Effects of Graphene Oxide Nanoparticles on the Immune System Biomarkers Produced by RAW 264.7 and Human Whole Blood Cell Cultures

Graphene oxide nanoparticles (GONPs) have attracted a lot of attention due to their many applications. These applications include batteries, super capacitors, drug delivery and biosensing. However, few studies have investigated the effects of these nanoparticles on the immune system. In this study, the in vitro effects of GONPs on the immune system was evaluated by exposing murine macrophages, RAW 264.7 cells and human whole blood cell cultures (to GONPs. The effects of GONPs on RAW cells were monitored under basal conditions. The whole blood cell cultures were exposed to GONPs in the presence or absence of the mitogens lipopolysaccharide (LPS) and phytohaemmagglutinin (PHA). A number of parameters were monitored for both RAW and whole blood cell cultures, these included cytotoxicity, inflammatory biomarkers, cytokines of the acquired immune system and a proteome profile analysis. The GONPs were cytotoxic to both RAW and whole blood cell cultures at 500 μg/mL. In the absence of LPS, GONPs elicited an inflammatory response from the murine macrophage, RAW and whole blood cell cultures at 15.6 and 5 μg/mL respectively. This activation was further corroborated by proteome profile analysis of both experimental cultures. GONPs inhibited LPS induced interleukin 6 (IL-6) synthesis and PHA induced interferon gamma (IFNγ) synthesis by whole blood cell cultures in a dose dependent manner. In the absence of mitogens, GONPs stimulated IL-10 synthesis by whole blood cell cultures. The current study shows that GONPs modulate immune system biomarkers and that these may pose a health risk to individuals exposed to this type of nanoparticle.


Introduction
Graphene (G) is a two-dimensional (2-D) carbon material, with a hexagonal structure (honeycomb lattice) consisting of sp 2 hybridized atoms, which has been isolated from its three-dimensional parent material, graphite [1][2][3]. Graphene oxide (GO) is a single carbon layer graphene derivative, containing oxygen-bearing functional groups such as carboxyl and hydroxyl groups [2,4,5]. These groups are acquired when graphene undergoes oxidization [2]. These oxidized graphene nanoparticles possess unique characteristics such as electronic and thermal conductivity, superior mechanical strength and optical properties [6]. Due to these exceptional physiochemical properties of graphene oxide nanoparticles (GONPs), it has a broad range of applications in a number of fields. These fields include electrochemistry, biomedicine, biosensing, drug delivery, high energy capacity batteries and super capacitors to name a few. However, these properties do not necessarily ensure that GONPs are good drug delivery carriers [2,5]. . Data represents mean ± SD with n = 9. Bars marked with symbols indicate significant difference to control (p < 0.01). Significance demarcated by: *-significantly different compared to negative control (p < 0.001), # -significantly different compared to positive control (p ≤ 0.005).

The Effects of GONPs on the MIPs Chemokines Using RAW 264.7 Cells
The Effects of GONPs on MIP-1α Using RAW 264.7 Cells GONPs in the range of 15.6-62.5 μg/mL in media only notably (p < 0.001) increased the synthesis of MIP-1α, compared to the culture control not exposed to GONP (Figure 1d). However, GONPs concentrations ≥ 125 μg/mL significantly decreased the synthesis of MIP-1α compared to the culture control which was not exposed to GONP. The results for control cultures exposed to media in the presence of a mitogen only (803.85 ± 353.70 ng/mL MIP-1α) are not included on the figure.
The Effects of GONPs on MIP-1β Using RAW 264.7 Cells Cultures exposed to GONP concentrations in the range of 15.6-31.25 significantly (p < 0.001) upregulated the amount of MIP-1β secreted by the RAW cells compared to the culture control not exposed to GONP (Figure 1e). GONP concentrations ≥ 62.5 μg/mL, on the other hand significantly Positive control not represented (307.39 ± 171.86 ng/mL MIP-2). Data represents mean ± SD with n = 9. Bars marked with symbols indicate significant difference to control (p < 0.01). Significance demarcated by: *-significantly different compared to negative control (p < 0.001), # -significantly different compared to positive control (p ≤ 0.005).

