TY - JOUR
AU - Bogatyrev, Vladimir A.
AB - Abstract A diagnostic test system was developed to determine the toxicity of nanomaterials to the saltwater microalga Dunaliella salina through evaluation of cell death and changes in the culture growth rate at various toxicant concentrations, providing LC50 and other toxicological metrics. The viability of cells was shown to decrease with decreasing chlorophyll absorption of red light by damaged cells. This correlation was confirmed by independent fluorescence microscopic measurements of live and dead cells in the population. Two standard colorless pollutants, hydrogen peroxide and formaldehyde, were used to validate the colorimetric method. The method’s performance is exemplified with three Ag-containing preparations (Ag nitrate, Ag proteinate, and 20-nm Ag nanoparticles) and with cetyltrimethylammonium bromide (CTAB) mixed with colloidal 15-nm Au and 20-nm Ag nanoparticles. The toxicity of the Ag-containing preparations to D. salina decreased in the order Ag nitrate ≥ Ag proteinate ≫ colloidal Ag. The toxicity of colloidal Au-CTAB mixtures was found to depend mostly on the content of free CTAB. The toxicity of colloidal Ag increased substantially in the presence of CTAB. The results suggest that our D. salina–based colorimetric test system can be used for simple and rapid preliminary screening of the toxicity of different nanomaterials. plasmon-resonant particles, CTAB, toxicology, spectrophotometry, fluorescence microscopy Owing to their unique optical properties from the surface plasmon resonance phenomenon, Au and Ag nanoparticles have found numerous applications in biomedical studies as optical probes (Dreaden et al., 2012; Dykman and Khlebtsov, 2012). In particular, colloidal Au (CG) nanoparticles of various shapes, sizes, and structures are in most common use because their plasmonic properties can be controllably tuned (Khlebtsov and Dykman, 2010) and because they are useful tools for surface functionalization (Sapsford et al., 2013). Even though the toxicity of CG is relatively low, some concerns have been expressed regarding the biodistribution of nanoparticles, including nanogold (Khlebtsov and Dykman, 2011), and related toxicity issues (Sharifi et al., 2012). Several common in vitro toxicological techniques are now available for nanotoxicity assessment (Alkilany and Murphy, 2010; Marquis et al., 2009). Toxicological studies made with animal and human cell cultures through standard methods present technological difficulties and are time consuming and costly. Yet, the mechanisms of protection against unfavorable factors are often similar for all living organisms. Therefore, it is desirable to develop a simple laboratory system for toxicity testing that would improve reproducibility and decrease costs while preserving the same level of informativity as reached with animal cell cultures. Synthesis of Au and Ag nanoparticles often involves the use of fairly toxic substances, e.g., cetyltrimethylammonium bromide (CTAB) (Kooij et al., 2012; Li et al., 2014). As the cell membrane is the common target for CTAB and other toxicants (Qiu et al., 2010), it is reasonable to look for convenient cells with a well-developed membrane apparatus. In this regard, the unicellular green microalga Dunaliella salina is a good candidate because, like animal cells, it lacks a cell wall (Masyuk, 1973). Algae of this genus grow intensely and are easy to cultivate in pure culture, as they normally live in salt lakes under hyperosmotic conditions. In addition, the viability of D. salina can be evaluated by the content of chlorophyll through photometry without the use of vital dyes. This makes D. salina an attractive object for nanotoxicological investigations, as reported recently in a study of the toxicity of PbS nanoparticles (Zamani et al., 2014). Some species of Dunaliella have already found application in toxicology (Masyuk et al., 2007) and apoptosis (Segovia et al., 2003) research; a recent example is the study of the toxicity of 50-nm colloidal Ag (CS) particles to D. tertiolecta by Oukarroum et al. (2012). Thus, published reports indicate that it is possible in principle to use D. salina as an all-purpose toxicological test object. The viability of D. salina, as that of animal cells, can be evaluated with microplate systems (Blaise and Férard, 2005) and photometric measurements (Zhu et al., 2010). Here, we describe a simple D. salina–based colorimetric test system for rapid screening of the toxicity of nanomaterials potentially applicable in medical research. The system was validated with 2 standard toxicants by comparing photometric data with direct fluorescence microscopy counting of live and dead cells. The applicability of the method to nanomaterials is exemplified with as-prepared and CTAB-containing 15-nm Au and 20-nm Ag nanoparticle colloids. MATERIALS AND METHODS Microalga. Dunaliella salina Teod. IPPAS D-294 was obtained from the microalgal collection maintained at the Russian Academy of Sciences’ Institute of Plant Physiology, Moscow. For plant maintenance and biomass accumulation, we used Ben-Amotz’s medium (Shaish et al., 1990) composed as follows: NaCl, 1.5 M; MgSO4 × 7H2O, 5 mM; CaCl2, 0.3 mM; KNO3, 25 mM; K2HPO4, 0.2 mM; EDTA, 30 μM; NaHCO3, 50 mM; FeCl3, 2 μM; MnCl2, 7 μM; CuCl2, 1 μM; ZnCl2, 1 μM; CoCl2, 1 μM; (NH4)6Mo7O24, 1 μM; pH 8. Immediately before inoculation, the solution was filtered through a 0.22-μm-pore-size Millipore filter. To produce biomass and maintain the culture, accumulation ones were grown in biological culture growth vessels of diameter 6.5 cm and height 7 cm. Five milliliters of inoculum was added to 50 ml of Ben-Amotz’s medium. A 4- to 5-day-old culture in the logarithmic growth phase was used. The microalgal concentration was estimated spectrophotometrically by using a Specord M40 spectrophotometer (Carl Zeiss Jena, Germany) and a Multiskan Ascent microplate