Heteroaggregation Reduces Antimicrobial Activity of Silver

Aug 1, 2014 - Chuanjia Jiang , Benjamin T. Castellon , Cole W. Matson , George R. Aiken ... Tiezheng Tong , Jean-François Gaillard , and Kimberly A. ...
1 downloads 0 Views 2MB Size
Download PDF
Letter pubs.acs.org/journal/estlcu

Heteroaggregation Reduces Antimicrobial Activity of Silver Nanoparticles: Evidence for Nanoparticle−Cell Proximity Effects Khanh An Huynh,† J. Michael McCaffery,‡ and Kai Loon Chen*,† †

Department of Geography and Environmental Engineering, Johns Hopkins University, Baltimore, Maryland 21218-2686, United States ‡ The Integrated Imaging Center, Department of Biology, Johns Hopkins University, Baltimore, Maryland 21218-2686, United States S Supporting Information *

ABSTRACT: Heteroaggregation between silver nanoparticles (AgNPs) and naturally occurring colloids is a critical process that controls the environmental impacts of AgNPs. In this study, the effect of heteroaggregation between citrate-coated AgNPs and hematite nanoparticles (HemNPs) on the antimicrobial activity of AgNPs toward Escherichia coli is investigated. Our experiments are conducted at a AgNP concentration of 2.2 mg/L, at which the dissolved silver concentration of 19 μg/L in the bulk solution is not high enough to inhibit E. coli growth. At these AgNP and dissolved silver concentrations, the bacteria grew in samples containing AgNP−HemNP heteroaggregates while growth was completely inhibited in samples containing only AgNPs, indicating that heteroaggregation can reduce the antimicrobial activity of AgNPs. Furthermore, the degree of reduction in the AgNP antimicrobial activity is elevated when the HemNP/AgNP number concentration ratio is raised from 0.4 to 13.0. At a high HemNP/AgNP ratio (13.0), most AgNPs are observed through cryogenic transmission electron microscopy to attach to HemNPs, and some are surrounded by multiple HemNPs. These observations suggest that heteroaggregation can decrease AgNP antimicrobial activity by physically preventing AgNPs from coming into direct contact or close proximity with E. coli cells.

■

to the ubiquity of NOCs.24−26 Thus, heteroaggregation is expected to be a key process that controls the environmental fate, transport, and toxicity of AgNPs.27 The effect of heteroaggregation on the antimicrobial activity of AgNPs, however, has not been investigated to date. In light of recent revelations that the antimicrobial properties of AgNPs are strongly dependent on the proximity between AgNPs and cells,21−23 it is conceivable that heteroaggregation may prevent the close proximity between AgNPs and bacteria and hence diminish the cytotoxicity of the nanoparticles. In this paper, we show for the first time that heteroaggregation between citrate-coated AgNPs and hematite nanoparticles (HemNPs), a model NOC, reduces the antimicrobial activity of AgNPs toward E. coli. Specifically, the bacteria were exposed to heteroaggregates composed of AgNPs and HemNPs, and the antimicrobial activity of AgNPs was found to depend on the conformation of these heteroaggregates. The structures of AgNP−HemNP heteroaggregates were also examined and quantified by cryogenic transmission electron microscopy (cryo-TEM), and a mechanism is proposed for the reduction of AgNP antimicrobial effects through heteroaggregation. The findings from this study will have important implications not only for the prediction of the

INTRODUCTION Through the use and disposal of silver-containing products, silver nanoparticles (AgNPs) are likely to be released into the environment1−5 and thus may cause adverse effects on the environment and on human health.6,7 While various mechanisms have been proposed for the cytotoxicity of AgNPs,6,8−17 the dissolution of AgNPs resulting in the release and cellular uptake of Ag+ ions has been shown to be a key mechanism for the inactivation of microorganisms.6,11,14,16,17 Once Ag+ ions are released,16,18,19 they can bind to proteins and DNA inside cells and/or on cell membranes.6,9,20 These interactions can affect cellular processes, such as respiration and division, and thus eventually lead to cell death.6,9,20 Recently, the close proximity between AgNPs and bacterial cells has been discovered to be a critical factor that controls the efficacy of AgNPs in inactivating bacteria.21−23 Bondarenko et al.21 demonstrated that the toxicity of AgNPs was significantly reduced when direct contact between AgNPs and Escherichia coli in a suspension was prevented by using a dialysis membrane.21 McQuillan et al.22 also postulated that the dissolution of AgNPs that were attached on the cell membranes resulted in a locally high Ag+ concentration at the nanoparticle−cell interface and hence caused the inactivation of the E. coli cells.22 When AgNPs are released into natural aquatic systems, they likely undergo heteroaggregation (aggregation between two types of colloids) with naturally occurring colloids (NOCs) due © XXXX American Chemical Society

