Article pubs.acs.org/Langmuir
Lateral Dimension-Dependent Antibacterial Activity of Graphene Oxide Sheets Shaobin Liu,† Ming Hu,† Tingying Helen Zeng,‡ Ran Wu,† Rongrong Jiang,† Jun Wei,∥ Liang Wang,⊥ Jing Kong,‡,§ and Yuan Chen*,† †
School of Chemical and Biomedical Engineering, Nanyang Technological University, Singapore 637459 Center for Excitonics, Research Laboratory of Electronics, and §Department of Electrical Engineering and Computer Sciences, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, United States ∥ Singapore Institute of Manufacturing Technology, Singapore 638075 ⊥ Department of Chemistry and Chemical Engineering, Tianjin University of Technology, Tianjin 300384, People’s Republic of China ‡
S Supporting Information *
ABSTRACT: Graphene oxide (GO) is a promising precursor to produce graphene-family nanomaterials for various applications. Their potential health and environmental impacts need a good understanding of their cellular interactions. Many factors may influence their biological interactions with cells, and the lateral dimension of GO sheets is one of the most relevant material properties. In this study, a model bacterium, Escherichia coli (E. coli), was used to evaluate the antibacterial activity of well-dispersed GO sheets, whose lateral size differs by more than 100 times. Our results show that the antibacterial activity of GO sheets toward E. coli cells is lateral size dependent. Larger GO sheets show stronger antibacterial activity than do smaller ones, and they have different time- and concentration-dependent antibacterial activities. Large GO sheets lead to most cell loss after 1 h incubation, and their concentration strongly influences antibacterial activity at relative low concentration (20 μm.42 The lateral dimension of GO sheets is relevant for many biological phenomena that depend on particle size, such as cell adhesion and spread on large GO sheets, cell intake of small GO sheets, and cell deformation caused by interacting with GO. Furthermore, recent studies21,43,44 suggest that the antibacterial activity of GO is related to the actual lateral dimension of GO sheets in various suspensions, which is strongly correlated with their aggregation. Well-dispersed GO sheets demonstrate the strongest antibacterial activity among several graphene-family nanomaterials.21 On the other hand, aggregated GO sheets in LB (Luria−Bertani) medium are found to act as a cellular growth enhancer.44 The bacterial toxicity mechanism related to the lateral dimension of GO needs to be clarified for their various potential applications.
In this work, we used a model bacterium, Escherichia coli (E. coli), to evaluate the antibacterial activity of well-dispersed GO sheets with different lateral size. GO sheets at different sizes were prepared by sonication. Their lateral dimension and chemical properties were characterized by atomic force microscope (AFM), scanning electron microscope (SEM), Xray photoelectron spectroscopy (XPS), and ultraviolet−visible (UV−vis) absorption spectroscopy, respectively. Their time and concentration antibacterial activity was quantified by the colony counting method. The effect of aggregation and oxidative capacity was evaluated. The different interaction modes between GO sheets and cells were studied by AFM.
2. EXPERIMENTAL SECTION 2.1. GO Preparation. GO sheets were prepared by the modified Hummer’s method.45,46 2 g of K2S2O8, 2 g of P2O5, and 6 mL of 98% H2SO4 were mixed in a 50 mL beaker, and then heated to 80 °C in a 12365
dx.doi.org/10.1021/la3023908 | Langmuir 2012, 28, 12364−12372
Langmuir
Article
Figure 2. Histogram of the weight distributions of GO sheets with different lateral sizes. (A) GO-0, (B) GO-10, (C) GO-30, (D) GO-50, (E) GO120, and (F) GO-240. At least 200 GO sheets were measured for each sample by AFM to obtain the distribution. water bath. One gram of graphite powder (Bay carbon) was added into the mixture, and kept at 80 °C for 4 h. Next, the mixture was diluted using distilled water, and filtered through 0.20 μm nylon membrane, followed by thorough washing with deionized water and drying. Afterward, the as-treated dry graphite powder was added into 92 mL of H2SO4 in an ice bath. Five grams of KMnO4 was added slowly with stirring. The mixture was heated to 35 °C under vigorous stirring, and kept for 2 h. Next, 184 mL of water was slowly added; 15 min later, 560 mL of water and 10 mL of H2O2 were added. Solid powders were collected by centrifugation from the mixture, and then washed with 1:10 HCl and water. Last, graphite oxide was suspended in deionized water, and metal ions and acids were removed by dialysis. The exfoliation of GO was performed by bath sonication for 30 min. This sample was marked as “GO-0”. The GO sheets with different lateral sizes were prepared by sonication for 10, 30, 50, 120, and 240 min using a cup-horn sonicator (SONICS, VCX-130) at 130 W. These samples were marked as “GO-10”, “GO-30”, “GO-50”, “GO-120”, and “GO-240”, respectively. GO suspensions (3 μL) were dropped on freshly cut mica followed by air drying for AFM analysis. AFM analysis was performed on an MFP3D microscope (Asylum Research, Santa Barbara, CA) with a cantilever (Arrow NC, Nanoworld) in AC mode. Lateral dimension distribution of GO samples was determined by analyzing AFM images using the Image J software (National Institutes of Health, U.S.). At least 200 GO sheets were measured for each GO sample. SEM images were taken by drying GO suspensions on clean silicon wafers, and then viewed on a JEOL field emission-SEM (JSM-6700F), working at 5 kV. XPS spectra of GO sheets were collected using a Kratos Axis Ultra DLD (delay line detector) spectrometer equipped with a monochromatic Al Ka X-ray source (1486.69 eV). 