Graphene-Based Antibacterial Paper
Graphene-Based Antibacterial Paper
G
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
atoms that are tightly packed into ABSTRACT Graphene is a monolayer of tightly packed carbon atoms that possesses many interesting
a two-dimensional crystal.1 This ex- properties and has numerous exciting applications. In this work, we report the antibacterial activity of two water-
Downloaded via UNIV OF SUNDERLAND on September 27, 2018 at 09:09:59 (UTC).
tremely thin nanomaterial possesses very dispersible graphene derivatives, graphene oxide (GO) and reduced graphene oxide (rGO) nanosheets. Such
high mechanical stiffness2⫺4 and extraordi-
graphene-based nanomaterials can effectively inhibit the growth of E. coli bacteria while showing minimal
nary electronic transport properties.5⫺10
cytotoxicity. We have also demonstrated that macroscopic freestanding GO and rGO paper can be conveniently
Since the seminal work of Geim and co-
workers on freestanding graphene in 2004,5 fabricated from their suspension via simple vacuum filtration. Given the superior antibacterial effect of GO and
various forms of graphene sheets have the fact that GO can be mass-produced and easily processed to make freestanding and flexible paper with low cost,
been actively explored11 with novel applica- we expect this new carbon nanomaterial may find important environmental and clinical applications.
tions such as transistors,12⫺14 solar cells,15⫺17
and sensors.18,19 While biological studies of KEYWORDS: graphene oxide · reduced graphene oxide · minimal cytotoxicity ·
graphene are relatively limited, significant antibacterial activity · graphene oxide paper
recent attention has been drawn toward in-
teractions between graphene derivatives
nanoparticles,28 titanium oxide nanoparti-
and bioorganisms.20⫺22 Here, we interrogate
cles,29 and carbon nanotubes (CNTs). CNTs
interactions between graphene oxide (GO)
are another type of carbon nanomaterials
nanosheets and bacterial and mammalian
and can be regarded as rolled graphene.1
cells and report our novel finding on the ex-
CNTs have been found to be cytotoxic,30⫺32
cellent antibacterial activity and minimal
unless via elaborate surface functionaliza-
cytotoxicity of GO nanosheets. Particularly,
tion,33 to both human cells and bacteria.
we demonstrate that macroscopic antibac-
Very recently, graphene paper has been
terial graphene-based paper can be conve-
found to be a biocompatible substrate for
niently fabricated with superior inhibition
adhesion and proliferation of L-929 cells,20
ability to bacteria growth, suggesting the
neuroendocrine PC12 cells, oligodendroglia
promising environmentally friendly applica-
cells, and osteoblasts,34 and graphene ox-
tions of these low-cost and highly effective ide has been employed as an effective
carbon nanomaterials. nanocargo to deliver water-insoluble drugs
Antibacterial materials are widely used into cells.21,22 This has motivated us to ex-
in daily life and effectively protect the pub- plore the cytotoxicity of graphene and its
lic health. A wide range of materials, includ- antiseptic properties.
ing antibiotics,23 metal ions,24 and quater-
nary ammonium compounds,25 have been RESULTS AND DISCUSSION
*Address correspondence to
known to prevent attachment and prolifera- Preparation and Characterization of GO [email protected],
tion of microbes on material surfaces. How- Nanosheets and Paper. GO nanosheets are a [email protected].
ever, these materials are known to be asso- Fax: (ⴙ) 86-21-3919-4702.
chemically modified graphene with sus-
ciated with concerns about antibiotic pended hydroxyl, epoxyl, and carboxyl Received for review May 19, 2010
resistance, environmental pollution, rela- functional groups,8 which enable them to and accepted June 25, 2010.
tively complex processing, and high be readily dispersed in water. GO
Published online July 1, 2010.
cost.26,27 More recently, antibacterial proper- nanosheets were prepared according to a 10.1021/nn101097v
ties of nanomaterials have been explored modified Hummer method, resulting in a
to meet these challenges, including silver brown colloidal suspension.35 The thickness © 2010 American Chemical Society
Figure 1. Characterization of GO and rGO nanosheets and paper. AFM images of (a) GO and (b) rGO sheets. (c) Photographs
of freestanding and flexible GO (top) and rGO paper (bottom) (inset of (c), the photos of GO (top) and rGO (bottom) paper
penetrated by white light). (d) Thickness of GO (top) and rGO (bottom) paper as measured via SEM.
