Gold nanoparticles (nAu) are used in drug delivery systems allowing for targeted cellular distribution. The effects of increased use and release of nanoparticles into the environment are not well known. A species sensitivity distribution (SSD) allows for the ecotoxicological hazard assessment of a chemical based on single species toxicity tests. Aquatic toxicity needs to be related to particle characterization in order to understand the effects. The behaviour of nAu in the medium changed as the concentration increased. The toxic potential of ionic gold and nAu was expressed as a hazardous concentration where 5% of species will be harmed (HC5). The HC5 for nAu was much higher (42.78 mg/L) compared to the ionic gold (2.44 mg/L). The differences between the hazard potentials of nAu and ionic gold were attributed to the nAu not releasing any Au ions into solution during the exposures and following an aggregation theory response. Exposures to ionic gold on the other hand followed a clear dose dependent response based on the concentration of the ionic metal. Although SSDs present an indication of the relative hazard potential of nanoparticles, the true worth can only be achieved once other nanoparticle characteristics and their behavior in the environment are also considered.
In recent years, gold nanoparticles (nAu) have been studied and developed within the biological and photothermal therapeutic contexts. The major clinical interest is within the application of novel drug delivery systems and targeted delivery into cells [
To manage environmental resources such as water quality, laboratory toxicity tests are used worldwide and are deemed as the first step in a tiered approach to set up guidelines for acceptable maximum concentrations of specific pollutants [
Species sensitivity distributions (SSDs) are being integrated into ecological risk assessments to evaluate the toxicity of particular chemicals [
The nAu stock solution (
The reconstituted water for
Physicochemical water quality parameters were measured according to standard test protocols [
Total gold concentrations were determined at the start of the toxicity bioassay (0 h) as well as at the conclusion of the assay (48 h or 96 h where relevant) to verify nominal concentrations in both ionic and nAu exposures. Samples (11 mL) were acidified with 3 mL Suprapur 30% hydrochloric acid (Merck) and 1 mL 65% HNO3 to ensure that all of the gold was in ionic form. Inductively coupled plasma atom emission spectroscopy (ICP-OES) (Spectro Arcos FSH12) techniques were used to analyse the water samples.
Based on the recommendations by von der Kammer et al. [
The arthropods
For nAu a concentration range of 0.5 mg/L, 2 mg/L, 5 mg/L, 10 mg/L, 15 mg/L, 20 mg/L, 25 mg/L, 35 mg/L, and 45 mg/L was used. A separate dispersant control (the citrate buffer in which the nAu particles were dispersed) was made up in the same concentration range as the nAu. The ionic gold exposure concentration range was 0.0005 mg/L, 0.005 mg/L, 0.5 mg/L, 1 mg/L, 2 mg/L, and 5 mg/L. A positive control consisting of 1 g/L potassium dichromate solution was also conducted. The test organisms were not fed during the assay.
Twenty-one neonates were used for each concentration and tests were carried out in triplicate with each of the three 50 mL beakers per concentration containing seven organisms. A 16 h light and 8 h dark cycle was applied for the duration of the test and the temperature was maintained at
Acute toxicity tests were conducted using two standard toxicity test fish species, that is,
Data from the acute bioassays were analysed to calculate LC50/EC50 values using the Probit or Trimmed Spearman-Kärber methods where appropriate [
The results obtained from the bioassays were used to compare the tolerances of standard test species and indigenous arthropod and fish species using an SSD approach [
The physicochemical water parameters were maintained within the OECD guidelines [
Physicochemical water quality means for selected species in toxicity exposure bioassays with nAu, chloroauric acid, and citrate buffer.
