Green conversion of graphene oxide to graphene nanosheets and its biosafety study

Chemical reduction of graphene oxide (GO) to graphene employs the use of toxic and environmentally harmful reducing agents, hindering mass production of graphene which is of tremendous technological importance. In this study we report a green approach to the synthesis of graphene, bio-reduced by crude polysaccharide. The polysaccharide reduces exfoliated GO to graphene at room temperature in an aqueous medium. Transmission electron microscopy image provides clear evidence for the formation of few layer graphene. Characterization of the resulting polysaccharide reduced GO by Raman spectroscopy, Fourier transform infrared spectroscopy and Energy dispersive X-ray analysis confirms reduction of GO to graphene. We also investigated the degree of biosafety of the reduced GO and found it to be safe under 100 μg/ml.


Introduction
Due to the electronic, mechanical, thermal and optical uniqueness, two-dimensional graphene nanosheets (GNS) have significantly transformed areas of nanoscience [1,2]. Even though recent research has produced some unique and versatile 2-D nanostructures [3], GNS still holds great promise for potential applications in nanoelectronics [4], sensors [5,6], nanocomposites [7] and other technological fields. Harnessing these characteristics, which renders GNS its potential, necessitates a large-scale production of the nanosheets. Micromechanical cleavage [1], epitaxial growth [8], solution-based chemical reduction [9] and a few other methods are mainly applied for production of GNS. Usually, chemical reduction of GO was carried out using hydrazine and its derivatives [10,11], but its high toxicity and instability makes the procedure potentially hazardous and asks for great care. GNS has a tendency of ππ stacking, making the bulk synthesis of it a key challenge. This can be overcome by the attachment of other molecules or polymers on to the nanosheets. Recently, scientists have reported synthesis of GNS under much milder conditions using molecules like ascorbic acid [12], reducing sugars a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 Physico-chemical characterization of crude polysaccharide Total carbohydrate, protein and phenol were determined using phenol-sulphuric acid, Bradford and Folin-Ciocalteau methods respectively and quantified using glucose, BSA and gallic acid as respective standards [45]. Total glucan, α-glucan and β-glucan were quantified using mushroom and yeast β-glucan assay kit (Megazyme Int.). Helical structure of the polysaccharide was analyzed by Congo red-polysaccharide reaction according to the method described by Qiu et al [46]. Sugar composition of the crude polysaccharide was partially determined by HPTLC [47] and GC-MS [48]. TFA-hydrolyzed polysaccharide was separated by HPTLC with chloroform, n-butanol, methanol, water and acetic acid (4.5: 12.5: 1.5: 1.5) as solvent system and 0.1% orcinol in 5% H 2 SO 4 as detection reagent. Monosaccharides were derivatized to alditol-acetates and subjected to GC-MS (Agilent technologies, USA, HP5MS 30 mm × 0.25 mm × 0.25 μm + 10 m Duraguard column) under defined conditions (80˚C for 2 min, then 15˚C /min to 200˚C for 2 min, then 4˚C /min to 240˚C for 2 min, then 15˚C /min to 280˚C for 5 min, pressure: 9.4 psi, constant flow rate of 1 ml /min, carrier gas: He). Chromatograms of the samples were analyzed by comparing results with standard carbohydrates as well as using NIST database stored in the system.

Preparation of graphene oxide and reduction of GO to PR-GO nanosheets
The graphite oxide was prepared from graphite powder by Hummers method [49]. The obtained GO was dispersed in deionized water (2 mg/ml) and exfoliated in an ultrasonicator for two successive 30 min periods. This exfoliated GO was concomitantly added to an aqueous solution (20 mg/ml) of crude water-soluble polysaccharide and stirred for 48 hours under room temperature. The solution was monitored at regular intervals. Both positive (crude polysaccharide solution of Pleurotus flabellatus) and negative control (Graphene oxide solution) were maintained under similar conditions. Then the solution was centrifuged and the pellet was used for further characterization.

