In Vitro Bacterial Cytotoxicity of CNTs: Reactive Oxygen Species

Dec 20, 2013 - A majority of the earlier reports attributed the bactericidal cytotoxicity of CNTs to bacterial cell .... Values are expressed as mean ...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Langmuir

In Vitro Bacterial Cytotoxicity of CNTs: Reactive Oxygen Species Mediate Cell Damage Edges over Direct Physical Puncturing Krishnamoorthy Rajavel,† Rajkumar Gomathi,‡ Sellamuthu Manian,‡,§ and Ramasamy Thangavelu Rajendra Kumar*,†,⊥ †

Advanced Materials and Devices Laboratory, Department of Physics, School of Physical Sciences and ‡Microbiology Laboratory, Department of Botany, School of Life Sciences, Bharathiar University, Coimbatore 641046, India § Research and Development Division, Manian Laboratories Pvt. Ltd., Peelamedu, Coimbatore 641004, India S Supporting Information *

ABSTRACT: Understanding the bacterial cytotoxicity of CNTs is important for a wide variety of applications in the biomedical, environmental, and health sectors. A majority of the earlier reports attributed the bactericidal cytotoxicity of CNTs to bacterial cell membrane damage by direct physical puncturing. Our results reveal that bacterial cell death via bacterial cell membrane damage is induced by reactive oxygen species (ROS) produced from CNTs and is not due to direct physical puncturing by CNTs. To understand the actual mechanism of bacterial killing, we elucidated the bacterial cytotoxicity of SWCNTs and MWCNTs against Gram-negative human pathogenic bacterial species Escherichia coli, Shigella sonnei, Klebsiella pneumoniae, and Pseudomonas aeruginosa and its amelioration upon functionalizing the CNTs with antioxidant tannic acid (TA). Interestingly, the bacterial cells treated with CNTs exhibited severe cell damage under laboratory (ambient) and sunlight irradiation conditions. However, CNTs showed no cytotoxicity to the bacterial cells when incubated in the dark. The quantitative assessments carried out by us made it explicit that CNTs are effective generators of ROS such as 1O2, O2•−, and • OH in an aqueous medium under both ambient and sunlight-irradiated conditions. Both naked and TA-functionalized CNTs showed negligible ROS production in the dark. Furthermore, strong correlations were obtained between ROS produced by CNTs and the bacterial cell mortality (with the correlation coefficient varying between 0.7618 and 0.9891) for all four tested pathogens. The absence of bactericidal cytotoxicity in both naked and functionalized CNTs in the dark reveals that the presence of ROS is the major factor responsible for the bactericidal action compared to direct physical puncturing. This understanding of the bactericidal activity of the irradiated CNTs, mediated through the generation of ROS, could be interesting for novel applications such as regulated ROS delivery in cancer therapy and the sanitation of potable water supplies.

I

pneumoniae leads to complications such as pneumonia, urinary infections, septicemia, and ankylosing spondylitis. Both of these pathogenic organisms inhabit water and are major sources of waterborne infection in humans. Generally, the bactericidal mechanism involves damage to the membrane integrity and its cellular components. Damage at any of these sites initiates a number of subsequent changes leading to bacterial cell death. The first report of SWCNTs exhibiting antibacterial activity, by developing direct contact and damaging the cell membrane, was made by Kang et al.7 Such a bacterial destruction process of CNTs was found to depend greatly on their surface morphologies and composition. Several structural parameters such as the size,10 length,11 chirality,12 and surface functional groups (−OH, −COOH, and −NH2)6 of SWCNTs and metal impurities13 were reported to govern the bacterial cytotoxicity. In particular, the length of

n recent years, carbon nanotubes (CNTs) have taken on a new dimension with a wide range of application in several biological systems. Their unique physical and surfacefunctionalization properties favor their utilization in the fabrication of biosensors,1 drug delivery,2 water purification,3 and the growth of mesenchymal stem cells.4 Owing to their large surface area and mesopore volume, CNTs are gaining a great deal of attention in the elimination of biological contaminants, particularly bacteria and viruses. These microbial contaminants include human pathogens such as Escherichia coli, Salmonella spp., Vibrio cholera, and rotavirus, whose mere presence impose a great threat to natural water habitats. Several researchers have reported CNTs exhibiting cytotoxicity over a wide range of microorganisms, including bacteria such as Escherichia coli,5 Salmonella typhimurium,6 Bacillus subtilis, Staphylococcus aureus, Micrococcus lysodikaticus,7 and Streptococcus mutans8 and viruses.9 However, the influence of CNTs on dreaded bacterial pathogens Shigella sonnei and Klebsiella pneumoniae have not been studied so far. S. sonnei causes bacillary dysentery, hemolytic anemia, and fever, and K. © 2013 American Chemical Society

Received: September 2, 2013 Revised: November 9, 2013 Published: December 20, 2013 592