2.2. Cell Preparation. E. coli K12 were grown in LB medium at 37 °C and harvested in the midexponential growth phase. Cultures were centrifuged at 6000 rpm for 10 min to pellet cells, and cells were washed three times with deionized water to remove residual macromolecules and other growth medium constituents. The pellets were then resuspended in deionized water. Bacterial cell suspensions were diluted to obtain cell samples containing 106−107 CFU/mL. 2.3. Cell Viability Test. E. coli cells were incubated with fresh GO suspensions in deionized water at 37 °C under 250 rpm shaking speed for 2 h. The viability of E. coli cells was evaluated by the colony counting method. Briefly, series of 10-fold cell dilutions (100 μL each) were spread onto LB plates, and left to grow overnight (12 h) at 37 °C. Colonies were counted, and compared to those on control plates to calculate changes in the cell growth inhibition. All treatments were
prepared in duplicate and repeated at least on three separate occasions. Loss of viability was calculated by the following formula: loss of viability % = (counts of control − counts of samples after incubation with suspensions)/counts of control. The cell visibility tests were also carried out in saline solution for comparison. 2.4. Cell Imaging by AFM.38 Cover glass slides were rinsed with alcohol, acetone, and deionized water in sequential steps. A cell sample of 10 μL was extracted from bacterial cell suspensions after incubation with GO for 2 h, and drop-cast directly on cleaned glass slides. To obtain AFM images, we dried and immobilized cells on glass slides under ambient air for 10 min. All samples were characterized within 30 min after drying. AFM images were obtained using a MFP3D microscope (Asylum Research, Santa Barbara, CA). To minimize lateral forces on cells during imaging, a silicon noncontact high resonance frequency cantilever (NCH type from Nanoworld with a spring constant of 42 N/m) was used to image cells during tapping mode. The scan rate was set to 1 Hz at various scan sizes, while all data resolutions were set to 256 pixels × 256 lines or 512 pixels × 512 lines. To remove the Z offset between scan lines, we applied the first-order flattening to AFM height profiles. Next, AFM height profiles were analyzed using the Igor Pro MFP-3D software to obtain root-meansquare (rms) roughness on 200 nm × 200 nm areas. Five E. coli cells and five different areas on the height profile of each cell were randomly selected to calculate their surface roughness. 2.5. Estimation of Large GO Sheet Quantity. The C−C bond length is 0.142 nm, and one GO hexagon unit contains 2 C atoms with an area of 0.052 nm2. We considered large GO sheets with size larger than 9 μm2 only, and they are more likely to cover an entire E. coli cell. Such GO sheets count for 21.2 wt % of total GO sheets in the GO-0 sample based on the analysis of AFM images similar to Figure 1A. The average size of these GO sheets is 13.06 μm2. The average number of C atoms in one such GO sheet is 5.023 × 108. Assuming the C/O ratio of GO sheets is 5:1,47 the number of O atoms in one such GO sheet is 1.005 × 108. The weight of one GO sheet is approximately 1.268 × 10−17 kg. Thus, the number of GO sheets larger than 9 μm2 in the GO0 suspension (5 μg/mL) is about 8.360 × 107 mL−1.
3. RESULTS AND DISCUSSION 3.1. GO Dispersions. The monolayer GO sheets were prepared by the modified Hummer’s method as described in the method section.45,46 As-prepared graphite oxide was first washed and dialyzed to remove metal ions and acid residues. Large GO sheets were then produced by sonicating graphite 12366
dx.doi.org/10.1021/la3023908 | Langmuir 2012, 28, 12364−12372
Langmuir
Article
energy of 283.17 eV suggests the existence of CC sp2 bonds in the GO sheets, while 285.21 eV results from C−O bonds (epoxy and hydroxyl groups), and a binding energy of 286.90 eV gives the evidence of CO (carbonyl) group formed during the oxidation. According to the integrated areas (ACC/ AC−O/ACO = 15 000/20 000/4500), the ratios of sp2-bonded carbon atoms versus carbon atoms in epoxy and hydroxyl groups and in carbonyl groups in the GO-0 sheets are roughly 3.33/4.44/1.48 Thus, the ratio of carbon atoms in the perfect graphene sheet to defects in the lattice is 1 to 1.63. The ACC/ AC−O/ACO ratios of GO-50 and GO-240 sheets are 3.29/ 3.94/1 and 3.60/4.67/1, respectively. XPS results suggest that our sonication can reduce the size of GO sheets without significantly changing their chemical properties. 3.2. Antibacterial Activity of GO. Antibacterial activity of GO sheets with different lateral sizes was evaluated using E. coli cells. GO aqueous suspensions (40 μg/mL) were first incubated with E. coli suspensions (106−107 CFU/mL) for 2 h under 250 rpm shaking speed. The viability of cells was evaluated by the colony counting method. Briefly, a series of 10-fold cell dilutions (100 μL each) were spread onto LB plates and left to grow overnight at 37 °C. Colonies were then counted and compared to those on control plates to calculate changes in the cell growth inhibition. Figure 3 shows the