Characterization of GO and PR-GO
Particle size of the PR-GO nanosheets was measured by laser diffractometry using a nano size particle analyzer (Zen 1600 Malvern USA) in the range between 0.6 nm and 6.0 μm. Transmission Electron Microscopy was performed on a Tecnai G 2 spirit Biotwin (FP 5018/40), operating at an accelerating voltage of 80 kV. XRD measurements of PR-GO was taken with a PW 3040/60 PANalytical X-ray diffractometer that operated at a voltage of 45 KV and current of 30 mA with Cu Kα radiation (λ 1.54443 Ǻ). The diffracted intensities were recorded from 35t o 90˚2θ angles. EDX analysis was carried out with a Hitachi S 3400N, Japan, to identify the elemental compositions of the particles. The polysaccharide (PS), GO and PR-GO, were all analysed by Fourier transform infrared (FTIR) spectroscopy (Shimadzu 8400S fourier transform infrared spectrophotometer). The system worked in a diffuse reflectance mode at a resolution of 4 cm -1 in KBr pellets. The scanning data were obtained from the average of 50 scans in the range of 4000 to 400 cm -1 . Raman spectra was monitored using 1.96 eV (633 nm) line of a He Ne laser in HORIBA-JOBIN-YVON Lab RAM HR 800 instrument by placing the sample solution into a semi-micro stopper cuvette with an exposure time of 1 s.

Biosafety evaluation of PR-GO nanosheets in human lymphocyte cells
Lymphocyte cells were isolated from fresh blood according to the method of Boyum [50], using Histopaque. The cells were washed with PBS and resuspended in RPMI-1640 media at a concentration of 10 6 cells/ml. The cells were incubated in RPMI-1640 media containing different concentrations of PR-GO nanosheets (0, 50, 100 and 250 μg/ml), for 3 h at 37˚C. Following treatment, the lymphocytes were processed for cytotoxicity and genotoxicity assays. The lymphocytes were also processed for FACS analysis for oxidative stress.
Evaluation of cytotoxicity of PR-GO nanosheets by MTT assay. The MTT assay allowed the quantitative determination of cell viability. The assay is based on the capability of viable cells to convert MTT to spectroscopically detectible formazan. The assay was performed according to a method previously published [39]. after treatment and incubation, cells were treated with 0.5 mg/ml solution of 3-(4, 5-dimethylthiazolyl-2)-2,5-diphenyltetrazolium bromide (MTT; 100 μl/well) at 37˚C for 3 h. Optical density (OD) was read on iMark™ Microplate Absorbance Reader (BIO-RAD, USA) at 570 nm, with 630 nm as a reference wavelength. The interference of nanosheets in the assays was eliminated by maintaining sample blanks for all the concentrations tested. The OD values obtained from the sample blanks were deducted from the OD values obtained after the assays. The values obtained thereafter had been expressed as percentage compared to control. All experiments were performed at least in triplicate on three separate occasions. Data have been presented as mean ± SD. Triton X-100 (0.1%, 10 min) was used as positive control for the experiments.
Evaluation of cytotoxicity of PR-GO nanosheets by resazurin assay. The resazurin system measures the metabolic activity of living cells [51]. Resazurin is reduced to resorufin (highly fluorescent) in the medium by cell activity, and a direct correlation exists between the reduction of resazurin in the growth medium and the metabolic activity of living cells. Cells after incubation were washed twice in PBS. Resazurin assay was performed using Resazurin based in vitro toxicology assay kit (TOX-8) according to methods previously reported [39]. To eliminate variability, three replicates per concentration were maintained. Triton X-100 (0.1%, 10 min) was used as positive control for the experiments.
