Grantee Research Project Results
Final Report: Transformations of Biologically-Conjugated CdSe Quantum Dots Released Into Water and Biofilms
EPA Grant Number: R831712Title: Transformations of Biologically-Conjugated CdSe Quantum Dots Released Into Water and Biofilms
Investigators: Holden, Patricia , Nadeau, Jay L.
Institution: University of California - Santa Barbara , McGill University
EPA Project Officer: Aja, Hayley
Project Period: October 1, 2004 through September 30, 2007
Project Amount: $332,099
RFA: Exploratory Research to Anticipate Future Environmental Issues: Impacts of Manufactured Nanomaterials on Human Health and the Environment (2003) RFA Text | Recipients Lists
Research Category: Nanotechnology , Safer Chemicals , Human Health
Objective:
Semiconductor nanocrystals (quantum dots) differ in important ways from bulk semiconductor materials. Their increased band gap means that they function as strong oxidizing and/or reducing agents, and their small size allows them to pass into living cells. Conjugation of biomolecules to the crystal surface can alter any or all of these properties. In preliminary experiments, we have observed that nucleobase-conjugated CdSe quantum dots were actively taken up by soil and aquatic bacteria (for example, Bacillus subtilis, and Escherichia coli). Effects on microbial viability attributed to the presence of the quantum dots included slower doubling times, heavy metal sequestration, and “blebbing” of metals into the environment. We propose here to quantify these effects using a variety of biologically-conjugated quantum dots and an assortment of microbial species, monitoring the process of quantum dot uptake and breakdown and characterizing the breakdown products that result from bacterial metabolism of these particles. Possible hazards to microbial populations with extrapolation to humans through contamination of soil and water with quantum dot breakdown products will be analyzed and quantified.
Summary/Accomplishments (Outputs/Outcomes):
Background: composition and use of QDs
Quantum dots (QDs) are photoluminescent semiconductor nanocrystals that promise an extraordinary advance in biological labeling at the nanoscale. Absorption spectra of QDs are broad and their emission spectra are narrow, allowing for multi-wavelength labeling that is easily distinguished (Fig. 1), and is controllable by varying the size of the particle. The outer surfaces of QDs can be readily bound to organic molecules1 2 3, resulting in a spectrum of biologically compatible labels that are all excited with a single wavelength. This feature has been used to create DNA microarrays4, to label fixed and permeabilized mammalian cells5, and to observe living slime molds6 and bacteria7.
Figure 1. CdSe QD photoluminescence in physiological saline at 25-37 °C. A, Visible absorbance spectra from 3 independent preparations of CdSe QDs. They are labeled according to their fluorescence emission (green, yellow, and red). B, Emission spectra of the samples in panel A, showing both the visible peaks and near-IR trap emission. Excitation, 400 nm. C, Epifluorescence image of triple labeling of S. aureus cultures with red, yellow, and green QDs (scale bar = 5μm). (All images adapted from data published in7).
The most common QD materials for biological use are cadmium and selenium, forming CdSe wurzite nanocrystals. These nanocrystals are often overlaid with another semiconductor material, such as ZnS or CdS, to increase the intensity of the light emission. Additionally, commercially available quantum dots are further coated with proprietary polymer and protein layers to render them biologically compatible and water-soluble (www.quantumdot.com Exit )(Fig. 2).
Figure 2. Chemical composition of the most commonly used QDs in biological applications. A, CdSe QDs (referred to as “bare QDs” in this proposal). Solubilization is accomplished by self-assembly of an alkanethiol (such as mercaptoacetic acid, shown) whose –SH bonds directly to the semiconductor, leaving the carboxylate group free to interact with aqueous solution. B, “Core-shell” QDs are overlaid with a 1-2 nm thick layer of ZnS or CdS. Solubilization is identical to the case of the bare crystals. In both cases, the free carboxylates can be covalently bonded to proteins or other organic molecules of choice. C, Commercially available QDs (Quantum Dot Corporation, Hayward, CA) consist of the CdSe/ZnS “core-shell” further coated with polymers and the protein streptavidin. Overall nanocrystal diameter is 2-to 3-fold that of the “core-shell” QDs.
Possible risks of QDs released into the environment
Obvious environmental risks of these particles are those associated with toxic concentrations of the heavy metals Cd, Se, and Zn. Cadmium has no known nutritional value to any living organisms, with the possible exception of a species of diatom8, and is considered a serious pollutant9. The primary health effect of cadmium exposure to humans is that it accumulates in the kidneys and causes renal tubular damage10. Selenium and zinc, although essential trace elements, can be toxic to birds and fish, as well as humans11, 12.
However, possible environmental risks of QDs arise not only from the heavy metals from which they are made, but from specific properties that are engendered by the nanometer scale of the particles. In particular, two properties of QDs are not possessed by bulk semiconductors: (1) their small size permits them to intercalate into or to cross cell membranes; (2) they are very strong oxidizing and reducing agents. Both of these issues were addressed in this research.
Because biological QDs are synthesized and stored in liquids, inhalation is of minor concern; greater risks may result from contamination of water and soil, from exposure to microorganisms, and by ingestion (either inadvertently or for medical procedures). QDs are light- and pH-sensitive and degrade rapidly into ionic components; such properties are well known7, 13 although studies of environmental effects have not been performed.
The laboratory of the Co-I Nadeau was the first to show that when living bacteria are exposed to QDs, internalization and/or degradation of the nanocrystals may occur even more rapidly than in ordinary solutions. The final endpoints were investigated in this research. In work performed and published at the very beginning of this project in the Nadeau lab, the rate and nature of the QD degradation depended on bacterial strain and metal-dependent enzymes; it did not occur with killed bacteria. Cells may take up Se as a nutrient and sequester or extrude Cd; often the toxic Cd is enclosed into “nanocrystals” of the cells’ own making. Similarly, bacteria exposed to Zn may create ZnS nanocrystals14, 15. These properties have been exploited to create semiconductor nanocrystals from microbial cultures16, but the environmental impact on soil or water bacteria which can be either planktonic or biofilm, or the impact on bacterial-eukaryotic symbioses, has not been addressed.
