APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Apr. 2011, p. 23252331 0099-2240/11/$12.00 doi:10.1128/AEM.02149-10 Copyright 2011, American Society for Microbiology.

. All Rights Reserved.

Vol. 77, No. 7

Antibacterial Activity and Mechanism of Action of Zinc Oxide Nanoparticles against Campylobacter jejuni
Yanping Xie,1 Yiping He,2* Peter L. Irwin,2 Tony Jin,3 and Xianming Shi1*
Joint Sino-U.S. Food Safety Research Center & Bor Luh Food Safety Center, School of Agriculture & Biology, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai, China 200240,1 and Joint U.S.-Sino Food Safety Research Center & Molecular Characterization of Foodborne Pathogens Research Unit2 and Residue Chemistry and Predictive Microbiology Research Unit,3 U.S. Department of Agriculture, Agricultural Research Service, Eastern Regional Research Center (USDA-ARS-ERRC), 600 East Mermaid Lane, Wyndmoor, Pennsylvania 19038
Received 10 September 2010/Accepted 28 January 2011

The antibacterial effect of zinc oxide (ZnO) nanoparticles on Campylobacter jejuni was investigated for inhibition and inactivation of cell growth. The results showed that C. jejuni was extremely sensitive to treatment with ZnO nanoparticles. The MIC of ZnO nanoparticles for C. jejuni was determined to be 0.05 to 0.025 mg/ml, which is 8- to 16-fold lower than that for Salmonella enterica serovar Enteritidis and Escherichia coli O157:H7 (0.4 mg/ml). The action of ZnO nanoparticles against C. jejuni was determined to be bactericidal, not bacteriostatic. Scanning electron microscopy examination revealed that the majority of the cells transformed from spiral shapes into coccoid forms after exposure to 0.5 mg/ml of ZnO nanoparticles for 16 h, which is consistent with the morphological changes of C. jejuni under other stress conditions. These coccoid cells were found by ethidium monoazide-quantitative PCR (EMA-qPCR) to have a certain level of membrane leakage. To address the molecular basis of ZnO nanoparticle action, a large set of genes involved in cell stress response, motility, pathogenesis, and toxin production were selected for a gene expression study. Reverse transcription-quantitative PCR (RT-qPCR) showed that in response to treatment with ZnO nanoparticles, the expression levels of two oxidative stress genes (katA and ahpC) and a general stress response gene (dnaK) were increased 52-, 7-, and 17-fold, respectively. These results suggest that the antibacterial mechanism of ZnO nanoparticles is most likely due to disruption of the cell membrane and oxidative stress in Campylobacter. Directly adding antimicrobial agents to food or into packaging materials during food processing is considered an effective means of controlling microbial contaminants in food and extending the shelf life of fresh produce and meat. In recent years, inorganic antimicrobial agents, such as metal oxides, have received increasing attention in food applications because they are not only stable under high temperatures and pressures that may occur in harsh food-processing conditions, but they are also generally regarded as safe for human beings and animals relative to organic substances (6, 24). Zinc oxide (ZnO) is listed as generally recognized as safe (GRAS) by the U.S. Food and Drug Administration (21CFR182.8991). As a food additive, it is the most commonly used zinc source in the fortication of cereal-based foods. Because of its antimicrobial properties, ZnO has been incorporated into the linings of food cans in packages for meat, sh, corn, and peas to preserve colors and to prevent spoilage. Nano-sized particles of ZnO have more pronounced antimicrobial activities than large particles, since the small size (less than 100 nm) and high surface-to-volume ratio of nanopar* Corresponding author. Mailing address for Yiping He: USDAARS-ERRC, 600 East Mermaid Lane, Wyndmoor, PA 19038. Phone: (215) 233-6422. Fax: (215) 836-3742. E-mail: yiping.he@ars .usda.gov. Mailing address for Xianming Shi: Mail box 49, School of Agriculture & Biology, Shanghai Jiao Tong University, 800 Dongchuan Road, Shanghai, China 200240. Phone and fax: 86-213420-6616. E-mail: xmshi@sjtu.edu.cn. Published ahead of print on 4 February 2011. 2325

ticles allow for better interaction with bacteria. Recent studies have shown that these nanoparticles have selective toxicity to bacteria but exhibit minimal effects on human cells (21). ZnO nanoparticles have been shown to have a wide range of antibacterial activities against both Gram-positive and Gram-negative bacteria, including major foodborne pathogens like Escherichia coli O157:H7, Salmonella, Listeria monocytogenes, and Staphylococcus aureus (13, 14), but currently there is no information available on their antibacterial effect against species of Campylobacter. Campylobacter jejuni is a leading cause of microbial foodborne illness worldwide. In fact, it has recently been shown that approximately 80% of poultry products are contaminated with this pathogen (11). Consumption of Campylobacter-contaminated food and water usually causes a mild to severe gastrointestinal infection in humans that can sometimes develop into a life-threatening disease called GuillainBarre syndrome (28). Therefore, it is important to focus on the use of ZnO particles as a potential food safety intervention technology to effectively control Campylobacter and other microbial contaminants in food. To make better use of ZnO nanoparticles in food products and to assist in the development of powerful, but nontoxic, antimicrobial derivatives, it is necessary to understand the mechanism of action of ZnO nanoparticles against bacteria, but to date, the process underlying their antibacterial effect is not clear. However, a few studies have suggested that the primary cause of the antibacterial function might be from the disruption of cell membrane activity (4). Another possibility

2326

XIE ET AL.

