Effects of Soluble Cadmium Salts Versus CdSe Quantum Dots on the

Feb 27, 2009 - ... Randall E. Mielke, Samuel M. Webb, Christopher Ehrhardt, Jin Ping .... Lauren M. Tom , Yvette M. Piceno , Gary L. Andersen , and Pa...
0 downloads 0 Views 3MB Size
Download PDF
Environ. Sci. Technol. 2009, 43, 2589–2594

Effects of Soluble Cadmium Salts Versus CdSe Quantum Dots on the Growth of Planktonic Pseudomonas aeruginosa JOHN H. PRIESTER,† PETER K. STOIMENOV,‡ RANDALL E. MIELKE,§ SAMUEL M. WEBB,| CHRISTOPHER EHRHARDT,⊥ JIN PING ZHANG,# GALEN D. STUCKY,‡ A N D P A T R I C I A A . H O L D E N * ,† Donald Bren School of Environmental Science & Management, Department of Chemistry and Biochemistry, Earth Sciences, and Materials Research Laboratory, University of California, Santa Barbara, California 93106, Center for Life Detection, Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California 91109, and Stanford Synchrotron Radiation Laboratory, Stanford Linear Accelerator Center, Menlo Park, California 94025

Received October 3, 2008. Revised manuscript received January 15, 2009. Accepted January 22, 2009.

With their increased use, engineered nanomaterials (ENMs) will enter the environment where they may be altered by bacteria and affect bacterial processes. Metallic ENMs, such as CdSe quantum dots (QDs), are toxic due to the release of dissolved heavy metals, but the effects of cadmium ions versus intact QDs are mostly unknown. Here, planktonic Pseudomonas aeruginosa PG201 bacteria were cultured with similar total cadmium concentrations as either fully dissolved cadmium acetate (Cd(CH3COO)2) or ligand capped CdSe QDs, and cellular morphology, growth parameters, intracellular reactive oxygen species (ROS), along with the metal and metalloid fates were measured. QDs dissolved partially in growth media, but dissolution was less in biotic cultures compared to sterile controls. Dose-dependent growth effects were similar for low concentrations of either cadmium salts or QDs, but effects differed above a concentration threshold of 50 mg/L (total cadmium basis) where (1) the growth of QD-treated cells was more impaired, (2) the membranes of QD-grown cells were damaged, and (3) QD-grown cells contained QD-sized CdSe cytoplasmic inclusions in addition to Se0 and dissolved cadmium. For most concentrations, intracellular ROS were higher for QDversus cadmium salts-grown bacteria. Taken together, QDs were * Corresponding author e-mail: [email protected]; tel: 805893-3195; fax: 805-893-7612. † Donald Bren School of Environmental Science & Management, University of California, Santa Barbara. ‡ Department of Chemistry and Biochemistry, University of California, Santa Barbara. § Center for Life Detection, Jet Propulsion Laboratory, California Institute of Technology. | Stanford Synchrotron Radiation Laboratory, Stanford Linear Accelerator Center. ⊥ Earth Sciences, University of California, Santa Barbara. # Materials Research Laboratory, University of California, Santa Barbara. 10.1021/es802806n CCC: $40.75

Published on Web 02/27/2009

 2009 American Chemical Society

more toxic to this opportunistic pathogen than cadmium ions, and were affected by cells through QD extracellular stabilization, intracellular enrichment, and cell-associated decay.

Introduction The rapid development of the engineered nanomaterials (ENMs) industry has raised concerns about ENM releases to the environment (1) where bacteria are abundant (2) and can catalyze essential nutrient-recycling reactions during growth (3) and influence ENM fates (4). Reported individual ENM-bacterial biophysical interactions include biosorption, ENM breakdown (5), and cellular uptake (6), with effects including membrane damage and toxicity (7, 8). However, such interactions are rarely evaluated in concert and over growth-associated time scales. This leaves questions about ENM fates and effects when bacteria are present, including the quantitative importance to toxicity end points of intact ENMs versus breakdown products. For heavy metal-composed ENMs such as cadmium selenide (CdSe) quantum dots (QDs), toxic metal ions may cause cellular toxicity (9). CdSe QDs are fluorescent semiconductor ENMs of interest in photovoltaics (10) and in diagnostics for stably labeling mammalian cells (11) and bacteria (12). CdSe QDs are often capped with ZnS to enhance fluorescence (13); core-shell configurations also stabilize Cd(II) surface atoms against dissolution which increases biocompatibility (9, 14). Still, there are concerns about cap loss (9, 15) and the subsequent toxicity of bare QDs and dissolved cadmium. Cd(II) causes cellular toxicity by several mechanisms including interfering with DNA repair (16) and metabolic proteins (17), membrane lipid peroxidation (18), substitution for physiological Zn(II), and reactive oxygen species (ROS) formation (19). In gram-negative bacteria, Cd(II) readily enters the cell through “open gate” Mg(II) transporters. Efflux through cadmium-induced membrane proteins is the primary resistance mechanism (20). QDs also generate ROS, which damage membranes (21); such damage may account for QDs nonspecifically entering bacteria in the dark (22) and in the light (12). Once in cells, QD toxicity may again be from either intact nanoparticles (12) or from released Cd(II) ions (23). However, still unknown are the relative toxicities and fates of Cd(II) ions versus either absorbed or assimilated intact QDs, especially when they co-occur in the bacterial growth environment. The objectives of this study were to quantify the effects to bacteria of ligand capped CdSe QDs and Cd(II) ions, taking into account that Cd(II) ions might be generated during bacterial growth if QDs dissolve, and that QDs may also be directly toxic. We asked: to what extent do QDs dissolve in bacterial culture? Are QDs and/or Cd(II) ions assimilated by cells? What is the toxicity of Cd(II) ions versus QDs? Assuming that QDs would be toxic due to dissolved cadmium, we grew a relatively cadmium-tolerant bacterial strain, Pseudomonas aeruginosa PG201, with measurably high total cadmium that was initially in the form of either Cd(II) ions or QDs. Citrate capped, as opposed to core-shell (i.e., metal capped), QDs were used to increase the likelihood that both Cd(II) and QDs would coexist in culture, thus providing the opportunity to differentiate ENM effects from heavy metal toxicity. It has been previously reported (15, 21) that capped QDs with no dissolved Cd(II) cause toxicity, but the relative effects of both intact QDs and dissolved Cd(II) have not been evaluated until now. Bacteria were also grown with soluble cadmium salts to separately quantify the effects of Cd(II) ions. While VOL. 43, NO. 7, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2589