In this work, we studied solution-based and bacterially-assisted QD degradation; to quantify the products generated and their impact upon bacterial growth and metabolism; and to quantify the oxidative and reductive properties of QDs. Bare and core-shell CdSe QDs were used in the studies. For bacterial studies, the soil microbe Pseudomonas aeruginosa (Gram negative) was used as a model system in aqueous cultures and in biofilms. To adequately mimic the soil environments inhabited by bacterial biofilms, biofilm-QD interactions were studied under unsaturated conditions. Column studies, while proposed, were not performed.
Research Scope and Approaches
I. Originally proposed questions and approaches
In the research proposal, either explicitly or implicitly, the following questions and associated approaches were described:
- What are the abiotic breakdown products of QDs exposed to a variety of harsh conditions, including room light; UV light (354 nm); pH from 4-14; and oxidizing or reducing conditions (resulting either from addition of chemicals, such as hydrogen peroxide, or from application of potentials using a potentiostat), oxygen and temperature extremes?. The presence of degradation products in solutions was to be monitored by ICP (inductively coupled plasma emission spectrometry), transmission electron microscopy and environmental scanning electron microscopy with elemental analyses.
- What are the effects of QDs on bacterial doubling times under normal and starved conditions? Two aerobes were proposed for study: Pseudomonas aeruginosa PG201 (a Gram negative organism) and Staphylococcus aureus (strain was to be determined, a Gram positive organism) by monitoring growth vs. fluorescence with a spectrophotometer, direct microscopic counts, and plate counts. The ratio of fluorescently labeled to unlabeled bacteria was to be monitored until fluorescence disappeared, and the rate of disappearance was to be compared with that expected from simple dilution due to doubling.
- What is the nature of the QDs and QD material found inside bacteria that have taken up the particles? In one relevant study (Kloepfer and Nadeau, 2005), B. subtilis that had taken up adenine-QDs were lysed, and the fluorescence spectrum of the resulting particles was measured. While this revealed that at least some of the particles remained intact inside the cells, no measurements were done on particles that were not fluorescent or that could not be resolved by TEM. In this research, we proposed to measure the concentrations of Cd and Se inside the cytoplasm of washed and lysed cells, and to characterize membrane-associated QDs, which show a different fluorescence spectrum from that of cytoplasmic QDs (Nadeau and Kloepfer, 2005).
- What are the breakdown products, and are they released from cells before or after cell death? Gram positive and Gram negative species were to be incubated with QDs, and the concentration of metals monitored in the medium over hours to weeks. The medium was also to be inspected by fluorescence microscopy and TEM.
- What is the evidence for DNA or oxidative damage to cells by QDs? Whereas fluorescence has been well studied, there have been few investigations of the electrochemical properties of QDs. We proposed to use time-correlated single photon counting (TCSPC) techniques to quantitatively measure the electron transfer properties and timescales of QDs. We also proposed to study the interactions between QDs and DNA molecules, rather than single base pairs, with in vitro DNA damage assays. Oxidative DNA damage may be assessed by looking at the most easily oxidized base, guanine, and its oxidation product, 8-oxoguanine25. A fluorescent label for 8oxoguanine exists which enters cells after they have been fixed and permeabilized; cells with DNA damage fluoresce (Kamiya Biomedical Company, Seattle, WA). We proposed to use this commercially-available reagent to test DNA damage in bacteria that have taken up conjugated QDs (e.g., B. subtilis and QD-adenine conjugates).
- How is the formation of bacterial biofilms altered in the presence of quantum dots? We proposed to address this using two strains and both under saturated and unsaturated conditions. We hypothesized that initial biofilm development will be negatively impacted by the effects of QDs on planktonic cells, but that the detrimental effects will be less for the Cd-resistant bacteria relative to sensitive bacteria. Further, we hypothesized total colonization would be reduced under conditions of QD, hence Cd, application during growth. Finally, we hypothesized that Cd-resistant bacteria would provide protection to Cd-sensitive bacteria in dual species biofilms, with the result of increasing development and growth of Cd-sensitive bacteria in mixed-species biofilms versus single-species biofilms.
- What is the toxicity of QDs to existing biofilms? How is the toxicity effect different for biofilms cultivated under saturated versus unsaturated conditions? We hypothesized that Cd-resistant biofilm cells would resist QD toxic effects as they do in liquid culture but that an additional level of resistance will be afforded by the effects of EPS on reducing QD transport through biofilms. Similarly, the sensitivity of Cd-sensitive cells to QDs would be decreased in the biofilm mode of growth.
- How do QDs interact with bacterial biofilms? We hypothesized that QDs would partition into the EPS matrix where they would breakdown into Cd and Se. Cd, being positively charged, will chelate with any number of the polyanionic biomolecules in the EPS. We hypothesized that the transport of QDs into the EPS matrix would be similar across species, but that the breakdown of QDs and subsequent binding of Cd to the EPS would be species-specific because the EPS matrix chemistry is not the same across bacterial species. Further, we hypothesized that Cd-resistant bacteria would harbor less Cd intracellularly, while Cd-sensitive (in that they lack an efflux system) would equilibrate Cd across the cell membrane.
- How do bacteria alter the fates of QDs in porous media? We hypothesized that the sequestration of QDs and thus Cd and Se in bacterial biofilms would reduce the mobility of these nanoparticles in saturated and unsaturated porous media. We also hypothesized that the reduction of mobility would be transient and more effective for Cd-resistant bacteria. Over the long term and especially for Cd-sensitive bacteria, we hypothesized that toxicity and normal turnover would enhance QD and thus Cd mobility in inoculated versus uninoculated porous media. We proposed to test the latter hypothesis by cultivating bacteria in porous media (saturated or unsaturated sand), applying QDs in solution, and after a suitable incubation time monitoring discharge of Cd and QDs under flow through conditions.
