Exploring the Toxicology of Depleted Uranium with Caenorhabditis elegans.

Depleted uranium (DU) is an emerging heavy metal pollutant with considerable environmental and occupational concerns. Its radiotoxicity is known to be low. However, its chemical toxicity should not be ignored. In order to explore the chemical toxicity of DU, the effects of uranyl nitrate, prepared from DU, on the model organism Caenorhabditis elegans were investigated. Chronic exposure to DU did not affect the lifespan or reproduction of the worm. DU had little effect on the physiological processes of C. elegans. Additionally, DU treatment did not make C. elegans more susceptible to UV, heat, or oxidative stress. Interestingly, chronic exposure of DU decreased the in vivo reactive oxygen species-scavenging ability through inhibiting the expression of antioxidant genes ctl-1, ctl-2, ctl-3, gst-7, and gst-10. Chronic but not acute exposure of DU induced a statistically significant degeneration of the dopaminergic (DAergic) neurons of treated worms and promoted the increase of α-synuclein aggregation and DAergic neurotoxicity. These findings may raise the public concerns regarding DU as an etiologic agent of Parkinson's disease and underline its potential neurotoxicity.

Introduction

Depleted uranium (DU) is a by-product left over when natural uranium is enriched by increasing the proportion of the isotope 235U for use in nuclear reactors and nuclear weapons.[1,2] The major use of DU is in the military as an alloy for armor and ammunition because of its extremely dense and pyrophoric properties.[3] Uranium may be released into the environment as it is mined, processed, and applied. While DU is less radioactive than natural uranium, it retains the chemical toxicity of uranium.[4,5] As a heavy metal,[6] the acute (≤24 h exposure) chemical toxicity of DU is well characterized, and the kidneys are its most vulnerable target.[5] However, the chronic (>24 h to several days exposure) DU toxicology has not been well studied, and the findings are not always consistent.[4] Whether or not DU causes neurodegenerative diseases is not clear.

In order to investigate the chronic toxicology of DU, herein, the nematode Caenorhabditis elegans was chosen as the model organism, which offers several advantages, including a completely sequenced genome,[7] freely available numerous multicolor reporter constructs,[8] transparent body, a rapid replication cycle, and the ease of growing and maintenance as well as manipulation.[9,10] In contrast to rodents which have 10,000–20,000 DAergic neurons, or humans which have greater than 40,000 DAergic neurons,[11]C. elegans have only eight DAergic neurons: two anterior deirid, four cephalic, and two posterior deirid neurons,[12] which makes the in situ investigation of the vulnerability of DAergic neurons upon DU exposure possible.

Results

DU has Little Effect on the Lifespan and Reproduction of C. elegans

The nematode, C. elegans, is an excellent model organism in the toxicological studies.[9,10] First, the lifespan was used as an endpoint to measure the effect of chronic DU exposure on the general health of C. elegans. DU concentrations of 0.01 mM (Figure A), 0.1 mM (Figure B) and 1 mM (Figure C) did not affect the lifespan of the wild-type N2 C. elegans compared to the untreated control. It should be pointed out that the highest DU concentration (1 mM) tested was relevant to the typical uranium contamination observed in soil (0.2–4.2 mM).[13]

Figure 1

Effects of DU on the lifespan and reproductive ability of C. elegans. (A–C) Survival curve of worms treated with DU, the curve is representative of three independent experiments. Worms were synchronized to the L4/young adult stage, and subsequently exposed to different concentrations of DU (0.01, 0.1, or 1 mM). (D) The number of offspring embryos of DU-treated worms. (E–G) Curves of the reproductive span of DU-treated worms. (H) Images of germ cells stained by AO. Error bars represented the SEM of three independent replicates. ns, not significant.

The effect of DU treatment on the wild-type N2 C. elegans reproduction was also investigated. DU did not affect the fertility (Figure D). The egg-laying pattern was not notably affected by treatment with 0.01 mM (Figure E), 0.1 mM (Figure F), or 1 mM (Figure G) DU. The effect of DU on the reproductive system was investigated further. As shown in Figure H, no signs of germ cell corpses were observed in the acridine orange (AO) staining, which showed that DU (1 mM) did not induce germ cell apoptosis of worms.

Together, these results demonstrate that DU does not affect the lifespan and reproduction of C. elegans.