reader (ThermoFisher Scientific). For determination of the number of live cells in the suspensions, control measurements were made by using a Bio-Rad TC20 cell counter (Bio-Rad Laboratories). Preparation of gold nanoparticles. Anionic gold nanoparticles (diameter of approximately 15 nm; CG-15) were prepared by the method of Frens (1973) by reducing HAuCl4 (Sigma-Aldrich) with sodium citrate (Fluka, Switzerland) at boiling. For stock solutions, the suspension (57.9mg Au/L) was concentrated by centrifugation (15 000 × g, 10 min) and removing of 90.4% supernatant. Ag preparations. CS nanoparticles (diameter of approximately 20 nm; CS-20) were prepared according to Fabrikanos et al. (1963) by reducing AgNO3 (Sigma-Aldrich) with the disodium salt of EDTA (Serva, Germany) at boiling (final concentration of 30mg Ag/L). Immediately before the experiment, the suspensions were concentrated to 440 mg Ag/L. Samples containing “ionic” Ag were prepared immediately before the experiment by adding 1.8 ml of the culture medium to 200 µl of aqueous AgNO3 (500 mg Ag/L). The Ag concentration in the stock solution was 50mg Ag/L. A solution of Ag proteinate [PrAg (8.3% [Ag]); Fluka, Switzerland] was made by dissolving 2.5 mg of powder in 5 ml of the culture medium. The Ag concentration in the stock solution was 41.5 mg/l. In calculating the final toxicant concentrations in the wells, we accounted for the dilution with the culture medium and inoculum. Toxicant solutions. A stock solution of formaldehyde was made by diluting 10% commercial formalin with the culture medium. H2O2 solutions were prepared in the same way except that 37% H2O2 was used as the initial solution (KhimMed, Russia). Working dilutions of CTAB (Sigma-Aldrich) were prepared from a 10 mM (3.64 g/L) stock solution. Toxicity test. Nanomaterial toxicity to D. salina was assessed in 96-well flat-bottomed plates (Greiner Bio-One, Austria) in 2 different types of assays: 1D and 2D assays were used to evaluate the effects of single toxicant and mixture exposure scenarios, respectively (Figure 1). In 1 plate of 1D assay, 2 different toxicants can be tested. Initially, decreasing from 150 to 0 μl portions of the toxicants were added to the wells in duplicated (in 2D type) or triplicated rows (in 1D type), with a fixed step of 10%–20%. Then, to each well was added the culture medium to a final volume of 150 μl. Before microalgal inoculation, the initial toxicant concentration in the first well of each row was 2-fold higher than the maximal concentration. The last wells of the rows received no toxicants and were used as negative controls. FIG. 1. View largeDownload slide One- (A) and 2-dimensional (B) toxicant dilution schemes. In the 1D experiment: “toxicant 1” (AgNO3), “toxicant 2” (colloidal silver). Rows A and E are triplicated (A, B, C) and (E, F, G). In the 2D experiment: “toxicant 1” (cetyltrimethylammonium bromide), “toxicant 2” (colloidal silver). Rows A-D are duplicated (A, E) (B, F) (C, G) (D, F). Arrows show dilution direction of toxicants. FIG. 1. View largeDownload slide One- (A) and 2-dimensional (B) toxicant dilution schemes. In the 1D experiment: “toxicant 1” (AgNO3), “toxicant 2” (colloidal silver). Rows A and E are triplicated (A, B, C) and (E, F, G). In the 2D experiment: “toxicant 1” (cetyltrimethylammonium bromide), “toxicant 2” (colloidal silver). Rows A-D are duplicated (A, E) (B, F) (C, G) (D, F). Arrows show dilution direction of toxicants. For inoculation, the microalgal cells were sedimented by centrifugation at 50 × g for 15 min and were redispersed in fresh culture medium so that after the addition of 150 μl of the microalgal suspension to the wells, the suspension absorbance D690 was approximately 0.1, corresponding to a cell concentration of about 5 × 106 ml−1. The plates were then incubated in a laboratory growth chamber (a glass cuvette 35 × 30 × 4 cm filled with water to a depth of approximately 0.5 cm); temperature 24°C ± 2°C. The growth chamber received constant illumination from below with daylight fluorescent lamps (light intensity of 80–100 μM m−2 s−1). The standard incubation time was 48 h. After 24 h of the experiment, the samples were analyzed with a DMI 3000 inverted microscope (Leica, Germany; Simbioz Center for the Collective Use of Research Equipment in the Field of Physical–Chemical Biology and Nanotechnology, IBPPM RAS) set to the phase contrast and luminescence modes (objective lens ×20, dichroic beam splitter D). Images were obtained with a DFC 420 digital camera and the LAS program (Leica) by recording the number of motile (cell-track image), live (red fluorescence of chlorophyll), and dead (blue fluorescence) cells in the bottom center of the well. Next, absorbance was recorded with the Multiskan Ascent reader, and the plates were put back in the growth chamber. After the entire exposure period had elapsed, the cells were fixed by adding 10 μl of 25% glutaraldehyde (Electron Microscopy Science) to each well and leaving the plates to stand overnight at 4 °C in the dark. The live and dead cells were counted as described above. Absorbance was measured at 690 nm by recording D0 (immediately after inoculation) and Dt(c), where t is the exposure time (h) and c is the toxicant concentration. For data analysis, the absorbance changes ΔDt(c)=Dt(c)−Dt=0(c) were normalized to the negative control value ΔDt(c=0)=Dt(c=0)−Dt=0(c=0):   δD=ΔDt(c)/ΔDt(c=0)≡Dt(c)−Dt=0(c)/Dt(c=0)−Dt=0(c=0) . (1) For evaluation of the joint effect of 2 toxicants (CTAB/CG and CTAB/CS), dilutions with a step of 0.1 or 0.2 mg CTAB/L were made in parallel rows with various concentrations of CG-15 or CS-20. The CG-15 concentrations used were 60, 30, 15, and 0 mg Au/L, and those of CS-20 were 10, 5, 2.5, and 0 mg Ag/L. The CG-15 or CS-20 concentration was constant in the rows, and that of CTAB was constant in the columns. The last, 12th column was toxicant free and was used as a control (Supplementary Data). This experimental design was called by us a 2D test system. For photometric evaluation, we used the change in absorbance that occurred on day 2 of the 2-day experiment:   ΔD=D48(c)−D24(c). (2) In determining the lethal/median effective concentration (LC50), and the absolute LC100, the effective concentration was that which caused a decrease in absorbance relative to the control and an increase in the number of dead (no red fluorescence) cells in the field of view of the fluorescence microscope. Each experiment was repeated at least twice at different times. Statistics. The obtained results were statistically processed by the standard procedures integrated in Excel 2007 software (Microsoft Corp). After the arithmetic mean and the SD for a given data sample had been found, we determined the SE of the arithmetic mean and its confidence limits with account of Student’s t coefficient (n, p) [number of measurements n = 3, significance level = 95% (P = .05)]. The significance of differences between individual samples was evaluated by a 2-sample unpaired Student’s t test with unequal variances. Differences were considered significant when the experimentally found Pexp value was ≤ .05. The reliability of the changes recorded in the results of each of the experiments described above for the entire set of toxicant concentrations examined was assessed by 1-way ANOVA by using Fisher’s ratio test. The dependences found were considered significant at F > Fcrit, where the critical value Fcrit at n = 3 and m = 4–5 (number of data sets) corresponded to P = .05 (with the number of degrees of freedom (df) lying between 4 and 14) and the effective value Peff was < .05. RESULTS Colorimetric Evaluation of the Microalgal Concentration Measurement of the algal suspension absorbance with the Multiskan Ascent reader at 690 nm indicated that the concentration of D. salina in the inoculation culture (end of the logarithmic growth phase) had been estimated correctly. In the D690 range 0–1.2, the dependence of absorbance on the suspension dilution coefficient proved almost linear when the wells contained 300 µl of the suspension (Supplementary Data). In the bottom center of the well (Supplementary Data), a correct estimation of cells with an inverted microscope is possible only up to absorbance values of ≤ 0.2, because at high concentrations, the images of individual cells overlap to a large extent. Nevertheless, changes in the number of dead cells can be reliably recorded in fluorescence microscopic images both in nonfixed cultures after a 24-h exposure and in fixed samples after a 48-h exposure at toxicant concentrations of ≤ LC50. Figure 2 shows the absorption spectra for a D. salina suspension in the culture medium at the end of the logarithmic growth phase and the absorption spectra for a dimethyl sulfoxide extract of the same cells. The spectra were taken with two instruments: Specord M40 and Multiskan Ascent. The circles show the absorbance values for the Multiskan reader at the corresponding filter passbands; absorbance was normalized to a 1-cm optical path. FIG. 2. View largeDownload slide Absorption spectra of a Dunaliella salina suspension that were taken on a Specord M40 (solid line) and a Multiskan Ascent (circles) instrument. Dashed line, dimethyl sulfoxide extract. FIG. 2. View largeDownload slide Absorption spectra of a Dunaliella salina suspension that were taken on a Specord M40 (solid line) and a Multiskan Ascent (circles) instrument. Dashed line, dimethyl sulfoxide extract. Between 500 and 700 nm, the spectra recorded with the 2 instruments coincide well. It should be noted that the filter passband of 690 nm is in the wing of the long-wavelength peak of suspension absorption and that the peaks of the chlorophyll extract are located in the regions between the standard light filters of the Multiskan Ascent reader. Toxicity of the Colorless Toxicants Figure 3 shows changes in the δD (1) of the D. salina suspension versus the hydrogen peroxide concentration in the incubation medium for different exposure times (24 and 48 h). At concentrations higher than 30 mg H2O2/L, the δD24 and δD48 values prove negative (Figure 3A), indicating that > 50% of the population died. Calculation of LC50 as the point of intersection with the zero-value curve δD shows an LC50 value of 30 ± 4.97 mg H2O2/L according to measurements of δD48. FIG. 3. View largeDownload slide Change in the δD of the D. salina suspension at different concentrations of H2O2 [D0 = 0.09, D24(0) = 0.12, D48(0) = 0.37] (A) and CH2O [D0 = 0.065, D24(0) = 0.11, D48(0) = 0.28] (C). The number of cells in the fluorescence microscope field in the bottom center of the well for H2O2 (B) and CH2O (D). Blue (dead cells) and red (live cells) fluorescence after a 24-h exposure. The experiments were conducted in triplicate, and results are shown as the means and SDs. FIG. 3. View largeDownload slide Change in the δD of the D. salina suspension at different concentrations of H2O2 [D0 = 0.09, D24(0) = 0.12, D48(0) = 0.37] (A) and CH2O [D0 = 0.065, D24(0) = 0.11, D48(0) = 0.28] (C). The number of cells in the fluorescence microscope field in the bottom center of the well for H2O2 (B) and CH2O (D). Blue (dead cells) and red (live cells) fluorescence after a 24-h exposure. The experiments were conducted in triplicate, and results are shown as the means and SDs. At 25 mg H2O2/L, the number of dead (no red fluorescence) cells increased significantly (from 4.5 ± 2.12 at 20 mg H2O2/L to 32.5 ± 4.95), and at 30 mg H2O2/L, the numbers of blue and red cells were approximately the same. Probably, this concentration can be considered median (LC50). The change in absorbance at this point (Figure 3A) indicated that the cell death and growth rates were approximately equal because the zero value of the absorbance increment was retained at both 24- and 48-h exposures. The total number of cells being