The research overall was aimed at quantifying the toxic effects of CdSe quantum dots on bacteria and to eliminate as much ambiguity as possible regarding the source of the toxicity. Thus, we overall set out to answer:
- Are quantum dots toxic to bacteria?
- If so, is this related to the presence of heavy metals?
- Must the quantum dots be degraded in order for the toxicity to be seen?
- Does the size of the quantum dots matter? If so, is that related to the redox properties of the particles?
- Do commercial polymer coatings reduce or eliminate quantum dot toxicity?
- What are bacterial mechanisms for sequestering or eliminating quantum dots?
- Do liquid cultures of bacteria process quantum dots differently than biofilms?
II. Actual scope of research
The actual scope of research matched well the overarching questions listed directly above. By the nature of conducting research, there were deviations from the proposed scope. Some of the deviations were due to the research progressing deeply in one direction at the expense of testing another condition specified in the original scope. Still, across our two research groups (Holden and Nadeau), we accomplished a great deal of what we set out to accomplish and the project has delivered useful understanding regarding the fate and toxicity of CdSe QDs in bacterial systems.
Actual results
The major results of this research regard our assessments of: 1) toxicity and fate of CdSe QDs with Pseudomonas aeruginosa in liquid culture, 2) effects of CdSe QDs on Pseudomonas aeruginosa biofilms, 3) photophysics and biological interactions of CdSe QDs including electron transfer. An additional study that was primarily supported from another project but that occurred and was enhanced by this one is also described.
I. Toxicity and Fate of CdSe QDs with Planktonic Pseudomonas aeruginosa
The results described herein are currently being drafted into a manuscript for publication. This work is responsive to the original objectives 1 – 4 (above) in the original proposal. The methodologies are described briefly, but are consistent with methods previously published by the investigators and collaborators on this project. In addition to Holden and Nadeau, other collaborators include John H. Priester (main researcher on this project in the Holden lab at UCSB), Peter K. Stoimenov (postdoc originally hired on this project, joint between Holden and Stucky groups at UCSB, and responsible for CdSe QD synthesis throughout this subproject), Randall E. Mielke (Jet Propulsion Laboratory for STEM), Samuel M. Webb (Stanford Synchrotron Radiation Laboratory for XANES measurements), Christopher Ehrhardt (graduate student who performed the XRD at UCSB), Jin Ping Zhang (research scientist at the Materials Research Lab at UCSB and STEM contributor), Galen D. Stucky (Professor of Chemistry and Materials at UCSB, and consultant to this project).
The methods used in this research involved cultivating Pseudomonas aeruginosa PG201 (hereafter called PG201) in rich media (Luria Broth or LB) with and without Cd(II) over a range of concentrations, and with or without CdSe QDs. Growth was monitored by measuring optical density. Growth endpoints (late exponential phase) were measured by counting cells (SYBR Gold epifluorescent direct counts) and quantifying intracellular DNA (iDNA, by SYBR Green fluorometry). The CdSe QDs were synthesized by Dr. Stoimenov. Bare CdSe QDs (5 nm nominal size) were chosen for study but were very weakly fluorescent and thus were only traceable by chemical analytical methods. Dose response with Cd(II) administered as cadmium acetate salts was assessed during growth (30° C, aerobic, dark) to determine toxicity thresholds of PG201 to cadmium. Based on PG201’s tolerance to Cd(II), appropriate concentrations of CdSe QDs were selected for comparative growth assays. The growth parameters of interest included growth rate, lag time, and yield.
QDs were anticipated to be metabolized to a degree, to cause some degree of toxicity, and to possibly dissolve with some of the toxicity being attributable to Cd(II) ions. A suite of experiments and measurements were performed to attempt to assess all of the possible fates, using mass balance-based approaches (Fig. 3).
Figure 3. Conceptual diagram of CdSe quantum dot (QD, green spheres with small sphere conjugated attached) interactions with a bacterial cell. Bioconjugated, bare or core-shell, CdSe QDs are subject to abiotic decomposition processes in the vicinity of cells that can lead to heavy metal and metalloid release; released metals and metalloids may be intracellularized, causing toxicity and possibly re-assembling inside cells where they are retained or expelled. Extracellular labeling by whole QDs can be nonspecific. Receptor mediated labeling may lead to whole nanoparticle uptake, especially through damaged membranes that are otherwise insufficiently porous to allow diffusive passage. Once inside cells, whole QDs may be expelled, or metabolized to various extents including conjugate loss and complete breakdown. (Source: Nadeau et al., 2008, in press, “In Grassian, V. H. “Nanoscience and Nanotechnology: Environmental and Health Impacts”, Wiley & Sons, NY.
Prior to each growth experiment, QDs were dialyzed to remove free Cd(II) ions. However, some dissolution (which was determined in this research to be independent of aqueous chemistry including LB, spent broth supernatant, and deionized water) occurred abiotically during culture growth. To discern between possible cell-induced breakdown and abiotic dissolution, and between intact versus dissolved metals from QDs, several different measurement techniques and approaches were needed. Table 1 summarizes the approaches used.