Received: March 18, 2014 Accepted: August 1, 2014

A

dx.doi.org/10.1021/ez5002177 | Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX

Environmental Science & Technology Letters

Letter

Figure 1. (a) Growth curves of E. coli in the presence of AgNPs and dissolved silver. All the experiments were performed at least three times. Error bars represent standard deviations. (b) Representative heteroaggregation profile obtained from a binary suspension containing both AgNPs (2.6 mg/ L) and HemNPs (6.0 mg/L) resulting in a HemNP/AgNP number concentration ratio of 2.2. Also presented are homoaggregation profiles obtained from suspensions containing either AgNPs (2.6 mg/L) or HemNPs (6.0 mg/L). (c) Representative heteroaggregation profiles obtained at different HemNP/AgNP ratios. (d) Expected predominant heteroaggregate structures at different HemNP/AgNP ratios.

nm, respectively. The electrophoretic mobilities of AgNPs and HemNPs in deionized (DI) water (Millipore) with no pH adjustment (about pH 5.5) were −2.68 × 10−8 and 1.11 × 10−8 m2 V−1 s−1, respectively (ZetaPALS, Brookhaven). Details are provided in the Supporting Information. Determination of AgNP and Dissolved Silver Concentrations. The concentrations of the AgNP and dissolved silver in any AgNP suspension were determined using centrifugal filtration and ICP-MS. Details are provided in the Supporting Information. Heteroaggregation of AgNPs and HemNPs. The heteroaggregation of AgNPs and HemNPs was investigated by preparing binary suspensions composed of citrate-coated AgNPs and HemNPs in DI water (about pH 5.5). Because AgNPs and HemNPs were oppositely charged under such conditions, they underwent favorable heteroaggregation. For the heteroaggregation experiments, the AgNP concentration was fixed at 2.6 mg/L (1.75 × 1012 particles/L) and three HemNP concentrations (1.5, 6.0, and 20.0 mg/L or 9.66 × 1011, 3.86 × 1012, and 1.29 × 1013 particles/L, respectively) were used. These concentrations were chosen in order to allow for heteroaggregation to take place at low, medium, and high HemNP/AgNP number concentration ratios or HemNP/ AgNP ratios (i.e., 0.6, 2.2, and 7.4, respectively), which could potentially result in different degrees of heteroaggregation.32 During the heteroaggregation process, the intensity-weighted hydrodynamic diameters of the heteroaggregates were meas-

ecological impacts of AgNPs and other inorganic nanoparticles that can dissolve in aqueous systems but also for the application of these nanomaterials. Importantly, heteroaggregation with NOCs can greatly reduce the adverse impacts of AgNPs on microorganisms in natural and engineered aquatic environments. At the same time, when AgNPs are utilized in environmental and biomedical applications, their heteroaggregation with other colloids should be prevented to maximize their antimicrobial activity.

■

MATERIALS AND METHODS

Preparation of Nanoparticles. Unmodified AgNPs were synthesized by ultrasonicating a mixture of Tollens’ reagent and glucose.28 To prepare citrate-coated AgNPs, the unmodified AgNPs were cleaned and resuspended in a 1 μM sodium citrate solution. The concentrations of AgNPs and dissolved silver in the stock suspension were determined to be 58.8 mg/L and 171 μg/L, respectively, through inductively coupled plasma mass spectrometry (ICP-MS, PerkinElmer Elan DRC II). HemNPs were synthesized through the forced hydrolysis of FeCl3.29−31 By gravimetric analysis, the mass concentration of HemNPs in the stock suspension was determined to be 4.4 g/ L. Our previous studies28,32 have shown, through TEM imaging, that both AgNPs and HemNPs synthesized using these methods were mostly spherical. The hydrodynamic diameters of AgNPs and HemNPs determined by dynamic light scattering (DLS, BI-200SM, Brookhaven) were 64.6 and 82.4 B

dx.doi.org/10.1021/ez5002177 | Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX

Environmental Science & Technology Letters

Letter

Figure 2. (a) Increase in optical density of bacterial suspensions at 600 nm after 8 h of incubation in the presence of heteroaggregates formed at different HemNP/AgNP ratios. Control suspension did not contain AgNPs or HemNPs. (b) Representative cryo-TEM images of heteroaggregates formed at a HemNP/AgNP ratio of 13.0. Arrows point to AgNPs. Heteroaggregates were prepared in DI water at AgNP and HemNP concentrations of 2.6 and 35.3 mg/L, respectively. (c) Dissolved silver concentrations in DM medium at the beginning (t = 0 h) and end (t = 8 h) of incubation in the absence of bacteria. (d) Proposed effects of heteroaggregation on antimicrobial activity of AgNPs. In panels (a) and (c), the AgNP number concentration was fixed at 1.49 × 1012 particles/L (or 2.2 mg/L), while the number concentrations of HemNPs were varied. All the experiments were performed at least three times. Error bars represent standard deviations. Asterisks (∗) indicate statistical significance (p < 0.05) between samples containing different HemNP concentrations.

