Combined effects of alpha particles and depleted uranium on Zebrafish (Danio rerio) embryos.

The combined effects of low-dose or high-dose alpha particles and depleted uranium (DU) in Zebrafish (Danio rerio) embryos were studied. Three schemes were examined-(i) [ILUL]: 0.44 mGy alpha-particle dose + 10 µg/l DU exposure, (ii) [IHUH]: 4.4 mGy alpha-particle dose + 100 µg/l DU exposure and (iii) [IHUL]: 4.4 mGy alpha-particle dose + 10 µg/l DU exposure-in which Zebrafish embryos were irradiated with alpha particles at 5 h post fertilization (hpf) and/or exposed to uranium at 5-6 hpf. The results were also compared with our previous work, which studied the effects of [ILUH]: 0.44 mGy alpha-particle dose + 100 µg/l DU exposure. When the Zebrafish embryos developed to 24 hpf, the apoptotic signals in the entire embryos, used as the biological endpoint for this study, were quantified. Our results showed that [ILUL] and [IHUL] led to antagonistic effects, whereas [IHUH] led to an additive effect. The effect found for the previously studied case of [ILUH] was difficult to define because it was synergistic with reference to the 100 µg/l DU exposure, but it was antagonistic with reference to the 0.44 mGy alpha-particle dose. All the findings regarding the four different schemes showed that the combined effects critically depended on the dose response to each individual stressor. We also qualitatively explained these findings in terms of promotion of early death of cells predisposed to spontaneous transformation by alpha particles, interacting with the delay in cell death resulting from various concentrations of DU exposure.

INTRODUCTION

Organisms living in the environment are simultaneously exposed to various stressors (e.g. ionizing radiation, heavy metals, etc.), and a good understanding of the combined effects is necessary for realistic risk assessments [1-7]. The combined effects can be more complicated than the simple sum of the effects from individual stressors. To date, the most commonly identified combined effects have included additive, synergistic and antagonistic effects. Interestingly, most investigations of combined effects were based on high-dose (HD) effects of individual stressors [3, 5, 8–13]. However, evidence has been accumulated showing that HD effects can be different from low-dose (LD) effects. For example, an hormetic response, characterized as a biphasic dose–response relationship exhibiting a low-dose stimulation and a high-dose inhibition, has been widely accepted as a universal phenomenon for various stressors [14-15].

In the present study, the combined effects of LD or HD exposure to uranium metal [in the form of depleted uranium (DU)] and exposure to alpha particles were examined. People are exposed to a range of doses of uranium, which occurs naturally in the environment, and its distribution may be influenced by human activities, such as the coal and phosphate industries, military application or nuclear fuel cycle. The man-made enrichment of 235U (which is the only naturally fissile isotope in the isotopic composition of uranium relative to the other isotopes, i.e. 238U and 234U) is a critical process in nuclear power generation and nuclear weapon production, and creates mixtures of U isotopes that may be more or less radioactive in terms of both alpha and gamma radiation types (see Ref. [16] for details). We note that DU is a byproduct of the enrichment process.

Interest in the chemical toxicity of DU has increased since the Gulf War because DU was extensively used in various armor munitions as the strengthening material [17]. High natural groundwater uranium concentrations found at some locations, e.g. in the USA [18] and Finland [19], may also contribute to the human population exposure. Alpha-particle irradiation is also ubiquitous in our natural environment, e.g. from the decay of radon progeny [20]. Extensive previous research has investigated the biological effects of radon on humans [21-22], and these studies have indicated that radon progeny in air could lead to health issues for humans, particularly lung cancers [23-25]. Besides affecting the human respiratory tract, radon can also be dissolved in blood and move within the human body through the blood circulatory system. It is now well established that alpha particles from radon progeny contribute the largest natural radiation dose to human beings [22, 23]. Taken together, it becomes apparent that people and animals that live in areas with high natural uranium concentrations in groundwater [18, 19] will experience exposures to both alpha particles and DU. Hence, the combined effects from DU and alpha-particle irradiation have real-life relevance.