Table 1: Approaches and Methods Used to Assess Toxicity and Fate of CdSe QDs in Pseudomonas aeruginosa Planktonic Cultures. |
||
Assessment Focus |
Treatment and Approach |
Measurement Method(s) |
Toxicity |
Cd(II) vs CdSe QDs vs Controls |
|
Growth rate |
Growth curve |
OD600 |
Lag time |
Growth curve |
OD600 |
Growth extent (yield) |
Growth curve |
OD600, iDNA, Direct counts |
Membrane damage |
Morphology assessment |
Quantitative analysis of STEM images |
Macromolecules |
Protein analysis |
Bradford (BioRad reagents) |
|
|
|
Biotic Fate |
Cd(II) vs CdSe QDs |
|
Cellular total cadmium |
Late exponential phase harvest and quantification for cells |
ICP-AES |
Cellular selenium |
Late exponential phase harvest and quantification for cells |
ICP-AES |
Cellular free cadmium |
Late exponential phase harvest and quantification for cells |
Measure-iT cadmium assay (Invitrogen) |
Cellular QDs |
Se oxidation state |
XANES |
|
Crystallography of cells |
XRD |
|
Intracellular particle sizing and metal/ metalloid localization |
STEM / EDS, quantitative analysis of STEM images |
Extracellular free Cd |
Late exponential phase harvest and quantification for supernatant |
Measure-iT cadmium assay (Invitrogen) |
|
|
|
Abiotic Fate |
CdSe QDs and Cd(II) |
|
QD dissolution |
Dialysis of previously-dialyzed, intact QDs in various media (water, LB, spent cell-free broth) |
Measure-iT cadmium assay (Invitrogen) for dissolved Cd(II) from CdSe QDs |
We observed expected toxicity of Cd(II) that followed a dose-response relationship (Fig. 4, Table 2). Based on the trends, two concentrations of (based on Cd equivalent) CdSe QDs were studied for their comparative effects and also fates: 37.5 mg/L and 75 mg/L. Growth rates with CdSe QDs fell on a dose-response relationship with Cd(II) (Fig. 5), implying that Cd(II) specific ion toxicity might be responsible for the toxicity of CdSe QDs to these bacteria. However, the growth rate at 75 mg/L of cadmium was significantly lower for CdSe QD-treated as compared to Cd(II)-treated cultures (Fig. 5). Thus, it appeared based on growth rate that CdSe QDs were more toxic than Cd(II). But the lag time was shorter for QDs (Table 2), and the total cells (by direct counts) were statistically similar at late exponential phase for QD- versus Cd(II)treated cells (Table 2). Thus, cells appeared to grow somewhat more slowly, yet to the same extents whether fed QDs or Cd(II) ions at the same total initial cadmium concentration.
Figure 4. P. aeruginosa PG201 growth curves for varying Cd(II) concentrations. Shown are representative plots for each treatment, selected from one of three replicates.
Table 2: Effect of Cd(II) and CdSe QDs on lag time and yield of P. aeruginosa cultures. |
|||
Treatment |
Growth Parameter (± SE) |
||
Lag Time (h) |
Max. OD600 |
Yield (cell #/ml)* |
|
Control |
4.80 ± 0.49a |
1.74 ± 0.04f |
6.91 × 109 ± 9.93 × 108 k |
20 mg/L Cd(II) |
5.69 ± 0.31b,c |
1.38 ± 0.12g,h |
NA |
75 mg/L Cd(II) |
5.96 ± 0.46c |
1.31 ± 0.07h |
3.05 × 109 ± 5.07 × 108 l,n |
150 mg/L Cd(II) |
11.64 ± 0.99d |
1.10 ± 0.17i |
1.79 × 109 ± 1.64 × 108 l |
37.5 mg/L CdSe |
6.99 ± 0.65e |
1.57 ± 0.03j |
2.84 × 109 ± 1.08 × 108 m,n |
75 mg/L CdSe | 5.49 ± 0.09e,b | 1.75 ± 0.04f | 2.51 × 109 ± 5.16 × 107 n |
* For 24 h cultures. NA indicates values not available for the particular treatment. |
Figure 5. Specific growth rate versus initial cadmium concentration. CdSe quantum dot treatments are shown as closed diamonds, while Cd(II) – alone treatments are shown as closed circles. Error bars are standard error of the mean (n=3). Values for the 75 mg/L Cd(II) and 75 mg/L QD treatments (*) are significantly different (t test, P < 0.05).
Because it is well understood that cadmium causes Pseudomonas to induce the production of proteins, we also measured protein intra- and extracellularly to determine if similar production was occurring with Cd(II) versus CdSe QD administration. Intracellular protein and extracellular protein increased with increasing Cd(II) concentration and the relationship was shared across the Cd(II) ion and QD treatments (Fig. 6.). This further implied that a similar physiological response to cadmium occurred with these cells whether the form of cadmium was nanoparticulate or dissolved.
Figure 6. Accumulated protein mass versus initial cadmium concentration. Both intracellular (closed symbols) and extracellular (open symbols) protein mass per cell showed positive correlations with cadmium concentration (ANOVA p = 0.03 and p = 0.02, respectively). This trend was apparent when cadmium was administered as cadmium acetate (squares) or CdSe quantum dots (diamonds). Control treatments (circles) had the lowest intra- and extra- cellular protein masses per cell. Error bars are standard error of the mean (n=3).
However, STEM images implied that there were different responses manifested in cellular morphology in that QD-treated cells frequently showed frequent membrane damage (Fig. 7) while Cd(II) treated cells showed less frequent damage (not shown). A large number of images from each treatment (75 mg/L cadmium equivalent) were analyzed for membrane aberrations. A total of 44 (81%; n = 54) QD – treated and 15 (33%; n = 46) Cd(II) – treated cells had membrane damage. Blebbing (Fig. 7B) was observed in QD – treated, but not Cd(II) – treated cells.
These toxicity results implied that QD-treated cells were impacted by both intact QDs and by cadmium. However, effects of QDs would require that QDs were intact and possibly crossing the cell membrane. Therefore, a number of studies were performed to determine the fates of QDs in cultures.
Figure 7. STEM micrographs of QD – treated cells showing membrane holes (A), blebbing (B), and a cell wall detached from the plasma membrane (C). White arrows indicate the region of damage.