oxide in a bath sonicator for 0.5 h. These GO sheets dispersed in water were marked as “GO-0”. Last, the large GO sheets (GO-0) in aqueous suspension were further sonicated in a cuphorn to yield smaller GO sheets. Sonication was conducted by placing GO suspensions in an ice bath, which avoids overheating samples.We used the sonication method to reduce the lateral size of GO sheets because this method results in negligible changes to the surface chemical properties of GO sheets. An aqueous suspension (3 μL) of each GO sample was dropped on a freshly cut mica for AFM analysis (Figure 1). AFM height profile in Figure 1 shows that these samples all have thicknesses of around 1 nm, confirming that monolayer GO sheets were prepared. It also shows no aggregations occurred during the sonication. Image J software (National Institutes of Health, U.S.) was applied to analyze AFM images to determine the lateral size of individual GO sheets. The average size of GO sheets is determined by measuring ∼200 GO sheets on at least 10 reprehensive AFM images. It is calculated by the equation: the average size (μm2) = the total area (μm2)/the total number of GO sheets. The average size of GO sheets in GO-0, GO-10, GO-30, GO-50, GO-120, and GO240 samples is 0.753, 0.127, 0.065, 0.035, 0.013, and 0.010 μm2, respectively. Although the difference in their average size among six samples is significant, the average size alone may not be adequate to describe their size difference. There are large numbers of small GO sheets, even though they only count for a small weight fraction. Lateral dimension distribution of GO samples was analyzed by the Image J software. Because most GO sheets are monolayer, the weight of each GO sheet is proportional to its lateral area. To compare their concentration-dependent antibacterial activity, we quantify their lateral dimension distribution based on the weight fraction of GO sheets with various lateral sizes in a sample, other than just counting the number of GO sheets. The weight distributions of GO sheets in different samples are presented in Figure 2. It also should be noted the average size of GO-0 and GO-10 samples is small because of the existence of many small GO sheets (for example, 0.4 μm2). The size of the largest GO sheets (GO-0) is several orders larger as compared to those in GO240. The weight fraction of GO sheets with smaller sizes also increases significantly with extending sonication time. This shows that GO samples with six different lateral sizes are prepared. SEM images were also taken on GO films prepared by drying GO aqueous suspensions on silicon wafers. As shown in Supporting Information, Figure S1, GO films are comprised of multilayer GO sheets forming circular rings, which are smooth at the center with small wrinkles at their edges. The diameter of the rings decreases with the decrease of GO sheet lateral sizes. The carbon-related chemical component species in GO samples were characterized by high-resolution XPS. Figure S2 in the Supporting Information shows survey scans (pass energy of 1100 eV and step size of 1 eV) of three GO samples (GO-0, GO-50, and GO-240). The peaks from impurities are small except for the major elements of carbon and oxygen, showing impurities were removed in the exfoliation oxidation. The impurities would have little influence on the physiochemical properties of GO sheets. High-resolution XPS analysis of carbon components in the GO sheets is shown in Figure S3 (pass energy of 20 eV and step size of 0.1 eV). A binding
Figure 3. The viability of E. coli cells (5 mL of 106−107 CFU/mL) after incubating with GO suspensions (5 mL of 80 μg/mL) for 2 h with 250 rpm shaking speed at 37 °C. Loss of viability was calculated by the following formula: loss of viability % = (counts of control − counts of samples after incubation with suspensions)/counts of control.
viability of E. coli cells decreased by 97.7 ± 1.0%, 91.5 ± 0.5%, 87.8 ± 2.09%, 71.2 ± 6.0%, 60.7 ± 8.8%, and 45.5 ± 5.5% after incubating with GO-0, GO-10, GO-30, GO-50, GO-120, and GO-240 aqueous suspensions, respectively. GO samples with larger lateral size show stronger antibacterial activity than do smaller ones. Parallel experiments were also carried out in saline solution under physiological circumstance. As shown in Figure S4 in the Supporting Information, the trend found about the size-dependent antibacterial activity is similar. The control experiments without GO sheets also suggest that culture solvents do not affect the E. coli cell survival in our tests. However, the antibacterial activity of GO sheets in saline solution is significantly lower than that obtained in deionized water, especially for larger size GO samples. This can be attributed to the aggregation of large GO sheets in saline solution. In our previous study,21 we have showed that aggregation can greatly affect the antibacterial activity of graphene-based materials. When GO sheets aggregate together, their antibacterial activity is decreased. Thus, we dispersed GO 12367
dx.doi.org/10.1021/la3023908 | Langmuir 2012, 28, 12364−12372
Langmuir
Article
Figure 4. (A) Time-dependent antibacterial activity of GO-0 and GO-240 suspensions. Five milliliters of GO-0 or GO-240 (80 μg/mL) was incubated with E. coli cells (106−107 CFU/mL, 5 mL) for 240 min (the concentration of GO in mixtures is 40 μg/mL). The loss of viability was measured at 0, 30, 60, 120, and 240 min, respectively. Deionized water without GO samples was used as control. (B) The time-dependent E. coli cell inactivation rates after incubating with GO-0 and GO-240 suspensions, which are extracted from Figure 5A with the unit of d(%)/d(min). (C) Concentration-dependent antibacterial activities of GO-0 and GO-240 suspensions. Five milliliters of GO-0 or GO-240 suspension (at 10, 20, 40, 80, and 160 μg/mL) was incubated with E. coli cells (106−107 CFU/mL, 5 mL) for 2 h. (D) The dependence of E. coli cell inactivation rates on GO concentration, which are extracted from Figure 5C with the unit of d(%)/d(μg/mL). Data in (A) and (C) are fitted with exponential curves. The fitting results are presented in the figures.