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601

Langmuir

Article

CNTs influences cytotoxic effects, and the variation of their activity as a function of varying length has been reported.12 Besides, the chemical functionalization is also found to influence the cytotoxic properties mediated by CNTs in biosystems.14 However, the exact mechanism of CNT-mediated bacterial cytotoxicity has not been clearly understood so far. Several reports claim that the bactericidal action of CNTs is through the direct physical puncturing10,15 and/or physical contact with the bacterial cell surface;11,12,16 however, extensive investigations on the reactive oxygen species (ROS) production of these bactericidal CNTs is lacking. Therefore, a detailed investigation of the ROS-producing capabilities of CNTs will shed light on the mechanism behind the bacterial cytotoxicity of CNTs. The present investigation is aimed at evaluating CNTmediated bactericidal activity in Gram-negative pathogenic microorganisms such as Escherichia coli, Shigella sonnei, Klebsiella pneumoniae, and Pseudomonas aeruginosa. The bacterial cell cytotoxicity activity was inferred from the bacterial growth measurements upon treatment with single-walled carbon nanotubes (SWCNTs) (0.4−1.2 μm long), multiwalled carbon nanotubes (MWCNTs) (0.5−1.5 μm long), and antioxidant TA-functionalized SWCNTs and MWCNTs. Interestingly, CNT (SWCNTs and MWCNTs) treatment under ambient light (laboratory condition) exhibited severe bactericidal activity whereas TA-functionalized CNTs showed no cytotoxicity to bacterial cells. However, when incubated in the dark, neither CNTs nor TA-functionalized CNTs showed any cytotoxicity. Furthermore, we have established through various quantitative experiments that naked CNTs are effective generators of ROS under irradiation conditions and showed negligible ROS production in the dark whereas TA-functionalized CNTs failed to generate ROS under both light and dark conditions. Hence, the observed absence of antibacterial activity by CNTs in the dark or TA functionalization revealed the fact that ROS governs the major factor responsible for bacterial cell death.

Table 1. Characterization of Carbon Nanotubes Used in the Study

a

parameter

SWCNTs

MWCNTs

methods metal impurities length diameter aspect ratio end G/D ratio

HiPCO nill 0.4−1.2 μm 0.9−1.2 nm 400−1000

pyrolysis nill 0.5−1.5 μm 10−50 nm 50−30 closed 0.855

a

0.824

No data.

software, see supportive information Figure S3). The EDAX spectrum of MWCNTs also registered the same pattern as observed for SWCNTs indicating the absence of metal impurities. Further, the TEM observation showed that the majority of the MWCNTs used in the present study were close ended (Figure 1 (d)). Bacterial Cytotoxicity of CNTs. The cytotoxicity of CNTs against four pathogenic microorganisms (E. coli, S. sonnei, K. pneumoniae and P. aeruginosa) was evaluated using optical density based growth curves (Material and Methods) and the number of colony forming units (cfu) over spread plates of agar media. The number of viable bacterial cells following CNT treatment was assessed employing a minimal medium containing only the essential nutrients for growth and a nutrient rich medium containing a complex of nutrients. As can be seen from Figure 2 (a-d), the growth curves indicated a remarkable cytotoxic activity of CNTs (90 μg/mL) against all the tested bacterial pathogens maintained in the minimal medium. In addition, the cell death upon treatment with SWCNTs and MWCNTs is also evident from the spread plates in terms of reduced number of colony forming units (cfu/ml) of the test bacteria (Table 2). The assessment of bacterial cytotoxicity activity of CNTs on nutrient rich medium also followed the same pattern as observed in the minimal medium (Supportive information, Figure S4 a-d and Table ST1). These results clearly establish the fact that the bacterial cytotoxicity of CNTs is nonspecies specific and unaffected by the nutrient richness of the growth medium. Bacterial cytotoxicity obtained for 90 μg/mL concentration of SWCNTs and MWCNTs in the present study is well in agreement with Arias6 and Yang et al.,11 reported for 100 μg/mL concentration of SWCNTs and MWCNTs respectively. The dimensions of SWCNTs 6 (diameter 1−1.5 nm, and length 1−5 μm) and MWCNTs11 (diameter 15−30 nm, and length of 1−5 μm) are fairly matches with the dimensions of CNTs used in the present study (Table 1). Generally, CNTs are found to act as ROS generators20 via a series of photoinduced chemical reactions in aqueous media and produce large amounts of singlet oxygen (1O2), superoxide anions (O2•−), and hydroxyl radicals (•OH) under visible light.21,22 To investigate the possible mechanism behind the bacterial cytotoxicity further, the CNTs were functionalized (Materials and Methods) using an antioxidant TA (characterization details in Supporting Information S5). An experiment was performed where the bacterial growth medium was supplemented with antioxidant (TA)-treated CNTs instead of CNTs. It is interesting from the optical-density-based growth curves that the functionalization of CNTs (SWCNTs and MWCNTs) with antioxidants ameliorated the bacterial growth inhibition in all of the tested bacterial species (E. coli, S. sonnei,

■

RESULTS AND DISCUSSION The nature of SWCNTs and MWCNTs (Figure S1a,b, Supporting Information) was confirmed by analyzing the G, D, and G| band vibrations of the Raman spectra. The diameters of SWCNTs have been determined from their radial breathing modes (RBM) using the expression ω(RBM)= A/dt + B, where A and B are constants. From the values A = 234 cm−1 and B = 10 cm−1 for the bundles of SWCNTs, the calculated diameters (dt) were in the range of 1.271 to 0.9512 nm.17 From the Raman spectrum, the obtained sample of SWCNTs was found to contain both semiconducting and metallic nanotubes that were indeed verified by previous reports.18,19 The details of the material characterization of CNTs used in this study are presented in Table 1. The TEM images of bundles of SWCNTs and MWCNTs are shown in Figure 1 (a-d). From the Figure 1(a), certain traces of amorphous carbon were seen on the SWCNTs. The length and diameter distribution of SWCNTs from the TEM images obtained were found to be 0.4 to 1.2 μm and 0.9 to 1.2 nm, respectively and the EDAX analysis confirmed the absence of metal impurities. The length distribution and EDAX spectrum of SWCNTs are given in supportive information (supportive information Figure S2). The length of MWCNTs was found to be 0.5 − 1.5 μm and their diameter were found to vary between 10 and 50 nm (calculated from TEM image using Image 1.32 J 593