Species | pH (min–max) | D.O. (mg/L) (min–max) | O2 (%) (min–max) | EC ( |
Temp. (°C) (min–max) | TDS (mg/L) |
---|---|---|---|---|---|---|
Citrate buffer | ||||||
|
7.85 | 5.43 | 64.6 | 339 | 20.5 | 171 |
(6.57–8.36) | (4.44–6.08) | (61.7–67.4) | (309–374) | (20–21.2) | (156–189) | |
|
7.70 | 7.26 | 77.0 | 310 | 19.3 | 155 |
(7.6–8.37) | (6.63–8.72) | (62.6–91.4) | (297–362) | (20–21.9) | (148–181) | |
|
8.26 | 4.77 | 69.6 | 708 | 23.3 | 350 |
(7.10–8.65) | (4.0–6.15) | (60.0–87.2) | (621–809) | (22.7–23.9) | (309–398) | |
|
7.98 | 5.04 | 66.9 | 708 | 21.4 | 365 |
(7.39–8.34) | (4.26–5.9) | (60.2–78.08) | (530–841) | (20.6–22.1) | (321–407) | |
|
7.74 | 4.85 | 67.7 | 748 | 22.2 | 373 |
(6.57–8.22) | (3.5–5.77) | (60–77.1) | (696–810) | (21.6–23.03) | (316–405) | |
|
6.93 | 5.62 | 71.7 | 732 | 23.1 | 366 |
(6.01–7.75) | (4.3–6.47) | (60–89.5) | (665–874) | (22.9–23.2) | (333–402) | |
|
6.57 | 4.85 | 67.9 | 674 | 22.1 | 372 |
(5.81–7.39) | (4.00–5.77) | (60–77.1) | (612–704) | (21.3–22.6) | (313–595) | |
|
7.06 | 5.93 | 67.0 | 680 | 22.5 | 346 |
(5.99–7.93) | (4.89–7.07) | (60.3–80.5) | (638–729) | (21.5–23) | (324–369) | |
|
||||||
nAu | ||||||
|
7.66 | 5.05 | 63.1 | 328 | 20.7 | 167 |
(7.05–8.7) | (4.4–6.1) | (60.7–72.6) | (208–420) | (20–21.9) | (104–220) | |
|
7.78 | 7.23 | 78.4 | 309 | 20.2 | 155 |
(7.74–8.39) | (6.39–9.28) | (71.7–97.8) | (293–374) | (20.5–22.2) | (146–206) | |
|
7.88 | 4.49 | 68.1 | 674 | 22.6 | 336 |
(7.13–8.42) | (4.04–6.26) | (60.1–98.7) | (642–765) | (22.4–23.9) | (321–412) | |
|
7.43 | 4.68 | 63.9 | 681 | 20.8 | 340 |
(7.13–7.97) | (4.11–6.24) | (60.0–82.9) | (621–827) | (20.1–21.1) | (310–414) | |
|
7.57 | 5.97 | 67.3 | 743 | 22.2 | 370 |
(6.31–8.19) | (4.84–8.26) | (60.0–79.6) | (658–824) | (21.3–22.6) | (329–412) | |
|
7.24 | 5.89 | 73.9 | 699 | 23.0 | 350 |
(6.23–7.79) | (5.14–6.76) | (60.8–92.20) | (580–762) | (22.9–23.2) | (328–379) | |
|
7.51 | 5.69 | 66.6 | 703 | 22.2 | 335 |
(6.85–8.01) | (4.37–8.26) | (60–79.6) | (617–754) | (21–22.7) | (308–378) | |
|
7.55 | 5.55 | 55.9 | 556 | 22.9 | 275 |
(6.29–7.90) | (4.08–7.08) | (60.0–80.7) | (628–762) | (21.6–23.1) | (314–381) | |
|
||||||
Chloroauric acid | ||||||
|
7.90 | 5.35 | 66.3 | 344 | 20.6 | 176 |
(6.93–8.95) | (4.35–6.13) | (62.7–71.5) | (198–382) | (20–22) | (104–267) | |
|
8.22 | 7.57 | 83.3 | 332 | 21.6 | 168 |
(7.59–8.43) | (6.4–9.78) | (62.5–99.7) | (296–381) | (20.4–22.4) | (148–259) | |
|
7.92 | 4.98 | 72.0 | 677 | 23.0 | 350 |
(6.24–8.67) | (4.2–6.044) | (62.8–85.3) | (621–739) | (21.8–23.7) | (311–485) | |
|
7.56 | 5.86 | 72.3 | 700 | 21.3 | 350 |
(6.45–7.98) | (4.23–7.85) | (60.2–84.8) | (620–806) | (20.6–21.6) | (310–403) | |
|
7.33 | 5.63 | 73.6 | 711 | 21.7 | 355 |
(6.26–8.27) | (4.8–7.67) | (60.9–81) | (616–838) | (20.9–22.4) | (308–419) | |
|
7.62 | 6.82 | 77.0 | 702 | 23.1 | 352 |
(7.36–8.04) | (4.49–8.26) | (62.1–99) | (603–736) | (22.3–23.5) | (301–400) | |
|
7.24 | 5.61 | 73.4 | 633 | 21.6 | 388 |
(6.26–8.14) | (4.6–7.67) | (62.3–81) | (545–792) | (21.0–22.1) | (308–436) | |
|
7.77 | 5.90 | 76.7 | 700 | 23.1 | 350 |
(7.36–8.04) | (4.46–8.24) | (61.8–99.4) | (603–794) | (22.3–23.6) | (301–398) |
To verify the concentrations added to each exposure group, total gold concentrations were measured. The nominal ionic gold concentrations and measured concentrations were in close agreement (Figure