APPL. ENVIRON. MICROBIOL.

plates to determine the bactericidal (bacteria-killing) or bacteriostatic (bacteriainhibiting) effect of ZnO nanoparticles. The MICs of ZnO nanoparticles for C. jejuni, E. coli O157:H7, and Salmonella were determined using a broth microdilution method reported previously (19). Briey, serial 2-fold dilutions of nanoparticles ranging from 0.00625 to 1.6 mg/ml were prepared in a 96-well microtiter plate using MH or LB broth. Freshly grown bacterial cells were inoculated into each well to give a nal concentration of 104 CFU/ml. After the inocula were incubated microaerobically for 24 h at 42C for C. jejuni or aerobically for 16 h at 37C for E. coli O157:H7 and Salmonella, cell growth in each well was monitored and compared with that of the positive-control well to which no ZnO nanoparticles were added. The MIC was recorded as the lowest concentration of ZnO nanoparticles that completely inhibited cell growth. Examination of cell morphology by scanning electron microscope (SEM). Mid-log-phase C. jejuni cultures were treated with 0.5 mg/ml of ZnO nanoparticles for 12 h under microaerobic conditions. Aliquots of 200 l of treated and untreated cell suspensions were deposited on glass coverslips. After being air dried for 1 h, the coverslips were xed with a primary xative solution containing 2.5% glutaraldehyde and 0.1 M imidazole buffer solution (pH 7.2) for 2 h. Subsequently, the xative solution was exchanged for 0.1 M imidazole buffer, followed by dehydration with a series of ethanol solutions (50%, 80%, and 100%), with three ethanol changes at each concentration. Finally, the coverslips were dried by liquid CO2-ethanol exchange in a DCP-1 Critical Point dryer (Denton Vacuum, Inc., Cherry Hill, NJ). The coverslips were mounted on SEM stubs with carbon adhesive tabs and then sputter coated with a thin layer of gold using a Scancoat Six sputter coater (BOC Edwards, Wilmington, MA). Digital images of the treated and untreated C. jejuni cells were acquired using a Quanta 200 FEG scanning electron microscope (FEI, Inc., Hillsboro, OR) at an accelerating voltage of 10 kV and instrumental magnications of 25,000. EMA treatment, DNA isolation, and EMA-qPCR assay. Ethidium monoazide (EMA) treatment of C. jejuni cells and a follow-up quantitative PCR (qPCR) were carried out as described before (10). Briey, 1 ml of freshly grown cells was treated with 20 mg/ml of EMA in the dark for 5 min and subsequently exposed to a 600-W halogen light for 1 min. Cells were then immediately washed with phosphate-buffered saline and subjected to DNA extraction using a DNeasy tissue kit (Qiagen, Valencia, CA) according to the manufacturers instructions. The hipO gene was selected as a target for detection in qPCR because the DNA sequence is unique to C. jejuni. The primers and TaqMan probe of hipO used in this experiment were reported in previous work (9). RNA preparation and RT-qPCR analysis of gene expression. To prepare total cellular RNA, 80 ml of the C. jejuni 81-176 culture in the late log phase of growth was treated or not treated with ZnO nanoparticles at 0.1 mg/ml for 30 min. The ZnO concentration used herein was determined from the cell-killing results shown in Fig. 1A. After the treatment, cells were harvested by centrifugation at 4,000 g for 10 min at 4C. RNA isolation was carried out using TRI Reagent according to the manufacturers instructions (Molecular Research Center, Inc., Cincinnati, OH). DNase I treatment and reverse transcription (RT) of the RNA samples were done as described before (8). Quantication of cDNA was performed on a 7500 real-time PCR system (Applied Biosystems, Foster City, CA). For PCR, all the listed primers were designed using Primer3 software (http: //frodo.wi.mit.edu/primer3). Each 20- l PCR mixture contained 0.25 EvaGreen dye (Biotium, Hayward, CA), 0.25 M each primer, 2 l of cDNA template, 5 units of Platinum Taq DNA polymerase, and buffer (Invitrogen, Inc., Carlsbad, CA). The amplication program was 50C for 2 min and 95C for 10 min, followed by 40 cycles at 95C for 15 s and 60C for 1 min. The gyrA gene was used as a reference for data normalization. tsf and 16S rRNA housekeeping genes were also included as controls to ensure data reliability. All the samples, including no-RT and no-template controls, were analyzed in CT triplicate. Data analysis was performed using the 2 method, where CT CT(treated sample) CT(untreated sample), CT CT(target gene) CT(gyrA), and CT is the threshold cycle value for the amplied gene (15).