The Effects of GONPs on the MIPs Chemokines Using RAW 264.7 Cells
The Effects of GONPs on MIP-1α Using RAW 264.7 Cells GONPs in the range of 15.6-62.5 µg/mL in media only notably (p < 0.001) increased the synthesis of MIP-1α, compared to the culture control not exposed to GONP (Figure 1d). However, GONPs concentrations ≥ 125 µg/mL significantly decreased the synthesis of MIP-1α compared to the culture control which was not exposed to GONP. The results for control cultures exposed to media in the presence of a mitogen only (803.85 ± 353.70 ng/mL MIP-1α) are not included on the figure.
The Effects of GONPs on MIP-1β Using RAW 264.7 Cells Cultures exposed to GONP concentrations in the range of 15.6-31.25 significantly (p < 0.001) upregulated the amount of MIP-1β secreted by the RAW cells compared to the culture control not exposed to GONP (Figure 1e). GONP concentrations ≥ 62.5 µg/mL, on the other hand significantly reduced MIP-1β synthesis compared to the culture control containing no nanoparticle. Results of media only exposed to LPS in the absence of GONP (1127.19 ± 468.69 ng/mL MIP-1β) are not included in the figure.
The Effects of GONPs on MIP-2 Using RAW 264.7 Cells GONP concentrations ≤ 62.5 µg/mL significantly (p < 0.001) increased MIP-2 synthesis compared to the control cultures exposed to media only without any nanoparticle (Figure 1f). MIP-2 synthesis was not affected by GONP concentrations ≥ 125 µg/mL. Media in the presence of LPS without GONP (307.39 ± 171.86 ng/mL MIP-2) is not included on the figure.
2.1.5. Secretory Cytokine and Chemokine Profile of RAW 264.7 Cells UponGONP Treatment Membranes exposed to media only, media in the presence of a mitogen and 15.6 µg/mL GONPs allowed for the analysis of various cytokines and chemokines expressed by the cells upon the relative exposure ( Figure 2). Quantification of the membranes showed that the simulated inflammatory response (media in the presence of LPS) allowed the cells to secrete certain cytokines and chemokines that were not synthesized by the cells when exposed to media only and 15.6 µg/mL GONPs. These proteins include: IFNγ-inducible protein 10 (IP-10); Granulocyte-macrophage colony-stimulating factor (GM-CSF); IL-6; macrophage colony-stimulating factor (M-CSF); interleukin 27 (IL-27); interleukin 1 receptor antagonist (IL-1ra); and IL-β. reduced MIP-1β synthesis compared to the culture control containing no nanoparticle. Results of media only exposed to LPS in the absence of GONP (1127.19 ± 468.69 ng/mL MIP-1β) are not included in the figure.
The Effects of GONPs on MIP-2 Using RAW 264.7 Cells GONP concentrations ≤ 62.5 μg/mL significantly (p < 0.001) increased MIP-2 synthesis compared to the control cultures exposed to media only without any nanoparticle (Figure 1f). MIP-2 synthesis was not affected by GONP concentrations ≥ 125 μg/mL. Media in the presence of LPS without GONP (307.39 ± 171.86 ng/mL MIP-2) is not included on the figure.
2.1.5. Secretory Cytokine and Chemokine Profile of RAW 264.7 Cells UponGONP Treatment Membranes exposed to media only, media in the presence of a mitogen and 15.6 μg/mL GONPs allowed for the analysis of various cytokines and chemokines expressed by the cells upon the relative exposure ( Figure 2). Quantification of the membranes showed that the simulated inflammatory response (media in the presence of LPS) allowed the cells to secrete certain cytokines and chemokines that were not synthesized by the cells when exposed to media only and 15.6 μg/mL GONPs. These proteins include: IFNγ-inducible protein 10 (IP-10); Granulocyte-macrophage colony-stimulating factor (GM-CSF); IL-6; macrophage colony-stimulating factor (M-CSF); interleukin 27 (IL-27); interleukin 1 receptor antagonist (IL-1ra); and IL-β.
However, the exposure of RAW cells to 15.6 μg/mL GONP in the absence of a mitogen affected the synthesis of certain proteins that significantly differed (p < 0.001) from both the negative (media only) and positive (media in the presence of LPS) controls. These cytokines and chemokines include: Granulocyte-colony stimulating factor (G-CSF); tumor necrosis factor α (TNF-α); monocyte chemoattractant protein 1 (MCP-1)/JE; intracellular adhesion molecule 1 (ICAM-1); MIP-1α; MIP-2; regulated on activation, normal T cell expressed and secreted (RANTES); and stromal cell-derived factor 1 (SDF-1) (Figure 3).  However, the exposure of RAW cells to 15.6 µg/mL GONP in the absence of a mitogen affected the synthesis of certain proteins that significantly differed (p < 0.001) from both the negative (media only) and positive (media in the presence of LPS) controls. These cytokines and chemokines include: Granulocyte-colony stimulating factor (G-CSF); tumor necrosis factor α (TNF-α); monocyte