of the GO sheets was ⬃1.1 nm as measured via atomic paper via a one-step vacuum filtration protocol.36,37
force microscopy (AFM), suggesting the formation of a Such graphene materials in the paper form are easy to
single-layer 2-D nanomaterial (Figure 1a). Hydrazine re- use and have potential practical applications. The GO
duction of GO led to a black rGO suspension, a more paper had a thickness of ⬃1.5 m and that of the rGO
conductive version of GO nanosheets with less surface paper was ⬃4.6 m, as characterized by scanning elec-
defects. AFM measurements revealed that rGO had a re- tron microscopy (SEM). Interestingly, the GO paper
duced sheet thickness of ⬃1.0 nm (Figure 1b), which looked lackluster while the rGO paper was lustrous (Fig-
was possibly attributed to partial removal of oxygen ure 1c,d), which was possibly due to the marked differ-
functional groups on the surface of GO nanosheets dur- ence in electronic properties of GO and rGO
ing the reduction process. GO and rGO nanosheets nanosheets.
were fairly polydispersed with lateral dimensions rang- Cellular Uptake and Cytotoxicity of GO Nanosheets. We first
ing from nanometers to micrometers. evaluated the cellular uptake and cytotoxicity of GO
Both GO and rGO nanosheets could be easily made nanosheets (85 g/mL) with a mammalian cell line,
into macroscopic, freestanding, robust, and flexible A549. Transmission electron microscopy (TEM) studies
Figure 2. Cellular uptake and cytotoxicity of GO nanosheets. (a) TEM images of A549 cells incubated with 85 g/mL GO
nanosheets for 6 h. GO nanosheets were internalized in A549 cells (indicated by arrows). “pm” and “num” stand for plasma
membrane and nuclear membrane. The light-colored parts marked with an arrow were ascribed to single- or few-layer GO
sheets, while the black parts were GO aggregates. (b) Viability of cells incubated with 20 and 85 g/mL GO nanosheets for 2 h
and 24 h. (c) Distribution of A549 cells without GO nanosheet treatment (left) and treated with 20 g/mL (middle) and 85
g/mL (right) GO for 24 h.
demonstrated that GO nanosheets were inside the en- plied that the observed small decrease in cell viability
dosome of the cytoplasm (Figure 2a and Figure S2 (Sup- might arise from GO-retarded cell cycles and thus
porting Information)), suggesting that GO nanosheets slightly decreased proliferation rates, rather than from
could be internalized within A549 cells via endocyto- apoptosis or death of cells. Therefore, we concluded
sis.21 The metabolic activity assays based on succinate that GO nanosheets were relatively biocompatible
dehydrogenase activity in the mitochondria showed nanomaterials with mild cytotoxicity.
that GO nanosheets (20 g/mL) exhibited no cytotoxic- Antibacterial Activity of GO Nanosheets. We then evalu-
ity to A549 within 2 h incubation and a slight decrease ated the antibacterial activity of GO nanosheets by in-
in cell viability (⬃20%) within 24 h. GO nanosheets of vestigating the interaction the E. coli DH5␣ cells with
higher concentration (85 g/mL) led to an increased cy- GO nanosheets. The metabolic activity of E. coli DH5␣
totoxicity (⬃50%) within 24 h (Figure 2b). cells in the presence of GO nanosheets was measured
Flow cytometric analysis provided a mechanistic via a luciferase-based ATP assay kit. After 2 h incubation
study of the interaction of GO nanosheets with A549 with GO nanosheets of 20 g/mL at 37 °C, the cell meta-
cells. Interestingly, we found that apoptosis did not oc- bolic activity for E. coli deceased to ⬃70% and to ⬃13%
cur in A549 cells treated with GO nanosheets of both 20 at a GO nanosheet concentration of 85 g/mL (Figure
and 85 g/mL for 24 h (Figure 2c). Cell cycle analysis 3a), suggesting the strong inhibition ability of GO
showed that the percentage of cells in the G0/G1 phase nanosheets to E. coli. We also employed a classic colony
slightly increased (Table S1 (Supporting Information)) counting method to measure the microbial viability of
and those in the G2/M phase almost multiplied, while E. coli treated with 85 g/mL GO for 2 h. Significantly,
those in the S phase significantly decreased, which sug- we found that GO almost completely suppressed the
gested that GO-treated A549 cells were kept in the G2 growth of E. coli, leading to a viability loss up to 98.5%