DU has Little Effect on the Physiological Processes of C. elegans

To investigate whether DU affects the physiological processes of C. elegans, the locomotive activity and the pharyngeal pumping of the wild-type N2 C. elegans were measured at days 3, 5, 7, and 9, where the L4 stage was defined as day 0. As the animal aged, its head began thrashing, which was quantitatively measured as a body movement, and it progressively decreased with aging (Figure A). Because 1 mM DU had little effect on the lifespan of C. elegans after more than 3 weeks of exposure, 0.01 and 0.1 mM DU that are lower than the typical uranium contamination observed in soil (0.2–4.2 mM)[13] were not tested in the following assays. DU (1 mM) treatment did not decrease the body bends of worms at days 3, 5, and 7, while it slightly decreased the body bends of worms at day 9 compared to the untreated control (Figure A). The pharynx is a neuromuscular organ that undergoes rhythmic contractions.[14] Pharyngeal pumping could be counted, and the rate displayed an age-related decline (Figure B). DU (1 mM) did not decrease the pharyngeal pumping rate of worms at all time points compared to the untreated control (Figure B). Taken together, DU showed little effect on the physiological processes of C. elegans.

Figure 2

Effects of DU on the physiological processes of C. elegans. (A) Body bends of worms treated with or without 1 mM DU. (B) Pharyngeal pumping rate of worms treated with or without 1 mM DU. All the worms in the experiment were synchronized with the L4/young adult stage and subsequently exposed to 1 mM DU. Error bars represented the SEM of three independent replicates. ns, not significant; *p < 0.05.

DU has Little Effect on C. elegans’ Response to UV, Heat, and Oxidative Stress

Because DU had little effect on the general health and physiological processes of C. elegans under normal conditions, whether DU (1 mM) could affect C. elegans under stress conditions was subsequently examined. DU treatment did not affect the survival curves of the wild-type N2 C. elegans upon exposure to UV (Figure A), heat stress (Figure B), or oxidative stress (Figure C) compared to the untreated control. These results indicate that DU does not affect C. elegans’ response to various kinds of unfavorable stress.

Figure 3

Effect of DU on C. elegans’ response to UV, heat, or oxidative stress. (A–C) Survival curve of N2 wild-type worms treated with DU under UV stress (A), heat stress (B), and oxidative stress (C). All the worms in the experiment were synchronized with the L4/young adult stage, and subsequently exposed to 1 mM DU. Stress assays were performed at day 3 postadult stage.

DU Increase of α-Synuclein Aggregation and DAergic Neurotoxicity

Uranium crosses the blood–brain barrier[15] and may result in neurotoxicity and neurodegeneration.[16] Because DU does not affect the general health and the physiological processes of C. elegans as well as their responses to unfavorable environmental stress, whether DU (1 mM) could affect the progression of Parkinson’s disease (PD) was tested. The NL5901 strain, which was created by inserting the human α-synuclein gene with an YFP fusion construct driven by the unc-54 promoter, was used as a PD model. As shown in Figure A, DU (1 mM) slightly shortened the lifespan of NL5901 worms when compared to the untreated control. NL5901 worms exhibited the aggregation of α-synuclein (Figure B). The YFP intensities were quantitatively analyzed and the mean fluorescence intensity in the control group was set to 1. DU (1 mM) treatment did not increase the YFP intensities at days 3 and 5, while it increased the α-synuclein-YFP at day 7 when compared to the untreated control (Figure C), which implies that chronic exposure of DU promotes the development of PD.

Figure 4

Effect of DU on the aggregation of α-synuclein in NL5901 worms. (A) Survival curve of NL5901 worms treated with DU. (B) Endpoint fluorescence microscopy image of α-synuclein aggregates in NL5901 worms treated with or without DU at day 3, 5, and 7 postadult stage. (C) Quantitative analysis of α-synuclein aggregates in NL5901 worms. All the worms in the experiment were synchronized with the L4/young adult stage, and subsequently exposed to 1 mM DU. Error bars represented the SEM of three independent replicates of total worms. ns, not significant; ***p < 0.001.