Exposure to uranium compounds in which the uranium isotopic composition is natural or depleted can produce effects in mammalian organs and organ systems (such as the kidney, the central nervous system, the lung and the liver), largely depending on the chemical dose [26]. There have been reports showing that the kidney is the most sensitive organ to uranyl compounds and is the critical target for acute uranium toxicity [27-29]. Uranium may interact with DNA in a variety of ways. Stearns et al. [30] reported the formation of uranium–DNA adducts and mutations in CHO EM9 mammalian cells after directly exposing the cells to a DU compound. Uranium could also react with DNA through hydrolysis of the DNA phosphate backbone [31]. Yazzie et al. [31] demonstrated that DU in uranyl acetate caused DNA single-strand breaks in the presence of vitamin C. Another study suggested that the formation of hydroxyl radicals were responsible for oxidative DNA damage in the presence of reactions with uranyl ion, hydrogen peroxide and ascorbate [32]. Moreover, DU induced genomic instabilities, such as delayed reproductive death and micronuclei formation in cells [33].

Recently, our group examined the combined effects from HD DU (concentration = 100 μg/l) and LD alpha-particle irradiation (dose = 0.44 mGy) on embryos of the Zebrafish, Danio rerio [34]. We demonstrated that the LD effect induced by the first stressor (alpha-particle irradiation) in terms of a reduction in the number of apoptotic signals, when combined with the HD effect of the second stressor (DU) in terms of an increase in the number of apoptotic signals, led to apparently even more apoptotic signals than the HD effect of the second stressor. We proposed to explain this combined effect by suggesting the alpha particle–induced early death of cells predisposed to spontaneous transformation [14] and DU-induced delayed cell death [33]. We further commented that explorations with different combinations of LD or HD exposures to alpha particles and DU would allow studies of the various manifestations of the combined effect of these two stressors and thus help elucidate the general mechanisms underlying these various manifestations.

Therefore, the present work aimed to further study the combined effects from LD or HD exposures to alpha particles and DU, i.e. [LD of alpha particles + LD of DU], [HD of alpha particles + HD of DU] and [HD of alpha particles + LD of DU]. We continued our use of embryos of the Zebrafish in our study of the combined effects. Our group had also previously employed the embryos of the Zebrafish for studying combined effects resulting from alpha particles and cadmium [35-37]. In recent years, there have been a growing number of research projects using Zebrafish or their embryos to investigate the toxicity of various environmental stressors. One of the major reasons for this is that the Zebrafish and human genomes share considerable homology, including conservation of most DNA repair–related genes [38].

In summary, the present work showed that [LD of alpha particles + LD of DU] led to an antagonistic effect, [HD of alpha particles + HD of DU] led to an additive effect and [HD of alpha particles + LD of DU] led to an antagonistic effect. In relation, the previously studied theme [LD of alpha particles + HD of DU] [34] led to a new undefined combined effect.

MATERIALS AND METHODS

Ethics statement

The animal studies were approved by the Department of Health, Government of the Hong Kong Special Administrative Region, Reference No. (13–7) in DH/HA&P/8/2/5 Pt.1, and were performed in accordance with the guidelines. The embryos were anesthetized using 0.0016 M tricaine before their apoptotic signals were counted under a fluorescent microscope.

Experimental animals

Adult male and female Zebrafish were mixed and kept in fish tanks where the water was kept at 28°C. To maintain a good and stable production of embryos, the fish were maintained with a light–dark cycle of 14 h of light followed by 10 h of dark periods. A specially designed plastic embryo collector was used to collect embryos when the 14 h photoperiod began. To ensure the synchronization of their developmental stages, the embryos were collected within a period of 30 min. All collected embryos were then kept in deionized water and in an incubator maintained at 28°C until they reached 4 h post fertilization (hpf). At 4 hpf, corresponding to the sphere stage of the blastula period, healthy and well developed embryos were selected under a stereomicroscope, and were then transferred into a new Petri dish with a thin agarose gel layer at the bottom filled with E3 medium (5 mM NaCl, 0.33 mM MgSO4, 0.33 mM CaCl2, 0.17 mM KCl and 0.1% methylene blue), for dechorionation.