with E. coli cells (106−107 CFU/mL) for 2 h with 250 rpm shaking speed at 37 °C. Figure 4C shows the loss of E. coli cell viability increase as the concentration of GO-0 or GO-240 suspensions increases. For large GO sheets (GO-0), it is 55.7 ± 7.7% at 5 μg/mL, and increases to 80.3 ± 4.8%, 90.1 ± 4.9%, 97.7 ± 1.0%, and 99.4 ± 0.4% at 10, 20, 40, and 80 μg/mL, respectively. In contrast, small GO sheets (GO-240) exhibit much weaker antibacterial activities of 7.8 ± 3.1%, 16.7 ± 5.8%, 29.4 ± 5.2%, 45.5 ± 5.5%, and 52.5 ± 6.7% at 5, 10, 20, 40, and 80 μg/mL, respectively. The dependence of E. coli cell inactivation rates on GO concentration is shown in Figure 4D. The concentration of large GO sheets strongly influences their antibacterial activity at relative low GO concentration (20 μg/mL), while the concentration of small GO sheets has less impact on their antibacterial activity. We fitted the data in Figure 4A and C with an exponential curve. The fitting results are shown in Figure 4. From the time-dependent experimental data, the fitted maximum inactivation rate of GO-0 is 103.4%, which is similar to the value obtained from the concentrationdependent experimental data at 97.41%. In contrast, the fitted maximum inactivation rate of GO-240 is 58.38% or 56.84%. The significant difference between the two GO samples indicates that the antibacterial activity of GO sheets is size dependent. 3.4. Aggregation and Oxidation Capacities. Our previous study shows that membrane and oxidative stress play important roles in the biological interactions between
sheets in deionized water in this study to minimize the influence of aggregation. 3.3. Time- and Concentration-Dependent Antibacterial Activity. We further examined the time-dependent antibacterial activity of GO samples. E. coli cells were incubated with GO-0 or GO-240 suspensions (40 μg/mL) for different time up to 4 h before cell viability tests were carried out by the colony counting method. The loss of cell viability after different incubation periods is shown in Figure 4A. It increases with the extending incubation time for both large (GO-0) and small (GO-240) GO sheets. However, they change in different trends. For large size sheets, the loss of E. coli cell viability increases sharply from 38.7 ± 4.8% after 30 min to 89.4 ± 3.3% after 1 h of incubation. It only increases slightly to 97.7 ± 1.0% after 2 h and 99.3 ± 0.6% after 4 h. The change after 1 h of incubation is due to the fact that the loss of cell viability has approached the complete inactivation (∼99%). In contrast, the loss of E. coli cell viability for small size GO sheets increases steadily. It is 18.5 ± 3.0% after 30 min and 33.1 ± 5.1%, 45.5 ± 5.6%, and 56.2 ± 5.1% after 1, 2, and 4 h of incubation, respectively. Figure 4B further illustrates the inactivation rate of E. coli cells. After interacting with large size GO sheets, it is much higher in the first hour, and sharply decreases after 2 h. In contrast, after interacting with small size GO sheets, it displays a moderate decreasing trend. Next, the concentration dependence of antibacterial activities was examined. GO-0 and GO-240 suspensions at various concentrations (5, 10, 20, 40, and 80 μg/mL) were incubated 12368
dx.doi.org/10.1021/la3023908 | Langmuir 2012, 28, 12364−12372
Langmuir
Article
graphene-family nanomaterials and bacterial cells.21 Graphenefamily nanomaterials may deposit on cells, which could disrupt and damage cell membranes, resulting in cell death. They may also induce oxidative stress on bacteria. Well-dispersed GO sheets are more likely to interact with bacterial cells in aqueous suspensions as compared to aggregated reduced GO. Conductive graphene or reduced GO would mediate more intensive oxidative stress as compared to insulating GO, if they are both in good contact with cells.21 However, in the current study, aggregation or dispersion of GO sheets unlikely contributes to the antibacterial activity difference shown in Figure 3, because GO sheets with various lateral sizes are all well dispersed in their mixtures with E. coli cells. Figure S5 in the Supporting Information shows clear GO sheet and cell suspensions after mixing them. After 2 h incubation, the suspensions of GO and E. coli mixtures remain transparent. No significant aggregation or solid particle precipitation was observed during the incubation for all tested GO samples.
Another possible reason for the different antibacterial activity among GO samples could be the diverse oxidation capacity owned by GO sheets with different lateral sizes. They might induce oxidative stress at different levels on E. coli cells. To evaluate this possibility, we compared the oxidation capacity of different GO samples toward glutathione (GSH). GSH is a tripeptide with thiol groups serving as a redox state mediator in bacteria. Thiol groups (−SH) in GSH can be oxidized by carbon nanomaterials to disulfide bond (−S−S−), which converts GSH to glutathione disulfide. Ellman’s assay was used to quantify the concentration of thiol groups in GSH, and the experimental details are described in the materials and methods session. GSH (0.4 mM) is oxidized in vitro by GO suspensions (40 μg/mL). Bicarbonate buffer (50 mM at pH = 8.6) without GO was used as a control. Figure 6 shows that about 22% of GSH is oxidized after exposure to GO. GO samples with different lateral sizes result in no significant difference on the loss of GSH, suggesting they have similar oxidation capacity. Furthermore, XPS results in Figures S2 and S3 show the ratios of C atoms in different functional groups on GO sheets with different size have small changes. Figure S6 in the Supporting Information shows the UV−vis spectra of different GO sheets are almost identical, confirming that their surface chemical properties are similar. Results here suggest that the oxidative capacity of GO sheets is more a function of their properties, such as conductivity or surface defect densities, other than the number of their edge sites related to their lateral dimension. 3.5. Interaction between GO and E. coli Cells. AFM was used to survey the interactions between GO sheets and E. coli cells. Figure 6 shows AFM images of E. coli cells after incubating with large GO (GO-0) and small GO (GO-240) sheets for 2 h. As a control, the cell in deionized water without GO sheets preserves its membrane integrity after 2 h of incubation (Figure 6A and B). E. coli cells were covered by large GO sheets (Figure 6C and D). In contrast, small GO sheets
Figure 5. Oxidation of glutathione by GO sheets with various lateral sizes. Loss of glutathione (0.4 mM) after in vitro incubation with 40 μg/mL of GO suspensions for 2 h. The bicarbonate buffer (50 mM at pH = 8.6) without GO was used as a control.