could be the induction of intercellular reactive oxygen species, including hydrogen peroxide (H2O2), a strong oxidizing agent harmful to bacterial cells (13, 22). It has also been reported that ZnO can be activated by UV and visible light to generate highly reactive oxygen species such as OH , H2O2, and O22 . The negatively charged hydroxyl radicals and superoxides cannot penetrate into the cell membrane and are likely to remain on the cell surface, whereas H2O2 can penetrate into bacterial cells (18). To better understand the nature of the inhibitory and lethal effects of ZnO nanoparticles on bacteria, we used C. jejuni as a model organism to investigate this mechanism. C. jejuni is a Gram-negative, spiral-shaped, highly motile, thermophilic and microaerophilic bacterium that grows optimally in a neutral pH and microaerobic environment at 42C. Unlike other major foodborne pathogens, such as E. coli O157:H7, Salmonella, and L. monocytogenes, C. jejuni has a low tolerance for oxygen but does require some for growth (i.e., it is microaerophilic). Due to the lack of some important oxidative stress response genes (soxRS and oxyR) and a global stationary-phase stress response gene (rpoS), C. jejuni is extremely sensitive to oxidative stress as well as to other environmental stresses. Exposure of this organism to different stresses results in a remarkable morphological shift from spiral-shaped cells to coccoid forms, which is associated with the loss of culturability (10, 25). Due to these distinguishing characteristics and sensitive stress responses, C. jejuni is highly suitable for studying the modes of action of ZnO nanoparticles against bacterial cells, especially in the assessment of cell membrane integrity and the reactive oxygen species-induced stress response. The purpose of this research was to evaluate the antibacterial effects and to investigate the mechanism of action of ZnO nanoparticles against C. jejuni by examining cell morphology, membrane permeability, and gene expression through the utilization of scanning electron microscopy as well as advanced molecular methods. The results were compared to those for other foodborne pathogens, including E. coli O157:H7 and Salmonella.
MATERIALS AND METHODS ZnO nanoparticles. ZnO nanoparticles (purity over 99.7%) with an average size of 30 nm and a Brunauer-Emmett-Teller (BET)-specic surface area of 35 m2/g were purchased from Inframat Advanced Materials LLC (Manchester, CT). A stock suspension was prepared by resuspending the nanoparticles in double-distilled water (ddH2O) to yield a nal concentration of 100 mg/ml; the suspension was kept at 4C. Immediately after the suspension was subjected to vigorous vortex mixing, aliquots of the suspension were added into Mueller-Hinton medium (MH; Becton Dickinson Co., Sparks, MD) for the following experiments. Bacterial culture conditions and antibacterial test. C. jejuni strains 81-176, ATCC 35918, and ATCC 49943 were grown at 42C in MH broth in a microaerobic workstation (Don Whitley Scientic Ltd., Shipley, United Kingdom) in which 5% O2, 10% CO2, 85% N2, and 82% relative humidity were maintained. A mixed culture of the three C. jejuni strains was prepared by combining equal volumes (1/3) of each pure culture growing at the late log phase. Salmonella enterica serovar Enteritidis ATCC 13076 and E. coli O157:H7 ATCC 43889 were aerobically grown at 37C in Luria-Bertani medium (Becton Dickinson). Bacterial growth inhibition was tested by inoculating ca.104 CFUs of C. jejuni cells onto each MH agar plate or into 20 ml of MH broth containing various concentrations (0, 0.025, 0.03, 0.04, 0.05, and 0.10 mg/ml) of ZnO nanoparticles. After the inocula were incubated for 16 h, inhibition of cell growth was determined by counting the numbers of CFUs on the plates or by the turbidities of the cell cultures. The inoculas that showed no cell growth were further veried for cell culturability by spreading 100- l aliquots of the cultures onto drug-free MH agar

RESULTS Growth inhibition of C. jejuni by ZnO nanoparticles. Growth inhibition of C. jejuni 81-176 was examined both on agar plates and in broth containing a range of concentrations (0, 0.025, 0.03, 0.04, 0.05, and 0.10 mg/ml) of ZnO nanoparticles. On the plates spread with 104 CFU/plate and in the broth inoculated with the equivalent number of cells, bacterial

VOL. 77, 2011

ANTIMICROBIAL MECHANISM OF ZnO NANOPARTICLES

2327

FIG. 2. Scanning electron microscopic images of C. jejuni. (A) C. jejuni cells in the mid-log phase of growth were treated with 0.5 mg/ml of ZnO nanoparticles for 12 h under microaerobic conditions. (B) Untreated cells from the same growth conditions were used as a control.