By STEM, cells treated with QDs versus Cd(II) appeared different in that the former contained many small particulate forms in the cytoplasm while the latter contained regions in the cells that were bright and particles that were mostly in the periplasm(Fig. 8). Also by STEM /EDS, both Se and Cd accumulated in cells treated with QDs while Cd-“regions” were apparent in Cd-treated cells (not shown). The mean particle diameters for the 75 mg/L QD and 75 mg/L Cd(II) treatments were 8.02 ± 0.24 nm and 14.99 ± 0.54 nm, respectively. These diameters were significantly different (T-test, P = 0.000). The implication was that intact QDs were inside cells. Consistently, the XRD analysis revealed a small wurtzite peak for cells treated with QDs. This would further imply that nonspecific update of QDs was occurring to some extent. However these two datasets were not conducive to performing a mass balance on applied metal.
Figure 8. STEM micrographs of QD – treated (A) and Cd(II) – treated (B) P. aeruginosa cells. Note that the bright particles in (A) are uniformly distributed throughout the cell, while the bright particles in (B) are confined to the periplasmic space.
To better understand the fate of QDs with the cells, the distribution of cadmium and Cd(II) ions was measured after 24 hours by ICP-AES and by Measure-iT methods for the 75 mg/L cadmium treatment. Because the Measure-iT method only measures free Cd(II) ions (determined in this study), the combination of these two methods allowed for, by difference, determining how much cadmium was still in the form of QDs by the growth experiment endpoint (late exponential phase). The mass balance on cadmium implied that QD (bound Cd) cadmium was still quite present in the culture supernantant and, while the amount of total cadmium in cells was the same for QD- versus Cd(II)-treated cultures, the there was less free cadmium in the culture supernatant for QD-treated versus Cd(II)-treated cultures (Fig. 9).
Figure 9. Free (black) and bound (grey) cadmium mass balance in CdSe quantum dot – treated (A) and cadmium acetate – treated (B) P. aeruginosa planktonic cultures. Initial values were measured immediately prior to inoculation; final values were measured at 24 h of growth. Bound cadmium is defined as the measured total cadmium mass (measured by ICP-AES) minus the free cadmium mass (measured by the Measure-iT Cadmium/Lead Assay, Invitrogen).
Also as indicated by Fig. 9 data, the fact that Cd(II) ion content increased in the supernatant of the QD-treated cultures (initial to final difference), there was some dissolution occurring in the cultures under these conditions. While not shown here, several dissolution studies have also been performed, and these studies indicate that dissolution is not enhanced in biotic cultures studied here. Rather, less dissolution of QDs occurs in cultures than in uninoculated control systems (i.e. with LB alone). This implies that while dissolution is occurring, as would be expected in an aqueous system, these bacteria tend to stabilize QDs extracellularly. Still, the evidence from STEM and XRD would imply that QDs enter cells. To further evaluate the fate of QDs in cells, beyond the STEM images that implied intact QDs in cells, we also performed XANES to determine the oxidation state of Se in cellular fractions. In this way, selenium was used as a tracer for the intactness of QDs. XANES revealed that the selenium in cells had an oxidation state more closely resembling elemental selenium than either bulk or QD CdSe (Fig. 10). This implied that QDs in cells were broken down. However, XANES is not quantitative. Also, small amounts of intact QDs, which could enter cells through damaged membranes, would not necessarily alter a XANES plot if the majority selenium content indeed elemental.
Figure 10. XANES spectra for the CdSe quantum dot treatment.
To further compare the possible contribution of QDs to both the cadmium and selenium “burden” in cells, an analysis that combined particle counts (Fig. 8) and ICP data (Fig. 9) was performed. XANES analysis of QD-treated cells showed that virtually all measured selenium was in an oxidized form, or in an organo-selenium compound, i.e. essentially no Se2- was measured but yet should occur with intact CdSe QDs. Additionally, all of the total cadmium (measured by ICP-AES) associated with cells was determined to be Cd2+ (measured by the Measure-iT Cadmium Assay), suggesting that all QDs had broken down when they came into contact with cells. From the STEM micrographs, however, there are a significant number (nearly 200,000) of nanoparticles per cell.
The mean number of nanoparticles per cell (from STEM images, as per Fig. 8) was calculated to be 1.87 × 105 ± 1.64 × 104. The total surface area covered by nanoparticles was 9.43 × 106 ± 8.26 × 105 nm2, and the mean surface area of a cell was 3.87 × 108± 3.86 × 107 nm2. Thus, approximately 2.60% ± 0.29% of the total surface area in the images of cells was covered by nanoparticles. When assuming that the nanoparticles are evenly distributed on the surface of the cells, the total mass of cadmium that could be contained in the nanoparticles was calculated to be 0.0038 pg. When it was assumed that the particles are evenly distributed throughout the cells, the total Cd mass contained in nanoparticles was calculated to be 0.0083 pg.
Thus, if nanoparticles observed in Fig. 8 were indeed intact CdSe QDs, a maximum of 0.0083 pg of cadmium associated per cell would occur as nanoparticles. However, a total of 0.14 pg of cadmium was measured per cell when cells were grown with 75 mg/L (as cadmium) of CdSe QDs. Thus, only a maximum of 5.9% of the cadmium could be bound in QDs, even if there are nearly 200,000 QDs per cell. Given also that there is between 1% and 10% error in the measurements in cell-associated Cd2+, then it is possible that all of the total intracellular cadmium would be measured as “free” Cd(II). The same could be said about the XANES analysis. Because at least 94.1% of the total cell- associated Se is not bound in QDs, any signal given by intact QDs may have been swamped by the larger signal from oxidized Se or organoselenium compounds. It is also interesting to note that a very weak XRD peak for CdSe was observed in cells treated with QDs. XRD is likely a more sensitive technique than ICP-AES and the fluorescent Cd assay (Measure-iT), and may have picked up a signal from the < 6% of intact QDs.