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601

Langmuir

Article

Figure 1. Representative high resolution TEM images of CNTs (0.1 mg/mL) (a) bundles of SWCNTs, (b) individual SWCNT indicated using arrow marks, (c) different diameter distribution of purified MWCNTs, and (d) multiple layered close ended MWCNTs.

generation of 1O2 in SWCNT and MWCNT suspensions (Figure 3a,d) is indicated by the depletion of furfuryl alcohol (FFA) contents that was proportional to the period of exposure of CNT suspensions to direct sunlight or ambient light under laboratory conditions. In the case of antioxidant (TA)functionalized CNT suspensions, however, the production of 1 O2 was suppressed significantly (p < 0.05). As for the production of O2•−, the formation of formazan, a reduction product of nitrobluetetrazolium (NBT), was monitored in CNT suspensions under irradiation conditions. The antioxidant-functionalized CNTs resulted in a significantly (p < 0.05) lower production of the same (Figure 3b,e). The decay of pchlorobenzoic acid (pCBA) under irradiation conditions (solar and ambient light under laboratory conditions) by CNT suspensions suggests the production of •OH, which was also remarkably suppressed in the suspension of the antioxidant (TA)-functionalized CNTs (Figure 3c,f). The overall production of ROS was found to be higher under direct solar irradiation than under laboratory conditions. Interestingly, the antioxidant-functionalized CNTs registered a significant reduction of ROS generation under both direct and ambient light conditions. CNT could act as photosensitizing species for the production of ROS. The functionalization of CNTs with antioxidants has resulted in a remarkable suppression of oxyradical production. It is noteworthy that the amount of •OH produced as recorded by the untreated CNTs was higher than the amounts of 1O2 and O2•−. The cyclic conversion of O2•− to •OH is also well known.25 It is presumed that the genesis of •OH in the

K. pneumoniae, and P. aeruginosa) (Figure 2a−d). Besides, surface functional groups tend to alter the antibacterial activity of CNTs through aggregation and dipersity14 but for our TAfunctionalized CNTs (SWCNTs and MWCNTs) no sign of aggregation is displayed when compared to naked CNTs. To assess the individual impact of antioxidant TA on the growth of bacteria, the culture medium was supplemented with TA. Even though good growth was observed, the supplementation of this natural antioxidant rendered a partially acidic environment (change in pH from 7 to 5) in the growth medium, thereby slightly depressing the growth of the bacteria. The decrease in the growth of bacteria in TA-supplemented media also may be due to the depression of the substrates required for normal microbial growth, ion deprivation, or the direct action on microbial metabolism through the inhibition of oxidative phosphorylation.23 However, a rapid multiplication of bacterial cells was observed in the control without any CNT or TA supplementation. We made a further attempt to quantify the free radical production (ROS) by SWCNTs and MWCNTs in an aqueous medium irradiated under solar (903 lm/m2) as well as ambient light conditions (laboratory conditions, 180 lm/m2) (Materials and Methods). The production of ROS such as singlet oxygen, superoxide, and hydroxyl radicals under a different time interval of exposure under ambient room light and sunlight is presented in Figure 3a−c,d−f, respectively. Because antioxidants are found to be effective scavengers of ROS,24 the antioxidant (TA)-functionalized CNTs were also investigated for the production of ROS species (1O2, O2•−, and •OH). The 594

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601

Langmuir

Article

Figure 2. Growth curve measurements of (a) E. coli, (b) S. sonnei, (c) K. pneumoniae and (d) P. aeruginosa in minimal broth showing antibacterial effects of SWCNTs/MWCNTs ((90 μg/mL) and their amelioration on the antioxidant (TA) coated CNTs (90 μg/mL). The cell density of various bacterial species used in the said treatments ranged between 1 and 1.6 × 107 cfu/mL.

Table 2. Effect of CNT Treatment on the Bacterial Cell Viability Assessed on the Minimal Agar Mediuma bacteria population (cfu/mL)b treatment

E. coli

S. sonnei

K. pneumoniae

P. aeruginosa

control SWCNTs (90 μg/mL) MWCNTs (90 μg/mL)

1.7 × 108 ± 1.5 × 107 c 1.2 × 104 ± 1 × 103 a 1.7 × 104 ± 1.5 × 103 b

1.5 × 108 ± 2 × 107 c 0.17 × 104 ± 0.2 × 103 b 0.13 × 104 ± 0.2 × 103 a

1.0 × 108 ± 2 × 107 c 1.0 × 104 ± 2 × 103 a 1.6 × 104 ± 0.6 × 103 b

1.8 × 108 ± 2.5 × 107 c 1.9 × 104 ± 2 × 103 a 2.0 × 104 ± 1 × 103 b

a Values are expressed as mean ± standard deviation (n = 3). Mean values followed by different superscript letters in a column indicate a significant statistical difference (p < 0.05), where a < b < c indicates the number of colony-forming units. bObserved 24 h after plating on a nutrient agar medium.