FIG. 1. Antibacterial activities of ZnO nanoparticles against C. jejuni, S. enterica serovar Enteritidis, and E. coli O157:H7. Freshly grown bacterial cultures (108 to 109 CFU/ml) were treated with a range of concentrations of ZnO nanoparticles. Culturable cell numbers were determined at the time intervals after treatment shown on the gure. The values for CFU/ml are the means of 12 replicates. Error bars indicate standard deviations of the means.

growth was completely inhibited at 0.03 mg/ml of ZnO nanoparticles. However, at a concentration of 0.025 mg/ml, ZnO nanoparticles had a modest effect on cell growth, resulting in the recovery of fewer viable cells than from the untreated control suspension. To determine if the growth inhibition was caused by an inhibitory or lethal effect of ZnO nanoparticles, 100- l aliquots of the treated-cell suspension were spread onto drug-free MH plates. The results showed that the cells treated with 0.03 mg/ml of the nanoparticles for 16 h were no longer culturable, suggesting a lethal effect of ZnO nanoparticles against C. jejuni. In addition, the MIC of ZnO nanoparticles for all three C. jejuni strains was determined to be between 0.05 and 0.025 mg/ml, which was 8- to 16-fold lower than that (0.4 mg/ml) for E. coli O157:H7 and S. enterica serovar Enteritidis, clearly indicating a higher level of susceptibility of C. jejuni to ZnO nanoparticles. Lethal effect of ZnO nanoparticles on C. jejuni. The lethal effect of ZnO nanoparticles on C. jejuni was further investigated using ca. 108 CFU/ml of freshly grown pure and mixed

cultures of C. jejuni strains 81-176, ATCC 35918, and ATCC 49943. Cell culturability of all three C. jejuni strains was affected by ZnO nanoparticles at all concentrations tested (Fig. 1A to D). Most signicantly, 0.5, 0.3, and 0.1 mg/ml of ZnO nanoparticles resulted in complete killing (100%) of 108 CFU/ml of C. jejuni cells in 3 h or less. The pure and mixed cultures of three C. jejuni strains showed a similar susceptibility to ZnO nanoparticles. In contrast to the strong and rapid bactericidal action against C. jejuni, 20- to 100fold-higher concentrations (10 mg/ml) of ZnO nanoparticles caused only a 1- or 2-log reduction in viable E. coli O157:H7 and S. enterica serovar Enteritidis cells after an 8-h exposure (Fig. 1E and F). These results demonstrated that ZnO nanoparticles are effective at killing C. jejuni even at low concentrations. Morphological analysis of C. jejuni. Effects of ZnO nanoparticles on C. jejuni cell morphology were examined by scanning electron microscopy. After a 12-h treatment with 0.5 mg/ml of ZnO nanoparticles in MH broth under microaerobic conditions, spiral-shaped C. jejuni cells underwent a dramatic change from spiral to coccoid morphological forms. The SEM image in Fig. 2A illustrates the dominance of coccoid forms in the treated cells and shows the formation of irregular cell surfaces and cell wall blebs in great detail. These coccoid cells remained intact and possessed sheathed polar agella. The image of the untreated cells clearly displays spiral shapes (Fig. 2B). Moreover, this ZnO nanoparticle-induced formation of coccoid cells was conrmed by confocal microscopic visualization (data not shown). Not surprisingly, a similar transformation in morphology was observed when C. jejuni was exposed to different environmental stresses, including oxidative stress (10, 25). To determine the bactericidal versus bacteriostatic effect of ZnO nanoparticles, cultures previously exposed to ZnO nanoparticles and exhibiting a coccoid cell morphology were spread plated. No growth of the coccoid cells was observed on drug-free MH plates, conrming that they were no longer culturable. Together these results suggest that ZnO nanoparticles cause not only cell morphology changes but also a lethal effect against C. jejuni. EMA-PCR assessment of cell membrane integrity. EMA selectively enters into membrane-compromised cells and binds

2328

XIE ET AL.

APPL. ENVIRON. MICROBIOL.

FIG. 3. EMA-qPCR of C. jejuni membrane integrity. Mid-logphase cells exposed to different concentrations of ZnO nanoparticles were briey treated with (black bars) and without (white bars) EMA. Inhibition of DNA amplication was quantied by real-time PCR of the hipO gene. Reduced DNA amplication in cells exposed to 0.3 and 0.5 mg/ml of ZnO nanoparticles indicates a certain degree of membrane leakage in the treated cells.