Mean measured gold concentrations (mg/L) ± standard deviation of the (a) ionic gold (HAuCl4) and (b) nAu exposures at all exposure concentrations used in this study.
The largest proportion of nAu (i.e., >90%) in the
Size distribution of gold nanoparticles across all concentrations in
nAu size | |||
---|---|---|---|
Concentration | nm | St. dev. | % intensity |
25 mg/L | 565.6 | 87.82 | 93.9 |
35 mg/L | 827.45 | 56.50 | 92.5 |
45 mg/L | 3175.5 | 443.35 | 93 |
In contrast to the nAu particle aggregation/agglomeration in
Raw data sheet showing two peaks of
The nanoparticle size distribution was determined at 0 h and again at 96 h to observe any changes in aggregation over time. Results are presented by indicating the main size distribution classes in terms of the percentage contribution after 0 h and 96 h. At 0 h nAu exposures there were two dominant size distribution classes, that is, sizes representing >60% intensity and less than 40% intensity (Figure
Particle size distribution (nm) of nAu with a percentage intensity of (a) more than 60% at 0 h and (b) less than 40% at 0 h as well as (c) more than 55% at 96 h and (d) less than 45% at 96 h.
At 0 h in the 5 mg/L concentration 89% of the sample had a size distribution of 150.7 nm and 12% of nanoparticles were 12.64 nm. After 96 h more than 90% of particles were 133.7 nm in size. At 10 mg/L all particles measured in the sample were around 75.79 nm while after 96 h the particles (82.7%) had agglomerated to 986.4 nm while 17.3% remained approximately 79.64 nm in diameter. At 15 mg/L, 89% were in the size range of 89.49 nm, while 11% were at 15.51 nm. After 96 h particles had all aggregated to 96.6 nm. At the onset of the 20 mg/L exposure 88% of the particles had aggregated to 127.9 nm and 12.2% to 17.12 nm in diameter. After 96 h 64% of the sample had agglomerated to 894.5 nm and 36% of the sample was 121.5 nm in size. At the 25 mg/L concentration just over half (59.6%) of the sample was 298.4 nm in size while just under half (40.4%) of the sample was 71.48 nm. After 96 h the same ratio was maintained where 54.5% of the sample had a size of 455.6 nm and 45.5% had a size of 96.89 nm. At 30 mg/L the prevalence was 72.3% distribution of 163.1 nm particles and 27.7% at 40.27 nm, while after 96 h all particles had agglomerated to 201.5 nm. The 35 mg/L concentration revealed that 91% of the sample had aggregated to 96.89 nm and 9% of the sample remained around 14 nm in size. After 96 h all particles had aggregated to 138.2 nm. The 40 mg/L concentration showed an immediate aggregation at 0 h where 57.4% of the sample was 464.9 nm and 42.6% of the sample was 57.55 nm in size. After 96 h all sample measured contained a uniform size of 226.9 nm. At the highest concentration, 45 mg/L, 79.9% of the sample had particles 82.52 nm in size and 15% remained around the original size of 14 nm. After 96 h the prevalence of particles at 288 nm was 87.7% and 12.3% were 63.05 nm in size. In the 40 mg/L and 45 mg/L concentration nAu had precipitated out of solution and were visible to the naked eye. These agglomerates were in the millimetre size range and could not be measured on the DLS instrument.