CdSe QDs did dissolve in bacterial growth media, dissolution was incomplete. After accounting for effects of Cd(II) ions, the residual toxicity from intact QDs was relatively greater than toxicity from cadmium salts alone. This work thus reveals the separable effects of QDs and their released heavy metals, and newly evaluates the effects of ENMs on an important opportunistic pathogen.

Experimental Section Chemicals and Bacterial Culturing. Pseudomonas aeruginosa PG201, a well-studied environmental strain (24), was cultured with either Cd(II) or CdSe QDs in Luria Bertrani (LB) broth. All chemicals were reagent grade or better (Sigma Chemical, St. Louis, MO; and Fisher Scientific, Hampton, NH). See Supporting Information for QD synthesis and culture details. Cultures were amended with either cadmium acetate (Cd(CH3COO)2 at 5, 10, 20, 37.5, 75, 115, and 150 mg/L as cadmium) or CdSe QDs (10, 20, 37.5, 50, 75, 100 and 125 mg/L as cadmium). Controls included medium-only (no cadmium) and uninoculated versions of each treatment. Five independent replicates were prepared for each treatment and control. Cultures were prepared in 200 µL volumes in 96-well microplates (see Supporting Information for details). At 6 h after inoculation, 3 replicates were subsampled by aseptically removing 2 µL for Cd(II) ion quantification (see below). To test the effects of citrate on growth (at concentrations at and above those present with the QDs), an additional treatment involved amending LB broth with 400, 800, or 1200 mg/L sodium citrate. Growth experiments with Cd(CH3COO)2 and CdSe QDs were repeated independently using larger volumes (7.5 mL) and selected cadmium concentrations (with at least 3 replicates) for microscopy, analyses of intra- and extracellular metals and macromolecules, and analysis of intracellular ROS. The optical density (OD) was monitored (600 nm) to determine if experimental scale-up created bias. Another growth experiment was for assaying acidification by measuring the pH at 0 and 24 h for the control and the Cd(CH3COO)2 and CdSe QD treatments amended with 75 mg/L total cadmium. Separate 7.5 mL cultures were also grown to study selenium-only effects using sodium selenite (Na2SeO3) over a concentration range up to 500 mg/L Se(IV). Cell Harvesting. To determine the distribution of Cd(II) ions versus intact QDs in bacterial culture, late exponential phase (24 h) cultures of the 75 mg/L (total cadmium) Cd(CH3COO)2 and CdSe QD treatments were harvested for quantifying total cellular metal and metalloid contents by inductively coupled plasma atomic emission spectroscopy (ICP-AES), analyzing Se oxidation state by X-ray absorption near edge spectra (XANES), electron microscopy, epifluorescence microscopy, and for determining the crystal structure of intracellular metal and metalloid by X-ray diffraction (XRD). Intracellular biomacromolecules were also assayed (see Supporting Information). Intracellular ROS. Total ROS in abiotic CdSe and cadmium acetate solutions, and in midexponential phase P. aeruginosa cultures, were quantified using the 2′,7′-dichlorofluorescein diacetate (DCFH-DA) assay (25, 26) (see Supporting Information for assay details). Total and Dissolved Cadmium Quantification. Dissolved cadmium was quantified by the Measure-iT kit (Invitrogen, Carlsbad, CA). Calibration standards were prepared using Cd(CH3COO)2 (0, 25, 50, 75, and 100 mg/L as cadmium) in all relevant aqueous conditions: nanopure H2O, sterile LB, and filter-sterilized (0.2 µm) late exponential phase (24 h) culture supernatant. ICP-AES with a TJA High Resolution IRIS instrument (Thermo Electron Corporation, Waltham, MA) was used to quantify total selenium and cadmium. 2590