Exposure protocols

The present work aimed to study the outcomes of various experimental schemes: [IXUY],which involved (up to 240 s irradiation period resulting in an alpha-particle dose level ‘X’ at 5 hpf) + (1 h exposure period to a waterborne uranium concentration level ‘Y’ from 5 to 6 hpf), where ‘X’ could be either ‘L’, representing a LD of alpha-particle irradiation (0.44 mGy), or ‘H’, representing a HD of alpha-particle irradiation (4.4 mGy); ‘Y’ could be either ‘L’, representing a LD exposure to uranium (10 µg/l), or ‘H’ representing a HD exposure to uranium (100 µg/l).

Using the same nomenclature system, the experimental scheme adopted in our previous study [34] would be [ILUH]. In the present work, the three different experimental schemes were: A total of three to four replicate experiments were performed for each condition. For each replicate experiment, 40 dechorionated embryos were employed. These embryos were separated into four groups, each having 10 embryos, and accommodated in four separate Petri dishes lined with a thin layer of biocompatible agarose on the bottom. The four groups were as follows.

Condition 1: [ILUL] (i.e. LD of alpha particles + LD of DU)

Condition 2: [IHUH] (i.e. HD of alpha particles + HD of DU)

Condition 3: [IHUL] (i.e. HD of alpha particles + LD of DU)

The Control group

In this group, ‘[C]’, the embryos were dechorionated and received no further treatment.

Alpha-particle–irradiated and uranium-dosed group

In this group, [IXUY], the dechorionated embryos received level X alpha-particle irradiation at 5 hpf + level Y uranium exposure for 1 h (from 5 to 6 hpf), as described above.

Alpha-particle–irradiated groups

In this group, [IX], the dechorionated embryos received level X alpha-particle irradiation at 5 hpf, as described above.

Uranium-dosed group

In this group, [UY], the dechorionated embryos received level Y uranium exposure for 1 h (from 5 to 6 hpf), as described above.

A volume of 3 ml of medium was used in each Petri dish. Since the DNA repair mechanism in Zebrafish embryos would start to operate after the cleavage stages (0.7–2.2 hpf) [39], exposures of embryos to alpha-particle radiation and/or DU started at 5 hpf, which was well after the cleavage stages. The embryos in the IX and UY groups were exposed to alpha-particle radiation or uranium, respectively, whereas the embryos in the IXUY group were exposed to both alpha-particle radiation and DU. The experimental flow for the treatment procedures of embryos in the IXUY group for the three different experimental schemes (Conditions 1–3) is shown in Fig. 1.

Fig. 1.

Experimental flow for treatment procedures of embryos. The experimental flow for the treatment procedures of embryos in the alpha-particle–irradiated and uranium-dosed group (IXUY) for the three different experimental schemes (Conditions 1 to 3).

Alpha-particle irradiation

In the present experiment, the setup for alpha-particle irradiation largely followed that designed by Yum et al. [40]. A biocompatible Mylar film (Dupont, Hong Kong) with a thickness of 3.5 μm was used as the support substrate during irradiation. The Mylar film was glued using an epoxy (Araldite Rapid, England) to the bottom of a Petri dish that had a hole with a diameter of 35 mm at the center. An 241Am source with alpha-particle energy of 5.49 MeV under vacuum and an activity of 4.26 kBq was employed in the present work. At 5 hpf, the embryos in the IX and IXUY groups were transferred onto the substrate in the irradiation dish and irradiated with alpha particles for 24 or 240 s, which corresponded to absorbed doses of ∼0.44 and ∼4.4 mGy, respectively. According to Yum et al. [41], the LD effect (∼0.44 mGy) and HD effect (∼4.4 mGy) of alpha particles in Zebrafish embryos were opposite in terms of the changes in the number of apoptotic signals within the embryos.

The two groups of embryos, (IL or IH) and (ILUL or IHUH or IHUL), were irradiated with alpha particles coming from below through the Mylar film, instead of coming from above (to avoid the problem of having different travelling distances of alpha particles in the medium before reaching the embryos). All embryos were orientated in such a way that all the cells of the embryos faced downwards towards the Mylar film so that the alpha particles would be directed towards the cells. Similar treatment and experimental procedures were followed for embryos in the UY group, except that the 241Am source was not used; this was referred to as ‘sham-irradiation’. After irradiation or sham-irradiation for the IXUY and UY groups, respectively, the embryos were then exposed to uranium.