The Effects of GONPs on Viability of Whole Blood Cell Cultures
GONP concentrations ≤ 50 μg/mL had no effect on cell viability ( Figure 4). However, the highest concentration of GONP evaluated, 500 μg/mL, significantly (p < 0.003) reduced viability compared to cultures exposed to media only in the absence of GONP. . Cell viability of whole blood cell cultures exposed to GONPs. Data represents mean ± SD with n = 4. Bars marked with symbol indicate significant difference to control (p < 0.01). Significance demarcated by: *-significantly different compared to negative control (p < 0.003).

The
Effects of GONPs on the Inflammatory System Biomarker IL-6 Using Whole Blood Cell Cultures Cultures exposed to GONPs alone significantly (p < 0.001) increased the IL-6 release from the cells at all the concentrations monitored when compared to the culture control exposed to media only ( Figure 5a). Conversely, GONPs concentrations in the presence of LPS significantly (p < 0.002) Figure 3. Quantification of cytokines secreted by RAW 264.7 cultures not stimulated with LPS after treatment with medium only or medium containing 15.6 µg/mL GONPs. Membranes subjected to chromogenic exposure. Data is represented as mean ± SD. Bars marked with symbols indicate significant differences (p < 0.01). Significance demarcated by: *-significantly different compared to the negative control (p < 0.001), # -significantly different compared to the positive control (p < 0.001).

The Effects of GONPs on Viability of Whole Blood Cell Cultures
GONP concentrations ≤ 50 µg/mL had no effect on cell viability ( Figure 4). However, the highest concentration of GONP evaluated, 500 µg/mL, significantly (p < 0.003) reduced viability compared to cultures exposed to media only in the absence of GONP.

The Effects of GONPs on Viability of Whole Blood Cell Cultures
GONP concentrations ≤ 50 μg/mL had no effect on cell viability ( Figure 4). However, the highest concentration of GONP evaluated, 500 μg/mL, significantly (p < 0.003) reduced viability compared to cultures exposed to media only in the absence of GONP. . Cell viability of whole blood cell cultures exposed to GONPs. Data represents mean ± SD with n = 4. Bars marked with symbol indicate significant difference to control (p < 0.01). Significance demarcated by: *-significantly different compared to negative control (p < 0.003).

The Effects of GONPs on the Inflammatory System Biomarker IL-6 Using Whole Blood Cell Cultures
Cultures exposed to GONPs alone significantly (p < 0.001) increased the IL-6 release from the cells at all the concentrations monitored when compared to the culture control exposed to media only ( Figure 5a). Conversely, GONPs concentrations in the presence of LPS significantly (p < 0.002) Figure 4. Cell viability of whole blood cell cultures exposed to GONPs. Data represents mean ± SD with n = 4. Bars marked with symbol indicate significant difference to control (p < 0.01). Significance demarcated by: *-significantly different compared to negative control (p < 0.003).

The Effects of GONPs on the Inflammatory System Biomarker IL-6 Using Whole Blood Cell Cultures
Cultures exposed to GONPs alone significantly (p < 0.001) increased the IL-6 release from the cells at all the concentrations monitored when compared to the culture control exposed to media only ( Figure 5a). Conversely, GONPs concentrations in the presence of LPS significantly (p < 0.002) reduced the secretion of IL-6 from the cells in a dose dependent manner compared to the control, exposed to media only containing the mitogen. reduced the secretion of IL-6 from the cells in a dose dependent manner compared to the control, exposed to media only containing the mitogen.
(a) (b) Figure 5. The inflammatory biomarker expression levels of whole blood cell cultures exposed to GO. NPs in the absence or presence of LPS: (a) IL-6 expression levels and (b) MIP-1β expression levels.

The Effects of GONPs on the Inflammatory Chemokine, MIP-1β Using Whole Blood Cell Cultures
MIP-1β was significantly (p < 0.001) upregulated by GONPs under unstimulated conditions at 5 and 50 μg/mL respectively, when compared to the culture control, not containing any nanoparticle ( Figure 5b). However, under a simulated inflammatory response, GONPs exhibited a prominent (p < 0.001) reduction in MIP-1β synthesis at concentrations ≥ 50 μg/mL when compared to the culture control, media only containing LPS.