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 7, 2009

Instrument calibration was against commercial Cd and Se standards (Sigma Chemical). QD Integrity by XRD and XANES; and Abiotic Dissolution. Abiotic QD dissolution (see Supporting Information) was measured. QD intracellular integrity was inferred in two ways: using X-ray diffraction for cell-associated QD crystal structure, and using XANES for cellular Se oxidation state. Se-K-edge XANES spectra were collected at the Stanford Synchrotron Radiation Laboratory (SSRL) beam line 11-2 under SPEAR3 to determine the oxidation state of Se, using overnight-shipped (dry ice) triplicate cell pellets from a CdSe QD (75 mg/L) treatment for 24 h cultures as described above. Detailed descriptions of the XANES and XRD methods are located in the Supporting Information. Microscopy and Image Analysis. Phase-contrast microscopy (Nikon E800 at 1000 × total magnification, with image acquisition) was used for measuring (in micrographs using Photoshop 5.5) the cell aspect ratios of ten cells per treatment for culture aliquots dispensed directly onto microscope slides. Total cell counts were determined by epifluorescence microscopy of cells from separately grown late exponential phase cultures (LB controls and 75 mg/L total cadmium as either QDs or Cd(II) ions) that were SYBR gold-stained (Invitrogen, Carlsbad, CA) and counted as before (27). High-resolution microscopy was used to assess membrane integrity and metal, metalloid, and QD localization in cells. Scanning transmission electron microscopy (STEM) and energy dispersive X-ray analysis (EDXA) were performed using 24 h cultures for the 75 mg/L (total cadmium) QD and Cd(II) treatments plus controls. See the Supporting Information for sample preparation and instrument details. STEM micrographs were analyzed in Adobe Photoshop for cell and nanoparticle dimensions, nanoparticle counts, and membrane damage frequency. Cell dimensions were measured from longitudinal and vertical cross sections (5 each), and cell surface areas and volumes were calculated assuming cylindrical geometry. Data Analysis. Cellular Se and Cd contents were normalized to cell counts as before (27). Statistical analyses were performed with SPSS 12.0.1 (SPSS Inc., Chicago, IL) or Microsoft Excel 2000 software. Means were compared by the Student t test. Where applicable, standard errors were propagated according to standard methods (28).

Results Bacterial Growth and Relationship to QD Dissolution. Growth of P. aeruginosa PG201 was inhibited by both Cd(CH3COO)2 (Figure S1) and CdSe QDs (Figure S2) in that increased total cadmium resulted in longer lag times, lower specific growth rates, and lower yields (Tables S1, S2). Growth parameters appeared related to total cadmium concentration similarly for QD-treated and Cd(CH3COO)2-treated cultures (Tables S1 and S2). The pH was constant during growth and averaged 7.4 ( 0.1. Neither sodium citrate (Figure S3), nor sodium selenite (data not shown) inhibited growth. Because growth with either QDs or Cd(CH3COO)2 appeared similar (Figure S1 and S2; Tables S1 and S2), complete QD dissolution was implied. Still, several tests were performed to quantify QD dissolution: (1) a dialysis study of initial dissolution kinetics in water (Supporting Information 2.2; Figure S4); (2) dialysis studies to determine aqueous chemistry effects on 24 h dissolution end points (Supporting Information 2.2; Figure S5); and (3) Cd(II) ion quantification in cultures at 6 h, i.e., entry into exponential phase for most cultures. Most relevant to growth, QDs were between 25 and 50% dissolved in cultures at 6 h (Figures S1 and S2 early exponential phase), and the relationship between the total and dissolved fraction was expectedly linear (Figure S6). Relating specific growth rates to the concentration of dissolved Cd(II) at 6 h (i.e., end of lag phase) revealed that

FIGURE 1. Specific growth rate of P. aeruginosa versus dissolved cadmium amended as either cadmium acetate (squares) or CdSe QDs (diamonds). As shown in the graph, all cadmium treatments result in reduced growth rates relative to the no-cadmium control (dark circle). At QD concentrations exceeding 50 mg/L (total cadmium basis), QDs remain relatively toxic while cultures appear to resist Cd(II) ions. QD- versus Cd(CH3COO)2-treated cultures were similarly affected at lower total cadmium concentrations (Figure 1). However, for dissolved cadmium concentrations exceeding approximately 20 mg/L (i.e., total QD cadmium 50 mg/L) growth rates of QD-treated cultures continued to decline steeply as a function of Cd(II) dose while growth rates for Cd(CH3COO)2-treated cultures were less sensitive to cadmium dose (Figure 1). When plotted against dissolved Cd(II), the other growth curve metrics of lag time and yield (maximum OD) also showed stronger effects with QDs in comparison to Cd(CH3COO)2 (Figure S7). Cellular Morphology and Estimated Intracellular Nanoparticles.Above50mg/Ltotalcadmium,QD-andCd(CH3COO)2grown cultures appeared to respond differently to the dissolved cadmium dose (Figure 1). Thus one concentration above this threshold was chosen for analyzing cellular size, morphology, and comparative metal, metalloid, and biomacromolecule contents for the QD and Cd(CH3COO)2 treatments. Using lower resolution (i.e., 1000×) phase contrast microscopy, QD-grown cells were shorter as indicated by lower aspect ratios (3.3 ( 0.2) compared to either cadmiumtreated (3.8 ( 0.3) or control (3.9 ( 0.2) cells. Using highresolution STEM, cells cultured with Cd(CH3COO)2 appeared similar to no-metal controls except for electron dense deposits in and near the membranes (Figure 2A vs B) that, by EDXA, were cadmium-rich (data not shown). Cells cultured with QDs also contained Cd- and Se-rich electron dense intracellular deposits (Figure S8) but were highly disfigured (Figure 2C). A total of 54 QD-grown and 46 Cd(CH3COO)2treated cells in STEM images were evaluated for membrane damage, as defined by holes, blebbing, or detachment of the plasma membrane from the cell wall (Figure S9). Most (81%; n ) 44) of the QD-treated and fewer (33%; n ) 15) of the Cd(CH3COO)2-treated cells had membrane damage (e.g., Figure 2). Blebbing (Figure S9B) was observed in QD-treated, but not for Cd(CH3COO)2-treated cells. Nanosized particles appeared throughout the QD-treated cells (Figure 2C), while in the Cd(CH3COO)2-treated cells particles were in the periplasmic space (Figure 2B). The mean particle diameters differed (t test, P ) 0.00), with 8.02 ( 0.24 nm and 14.99 ( 0.54 nm for the QD and Cd(CH3COO)2 treatments, respectively. Because the nanosized particles in the QD treatments were similar in size to the administered QDs (5 nm), these particles were counted, resulting in an

