Graphene Oxide (GO) as Antimicrobial Property

Over the past several years scientists have studied graphene materials (graphene, graphene oxide and reduced graphene oxide) for its antimicrobial properties and its future application in the biomedical field. To date, the exact mechanism for why graphene and its derivatives have antibacterial properties has not been fully understood due to experimental design variables. Scientists have agreed that oxidative stress, phospholipid extraction, and nanoknives all play an important role in the antibacterial properties of graphene materials. There are several techniques such as scanning electron microscopy, transmission electron spectroscopy, and mass spectroscopy that have been used to monitor change in morphology and cell death after exposure to graphene and its derivatives. Through these techniques scientist have been able to confirm that graphene materials are able to promote cellular death in both gram-positive and gram-negative bacteria, leading to its promising future also in pharmaceuticals. Graphene is a two-dimensional hexagonal structure, consisting of a basal plane (along the surface) and lateral edges, that are made up of sp² hybridized carbons which form a conjugated Ï€ system. This Ï€-conjugated system makes graphene very hydrophobic and able to interact with the hydrophobic cell membranes of bacteria. Graphene oxide (GO) is most commonly prepared using the Hummers method, which consists of a multi-step synthesis. First the graphite material is oxidized to form graphite oxide and then this is exfoliated via sonication to form a single layer of material called GO. Through the oxidation process oxygenated functional groups are introduced into the basal surfaces and lateral edges of GO. On the basal surface of GO, the functional groups of epoxides and hydroxyl groups are present, whereas, the larger groups such as carboxyl and carbonyl groups form along the lateral edges. It is important to note that with the introduction of the oxygenated functional groups, the basal plane and lateral edges will need to adopt sp³ hybridized carbons, which results in the formation of defect regions where the Ï€-conjugated system is broken. The presence of these defects will be explored in more detail below as these areas are critical in the antibacterial properties of GO.

In the literature, GO and other graphene materials have demonstrated through both physical and chemical mechanisms they have the ability to reduce bacterial cell count of both gram-negative and gram-positive bacteria¹. Several pivotal studies have been conducted in order to determine the key mechanism behind the antibacterial properties of GO. Nada et.al studied the antibacterial mechanisms using Raman spectroscopy², Zhang et.al studied the antibacterial activity using mass spectroscopy³, and Yusong et.al examined the effects of extraction of phospholipids from E-coli by graphene nanosheets⁴.These studies have led to the development of several mechanisms including the cutting of the membrane wall by the sharp edges of the defect sites (nanoknives), phospholipid extraction and oxidative stress. The various experiments to date regardless of experimental design demonstrated that it is not a single mechanism that is responsible for the antibacterial properties of GO but rather a combination of several mechanisms that can contribute to the overall antibacterial properties. Ultimately, it is the breakdown in the cell membrane that causes intracellular leakage of vital proteins, nucleic acids, and cytoplasmic material that results in the loss of reproduction and cell death.

The cutting mechanism also referred to in the literature as nanoknives is one of the most important mechanisms that lead to the antibacterial properties of GO. The sharp lateral edges of GO are able to penetrate into the cell membrane’s inner and outer layer. This leads to the loss of cell membrane integrity and leakage of intracellular materials such as the cytoplasm, nucleic acids, proteins and amino acids.

Nada et al used a variety of techniques, such as Raman spectroscopy and scanning electron microscopy (SEM) to observe the cellular changes (deformation or loss of cell membrane integrity) of bacteria, such as E Coli when exposed to varying concentrations of GO material. Raman spectroscopy was used to determine how E.coli reacts to an increasing concentration of GO. From the research done by Nada and team the bands with the greatest change were those of adenine (a nucleic acid) (729cm^-1), the S-S stretching vibrations (490cm^-1) of the disulphide bonds (found on the surface of the cell) and the amide group (found in proteins) bending vibration at 610cm^-1.² The intensity of each of these bands increased with the increase in GO concentration. This positively confirms that GO had penetrated into the cell membrane, causing the loss integrity and allowing the intracellular material such as adenine and other proteins to leak from the cell. Once this material was leaked from the cell, its vibrational absorption increase the signal in the Raman spectrum intensified. The examined the effects of GO on E. coli using SEM was used to observe the morphological changes in cells after exposure to GO. Figure 1 below, shows the morphological effects on E. coli with increasing GO concentration². Image A represents the control of E. coli cells. In image B, E. coli was treated with low concentration of GO causing morphological changes seen by the distortion (flattening or squishing) of the rod shape. As the concentration of the GO treatment increases the sharp edges of the GO material are able to penetrate into the cell membrane (as seen in the red arrow in image c) and lead to a loss of cell membrane integrity resulting in the leakage of the intracellular material. ² This effect was not isolated to only gram-negative bacteria but also gram-positive bacteria as well, in which the thickness of the cell membrane is different however similar results were observed. Therefore, the thickness of the cell membrane does not play a critical role in the nano-knife cutting mechanism as GO was able to penetrate both the inner and outer cell membrane walls.