recorded is maximal at this point, and when the toxicant concentration is raised, the number of viable cells declines to the point of disappearance. At 35 mg H2O2/L, the number of viable cells was maximal, decreasing slightly (possibly owing to cell lysis) as the H2O2 concentration was further increased. Supplementary Data provides images of D. salina cells after a 24-h exposure to different concentrations of H2O2. Live cells of D. salina were very motile, swimming actively all over the well. At a photoexposure time of approximately 0.3 s, they left track images of up to 4 times their own sizes. Therefore, account was taken of only nonmotile cells, which were in the bottom center of the well and did not leave tracks. The images of substrate-attached live cells making active vibratory movements proved blurred (Supplementary Data). Figures 3B and D show the results of photometric and microscopic measurements of microalgal suspensions at 0–100 mg CH2O/L. The extrapolation curves for 24- and 48-h exposures cross the x-axis (δD = 0) between 3 and 5 mg CH2O/L, respectively (Figure 3B). The microscopic results (Figure 3D) also show that the 24-h LC50 for formaldehyde ranges from 3 to 4 mg CH2O/L. The number of dead cells increased significantly (from 10 ± 1.41 at 1.4 mg CH2O/L to 19.5 ± 1.41 at 2.8 mg CH2O/L). At the same concentration, the number of live cells increased considerably (from 61 ± 7.07 to 218 ± 67.9), possibly owing to the loss of motility. In the next wells (5.6 mg CH2O/L), the maximum total number of cells was recorded, with an approximately 85% cell death, whereas at 11.2 mg CH2O/L, almost all of the cells died. Thus, the CH2O concentrations acting on D. salina lie in the range 2–12 mg CH2O/L, with a median value of about 4 mg CH2O/L. Photometric and microscopic experiments with CTAB (Supplementary Data) showed that the median LC50 value is about 0.6 mg CTAB/L. Thus, determination of the toxicity of the colorless pollutants with the proposed test system presents no particular problems. The values of LC100 and LC50 can be found from plots of spectrophotometric measurements: to LC100 there correspond the points at which the extrapolation curve δDt(c) reaches a plateau in the region of negative values, and the concentration corresponding to LC50 is found from the point of intersection with the zero-value curve δD. Toxicity of the Ag Preparations A slightly more complicated case is that of Ag preparations that possess the plasmon resonance of metallic nanoparticles and the photosensitivity of salts, presupposing that changes in toxicant absorbance will make a solid contribution to the total extinction of light by the suspension being tested. The measured results for the Ag samples used are given in Figure 4. FIG. 4. View largeDownload slide Change in the δD of the Dunaliella salina suspension at different concentrations of AgNO3 [D0 = 0.08, D24(0) = 0.21, D48(0) = 0.45] (A), PrAg [D0 = 0.08, D24(0) = 0.23, D48(0) = 0.44] (C), and CS [D0 = 0.08, D24(0) = 0.22, D48(0) = 0.45] (E). The number of cells in the fluorescence microscope field in the bottom center of the well for AgNO3 (B), PrAg (D), and CS (F). Blue (dead cells) and red (live cells) fluorescence after a 24-h exposure. The experiments were conducted in triplicate, and results are shown as the means and SDs. FIG. 4. View largeDownload slide Change in the δD of the Dunaliella salina suspension at different concentrations of AgNO3 [D0 = 0.08, D24(0) = 0.21, D48(0) = 0.45] (A), PrAg [D0 = 0.08, D24(0) = 0.23, D48(0) = 0.44] (C), and CS [D0 = 0.08, D24(0) = 0.22, D48(0) = 0.45] (E). The number of cells in the fluorescence microscope field in the bottom center of the well for AgNO3 (B), PrAg (D), and CS (F). Blue (dead cells) and red (live cells) fluorescence after a 24-h exposure. The experiments were conducted in triplicate, and results are shown as the means and SDs. With AgNO3, we observed an extended zone of sublethal concentrations (4–18 mg Ag/L; Figure 4A). The photometric data fluctuated at positive values close to 0. To an LC50 value of about 4 to 6 mg Ag/L, found by counting live and dead cells (Figure 4B), there corresponded a critical break of the photometric curves and achievement of a plateau. A similar picture emerged in experiments with PrAg that were run with a linear step of 1 mg Ag/L (Figs. 4C and D). However, the LOEC was slightly higher (approximately 2 mg Ag/L) and the zone of half-lethal concentrations was broader (4–11 mg Ag/L). The LC100, determined in an experiment with a large step, was > 18 mg Ag/L. Significant fluctuations were observed for CS-20 (Figs. 4E and F). It can also be noted that photometry yielded a slightly inflated estimate of toxicity when compared with microscopy. Specifically, δD24 and δD48 values of ≤ 0 were observed at about 100 and 140 mg Ag/L, respectively, whereas fluorescence microscopy revealed insignificant changes in the numbers of live and dead cells across the range 40–200 mg Ag/L (Figure 4E). Possibly, these differences in estimating the LC50 were due either to the uncontrolled contribution of Ag nanoparticle aggregates to light absorption or to an insufficiently correct microscopic estimate of the viability of live cells exposed to Ag for 24 h. The values of SDs indicate a large uncertainty in the measuring accuracy of the LC50. Therefore, we should only note the semiquantitative analysis in this work. Nevertheless, such an approach is considered to be acceptable for comparison of the effects of allied toxicants. To test this assumption, we ran a comparative experiment to investigate the toxicity of CS-20 and its supernatant liquid from centrifugation at 13 000 × g for 15 min. Atomic absorption spectroscopy data on the content of Ag in stock solutions showed that the supernatant liquid contained an amount of Ag that was approximately 13% of that in the initial CS-20 solution. In this case, microscopic evaluation was done