solution, 10mMβ10 \mathrm{mM} \beta glycerophosphate (Sigma-Aldrich, St. Louis, MO), 10 nM dexamethasone (Sigma-Aldrich) and 10 nM calcitriol ( 1α1 \alpha, 25-dihydroxyvitamin D3, SigmaAldrich) supplemented with 10%10 \% fetal bovine serum (FBS). The OB media was changed every 2 to 3 days. Similar cells were grown in standard growth media as negative controls.

Cytochemical staining

Staining of a mineralized matrix with alizarin red SS Phosphate buffer saline (PBS) was used to wash and detach the cell layer from the scaffold. The cell layer was then fixed using 4%4 \% paraformaldehyde (PFA) at
room temperature for 15 min , followed by the aspiration of PFA. The cells were then rinsed in distilled water three times and stained with a commercially available 2%2 \% alizarin red S (ARS) staining kit (cat. no. 0223, ScienCell, Research Laboratories, Carlsbad, CA, USA) for 20−30 min20-30 \mathrm{~min}. The samples were then washed three to five times to remove the excess dye, and they were stored in water to prevent them from drying out. An inverted microscope (Zeiss, Thornwood, NY, USA) was used to acquire the images.

OsteoImage mineralization assay

An assay kit (cat. no. PA-1503, Lonza, USA) for osteoImage mineralization was used to quantify the mineralized matrix. Differentiated cells were washed using PBS followed by fixation for 20 min using icecold ethanol ( 70%70 \% ). From this stage onward, all the procedures were executed in a dark room. Firstly, the diluted staining reagent was added to the plate containing the cells for half an hour at room temperature. Subsequently, the cells were washed and the fluorescence was read at an excitation wavelength of 492 nm , while the emission wavelength was set at 520 nm for imaging and stain quantitation.

Stromal cell attachment and morphological assessment

The specimens were seeded with CL1 stromal cells and their attachment was observed after three days using scanning electron microscopy (SEM) (JSM-6360 LV, JEOL, Tokyo, Japan). To achieve this, hBMC-CL1 cells were grown onto two different scaffolds at day 0 . To achieve improved cell attachment, the scaffolds were coated with fetal calf serum for 12 h . Before the addition of cells, the specimens were washed once with growth media, followed by the seeding of 1×1051 \times 10^{5} cells on each specimen. On the final day, the samples were prepared for SEM imaging by washing them with PBS followed by fixation at 4∘C−6∘C4^{\circ} \mathrm{C}-6^{\circ} \mathrm{C} with 1%1 \% glutaraldehyde (Sigma-Aldrich, MO, USA) buffered in 0.1 M sodium cacodylate (Agar Scientific, Essex, UK). After glutaraldehyde fixation, osmium tetroxide (1%) (Agar Scientific, Essex, UK) was used to fix the cells according to the manufacturer’s guidelines. Every specimen was dried, and gold was coated using sputtering before examination with a Carl Zeiss Sigma VP Oxford Micro-analysis S800 system, as reported earlier [26]. Images were acquired at different magnifications using SEM for each specimen.

Real-time qRT-PCR

Total ribonucleic acid (RNA) was extracted using a PureLink RNA mini isolation kit (cat no. 12183018A, Ambion ®{ }^{\circledR}, Thermo Fisher Scientific, MA, USA) according to the manufacturer’s directives. A nanodrop spectrophotometer (Nanodrop 2000, Thermo Fisher Scientific, MA, USA) was used to quantify the total RNA. One μg\mu \mathrm{g} of the RNA specimen was used to