average of 1.87 ( 0.16 × 105 nanoparticles per cell. Intra- and extracellular quantum dot concentrations were calculated using the latter counts as well as the measured extracellular concentrations of total and dissolved cadmium (Supporting Information 2.3). No further evaluation was made of the cadmium-rich inclusions within Cd(CH3COO)2-treated cells. Metal, Metalloid, and Biomacromolecule Contents. For cultures grown with 75 mg/L cadmium, total intracellular cadmium (by ICP-AES) at 24 h (early stationary phase) was similar for QD- and Cd(CH3COO 2)2-treated cultures and averaged 0.15 ( 0.02 pg cell-1, which was approximately 4% of that administered. Assuming cell dimensions (from STEM images) and counts (from epifluorescent microscopy), the intracellular concentration of this dissolved Cd(II) was then approximately 335 g/L, which was 4460 times higher than the medium. Cells grown with QDs (75 mg/L) had an intracellular Se content (by ICP-AES) of 0.08 ( 0.01 pg cell-1, which appeared, by XANES, to be mainly Se0 or other reduced organoselenium compounds (Figure S10). By both ICP-AES and the Measure iT assay, all of the intracellular Cd(II) appeared to be dissolved (Table S3). However, a relatively weak XRD peak at 2θ ≈ 24° supported that cells contained at least some intact CdSe crystals (Figure S11). This peak is consistent with the dominant peak for CdSe QDs synthesized according to the same methods that we used (29). However, this peak was only slightly above background, and other identifying peaks were likely obscured by the dominant Al peaks. QD and Cd(CH3COO)2-treated cells had greater amounts of intracellular protein, DNA, and carbohydrates as compared to control cultures (Supporting Information 2.4). Also, compared to sterile controls, QDs in cultures were less dissolved (68.5% vs 59.7%, respectively) after 24 h, which implied some extracellular QD stabilization in cultures (Table S3). Intracellular ROS and Media ROS. After correcting for background fluorescence (from DCFH, but not from either cells or QDs which did not interfere), and converting DCFH fluorescence to H2O2 equivalents, neither QDs nor cadmium salts at 10 mg/L total cadmium resulted in measurable intracellular ROS (after 15 h of growth). However, at cadmium concentrations spanning 20 through 125 mg/L, cellular ROS was significantly greater for QD- versus Cd(II)-grown cells (Figure 3). Abiotic ROS generation was greater for QDs in LB than for QDs in water (detailed in Supporting Information 2.5). LB alone also generated ROS, but at a much lower concentration than for either LB or water plus QDs (Supporting Information 2.5). After correcting for background fluorescence (as above) and again converting fluorescence to H2O2 equivalents, sterile LB broth plus DCFH yielded an ROS concentration of 534 ( 183 mg/L H2O2. When added to either LB or to water, CdSe QDs also generated ROS abiotically, albeit to a much lesser extent in water as compared to LB broth (Figure S12) and to a much greater extent than LB alone (above). Cadmium salts added to either sterile LB or water did not result in ROS formation. The pattern of ROS versus cadmium concentration was similar when comparing the biotic (Figure 3) and abiotic (Figure S12) QD treatments, with the exception of the decrease at 125 mg/L cadmium for the abiotic treatments.

Discussion Prior reports for CdSe QDs suggest that surface cap and solution chemistry modulate toxicity from cadmium ions (9, 15). Yet ZnS-capped, i.e., core shell, CdSe (15) and CdTe (21) QDs that did not release cadmium ions were still toxic, and PEG-modified CdSe QDs were toxic to mammalian cells in a dose-dependent fashion corresponding to intracellularized QDs (23). In the environment, various abiotic and biotic fate-related processes affecting ENM phase distribution VOL. 43, NO. 7, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2591

FIGURE 2. STEM images of P. aeruginosa grown in LB either without metals (A), or with 75 mg/L total cadmium in the form of either cadmium acetate (B) or CdSe QDs (C), where the latter treatment is associated with cells exhibiting damaged membranes. Scale bar is 1 µm.