on a culture fixed with glutaraldehyde after a 48-h exposure. The photometric and microscopic data are given in Figure 5. FIG. 5. View largeDownload slide Change in the δD of the Dunaliella salina suspension at different concentrations of CS-20 [D0 = 0.1, D24(0) = 0.17, D48(0) = 0.39] (A) and the supernatant liquid [D0 = 0.1, D24(0) = 0.14, D48(0) = 0.32] (B). The number of cells in the fluorescence microscope field in the bottom center of the well for CS-20 (C) and the supernatant liquid (D). Blue (dead cells) and red (live cells) fluorescence after exposure for 48 h and fixation with glutaraldehyde. The experiments were conducted in triplicate, and results are shown as the means and SDs. FIG. 5. View largeDownload slide Change in the δD of the Dunaliella salina suspension at different concentrations of CS-20 [D0 = 0.1, D24(0) = 0.17, D48(0) = 0.39] (A) and the supernatant liquid [D0 = 0.1, D24(0) = 0.14, D48(0) = 0.32] (B). The number of cells in the fluorescence microscope field in the bottom center of the well for CS-20 (C) and the supernatant liquid (D). Blue (dead cells) and red (live cells) fluorescence after exposure for 48 h and fixation with glutaraldehyde. The experiments were conducted in triplicate, and results are shown as the means and SDs. In this experiment, the LC50 estimates were not significantly different between photometry and microscopy: 75 mg Ag/L for CS-20 and 20 mg Ag/L for the supernatant liquid. These data indicate that the highly disperse and complex ionic Ag species in the supernatant liquid are much more toxic than the colloidal disperse CS-20. Toxicity of the CTAB/CG And CTAB/CS Mixtures Visual evaluation of the results of the 2D experiment at 0–2 mg CTAB/L and 0–60 mg Au/L showed that after a 48-h exposure, the green hue persisted in all rows, beginning with column 6—that is, at concentrations lower than 1 mg CTAB/L (Supplementary Data). The standard photometric evaluation adopted in this study proved noninformative owing to considerable fluctuations of D0(c) (Supplementary Data), caused by the rapid aggregation of CG-15 in the salt environment and by the associated optical changes. The change in absorbance on day 2 of the experiment, ΔD=D48(c)−D24(c), was used as an evaluation characteristic (Figure 6). FIG. 6. View largeDownload slide The absorbance increase ΔD = D48(c) – D24(c) on day 2 of the experiment versus the cetyltrimethylammonium bromide concentration at different concentrations of CG-15. The experiments were conducted in duplicate, and results are shown as the means and SDs. FIG. 6. View largeDownload slide The absorbance increase ΔD = D48(c) – D24(c) on day 2 of the experiment versus the cetyltrimethylammonium bromide concentration at different concentrations of CG-15. The experiments were conducted in duplicate, and results are shown as the means and SDs. Presenting photometric characteristics as an absorbance increase on the final day of the experiment enables the data scatter to be decreased greatly. This is due to the characteristic time of CG aggregation being much shorter than the characteristic time of the optical changes caused solely by the cell’s contribution. Analysis of the ΔD change showed that at all CG-15 concentrations, the extrapolation curves cross the line of zero ΔD values at 0.45–0.55 mg CTAB/L. With increasing CG-15 concentration, the intersection points shift slightly to lower CTAB concentrations. However, the data obtained by cell counting in the fluorescence microscope field showed that all mixtures produced approximately the same effect (Supplementary Data). With account taken of the data scatter, it can be concluded that CG exerts no synergistic effect on CTAB toxicity to D. salina when the alga is grown with high NaCl. A different picture was observed at 0–1 mg CTAB/L and 0–10 mg Ag/L of CS-20 (Figure 7). As with CTAB/CG, LC50 points are observed in the vicinity of the zero values of the y-axis. From the counts of live and dead cells after a 48-h exposure (fixed cultures), viability (Vi) was 57% ± 9% at 10:0.3 mg Ag/L:mg CTAB/L, 46% ± 1.5% at 5:0.4 mg Ag/L:mg CTAB/L, 54% ± 13% at 2.5:0.5 mg Ag/L:mg CTAB/L, and 61% ± 40% at 0:0.7 mg Ag/L:mg CTAB/L (Figure 7, arrows). Thus, both components of the CTAB/CS-20 mixture exerted a pronounced unidirectional toxic effect on D. salina. FIG. 7. View largeDownload slide The absorbance increase ΔD = D48(c) – D24(c) on day 2 of the experiment versus the cetyltrimethylammonium bromide concentration at different concentrations of CS-20. The experiments were conducted in duplicate, and results are shown as the means and SDs. Arrows indicate viability from the counts of dead and live cells. FIG. 7. View largeDownload slide The absorbance increase ΔD = D48(c) – D24(c) on day 2 of the experiment versus the cetyltrimethylammonium bromide concentration at different concentrations of CS-20. The experiments were conducted in duplicate, and results are shown as the means and SDs. Arrows indicate viability from the counts of dead and live cells. A comparison of the photometric and microscopic data indicated that the half-lethal concentrations of CTAB were effective at minimal positive ΔD(48–24) values for 10 and 5 mg Ag/L and at minimal negative ΔD(48-24) values for 2.5 and 0 mg Ag/L. In addition, the total number of cells recorded at Vi ≤ 50% was about 1.5-fold smaller for CTAB than it was for the CTAB/CS. We do not quite understand why this should be so and can only note that at a CTAB concentration of ≥ LC50 in Ag-free samples, the shape and size of viable cells changed dramatically: they became much smaller and highly elongated, with less cell volume occupied by chloroplast. Conversely, in CTAB/CS samples, very large, roundish cells were often observed among the surviving cells (Supplementary Data). DISCUSSION The absorption spectrum for D. salina suspensions is determined mostly by the content of chlorophyll a found in the cup-shaped