PD is typically associated with degeneration of DAergic neurons.[17] Therefore, whether DU (1 mM) could lead to DAergic neurodegeneration was investigated. The BZ555 strain, in which all DAergic neurons are tagged with GFP by fusing with DAT-1, was used to visualize the bodies and processes of DAergic neurons. BZ555 worms were treated with DU (1 mM) and their green fluorescence was analyzed in the nerve ring, which contains GFP-tagged DAergic neurons (Figure A). DU (1 mM) treatment did not affect the soma of DAergic neurons when compared to the untreated control at days 1 and 2 (Figure B), while it induced a statistically significant shrinkage of the soma of DAergic neurons of treated worms at day 3 (Figure B), which indicates that chronic exposure of DU leads to the degeneration of DAergic neurons. To confirm the generality of the results from the transgene BZ555 worms, the effect of DU on the functionality of DAergic neurons was further evaluated by the slowing response assay in wild-type N2 worms. As shown in Figure C, DU (1 mM) treatment did not affect the functionality of DAergic neurons at day 1, as revealed by the difference in the average number of body bends per 20 s between N2 worms in OP50-seeded plates and plates without food, while it induced statistically significant inhibition of the functionality of DAergic neurons of the DU-treated worms at days 2 and 3. Together, these results demonstrate that chronic DU exposure induces DAergic neuron degeneration and therefore impairs their functionality.

Figure 5

Effect of DU on the DAergic neurons. (A) Endpoint fluorescence microscopy image of DAergic neurons in BZ555 worms treated with or without DU at day 1, 2, and 3 postadult stage. All the worms in the experiment were synchronized with the L1 larval stage, and subsequently exposed to 1 mM DU. (B) Quantitative analysis of the intensity of DAergic neurons shown in (A). (C) Difference in the average number of body bends per 20 s between N2 worms in OP50-seeded plates and plates without food. N2 worms were synchronized with the L1 larval stage, and subsequently started to expose to 1 mM DU for 1–3 days. 6-OHDA treated N2 worms were used as positive control. Data were expressed as mean ± SEM. ns, not significant; **p < 0.01; ***p < 0.001.

DU Decreases the Expression of Reactive Oxygen Species-Scavenging Genes

Reactive oxygen species (ROS), which are key modulators in the development of PD, cause serious damage to and death of DAergic neurons.[18] In order to better understand how DU affects PD development, the effect of DU on in vivo antioxidant genes was investigated. Superoxide dismutases (SODs),[19] glutathione S-transferases (GSTs),[20] and catalases (CTLs)[21] are the main ROS-scavenging enzymes involved in cell detoxification processes.[22] Therefore, the effects of DU on the expression of SODs, GSTs, and CTLs were measured. Chronic exposure of DU decreased the CTL family (ctl-1, ctl-2, and ctl-3) mRNA expression (Figure A). In the GST family, DU treatment reduced the mRNA expressions of gst-7 and gst-10, while DU did not affect gst-1, gst-4, and gst-20 gene expression (Figure B). In contrast, DU did not affect the SOD family (sod-1, sod-2, sod-3, sod-4, and sod-5) mRNA expression (Figure C). These data show that DU decreases the in vivo ROS-scavenging ability by inhibiting the expression of antioxidant genes ctl-1, ctl-2, ctl-3, gst-7, and gst-10, which would cause the death of DAergic neurons and promote the PD development.

Figure 6

Effect of DU on the expression of ROS-scavenging genes. The mRNA expressions of the superoxide dismutase (SOD) family (A), catalase (CAT) family (B), and GST family (C) were measured by qRT-PCR. Error bars represent the SEM of three independent replicates. ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001.

Discussion

Natural uranium consists of three isotopes, 238U (99.27%), 235U (0.72%), and 234U (0.0054%).[23] DU is what remains after the removal of enriched 235U, and may also be produced from the reprocessing of spent nuclear reactor fuel.[24] Compared with natural uranium, DU contains a less fissile isotope 235U. Therefore, DU has low radioactivity and is not usually considered to exert significant radiotoxicity.[3] In order to systemically characterize the chemical toxicity of uranium, the effects of uranyl nitrate, prepared from DU, on the model organism C. elegans were investigated.

DU had little effects on the lifespan, reproduction, and physiological processes of C. elegans under normal conditions. DU treatment did not make C. elegans more susceptible to UV, heat, or oxidative stress, either. They appear to be nonhazardous for C. elegans upon exposure to DU.