FIGURE 3. H2O2 equivalent intracellular reactive oxygen species (ROS) concentration for P. aeruginosa grown in LB with 75 mg/ L total cadmium in the form of either cadmium acetate or CdSe QDs. ROS is normalized to intracellular DNA to account for slight differences in culture yield. could enable exposure to Cd(II) ions and intact ENMs simultaneously. However, as emphasized in a recent review (30), the relative contributions to toxicity of co-occurring ENMs and their dissolved phases are mostly unknown. We addressed this issue by growing a model bacterial strain with either QDs or cadmium ions, separately studying QD dissolution and metal/metalloid fates with cells, and relating growth inhibition to the dissolved cadmium concentration. While aqueous phase cadmium speciation was not directly studied, published relationships encompassing the medium pH (7.4) and cadmium concentrations (e150 mg/L) studied here would suggest that dissolved cadmium was predominantly Cd(II) (31). Additionally, our report is one of the first of effects of ENMs on an opportunistic pathogen. CdSe QD toxicity to P. aeruginosa clearly exhibited two phases, i.e., a low concentration phase that mimicked the growth inhibition from cadmium salts, and a higher concentration phase that inhibited growth more extensively relative to cadmium salts. Previously, the minimum inhibitory concentration (MIC) for Cd(II) was reported as 500-700 mg/L for P. aeruginosa in LB (32, 33). In our study, above approximately 50 mg/L total cadmium, cells intoxicated by cadmium salts became resistant whereas cells treated with QDs did not, thus implying that toxicity from intact QDs was quantitatively important above this threshold. The dissolution data confirmed that both intact QDs and dissolved Cd(II) were co-occurring, and thus further examination of cellular effects and metalloid fates was performed for the 75 mg/L QD and cadmium salts treatment, i.e., for a concentration at which a nanoparticle2592

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 7, 2009

specific effect was clearly implied. Particularly with uncapped nanoparticles including bare CdSe QDs, and metal oxides such as ZnO and CuO (34) that can also easily dissolve, the approach used here for measuring dissolved cadmium, which was adopted from Cho et al. (35), was essential to revealing a potent “nanoparticle effect” of intact QDs. Selenium, added as sodium selenite, was not observed to be toxic to the strain studied here. Nevertheless, tracking the oxidation state of cell-associated selenium by XANES provided evidence of QD breakdown. It has been proposed (9) that oxidation of CdSe QD surfaces leads to QD dissolution and the formation of SeO2. Had a similar process occurred here with P. aeruginosa, a mixture of Se2- and Se4+ would have been observed by XANES. Instead, the predominant form of selenium appeared to be elemental selenium, or a mixture of elemental selenium and organoselenium compounds (Figure S10). It is possible that the above proposed process is an extracellular phenomenon, or that an undetermined biotic process converted SeO2 to elemental selenium. This has been observed with P. aeruginosa and Na2SeO3 (data not shown) but has yet to be evaluated with SeO2. At 75 mg/L total cadmium, ROS generation was enhanced for QD- versus Cd(II)-grown bacteria. ROS formation and subsequent cellular effects, an emerging paradigm for cellular toxicity by nanoparticles (36), have been used to explain toxicity from CdTe QD-treated mammalian cells (37) and E. coli bacteria (21). ROS formation does not require irradiation (21, 37) and occurs in rich growth media (38). In this study, the growth medium interacted with QDs to initiate ROS formation. However, QDs enhanced ROS concentrations significantly above the medium-only controls and relative to abiotic and biotic treatments with cadmium salts. ROS damaged membranes and facilitated QD entry, as described before (12), which may also explain a prior report of similarly formulated QDs entering E. coli (22). Membrane damage is widely observed with other ENMs (30) such as MgO (8), ZnO (6, 39), Ag (7, 40, 41), and single walled carbon nanotubes (SWNTs) (42). In many cases, ENMs associate with membranes (8, 30, 39) and direct contact is required for toxicity (5, 30, 42) because ROS is produced on the cell surface (21, 30). However, direct association of QDs with the cell envelope, a previously described prerequisite for ROS-mediated membrane damage by CdTe QDs (21), was not observed in our STEM images. While cadmium ions can enter bacteria passively (see Introduction), CdSe QDs are too large to diffuse into cells (12) and thus in the absence of receptor-specific QD bioconjugates (12), membrane damage can facilitate QD entry (30). Some QD entry is implied from the end points studied here, but the same end points could also arise from extracellular QD breakdown, uptake of metal/ metalloids, and then reformation of nanocrystals in cells. Still, when