illuminated aqueous suspension system of CNTs has contributed to the bacterial cytotoxic activity because •OH is considered to be the most deleterious ROS in biological systems.24 Among the ROS species, the superoxide anion (O2•−), though harmful to cellular components,21 cannot directly initiate lipid oxidation.26 However, it is a potential precursor of highly reactive species such as the hydroxyl radical (•OH) and peroxynitrite.27 Hydroxyl radicals are considered to be one of the quick initiators of the lipid peroxidation process, which involves an instant reaction with the polyunsaturated fatty acids, proteins, and sugars in biological material.28

It is evident from the above-mentioned investigation (Figure 3) that the antimicrobial activity of CNTs is mediated through the formation of ROS. A strong linear correlation was obtained between the ROS generated by untreated/TA-functionalized CNTs (Figure 3) and the bacterial cytotoxicity as observed from the growth curves (Figure 2). As can be seen from Table 3, their R2 values varied between 0.7618 and 0.9687, indicating that ROS production is directly responsible for the bactericidal activity (Supporting Information Figures S6−S8). There was also an observed strong correlation between the bactericidal activity and the production of ROS by metals/metal oxide 595

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601

Langmuir

Article

Figure 3. Production of ROS at different exposure times under direct sunlight (903 lm/m2) and ambient light under laboratory conditions (180 lm/ m2) for CNT suspensions (100 μg/mL): (a, d) for FFA (0.2 M) reduction, (b, e) for NBT (0.2 mM) reduction, and (c, f) for pCBA (2 μM) reduction.

correlation between ROS generation and bacterial cell viability was obtained, certain questions remain unanswered. (i) Is ROS the only factor responsible for bacterial cell death? (ii) Do both the ROS and direct physical puncturing cause cell death? To answer these questions, CNT (SWCNTs, MWCNTs, TAfunctionalized SWCNTs, and MWCNTs) samples were evaluated for their ROS production and antibacterial activity in the dark. It is evident from Figure 4 that in the dark all samples (naked and TA-functionalized SWCNTs and MWCNTs) exhibited similar behavior and the ROS production of the samples was completely suppressed. Furthermore, from the optical-density-based growth curves for K. pnuemoniae and S. sonnei incubated in the dark (experimental procedure in Material and Methods), surprisingly, CNTs showed no cytotoxicity for the bacterial cells in the dark and the bacterial

Table 3. Correlation Analysis of Bacterial Cell Viability versus ROS Production of Untreated/TA-Functionalized CNTs under Ambient Light (Laboratory Condition) R2 value singlet oxygen radical superoxide radical hydroxyl radical

E. coli

S. sonnei

K. pnuemoniae

P. aeruginosa

0.9687 0.837 0.7618

0.8899 0.8783 0.9512

0.9096 0.9733 0.9419

0.8711 0.9891 0.9457

nanoparticles under UV light illumination in a previous study.29,30 A clear understanding of the factors governing the CNTinduced bacterial cell death is mandatory for a wide range of applications. In the present study, even though a strong

Figure 4. Production of ROS at different exposure times for CNTs/TA-functionalized CNTs (100 μg/mL) in the dark as assessed by (a) FFA (0.2 M) reduction, (b) NBT (0.2 mM) reduction, and (c) pCBA (2 μM) reduction. 596

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601

Langmuir

Article

Figure 5. Optical-density-based growth curves under dark incubation. (a) S. sonnei and (b) K. pneumoniae showing normal growth in the presence of 90 μg/mL CNTs/TA-functionalized CNTs. The cell densities of the bacterial species under this treatment were ∼1.5 × 107 and ∼1.2 × 107 cfu/mL, respectively, for S. sonnei and K. pneumoniae.

Figure 6. Quantitative assessment of lipid peroxidation on bacterial cells treated with both CNTs/TA-functionalized CNTs (90 μg/mL). (a) S. sonnei and (b) K. pneumoniae tested under dark and ambient light (laboratory condition) incubation. The cell densities of the bacterial species in this treatment were ∼1.5 × 107 and ∼1.2 × 107 cfu/mL for S. sonnei and K. pneumoniae, respectively.

Figure 7. Release of cellular contents from the bacterial cells upon incubation in 90 μg/mL CNTs (SWCNTs and MWCNTs) and TAfunctionalized CNTs (SWCNTs and MWCNTs) under dark and ambient light (laboratory condition) exposure of (a) S. sonnei and (b) K. pneumoniae at cell densities of ∼1.5 × 107 and ∼1.2 × 107 cfu/mL, respectively. *p < 0.05 is a significant difference compared to the control.

species registered normal growth for both naked and TAfunctionalized SWCNTs and MWCNTs (Figure 5a,b). Our results are very interesting because CNTs employed in the present investigation were cytotoxic under sunlight/ambient light conditions and noncytotoxic in the dark. Such an absence

of antibacterial activity in both naked and functionalized CNTs in the dark made it clear that the production of ROS is responsible for bacterial cell death and not direct physical puncturing. However, for further analysis, we assessed (i) lipid peroxidation (Figure 6) and (ii) DNA and RNA release (Figure 597

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601

Langmuir

Article

Figure 8. Schematic representation of oxidative stress-mediated cell death and its amelioration by antioxidant treatment. (A) Oxidative stress (ROS)mediated bacterial cell death. (B) Protection of bacterial cells against oxidative stress by an antioxidant.