At 0 h there was a clear increase in size of nAu aggregations and agglomeration at concentrations of 25 mg/L and 40 mg/L when compared to other concentrations. But after 96 h the 20 mg/L and 25 mg/L concentrations still showed agglomerations increasing in size and dispersion in the media while concentrations 30 mg/L and above showed no or lower measured sizes, which is attributed to particle precipitation.
The EC differed for the
Zeta potential (mV) of the nAu in the
nAu | 1 mg/L | 5 mg/L | 15 mg/L | 25 mg/L | 35 mg/L | 45 mg/L |
---|---|---|---|---|---|---|
Zeta potential [mV] | ||||||
0 h |
|
|
|
|
|
|
48 h |
|
|
|
|
|
|
For the fish medium there was much greater variation (as observed in the increase in the standard deviation) of the measured zeta potential. The standard deviation (Figure
(a) The zeta potential (mV) average peak reading across all concentrations. (b) The zeta potential (mV) reading where the peak had greater-than-seventy-percent distribution measured across all concentrations.
The FTIR results revealed two distinct peaks (Figure
(a) The FTIR spectra results of the highest concentration of gold nanoparticle in an aqueous solution. (b) FTIR spectra results of the dried gold nanoparticle sample in the highest concentration to eliminate OH interference from the solution.
The TEM results revealed a better understanding of the size distribution patterns. As defined by Klaine et al. [
(a) A TEM micrograph of a drop of the concentrated citrate capped 1000 mg/L concentration gold nanoparticle stock solution. (b) An FFT image of the nAu to confirm the presence of a crystal lattice. ((c) and (d)) A TEM micrograph of the 5 mg/L gold nanoparticle concentration showing average aggregations and measurements. ((e) and (f)) A TEM micrograph of the 10 mg/L gold nanoparticle concentration showing average aggregations and measurements.
In the 20 mg/L exposure there were two major types of agglomerates noted; these were skeletal-like agglomerates which reached approximately 352.1 nm in length (Figure
((a) and (b)) A TEM micrograph of the 20 mg/L nAu concentration showing average aggregations and measurements. ((c) and (d)) A TEM micrograph of the 30 mg/L gold nanoparticle concentration showing average aggregations and measurements. ((e) and (f)) A TEM micrograph of the 40 mg/L gold nanoparticle concentration showing average aggregations and measurements.
No mortalities were recorded in the controls and the dispersant controls (i.e., mortalities were below 10%). It was not possible to calculate LC50 values for the 48 h
Median lethal concentrations for 50% (LC50) effect and the corresponding 95% confidence intervals for test organisms exposed to nAu and chloroauric acid.
Species | Duration/endpoint | LC50 (mg/L) (95% CI) | NOEC | LOEC | |
---|---|---|---|---|---|
Mean | Upper & lower confidence limits (mg/L) | ||||
nAu | |||||
|
48 h | 75.31 |
|
1 | 10 |
|
48 h | Could not be calculated | 0 | 5 | |
|
96 h | Could not be calculated | <15 | >15 | |
|
96 h | No mortalities | |||
|
96 h | 52.57 |
|
<35 | 35 |
|
96 h | Could not be calculated | 5 | 10 | |
|
96 h | 12.08 |
|
5 | 25 |
|
96 h | Could not be calculated | 1 | 5 | |
|
|||||
Chloroauric acid | |||||
|
48 h |
|
|
<0.0005 | 0.0005 |
|
|||||
|
48 h |
|
|
<0.005 | 0.01 |
|
|||||
|
96 h |
|
|
<1 | 5 |
|
|||||
|
96 h |
|
|
<1 | 5 |
|
|||||
|
96 h |
|
|
<10 | 10 |
|
|||||
|
96 h |
|
|
<10 | 10 |
|
|||||
|
96 h |
|
|
1.5 | 2 |
|
|||||
|
96 h |
|
|
<1 | 1 |
|
The cumulative mortality rates of the organisms over the 48 h and 96 h exposure periods for both nAu and ionic gold are depicted in Figures
Cumulative mortalities of test organisms exposed to nAu over 48 h (daphnids) and 96 h (fish).