comparing the estimated cadmium mass in visualized intracellular 8 nm CdSe particles against the 20-fold higher total cadmium concentration quantified by ICP-AES, few cellular QDs were intact. Extensively broken-down cellular QD material is also supported by the weak XRD signal, the XANES showing selenium as Se(0) or organoselenium compounds, the fact that the total and dissolved cellular cadmium levels were quantitatively the same, and the fact that overall cadmium enrichment in cells was 4460× when intact QD enrichment appeared to be 30×. Thus, we conclude that QDs enter cells where they mostly decompose, through oxidative, ROS-forming processes that enhance the toxicity of intracellularized QDs relative to intracellular cadmium. The impression here that QDs were more reactive, and thus perhaps more damaging inside versus outside of cells, is compelling and worthy of further evaluation. To evaluate, we compared the levels of ROS per intact QD under abiotic conditions (in culture media) versus the ROS per intracellular QD. Using the ROS data in Figure 3, the measured intracellular DNA (iDNA) per cell, and the estimated number of QDs per cell, we calculated (for the 75 mg/L cadmium concentration) that the intracellular ROS per cell-associated QD was approximately 6.4 × 10-13 mg/L H2O2 equivalents per QD. Then using dissolution data (Figure S6) to determine the amount of QD cadmium in LB, and the estimated 109 cadmium atoms per QD (see Results, and Kasuya et al. (43)) plus the ROS per QD (Figure S12), we calculated that there were 1.7 × 10-15 mg/L H2O2 equivalents per QD in sterile LB. Comparing the two numbers, it would appear that ROS per QD were nearly 400 times greater inside cells than outside. The two numbers could converge either if our estimate of intact intracellular QDs was low, or if intracellular Cd(II) also generated ROS, the latter of which is possible based on mammalian cell culture (44). But in our studies, intracellular ROS from Cd(II) ions was very low (Figure 3) and thus there were either more QDs in cells than we estimated from STEM images or QDs were relatively more potent toxicants once inside cells. Overall, a plausible scenario for the QDassociated toxicity observed here is that (1) cadmium ions induced efflux pump-mediated resistance that became overwhelmed, leading to intracellular Cd(II) ion accumulation, (2) QDs produced ROS extracellularly which damaged membranes and allowed QDs to enter cells, (3) once inside cells, QDs, and to a lesser extent Cd(II), continued to generate ROS, thus exacerbating toxicity processes that began extracellularly but also facilitated QD breakdown inside cells, and (4) membrane damage inhibited performance of higher concentration-related cadmium resistance functions, thus further impairing cells when intact QDs were present. Still unknown, but of considerable interest in future research, would be establishing how cells may participate in enhancing intracellular ROS production from QDs, and if the process also catalyzes QD decay in bacteria as observed here. The cellular fates of QDs occurring in this study (Figure 4) may be environmentally relevant, particularly as they were studied for an environmentally important opportunistic pathogen. Cadmium ions hyperaccumulated in cells and QDs appeared relatively broken down inside cells, but QDs were marginally stabilized against extracellular dissolution. As above, intact QDs were in higher concentration inside cells relative to outside (Figure 4). Many bacteria are known to hyperaccumulate heavy metals (45) including cadmium (46), and it has been suggested that metal-laden bacteria could mediate metal transport in the environment (47). In our prior work with Cr(VI) and P. putida biofilms, extracellular DNA and Cr(III) appeared to stabilize each another (27). Here, protein, total carbohydrate as well as intracellular DNA were elevated in bacteria grown with either QD or cadmium salts (see Supporting Information) which may imply biomacromolecular involvement in intracellular, as well as extracellular

FIGURE 4. Model of citrate-stabilized CdSe quantum dot (QD) interactions with planktonic P. aeruginosa. QDs dissolve partially extracellularly, but dissolution is greater in the absence of cells. QDs generate extracellular ROS that permeabilize membranes, facilitating cellular entry, wherein enhanced ROS production per QD is observed. QDs are enriched 30× in cells compared to the extracellular environment, but most QDs in cells decay to metal/metalloids and cadmium enrichment is 4460×. (48), metal hyperaccumulation. A plausible environmental scenario is that these bacteria stabilize QDs extracellularly, but intracellularly sequester QDs, cadmium, and selenium. These outcomes could reduce long-range transport of ENMs and their constituents in the environment, but might allow long-term effects on microbial communities (49-51).

Acknowledgments This research was supported by the U.S. EPA STAR Awards R831712 and R833323, by the University of California Toxic Substances Research and Training Program: Lead Campus program in Nanotoxicology, and by the Office of Science (BER), U.S. Department of Energy, Grant DE-FG0205ER63949. Publication of this material is supported by the National Science Foundation and the Environmental Protection Agency under Cooperative Agreement EF 0830117. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation or the Environmental Protection Agency. This work has not been subjected to EPA review and no official endorsement should be inferred. This work made use of UCSB MRL Central Facilities supported by the MRSEC Program of the National Science Foundation under award DMR05-20415. Portions of this research were carried out at the Stanford Synchrotron Radiation Laboratory, a national user facility operated by Stanford University on behalf of the U.S. Department of Energy, Office of Basic Energy Sciences.

Supporting Information Available Detailed descriptions of the materials and methods, as well as the complete results of the effects of CdSe QDs and cadmium acetate on growth, the abiotic QD dissolution, and biomacromolecule contents.This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Maynard, A. D. Nanotechnology: A Research Strategy for Addressing Risk; Woodrow Wilson International Center for Scholars: Washington, DC, July 2006; p 45. (2) Whitman, W. B.; Coleman, D. C.; Wiebe, W. J. Prokaryotes: the unseen majority. Proc. Natl. Acad. Sci. USA 1998, 95, 6578– 6583. (3) Schlesinger, W. H. Biogeochemistry: An Analysis of Global Change; Academic Press: San Diego, CA, 1997; p 588. VOL. 43, NO. 7, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