Evaluation of cytotoxicity of PR-GO nanosheets by neutral red uptake assay. Neutral red uptake assay was done to determine the accumulation of neutral red dye in the lysosomes of viable, uninjured cells [52]. Neutral red retention assay was performed according to the methods described previously [39]. Following incubation with PR-GO, the cells were washed carefully with PBS and incubated with neutral red solution (4 μg ml −1 in PBS) for 3 h. Following incubation cells were rinsed (thrice) with PBS to remove all dye not incorporated within cellular lysosomes. 200 μl of acidified ethanol (1% acetic acid, 50% ethanol, 49% D H 2 O) was added to each well and incubated in dark for 15 min, before being read at 540 nm in the plate reader (iMark™ Microplate Absorbance Reader (BIO-RAD, USA). All experiments were performed at least in triplicate on three separate occasions. Triton X-100 (0.1%, 10 min) was used as positive control for the experiments.
Evaluation of cytotoxicity of PR-GO nanosheets by flow cytometric estimation of propidium iodide (PI) uptake. Following treatment with PR-GO (0, 50, 100 and 250 μg/ml) for 3 h at 37˚C, cells were washed twice with PBS and re-suspended in 500 μl PBS (2 × 10 6 cells, 25 μM PI). A set of three replicates per concentration were maintained for the experiment. After a fifteen-minute incubation at room temperature in the dark, the PI fluorescence in cells was measured by flow cytometry (FACS Aria III flow cytometer, BD Biosciences, USA; excitation λ = 488 nm, emission λ = 530-540 nm) and 20,000 events were acquired for analysis. Triton X-100 (0.1%, 10 min) was used as positive control. Results were expressed as the mean fluorescence intensity. The interference of nanosheets in the assays was eliminated by maintaining sample blanks for all the concentrations tested.
Evaluation of genotoxicity of PR-GO by comet assay. Human lymphocytes incubated with PR-GO nanoparticles (0, 50, 100 and 250 μg/ml) for 3 h at 37˚C were processed for DNA damage studies following the method of Singh et al. [53], with modifications [41]. Slides were prepared in triplicates per concentration. Slides were immersed in cold lysis solution (pH 10; 2.5 M NaCl, 100 mM Na 2 EDTA, 10 mM Trizma base, 1% Triton X-100, 10% DMSO) and kept at 4˚C for 60 min. After lysis the DNA was allowed to unwind in the electrophoresis buffer (300 mM NaOH: 1mM Na 2 EDTA at pH 13.5) for 20 min at 4˚C. This was followed by electrophoresis conducted at a constant voltage of 26 V and 300 mA at 4˚C. Slides were neutralized in 0.4 M Tris (pH 7.5) for 5 min and finally rinsed in water. H 2 O 2 (25 μM, 10 min, 4˚C in dark) treated cells were used as positive control for the experiment. Each experiment was repeated twice.
The slides were stained with EtBr (20 μg/ml) and rinsed in water to wash off excess stain. Slides were scored using image analysis system (Kinetic imaging; Andor Technology, Nottingham, UK) attached to a fluorescence microscope (Leica, Wetzlar, Germany) equipped with appropriate filters (N2.1). The microscope was connected to a computer through a chargecoupled device (CCD) camera to transport images to software (Komet 5.5) for analysis. The This study was carried out in strict accordance with the recommendations and guidelines of the Bioethics Committee for Animal and Human Research Studies, University of Calcutta (Sanction No. BEHR/AM/180609). Fresh blood samples for lymphocyte culture were obtained from healthy volunteers after obtaining written informed consent. Subsequent experiments were performed in vitro and did not involve any form of human intervention or studies in animal model and hence no additional clearance was required. The Bioethics Committee for Animal and Human Research Studies, University of Calcutta was fully aware of and sanctioned this present research work.