The size of the GO also plays a key role in the cutting ability of GO due to the changes in the basal plane and lateral edges. A study lead by Perreault examined how altering the size of GO changed the cell viability of E. coli. GO sheets were altered by varying the sonication time and power in order to obtain GO sizes ranging from 0.65 µm² down to 0.01 µm².⁵ Using live/dead fluorescence staining it was determined that there was no change in cell viability (%living) when the bacteria was not in contact with GO as shown in the control of figure 2.⁵ The green dots represents the number of colony forming units (CFU) alive and the red dots are dead CFUs. The smaller sheet size of 0.01 µm² has significantly lower cell viability (30%) when compared to the 0.65 µm² sheet (73%).⁵ This phenomenon is explained by the fact that as the size of the GO material is decreased the amount of defects in the sp² hybridized carbons of the basal plane and edges increases. The defects are caused by either the addition of oxygen groups or shrinkage in size that destroys a portion of the basal plane. This effect creates additional sharp edges in both the lateral edges and basal plane leading to more nanoknives that are capable of cutting through the cell membrane. This allows for increase areas of interaction with the bacterial. This was also confirmed using SEM where larger sheet sizes (0.65µm²) showed very little deformation and almost normal cell structure whereas smaller sheets (0.01 µm²) appeared to be flattened and deformed leading to a compromised cell structure⁵. It is interesting to note that the amount of surface defects also plays a critical role in the formation of reactive oxygen species (ROS) that will create oxidative stress to the cell membrane and additional loss of membrane integrity. Therefore, it is impossible to determine if only the physiochemical mechanism of cutting is the single mechanism that impacts cell viability or multiple mechanisms do. In reality, with smaller GO size the loss of cell membrane integrity and intracellular leakage can be attributed to both the nano-knife effect and oxidative stress due to the increase in defects. This mechanism also has been shown to work in combination with other mechanisms such phospholipid extraction in which after cutting into the cell membrane the phospholipids are better able to be extracted by GO. This effect is due to hydrophobic attraction between the phospholipid tails and the Ï€-conjugated sp² carbons of the basal plane. This will be discussed in further details in the later sections.

GO is highly hydrophobic due to the Ï€-conjugation of the basal surface which allows it to interact with the phospholipid layer of the bacterial cell membrane. The oxygenated groups on GO provides a hydrophilic portion of the compound that is able to interact with the polar heads of the phospholipids. Yusong et al performed a detailed experiment on this and showed that extraction of the phospholipid molecules via hydrophobic interactions causes cell membrane deformation and collapse.⁵

There are two main mechanisms for the extraction of the phospholipids depending on the size of the GO material. With larger GO materials, the nanosheet will lie along the cell membrane causing a disruption of the phospholipid membrane; allow the GO material to embed itself into the lipophilic portions of the membrane¹. For smaller GO materials, the sheet will lie perpendicular to the membrane and penetrate through the cell membrane (either partially or completely) via the nano-knife mechanism. This further supports the theory that it is not a single mechanism that allows for the antibacterial properties but a combination of several.

The extraction of phospholipids occurs in several steps (see Figure 3). The first step involves a slight shifting of phospholipids due to the docking of the GO nanosheet as described by Yuesong (Image a).⁴ Next the phospholipid head will begin to break through the cell membrane and “climb” up the GO sheet (image b). This climbing effect is observed as result of the strong attractive forces between the hydrophobic regions of the basal plane and the hydrophobic tails of the phospholipids. This extraction process will continue until the GO material is covered completely with phospholipids with several layers of phospholipids climbing at the same time (image d). The phospholipids will evenly spread over the surface in such a way that the hydrophobic tails will align with the hydrophobic regions of the basal surface and the polar heads will align with the oxygenated functional groups of GO (images e and f).⁴ The loss of the phospholipids from the cell membrane causes a loss in membrane density which weakens the cell membrane. Due to the loss of density and the penetration of the GO into the cell, cell membrane integrity is lost resulting in the leakage intracellular fluids. This effect can be seen with the aid of TEM as morphological changes are easily identified using this method of analysis (see figure 4)⁴. Image A represents the initial state in which no morphological changes have occurred when E-coli was treated with GO.⁴ Over time as the phospholipid extraction occurs, the cell membrane density will decrease as shown in images b and c seen by the Type B arrows. The last step in the process is when cell integrity is lost and the intracellular material begins to leak. This effect is also observed in the images d-f where the cells are transparent due to loss of the cytoplasm⁴. This confirms that the phospholipid extraction occurs with the aid of the nanoknives leading to loss in cell viability.