Figure 1. Raman Spectroscopy of pure GO.
synthesize complementary deoxyribonucleic acid (cDNA) using a high-capacity cDNA reverse transcription kit (Applied Biosystems, Thermo Fisher Scientific, MA, USA) and a Labnet, Multigene Thermocycler (Labnet International, NJ, USA) according to the manufacturer’s instructions. The cDNA was used to determine the relative levels of messenger RNA (mRNA) via a real-time polymerase chain reaction (qRT-PCR) detection system (Applied Biosystem, Thermo Fisher Scientific, MA, USA) with a Power SYBR Green PCR kit (Applied Biosystems, Thermo Fisher Scientific, MA, USA), or with a TaqMan universal Master Mix II, AmpErase Uracil N-Glycosylase (UNG) (Applied Biosystems, Thermo Fisher Scientific, CA, USA), based on the manufacturer’s directives. After normalization with a reference gene (GAPDH), a comparative Ct method was used for the quantification of the gene expression, where ΔCT\Delta \mathrm{CT} is the difference between the CT values of the target and reference gene. All primers used in this study were acquired from Applied Biosystems.

Statistical analyses

A minimum of three samples were tested for characterization. Statistical analyses were performed with an SPSS (V21.0, IBM, New York, USA). One-way analysis of variance (ANOVA) was applied with Tukey’s post hoc analysis. Neat PMMA was considered as the control throughout the study. PP-values less than 0.05 were considered significant.

Results

Raman spectroscopy

Raman spectroscopy was performed to characterize the pure GO. Characteristic peaks were obtained at around 1600 cm−11600 \mathrm{~cm}^{-1} (G-band) and at around 1360 cm−11360 \mathrm{~cm}^{-1} (D-band) (figure 1).

Surface analysis

The roughness of the surface at the microscopic scale was smallest for the specimens of the control group. At GO loading of 0.024wt/wt%0.024 \mathrm{wt} / \mathrm{wt} \% (group GO1) in the PMMA, the roughness was significantly higher
( p=0.012p=0.012 ) in comparison with the control group. A similar trend ( p=0.002p=0.002 ) was observed when a GO load of 0.048wt/wt%0.048 \mathrm{wt} / \mathrm{wt} \% was incorporated into the PMMA (GO2 group). However, insignificant differences in surface roughness values were noticed between the groups GO1 (0.266μ m)(0.266 \mu \mathrm{~m}) and GO2 (0.288(0.288 μm)\mu \mathrm{m}) (figure 2).

The incorporation of GO into PMMA resulted in pronounced hydrophobicity. The specimens from group C yielded a water contact angle of 63.24∘±1.38∘63.24^{\circ} \pm 1.38^{\circ} (figure 3(a)), whereas the specimens from the GO1 and GO2 groups yielded higher water contact angles, which were equal to 68.06∘±1.63∘68.06^{\circ} \pm 1.63^{\circ} and 73.12∘±1.74∘73.12^{\circ} \pm 1.74^{\circ}, respectively. The values for the specimens from GO1 and GO2 were statistically higher compared to that of C. Additionally, a notable statistical difference ( p=0.001p=0.001 ) was observed between GO1 and GO2 (figure 3).

Cell viability

The proliferation of hBMSCs was tested (C, GO1, and GO2 groups) over a 10 day period. It was observed that throughout the time periods, the viability of the cells seeded on the GO2 specimens remained significantly higher compared to C and GO1. Cell viability increased for all the specimens at day 10. A significant difference between C and GO1 was only observed at day 10 (figure 4).

The hBMSCs demonstrated a preference for GO2 (figure 5©), as observed from the SEM images (figure 5). A dense extracellular matrix (ECM) formation was observed for GO2 specimens when seeded with hBMSCs. Minimum matrix deposition was noted on the C specimens (figure 5(a)). Spherical cell morphology was observed on the GO1 and GO2 specimens only (figures 5(b) and ©). By contrast, disrupted cell morphology was observed in the C specimens (figure 5(a)).

Matrix mineralization

Quantitative colorimetric mineralization of the ECM was performed. A highly significant degree of mineralization was recorded for the GO1 and GO2 specimens compared to C. However, GO2 demonstrated an

Figure 2. Noncontact surface profilometer images for specimens from groups (a) C, (b) GO1 and © GO2. (d) A bar chart showing the mean surface roughness values (Sa)\left(S_{a}\right) of the tested groups (error bars denote standard deviations).