increase in the optical density values was observed for CNTs exposed to ambient light under laboratory conditions. These results clearly show that the bacterial cell death in CNT-treated cultures exposed to light is due to ROS-induced, peroxidationmediated cell wall damage, resulting in the leakage of nucleic acids (DNA and RNA) and other cell contents. The functionalization of CNTs with antioxidant TA that curtailed ROS generation, however, protected the cells from peroxidation and subsequent death. Previous studies10,11,16 report that the cytotoxicity of CNTs to bacteria is due to cell membrane damage by direct contact. The antibacterial effect is further related to the expression of a higher level of stress-related gene products in CNT-treated cells.10 We observed the oxidative stress (ROS)-mediated lipid peroxidation (Figure 6) and release of cellular contents (Figure 7) in the test bacteria in the aqueous medium upon treatment with naked CNTs under light. However, the CNTs failed to produce the lethal effect in the target bacteria under dark incubation (Figure 5) in the absence of ROS generation (Figure 4). As such, it clear that the exposure of CNTs to light is crucial to their being lethal rather than their direct contact with the bacterial cells. Interestingly, the CNTs lose their cytotoxic potential when functionalized with antioxidant TA. The recorded protective effect of the antioxidant (TA) functionalization of CNTs (Figure 2, Table 2) could be attributed to the prevention/scavenging of ROS production by TA (Figure 3). Furthermore, TA functionalization with CNTs likely to reduce the interaction probability of CNTs and bacterial cells. On the basis of the aforementioned findings, we propose a three-stepped mechanism of cytotoxicity as follows: (i) Generation of ROS from the surface of CNTs, (ii) oxidative stress on the cell surface, and (iii) membrane damage and ultimate cell death (Figure 8). The apparent cell damage may

7) in the bacterial cells incubated in the presence of TAuntreated and TA-treated CNTs in the dark. These experiments also yielded results similar to the bacterial cytotoxicity tested in the dark because there was no evidence of significant bacterial cell damage owing to lipid peroxidation and/or DNA and RNA release when exposed in the dark. Lipid peroxidation is the major degenerative process affecting the cell membrane and other lipid-containing structures under the conditions of oxidative stress. Because the bactericidal mechanism involves the loss of membrane integrity, the extent of lipid peroxidation in the cells of two bacterial pathogens (S. sonnei and K. pneumoniae) exposed to CNTs was evaluated using the TBA (thiobarbituric acid) test (Material and Methods). Figure 6a,b depicts the formation of malondialdehyde, a product of lipid peroxidation, in the samples of bacterial cultures treated with both naked CNTs and TA-functionalized CNTs (SWCNTs and MWCNTs). CNT-treated samples exhibited a higher rate of lipid peroxidation than the samples treated with TA-functionalized CNTs when exposed to ambient light. However, no significant production of malondialdehyde was observed for all of the tested samples treated with CNTs and TA-functionalized CNTs under dark incubation. Therefore, the ROS-mediated oxidative stress results in lipid peroxidation on the bacterial cell surface, which might eventually lead to the efflux of cell contents (Figure 7a,b) and cell death. Figure 7 shows the optical density values assessed for the leakage of cellular contents such as DNA and RNA at an absorption maximum of 260 nm12 (Materials and Methods). Under dark incubation, observable variations in the optical density values were seen in all of the tested samples (90 μg/mL of a CNT suspension of S. sonnei at ∼1.5 × 107 cfu/mL and K. pneumoniae at ∼1.2 × 107 cfu/mL); however, a remarkable 598