to cellular DNA, which subsequently inhibits PCR amplication of the DNA. The reduction in PCR amplication has been used as an indicator of cell membrane leakage (17). To assess the membrane integrity of the coccoid cells, EMA-qPCR was performed on the C. jejuni cultures treated with 0, 0.1, 0.3, and 0.5 mg/ml of ZnO nanoparticles for 12 h. The results in Fig. 3 show that the cells treated with 0.3 and 0.5 mg/ml of ZnO nanoparticles had a 10-fold (1-log) reduction in DNA amplication, indicating increased penetration of EMA into the treated cells. This result demonstrates that the treatment of C. jejuni with ZnO nanoparticles increases cell membrane permeability (i.e., damages membrane integrity). Gene expression prole of C. jejuni in response to ZnO nanoparticle treatment. To understand the molecular basis of ZnO nanoparticle action against bacterial cells, a set of C. jejuni genes involved in general and oxidative stress responses, motility, pathogenesis, and toxin production were selected for a gene expression study (Table 1). After late-log-phase cells were exposed to 0.1 mg/ml of ZnO nanoparticles for 30 min, the transcripts of these genes were quantied by RT-qPCR. Most signicantly, two oxidative stress genes, katA (encoding catalase) and ahpC (encoding alkyl hydroperoxide reductase), and one general stress gene, dnaK (encoding a chaperone), were found to be upregulated 52-, 7-, and 17-fold, respectively, in response to the treatment. The transcription levels of other stress response genes as well as the analyzed virulence genes were not signicantly up- or downregulated ( 3-fold) (Fig. 4). As expected, the expression levels of three housekeeping genes (gyrA, which encodes gyrase subunit A; tsf, which encodes elongation factor TS; and 16S rRNA) were not changed regardless of the treatment (Fig. 4). Similar gene expression results were obtained from cells treated with 0.05 mg/ml of ZnO nanoparticles for the same length of time (data not shown). All these results suggest that the antibacterial mechanism of ZnO nanoparticles is likely due to oxidative stress in C. jejuni cells. DISCUSSION ZnO nanoparticles have a broad spectrum of antibacterial activities. At concentrations higher than 0.24 mg/ml, they inhibit the growth of E. coli O157:H7, L. monocytogenes, and S.

enterica serovar Enteritidis (12, 14). An inhibitory effect of ZnO nanoparticles on Bacillus subtilis, Staphylococcus aureus, Staphylococcus epidermidis, Streptococcus pyogenes, and Enterococcus faecalis has been reported as well (13). In this study, the antibacterial properties of ZnO nanoparticles were rst investigated in C. jejuni, the most common foodborne pathogen. Our results showed that C. jejuni is extremely sensitive to ZnO nanoparticles, with a MIC 8- to 16-fold lower than those for E. coli O157:H7 and Salmonella. Antibacterial tests both on agar plates and in broth showed that 0.03 mg/ml of ZnO nanoparticles was sufcient to inactivate C. jejuni, whereas the concentration of nanoparticles needed for 100% inhibition of E. coli O157:H7 growth was between 0.24 and 0.98 mg/ml (4, 14), approximately 8 to 32 times higher than the lethal dosage for C. jejuni. It has previously been reported that the antibacterial activity of ZnO nanoparticles increases with reduction in particle size (16). The number of bacterial cells and the growth media used could also contribute to variation in results. For data consistency, we used 30-nm (average-sized) ZnO nanoparticles to test similar numbers of bacterial cells. Previously, it was unclear whether ZnO nanoparticles function as a bactericidal or bacteriostatic agent, except for a few reports on the inhibition of bacterial growth (13, 14). In this study, we demonstrated that the action of ZnO nanoparticles on C. jejuni was bactericidal, not bacteriostatic, by showing no recovery of the treated cells on drug-free MH plates as well as by the rapid killing of 108 CFU/ml of freshly grown cells of three different C. jejuni strains. The effectiveness of ZnO nanoparticles in inactivating C. jejuni was also compared with their effectiveness against other major foodborne pathogens. For E. coli O157:H7 and Salmonella, 20- to 100-fold-higher concentrations of ZnO nanoparticles were needed to reduce 1 to 2 logs of viable cells (Fig. 1). Therefore, the bactericidal action of ZnO nanoparticles against C. jejuni was extremely effective. Although the antibacterial mechanism of ZnO nanoparticles is still unknown, the possibilities of membrane damage caused by direct or electrostatic interaction between ZnO and cell surfaces, cellular internalization of ZnO nanoparticles, and the production of active oxygen species such as H2O2 in cells due to metal oxides have been proposed in earlier studies (6, 24). The generation of H2O2 in ZnO slurries was determined by oxygen electrode analysis and spectrophotouorometry (23, 27). In examining cell morphology, membrane integrity, and gene expression in C. jejuni, we found that all of these aspects were affected by ZnO nanoparticles. A dramatic change in C. jejuni cell morphology was revealed by SEM by showing the dominance of coccoid forms in the treated cells while the untreated cells remained spiral. This considerable alteration in cell morphology was not only observed in C. jejuni cells treated with ZnO nanoparticles but has also been found in Campylobacter cells and in cells of the closely related genus Helicobacter when cells were exposed to different stresses (1, 5, 10). It might be specic to spiral bacteria, as no signicant changes in cell shape were found in E. coli O157:H7 after exposure to ZnO, but the nanoparticles adhered to the cell surface (29). In addition to changing the structure of C. jejuni cells, ZnO nanoparticles resulted in the formation of irregular cell surfaces and membrane blebbing and an increase in membrane permeability. This induced membrane leakage was also consistently observed in E. coli O157:H7 by transmission electron microscopy

VOL. 77, 2011

ANTIMICROBIAL MECHANISM OF ZnO NANOPARTICLES TABLE 1. Genes and primers selected for RT-qPCR

2329

Gene function/protein encodeda

Gene

Primer

Sequence (5 33 )