Results and discussion
Physico-chemical characterization of crude polysaccharide the crude poloysaccharide that nwas isolated from the dried basidiocarps of P. flabellatus, gave a yield of 4.5 ± 0.51% dry weight. it consisted mainly of carbohydrate (43.75 ± 3.25%) with low amount of protein (12.5%). No phenols were detected in the polysaccharide. Total glucan content of the polysaccharide was 17.68 ± 1.89% among which 1.29 ± 0.31% were α-glucans and 16.39 ± 2.2% were β-glucans. To further investigate the physical structure of polysaccharide, Congo red assay was performed which showed a bathocromic shift from 491 nm to 500 nm (Fig 1A)indicating the presence of triple helix of β-1, 3-1, 6-glucans as Congo red solely interacts with triple-helical polysaccharides [54]. HPTLC (Fig 1B) and GC-MS (Fig 1D) analysis indicated that the fraction was heteroglucan. Mannose, glucose and galactose residues were detected in the ratio of 18: 49.3: 32.7 from GC-MS analysis data.  XRD analysis of the PR-GO. X-ray diffraction study was used to confirm the reduction of graphene oxide to crystalline graphene. XRD pattern that was obtained has been represented in the Fig 5. The XRD patterns clearly showed that the PR-GO nanosheets were formed by the bioreduction. The XRD pattern of pristine graphite exhibited a basal 002 reflection peak at 2θ = 26.6˚(d-spacing = 0.335 nm). However, after oxidation of pristine graphite to GO, the 002 reflection peak shifted to the lower angle at 2θ = 11.32˚(corresponding to a d spacing of 0.7816 nm), where the d-spacing increases due to the intercalation of oxygen functionalities in between the basal plane of graphite [57,58]. However, in our experiment, the appearance of a broad peak centered at 2θ = 23.5˚indicated the presence of stacked graphene layers [59], validating the formation of few layer graphene [60].
EDX observation of PR-GO. The Energy-dispersive X-ray (EDX) spectrum (Fig 6) recorded in spot-profile mode showed strong signals from the carbon atoms in the nanosheets. Sharp optical absorption peaks of carbon and oxygen in the range of 0 to 1 keV signified the presence PR-GO nano-crystallites. Each spectra showed an intense band at around~3430 cm -1 due to the presence of O-H stretching [61]. The presence of a new peaks at around 2920 cm -1 were owing to the absorption of the polysaccharide extract containing phenolic compounds on the surface of graphene. So, from the above observation it is clear that the increased reduction time leads to a greater removal of the oxygen functionalities from the GO surface, indicating a successful reduction of GO to PRGO [62]. However, the peak at 1625 cm -1 attributed to the aromatic C = C groups present in all three spectra. This suggests that the frame of sp 2 carbon atoms after reduction by crude polysaccharides was retained well, as before [63]. The characteristic peaks for C = O stretching vibration appeared at 1746 cm -1 in case of GO, while they were completely absent in the spectra of PR-GO as well as crude polysaccharides [63]. The FTIR peaks at around 1250 cm -1 and 1060 cm -1 , found in the spectra of GO and PR-GO nanoparticles, signified the epoxy C-O and alkoxy C-O stretching vibration, respectively [64]. However, the peak positions and intensities observed for GO changed significantly in case of PR-GO. This can also be attributed to the partial reduction of GO to graphene [56,60]. Bands observed at around 1420 cm -1 was due to N-H deformation [65] found in the spectra of crude polysaccharide, which shifted significantly to 1395.8 cm -1 (PR-GO spectra) due to the symmetric stretching of the carboxyl side groups of the polysaccharide molecules [66]. This indicated a strong interaction between the crude polysaccharide and the PR-GO nanoparticles. The spectral band around 1077 cm -1 of the polysaccharide spectrum indicated C-C and C-O stretching vibrations in pyranoid rings indicating pyranose form of sugar [67,68]. The bands visible between 500 and 749 cm -1 signified the presence of R-CH groups [69]. In the crude polysaccharide, a weak band at around 800 cm -1 signifies the presence of β-glycosidic bond and therefore indicated the existence of β-glucan. These analyses further corroborated our inference that the polysaccharide consists mainly of β-structures of pyranose sugars and that the interaction of GO with the crude polysaccharide formed PR-GO nanosheets.
Raman spectra of PR-GO nanosheets. Raman scattering from a material strongly depends on its electronic structure. Hence, it is very useful for characterization of materials and has been extensively used to investigate the structures of graphite and graphene like materials. In particular, Raman spectroscopy has been used to study the extent of structural changes undergone during reduction of GO. The relative intensities of the two fundamental vibrational bands between 1100 and 1700 cm -1 in the Raman scattered light observed for both GO and PR-GO are crucial for this purpose (Fig 8). The band with the lower Raman shift is known as the D-band and the other one near 1600 cm -1 is the G-band. The D-band originates from a  breathing mode vibration involving κ-point mode phonons of A 1g symmetry, while the Gband arises from first order scattering by vibrational modes of the sp2 carbons of GO and PR-GO with E 2g symmetry. Thus, the D-band arises from vibration of sp2 carbon clusters, and hence the relative increase of the intensity of the D-band compared to the G-band in a sample would indicate the presence of greater portions of such sp2 bonded carbon clusters without defects. In the context of comparison of the Raman spectra of GO and PR-GO, the increase of purely graphene like domains in the PR-GO was expected to cause an increase of the ratio of the D and G-band intensities because of the removal of oxygen containing moieties from GO. This has actually been observed in our samples. The accompanying figure shows the fundamental Raman bands in both the GO and the PR-GO, shifted vertically by an arbitrary amount for clarity. The D-band for GO occurs at 1327 cm -1 , and in case of PR-GO it shows a shift to 1334 cm -1 . For the G band the positions of the vibrational band peaks are 1584 cm -1 for GO and 1587 cm -1 for PR-GO. Also, the D-band becomes relatively more prominent in the Raman spectrum of PR-GO. The ratio of the peak intensities of the two bands increases from 1 in GO to about 1.15 in PR-GO. The lower ratio for GO indicates the destruction of the sp2 orbitals due to the linkage of the C atoms with oxidizing groups. Our observation was also consistent with those obtained by other workers using various other green reduction techniques [56,70]. In the work of Xing-Hua Xia et. al., [71] it was shown that the 2D peak becomes weak and  Green synthesis of graphene nanosheets and biosafety study broadened as we go from graphite to reduced graphene oxide. In our work, also this line became so weak and diffused that no 2D peak at 2750 cm -1 region could be observed.