The Effects of GONPs on the Humoral Immune System Biomarker IL-10 Using Whole Blood Cell Cultures
GONP concentrations ≤ 50 μg/mL in the absence of PHA significantly (p < 0.002) upregulatednIL-10 production compared to the culture control, not containing nanoparticle ( Figure  6a). However, the highest concentration (500 μg/mL) of the unstimulated cultures had no effect on the adaptive immune response. On the other hand, GONPs concentrations in the presence of PHA significantly (p < 0.001) inhibited IL-10 secretion in a dose dependent manner compared to the culture control, which only contained PHA.
(a) (b) Figure 5. The inflammatory biomarker expression levels of whole blood cell cultures exposed to GO. NPs in the absence or presence of LPS: (a) IL-6 expression levels and (b) MIP-1β expression levels. Data represents mean ± SD with n = 4. Bars marked with symbols indicate significant differences (p < 0.01). Significance demarcated by: *-significantly different compared to negative control (p < 0.001), # -significantly different compared to positive control (p < 0.001).

The Effects of GONPs on the Inflammatory Chemokine, MIP-1β Using Whole Blood Cell Cultures
MIP-1β was significantly (p < 0.001) upregulated by GONPs under unstimulated conditions at 5 and 50 µg/mL respectively, when compared to the culture control, not containing any nanoparticle ( Figure 5b). However, under a simulated inflammatory response, GONPs exhibited a prominent (p < 0.001) reduction in MIP-1β synthesis at concentrations ≥ 50 µg/mL when compared to the culture control, media only containing LPS.

The Effects of GONPs on the Humoral Immune System Biomarker IL-10 Using Whole Blood Cell Cultures
GONP concentrations ≤ 50 µg/mL in the absence of PHA significantly (p < 0.002) upregulatednIL-10 production compared to the culture control, not containing nanoparticle (Figure 6a). However, the highest concentration (500 µg/mL) of the unstimulated cultures had no effect on the adaptive immune response. On the other hand, GONPs concentrations in the presence of PHA significantly (p < 0.001) inhibited IL-10 secretion in a dose dependent manner compared to the culture control, which only contained PHA. GONP concentrations ≤ 50 μg/mL in the absence of PHA significantly (p < 0.002) upregulatednIL-10 production compared to the culture control, not containing nanoparticle ( Figure  6a). However, the highest concentration (500 μg/mL) of the unstimulated cultures had no effect on the adaptive immune response. On the other hand, GONPs concentrations in the presence of PHA significantly (p < 0.001) inhibited IL-10 secretion in a dose dependent manner compared to the culture control, which only contained PHA.
(a) (b) Figure 6. The acquired immune system biomarker expression levels of whole blood cell cultures exposed to GONPs in the absence or presence of PHA: (a) IL-10 expression levels and (b) IFNγ Figure 6. The acquired immune system biomarker expression levels of whole blood cell cultures exposed to GONPs in the absence or presence of PHA: (a) IL-10 expression levels and (b) IFNγ expression levels. Data represents mean ± SD with n = 4. Bars marked with symbols indicate significant differences (p < 0.01). Significance demarcated by: *-significantly different compared to negative control (p < 0.001), # -significantly different compared to positive control (p < 0.001).

Secretory Cytokine and Chemokine Profile of Whole Blood Cell Cultures Upon GONP Treatment
The proteome profile of whole blood cell cultures exposed to media only (negative control), media containing the mitogen, LPS (positive control) and 5 µg/mL GONPs in the absence of a stimulus revealed notable changes in the synthesis or inhibition of a number of cytokines and chemokines (Figure 7). GONP concentrations did not affect the cell mediated cytokine, IFNγ under normal conditions (i.e., in the absence of PHA) (Figure 6b). However, GONPs notably (p < 0.001) reduced the synthesis of IFNγ in a dose dependent manner in PHA stimulated cultures compared to media only containing PHA (culture control). This dose dependent decrease of IFNγ mimicked the response of the other adaptive immune cytokine, IL-10 ( Figure 6a).