Taken together, these results imply that QDs are somehow stabilized by cells such that QDs do not break down as much as they would under abiotic, similar solution chemistry conditions. However, QDs cause membrane damage not observed with Cd(II) and, because they do not break down significantly in culture, they are more potent toxicants than Cd(II). The fact that growth rates and other physiological indices (protein induction, lag time and yield) are similar with Cd(II) is perhaps misleading: QDs are not equally toxic as Cd(II), they are in fact more toxic. QDs enter cells nonspecifically. The constituent metal and metalloid are in cells, but it is not known if QDs are broken down inside cells or if they are dissolved outside of cells wherein constituents diffuse into cells and accumulate. However, intact QDs are observed in cells. Thus it is possible that breakdown could be associated with cells. These results are still in analysis and to be included in a manuscript for publication.
II. Toxicity of CdSe QDs to Biofilm Pseudomonas aeruginosa
Similar experiments (and methods) to those described above were conducted with PG201 cultivated on membranes overlaying solid media formulated with LB (as in liquid culture). The methods for both culturing and imaging biofilms were furthered, at the beginning of this project, by two complementary studies (see Publications, Priester et al., 2006; Priester et al., 2007). Growth of biofilms PG201 followed very similar dose-response relationships to those observed with planktonic PG201 (Fig. 11). Bright “dots” appeared n ESEM images of biofilms, implying that QDs were accumulating in the extracellular polymers (Fig. 12). Consistently, cadmium ions became very concentrated under membranes of Cd(II)-treated biofilms, as did QDs under QD-treated biofilms (by ICP-AES, Fig. 13). These data imply that QDs cause toxicity similarly to levels observed in planktonic cells, but QDs and Cd(II) tend to hyper-accumulate around cells This suggests that biofilms could be zones of biomagnification in unsaturated systems.
Figure 11. Lag time (top) and specific growth rate (bottom) of PG201 with either Cd(II) or CdSe QDs added to either planktonic (solid symbols) or biofilms (open symbol) cultures.
Figure 12. Accumulation of QDs in extracellular polymers of biofilm PG201
Figure 13. Depiction of observed accumulation of Cd(II ) (top) and CdSe QDs (bottom) in solid media beneath membranes with associated biofilms. ICP-AES analysis of agar sampled under the membrane and between membranes is the basis for this depiction.
III. Electron transfer studies
Many microbes use metals as terminal electron acceptors, thereby reducing them and solubilizing solid surfaces including metal oxides, e.g. Shewanella sp. decomposing magnetite nanocrystals. The decrease in metabolic activity of mammalian cells with photoactivation of quantum dots is suggestive of electron transfer from cellular energy-generating systems to the nanoparticles. This observation was exploited by Clarke et al. who modified CdSe quantum dots with a dopamine bioconjugate that shuttles electrons to the quantum dot from cells (see Publications). Phototoxicity is observed, but is controlled with the administration of antioxidants. The importance of this work is mainly that fluorescent biosensors are shown to be highly redox sensitive within cells, thus fluorescing maximally where cellular subcompartments are most oxidizing and thus electrons are flowing to the bioconjugated quantum dot. However, this work was also applied to investigating the possibility of electron transfer between membranes of actively-respiring bacteria and nanoparticles, as has been shown recently for TiO2 and E. coli.
In this part of the research, both Escherichia coli and PG201 were studied for their possible electron transfer with QDs. CdSe/ZnS QDs were synthesized using a method adapted from the literature based on the noncoordinating solvent 1-octadecene (ODE). Extensive outer labeling of the bacteria was observed by ESEM (Fig. 14 A, B) of E. coli with dopamineconjugated core shell (ZnS) CdSe QDs. QDs also appear clustered away from cells (Fig. 14A). Analysis of image variety (e.g. as in Priester et al., 2007) shows concentrations of bright pixels, indicative of intense metal labeling, on the perimeters of cells (Fig. 14B), suggesting close physical association between cells and QDs. Lifetime fluorescence measurements (Fig. 15) using time resolved single photon counting (TRSPC, i.e. laser spectcroscopy) similarly to Nadtochenko et al., is suggestive of fluorescence enhancement for QDs associated with cells. This was not observed to a similar extent with either NaN3-treated cells or with P. aeruginosa, nor with live, unstained cells (data not shown). While this preliminary data are suggestive of electron transfer between cells and dopamine-conjugated QDs, additional experimentation is needed to rule out confounding effects of other, perhaps media-associated, constituents. This work is being completed as of the time of the writing of this report.
IV. Other related research
In addition to the experimental work that was responsive to the original proposal, there was also a project in the social sciences that was completed while this research was performed. While no funds from this grant were used to directly support that research, page charges for publication were used and this research undoubtedly made it possible for the research to be awarded to UCSB. The subject was surveying nanomaterials organizations worldwide for their EH&S practices. The work is in press in ES&T (Conti et al., 2008) and provides useful insights for policy in that area.
Figure 14: Association of dopamine-conjugated core-shell (ZnS) CdSe QDs with washed, exponential phase Escherichia coli cells as visualized with ESEM, showing A) clusters of QDs (bright regions) around cells, B) enhanced image variety (see Priester et al. in Publications, 2007) at the cell and QD cluster perimeters, C) the relative absence of electron-dense regions in control (no QD) cells, and D) lack of image variety with control cells. (unpublished data, Priester et al.).
Figure 15: Fluorescence decay (lifetime fluorescence) of solutions containing dopamine-conjugated green CdSe-ZnS core shell quantum dots incubated (30 min) either with washed, exponential phase Escherichia coli cells (QDDAEcoNorm) or without cells (QD+DA). Excitation was with a 100 fs laser pulse (5 to 13 mW, 400 nm) and the buffer is phosphate (Sigma) (unpublished data, J. Nadeau, S. Clarke, J. Priester, P. Holden, et al.)