Another group of scientist lead by Zhang’s used mass spectroscopy to map the metabolites of E-coli before and after exposure to GO.³ A typical mass spectrum of E-coli without GO treatment is shown Figure 5a. The most important peaks are at m/z 306 which corresponds to glutathione (GHS), which is an antioxidant commonly found in E. coli, and the membrane phospholipids phosphatidylethanolamine (PE) and phosphatidylglycerol (PG), at approximately m/z 600-800³. As the concentration of GO increased the overall intensities of the GSH, PE and PG all decreased (see fig 5b). This aligns with the previous studies of Yusong in which the extraction of the phospholipids leads to a decrease in cell membrane density. This was clearly observed by the decrease in signal intensities of both PE and PG.

In the mass spec results obtain by Zhang’s team there was also a decrease in GSH intensity. GSH is a critical antioxidant found in E. coli and will act as a scavenger for any free oxygen radicals that may be present in the bacteria’s environment. GSH prevents the oxidative breakdown of the phospholipid membrane that can also lead to a loss of cell membrane integrity and intracellular leakage. This finding confirms the previous hypothesis that there are many mechanisms running in parallel that ultimately attribute to the overall antibacterial properties of GO. In this very example, the nano-knives are able to cut into the phospholipid membrane and start the chain reaction of phospholipid extraction. This extraction causes a decrease in phospholipid density which in parallel with the cutting of the membrane by the nanoknives and oxidative stress leads to a loss in cell membrane integrity. This loss in cell membrane integrity leads to a loss of intracellular material such cytoplasmic material, nucleic acids and proteins..

As demonstrated by the mass spectrum results obtained by Zhang et al, oxidative stress occurs when there is a decrease in antioxidant concentration (such as GSH)³. The consumption of GSH occurs in the presence of reactive oxygen species (ROS) which are typically composed of hydrogen peroxide (H₂O₂), superoxide anions (O₂^*), hydroxyl radical (OH^*) or singlet molecular oxygen species(¹O₂)¹. When the antioxidants (such as GSH) are depleted the ROS compounds will then cause oxidation of key lipids in the cell membrane, proteins or even nucleic acids. The formation of the ROS compounds occurs due to the very nature of GO. When graphite is oxidized to form graphite oxide, the basal plane obtains additional defects as the result of the loss of Ï€-conjugation. This loss of Ï€-conjugation occurs due to the introduction of the oxygenated functional groups such as epoxides and hydroxyl groups which creates sp³ hybridized carbons. The introduction of the sp³ carbons alters the planer geometry of the basal plane by introducing tetrahedral carbons, creating defects in the lateral edges and basal plane. Additional defects can also be introduced as mentioned in the previous section when the size of the GO material is decreased. As the size of the sheets decreases the ability of the GO to stay as a perfect hexagonal structure also diminishes as the structure gets distorted to accommodate the smaller size.

The process of oxidative stress begins with the formation of ROS compounds. Oxygen is absorbed to the surface of the basal plane or lateral edges at the defects sites to form surface oxides. These surface oxides release ROS species in the presence of antioxidants as a result of a redox reaction. In the presence of GSH (an antioxidant found in E. coli), the surface oxides found on GO are reduced by an electron transfer from GSH. Another GSH molecule will then donate a proton to the ROS species, causing its reduction and subsequent release of a water molecule.⁵ The deprotonated GSH molecule will then react with another GSH molecule to form glutathione disulphide. If GSH is not available, the ROS species will then oxidize the lipids in the cell membrane to form as a lipid peroxide radical causing oxidative stress to the cell membrane. This oxidative stress will continue down the cell membrane until the lipid peroxide radical encounters an antioxidant bound to the cell membrane such as Vitamin E. It is important to keep in mind that with an increased amount of free oxygen radicals the extent of oxidative stress increases and the overall cell integrity decreases.