2593

(4) Wiesner, M. R.; Lowry, G. V.; Alvarez, P.; Dionysiou, D.; Biswas, P. Assessing the risks of manufactured nanomaterials. Environ. Sci. Technol. 2006, 40 (14), 4336–4345. (5) Thill, A.; Zeyons, O.; Spalla, O.; Chauvat, F.; Rose, J.; Auffan, M.; Flank, A. M. Cytotoxicity of CeO2 nanoparticles for Escherichia coli. Physico-chemical insight of the cytotoxicity mechanism. Environ. Sci. Technol. 2006, 40 (19), 6151–6156. (6) Brayner, R.; Ferrari-Iliou, R.; Brivois, N.; Djediat, S.; Benedetti, M. F.; Fievet, F. Toxicological impact studies based on Escherichia coli bacteria in ultrafine ZnO nanoparticles colloidal medium. Nano Lett. 2006, 6 (4), 866–870. (7) Pal, S.; Tak, Y. K.; Song, J. M. Does the antibacterial activity of silver nanoparticles depend on the shape of the nanoparticle? A study of the gram-negative bacterium Escherichia coli. Appl. Environ. Microbiol. 2007, 73 (6), 1712–1720. (8) Stoimenov, P. K.; Klinger, R. L.; Marchin, G. L.; Klabunde, K. J. Metal oxide nanoparticles as bactericidal agents. Langmuir 2002, 18 (17), 6679–6686. (9) Derfus, A. M.; Chan, W. C. W.; Bhatia, S. N. Probing the cytotoxicity of semiconductor quantum dots. Nano Lett. 2004, 4 (1), 11–18. (10) Huynh, W. U.; Dittmer, J. J.; Alivisatos, A. P. Hybrid nanorodpolymer solar cells. Science 2002, 295 (5564), 2425–2427. (11) Chan, W. C. W.; Nie, S. Quantum dot bioconjugates for ultrasensitive nonisotopic detection. Science (Washington, DC) 1998, 281 (5385), 2016–2018. (12) Kloepfer, J. A.; Mielke, R. E.; Nadeau, J. L. Uptake of CdSe and CdSe/ZnS quantum dots into bacteria via purine-dependent mechanisms. Appl. Environ. Microbiol. 2005, 71 (5), 2548–2557. (13) Hines, M. A.; Guyot-Sionnest, P. Synthesis and characterization of strongly luminescing ZnS-Capped CdSe nanocrystals. J. Phys. Chem. 1996, 100 (2), 468–471. (14) Ryman-Rasmussen, J. P.; Riviere, J. E.; Monteiro-Riviere, N. A. Surface coatings determine cytotoxicity and irritation potential of quantum dot nanoparticles in epidermal keratinocytes. J. Invest. Dermatol. 2007, 127 (1), 143–153. (15) Hardman, R. A toxicologic review of quantum dots: Toxicity depends on physicochemical and environmental factors. Environ. Health Perspect. 2006, 114 (2), 165–172. (16) Hartwig, A.; Asmuss, M.; Ehleben, I.; Herzer, U.; Kostelac, D.; Pelzer, A.; Schwerdtle, T.; Burkle, A. Interference by toxic metal ions with DNA repair processes and cell cycle control: Molecular mechanisms. Environ. Health Perspect. 2002, 110, 797–799. (17) Zimenkov, Y. V.; Salminen, A.; Efimova, I. S.; Lahti, R.; Baykov, A. A. Cd2+-induced aggregation of Escherichia coli pyrophosphatase. Eur. J. Biochem. 2004, 271 (14), 3064–3067. (18) Smeets, K.; Cuypers, A.; Lambrechts, A.; Semane, B.; Hoet, P.; Van Laere, A.; Vangronsveld, J. Induction of oxidative stress and antioxidative mechanisms in Phaseolus vulgaris after Cd application. Plant Physiol. Biochem. 2005, 43 (5), 437–444. (19) Nies, D. H. Microbial heavy-metal resistance. Appl. Microbiol. Biotechnol. 1999, 51 (6), 730–750. (20) Nies, D. H.; Silver, S. Ion Efflux Systems Involved In Bacterial Metal Resistances. J. Ind. Microbiol. 1995, 14 (2), 186–199. (21) Lu, Z. S.; Li, C. M.; Bao, H. F.; Qiao, Y.; Toh, Y. H.; Yang, X. Mechanism of antimicrobial activity of CdTe quantum dots. Langmuir 2008, 24 (10), 5445–5452. (22) Hirschey, M. D.; Han, Y. J.; Stucky, G. D.; Butler, A. Imaging Escherichia coli using functionalized core/shell CdSe/CdS quantum dots. J. Biol. Inorg. Chem. 2006, 11 (5), 663–669. (23) Chang, E.; Thekkek, N.; Yu, W. W.; Colvin, V. L.; Drezek, R. Evaluation of quantum dot cytotoxicity based on intracellular uptake. Small 2006, 2 (12), 1412–1417. (24) Steinberger, R. E.; Holden, P. A. Extracellular DNA in singleand multiple-species unsaturated biofilms. Appl. Environ. Microbiol. 2005, 71 (9), 5404–5410. (25) Foucaud, L.; Wilson, M. R.; Brown, D. M.; Stone, V. Measurement of reactive species production by nanoparticles prepared in biologically relevant media. Toxicol. Lett. 2007, 174 (1-3), 1–9. (26) Wang, H.; Joseph, J. A. Quantifying cellular oxidative stress by dichlorofluorescein assay using microplate reader. Free Radical Biol. Med. 1999, 27 (5-6), 612–616. (27) Priester, J. H.; Olson, S. G.; Webb, S. M.; Neu, M. P.; Hersman, L. E.; Holden, P. A. Enhanced exopolymer production and chromium stabilization in Pseudomonas putida unsaturated biofilms. Appl. Environ. Microbiol. 2006, 72 (3), 1988–1996. (28) Taylor, J. R., An Introduction to Error Analysis: The Study of Uncertainties in Physical Measurements, 2nd ed.; University Science Books: Sausalito, CA, 1997. (29) Rogach, A. L.; Kornowski, A.; Gao, M. Y.; Eychmuller, A.; Weller, H. Synthesis and characterization of a size series of extremely

2594

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 7, 2009

(30) (31) (32) (33) (34)

(35)

(36) (37)

(38) (39)

(40) (41) (42) (43)

(44)

(45) (46) (47) (48)

(49)