Figure 3. Water drop images during contact angle measurements for specimens from groups (a) C, (b) GO1 and © GO2. (d) A bar chart showing the average contact angles of the tested groups (error bars denote standard deviations).
increased and significant degree of mineralization in contrast with GO1 (figure 6).

Further analysis of mineralization was performed with alizarin red staining (figures 7(a)-(f)) at day 14 and was compared with the control of the respective specimens. Qualitatively, specimen GO2 exhibited excellent mineralization as justified by the dark pigmentation observed in its central region. C and GO1 demonstrated no differences based on qualitative visual inspections.

Moreover, confocal microscopy shows a detrimental effect on the C specimens when compared to

GO1 and GO2. Both the GO1 and GO2 exhibited greater cell viability in comparison to C (figures 7(g)-(l)).

Expression of osteogenic marker genes

The effects of the incorporation of GO in the PMMA matrix were observed for anabolic (COL1A1, BMP4, BMP2, RUNX2 and ALP) and catabolic (MMP9 and MMP2) genes when seeded with the hBMSCs (figure 8). It was observed that the incorporation of GO had a stimulatory effects on all the anabolic genes.

Figure 4. Cell viability of hBMSCs tested over 10 days for groups C, GO1 and GO2, using the Alamar blue dye.

Figure 5. Scanning electron microscopy images of (a) C, (b) GO1 and © GO2 after 14 days of in vitro culture at a magnification of 200×(200 \times( scale bar =100μ m)=100 \mu \mathrm{~m}). The respective insets are at a magnification of 1000×(1000 \times( scale bar =10μ m)=10 \mu \mathrm{~m}).

By contrast, an inhibitory effect was observed for the catabolic genes owing to the incorporation of GO. Significantly pronounced effects of GO incorporation were observed for COL1A1 and BMP4 with respect to the percentage of GO rooted in the PMMA matrix (figures 8(a) and 7(b)). The genes BMP2 and RUNX2 were expressed in GO2 specimens only, and no expression was detected in the C and GO1 specimens (figures 8© and (d)).

The expressions for MMP9 and MMP2 demonstrated similar trends toward the incorporation of GO. Higher GO content ( 0.5wt%0.5 \mathrm{wt} \% ) significantly reduced the catabolic gene expression. The expression of ALP was significantly higher for the GO2 specimen in comparison to the control; however, insignificant differences were noticed when GO1 and GO2 were equated (figure 8(g)).

Discussion

The current study appraised the biological compatibility and cell proliferation when GO was incorporated in PMMA at different compositions. Raman spectroscopy confirms that oxidized graphene is used rather than pure graphene or graphite. Similar results were presented previously by other research groups depicting the characteristic peaks of the Raman shift at about 1600 cm−11600 \mathrm{~cm}^{-1} (G-band) and at around 1360 cm−11360 \mathrm{~cm}^{-1} (D-band) [27−29][27-29].

Surface parameters, such as S4S_{4}, were used to evaluate the surface roughness over a specific area. The statistically significant surface roughness values in the GO1 and GO2 specimens suggested that the incorporation of GO nanosheets enhanced the roughness of the surface in the specimens (figure 2(d)). Surface roughness measurements further complemented the water contact angle outcomes. Neat PMMA was characterized in part by a water contact angle of 63.24∘±1.38∘63.24^{\circ} \pm 1.38^{\circ} (figure 3(a)) in accordance with previous studies [30, 31]. However, the statistically significant values of the water contact angles in the cases of the GO1 and GO2 specimens may be attributed to the hydrophobic nature of GO and its protrusion from the surface during the GO-PMMA composite fabrication (figure 3(d)) [31, 32]. Higher surface roughness is attributed to the improved anchorage and increased cell attachment [33]. The improved surface attributes can lead to improved osseo-integration results. Accordingly, the synthesis of a rough surface is of paramount importance for better osseo-integration. In fact, this is a commonly used method for the improvement of implant attachment and osteogenic cell behaviors on rough implant surfaces [34-37].