Secretory Cytokine and Chemokine Profile of Whole Blood Cell Cultures Upon GONP Treatment
The proteome profile of whole blood cell cultures exposed to media only (negative control), media containing the mitogen, LPS (positive control) and 5 μg/mL GONPs in the absence of a stimulus revealed notable changes in the synthesis or inhibition of a number of cytokines and chemokines ( Figure 7). Figure 8 represents the quantification of the membranes. The whole blood cell cultures exposed to GONPs caused the significant upregulation of macrophage migration inhibitory factor (MIF), MCP-1 and interleukin 8 (IL-8) compared to both controls. The GONPs also notably upregulated serpin E1, RANTES and ICAM-1 compared to the negative control. Cytokines and chemokines that were prominently down regulated compared to the positive control include IL-1ra, MIP-1α/β, IL-6 and IL-1β (Figure 8). Whole blood cells exposed to media only (negative control) did not synthesize the following cytokines: IL-1ra, MCP-1, MIP-1α/β, IL-6, IL-8 and IL-1β.   E1, RANTES and ICAM-1 compared to the negative control. Cytokines and chemokines that were prominently down regulated compared to the positive control include IL-1ra, MIP-1α/β, IL-6 and IL-1β (Figure 8). Whole blood cells exposed to media only (negative control) did not synthesize the following cytokines: IL-1ra, MCP-1, MIP-1α/β, IL-6, IL-8 and IL-1β.

Discussion
Graphene has attracted lots of attention due its numerous potential industrial and medical applications. However, very little is known about the effects of these nanoparticles on the immune system. The exposure of RAW cells to GONPs induced cytotoxicity at the highest concentration (500 μg/mL). Similar data was obtained by Li et al. who exposed pristine graphene to RAW 264.7 cells [17]. They proposed that GONPs exerted its cytotoxic effects by depleting the mitochondrial membrane potential and increased the production of ROS. The decrease in cell viability could also be attributed to the higher capacity of macrophages to internalize GO and this could have affected the assembly of actin within the cell [18,19]. A reduction in cell viability was also reported when exposing human THP-1 cells to GONPs, indicating consistent cytotoxic effects against macrophages and monocytes regardless of species [4].
In the current study, inflammatory and proteome profiling data for GONPs, at basal conditions, increased the synthesis of inflammatory cytokines and chemokines such as NO, IL-6, TNF-α, MIPs and RANTES. These findings were corroborated by Orecchioni et al. as it was reported that, GONPs significantly stimulated the secretion of Th1/Th2 cytokines, such as IL-1a, IL-6, IL-10, TNF-α and GM-CSF, as well as chemokines such as MCP-1, MIP-1α, MIP-1β and RANTES in primary and immortalized murine macrophages [20]. This indicates activation of the macrophages by GONPs and could potentially be attributed to the interaction of GONPs with the toll-like receptors [21,22]. Feito et al. found that RAW 264.7 cells exposed to poly(ethylene glycol-amine) (PEG) functionalized GO upregulated TNF-α under both basal and LPS stimulated conditions [18]. IL-6 was found to be down regulated at 25 and 100 μg/mL GO in a simulated inflammatory response (+LPS). No differences were seen in IL-6 levels under basal conditions. The TNF-α result obtained by Feito et al. is consistent with what was found in the current study using proteome profiling [18]. The inflammatory cytokine, IL-6 was significantly upregulated at 15.6 and 31.25 μg/mL in the absence of a mitogen in the current study. This is contradictory to what was found by Feito et al but could be attributed to the functionalization of GO [18]. Functionalization of nanoparticles with biocompatible polymers increases their stability under physiological conditions. It also minimizes their interactions with other biomolecules and reduce immunological responses [8,18]. The effects of GONPs on RAW cells, under LPS stimulated conditions, were not evaluated in the current study. Figure 8. Quantification of cytokines secreted by whole blood cell cultures not stimulated with LPS after treatment with medium only or medium containing 5 µg/mL GONPs. Membranes subjected to chromogenic exposure. Data is represented as mean ± SD. Bars marked with symbols indicate significant differences (p < 0.01). Significance demarcated by: *-significantly different compared to the negative control (p < 0.001), # -significantly different compared to the positive control (p < 0.001).