Figure 6. AFM amplitude and 3D images of E. coli cells after incubation with GO sheets. (A,B) E. coli incubation with deionized water for 2 h, (C,D) E. coli incubation with the 40 μg/mL GO-0 suspension for 2 h, and (E,F) E. coli after incubation with the 40 μg/mL GO-240 suspension for 2 h. The scale bars are 1 μm. 12369
dx.doi.org/10.1021/la3023908 | Langmuir 2012, 28, 12364−12372
Langmuir
Article
(Figure 6E and F) adhere to the surface of cells, and do not fully cover the cell surface. The AFM surface roughness analysis corroborates with images shown in Figure 6. The surface roughness values of E. coli cells shown in Figure 6A, C, and E are 4.78 ± 0.82, 2.18 ± 0.46, and 14.80 ± 3.19 nm, respectively. After incubation with large GO-0 sheets, the cell surface roughness decreases (Figure 6C), which can be credited to the smooth GO sheets covering cell surfaces. However, after incubation with small size GO nanosheets (GO-240), cell surface roughness significantly increases, which is credited to the adhesion of many small GO fragments. Figure S7 in the Supporting Information shows larger field AFM images including more cells. Overall, the observed difference between GO sheets of different lateral dimension is statistically significant, other than only happening on one or two isolated cells. Monolayer graphene sheets are impermeable atomic membranes to many molecules.49 Akhavan et al. proposed that graphene sheets can wrap bacteria, biologically isolating them from growth medium.43 Therefore, cells can neither consume the nutrients nor proliferate. We find that this proposed mechanism could be applicable with the findings in our current study. Although GO sheets may have higher permeability as compared to graphene sheets, layers of GO sheets can surely isolate E. coli cells, as shown in Figure 6C and D. If the surface of an E. coli cell is covered by sufficient GO sheets, the covered area may become biologically inactive (i.e., transport though cell membrane is blocked), and the cell cannot proliferate, resulting in the cell viability loss observed in the followed colony counting test. Previous studies have shown that hydrophobic surfaces should attract bacteria more strongly than hydrophilic surfaces.50,51 Large GO sheets have more hydrophobic sp2 carbon, which might have a stronger interaction with the lipid bilayer of cell membranes. Akhavan et al. reported that about one-third of the bacteria could be reactivated after 48 h inactivation of the bacteria within the aggregated graphene sheets.43 We tried to release cells from the confinement of GO sheets by bath sonication as pervious researchers did,43 and then tested their viability loss again after sonication. As shown in Figure S8 in the Supporting Information, large GO sheets wrap cells firmly, and no live cells are released from the confinement of GO sheets. For cells interacting with small GO sheets, the bath sonication for 1−5 min has no effect on the cell viability loss, suggesting no live cells are released. However, if the sonication time is extended to 15 min, much more cells are killed, which may be caused by mechanical shear forces induced by the excessive sonication. These results show that for both the smallest and the largest GO sheets that in neither case the cells can be easily released, suggesting the GO−bacterial cell interaction is very strong. We propose the lateral dimension-dependent antibacterial activity of GO sheets can be credited to their efficiency in covering E. coli cell surface. It is easier for larger GO sheets to cover most of the cell surface during the 2 h incubation as compared to smaller GO sheets. This active-site blocking mechanism mainly depends on the lateral dimension of GO sheets, because the incubation condition is the same for all samples. Besides, all GO sheets are well dispersed in the suspension mixtures, as shown in Figure S5. Figure 7 shows a correction between the loss of E. coli cell viability and the average size of GO sheets. The loss viability increases rapidly when the average GO size increases from 0.010 to 0.127 μm2. Afterward, even though there are many more large GO sheets
Figure 7. Correlation between the loss of E. coli cell viability and the average size of GO sheets.
existing in the GO-0 sample, the loss viability only shows a minor increase. Figure 2 show that both GO-10 and GO-0 samples contain a significant fraction of large GO sheets. In all of our tests, the number of GO sheets is far more than the number of E. coli cells, even at the lowest GO concentration of 5 μg/mL. GO-0 suspension at 5 μg/mL contains an estimated 8.36 × 107 mL−1 large GO sheets with lateral size larger than 3 μm × 3 μm (see the detailed calculation in the materials and methods section). This is 10 times the number of E. coli cells at 106−107 mL−1. A GO sheet with lateral size larger than 9 μm2 can easily cover an entire E. coli cell, considering most E. coli cells are smaller than 3 μm in one dimension as shown in Figure 6. Once cells are fully covered by GO sheets, they cannot proliferate. The colony counting method measures the proliferation of cells, and thus its results are direct correlated with the entrapment of cells. It also should be noted that small GO fragments exist in all six samples, and their total number is much higher than that of large GO sheets as shown in Figure 2. The small GO fragments also contribute to the observed antibacterial activity, which is evidenced by the more than 50% inactivation of E. coli by the GO-240 sample after 2 h incubation. The fitted maximum inactivation rate of GO-240 is ∼57% shown in Figure 4. If we assume small GO fragments play similar roles in other GO samples, the observed lateral dimension-dependent antibacterial activity of GO sheets is likely contributing to the killing of the additional ∼40% of cells. Considering the number of large GO sheets is much less than that of small GO sheets, this suggests that large GO sheets are much more efficient in killing E. coli cells. It is still unclear whether the cells would be immediately killed once they are covered by GO sheets or over some time. Cells could be eventually killed over some time because of several mechanisms, such as membrane stress, oxidative stress, or lack of nutrients. This would require further studies.The timeand concentration-dependent antibacterial activity shown in Figure 4 can also be explained on the basis of the proposed mechanism. It is easier for larger size GO sheets to cover the whole surface of cells; thus most of the cells have been wrapped by the large GO-0 sheets after 1 h of incubation. With extending the incubation time to 4 h, only a minor increase in the loss of cell viability is observed. In contrast, smaller size GO sheets (GO-240) continue causing the death of more cells with the increase of incubation time, thus resulting in a continuous increase in cell viability loss. For the dependence on GO concentration, because the number of GO sheets is much larger than the number of cells in suspensions, the majority of cells are 12370
dx.doi.org/10.1021/la3023908 | Langmuir 2012, 28, 12364−12372
Langmuir
■
covered by large GO sheets at relative low concentration, and thus minor changes could be observed with the increase of GO concentration. On the other hand, for small GO sheets, increasing GO concentration increases the chance of interactions between GO sheets and cells, resulting in a steady increase in the loss of cell viability. Several previous studies have also proposed that nanomaterials with high aspect ratio, such as carbon nanotubes, may frustrate certain cells because cells cannot ingest long materials.34,52−54 The studies on mammalian cells may have no relevance to bacterial cells. Nevertheless, if E. coli cells do uptake GO sheets, and ingesting frustration is a cause of cell death, our results did not show it would play a significant role in our study. In our AFM study, no large GO sheets that are partially untaken by E. coli cells were observed. Some small GO sheets seem partly plunged in cell membranes. However, our AFM study cannot confirm whether there are small GO sheets inside cells. If they are inside cells, whether they are toxic to cells also requires further studies.