Oxidative stress response genes Peroxide-sensing regulator Flavodoxin I Iron-binding protein Superoxide dismutase (Fe) Alkyl hydroperoxide reductase Catalase Nonheme iron-containing ferritin Ferredoxin Rubrerythrin Ferric uptake regulator Carbon storage regulator General stress response genes Probable thiol peroxidase Cochaperonin Chaperonin Chaperone Cochaperone Bacterioferritin comigratory protein homolog RelA/Spot family protein Rod shape-determining protein RND efux system; membrane fusion protein RND efux system; inner membrane transporter RND efux system; outer membrane lipoprotein Major outer membrane protein Inner membrane protein ATP-dependent Clp protease ATP-binding subunit Virulence factors and toxins Cytolethal distending toxin cluster

perR dA dps sodB ahpC katA cft fdxA rbr fur csrA

Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse

TTCAATCTCTTTAGCGACGG CACATTTGGCGCAAACAACA TCCACCAAGACTAAGCCCTG CAGAAGGAGCGGCTAATACAA CCACTAATGTTAATATGCGTTCC TGTGCTTGATAATCTTGCGACAA TGGCGGTTCATGTCAAAGTA ACCAAAACCATCCTGAACCA AGTTGCCCTTCGTGGTTCGT ATCGCCCTTATTCCATCCTG ACCGTTCATGCTAAGGGAAG CCTACCAAGTCCCAGTTTCC TTCTTCTTCGTGTTGTTCGC GCTGGAGCCTTCTTGTTTGC CCCCACTTCTCATATCAGCG ATGCGTTGAATGCGTAGGAC TGCAGCAGTTACTAGGTTTT AGACATTTTAGAGAAGCGGC CCATTTCTTTTGGTTCAGCAG TGCAATCAAGGCTTGCTGTC TCAAAGTCGTTCAAACAGGG TCATTCTGAACAACAGAATGC GCCAGTTACAATGGTGCTGA TTTGCCACAAAATCACTTGC AAACAACAGCCTCAGGCATAA TTCTGTTCCACCGTATTTAGCA GCAGGCGATGGAACAACTAC TCCATACCGCGTTTTACCTC CGGTATGCCACAAATCGAAG GCTAAGTCCGCTTGAACCTG TTTAAAAGGCGGTGGATTTG TTTTCTACGACGCGATGATG ACCCCAGGTTGTACTACAGAAG AGCAATCTTACCTGTTTCATCG GCCCCAATAGCCCATAGAC ACCCCAAGCAAATCAAGAAC GAGCCTTCTGTTGTGGCAGTT AGCGGATCATTTTTTCAGTCAT TATTACGCCGCTAACTTGAG CAGCAAAGAAGAAGCACCAA TAATCCAGGTATGGGAGGTA GGAAAGATAGAAATGTAAGCG GGACGTTGAAGCAAGATGGT AGTTGGCGCTGTAGGTGAAT TTGATAGCGAACTTGATGAT ATACGAAGTCAGCACCAACG CTATTTCCATACCCCACAGC CCTTTAATTGCAGAAGTTCC GTAGGAGCTGGAAGCACAGG ACGGCGACTTAGGGGTTTAT TCCACATTTGTGCGTGATTG GATTTGGCGATGCTAGAGTTT AAAGCATCATTTCCATTGCG ACCAAGAACAGCCACTCCAA CCAAAAGGAAGTTCATCAGC AGCCTTTGCAACTCCTACTG CTCATCATTTGGAACGACTTG AATTATACTCATGCGGTGGC CCAATGTCGGCTCTGATTTG GCGCAGGAAGTGGATTTTC CCGTTTCCATCACCATCTTC ACACGCTTTGAAACAGGAGG Continued on following page

tpx groES groEL dnaK dnaJ BCP spoT mreB cmeA cmeB cmeC porA yagU clpA

cdtA cdtB cdtC

Invasion antigen B Flagellin A Flagellin B

ciaB aA aB

2330

XIE ET AL. TABLE 1Continued

Gene function/protein encodeda Gene Primer

APPL. ENVIRON. MICROBIOL.

Sequence (5 33 )

Outer membrane bronectin-binding protein N-acetylneuraminic acid synthetase Vacuolating cytotoxins Housekeeping genes Gyrase subunit A Elongation factor TS 16S rRNA
a

cadF neuB vacB

Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse Forward Reverse

TGCTTGTGGAGCTGGACGAG TAAAAGCGGTGGATTTGGAC TGTTGAAGTGGGACTAAGCG TCTAACTTGCCATCGCCTAA AAGGACGAGGTAGCATAGGT CAAACGGCGATAGTGTTGAT TGCTAAAGTGCGTGAAATCG GCATTGGTGCGTTTTCCTAT GAACTCCGCGAAAGTACAGG TTGCCACAAAATCTGTTTCG GCTCGTGTCGTGAGATGTTG TCACCGTAGCATGGCTGAT

gyrA tsf 16S rRNA gene

RND, resistance nodulation cell division family.