Quality assurance measures
The QA measures in the proposal were mostly followed. An exception was to the first major measure outlined. Because of the geographical separation between the Nadeau and Holden labs, inter-lab coordination on independent experiments was not feasible. Rather, close coordination and appropriate division of labor was performed during the collaborative experiments in electron 22 transfer (see III above). Also, Nadeau did not provide QDs to the Holden group for the planktonic and biofilms studies. Early on, the team decided to perform basic toxicity studies with bare QDs and these are fairly easy to make at UCSB where conditions of synthesis can be controlled. Both Nadeau and Holden collaborate with the Stucky group and are thus both confident that proper procedures for QD synthesis were followed. There was little batch to batch variability as determined by numerous growth studies with numerous batches of QDs.
Consistent with the original QA plan, in every experiment testing living bacteria with quantum dots, controls included: living bacteria without quantum dots; living bacteria without quantum dots, but with heavy metal salts equivalent to the concentration found in the quantum dots. Killed bacteria were not used. Whenever possible, the analysis of the sample was conducted with the researcher “blind” to the sample’s identity.
Also, no experiment was considered for presentation or publication unless it had been repeated at least 3 times with consistent results each time. “Consistency” can be a subjective measure (e.g., whether quantum dots are taken up or not); when it is objective, the criterion for sufficient statistics was either (a) <10% variation in the parameter being measured over at least 3 trials; or (b) significance vs. controls using Student’s t test.
Further, specific equipment: spectrometers were calibrated with dyes of known spectrum and quantum yield (coumarin and tetramethylrhodamine). For electron microscopy with energy-dispersive X-ray spectroscopy (EDS) to determine elemental ratios, the appropriate controls were prepared. For unfixed, un-embedded samples, the controls were bare electron microscopy grids. For embedded, thin-sectioned samples, controls consisted of embedded control bacteria as well as simply the embedding resin alone.
For abiotic experiments involving quantum dots, quality control was obtained by using quantum dots from a single preparation for all experiments. Samples were stored in the dark. For biotic studies, culturing was controlled by using the same age of inoculum prepared identically for each experiment, growing liquid cultures to exactly the same cell density and cell growth phase (e.g., exponential vs. plateau), and using identical culture medium, culture temperature, and exposure to light. Absorbance measurements were used to track bacterial growth so that cultures were standardized and time since inoculation was used as a reliable measure of cell density and growth phase. Statistical tests included calculation of mean, standard error, and standard deviation and Student’s t. For all efforts, samples were prepared and examined within a few hours, and were carefully controlled for time since exposure. Preservation for analysis did occur by freezing for ICP-AES and for other biochemical analysis using wet chemistry. For electron microscopy where samples needed to be preserved, this took place by exposure to 1% glutaraldehyde and subsequent storage at 4 °C in the dark.
The software used was Systat for simple statistics. No computer modeling was required for this research. In general, “outlying” data was not found or it would have been appropriately evaluated in the context of other data.
As the data were collected, preliminary results were presented at conferences and submitted to peer-reviewed journals. The results of the early stages of the study determined which experiments were to be repeated, and the number of times repetition was necessary to achieve statistical significance.
References
- Mattoussi, H. et al. Self-assembly of CdSe-ZnS quantum dot bioconjugates using an engineered recombinant protein. J. Am. Chem. Soc. 122, 12142-12150 (2000).
- Chan, W.C. & Nie, S. Quantum dot bioconjugates for ultrasensitive nonisotopic detection. Science 281, 2016-2018. (1998).
- Bruchez, M., Jr., Moronne, M., Gin, P., Weiss, S. & Alivisatos, A.P. Semiconductor nanocrystals as fluorescent biological labels. Science 281, 2013-2016 (1998).
- Han, M., Gao, X., Su, J.Z. & Nie, S. Quantum-dot-tagged microbeads for multiplexed optical coding of biomolecules. Nat Biotechnol 19, 631-635 (2001).
- Wu, X. et al. Immunofluorescent labeling of cancer marker Her2 and other cellular targets with semiconductor quantum dots. Nat Biotechnol 21, 41-46 (2003).
- Jaiswal, J.K., Mattoussi, H., Mauro, J.M. & Simon, S.M. Long-term multiple color imaging of live cells using quantum dot bioconjugates. Nat Biotechnol 21, 47-51 (2003).
- Kloepfer, J.A. et al. Quantum Dots as Strain- and Metabolism-Specific Microbiological Labels. Appl Environ Microbiol 69, 4205-4213 (2003).
- Lane, T.W. & Morel, F.M.M. A biological function for cadmium in marine diatoms. Proc. Natl. Acad. Sci. U. S. A. 97, 4627-4631 (2000).
- Jarup, L., Berglund, M., Elinder, C.G., Nordberg, G. & Vahter, M. Health effects of cadmium exposure - a review of the literature and a risk estimate. (vol 24, pg 1, 1998). Scand. J. Work Environ. Health 24, 240-240 (1998).
- Jarup, L. Cadmium overload and toxicity. Nephrol. Dial. Transplant. 17, 35-39 (2002).
- Walsh, C.T., Sandstead, H.H., Prasad, A.S., Newberne, P.M. & Fraker, P.J. Zinc - Health-Effects and Research Priorities for the 1990s. Environ. Health Perspect. 102, 5-46 (1994).
- Hoffman, D.J. Role of selenium toxicity and oxidative stress in aquatic birds. Aquat. Toxicol. 57, 11-26 (2002).
- Aldana, J., Wang, Y.A. & Peng, X.G. Photochemical instability of CdSe nanocrystals coated by hydrophilic thiols. J. Am. Chem. Soc. 123, 8844-8850 (2001).
- Bae, W.O., Abdullah, R., Henderson, D. & Mehra, R.K. Characteristics of glutathione-capped ZnS nanocrystallites. Biochem. Biophys. Res. Commun. 237, 16-23 (1997).
- Bae, W., Abdullah, R. & Mehra, R.K. Cysteine-mediated synthesis of CdS bionanocrystallites. Chemosphere 37, 363-385 (1998).