In composite systems, the matrix phase embeds and covers the fillers. This is similar to the case of bone cements where the surface of the material is in contact with the bone, which is covered with a resin matrix, regardless of the filler type used in the material.

Figure 6. Mineralization caused by graphene incorporation on C,GO1\mathrm{C}, \mathrm{GO} 1 and GO 2 specimens when seeded with hBMSCs, as determined by relative absorbance at day 14 .

Figure 7. (a)-(l) hBMSCs seeded on C, GO1 and GO2 specimens. (a)-(f) Osteoblast differentiation of hBMSCs measured with the use of alizarin red staining for the control specimen and at day 14. (g)-(l) Live/dead staining of differentiated hBMSCs for the control specimen and at day 14 (scale bar =250 mm=250 \mathrm{~mm} ).

However, in applications where fillers are intended to influence the surface characteristics, it is imperative to have the fillers exposed. The current study explains that the polymer matrix was cured using the free radical polymerization of MMA, and the substrate surface was cut with a diamond saw. In this respect, the process may have exposed the GO fillers. Incorporation of GO into the PMMA system enhanced the micro roughness of the surface, yet lowered its wettability, as justified by contact angle analyses. Conflicting evidence was reported by our group earlier, based on which some fillers may remain protruding from the matrix surface and cause increased hydrophobicity [13]. However, regarding the substrate parameters of
the present study, it was shown that there was a trend based on which the increased surface roughness had a positive impact on the osteogenic cell line behavior.

To unravel further information regarding cell behavior and ascertain surface changes owing to GO incorporation, various biological tests were also performed. Wu et al hypothesized that the chemical groups in GO, such as the OH−\mathrm{OH}- and COO- groups, are probably among the prime factors needed to precisely influence stem cell differentiation by triggering the Wnt-related signaling route [38]. In the current research, the specimens modified with GO showed improved osteogenic cell differentiation in bone formation, which demonstrates the constructive role of

Figure 8. Gene expressions represented as fold changes of (a) COL1A1, (b) BMP4, © BMP2, (d) RUNX2, (e) MMP9, (f) MMP2 and (g) ALP for hBMSCs seeded on C, GO1 and GO2 at day 14 compared to GAPDH, which was used as a housekeeping gene.

GO to enhance biological compatibility and cell proliferation, as justified by the Alamar blue and gene expression outcomes.

The results of the current study demonstrated spherical cell morphologies when they were seeded on the GO1 and GO2 specimens. By contrast, disrupted cell morphologies were observed in the control specimens. Moreover, higher osteoblast adhesion was observed in graphene-coated substrates compared with the substrate alone [39], and the current study demonstrated the same trend. As the concentration of GO increased, the effect of mineralization became more prominent. Similarly, Li et al reported better cell viability and cellular morphology. However, con-centration-dependent results in the same study revealed a detrimental effect on GO incorporation of greater than 2%2 \% [40]. The results of the current study demonstrated identical morphological changes with confocal microscopy, based on which detrimental morphological effects were observed in the control group specimens compared to GO1 and GO2.

Furthermore, the qualitative calcium staining pattern was substantiated quantitatively based on the dilution of ARS derived from the stained monolayer of the cells [41]. In the current study, an excellent mineralization pattern was observed based on the dark pigmentation observed in the central region of the GO2 specimen compared to the control group and GO1,
which demonstrated no difference. The indication of osteogenesis, reserved as a marker of bone regeneration, is the deposition of calcium phosphate in the ECM. In bone engineering, the bioactivity of the scaffold has been extensively assessed by osteoblasts cultured in the scaffold based on an in vitro mineralization study [42]. Moreover, the present study elaborates on the effects of anabolic and catabolic genes by the incorporation of GO. In a separate study, Depan et al showed that biomineralization, proliferation and the spreading of osteoblast cells was found to increase on HAP-CS-GO substrates, whereas osteoblasts were slow to spread on pure GO [43]. Our results also revealed that PMMA-GO could have promoted the attachment and spreading of hBMSCs and may have resulted in the formation of a dense ECM matrix compared to decreased matrix deposition, as observed on the controls [43].