Discussion
Graphene has attracted lots of attention due its numerous potential industrial and medical applications. However, very little is known about the effects of these nanoparticles on the immune system. The exposure of RAW cells to GONPs induced cytotoxicity at the highest concentration (500 µg/mL). Similar data was obtained by Li et al. who exposed pristine graphene to RAW 264.7 cells [17]. They proposed that GONPs exerted its cytotoxic effects by depleting the mitochondrial membrane potential and increased the production of ROS. The decrease in cell viability could also be attributed to the higher capacity of macrophages to internalize GO and this could have affected the assembly of actin within the cell [18,19]. A reduction in cell viability was also reported when exposing human THP-1 cells to GONPs, indicating consistent cytotoxic effects against macrophages and monocytes regardless of species [4].
In the current study, inflammatory and proteome profiling data for GONPs, at basal conditions, increased the synthesis of inflammatory cytokines and chemokines such as NO, IL-6, TNF-α, MIPs and RANTES. These findings were corroborated by Orecchioni et al. as it was reported that, GONPs significantly stimulated the secretion of Th 1 /Th 2 cytokines, such as IL-1a, IL-6, IL-10, TNF-α and GM-CSF, as well as chemokines such as MCP-1, MIP-1α, MIP-1β and RANTES in primary and immortalized murine macrophages [20]. This indicates activation of the macrophages by GONPs and could potentially be attributed to the interaction of GONPs with the toll-like receptors [21,22]. Feito et al. found that RAW 264.7 cells exposed to poly(ethylene glycol-amine) (PEG) functionalized GO upregulated TNF-α under both basal and LPS stimulated conditions [18]. IL-6 was found to be down regulated at 25 and 100 µg/mL GO in a simulated inflammatory response (+LPS). No differences were seen in IL-6 levels under basal conditions. The TNF-α result obtained by Feito et al. is consistent with what was found in the current study using proteome profiling [18]. The inflammatory cytokine, IL-6 was significantly upregulated at 15.6 and 31.25 µg/mL in the absence of a mitogen in the current study. This is contradictory to what was found by Feito et al but could be attributed to the functionalization of GO [18]. Functionalization of nanoparticles with biocompatible polymers increases their stability under physiological conditions. It also minimizes their interactions with other biomolecules and reduce immunological responses [8,18]. The effects of GONPs on RAW cells, under LPS stimulated conditions, were not evaluated in the current study.
Similarly to the RAW cells, GONPs also induced cytotoxicity in whole blood cell cultures at 500 µg/mL, and Orecchioni et al. exposed GONPs to peripheral blood mononuclear cells (PBMCs) and found that it induced its cytotoxic effects by inducing ROS-mediated apoptosis [6,23]. Zhi et al. also found that GO induced apoptosis in T lymphocytes in a dose dependent manner [24].
Exposing whole blood cell cultures to GONPs also induced the upregulation of the inflammatory cytokines IL-6 and MIP-1β under unstimulated/basal conditions. This activation of the inflammatory response could be attributed to the activation of the complement system which results in the production of anaphylatoxins (proinflammatory mediators). This may occur as GO may be sensed by components of the complement system as soon as GO is introduced into a biological system [25]. The activation of the complement system was studied by Wibroe et al. who found that there was an increase in C5a, which is a marker of the activation of the complement system when exposing human serum to GO [26]. Wibroe et al. also found that GO reduced the synthesis of IL-6 when human serum was stimulated by LPS, which is consistent with what was found in this study [26]. Proteome profiling data obtained in the current study confirmed that inflammatory cytokines and chemokines were activated when whole blood cell cultures were exposed to 5 µg/mL GONPs. Notably, IL-1ra, MCP-1, MIP-1α/β and IL-8 were upregulated by the GONPs.
GONPs not only inhibited an inflammatory response in a dose dependent manner but also modulated cytokines regulating the adaptive immune responses. The GONPs stimulated the synthesis of the humoral immune regulating cytokine, IL-10 but did not affect the cell mediated response regulating cytokine, IFNγ, from whole blood cell cultures under unstimulated/basal conditions. This indicates that GONPs would induce the differentiation of Th 0 cell to Th 2 cells. The B-lymphocytes are then stimulated to produce antibodies by interleukin 4 and 10 (IL-4 and IL-10) to combat the exposure to GONPs [27]. However, both IL-10 and IFNγ responses by whole blood cell cultures under PHA stimulated conditions were down regulated by GONPs.
This study clearly indicates that GONPs are cytotoxic at high concentrations and stimulates the activation of inflammatory responses under unstimulated/basal conditions. This activation/stimulation of an inflammatory response could cause autoimmunity to individuals exposed to GONPs [28]. Under unstimulated/basal conditions GONPs would also stimulate a humoral immune response in individuals exposed to them. However, under stimulated conditions mimicking an infection, both humoral and cell mediated responses were down regulated, indicating potential immunosuppression by GONPs exposure to individuals with an infection.
The data obtained in the current study is crucial for risk management of people upon GONP exposure, as these nanoparticles could be used in vivo as drug delivery systems, among other applications. GONPs may also be released into the environment and can potentially impact on human health.