and Raman spectroscopy when cells were treated with ZnO nanoparticles (14). When extracellular environments change, bacteria adopt mechanisms that quickly regulate the synthesis of defensive proteins in response to stress. In Campylobacter, a number of genes/proteins that play critical roles in protecting cells from different stresses, especially oxidative stress, have been identied (26). Most importantly, to eliminate reactive oxygen species and to assist the organism in defense against oxidative stress, superoxide dismutase (SodB) breaks down O 2 to H2O2 and O2, catalase (KatA) inactivates H2O2 and interrupts the formation of toxic intermediates, and alkyl hydroperoxide reductase (AhpC) destroys toxic hydroperoxide intermediates and repairs damage to molecules caused by oxidation (3, 20). In addition to these oxidative stress response proteins, general stress response proteins (DnaK, DnaJ, GroES, and GroEL), which act as molecular chaperones and play a critical role in preventing protein aggregation and refolding, are also important for cell survival (2). Analysis of ZnO nanoparticle-modulated stress gene expression showed that the transcription levels of two oxidative stress genes (ahpC and katA) and one general stress response gene (dnaK) were signicantly increased, up to 7- to 52-fold, while another 4 stress response genes (sodB, dps, groEL, and groES) were expressed at approximately 2- to 3-times-higher levels. Expression of all other

stress response genes was either unchanged or downregulated. Because KatA is a single catalase enzyme whose expression level increases in C. jejuni upon exposure to H2O2 (7), the 52-fold induction of KatA expression suggests a high probability that more intercellular H2O2 is produced in response to the ZnO nanoparticles. From these experiments and the role of the oxidative stress regulatory system in Campylobacter, we can conclude that the antibacterial mechanism of ZnO nanoparticles is very likely through increased levels of oxidative stress in bacterial cells. Furthermore, the expression of a number of virulence factors related to cell motility, toxin production, and adhesion to host cells was also examined in response to ZnO nanoparticles. All of the analyzed virulence genes were found to be downregulated, suggesting decreased pathogenicity of the bacterium after treatment. In summary, ZnO nanoparticles exhibited remarkable antibacterial activity and demonstrated a lethal effect against C. jejuni, even at low concentrations. ZnO nanoparticles induced signicant morphological changes, measurable membrane leakage, and substantial increases (up to 52-fold) in oxidative stress gene expression in C. jejuni. Based on these phenomena and cell responses, a plausible mechanism of ZnO inactivation of bacteria involves the direct interaction between ZnO nanoparticles and cell surfaces, which affects the permeability of membranes where nanoparticles enter and induce oxidative

FIG. 4. Relative gene expression levels of ZnO nanoparticle-treated and untreated C. jejuni. C. jejuni cells in the late log phase of growth were treated with 0 or 0.1 mg/ml of ZnO nanoparticles for 30 min. Transcripts of the selected genes were quantied by RT-qPCR, and data were analyzed using the comparative critical threshold ( CT) method. The relative expression ratio for each gene is presented as a log2 value in the histogram. A ratio greater than zero ( 0) indicates upregulation of gene expression and a ratio below zero ( 0) indicates downregulation. Error bars indicate standard deviations for three replicates.

VOL. 77, 2011

ANTIMICROBIAL MECHANISM OF ZnO NANOPARTICLES

2331

stress in bacterial cells, subsequently resulting in the inhibition of cell growth and eventually in cell death.
ACKNOWLEDGMENTS This research was jointly supported by the Agriculture Research Service, U.S. Department of Agriculture, the Ministry of Science and Technology of China (2009BADB9B01 and 2009BAK43B31), and the Science and Technology Commission of Shanghai Municipality (09DZ0503300). We thank Guoping Bao of the Microscopy Imaging Facility of the USDA, ARS, ERRC, for technical assistance in acquiring the SEM images and George Paoli of the Molecular Characterization of FoodBorne Pathogens Research Unit of the USDA, ARS, ERRC, for providing the C. jejuni ATCC strains.
REFERENCES 1. Adams, B. L., T. C. Bates, and J. D. Oliver. 2003. Survival of Helicobacter pylori in a natural freshwater environment. Appl. Environ. Microbiol. 69: 74627466. 2. Arsene, F., T. Tomoyasu, and B. Bukau. 2000. The heat shock response of ` Escherichia coli. Int. J. Food Microbiol. 55:39. 3. Baillon, M. L., A. H. van Vliet, J. M. Ketley, C. Constantinidou, and C. W. Penn. 1999. An iron-regulated alkyl hydroperoxide reductase (AhpC) confers aerotolerance and oxidative stress resistance to the microaerophilic pathogen Campylobacter jejuni. J. Bacteriol. 181:47984804. 4. Brayner, R., et al. 2006. Toxicological impact studies based on Escherichia coli bacteria in ultrane ZnO nanoparticles colloidal medium. Nano Lett. 6:866870. 5. Catrenich, C. E., and K. M. Makin. 1991. Characterization of the morphologic conversion of Helicobacter pylori from bacillary to coccoid forms. Scand. J. Gastroenterol. Suppl. 181:5864. 6. Fu, G., P. S. Vary, and C. T. Lin. 2005. Anatase TiO2 nanocomposites for antimicrobial coatings. J. Phys. Chem. B 109:88898898. 7. Grant, K., and S. Park. 1995. Molecular characterization of katA from Campylobacter jejuni and generation of a catalase-decient mutant of Campylobacter coli by interspecic allelic exchange. Microbiology 141:13691376. 8. He, Y., J. G. Frye, T. P. Strobaugh, and C. Y. Chen. 2008. Analysis of AI-2/LuxS-dependent transcription in Campylobacter jejuni strain 81-176. Foodborne Pathog. Dis. 5:399415. 9. He, Y., et al. 2010. Simultaneous detection and differentiation of Campylobacter jejuni, C. coli, and C. lari in chickens using a multiplex real-time PCR assay. Food Anal. Methods 137:168174. 10. He, Y. P., and C. Y. Chen. 2010. Quantitative analysis of viable, stressed and dead cells of Campylobacter jejuni strain 81-176. Food Microbiol. 27:439446. 11. Jacobs-Reitsma, W. F., U. Lyhs, and J. Wagenaar. 2008. Campylobacter in the food supply, p. 627644. In I. Nachamkin, C. M. Szymanski, and M. J. Blaser (ed.), Campylobacter, 3rd ed. ASM Press, Washington, DC.