AUTHOR INFORMATION
Corresponding Author
*Phone: (65) 63168939. Fax: (65) 67947553. E-mail:
[email protected]. Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS This work was supported by the National Research Foundation, Singapore (NRF-CRP2-2007-02), the Ministry of Education, Singapore (MOE2011-T2-2-062), and the Center for Center for Excitonics at Massachusetts Institute of Technology, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Basic Energy Sciences.
■
REFERENCES
(1) Paredes, J. I.; Villar-Rodil, S.; Martinez-Alonso, A.; Tascon, J. M. D. Graphene Oxide Dispersions in Organic Solvents. Langmuir 2008, 24, 10560−10564. (2) Compton, O. C.; Nguyen, S. T. Graphene Oxide, Highly Reduced Graphene Oxide, and Graphene: Versatile Building Blocks for Carbon-Based Materials. Small 2010, 6, 711−723. (3) Zhu, Y.; Murali, S.; Cai, W.; Li, X.; Suk, J. W.; Potts, J. R.; Ruoff, R. S. Graphene and Graphene Oxide: Synthesis, Properties, and Applications. Adv. Mater. 2010, 22, 3906−3924. (4) Geim, A. K.; Novoselov, K. S. The Rise of Graphene. Nat. Mater. 2007, 6, 183−191. (5) Li, D.; Kaner, R. B. Graphene-Based Materials. Science 2008, 320, 1170−1171. (6) Li, D.; Muller, M. B.; Gilje, S.; Kaner, R. B.; Wallace, G. G. Processable Aqueous Dispersions of Graphene Nanosheets. Nat. Nanotechnol. 2008, 3, 101−105. (7) Park, S.; Ruoff, R. S. Chemical Methods for the Production of Graphenes. Nat. Nanotechnol. 2009, 4, 217−224. (8) Dikin, D. A.; Stankovich, S.; Zimney, E. J.; Piner, R. D.; Dommett, G. H. B.; Evmenenko, G.; Nguyen, S. T.; Ruoff, R. S. Preparation and Characterization of Graphene Oxide Paper. Nature 2007, 448, 457−460. (9) Gilje, S.; Han, S.; Wang, M.; Wang, K. L.; Kaner, R. B. A Chemical Route to Graphene for Device Applications. Nano Lett. 2007, 7, 3394−3398. (10) Park, S.; An, J. H.; Jung, I. W.; Piner, R. D.; An, S. J.; Li, X. S.; Velamakanni, A.; Ruoff, R. S. Colloidal Suspensions of Highly Reduced Graphene Oxide in a Wide Variety of Organic Solvents. Nano Lett. 2009, 9, 1593−1597. (11) Cote, L. J.; Kim, F.; Huang, J. X. Langmuir-Blodgett Assembly of Graphite Oxide Single Layers. J. Am. Chem. Soc. 2009, 131, 1043− 1049. (12) Kim, F.; Cote, L. J.; Huang, J. X. Graphene Oxide: Surface Activity and Two-Dimensional Assembly. Adv. Mater. 2010, 22, 1954− 1958. (13) Kim, J.; Cote, L. J.; Kim, F.; Yuan, W.; Shull, K. R.; Huang, J. X. Graphene Oxide Sheets at Interfaces. J. Am. Chem. Soc. 2010, 132, 8180−8186. (14) Cote, L. J.; Kim, J.; Tung, V. C.; Luo, J. Y.; Kim, F.; Huang, J. X. Graphene Oxide as Surfactant Sheets. Pure Appl. Chem. 2011, 83, 95− 110. (15) Chen, F. M.; Liu, S. B.; Shen, J. M.; Wei, L.; Liu, A. D.; ChanPark, M. B.; Chen, Y. Ethanol-Assisted Graphene Oxide-Based Thin Film Formation at Pentane-Water Interface. Langmuir 2011, 27, 9174−9181. (16) Wang, Y.; Li, Z. H.; Wang, J.; Li, J. H.; Lin, Y. H. Graphene and Graphene Oxide: Biofunctionalization and Applications in Biotechnology. Trends Biotechnol. 2011, 29, 205−212. (17) Feng, L. Z.; Liu, Z. A. Graphene in Biomedicine: Opportunities and Challenges. Nanomedicine 2011, 6, 317−324.
4. SUMMARY AND CONCLUSIONS Six GO suspensions were prepared by the sonication method. AFM results show their lateral size differs by more than 100 times with distinct size distributions. XPS and UV−vis results suggest their surface chemical properties are similar. Antibacterial activities of GO sheets toward E. coli cells are lateral size dependent. Larger GO sheets show stronger antibacterial activity than do smaller ones. They also display different timeand concentration-dependent antibacterial activity. Large GO sheets lead to most cell loss in 1 h incubation. In contrast, the inactivation rate of E. coli cells increases continuously when they are incubated with small GO sheets up to 4 h. The concentration of large GO sheets strongly influences their antibacterial activity at relative low GO concentration (20 μg/ mL). In contrast, the concentration of small GO sheets influences their antibacterial activity over all tested concentration. Our results suggest the size-dependent antibacterial activity of GO sheets comes from neither their different aggregation states, nor oxidation capacity, because GO sheets with different lateral sizes are all well dispersed, and their oxidation capacity toward GSH is similar. AFM analysis of GO sheets and cells suggests that GO sheets interact strongly with cells and easily cover cell surfaces. We propose that large GO sheets more easily cover cells, which may block their active sites on membranes. Cells cannot proliferate once fully covered, leading to the cell viability loss in the followed colony counting test. In contrast, small GO can adhere to the bacterial surfaces, which cannot effectively isolate cells from environment, resulting in weaker antibacterial activity. This study highlights the importance of tailoring the lateral dimension of GO sheets to optimize the application potential with minimal risks for environmental health and safety.