Synthesis and Characterization of Graphene Oxide Nanoparticles (GONPs)
The graphene oxide nanoparticles were synthesized according to a modified Hummers' method [29], characterized and obtained from the University of Missouri. Graphene sheets were visualized via transmission electron microscopy (TEM, JEOL 1400, Peabody, MA, USA). Further characterization revealed the particles had a maximum absorbance at 227 nm with ultraviolet-visible spectrophotometry (UV-vis) (Lab Tech, UV 8100B, Hopkinton, MA, USA), with a scanning range 200-600 nm. A number of surface functional groups (i.e., hydroxyl and carboxyl groups) were detected by Fourier-transform infrared spectroscopy (FT-IR) in a Nicolet 4700 FT-IR spectrophotometer (Thermo scientific, Waltham, MA, USA) and the spectra collected over a wave number range from 400 to 4000 cm −1 . The electrophoretic mobility (EPM) of the nanomaterials was measured with a ZetaSizer Nano ZS (Malvern Instruments, Worcestershire, UK) in various electrolyte solutions by laser Doppler velocimetry. EPMs were converted to zeta-potential using the Smoluchowski equation (Equation (1)) [30]: where U is the electrophoretic mobility, ε is the dielectric constant of the solution, µ is its viscosity and ζ is the zeta potential. Disposable folded capillary cells were employed. Zeta potential indicated a surface potential of −49.2 mV at pH 7 and background ionic strength of 10 mM, NaCl (see Appendix A).  The RAW 264.7 cells (1 × 10 5 cells/mL) were cultured in cell culture treated 48 well plates and incubated in a humidified atmosphere of 5% CO 2 at 37 • C for approximately 48 h until the cells reached 80-90% confluence. After the incubation period, media was removed and replace with media containing 2.5% FBS. The subsequent procedures occurred in serum free media. The cells were pre-exposed for 2 h to various concentrations of GONPs. Thereafter the cells were left unstimulated and a positive control was also present. The positive control were cells only stimulated by lipopolysaccharide (LPS) (1 µg/mL) without the presence of nanoparticles and this would mimic an inflammatory immune response. The final concentration of FBS/well was 0.5%. Cultures were incubated overnight (~18 h) under standard tissue culture conditions. Culture supernatants were collected for nitric oxide (NO), interleukin 6 (IL-6), macrophage inflammatory protein 1α (MIP-1α), MIP-1β, MIP-2 and proteome profiling analysis.

NO Determination
After the overnight incubation of the RAW 264.7 cells, the amount of nitrite that was produced by the cells was measured in the culture supernatant as an indication of NO production. The NO assay is based on the Griess reaction [31]. The amount of NO production was measured against a doubling dilution range of a 100 µM nitrite standard (Sigma-aldrich). Nitrite standards or culture supernatant collected (100 µL) were mixed with 100 µL of Griess reagent (1:1 of 1% sulfanilamide and 0.1% naphtylethlemidimine-dihydrochloride in 2.5% phosphoric acid) (all reagents obtained from Sigma-aldrich). Thereafter, the plate was incubated at room temperature for 15 min. The absorbance was read at 540 nm by a microplate reader (Multiskan Ex, Thermo Electron Corporation) and the amount of NO produced by the RAW cells quantified. The MIP-1α samples were assayed at 1/270 and the LPS stimulated supernatant at 1/10,000. For the MIP-1β ELISA, the unstimulated culture supernatants were assayed at 1/20 while the LPS stimulated supernatants were assayed at 1/5000. The MIP-2 ELISA unstimulated supernatants were assayed at 1/100 and the mitogen stimulated supernatant at 1/1000.

Mouse Proteome Profiling Assay
A commercially available antibody array kit (Proteome Profiler, Mouse cytokine Array Panel A, R & D Systems) which was coated with 40 capturing antibodies in duplicate on a nitrocellulose membrane (dot blot) was used. The kit contained all the reagents for the assay and was performed as per the manufacturer's instructions. This cytokine and chemokine antibody array was used to determine the effects of GONPs exposure on cytokine and chemokine by RAW 264.7 macrophage cells.The assay required 500 µL of cell culture supernatants (unstimulated 0 µg/mL GONPs, LPS stimulated 0 µg/mL GONPs and unstimulated 15.6 µg/mL GONPs). Membranes were subjected to an ultra-sensitive chromogenic 3,3 ,5,5 -Tetramethylbenzidine (TMB) membrane substrate (Thermo Scientific, Waltham, MA, USA) to reveal sample-antibody complexes labeled with streptavidin-HRP. Photographs were taken of the blots after the exposure to the substrate.