chloroplast, which occupies the greater part of a drop-shaped vegetative cell. The absorption peaks for chlorophyll a are observed at 430 and 670 nm. However, owing to the presence of protein-pigment complexes (photosystems), the absorption spectra for green plants are broadened considerably. It follows from Figure 2 that the chlorophyll absorption peaks were in the regions between the standard light filters of the Multiskan Ascent reader. Nevertheless, despite the substantial discrepancies between the data for the short-wavelength part of the spectrum, the absorbance values at 540, 620, and 690 nm, recorded with the different instruments, were fairly close. Considering that the contribution of light scattering to the total extinction of light by a cell suspension increases with increasing number of lysed cells, we chose the filter passband of 690 nm as the major one to construct plots of absorbance change versus toxicant concentration. Overall, our results agree well with the data from Mecozzi et al. (2008) vis spectroscopic measurements of suspensions and acetone extracts of D. tertiolecta, showing that the short-wavelength absorption peak of a microalgal suspension is only slightly more accurate in determining the cell concentration than is the long-wavelength peak, as it shows a smaller SE of determination. The microplate variant of measuring microalgal viability by chlorophyll absorption enables one to avoid extraction of this pigment and to evaluate the kinetics of its photodegradation, which accompanies cell death. Intact cells retain their color, and the green pigment concentration is proportional to the number of viable cells. This allows one to judge the number of living organisms and the kinetics of population development. Cell death, recorded photometrically, can be readily seen as early as after 24 h of observations. In wells with dead cells, the suspension loses its color. The dead cells are observed with an inverted fluorescence microscope as objects with blue (and without red) fluorescence under appropriate exciting light, and their number increases as the toxicant concentration is raised. Correct counting of live cells is possible only in chemically fixed cultures after complete sedimentation. Another difficulty is that in control samples, the cells form > 1 layer at the bottom of the well. Even in such images, however, individual dead cells can be differentiated with sufficient reliability and can, therefore, be counted. To make photometric measurements correct, it is necessary that the following conditions be met: (1) the spectral properties of chromophores should be unchanged and (2) the uncontrolled contribution of light scattering to the total extinction of the light flux should be minimized. To check the correctness of the microplate method of absorbance measurement, we performed a series of parallel measurements of the same D. salina suspension diluted successively with the culture medium. For the microplate reader, the measured results indicated that absorbance depended almost linearly on the extent of dilution (Supplementary Data). Determining the concentration with the Bio-Rad TC20 cell counter proved much more labor-consuming and less rapid. In addition, such determination involves many systemic and accidental artifacts—suffice it to say that aggregates of Au and Ag nanoparticles often give a false-positive signal. To check that the spectral characteristics of the system remained uncharged, we conducted experiments with 2 colorless toxicants: formaldehyde and hydrogen peroxide. It is evident that the zero increment in absorbance as a first approximation should attest that a stationary process is taking place—that is, the rate of cell growth is equal to that of cell death. Exceeding the control values at subthreshold toxicant concentrations (lower than the minimum effective ones) indicates the effect of hormesis (growth promotion). Such behavior of D. tertiolecta populations was noted by Mecozzi et al. (2008) for standard pollutants and by Morelli et al. (2013) in their study of microalgal response to exposure to CdSe/ZnS quantum dots. Finally, negative values of absorbance change indicate that population death exceeds population growth. In accordance with the material safety data sheet for hydrogen peroxide, it is toxic to Chlorellavulgaris at a 72-h IC50 of 4.2 mg H2O2/L and its NOEC is 0.1 mg H2O2/L. From the data available in the IUCLID database, the 240-h LC/EC50 for C. emersonii is 17 mg H2O2/L. From the safety data sheet of ZEP Inc and from other data sheets, the 72-h EC50 is 1.2 mg H2O2/L for D. tertiolecta. Our 48-h LC50 value of approximately 30 mg H2O2/L for D. salina is much greater than the values available from other sources. This could be related to the higher tolerance of the hyperhalophilic alga D. salina for oxidative stress toxicants, as compared with that of other species of this genus (Arun et al., 2014). Formaldehyde, which acts as a chemical fixer when used at high concentrations, was chosen by us as another colorless xenobiotic. From the Merk material safety data sheet, formaldehyde is toxic to the green alga Pseudokirchneriella subcapitata at a 96-h EC50 of approximately 22 mg CH2O/L. Formaldehyde toxicity to D. salina determined in our test system was a 24-h LC50 of 6.8 mg CH2O/L. Elucidating the mechanisms governing the toxicity of various pollutants and the causes for the different tolerances of organisms belonging to different taxons will be the subject of our future research. For now, we can only say that the toxicological metrics are similar within one order of magnitude. CTAB can also be considered a colorless toxicant. From close analogies, we succeeded in finding data on the toxicity of cetyltrimethylammonium chloride (CTAC) to D. bardawil, with a 240-h IC50 of 2.8 ± 1.49 mg CTAC/L (Qv, Jiang, 2013). CTAB and CTAC have approximately the same levels of toxicity, at least