■
Article
ASSOCIATED CONTENT
S Supporting Information *
SEM images, XPS spectra, cell viability tests in saline solution, photographs of GO and cell suspensions, UV−vis absorption spectra of GO sheets, large field AFM images of cells, and cell viability after sonication treatments. This material is available free of charge via the Internet at http://pubs.acs.org. 12371
dx.doi.org/10.1021/la3023908 | Langmuir 2012, 28, 12364−12372
Langmuir
Article
(18) Kuila, T.; Bose, S.; Khanra, P.; Mishra, A. K.; Kim, N. H.; Lee, J. H. Recent Advances in Graphene-Based Biosensors. Biosens. Bioelectron. 2011, 26, 4637−4648. (19) Ratinac, K. R.; Yang, W. R.; Gooding, J. J.; Thordarson, P.; Braet, F. Graphene and Related Materials in Electrochemical Sensing. Electroanalysis 2011, 23, 803−826. (20) Akhavan, O.; Ghaderi, E. Toxicity of Graphene and Graphene Oxide Nanowalls Against Bacteria. ACS Nano 2010, 4, 5731−5736. (21) Liu, S.; Zeng, T. H.; Hofmann, M.; Burcombe, E.; Wei, J.; Jiang, R.; Kong, J.; Chen, Y. Antibacterial Activity of Graphite, Graphite Oxide, Graphene Oxide, and Reduced Graphene Oxide: Membrane and Oxidative Stress. ACS Nano 2011, 5, 6971−6980. (22) Hu, W. B.; Peng, C.; Luo, W. J.; Lv, M.; Li, X. M.; Li, D.; Huang, Q.; Fan, C. H. Graphene-Based Antibacterial Paper. ACS Nano 2010, 4, 4317−4323. (23) Bao, Q.; Zhang, D.; Qi, P. Synthesis and Characterization of Silver Nanoparticle and Graphene Oxide Nanosheet Composites as a Bactericidal Agent for Water Disinfection. J. Colloid Interface Sci. 2011, 360, 463−470. (24) Das, M. R.; Sarma, R. K.; Saikia, R.; Kale, V. S.; Shelke, M. V.; Sengupta, P. Synthesis of Silver Nanoparticles in an Aqueous Suspension of Graphene Oxide Sheets and Its Antimicrobial Activity. Colloids Surf., B 2011, 83, 16−22. (25) Zhang, D. H.; Liu, X. H.; Wang, X. Green Synthesis of Graphene Oxide Sheets Decorated by Silver Nanoprisms and Their Anti-Bacterial Properties. J. Inorg. Biochem. 2011, 105, 1181−1186. (26) Ma, J. Z.; Zhang, J. T.; Xiong, Z. G.; Yong, Y.; Zhao, X. S. Preparation, Characterization and Antibacterial Properties of SilverModified Graphene Oxide. J. Mater. Chem. 2011, 21, 3350−3352. (27) Xu, W. P.; Zhang, L. C.; Li, J. P.; Lu, Y.; Li, H. H.; Ma, Y. N.; Wang, W. D.; Yu, S. H. Facile Synthesis of Silver@Graphene Oxide Nanocomposites and Their Enhanced Antibacterial Properties. J. Mater. Chem. 2011, 21, 4593−4597. (28) Akhavan, O.; Ghaderi, E. Photocatalytic Reduction of Graphene Oxide Nanosheets on TiO(2) Thin Film for Photoinactivation of Bacteria in Solar Light Irradiation. J. Phys. Chem. C 2009, 113, 20214− 20220. (29) Sreeprasad, T. S.; Maliyekkal, M. S.; Deepti, K.; Chaudhari, K.; Xavier, P. L.; Pradeep, T. Transparent, Luminescent, Antibacterial and Patternable Film Forming Composites of Graphene Oxide/Reduced Graphene Oxide. ACS Appl. Mater. Interfaces 2011, 3, 2643−2654. (30) Santos, C. M.; Tria, M. C. R.; Vergara, R.; Ahmed, F.; Advincula, R. C.; Rodrigues, D. F. Antimicrobial Graphene Polymer (PVK-GO) Nanocomposite Films. Chem. Commun. 2011, 47, 8892−8894. (31) Park, S.; Mohanty, N.; Suk, J. W.; Nagaraja, A.; An, J. H.; Piner, R. D.; Cai, W. W.; Dreyer, D. R.; Berry, V.; Ruoff, R. S. Biocompatible, Robust Free-Standing Paper Composed of a TWEEN/Graphene Composite. Adv. Mater. 2010, 22, 1736−1740. (32) Cai, X.; Tan, S. Z.; Lin, M. S.; Xie, A.; Mai, W. J.; Zhang, X. J.; Lin, Z. D.; Wu, T.; Liu, Y. L. Synergistic Antibacterial Brilliant Blue/ Reduced Graphene Oxide/Quaternary Phosphonium Salt Composite with Excellent Water Solubility and Specific Targeting Capability. Langmuir 2011, 27, 7828−7835. (33) Vecitis, C. D.; Zodrow, K. R.; Kang, S.; Elimelech, M. Electronic-Structure-Dependent Bacterial Cytotoxicity of SingleWalled Carbon Nanotubes. ACS Nano 2010, 4, 5471−5479. (34) Yang, C. N.; Mamouni, J.; Tang, Y. A.; Yang, L. J. Antimicrobial Activity of Single-Walled Carbon Nanotubes: Length Effect. Langmuir 2010, 26, 16013−16019. (35) Kang, S.; Herzberg, M.; Rodrigues, D. F.; Elimelech, M. Antibacterial Effects of Carbon Nanotubes: Size Does Matter. Langmuir 2008, 24, 6409−6413. (36) Wang, X.; Jia, G.; Wang, H.; Nie, H.; Yan, L.; Deng, X. Y.; Wang, S. Diameter Effects on Cytotoxicity of Multi-Walled Carbon Nanotubes. J. Nanosci. Nanotechnol. 2009, 9, 3025−3033. (37) Kang, S.; Mauter, M. S.; Elimelech, M. Physicochemical Determinants of Multiwalled Carbon Nanotube Bacterial Cytotoxicity. Environ. Sci. Technol. 2008, 42, 7528−7534.