Quantification of Pixel Density for Cytokine and Chemokine Membranes
Membrane images were quantified using image processing and analysis Java software(version 1.6.0_24, Oracle Corporation, Redwood city, CA, USA), ImageJ (version 1.4.3.67, National Institutes of Health, Bethesda, MD, USA). Levels of cytokines and chemokines were expressed as a percentage of the reference spot. Microsoft Excel was used to calculate the percentages which is expressed as mean ± standard deviation (SD).

Blood Collection
Blood was collected by a doctor/nurse from a healthy male who was not using any medication. The blood was collected using venipuncture directly into 3.2% sodium citrate vacuum tubes (Greiner bio-one). The blood was processed immediately. The whole blood cell cultures were performed under sterile conditions. Ethical clearance was obtained from the University of the Western Cape (Ethics No. 10/9/43). Informed consent was also obtained from the participant.

Cell Culture
Human whole blood was diluted with Roswell Parks Memorial Institute (RPMI) 1640 media (Sigma-aldrich) to give rise to a 10% (v/v) mixture. Blood was either left unstimulated or stimulated with LPS (Sigma-aldrich) (0.1 µg/mL) or phytohaemmagglutinin (PHA) (Sigma-aldrich) (1.6 µg/mL). The stimulated conditions would mimic an inflammatory and acquired immune response respectively. Unstimulated or stimulated whole blood cell cultures were incubated overnight at 37 • C with high, intermediate and low concentrations of GONPs in a 24 well tissue culture treated plate (Nunc, Waltham, MA, USA). Along with the nanoparticle exposure, a positive control of 0.01% Tween20 (Merck, Modderfontein, South Africa) was also present. After the incubation period, culture supernatants were screened for cytotoxicity by a lactate dehydrogenase (LDH) assay, inflammatory markers IL-6 and MIP-1β, interleukin 10 (IL-10) and interferon gamma (IFNγ).

LDH Assay
After the overnight incubation, an aliquot of the unstimulated whole blood cell culture supernatants were screened for cytotoxicity and measured the release of LDH from the cells (LDH-cytotoxicity colorimetric kit II, BioVision, Milpitas, CA, USA). The kit contained all the reagents for the assay and was performed as per the manufacturer's instructions.

Cytokine Analysis Using DAS ELISAs
Commercially available kits (e-Bioscience, Ready-Set-Go) were used to analyze the level of cytokine secretion from the whole blood cell cultures. The kits were used as per the manufacturer's instructions and contained all the reagents to complete the assay. The unstimulated and LPS stimulated samples were analyzed using a 1/10 dilution for the IL-6 assay. While the unstimulated and PHA stimulated samples were assayed neat for IL-10 and IFNγ analysis. The same protocol was used as previously described for the mouse cytokine ELISA.

Human MIP-1β DAS ELISAs
A human MIP-1β ELISA (R & D Systems) was performed on the unstimulated and LPS stimulated culture supernatants of the whole blood cell cultures. The samples were diluted 1/10 in reagent diluent, 0.1% bovine serum albumin (BSA) (Sigma-aldrich). The same protocol was followed as the mouse MIPs ELISAs.

Human Proteome Profiling
A commercially available antibody array kit (Proteome Profiler, Human Cytokine Array Kit, R & D Systems) which was coated with 36 capturing antibodies in duplicate on a nitrocellulose membrane (dot blot) was used. The kit contained all the reagents for the assay and was performed as per the manufacturer's instructions. This cytokine and chemokine antibody array was used to determine the effects of GONPs on cytokine and chemokine secretion when exposed to whole blood cell cultures. The assay required 500 µL of cell culture supernatants (unstimulated 0 µg/mL GONPs, LPS stimulated 0 µg/mL GONPs and unstimulated 5 µg/mL GONPs). The subsequent steps were carried out as described for the mouse cytokine and chemokine proteome profiling.

Statistical Analysis
All experiments were performed in triplicate and the data was calculated using Microsoft Excel. Data is presented as mean ± standard deviation (SD). One way analysis of variance (ANOVA) using SigmaPlot 12.0 (Systat Software Inc., San Jose, CA, USA) was used to assess statistical differences with p < 0.01 being deemed significant. Figure A1. Fourier-transform infrared spectroscopy (FTIR) analysis of GONPs. Figure A2. Ultraviolet-visible spectrophotometry (UV-vis) of GONPs, with maximum absorbance at 227 nm. Figure A2. Ultraviolet-visible spectrophotometry (UV-vis) of GONPs, with maximum absorbance at 227 nm.