to C. vulgaris, as found in a study of the toxicity of quaternary ammonium compounds to this microalga (Zhu et al., 2010). In this work, D. salina was much more sensitive to CTAB (48-h LC50 of about 0.7 mg CTAB/L), possibly owing to the hyperosmotic conditions for D. salina growth, which cause osmotic stress when the barrier function of the cytoplasmic membrane is disrupted. In evaluating the toxicity of the Ag samples, we expected that the photometric evaluation of the experimental results would not be devoid of complications. We conducted experiments with 3 colloidal Ag preparations: CS-20, PrAg, and AgNO3 (in a NaCl environment). CS-20 is an unprotected (electrostatically stabilized) hydrosol of metallic Ag with a particle size of about 20 nm (a product of reduction of Ag nitrate with EDTA in a weakly alkaline environment). It retains its aggregation stability only when stored in hermetically sealed flasks under low ionic force. The plasmon resonant peak of CS-20 is near 400 nm. When rapid aggregation occurs, the spectra of CS-20 appear considerably broadened, ultimately achieving an almost neutral spectral character. PrAg is a commercial Ag oxide product stabilized with serum albumin, with a powder Ag content of 8.3%. Aqueous solutions of PrAg are resistant to aggregation and sedimentation in a broad range of acidity and ionic force values. It is evident that at 1.5 M NaCl, AgNO3 forms an AgCl suspension that does not possess plasmon resonant properties and, consequently, does not absorb in the visible part of the spectrum. Overall, our data on nano-Ag toxicity agree nicely with the recent findings of Hazani et al. (2013), who reported a CS LC50 value of approximately 50 mg Ag/L for C. vulgaris and D. tertiolecta. The major results for the toxicity of plasmon resonant nanoparticles to various organisms have been reviewed by Khlebtsov and Dykman (2011). In this study, we were first interested in the limitations imposed on the colorimetric detection system and associated with the use of particles with a pronounced plasmon resonance. Account was taken of the difficulties involved in evaluating spectra for such particles when they aggregate (Khlebtsov et al., 2004). The most illustrative in this respect are electrostatically stabilized CG sols, which change their color as a result of both salt and surfactant-induced aggregation. Nevertheless, evaluation of the absorbance change that occurred on day 2 of exposure of D. salina to CTAB/CG showed good agreement with the microscopic counts of live and dead cells. Moreover, we were unable to detect any statistically significant differences in determinations of the LC50 of CTAB/CG-15 with 0 to 60 mg Au/L. In the 2D CTAB/CS test system, the picture was radically different. The CS-20 concentration of 10 mg Ag/L, which had only a minimal effect (Figs. 4 and 5), substantially increased the toxicity to D. salina when CS-20 was mixed with CTAB. The half-lethal effect was achieved with 0.3-mg CTAB/L for mixtures with Ag and with 0.7-mg CTAB/L for mixtures without Ag. Such behavior of the system is quite understandable if one considers that complex ionic Ag species are much more toxic than colloidal disperse species (Figure 5) and that the solubility of metallic Ag increases in solutions containing CTAB (Pal et al., 1997). CONCLUSIONS We have developed a fairly simple and rapid system for testing the toxicity of solvable and nanomaterial toxicants. The best reproducibility is achieved when the growth rate of the control is the greatest possible (a 6-fold increase in absorbance within 48 h). In parallel experiments of the given run, the test system demonstrates good reproducibility. The results of independent experiments carried out at different times, may, however, differ. Initially, the aggregation of nanoparticles and the associated color changes seemed to us as a limiting factor for our colorimetric test system. However, the selected algorithms of measurement of the relative changes in absorbance allowed us to colorimetrically evaluate cell death in the case of products containing silver, which had a spectrally neutral color changes in the red band of chlorophyll absorption. In the case of gold nanoparticles, when the condition of the spectral neutrality is not met, it was still possible to estimate the cell death colorimetrically but only on the second day of the 48-h experiment. The most promising appears to be the 2D test system for analyzing toxicant mixtures, which can be used in searching for both effective growth-suppressing mixtures and antidotes. The test system can be improved further with the aid of an in-depth cytological analysis and study of population and organismal changes taking place in Dunaliella microalga under the effect of various toxicants. ACKNOWLEDGMENTS The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper. Funding This work was supported by the Russian Scientific Foundation (project no. 14-13-01167). We are grateful to Prof. Y.V. Balnokin for consultations on the cultivation of microalgae. We are grateful to Prof. S. Y. Shchyogolev for statistical consultation. We also thank Mr. D. N. Tychinin for his help in preparation of the manuscript. The work by L.A.D. was supported by RFBR grant no. 14-04-00114. REFERENCES Alkilany A. M. Murphy C. J. ( 2010). Toxicity and cellular uptake of gold nanoparticles: what we have learned so far. J Nanopart Res 12, 2313– 2333. Arun N. Laxmi, V. Singh D. P. ( 2014). Chromium (VI) induced oxidative stress in halotolerant alga Dunaliella salina and D. tertiolecta isolated from sambhar salt lake of Rajasthan (India). Cell Mol Biol  60, 90– 96. Google Scholar PubMed  Blaise C Vasseur P. ( 2005). Algal microplate toxicity test. In Small-scale Freshwater Toxicity Investigations (C. Blaise, and J.