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601

Langmuir

Article

distilled water were added to 200 μL of the desired CNT suspension in tubes. The bacterial cell suspension added to 200 μL of distilled water served as the control. All of the tubes were vortex mixed for 30 min and allowed to stand at room temperature for another 30 min, for a total of 1 h of exposure to CNTs prior to inoculation of the nutrient broth.6 The antibacterial activity of CNTs was evaluated in minimal as well as in nutrient broths adjusted to pH 7.0. The minimal broth had 0.5% dextrose, 0.3% K2HPO4, 0.3% KH2PO4, 0.2% NH4Cl, 0.005% MgSO4, and 0.005% FeSO4, whereas the nutrient broth contained a 0.5% peptic digest of animal tissues, 0.5% NaCl, 0.15% beef extract, and 0.15% yeast extract. For the evaluation of the antibacterial activity of CNTs in the minimal medium, the microbial inocula were grown in the same medium prior to treatment with CNTs. After 1 h of exposure to CNT as mentioned earlier, the CNT−bacterium mixture was aseptically transferred to 50 mL of minimal/nutrient broth and incubated at 37 °C under constant agitation (175 rpm) in a rotatory shaker. At 1 h intervals, 2.5 mL of the culture broth was withdrawn and the optical density (OD) was measured at 600 nm using a spectrophotometer (Shimadzu, UV-1601, Germany). Cell growth was monitored until the end of the exponential growth phase, and growth curves were obtained by plotting OD values against growth time. The antibacterial activity of CNTs was also determined by dilution plating of the above cultures where 0.1 mL of the diluted (1:10) culture was spread plated onto the minimal/nutrient agar plates. Colonies were counted after 24 h of incubation at room temperature and expressed as the number of viable cells (colonyforming units) per milliliter (cfu/mL). Amelioration of Bactericidal Activity of CNTs by Functionalization with Antioxidant. CNT−antioxidant suspensions were prepared as reported earlier.31 Briefly, CNTs dispersed in distilled water (1 mg/mL) were added to antioxidant TA at a rate of 1 mg/mL and kept in a shaker for 3 days. The mixtures were then sonicated for 1 h, centrifuged at 3000 rpm for 15 min, and passed through a PTFE (poly(tetrafluoroethylene)) membrane filter to remove unadsorbed excess antioxidants. The supernatants containing antioxidant-adsorbed CNTs were subjected to FT-IR characterization and used for further experiments. The antibacterial activity of the antioxidant-adsorbed CNTs was assessed in the minimal broth. The CNTs not treated with any antioxidant served as the control. Quantification of Reactive Oxygen Species. The ability of CNTs to generate ROS by a photoinduced reaction in solution was measured as described by Chen and Jafvert.22 The production of singlet oxygen (1O2) and the hydroxyl radical (•OH) was monitored by measuring the loss of, respectively, furfuryl alcohol (FFA, 0.2 M) and p-chlorobenzoic acid (pCBA, 2 μM) in a 100 μL CNT suspension incubated under direct sunlight (903 lm/m2) and ambient light (180.6 lm/m2) under laboratory conditions and in the dark. The light intensity was measured using a TES 1332 digital lux meter. The superoxide radicals (O2•−) that formed were determined using 100 μL of CNTs in 50 mL of a nitroblue tetrazolium salt solution (NBT, 0.2 mM) exposed to the above-mentioned light/dark conditions. In control tubes, the CNT suspension was replaced with distilled water. The samples were withdrawn at different time intervals (0.1, 0.3, 1, 2, 3, and 4 h) and passed through a 0.2 μm poly(tetrafluoroethylene) membrane filter, and the OD values were determined at 217, 237, and 530 nm for FFA, pCBA, and NBT, respectively. Assessment of CNT-Induced Lipid Peroxidation. The extent of lipid peroxidation induced by CNTs was evaluated using the thiobarbituric acid (TBA) test as described by Ohkawa et al.32,33 with slight modifications. Briefly, 0.5 mL of a 10% (w/v) bacterial cell suspension (S. sonnei and K. pneumoniae) was added to a 100 μg/mL CNT suspension, made up to 1 mL with distilled water, and incubated for 30 min at room temperature under dark and laboratory conditions. The control received 100 μL of distilled water in place of the CNT suspension. Then, 1.5 mL of 20% acetic acid (pH 3.5), 1.5 mL of 0.8% TBA (in 1.1% SDS), and 0.05 mL of 20% TCA were added in sequence, and the tubes were vortex mixed to ensure appropriate mixing. After incubation for 60 min at 100 °C in a water bath, the mixture was cooled to room temperature, added to 5 mL of butanol,

lead to the leakage of cytoplasmic contents including nucleic acid and cell death.27 Understanding the factors governing CNT-mediated cell death is important for wide range of applications. Previous reports10−12,15 claimed that bacterial cell death was due to the direct physical puncturing by CNTs without studying the ROS generation. This study differs from previous reports because we showed that ROS is responsible for bacterial cell death and not direct physical puncturing. Our results indicate that ambient light could be a critical factor influencing the bactericidal properties of CNTs in the case of ROS-induced cell damage. The fact that CNTs, in spite of direct contact with the bacterial cells, remain nontoxic in the dark but capable of generating ROS upon irradiation could be interesting for novel applications such as response to stimuli and local delivery of ROS, for instance, in the targeted killing of tumor cells by external photoactivation.

■

CONCLUSIONS This study demonstrated that CNTs could act as effective oxidants. Both SWCNTs and MWCNTs exhibited excellent cytotoxicity under ambient light/sunlight against pathogenic bacterial species owing to ROS production. The observed cytotoxicity was indeed light-dependent because it was nontoxic under dark incubation. A strong correlation was also found between the bacterial cytotoxicity and the ROS generation by CNTs exposed to irradiation. Our results illustrate that the production of ROS is the major factor responsible for bacterial cell death and not direct physical puncturing of CNTs. Our study brings out the importance of light as one of the decisive factors in the lethality of CNTs in both in vivo and in vitro applications. It could also shed more light on the novel applications of CNTs such as response to stimuli, antibacterial control systems, membrane filters, and the targeted destruction of tumor cells.

■

MATERIALS AND METHODS

Material and Chemicals. SWCNTs and MWCNTs were procured from Nano Lab (Waltham, MA). SWCNTs were prepared from a high-pressure carbon monoxide (HiPCO) method, and MWCNTs were synthesized by pyrolysis methods. Dextrose, MgSO4, K2HPO4, KH2PO4, NH4Cl, FeSO4, sodium dodecyl sulfate (SDS), nutrient broth, agar, and analytical-grade butanol were obtained from Himedia, India. Tannic acid (TA), H2O2, acetic acid, thiobarbituric acid (TBA), and trichloroacetic acid (TCA) were purchased from Sigma-Aldrich. TEM Characterization. A purified long-length carbon nanotube network was deposited on the carbon-coated copper grid for TEM images (JEOL 2100). CNT samples (0.1 mg) were dispersed in 1 mL of ethanol by bath sonication for 15 min. The obtained CNT solution was used to coat a carbon-coated copper grid and then allowed to dry on a hot plate. A CNT-coated copper grid was mounted on the TEM substrates for structural evaluation. Antibacterial Activity of a CNT Suspension. A CNT suspension of 1 mg/mL was prepared in distilled water by sonication for 1 h at room temperature using an ultrasonic bath (Rivotek, Mumbai, India). CNT suspensions were stable after 30 min of sonication. Bacterial cultures of E. coli, S. sonnei, K. pneumoniae, and P. aeruginosa (Microbiological Laboratory, India) were grown in nutrient broth (NB) at 37 °C for 12 h with constant agitation at 175 rpm. Cells were harvested by centrifuging the culture broth at 6000 rpm for 5 min. The pellet was washed three times and resuspended in 5 mL of sterile distilled water. A working concentration of 107 colony-forming units (cfu)/mL was prepared by suitably diluting the bacterial suspensions. Two milliliter aliquots of a bacterial cell suspension in 599