phase (mitosis metaphase) rather than passing the (Figure 3b). A control study was performed with GO-
check point to replicate DNA. We further counted the free supernatant that removed GO nanosheets via cen-
numbers of A549 cells treated with GO nanosheets for trifugation, which did not induce the apparent bacterial
24 h. Interestingly, the number of untreated cells was suppression effect with overnight incubation at 37 °C
3.11-fold that of seeded cells (Table S1), while the num- (Figure S3 (Supporting Information)). These results fur-
ber of cells treated with GO nanosheets of 85 g/mL ther confirmed that GO nanosheets were responsible
proliferated 2.78-fold. Overall, these data strongly im- for the observed strong antibacterial effect. TEM
www.acsnano.org VOL. 4 ▪ NO. 7 ▪ 4317–4323 ▪ 2010 4319
ARTICLE
Figure 4. Antibacterial activity and cytotoxicity of rGO nanosheets. (a) Metabolic activity of E. coli treated with 85 g/mL
GO and rGO nanosheets, respectively. (b) Antibacterial activity of 85 g/mL GO and rGO nanosheets against E. coli. (c) TEM
image of E. coli exposed to 85 g/mL rGO nanosheets at 37 °C for 2 h. (d) Viability of A549 cell incubated with 20 and 85
g/mL rGO nanosheets, respectively.
studies revealed that E. coli largely lost cellular integrity, combines flexibility and mechanical stiffness (Figure
with the cell membrane being severely destroyed and the 1c,d).46,47 The antibacterial activity of GO and rGO paper
cytoplasm flowing out (Figure 3c,d). Such irreversible GO- was determined by using airborne bacteria tests. Aero-
induced cellular damage of E. coli might arise from the ef- solized, airborne bacteria were imitated to sow the E. coli
fects of either oxidative stress or physical disruption that DH5␣ cells on the paper, on which cold Luria⫺Bertani
have been observed in cellular effects of related carbon (LB) growth medium (with 1.5% agar) was then spread.
nanomaterial (CNTs and fullerene).38⫺44 After overnight incubation at 37 °C, we could not find any
Antibacterial Activity and Cytotoxicity of rGO Nanosheets. cell growth on the GO paper (Figure 5a) and only a lim-
The reduced form of GO nanosheets, rGO nanosheets, ited number of E. coli colonies on the rGO paper (Figure
also exhibited high antibacterial effects. The metabolic 5b), implying the superior antibacterial effect of such
activity of E. coli DH5␣ cells was reduced to ⬃24% on graphene-based papers. In contrast, control studies in
treatment with rGO nanosheets of 85 g/mL at 37 °C for the absence of either GO or rGO paper led to a great num-
2 h (Figure 4a), and colony counting showed that less ber of colony-forming units (CFU). SEM studies further
than 10% of E. coli survived (Figure 4b). TEM studies re- confirmed that E. coli cells on the paper lost the integ-
vealed that rGO nanosheets destroyed the membrane rity of membranes (Figure 5c,d), which was responsible
of E. coli in a way similar to that of GO nanosheets (Fig- for the bacteria-killing effect of the graphene-based
ure 4c). However, the cell viability of A549 was re- paper. It is worth noting that a very recent study
duced to ⬃47% and ⬃15% with rGO nanosheets of 20 showed that bacteria did not grow on freestanding pa-
and 85 g/mL, respectively (Figure 4d). Therefore, rGO per composed of a Tween/rGO composite,48 an effect
nanosheets possessed antibacterial properties that arising from the surfactant Tween-based prevention of
were only slightly lower than those of GO nanosheets, nonspecific binding (NSB) of bacteria rather than the
while their cytotoxicity was significantly higher than graphene-based bacterial killing as reported in this
GO’s. Such difference in cytotoxicity might arise from work.
different surface charges and functional groups of GO
and rGO nanosheet surfaces (Figure S1).42,45 CONCLUSIONS
Antibacterial Activity of GO and rGO Paper. Freestanding In summary, graphene-based nanomaterials have
graphene-based paper is a highly useful material that been found to be excellent antibacterial materials
4320 VOL. 4 ▪ NO. 7 ▪ HU ET AL. www.acsnano.org
ARTICLE
Figure 5. Antibacterial activity of GO and rGO paper. Photographs of E. coli growth on GO (a) and rGO (b) paper (overnight
incubation at 37 °C). SEM images of E. coli attached to GO (c) and rGO (d) paper (12 h incubation at 37 °C).