Uranium exposure

In the present project, uranium exposure was achieved using uranyl acetate UO2(CH3COO)2·2H2O (Electron Microscopy Sciences). For each set of experiments, a new uranium stock solution was prepared to avoid uncertainties in the uranium concentration in different sets of experiments due to potential precipitation. Although making the stock solution in an acid like HCl could have kept the uranium concentration more stable, the acidic conditions could have affected the development of the Zebrafish embryos. To ensure that all the uranyl acetate was dissolved, the stock solution was prepared one day before performing each set of experiments. The uranyl acetate stock solution, with a concentration of between 0.15 and 0.30 g/l, was prepared by dissolving the uranyl acetate with MilliQ water. Since uranyl acetate was sensitive to light and would precipitate if exposed, all stock solution was kept at 4°C in the dark. On the day of the experiment, the stock solution was then further diluted to the desired concentration, which was either 10 µg/l (for Conditions 1 and 3) or 100 µg/l (for Condition 2) of uranium (in this study).

A volume of 3 ml of uranyl acetate working solution, which was sufficient to cover all embryos accommodated in each dish, was prepared in two new Petri dishes. Immediately after the IXUY and UY groups were irradiated and sham irradiated with alpha particles, respectively, the embryos were removed carefully with a glass dropper from the medium and transferred to a uranium solution to receive the uranium dosage. The embryos were kept in the uranium solution for 1 h in dark. After 1 h, the embryos were removed from the uranium solution. After washing with 6 ml of clean medium, the embryos were transferred to new Petri dishes with 3 ml of clean medium. All four groups of embryos (C, IXUY, IX and UY) were then returned to the 28°C incubator for further development.

Considering the specific activity of DU, the ranges in water of the alpha particles emitted by the relevant uranium isotopes and daughters, as well as the volume of the sensitive cells in the 5-hpf Zebrafish embryos, etc., the radiation dose received by the Zebrafish embryos due to DU exposure was many orders of magnitude lower than the radiation dose due to alpha particles emitted from the 241Am source described above. As such, the radiation dose from DU exposures could be safely neglected.

Quantification of apoptosis by vital dye staining

In the present project, apoptosis was chosen as the biological endpoint. Since the apoptotic activity in Zebrafish embryos was high before 24 hpf (due to organogenesis processes [42]), and pigment development on the Zebrafish embryos after 24 hpf presented problems for observing the signals from the apoptotic cells [43], we chose to examine the apoptotic signals in Zebrafish embryos at 24 hpf in the present work. When the Zebrafish embryos had developed to 24 hpf, the apoptotic signals in the entire embryos were quantified. Such an endpoint has been widely adopted for quantifying radiation effects in Zebrafish embryos [43, 44]. The staining procedure was as previously described by Choi et al. [45]. The four groups of embryos in the present study were transferred into a medium containing 2 µg/ml of a vital dye acridine orange (AO; Sigma, St Louis, MO, USA) (which has been commonly employed to quantify the level of apoptosis in Zebrafish embryos [46-48]) to stain in the dark for 60 min; they were then thoroughly washed twice in the culture medium. AO is a nucleic-acid selective fluorescent cationic dye. AO can cross the plasma membrane of viable and early apoptotic cells [49] and can interact with DNA and RNA by intercalation or electrostatic attractions, respectively. Hence, after AO staining, all nuclei of viable and early apoptotic cells would appear in green under a fluorescent microscope; apoptotic cells give much more intense green fluorescence as a result of chromatin condensation, which is a hallmark event for a cell undergoing apoptosis [50, 51]. In the present study, ‘apoptosis signals’ referred to the observed numbers of cells that were undergoing apoptosis. The apoptotic signals of the embryos were then counted under a fluorescent microscope after anesthetizing the embryo with 0.0016 M tricaine (Sigma, St Louis, MO, USA). For every single embryo, three images focusing on different sections of the embryos were captured under the fluorescent microscope (using a filter for green fluorescence and a magnification of ×40), and these three images were then combined into one signal image for quantification of apoptosis signals. All images were captured using identical exposure parameters. Only bright green spots within the embryos were scored. The computer program Particle Counting 2.0 (developed by J. Zhang) was employed to count the apoptotic signals in an embryo.