Furthermore, Duan et al reported that type 1 collagen can be observed around the implants after two weeks of implantation [25]. As a result, type 1 collagen was progressively synthesized in the implants for prolonged implantation times. Additionally, graphene was shown to be capable of promoting the synthesis of type 1 collagen [25]. However, in the current study, a higher concentration of GO exhibited significantly greater COL1A1 gene expression. Furthermore, Kim et al concluded that gene markers of osteogenic

potential, like ALP, RUNX2, and BMP2, were highly expressed when the media were conditioned for chondrocyte priming and GO substrate culturing. Similarly, Lee et al also reported enhanced osteogenic responses in association with calcium deposition when cells were seeded on GO-coated slides [26]. Likewise, the current study suggests an evident increase in calcium deposition owing to the expression and upregulation of the BMP2, BMP4, and RUNX2 genes when these were incorporated in the GO. Moreover, ALP is the gene of a secreted enzyme, which may lead to instantaneously increased levels within the growth media. However, the existence of a correlation between the activity and proliferation of the cells remains unknown. Jong Ho Lee et al reported that during a period spanning 7-14 days, the proliferation of osteoblasts is greatly enhanced and later begins to agglomerate in the ECM. Eventually, early markers of osteodifferentiation are expressed together with ALP after 7 days [41]. In the current study, a similar trend was observed. The ALP is considered as an early marker for osteodifferentiation, while its activity reduces with the maturation of the matrix. By contrast, calcium can be considered as a late-stage marker of osteodifferentiation, and its activity increases with matrix maturation [41]. In a separate study, Zhang et al used the PCR to identify the changes in expression of bone marker genes and showed that the activities of ALP and COL1 increased effectively owing to the presence of GO [44]. Our findings support those reported by Zhang et al, as ALP and COLA1 significantly increased over a period of 14 days compared to the housekeeping gene (GAPDH). Furthermore, in the present study, degradation was observed in association with the inhibitory effects of the MMP2 and MMP9 genes. The regulation of the mechanism, pertaining to (among others) the degradation of collagen and the extracellular matrix, is linked to the expression of MMP2. Moreover, bone resorption involves the main functions of MMPs and tissue inhibitors of MMP (TIMPs) within the bone. Additionally, graphene possesses the properties of promoting bone cell and ECM formation, whereas MMP results in the digestion of ECM components [45].

Conclusions

This study presents new insights into the applicability/ suitability of GO, and successfully outlines the effects of GO-PMMA when these are seeded with hBMSCs. The presented results reveal improved cell viability and osteogenic differentiation hBMSC responses. Throughout osteogenic differentiation, the cells exhibited mineralization phenotypes and were biocompatible with the tested composites.

It is concluded that all the anabolic genes (COL1A1, BMP4, BMP2, RUNX2, and ALP) are associated with stimulatory effects, while the catabolic genes (MMP2 and MMP9) exhibit inhibitory effects.

The available data might favor a new promising GO-PMMA-modified bone cementing material.

Acknowledgments

The authors are grateful to the Deanship of Scientific Research, King Saud University, for the funding received through the Vice Deanship of Scientific Research Chairs.

Author contributions

Conceptualization, E H M, P K V and A A K; methodology, B A M, R AL F and A M; formal analysis, M M K, E H M, A A K and S A S; resources, R Al FA, AL K and A M; original draft preparation, A A K and E H M; writing, review, and editing, F F, M M K, M A and S A S; supervision, P K V, A M and A AL K; funding acquisition, A AL K.

Declaration of conflicting interests

The authors declare that there are no conflicts of interest.