with mild cytotoxicity. We have also demonstrated nanosheets can be mass-produced and easily pro-
that macroscopic GO and rGO paper can be conve- cessed to make freestanding and flexible paper with
niently fabricated from their suspension via simple low cost, we expect this new carbon nanomaterial
vacuum filtration. Given the superior antibacterial ef- could offer new opportunities for the development
fect of GO nanosheets and the fact that GO of antibacterial materials.
MATERIALS AND METHODS Then cells were lysed with 20% acid sodium dodecyl sulfate
Preparation and Characterization of the Graphene Nanomaterials. GO (SDS, Sigma, USA) solution. After centrifugation, the absorbance
was prepared from purified natural graphite by the modified of the supernatants was measured at 570 nm by using a micro-
Hummers method,35 resulting in a colloidal suspension of GO plate reader (Bio-Rad 680, USA). Alternatively, treated cells were
sheets with a concentration of 0.85 mg/mL. The GO was reduced collected and counted with a blood counting chamber after be-
to rGO by using hydrazine hydrate.49 In a typical procedure, GO ing stained by trypan blue. In flow cytometry assays, treated cells
(0.85 mg/mL, 100 mL) was loaded in a 250 mL round-bottom were washed three times with ice-cold phosphate buffer (PBS)
flask, to which hydrazine hydrate (1 mL, 50 wt %) was added and and then stained with propidium iodide (PI) and FITC labeled An-
then heated in an oil bath at 80 °C under a water-cooled con- nexin V by using an Apoptosis Detection Kit (Bipec, USA). Each
denser for 24 h. Residual hydrazine hydrate was removed via di- experiment was independently performed at least three times.
alysis with distilled water. Bacterial Culture. E. coli DH5␣ (Clontech) cells were cultivated
Macroscopic GO paper and rGO paper were made by filtra- at 37 °C and maintained on LB plates (Luria⫺Bertani broth, Len-
tion of the suspension through a PVDF filter membrane (47 mm nox modification, with 1.5% agar). E. coli DH5␣ cells were grown
in diameter, 220 nm pore size) via vacuum at room tempera- overnight in LB medium at 37 °C and then harvested at the ex-
ture.46 Paper could be easily peeled off from the filter paper. The ponential growth phase via centrifugation. E. coli cells were
thickness of the paper was controlled by adjusting the volume washed twice to remove residual macromolecules and other
of the colloidal suspension. growth medium constituents and then resuspended in sterile sa-
The thickness of GO and rGO sheets was measured via line solution (0.9% NaCl). The E. coli cells were quantified via
atomic force microscopy (AFM), and the corresponding paper OD600 measurements.
was measured via scanning electron microscopy (SEM). Cell Viability Test of E. coli with GO and rGO Suspensions. E. coli DH5␣
Cell Culture and Cytotoxicity Test. A549 cells were grown in PRIM- cells were inoculated in saline solution containing 85 g/mL
1640 (Invitrogen, USA) with 10% heat-inactivated fetal bovine se- GO or rGO, with a final cell concentration of 107/mL. The mix-
rum and antibiotics (100 g/mL of streptomycin and 100 U/mL ture was incubated with gentle shaking for 2 h at 37 °C. The mix-
of penicillin) at 37 °C in 5% CO2. Cells were seeded in 6- or 24-well ture was diluted to 106-fold with a gradient method and then
plates and grown overnight prior to studies. Cells of ⬃80% con- applied uniformly on three LB culture medium plates per gradi-
fluent were incubated with fresh media containing GO or rGO ent solution (with a blank control). These sheets were incubated
suspensions with different concentrations (0, 20, and 85 g/mL). at 37 °C for 12 h. The colony-forming units (CFU) were counted,
After 24 h, 50 L of a 5 mg/mL thiazolyl blue tetrazolium bro- and the percentage of activated cells was determined from the
mide (MTT, Sigma-Aldrich, USA) solution was added to each well ratio of the number of cells in the mixture divided by the num-
of the 24-well plate, followed by incubation at 37 °C for 4 h. ber of cells at the beginning of experiments.