Emerging bodies of evidence suggest that heavy metal toxicity promotes the progress of neurodegenerative diseases.[25,26] PD is characterized by degeneration of dopaminergic neurons containing the aggregated α-synuclein protein.[27] Currently, multiple C. elegans PD models, notably the dopaminergic neurons and α-synuclein-related transgenic C. elegans PD models, have been used to study the effects of compounds on PD.[28] Uranium is a heavy metal[29] and the impact of DU on the progression of PD was investigated with C. elegans. Chronic exposure of DU (1 mM) decreased the expression of ROS-scavenging genes, led to the degeneration of DAergic neurons, and promoted the increase of α-synuclein aggregation and DAergic neurotoxicity. These results are different from a previous study where Jiang et al reported that DU had low neurotoxic potential and did not show any significant neurodegeneration following DU (1 mM) treatment.[30] It should be pointed out that a single end-point (24 h) measurement was made in Jiang et al’s study. Herein, 7 days, but not 3 days or 5 days, of DU (1 mM) treatment increased α-synuclein-YFP in NL5901 worms; 3 days, but not 1 day or 2 days, of DU (1 mM) treatment caused the shrinkage of the soma of DAergic neurons in BZ555 worms; 2 and 3 days, but not 1 day, of DU (1 mM) treatment impaired the functionality of DAergic neurons in wild-type N2 worms. These were in agreement with Jiang et al’s observation of low acute (24 h) neurotoxic potential of DU. Multigenerational tests of uranium exposure on C. elegans showed rapid phenotypic changes.[31] Most likely, more significant effects on the neuronal damage would be expected for C. elegans after multigenerational exposure to DU, which is worthy of further investigation.

This study found that chronic exposure of uranium caused the degeneration of DAergic neurons, which is consistent with the epidemiologic investigation that the Gulf War veterans exposed to DU had a highly increased prevalence of PD.[32] It should be noted that the sporadic PD is a slowly progressive disorder, which usually takes years/decades to show significant clinical signs.[33] Most likely, the chronic exposure but not the acute toxicology of DU is closely associated with the development of PD.

Conclusions

Overall, this study found that DU did not affect the lifespan and reproduction of C. elegans. Additionally, DU had little effect on the physiological processes of C. elegans. Moreover, DU did not affect the C. elegans’ response to UV, heat, or oxidative stress. However, chronic exposure of DU led to the degeneration of DAergic neurons and promoted the increase of α-synuclein aggregation and DAergic neurotoxicity. These findings may raise public concerns regarding DU as an etiologic agent of PD and underline its potential neurotoxicity. However, all these data are only suggestive until they can be tied to meaningful human research. Therefore, the effects of chronic DU exposure on the central nervous system and neurodegenerative diseases warrant further exploration.

Experimental Section

Chemicals

DU (UO2(NO3)2·6H2O, ≥99.9%) was purchased from Chushengwei Chemical Co., Ltd. (Wuhan, Hubei, China). The radioactivity of UO2(NO3)2·6H2O was checked with a Geiger counter to be in the range of 0.2–0.3 μSv/h, which was not significantly different from the in-house background radioactivity.

Worm Strains, Culture, and Synchronization

The worms were grown on a solid nematode growth medium (NGM) plate in an incubator. To synchronize, 50 worms at day 1 of the postadult period were picked in a new NGM plate and incubated for 6 h at 20 °C to lay eggs. The following strains were used in this work: Bristol strain N2 was used as the wild-type NL5901 [unc-54p::alpha synuclein::YFP + unc-119(+)], BZ555[dat-1p::GFP]. All the organisms were obtained from the Caenorhabditis Genetics Center (CGC) at the University of Minnesota. All strains were grown and incubated at 20 °C on NGM plates with Escherichia coli (E.coli) OP50.

Lifespan Measurement

Healthy worms at the young adult stage were placed on NGM plates with E. coli OP50 and different concentrations of DU. To prevent the production of offspring, 5-fluorodeoxyuridine was added. Each group contained 50 nematodes. Worms were transferred to new plates every two days. The number of survivors was recorded daily until all the worms died. The lifespan is defined as the time period between the young adult stage and death. The experiment was repeated three times independently.

Reproduction Assay

For reproduction assay, age-synchronized L1 stage nematodes were transferred to fresh plates, which was followed by DU exposure. At the L4 larval stage, the parent worm was transferred to a fresh exposure plate every 12 h and the number of offspring at all stages is counted and compared with the control group. The experiment was independently repeated three times for the reproduction test. Thirty worms were examined with three replicates.