nanoparticles, during incubation.33−35 The AgNP−HemNP heteroaggregates were prepared in DI water at the desired HemNP/AgNP ratios 15 min before they were incubated with the bacteria in DM medium. Separate experiments showed that both nanoparticles do not undergo significant homoaggregation in DM medium. Heteroaggregation was also much slower in the DM medium compared to DI water. The adsorption of citrate on HemNPs likely resulted in the charge reversal of the nanoparticles and hence in the slow heteroaggregation between both nanoparticles. During incubation, the optical density of the bacterial suspensions at a wavelength of 600 nm (OD600) was measured hourly using a UV−vis spectrophotometer (UV 1800, Shimadzu). All the experiments were performed at least three times. Additionally, the concentrations of viable bacterial cells in the suspensions after 2 h of incubation with AgNP−HemNP

ured every 15 s through time-resolved DLS at a scattering angle of 90° and a wavelength of 488 nm. Details are provided in the Supporting Information. Evaluation of Effects of Heteroaggregation on Antimicrobial Activity of AgNPs. E. coli K-12 bacteria (ATCC 25404) were used for the investigation of the effects of heteroaggregation on the antimicrobial activity of AgNPs. The bacteria were incubated with (i) AgNPs (2.2 mg/L), (ii) heteroaggregates composed of AgNPs (2.2 mg/L) and HemNPs (1, 5, and 30 mg/L), and (iii) AgNO3 (20, 31, and 43 μg/L) in a phosphate-reduced Davis minimal medium33−35 (herein referred to as DM medium, composition in Supporting Information) for 8 h at 37 °C and 250 rpm shaking speed. The DM medium was used because it has a relatively low ionic strength (about 43 mM) and can minimize the homoaggregation of the nanoparticles, as shown for C60 and TiO2 C

dx.doi.org/10.1021/ez5002177 | Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX

Environmental Science & Technology Letters

Letter

was raised to 2.2, conversely, the hydrodynamic diameter increased quickly. At a high ratio of 7.4, the hydrodynamic diameter increased to 134 nm after 6.5 min and then reached a plateau. The predominant structures of the heteroaggregates formed at these HemNP/AgNP ratios are expected to be similar to heteroaggregates composed of oppositely charged colloids37−39 (Figure 1d). At low HemNP/AgNP ratios (7.4), HemNPs outnumbered AgNPs, thus likely favoring the formation of small stable heteroaggregates that were composed of several HemNPs attached to each AgNP.38 On the basis of these results, HemNP/AgNP ratios of 0.4, 2.2, and 13.0 were employed to evaluate the effects of heteroaggregation on the antimicrobial activity of AgNPs. The AgNP concentration used was 2.2 mg/L, while the HemNP concentrations were 1.0, 5.0, and 30.0 mg/L. Additional experiments were conducted to verify that HemNPs even at the highest concentration of 30.0 mg/L did not inhibit the growth of the E. coli cells (Figure S3, Supporting Information). The increases in the optical density (ΔOD600) of the bacterial suspensions in the presence of the AgNP− HemNP heteroaggregates after 8 h of incubation are shown in Figure 2a. These data were extracted from the growth curves presented in Figure S4 of the Supporting Information. By performing unpaired one-sided t-tests with a significance level of 0.05 on the ΔOD600 data obtained at different HemNP/ AgNP ratios, it was found that heteroaggregation at the high and low HemNP/AgNP ratios of 13.0 and 0.4, respectively, reduced the antimicrobial activity of AgNPs as the ΔOD600 after 8 h of incubation at these ratios were higher than that in the absence of HemNPs (P-value = 0.012 and 0.009, respectively) (Figure 2a). Due to the higher variability of the experimental data at the medium HemNP/AgNP ratio of 2.2, heteroaggregation at this ratio was found to reduce the AgNP antimicrobial activity at a significance level of 0.12. A similar experiment with a different batch of AgNP stock suspensions also showed that heteroaggregation reduced the antimicrobial properties of AgNPs (Figure S5, Supporting Information). In addition, Figure 2a shows that heteroaggregation at the high HemNP/AgNP ratio of 13.0 resulted in a larger reduction in the antimicrobial activity of AgNPs than heteroaggregation at the low HemNP/AgNP ratio of 0.4 (P-value = 0.015). Furthermore, an increasing trend was observed between the number of viable bacterial cells in the heteroaggregate suspensions after 2 h of incubation and the HemNP/AgNP ratio (Figure S6b, Supporting Information). This result is in agreement with the results obtained from OD600 measurements after 2 h of incubation (Figure S6a, Supporting Information). Heteroaggregation Inhibits Direct Contact or Close Proximity between AgNPs and Bacterial Cells. Because the antimicrobial activity of AgNPs was considerably reduced in the presence of heteroaggregates formed at a high HemNP/ AgNP ratio, cryo-TEM was employed to examine the heteroaggregate structures at the ratio of 13.0. Representative heteroaggregates observed under cryo-TEM are presented in Figure 2b. AgNPs are differentiated from HemNPs in the TEM images by the presence of crevices on the surface of