As mentioned earlier, the size and shape of the GO plays a key role in determine the extent of the oxidative stress. As the size of the GO material decreases from 0.065 to 0.01µm² the amount of surface defects increases. Again this is due to the loss of the conjugation and destruction of the basal plane. With the smaller GO material the presence of additional defects allows for the formation of many more surface oxides that can later be reduced to form ROS compounds. These ROS compounds then will continue to deplete the antioxidants and lead to further lipid peroxidation. This effect will continue until enough oxidative stress exists to cause the membrane to collapse and bacterial death to occur.

Perreault and team demonstrated this effect when they examined through Raman spectroscopy and cell viability testing that the size of the GO material plays a critical role in the ability of the material to induce oxidative stress. Figure 6 shows that as the size of the sheet decreases, the amount of disorder due to the defects (D band) increases. Therefore there is a direct correlation to number of defects and the ability of GO to form ROS species that interact with GSH. This is further confirmed with the graph in figure 6a which shows the %loss of GSH decreases with the increase in GO size. Therefore it not only the nanoknives that that are influenced by the size of GO but also oxidative stress due to the adsorbed oxygen and ROS formation.

As demonstrated through the various studies conducted by Nada et al, Zhang et.al, and Yusong et al the key antibacterial mechanisms of GO are cutting via nanoknives, phospholipid extraction and oxidative stress that contribute to the overall antibacterial properties of GO. The research to date indicates it is not a single mechanism that creates the antibacterial properties of GO but a combination of all factors. How much antibacterial is dependent on the physical size of GO plays and the number of defects in the basal surface and lateral edges. As the size of GO decreases a larger degree of surface defects forms. These defects play two important roles in the antimicrobial properties of GO. First, an increase in surface defects creates an increase in the number of sharp edges that causes a loss of cellular membrane integrity. Directly, the cell membrane is compromised by the penetration of the nano-knives resulting in intracellular leakage of the cytoplasm, amino acids, proteins and nucleic acids, Indirectly, once the nano-knives cut into the cell membrane it becomes embedded through attractive force between the hydrophobic tails of the cell membrane’s phospholipids and basal plane promotes phospholipid extraction. This extraction causes a decrease in cell membrane density which further breaks down the membrane causing additional intracellular leakage of key nucleic acids, cytoplasm and proteins. Lastly, the increase in surface defects of the basal plane increases the presence of reactive oxygen species. These species will react with any antioxidant present in the bacterial cell depleting their levels. Once depleted or reduced, the ROS will cause oxidation of the cell membrane lipids leading to a chain reaction of oxidative stress through the cell membrane. This also will lead to a loss of cell membrane integrity and intracellular leakage of key proteins and nucleic acids responsible for cell growth and replication. Without this material the bacteria cell will die leading to the antibacterial properties observed in GO.

With the development of drug resistant bacteria, new and inventive ways to treat bacterial infections need to be explored. GO with its unique chemical and physical properties show much promise as the next antibacterial treatment. Additional studies need to be conducted to determine the cytotoxicity of GO in animals and humans in order to development new treatment therapies.

1. Zou,X.; Zhang, L.; Wang, Z., Luo, Y. Mechanisms of the Antimicrobial Activities of Graphene Materials. J. Am. Chem. Soc. 2016, 138, 2064-2077
2. Nada, S.S; Yi, D.K; Kim, K. Study of antibacterial mechanisms of graphene oxide using Raman spectroscopy. Sci. Rep. 2016, 6, 28443; doi: 10.103,srep28443
3. Zhang, N.; Hou, j.; Chen, S.; Xiong, C.; Liu, H.; Jin, Y.; Wang, J.; He, Q.; Zhao, R.; Nie, Z.. Rapidly probing antibacterial activity of Graphene oxide by Mass Spectrometry-based metabolite fingerprinting Sci. Rep. 2016, 6, 28045; doi: 10.1038,srep28085
4. Yusong, T.; Lv, M.; Xiu; Huynh, T.; Zhang, M.; Castelli, M.; Liu, Z.; Huang, Q.; Fan, C.; Fang, H.; Zhou, R. Destructive extraction of phospholipids from Escherichia coli membranes by graphene nanosheets. Nat. Nanotechnol. 2013, 8, 594-601, doi:10.1038/nnano.2013.125
5. Perreault, F.; Fonseca de Faria, A.; Nejati, S.; Elimelech, M. Antimicrobial Properties of Graphene Oxide Nanosheets: Why Size Matters. ACS Nano. 2015, 7, 7226-7236