Journal Articles on this Report : 16 Displayed | Download in RIS Format
Other project views: | All 50 publications | 18 publications in selected types | All 17 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Clarke SJ, Hollmann CA, Zhang Z, Suffern D, Bradforth SE, Dimitrijevic NM, Minarik WG, Nadeau JL. Photophysics of dopamine-modified quantum dots and effects on biological systems. Nature Materials 2006;5(5):409-417. |
R831712 (2006) R831712 (Final) |
Exit Exit |
|
Holsapple MP, Farland WH, Landry TD, Monteiro-Riviere NA, Carter JM, Walker NJ, Thomas KV. Research strategies for safety evaluation of nanomaterials, Part II: toxicological and safety evaluation of nanomaterials, current challenges and data needs. Toxicological Sciences 2005;88(1):12-17. |
R831712 (Final) R831715 (2005) |
Exit Exit Exit |
|
Leduc PR, Wong MS, Ferreira PM, Groff RE, Haslinger K, Koonce MP, Lee WY, Love JC, McCammon JA, Monteiro-Riviere NA, Rotello VM, Rubloff GW, Westervelt R, Yoda M. Towards an in vivo biologically inspired nanofactory. Nature Nanotechnology 2007;2(1):3-7. |
R831712 (Final) R831715 (2006) |
Exit Exit |
|
Monteiro-Riviere NA, Inman AO. Challenges for assessing carbon nanomaterial toxicity to the skin. Carbon 2006;44(6):1070-1078. |
R831712 (Final) R831715 (2005) R831715 (2006) |
Exit |
|
Priester JH, Olson SG, Webb SM, Neu MP, Hersman LE, Holden PA. Enhanced exopolymer production and chromium stabilization in Pseudomonas putida unsaturated biofilms. Applied and Environmental Microbiology 2006;72(3):1988-1996. |
R831712 (2006) R831712 (Final) |
Exit Exit Exit |
|
Priester JH, Horst AM, Van De Werfhorst LC, Saleta JL, Mertes LAK, Holden PA. Enhanced visualization of microbial biofilms by staining and environmental scanning electron microscopy. Journal of Microbiological Methods 2007;68(3):577-587. |
R831712 (2006) R831712 (Final) |
Exit Exit Exit |
|
Rochira JA, Gudheti MV, Gould TJ, Laughlin RR, Nadeau JL, Hess ST. Fluorescence intermittency limits brightness in CdSe/ZnS nanoparticles quantified by fluorescence correlation spectroscopy. Journal of Physical Chemistry C2007;111(4):1695-1708. |
R831712 (2006) R831712 (Final) |
Exit Exit |
|
Rouse JG, Yang J, Barron AR, Monteiro-Riviere NA. Fullerene-based amino acid nanoparticle interactions with human epidermal keratinocytes. Toxicology In Vitro 2006;20(8):1313-1320. |
R831712 (Final) R831715 (2006) |
Exit Exit |
|
Rouse JG, Yang J, Ryman-Rasmussen JP, Barron AR, Monteiro-Riviere NA. Effects of mechanical flexion on the penetration of fullerene amino acid-derivatized peptide nanoparticles through skin. Nano Letters 2007;7(1):155-160. |
R831712 (Final) R831715 (2006) |
Exit |
|
Ryman-Rasmussen JP, Riviere JE, Monteiro-Riviere NA. Penetration of intact skin by quantum dots with diverse physicochemical properties. Toxicological Sciences 2006;91(1):159-165. |
R831712 (Final) R831715 (2006) |
Exit Exit Exit |
|
Ryman-Rasmussen JP, Riviere JE, Monteiro-Riviere NA. Surface coatings determine cytotoxicity and irritation potential of quantum dot nanoparticles in epidermal keratinocytes. Journal of Investigative Dermatology 2007;127(1):143-153. |
R831712 (Final) R831715 (2006) |
Exit Exit Exit |
|
Walker GM, Monteiro-Riviere NA, Rouse J, O'Neil AT. A linear dilution microfluidic device for cytotoxicity assays. Lab on a Chip 2007;7(2):226-232. |
R831712 (Final) R831715 (2006) |
Exit |
|
Wei W, Sethuraman A, Jin C, Monteiro-Riviere NA, Narayan RJ. Biological properties of carbon nanotubes. Journal of Nanoscience and Nanotechnology 2007;7(4-5):1284-1297. |
R831712 (Final) R831715 (2006) |
Exit |
|
Witzmann FA, Monteiro-Riviere NA. Multi-walled carbon nanotube exposure alters protein expression in human keratinocytes. Nanomedicine: Nanotechnology, Biology, and Medicine 2006;2(3):158-168. |
R831712 (Final) R831715 (2006) |
Exit |
|
Xia X-R, Monteiro-Riviere NA, Riviere JE. Trace analysis of fullerenes in biological samples by simplified liquid-liquid extraction and high-performance liquid chromatography. Journal of Chromatography A 2006;1129(2):216-222. |
R831712 (Final) R831715 (2006) |
Exit Exit |
|
Zhang LW, Zeng L, Barron AR, Monteiro-Riviere NA. Biological interactions of functionalized single-wall carbon nanotubes in human epidermal keratinocytes. International Journal of Toxicology 2007;26(2):103-113. |
R831712 (Final) R831715 (2006) |
Exit |
Supplemental Keywords:
RFA, Scientific Discipline, TREATMENT/CONTROL, Sustainable Industry/Business, Environmental Chemistry, Sustainable Environment, Technology, Technology for Sustainable Environment, Biochemistry, New/Innovative technologies, Chemistry and Materials Science, Environmental Engineering, biofilm, quantum dots, DNA damage, heavy metal sequestration, nanotechnology, environmental sustainability, engineering, environmentally applicable nanoparticles, semiconductor nanocrystals, sustainability, innovative technologyProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.