heteroaggregates were determined through a spread plating technique. Details are provided in the Supporting Information. Cryo-TEM Imaging of Heteroaggregates. The AgNP− HemNP heteroaggregates were prepared in DI water for about 25 min before being observed through cryo-TEM. The procedure is described elsewhere32 and is also provided in the Supporting Information.

■

RESULTS AND DISCUSSION AgNPs Completely Inhibit Bacterial Growth at SubLethal Concentration of Dissolved Silver in Bulk Solution. The growth curves of E. coli in the absence and presence of 2.2 mg/L citrate-coated AgNPs are presented in Figure 1a. This AgNP concentration was the lowest nanoparticle concentration that can completely inhibit bacterial growth (Figure S1, Supporting Information). The initial dissolved silver concentration in the DM medium containing 2.2 mg/L AgNPs (t = 0 h) was determined to be 19 ± 4 μg/L (expressed as mass of Ag). In a separate experiment, 2.2 mg/L AgNPs was incubated in a DM medium that did not contain any E. coli cells. After 8 h of incubation, the dissolved silver concentration was 18 ± 6 μg/L, which was similar to the initial dissolved silver concentration (19 μg/L) thus indicating that AgNPs did not to undergo observable dissolution during the 8 h of incubation. Citrate in the DM medium was likely to adsorb on the AgNP surface and inhibit nanoparticle dissolution.36 To determine whether the bacterial growth was inhibited by the dissolved silver present in the DM medium containing AgNPs, additional E. coli growth curves were obtained in the presence of AgNO3 at dissolved silver concentrations of 20, 31, and 43 μg/L (Figure 1a). Procedures for determining these concentrations are presented in the Supporting Information. The results showed that the dissolved silver concentration of 20 μg/L was not high enough to retard bacterial growth while growth was completely inhibited at 43 μg/L. Therefore, the complete inhibition of bacterial growth observed in the presence of 2.2 mg/L AgNPs cannot be solely explained by the dissolved silver in the DM media (19 μg/L). Heteroaggregation with HemNPs Can Reduce Antibacterial Activity of AgNPs. Figure 1b presents the hydrodynamic diameters of AgNPs and HemNPs when prepared separately in DI water, as well as the hydrodynamic diameters when AgNPs and HemNPs were mixed in DI water at a HemNP/AgNP ratio of 2.2. When AgNPs and HemNPs were prepared separately, their hydrodynamic diameters remained stable with time. Because both nanoparticles are charged, they were stable to homoaggregation due to electrostatic repulsion.28,32 Conversely, when AgNPs and HemNPs were mixed to form a binary suspension, the hydrodynamic diameter increased quickly. Because both nanoparticles are oppositely charged, they underwent fast heteroaggregation through electrostatic attraction. The hydrodynamic diameters measured during the heteroaggregation of AgNPs and HemNPs at HemNP/AgNP ratios of 0.6, 2.2, and 7.4 are presented in Figure 1c and Figure S2 of the Supporting Information. At these HemNP/AgNP ratios, the amount of light scattered by AgNPs was comparable to that by HemNPs (Table S3, Supporting Information). Hence, the occurrence of heteroaggregation between AgNPs and HemNPs will result in an observable increase in the intensity-weighted hydrodynamic diameter. At a low HemNP/AgNP ratio of 0.6, the hydrodynamic diameter increased slowly. When the ratio D

dx.doi.org/10.1021/ez5002177 | Environ. Sci. Technol. Lett. XXXX, XXX, XXX−XXX

Environmental Science & Technology Letters

Letter

HemNPs32,40 while AgNPs were relatively smooth.28 Additionally, because silver is more electron-dense than iron due to its higher atomic number (47 vs. 26), AgNPs appeared significantly darker than the larger (and thicker) HemNPs. In evaluating nearly 40 heteroaggregates, it was found that one AgNP could associate with 1 (∼24%), 2 (∼50%), 3 (∼24%), and 4 (