12. Jin, T., D. Sun, J. Y. Su, H. Zhang, and H. J. Sue. 2009. Antimicrobial efcacy of zinc oxide quantum dots against Listeria monocytogenes, Salmonella Enteritidis, and Escherichia coli O157:H7. J. Food Sci. 74:M46M52. 13. Jones, N., B. Ray, K. T. Ranjit, and A. C. Manna. 2008. Antibacterial activity of ZnO nanoparticle suspensions on a broad spectrum of microorganisms. FEMS Microbiol. Lett. 279:7176. 14. Liu, Y., et al. 2009. Antibacterial activities of zinc oxide nanoparticles against Escherichia coli O157:H7. J. Appl. Microbiol. 107:11931201. 15. Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene expression C T method. Methods data using real-time quantitative PCR and the 2 25:402408. 16. Nair, S., et al. 2009. Role of size scale of ZnO nanoparticles and microparticles on toxicity toward bacteria and osteoblast cancer cells. J. Mater. Sci. Mater. Med. 20(Suppl. 1):S235S241. 17. Nogva, H. K., S. M. Dromtorp, H. Nissen, and K. Rudi. 2003. Ethidium monoazide for DNA-based differentiation of viable and dead bacteria by 5 -nuclease PCR. Biotechniques 34:804813. 18. Padmavathy, N., and R. Vijayaraghavan. 2008. Enhanced bioactivity of ZnO nanoparticlesan antimicrobial study. Sci. Technol. Adv. Mater. 9:17. 19. Petersdorf, R. G., and J. C. Sherris. 1965. Methods and signicance of in vitro testing of bacterial sensitivity to drugs. Am. J. Med. 39:766779. 20. Purdy, D., S. Cawthraw, J. H. Dickinson, D. G. Newell, and S. F. Park. 1999. Generation of a superoxide dismutase (SOD)-decient mutant of Campylobacter coli: evidence for the signicance of SOD in Campylobacter survival and colonization. Appl. Environ. Microbiol. 65:25402546. 21. Reddy, K. M., et al. 2007. Selective toxicity of zinc oxide nanoparticles to prokaryotic and eukaryotic systems. Appl. Phys. Lett. 90:21390212139023. 22. Sawai, J. 2003. Quantitative evaluation of antibacterial activities of metallic oxide powders (ZnO, MgO and CaO) by conductimetric assay. J. Microbiol. Methods 54:177182. 23. Sawai, J., et al. 1996. Detection of active oxygen generated from ceramic powders having antibacterial activity. J. Chem. Eng. Jpn. 29:627633. 24. Stoimenov, P. K., R. L. Klinger, G. L. Marchin, and K. J. Klabunde. 2002. Metal oxide nanoparticles as bactericidal agents. Langmuir 18:66796686. 25. Tangwatcharin, P., S. Chanthachum, P. Khopaibool, and M. W. Grifths. 2006. Morphological and physiological responses of Campylobacter jejuni to stress. J. Food. Prot. 69:27472753. 26. van Vliet, A. H., J. M. Ketley, S. F. Park, and C. W. Penn. 2002. The role of iron in Campylobacter gene regulation, metabolism and oxidative stress defense. FEMS Microbiol. Rev. 26:173186. 27. Yamamoto, O., M. Komatsu, J. Sawai, and Z. E. Nakagawa. 2004. Effect of lattice constant of zinc oxide on antibacterial characteristics. J. Mater Sci. Mater. Med. 15:847851. 28. Yuki, N., and M. Koga. 2006. Bacterial infections in Guillain-Barre and Fisher syndromes. Curr. Opin. Neurol. 19:451457. 29. Zhang, L. L., Y. H. Jiang, Y. L. Ding, M. Povey, and D. York. 2007. Investigation into the antibacterial behavior of suspensions of ZnO nanoparticles (ZnO nanouids). J. Nanopart. Res. 9:479489.