(50) (51)

small thiol-stabilized CdSe nanocrystals. J. Phys. Chem. B 1999, 103 (16), 3065–3069. Neal, A. L. What can be inferred from bacterium-nanoparticle interactions about the potential consequences of environmental exposure to nanoparticles? Ecotoxicology 2008, 17 (5), 362–371. Baes, C. F.; Mesmer, R. E. The Hydrolysis of Cations; John Wiley & Sons, Inc.: New York, 1976; p 489. Raja, C. E.; Anbazhagan, K.; Selvam, G. S. Isolation and characterization of a metal-resistant Pseudomonas aeruginosa strain. World J. Microbiol. Biotechnol. 2006, 22 (6), 577–585. Raja, C. E.; Sasikumar, S.; Selvam, G. S. Adaptive and cross resistance to cadmium(II) and zinc(II) by Pseudomonas aeruginosa BC15. Biologia 2008, 63 (4), 461–465. Heinlaan, M.; Ivask, A.; Blinova, I.; Dubourguier, H.-C.; Kahru, A. Toxicity of nanosized and bulk ZnO, CuO and TiO2 to bacteria Vibrio fischeri and crustaceans Daphnia magna and Thamnocephalus platyurus. Chemosphere 2008, 71 (7), 1308–1316. Cho, S. J.; Maysinger, D.; Jain, M.; Roder, B.; Hackbarth, S.; Winnik, F. M. Long-term exposure to CdTe quantum dots causes functional impairments in live cells. Langmuir 2007, 23 (4), 1974–1980. Nel, A.; Xia, T.; Madler, L.; Li, N. Toxic potential of materials at the nanolevel. Science 2006, 311 (5761), 622–627. Lovric, J.; Cho, S. J.; Winnik, F. M.; Maysinger, D. Unmodified cadmium telluride quantum dots induce reactive oxygen species formation leading to multiple organelle damage and cell death. Chem. Biol. 2005, 12 (11), 1227–1234. Halliwell, B. Oxidative stress in cell culture: an under-appreciated problem. FEBS Lett. 2003, 540 (1-3), 3–6. Zhang, L. L.; Jiang, Y. H.; Ding, Y. L.; Povey, M.; York, D. Investigation into the antibacterial behaviour of suspensions of ZnO nanoparticles (ZnO nanofluids). J. Nanoparticle Res. 2007, 9 (3), 479–489. Shrivastava, S.; Bera, T.; Roy, A.; Singh, G.; Ramachandrarao, P.; Dash, D. Characterization of enhanced antibacterial effects of novel silver nanoparticles. Nanotechnology 2007, 18 (22), 225103. Sondi, I.; Salopek-Sondi, B. Silver nanoparticles as antimicrobial agent: a case study on E. coli as a model for Gram-negative bacteria. J. Colloid Interface Sci. 2004, 275 (1), 177–182. Kang, S.; Pinault, M.; Pfefferle, L. D.; Elimelech, M. Single-walled carbon nanotubes exhibit strong antimicrobial activity. Langmuir 2007, 23 (17), 8670–8673. Kasuya, A.; Sivamohan, R.; Barnakov, Y. A.; Dmitruk, I. M.; Nirasawa, T.; Romanyuk, V. R.; Kumar, V.; Mamykin, S. V.; Tohji, K.; Jeyadevan, B.; Shinoda, K.; Kudo, T.; Terasaki, O.; Liu, Z.; Belosludov, R. V.; Sundararajan, V.; Kawazoe, Y. Ultra-stable nanoparticles of CdSe revealed from mass spectrometry. Nat. Mater. 2004, 3 (2), 99–102. Kim, S. J.; Jeong, H. J.; Myung, N. Y.; Kim, M. C.; Lee, J. H.; So, H. S.; Park, R. K.; Kim, H. M.; Um, J. Y.; Hong, S. H. The protective mechanism of antioxidants in cadmium-induced ototoxicity in vitro and in vivo. Environ. Health Perspect. 2008, 116 (7), 854– 862. Mullen, M. D.; Wolf, D. C.; Ferris, F. G.; Beveridge, T. J.; Flemming, C. A.; Bailey, G. W. Bacterial sorption of heavy metals. Appl. Environ. Microb. 1989, 55 (12), 3143–3149. Vecchio, A.; Finoli, C.; Di Simine, D.; Andreoni, V. Heavy metal biosorption by bacterial cells. Fresenius J. Anal. Chem. 1998, 361 (4), 338–342. Chen, J.-H.; Lion, L. W.; Ghiorse, W. C.; Shuler, M. L. Mobilization of adsorbed cadmium and lead in aquifer material by bacterial extracellular polymers. Water Res. 1995, 29 (2), 421–430. Lamelas, C.; Benedetti, M.; Wilkinson, K. J.; Slaveykova, V. I. Characterization of H+ and Cd2+ binding properties of the bacterial exopolysaccharides. Chemosphere 2006, 65 (8), 1362– 1370. Lorenz, N.; Hintemann, T.; Kramarewa, T.; Katayama, A.; Yasuta, T.; Marschner, P.; Kandeler, E. Response of microbial activity and microbial community composition in soils to long-term arsenic and cadmium exposure. Soil Biol. Biochem. 2006, 38 (6), 1430–1437. Wang, A. Y.; Chen, J.; Crowley, D. E. Changes in metabolic and structural diversity of a soil bacterial community in response to cadmium toxicity. Biol. Fertil. Soils 2004, 39 (6), 452–456. Renella, G.; Mench, M.; van der Lelie, D.; Pietramellara, G.; Ascher, J.; Ceccherini, M. T.; Landi, L.; Nannipieri, P. Hydrolase activity, microbial biomass and community structure in longterm Cd-contaminated soils. Soil Biol. Biochem. 2004, 36 (3), 443–451.

ES802806N