Data analysis

For each experimental scheme (i.e. Conditions 1 to 3 described in the section ‘Experiment protocols’), three to four sets of experiments, each with 40 Zebrafish embryos, were carried out on different days. The number of apoptotic signals for each whole Zebrafish embryo was counted as described above. For each condition, NC was interpreted as the average background apoptotic signal for the embryos in the corresponding set of experiments, and thus the net apoptotic signals for the IXUY, IX and UY groups could then be described as NIXUY Net = (NIXUY – NC), NIX Net = (NIX – NC) and NUY Net = (NUY – NC), respectively. Thus the normalized net apoptotic signal for these groups of the embryos could be determined by NIXUY † = [NIXUY Net/NC], NIX † = [NIX Net/NC] and NUY † = [NUY Net/NC]. The normalized net data for the all replicates were then grouped for further analyses. The data for each group (IXUY, IX and UY) were analyzed in turn as follows. As a first step, the statistical significance of the differences between the four conditions and the control (adding up to five conditions) was assessed using one-way analysis of variance (ANOVA). Cases with P values ≤0.05 were considered to correspond to statistically significant differences between at least two of the compared conditions. In such cases, post-hoc t-tests were then performed to assess the statistical significance of the difference between the control and each studied condition. A P value ≤0.005 (i.e. 0.05/10, where 10 was the total number of possible combinations of two-condition comparisons out of five conditions) was considered to correspond to a statistically significant difference between the control group and the studied condition. The combined effects due to the exposure to both alpha-particle radiation and uranium was characterized by comparing the IXUY group with the corresponding ‘reference’ groups through the Student's t-test, where the ‘reference’ groups were constructed based on the additive approach discussed in the following section. Cases with P values ≤0.05 were considered to correspond to statistically significant differences between the compared groups.

RESULTS

Three different experimental schemes, i.e. Condition 1 [ILUL], Condition 2 [IHUH] and Condition 3 [IHUL], were examined in the present work. The normalized net mean number of apoptotic signals (N ± SE) obtained from embryos in the C, IXUY, IX and UY groups in the three to four replicates of each experimental scheme have been plotted in Fig. 2 and recorded in Table 1 (where the results are represented as N ± SE, where SE is the standard error of the mean). Student's t-tests were used to determine the significance of the extent to which the IXUY, IX and UY groups differed from the corresponding C group of embryos under each set of experimental conditions. Cases with P ≤ 0.05 were considered statistically significant. The normalized net mean number of apoptotic signals (N ± SE) obtained in our previous study was also included for reference (in Fig. 2 and Table 1). In addition, results from one-way ANOVA for the five conditions (Control + Conditions 1–4) for each group (IXUY, IX and UY) and results from post-hoc t-tests (if P values were ≤0.05 from the one-way ANOVA) between the control and each studied Condition (1–4) were recorded in Table 2.

Fig. 2.

Normalized net mean numbers of apoptotic signals obtained for different groups of embryos. The normalized net mean numbers of apoptotic signals (N ± SE) obtained for embryos in the C, IXUY, IX and UY groups for the three to four replicates of the three experimental schemes (Condition 1 [ILUL], Condition 2 [IHUH] and Condition 3 [IHUL]). The results (for [ILUH]) from our previous study [34] were also included for comparison. *Cases with P ≤ 0.05 were considered statistically significant using Student's t-test by comparing with the corresponding C group of embryos under each experimental condition. aData extracted from Ref. [34].

Table 1.

Basal toxicity of DU and alpha particle exposures on embryos under each experiment condition

Condition 1 [ILUL]	Na = 4	NILUL†	NIL†	NUL†
		–0.15 ± 0.04	–0.28 ± 0.03	–0.29 ± 0.03
		nb = 36	n = 32	n = 37
		Pc = 0.008*	P = 2.96 × 10−7*	P = 3.61 × 10−7*
Condition 2 [IHUH]	N = 4	NIHUH†	NIH†	NUH†
		1.27 ± 0.12	0.59 ± 0.06	0.53 ± 0.09
		n = 37	n = 37	n = 33
		P = 1.70 × 10−13*	P = 1.26 × 10−10*	P = 3.32 × 10−6*
Condition 3 [IHUL]	N = 3	NIHUL†	NIH†	NUL†
		–0.01 ± 0.04	0.36 ± 0.09	–0.25 ± 0.05
		n = 29	n = 27	n = 27
		P = 0.424	P = 8.83 × 10−4*	P = 5.88 × 10−4*
Previous study [ILUH]d	N = 3	NILUH†	NIL†	NUH†
		0.41 ± 0.08	–0.16 ± 0.05	0.24 ± 0.05
		n = 30	n = 27	n = 30
		P = 7.65 × 10−5*	P = 0.019*	P = 0.001*