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601

Langmuir

Article

and centrifuged at 4000 rpm for 15 min. The pink chromogen, indicating the formation of free malondialdehyde in the TBA reaction, was spectroscopically measured at 532 nm. The % of lipid peroxidation for dark and light measurements was calculated using formulas 1 and 2, respectively.

⎥ ⎢ sample OD production(%) = ⎢ × 100⎥ ⎦ ⎣ control OD

(1)

⎥ ⎢ control OD − sample OD production(%) = ⎢ × 100⎥ ⎦ ⎣ control OD

(2)

(3) Upadhyayula, V. K. K.; Deng, S.; Mitchell, M. C.; Smith, G. B. Application of Carbon Nanotube Technology for Removal of Contaminants in Drinking Water: A Review. Sci. Total Environ. 2009, 408, 1−13. (4) Mooney, E.; Dockery, P.; Greiser, U.; Murphy, M.; Barron, V. Carbon Nanotubes and Mesenchymal Stem Cells: Biocompatibility, Proliferation and Differentiation. Nano Lett. 2008, 8, 2137−2143. (5) Kang, S.; Pinault, M.; Pfefferle, L. D.; Elimelech, M. SingleWalled Carbon Nanotubes Exhibit Strong Antimicrobial Activity. Langmuir 2007, 23, 8670−8673. (6) Arias, L. R.; Yang, L. Inactivation of Bacterial Pathogens by Carbon Nanotubes in Suspensions. Langmuir 2009, 25, 3003−3012. (7) Nepal, D.; Balasubramanian, S.; Simonian, A. L.; Davis, V. A. Strong Antimicrobial Coatings: Single-Walled Carbon Nanotubes Armored with Biopolymers. Nano Lett. 2008, 8, 1896−1901. (8) Akasaka, T.; Watari, F. Capture of Bacteria by Flexible Carbon Nanotubes. Acta Biomater. 2009, 5, 607−612. (9) Brady-Estévez, A. S.; Kang, S.; Elimelech, M. A Single-WalledCarbon-Nanotube Filter for Removal of Viral and Bacterial Pathogens. Small 2008, 4, 481−484. (10) Kang, S.; Herzberg, M.; Rodrigues, D. F.; Elimelech, M. Antibacterial Effects of Carbon Nanotubes: Size Does Matter! Langmuir 2008, 24, 6409−6413. (11) Yang, C.; Mamouni, J.; Tang, Y.; Yang, L. Antibacterial Activity of Single-Walled Carbon Nanotubes: Length Effect. Langmuir 2010, 26, 16013−16019. (12) Liu, S.; Wei, L.; Hao, L.; Fang, N.; Chang, M. W.; Xu, R.; Yang, Y.; 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. (13) Kang, S.; Mauter, M. S.; Elimelech, M. Physicochemical Determinants of Multiwalled Carbon Nanotube Bacterial Cytotoxicity. Environ. Sci. Technol. 2008, 42, 7528−7534. (14) Pasquini, L. M.; Hashmi., S. M.; Sommer, T. J.; Elimelech, M.; Zimmerman, J. B. Impact of Surface Functionalization on Bacterial Cytotoxicity of Single-Walled Carbon Nanotubes. Environ. Sci. Technol. 2012, 46, 6297−6305. (15) Lyon, D. Y.; Brunet, L.; Hinkal, G. W.; Wiesner, M. R.; Alvarez, P. J. Antibacterial Activity of Fullerene Water Suspensions (nC60) is Not Due to ROS-Mediated Damage. Nano Lett. 2008, 8, 1539−1543. (16) 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. (17) Jorio, A.; Pimenta, M. A.; Souza Filho, A. G.; Saito, R.; Dresselhaus, G.; Dresselhaus, M. S. Characterizing Carbon Nanotube Samples with Resonance Raman Scattering. New J Phys. 2003, 3, 139.1−139.17. (18) Dresselhaus, M. S.; Dresselhaus, G.; Saito, R.; Jorio, A. Raman Spectroscopy of Carbon Nanotubes. Phys. Rep. 2005, 409, 47−99. (19) Jorio, A.; Saito, R.; Dresselhaus, G.; Dresselhaus, M. S. Determination of Nanotubes Properties by Raman Spectroscopy. Phil. Trans. R. Soc., A 2004, 362, 2311−2336. (20) Joshi, A.; Punyani, S.; Bale, S. S.; Yang, H.; Borca-Tasciuc, T.; Kane, R. S. Nanotube-Assisted Protein Deactivation. Nat. Nanotechnol. 2008, 3, 41−45. (21) Chen, C. Y.; Jafvert, C. T. Photoreactivity of Carboxylated Single-Walled Carbon Nanotubes in Sunlight: Reactive Oxygen Species Production in Water. Environ. Sci. Technol. 2010, 44, 6674− 6679. (22) Chen, C. Y.; Jafvert, C. T. The Role of Surface Functionalization in the Solar Light-Induced Production of Reactive Oxygen Species by Single-Walled Carbon Nanotubes in Water. Carbon 2011, 49, 5099− 5106. (23) Scalbert, A. Antimicrobial Properties of Tannins. Phytochemistry 1991, 30, 3875−3883. (24) Nie, Z.; Liu, K. J.; Zhong, C.-J.; Wang, L.-F.; Yang, Y.; Tian, Q.; Liu, Y. Enhanced Radical Scavenging Activity by AntioxidantFunctionalized Gold Nanoparticles: A Novel Inspiration for Develop-