-F. F象rd, Eds.),  Vol. 1 pp. 137-179. Springer, Dordrecht, The Netherlands. Dreaden E. C. Alkilany M. A. Huang X. Murphy C. J. El-Sayed M. A. ( 2012). The golden age: gold nanoparticles for biomedicine. Chem Soc Rev  41, 2740– 2779. Google Scholar CrossRef Search ADS PubMed  Dykman L. A. Khlebtsov N. G. ( 2012). Gold nanoparticles in biomedical applications: recent advances and perspectives. Chem Soc Rev  41, 2256– 2282. Google Scholar CrossRef Search ADS PubMed  Fabrikanos A. Athanassiou S. Lieser K. H. ( 1963). Darstellung stabiler Hydrosole von Gold und Silber durch Reduktion mit Äthylendiamintetraessigsäure. Z Naturforschg  18, 612– 617. Frens G. ( 1973). Controlled nucleation for the particle size in monodisperse gold suspensions. Nature Phys Sci  241, 20– 22. Google Scholar CrossRef Search ADS   Hazani A. A. Ibrahim M. M. Shehata A. I. El-Gaaly G. A. Daoud M. Fouad D. Rizwana H. Moubayed N. M. S. ( 2013). Ecotoxicity of Ag-nanoparticles on two microalgae, Chlorella vulgaris and Dunaliella tertiolecta. Arch Biol Sci  65, 1447– 1457. Google Scholar CrossRef Search ADS   Khlebtsov N. G. Dykman L. A. ( 2010). Optical properties and biomedical applications of plasmonic nanoparticles. J Quant Spectrosc Radiat Transfer  111, 1– 35. Google Scholar CrossRef Search ADS   Khlebtsov N. G. Dykman L. A. ( 2011). Biodistribution and toxicity of engineered gold nanoparticles: a review of in vitro and in vivo studies. Chem Soc Rev  40, 1647– 1671. Google Scholar CrossRef Search ADS PubMed  Khlebtsov N. G. Melnikov A. G. Dykman L. A. Bogatyrev V. A. ( 2004). Optical properties and biomedical applications of nanostructures based on gold and silver bioconjugates. In Photopolarimetry in Remote Sensing  ( Videen G. Yatskiv Y. Mishchenko M., Eds.), pp. 265– 308. Kluwer Academic Publishers, Dordrecht. Kooij E. S. Ahmed W. Hellenthal C. Zandvliet H. J. W. Poelsema B. ( 2012). From nanorods to nanostars: Tuning the optical properties of gold nanoparticles. Colloids Surf a  413, 231– 238. Google Scholar CrossRef Search ADS   Li N. Zhao P. Astruc D. ( 2014). Anisotropic gold nanoparticles: synthesis, properties, applications, and toxicity. Angew Chem Int Ed  53, 1756– 1789. Google Scholar CrossRef Search ADS   Marquis B. J. Love S. A. Braun K. L. Haynes C. L. ( 2009). Analytical methods to assess nanoparticle toxicity. Analyst  134, 425– 439. Google Scholar CrossRef Search ADS PubMed  Masyuk N. P. ( 1973). Morphology, Systematics, Ecology, Geographical Distribution of Dunaliella Teod. Genius and Trends of its Practical Application. Naukova Dumka, Kiev, Ukraine (in Russian). Masyuk N. P. Posudin Y. I. Lilitskaya G. G. ( 2007) Photomovement of the Cells of Dunaliella Teod. (Dunaliellales, Chlorophyceae, Viridiplantae). National Agrarian Univ, Kiev, Ukraine. (in Russian). Mecozzi M. Onorati F. Oteri F. Sarni A. ( 2008). Characterisation of a bioassay using the marine alga Dunaliella tertiolecta associated with spectroscopic (visible and infrared) detection. Int J Environ Pollut  32, 104– 120. Google Scholar CrossRef Search ADS   Morelli E. Salvadori E. Bizzarri R. Cioni P. Gabellieri E. ( 2013). Interaction of CdSe/ZnS quantum dots with the marine diatom Phaeodactylum tricornutum and the green alga Dunaliella tertiolecta: A biophysical approach. Biophys Chem  182, 4– 10. Google Scholar CrossRef Search ADS PubMed  Oukarroum A. Bras S. Perreault F. Popovic R. ( 2012). Inhibitory effects of silver nanoparticles in two green algae, Chlorella vulgaris and Dunaliella tertiolecta. Ecotoxicol Environ Saf  78, 80– 85. Google Scholar CrossRef Search ADS PubMed  Pal T. Sau T. K. Jana N. R. ( 1997). Reversible formation and dissolution of silver nanoparticles in aqueous surfactant media. Langmuir  13, 1481– 1485. Google Scholar CrossRef Search ADS   Qiu Y. Liu Y. Wang L. Xu L. Bai R. Ji Y. Wu X. Zhao Y. Li Y. Chen C. ( 2010). Surface chemistry and aspect ratio mediated cellular uptake of Au nanorods. Biomaterials  31, 7606– 7619. Google Scholar CrossRef Search ADS PubMed  Qv X. -Y. and Jiang J. -G. ( 2013). Toxicity evaluation of two typical surfactants to Dunaliella bardawil, an environmentally tolerant alga. Environ Toxicol Chem  32, 426– 433. Google Scholar CrossRef Search ADS PubMed  Segovia M. Haramaty L. Berges J. A. Falkowski P. G. ( 2003). Cell death in the unicellular chlorophyte Dunaliella tertiolecta. A hypothesis on the evolution of apoptosis in higher plants and metazoans. Plant Physiol  132, 99– 105. Google Scholar CrossRef Search ADS PubMed  Sapsford K. E. Algar W. R. Berti L. Gemmill K. B. Casey B. J. Oh E. Stewart M. H. Medintz I. L. ( 2013). Functionalizing nanoparticles with biological molecules: developing chemistries that facilitate nanotechnology. Chem Rev  113, 1904– 2074. Google Scholar CrossRef Search ADS PubMed  Shaish A. Avro M. Ben-Amotz A. ( 1990). Effect of inhibitors on the formation of stereoisomers in the biosynthesis of β-carotene in Dunaliella bardawii. Plant Cell Physiol  31, 689– 696. Sharifi S. Behzadi S. Laurent S. Laird Forrest M. Stroeve P. Mahmoudi M. ( 2012). Toxicity of nanomaterials. Chem Soc Rev  41, 2323– 2343. Google Scholar CrossRef Search ADS PubMed  Zamani H. Jahromi A. M. Dehdashti H. Sheikhi M. H. ( 2014). Influence of PbS nanoparticle polymer coating on their aggregation behavior and toxicity to the green algae Dunaliella salina. Aquat Toxicol  154, 176– 183. Google Scholar CrossRef Search ADS PubMed  Zhu M. Fei G. Zhu R. Wang X. Zheng X. ( 2010). A DFT-based QSAR study of the toxicity of quaternary ammonium compounds on Chlorella vulgaris. Chemosphere  80, 46– 52. Google Scholar CrossRef Search ADS PubMed  © The Author 2016. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com
TI - Colorimetric Evaluation of the Viability of the Microalga Dunaliella Salina as a Test Tool for Nanomaterial Toxicity
JF - Toxicological Sciences
DO - 10.1093/toxsci/kfw023
DA - 2016-02-10
UR - https://www.deepdyve.com/lp/oxford-university-press/colorimetric-evaluation-of-the-viability-of-the-microalga-dunaliella-AVvkO9jZ8D
SP - 115
EP - 125
VL - 151
IS - 1
DP - DeepDyve
ER -