(38) Liu, S.; Ng, A. K.; Xu, R.; Wei, J.; Tan, C. M.; Yang, Y.; Chen, Y. Antibacterial Action of Dispersed Single-Walled Carbon Nanotubes on Escherichia coli and Bacillus subtilis investigated by atomic force microscopy. Nanoscale 2010, 2, 2744−2750. (39) Liu, S. B.; Wei, L.; Hao, L.; Fang, N.; Chang, M. W.; Xu, R.; Yang, Y. H.; Chen, Y. Sharper and Faster “Nano Darts” Kill More Bacteria: A Study of Antibacterial Activity of Individually Dispersed Pristine Single-Walled Carbon Nanotube. ACS Nano 2009, 3, 3891− 3902. (40) Arias, L. R.; Yang, L. J. Inactivation of Bacterial Pathogens by Carbon Nanotubes in Suspensions. Langmuir 2009, 25, 3003−3012. (41) Lyon, D. Y.; Brown, D. A.; Alvarez, P. J. J. Implications and Potential Applications of Bactericidal Fullerene Water Suspensions: Effect of nC(60) Concentration, Exposure Conditions and Shelf Life. Water Sci. Technol. 2008, 57, 1533−1538. (42) Sanchez, V. C.; Jachak, A.; Hurt, R. H.; Kane, A. B. Biological Interactions of Graphene-Family Nanomaterials: An Interdisciplinary Review. Chem. Res. Toxicol. 2012, 25, 15−34. (43) Akhavan, O.; Ghaderi, E.; Esfandiar, A. Wrapping Bacteria by Graphene Nanosheets for Isolation from Environment, Reactivation by Sonication, and Inactivation by Near-Infrared Irradiation. J. Phys. Chem. B 2011, 115, 6279−6288. (44) Ruiz, O. N.; Fernando, K. A. S.; Wang, B.; Brown, N. A.; Luo, P. G.; McNamara, N. D.; Vangsness, M.; Sun, Y.-P.; Bunker, C. E. Graphene Oxide: A Nonspecific Enhancer of Cellular Growth. ACS Nano 2011, 5, 8100−8107. (45) Hummers, W. S.; Offeman, R. E. Preparation of Graphitic Oxide. J. Am. Chem. Soc. 1958, 80, 1339−1339. (46) Kovtyukhova, N. I.; Ollivier, P. J.; Martin, B. R.; Mallouk, T. E.; Chizhik, S. A.; Buzaneva, E. V.; Gorchinskiy, A. D. Layer-By-Layer Assembly of Ultrathin Composite Films from Micron-Sized Graphite Oxide Sheets and Polycations. Chem. Mater. 1999, 11 (3), 771−778. (47) Mkhoyan, K. A.; Contryman, A. W.; Silcox, J.; Stewart, D. A.; Eda, G.; Mattevi, C.; Miller, S.; Chhowalla, M. Atomic and Electronic Structure of Graphene-Oxide. Nano Lett. 2009, 9, 1058−1063. (48) Dreyer, D. R.; Park, S.; Bielawski, C. W.; Ruoff, R. S. The Chemistry of Graphene Oxide. Chem. Soc. Rev. 2010, 39, 228−240. (49) Bunch, J. S.; Verbridge, S. S.; Alden, J. S.; van der Zande, A. M.; Parpia, J. M.; Craighead, H. G.; McEuen, P. L. Impermeable Atomic Membranes from Graphene Sheets. Nano Lett. 2008, 8, 2458−2462. (50) Shao, W.; Zhao, Q.; Abel, E. W.; Bendavid, A. Influence of Interaction Energy between Si-Doped Diamond-Like Carbon Films and Bacteria on Bacterial Adhesion under Flow Conditions. J. Biomed. Mater. Res., Part A 2010, 93, 133−139. (51) Teixeira, P.; Azeredo, J.; Oliveira, R.; Chibowski, E. Interfacial Interactions between Nitrifying Bacteria and Mineral Carriers in Aqueous Media Determined by Contact Angle Measurements and Thin Layer Wicking. Colloids Surf., B 1998, 12, 69−75. (52) Brown, D. M.; Kinloch, I. A.; Bangert, U.; Windle, A. H.; Walter, D. M.; Walker, G. S.; Scotchford, C. A.; Donaldson, K.; Stone, V. An in vitro Study of the Potential of Carbon Nanotubes and Nanofibres to Induce Inflammatory Mediators and Frustrated Phagocytosis. Carbon 2007, 45, 1743−1756. (53) Poland, C. A.; Duffin, R.; Kinloch, I.; Maynard, A.; Wallace, W. A. H.; Seaton, A.; Stone, V.; Brown, S.; MacNee, W.; Donaldson, K. Carbon Nanotubes Introduced into the Abdominal Cavity of Mice Show Asbestos-Like Pathogenicity in a Pilot Study. Nat. Nanotechnol. 2008, 3, 423−428. (54) Lin, S. C.; Shih, C. J.; Strano, M. S.; Blankschtein, D. Molecular Insights into the Surface Morphology, Layering Structure, and Aggregation Kinetics of Surfactant-Stabilized Graphene Dispersions. J. Am. Chem. Soc. 2011, 133, 12810−12823.
12372
dx.doi.org/10.1021/la3023908 | Langmuir 2012, 28, 12364−12372