ORCID iDs

Eraj Humayun Mirza (https://orcid.org/0000-0002-6057-1679
Aftab Ahmed Khan (https://orcid.org/0000-0002-9743-1979

References

[1] Kim H, Abdala A A and Macosko C W 2010 Graphene/ polymer nanocomposites Macromolecules 43 6515-30
[2] Sun X et al 2008 Nano-graphene oxide for cellular imaging and drug delivery Nano Res. 1205-12
[3] Liu J, Cui L and Losic D 2013 Graphene and graphene oxide as new nanocarriers for drug delivery applications Acta Biomater. 99243-57
[4] Shen H et al 2012 Biomedical applications of graphene Theranostics 2283-94
[5] Dinescu S et al 2014 In vitro cytocompatibility evaluation of chitosan/graphene oxide 3D scaffold composites designed for bone tissue engineering Bio-Med. Mater. Eng. 242249-56
[6] Zhang C et al 2016 The surface grafting of graphene oxide with poly (ethylene glycol) as a reinforcement for poly (lactic acid) nanocomposite scaffolds for potential tissue engineering applications J. Mech. Behav. Biomed. Mater. 53 403-13
[7] Konkena B and Vasudevan S 2012 Understanding aqueous dispersibility of graphene oxide and reduced graphene oxide through pK\mathrm{p} K a measurements J Phys. Chem. Lett. 3867-72
[8] Anehosur G V et al 2016 Antifungal effect on denture base materials using visible light activated titanium oxide synthesized by two different techniques J. Adv. Oral Res. 712-9
[9] Mirza E H et al 2017 Physical, mechanical, thermal, and dynamic characterization of carbon nanotubes incorporated poly(methyl methacrylate)-based denture implant J. Compos. Mater. 513931-40
[10] Puska M A et al 2005 Exothermal characteristics and release of residual monomers from fiber-reinforced oligomer-modified acrylic bone cement J. Biomater. Appl. 20 51-64