aN = number of replicates involved. bn = total number of embryos employed in that experimental group under each experimental condition. cP = P value obtained using Student's t-test by comparing with the corresponding C group of embryos under each experimental condition. dReference data obtained in Ref. [34]. *Cases with P ≤ 0.05 were considered statistically significant.

Table 2.

Results from one-way ANOVA on the five conditions (Control + Conditions 1 to 4) for each group (IXUY, IX and UY), and results from post-hoc t-tests (if P values ≤ 0.05 from the one-way ANOVA) between the control and each studied Condition (1 to 4)

	Condition 1 [ILUL]	Condition 2 [IHUH]	Condition 3 [IHUL]	Condition 4 [ILUH]a
IX	ANOVA: P = 8.17 × 10–30*
	P = 2.57 × 10–10**	P = 1.11 × 10–10**	P = 7.48 × 10–3**	P = 4.76 × 10–4**
UY	ANOVA: P = 2.10 × 10–26*
	P = 1.20 × 10–9**	P = 4.56 × 10–6**	P = 2.40 × 10–5**	P = 3.97 × 10–5**
IXUY	ANOVA: P = 1.87 × 10–49*
	P = 0.002**	P = 2.57 × 10–13**	P = 0.390	P = 1.43 × 10–5**

For cases with P ≤ 0.05 for the ANOVA, post-hoc t-tests were performed to further assess the differences between the control and each studied condition (1 to 4).

aReference data obtained in Ref. [34]. *Cases with P ≤ 0.05 were considered statistically significant. **Cases with P ≤ 0.005 (i.e. 0.05/10) were considered statistically significant.

Effects of high and low uranium exposures

In order to study the outcomes of the experimental schemes for [ILUL] (Condition 1) and [IHUL] (Condition 3), it was necessary to study the outcomes of the experimental scheme for [UL] to provide references for comparison (see the section ‘Combined effects of alpha particles and uranium’ below). It was interesting to observe that the normalized net mean number of apoptotic signals for the UL group of embryos was significantly smaller than that of the C group of embryos under both Conditions 1 and 3 (Table 1). Therefore, the LD effect of uranium (i.e. exposures to a low concentration of 10 μg/l) on Zebrafish embryos was similar to the LD effect of alpha particles.

Similarly, in order to study the outcomes of the experimental scheme for [IHUH] (Condition 2), it was necessary to study the outcomes of the experimental scheme for [UH] to provide references for comparison (see the section ‘Combined effects of alpha particles and uranium’ below). By comparing the C and UH groups, the effects of exposing Zebrafish embryos to a high uranium concentration were studied. It was found that the number of apoptotic signals of the UH group of embryos was significantly larger than that of the C group of embryos. A similar result was also obtained in our previous study [34], in which Zebrafish embryos at the same developmental stage were exposed to the same concentration of uranium for 1 h. Therefore, it could be concluded that exposing Zebrafish embryos to a high concentration of uranium (100 μg/l) from 5 to 6 hpf would lead to an increase in the number of apoptotic signals within the embryos.

Combined effects of alpha particles and uranium

Condition 1: [ILUL]

As described above, the normalized net apoptotic signal for these groups of the embryos could be determined by NILUL † = [NILUL/NC], NIL † = [NIL/NC] and NUL † = [NUL/NC]. All the normalized net data for the four sets of experiments were then grouped for further analyses. The combined effects were compared with the effects from individual stressors, as described by Ng et al. [37]. Briefly, two methods were employed to construct the expected mean number of apoptotic signals based on the effects contributed by individual stressors: The mean numbers of apoptotic signals obtained for these two groups, i.e. N L and N L, were equal by definition. To compare whether the UL or IL