Efflux Content Measurements. The bacterial cell membrane integrity was examined by UV spectroscopy at 260 nm. The bacterial membrane disruption can be assessed from the release of cell cytoplasmic contents. The amount of DNA and RNA released from the cytoplasm can be estimated from the measured absorbance at 260 nm.12 After 2 h of incubation with 200 μL of SWCNTs to 2 mL of bacterial suspensions (S. sonnei, ∼1.5 × 107 cfu/mL and K. pneumoniae, ∼1.2 × 107 cfu/mL), the solutions were then immediately filtered with a 0.22 μm PTFE membrane filter to remove the bacteria. The supernatant was then diluted appropriately, and the optical density at 260 nm was recorded. Statistical Analysis. All analyses were carried out in triplicate (n = 3) and expressed as the mean ± standard deviation. An ANOVA test (using SPSS 17.0 statistical software; Stat Soft Inc., Tulsa, OK) was used to compare the mean values of experiments. The statistical difference between means was determined with a Duncan multiple range test (p < 0.05) and a Dunnett t test.

■

ASSOCIATED CONTENT

S Supporting Information *

Description of the material and characterization of CNTs including Raman and length and diameter distributions. Bacterial growth curve in a nutrient medium for MWCNTs and SWCNTs. Antioxidant conformation from FT-IR. Correlation analysis of cell viability and generated ROS (singlet oxygen radical, superoxide radical, and hydroxyl radical). CNT treatment for bacterial cell viability. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Present Address ⊥

Department of NanoScience and Technology, Bharathiar University, Coimbatore 641046, India Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS K.R. acknowledges Bharathiar University for the award of a university research fellowship and expresses his sincere thanks to CSIR, New Delhi, for the award of a senior research fellowship. R.G. acknowledges the financial support provided by Department of Science and Technology, Government of India within the INSPIRE programme.

■

REFERENCES

(1) García-Aljaro, C.; Cella, L. N.; Shirale, D. J.; Park, M.; Muñoz, F. J.; Yates, M. V.; Mulchandani, A. Carbon Nanotubes-Based Chemiresistive Biosensors for Detection of Microorganisms. Biosens. Bioelectron. 2010, 26, 1437−1441. (2) Bianco, A.; Kostarelos, K.; Prato, M. Applications of Carbon Nanotubes in Drug Delivery. Curr. Opin. Chem. Biol. 2005, 9, 674− 679. 600

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601

Langmuir

Article

ment of New Artificial Antioxidants. Free Radic. Biol. Med. 2007, 43, 1243−1254. (25) Brunet, L.; Lyon, D. Y.; Hotze, E. M.; Alvarez, P. J. J.; Wiesner, M. R. Comparative Photoactivity and Antibacterial Properties Of C60 Fullerenes and Titanium Dioxide Nanoparticles. Environ. Sci. Technol. 2009, 43, 4355−4360. (26) Nordberg, J.; Arnér , E. S. Reactive Oxygen Species, Antioxidants, and the Mammalian Thioredoxin System. Free Radic. Biol. Med. 2001, 31, 1287−1312. (27) Kappus, H. Lipid Peroxidation: Mechansim and Biological Significance. In Free Radicals and Food Additives; Aruoma, O. I., Halliwell, B., Eds.; Taylor & Francis: London, 1991; pp 59−75. (28) 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. (29) Li, Y.; Zhang, W.; Niu, J.; Chen, Y. Mechanism of Photogenerated Reactive Oxygen Species and Correlation with the Antibacterial Properties of Engineered Metal-Oxide Nanoparticles. ACS Nano 2012, 6, 5164−5173. (30) Applerot, G.; Lellouche, J.; Lipovsky, A.; Nitzan, Y.; Lubart, R.; Gedanken, A.; Banin, E. Understanding the Antibacterial Mechanism of CuO Nanoparticles: Revealing the Route of Induced Oxidative Stress. Small 2012, 8, 3326−37. (31) Lin, D.; Xing, B. Tannic Acid Absorption and Its Role for Stabilizing Carbon Nanotube Suspensions. Environ. Sci. Technol. 2008, 42, 5917−5923. (32) Ohkawa, H.; Ohishi, N.; Yagi, K. Assay for Lipid Peroxides in Animal Tissues by Thiobarbituric Acid Reaction. Anal. Biochem. 1979, 95, 351−358. (33) Reddy, A. R.; Reddy, Y. N.; Krishna, D. R.; Himabindu, V. Multi wall Carbon Nanotubes Induce Oxidative Stress and Cytotoxicity in Human Embryonic Kidney (HEK293) Cells. Toxicology 2010, 272, 11−16.

601

dx.doi.org/10.1021/la403332b | Langmuir 2014, 30, 592−601