Bio-compatibility of PR-GO nanosheets
While carbon based nanomaterials, including graphene oxide are potential candidates for biomedical applications, their bio-compatibility is of prime concern, as several studies have reported the general toxicity of graphene based nanomaterials [28]. Hence, in addition to synthesis of PR-GO nanosheets, considerable emphasis has been given to safety assessment of these particles. A battery of tests has been used to determine the toxicity of the synthesized particles in human PBMCs.
Cytotoxic effect of PR-GO nanosheets on human PBMCs. The cytotoxic potential of PR-GO nanosheets in human PBMCs was evaluated using more than one endpoint, to avoid false positive/ negative results [72]. MTT assay was used to study alteration in mitochondrial activity [73] in cells exposed to PR-GO nanosheets. From the results of MTT assay, no significant change in mitochondrial activity could be observed (Fig 9). The results of neutral red uptake assay indicated loss of lysosomal integrity, induced by PR-GO nanosheets at concentrations of 100 μg/ml and above (Fig 9). [74,75]. The effect of PR-GO on cell membrane integrity was studied using flow cytometry and confocal microscopy (Fig 10). PI uptake increased significantly at the highest concentration tested (250 μg/ml), indicating loss of membrane integrity. In resazurin assay, magnitude of dye reduction gives an overview of the cell's metabolic condition. In the present study resazurin assay revealed significant alteration in metabolic activity at the highest treatment concentration (250 μg/ml) as compared to control (Fig 11). Considering at least one of the tests indicated toxic response, with respect to membrane integrity, mitochondrial function, lysosome integrity, and metabolic activity; PR-GO could be considered safe below concentrations of 100 μg/ml. DNA damage induced by PR-GO in human PBMCs. In addition to cytotoxicity, genotoxicity of the particles was evaluated using comet assay. Genotoxicity assessment of particle could provide valuable information regarding DNA damage that might lead to better understanding of mutation and cancer. Genotoxicity of various graphene oxides has been reported by some authors [76][77][78]. In the present study, DNA fragmentation induced by PR-GO in PBMCs was statistically significant (P 0.05) at treatment dose 250 μg/ml. Treatment doses 50 and 100 μg/ml induced DNA damage, but was not statistically significant when compared to control (Fig 11). Significant DNA damage was induced by H 2 O 2 (% Tail DNA: 55±8) and was used as a positive control for the experiment.
Generation of ROS in human PBMCs. In addition to direct DNA damage, certain nanoparticles are known to induce oxidative stress. Flow cytometric analysis indicated concentration dependent increase in ROS production (Fig 12). There was a *1.8-fold increase in ROS generation at the highest concentration tested (250 μg/ml), and *6-fold increase in the positive control (H 2 O 2 ). In a study by Chang et al. (2011), the authors suggested of a dose-dependent oxidative stress in A549 cells, despite little/absence of cytotoxic response. In another study [21] the authors revealed of GO internalization, cell-cycle alterations, apoptosis and oxidative stress.

Conclusion
In conclusion, a green approach to the synthesis of graphene nanosheets is reported using exfoliated graphene oxide as the precursor and a crude fungal polysaccharide as the reducer. The use of an environment-friendly reducing and capping agent is what this method derives its merit from. The method should find practical applications in bulk-synthesis of graphene nanosheets. Also, from all the toxicity endpoints studied in human PBMCs, this newly synthesized nanoparticle could be considered biologically safe at concentrations below 100 μg/ml.

Author contributions
Conceptualization: KA AM DC.