[11] Alhalawani A M and Towler M R 2013 A review of sternal closure techniques J. Biomater. Appl. 28 483-97
[12] Rakhshan V 2015 Marginal integrity of provisional resin restoration materials: a review of the literature Saudi J. Dent. Res. 633-40
[13] Khan A A et al 2017 Single and multi-walled carbon nanotube fillers in poly(methyl methacrylate)-based implant material J. Biomater. Tissue Eng. 7798-806
[14] Hautamäki M et al 2010 Osteoblast response to polymethyl methacrylate bioactive glass composite J. Mater. Sci., Mater. Med. 21 1685-92
[15] Hautamäki M P et al 2008 Repair of bone segment defects with surface porous fiber-reinforced polymethyl methacrylate (PMMA) composite prosthesis: histomorphometric incorporation model and characterization by SEM Acta orthopaedica 79555-64
[16] Hautamäki M et al 2013 Surface modification of fiber reinforced polymer composites and their attachment to bone simulating material J. Mater. Sci., Mater. Med. 24 1145-52
[17] Yadav N S and Elkawash H 2011 Flexural strength of denture base resin reinforced with aluminum oxide and processed by different processing techniques J. Adv. Oral Res. 233-6
[18] Khan A A et al 2018 Physical, mechanical, chemical and thermal properties of nanoscale graphene oxide-poly methylmethacrylate composites J. Compos. Mater. 52 2803-13
[19] Díaz-Arnold A M et al 2008 Flexural and fatigue strengths of denture base resin J. Prosthet. Dent. 100 47-51
[20] Puska M et al 2006 Glass fibre reinforced porous bone cement implanted in rat tibia or femur: histological and histomorphometric analysis Key Eng. Mater. 309 809-12
[21] Mattila R et al 2009 Bone attachment to glass-fibre-reinforced composite implant with porous surface Acta Biomater. 5 1639−461639-46
[22] Lewis G 1997 Properties of acrylic bone cement: state of the art review J. Biomed. Mater. Res. A 38 155-82
[23] Chia J S Y et al 2014 A novel one step synthesis of graphene via sonochemical-assisted solvent exfoliation approach for electrochemical sensing application Chem. Eng. J. 249 270-8
[24] Khan A A et al 2018 Physical, mechanical, chemical and thermal properties of nanoscale graphene oxide-poly methylmethacrylate composites J. Compos. Mater. 52 2803-13
[25] Duan S et al 2015 Enhanced osteogenic differentiation of mesenchymal stem cells on poly (l-lactide) nanofibrous scaffolds containing carbon nanomaterials J. Biomed. Mater. Res. A 103 1424-35
[26] Kim J et al 2018 Enhanced osteogenic commitment of murine mesenchymal stem cells on graphene oxide substrate Biomater. Res. 221
[27] Li B et al 2016 A simple approach to preparation of graphene/ reduced graphene oxide/polyallylamine electrode and their electrocatalysis for hydrogen peroxide reduction J. Nanosci. Nanotechnol. 16 12805-10
[28] Ikhsan N I et al 2015 Facile synthesis of graphene oxide-silver nanocomposite and its modified electrode for enhanced electrochemical detection of nitrite ions Talanta 144908-14
[29] Zhu Y et al 2010 Graphene and graphene oxide: synthesis, properties, and applications Adv. Mater. 22 3906-24
[30] Hashem M et al 2017 Influence of titanium oxide nanoparticles on the physical and thermomechanical behavior of poly methyl methacrylate (PMMA): a denture base resin Sci. Adv. Mater. 9558-44
[31] Khan A A et al 2017 Single and multi-walled carbon nanotube fillers in poly (methyl methacrylate)-based implant material J. Biomater. Tissue Eng. 7798-806
[32] Ma Y et al 2007 Fabrication of super-hydrophobic film from PMMA with intrinsic water contact angle below 90 Polymer 48 7455-60
[33] Deligianni D D et al 2000 Effect of surface roughness of hydroxyapatite on human bone marrow cell adhesion, proliferation, differentiation and detachment strength Biomaterials 2287-96
[34] Gotfredsen K et al 1992 Histomorphometric and removal torque analysis for TiO2hyphen; blasted titanium implants. An experimental study on dogs Clin. Oral Impl. Res. 377-84
[35] Ballo A M et al 2008 Osteoblast proliferation and maturation on bioactive fiber-reinforced composite surface J. Mater. Sci., Mater. Med. 193169
[36] Schneider G B et al 2004 Differentiation of preosteoblasts is affected by implant surface microtopographies J. Biomed. Mater. Res. A 69 462-8
[37] Schneider G et al 2003 Implant surface roughness affects osteoblast gene expression J. Dent. Res. 82372-6
[38] Wu C et al 2015 Graphene-oxide-modified β\beta-tricalcium phosphate bioceramics stimulate in vitro and in vivo osteogenesis Carbon 93116 - 29
[39] Ding X, Liu H and Fan Y 2015 Graphene-based materials in regenerative medicine Adv. Healthcare Mater. 41451-68
[40] Li M et al 2014 Graphene oxide/hydroxyapatite composite coatings fabricated by electrophoretic nanotechnology for biological applications Carbon 67 185-97
[41] Lee J H et al 2015 Enhanced osteogenesis by reduced graphene oxide/hydroxyapatite nanocomposites Sci. Rep. 518833
[42] Polo-Corrales L, Latorre-Esteves M and Ramirez-Vick J E 2014 Scaffold design for bone regeneration J. Nanosci. Nanotechnol. 1415−561415-56
[43] Depan D, Pesacreta T C and Misra R D K 2014 The synergistic effect of a hybrid graphene oxide-c chitosan system and biomimetic mineralization on osteoblast functions Biomater. Sci. 2264-74
[44] Zhang L et al 2017 Dual-functionalized graphene oxide based siRNA delivery system for implant surface biomodification with enhanced osteogenesis ACS Appl. Mater. Interface 934722-35
[45] Varghese S 2006 Matrix metalloproteinases and their inhibitors in bone: an overview of regulation and functions Front Biosci. 1166