Cumulative mortalities of test organisms exposed to ionic gold over 48 h (daphnids) and 96 h (fish).
Seven data points are generally required to generate a meaningful SSD for any given chemical [
Acute toxicity data from the published literature that were included used to derive the SSDs.
Test species | Endpoint | Duration (h) | LC50 |
Reference |
---|---|---|---|---|
Arthropods | ||||
|
LC50 | 48 | 2 | Li et al. [ |
|
LC50 | 48 | 0.64 | Nam et al. [ |
|
LC50 | 48 | 0.15 | This study |
|
LC50 | 48 | 0.01 | This study |
|
LC50 | 48 | 0.62 | Nam et al. [ |
|
||||
Fish | ||||
|
LC50 | 96 | 4.85 | This study |
|
LC50 | 96 | 20.58 | This study |
|
LC50 | 96 | 10.78 | This study |
|
LC50 | 96 | 7.53 | This study |
|
LC50 | 96 | 11.3 | This study |
|
LC50 | 96 | 0.93 | This study |
|
LC50 | 96 | 14.4 | Nam et al. [ |
|
LC50 | 96 | 10.7 | Nam et al. [ |
|
LC50 | 96 | 14.1 | Nam et al. [ |
Species sensitivity distributions and confidence limits for nAu based on combined data from this study and the published literature.
Species sensitivity distributions and confidence limits for ionic gold based on acute toxicity data from this study.
Species sensitivity distributions and confidence limits for ionic gold based on toxicity data from this study and the published literature [
Nanoparticle ecotoxicity involves several factors to consider, the initial size of the particles, the capping agent, how particles aggregate in an aquatic medium, and the charge that exists. All these affect the toxicity and ability for particle uptake in organisms [
When there is a change in any of these factors the behaviour of the nAu in an aquatic medium would also change. In
As all citrate capped gold nanoparticles in the stock solution were around the 14 nm range it is evident that from the moment the particles were diluted in the respective environmental media they started to agglomerate to one another. This could also reflect the 10 mg/L loss of nanoparticle concentrations measured due to aggregations as the entire sample was not read but only a portion of it. The citrate capping is soluble and by the hydrolysis of water is converted to citric acid [
Functional groups present on the nAu surface play a role in tissue distribution [
The nAu SSD revealed that fish were more sensitive to nAu with an LC50 (when calculable) of between 12.08 mg/L and 52.6 mg/L, while daphnids were less sensitive with an LC50 ranging from approximately 70 mg/L to 75.31 mg/L. Since arthropods are able to undergo molting they would be able to cope with the nAu particles adhered onto their carapace. Literature to date shows no evidence of nAu uptake in daphnids while gold ions would be taken up [
When comparing the SSD plots it becomes apparent that the nAu and ionic gold have different distributions in toxicity. Smaller organisms, like
The SSD plots for the nAu and the ionic gold clearly indicate that there was a difference in the hazard potential of the two substances with the latter being an order of magnitude more toxic. This is in contrast to studies where SSDs comparing nano metals (i.e., CuO, ZnO, nAg, and nAl2O3) to their bulk metal (ionic) equivalents revealed that the hazard potentials were very similar [
At this stage the toxicity and therefore hazard assessment of nanoparticles are still very much based on dose-response (i.e., exposure concentration) relationships. Only when more data becomes available on the relationship between nanoparticle characteristics (e.g., size, shape, charge, and functional group) and the exposure medium will predictors of toxicity other than the release of metal ions become apparent. Notwithstanding the limitations of using only concentration-based toxicity data, the information provided by the SSDs for nAu provides useful ecotoxicological comparisons with other nanomaterials.
The authors declare that there is no conflict of interests regarding the publication of this paper.
The research is based upon work supported by the Department of Science and Technology. The authors wish to thank MINTEK for supplying the nAu stock solutions.