AO Staining

Germ cell corpses were measured by AO staining. Age-synchronized L1 stage nematodes were transferred to fresh plates, which was followed by DU exposure. At day 1 postadult stage, the treated worms were stained for 1 h in the dark at 20 °C by transferring worms into a plate containing 500 μL of 25 μg/mL AO and OP50 in M9 buffer and then transferred to the NGM plate for 40 min for recovery on bacterial lawns.

Worms were then mounted onto 2% agarose pads and immobilized with 2 mM levamisole. To monitor the germ cell, worms were microscopically visualized and photographed using a fluorescence microscope. The apoptotic cells appeared yellow or yellow-orange after AO staining, representing increased DNA fragmentation, while intact cells were uniformly green in color.

Locomotion and Feeding Behavior Assay

Locomotion behavior was quantified by monitoring body thrashing. To assess body thrashing, the wild-type C. elegans (N2) were incubated for 6 h at 20 °C and allowed to lay eggs, then the synchronized worms at the L4 larval stage were treated with or without 1 mM DU. Worms at days 3, 5, 7, and 9 after the adult stage treated with or without DU were washed with M9 buffer and subsequently transferred to a slide glass containing 100 μL of M9 buffer. After 1 min recovery, body thrashes were counted for 1 min. A movement of the worm that swings its head and/or tail to the same side is counted as one thrash. Thirty worms were examined with three replicates. Feeding behavior was evaluated by the pharyngeal pumping rate, which was measured by counting the number of pharyngeal contractions for 30 s. The test was repeated three times with 10 randomly selected worms per treatment.

Stress Resistance Assays

To assess the stress resistance in solid culture medium, the wild-type N2 worms were incubated for 6 h at 20 °C and allowed to lay eggs, then the synchronized worms at the L4 larval stage were treated with DU until day 3 postadult stage. For oxidative stress, worms from each group were next transferred to plates containing 200 μM juglone (Sigma-Aldrich, St. Louis, MO, USA) in a NGM plate. To assess UV irradiation resistance, worms from each group were exposed to UV irradiation (254 nm) at a dose of 1000 J/m2 for 30 min and immediately transferred to an incubator set to 20 °C. To assess the heat shock resistance, the worms were transferred from an incubator set to 20 °C to the one set to 35 °C. The viable nematodes were scored every hour until all animals died. At least 150 worms were examined with three replicates.

α-Synuclein Aggregation Measurement

A transgenic strain, NL5901 [unc-54p::alpha synuclein::YFP + unc-119(+)], which stably expresses an unc-54:YFP fusion protein, was used to assess the PD in the worm induced by DU exposure. Briefly, NL5901 strain nematodes were exposed to DU from the young adult stage to days 3, 5, and 7 after the adult stage. At the end of exposure, worms were mounted onto 2% agarose pads and immobilized with 2 mM levamisole. To monitor the a-synuclein aggregation, the YFP protein was microscopically visualized and photographed using a Nikon Ts2-FL fluorescence microscope. At least 30 worms examined with three replicates were imaged, and the fluorescent signals were quantified in each worm using ImageJ software.

DAergic Neurodegeneration Measurement

A transgenic strain, BZ555, which expresses GFP in DAergic neurons through the dat-1::GFP reporter system, was used to assess the effect of DU on DAergic neurons; the worms were exposed to different concentrations of DU from the egg stage to day 1, 2, and 3 postadult stage on OP50 plates. At the end of exposure, worms were mounted onto 2% agarose pads and immobilized with 2 mM levamisole, and imaged with a Nikon Ts2-FL fluorescence microscope. DAergic neurons were counted by inspecting the GFP fluorescence, which could be quantified with ImageJ. At least 20 worms were examined with two replicates.

Basal Slowing Response

This assay was used to assess the effect of DU on the dopaminergic neuronal function. N2 worms were exposed to different concentrations of DU from the egg stage to day 1, 2, and 3 postadult stage on OP50 plates. At the end of exposure, worms were collected by washing the plate with M9 buffer; ten worms were transferred to OP50-seeded plates with ring-shaped OP50 lawn or to unseeded plates. After 5 min, the number of body bends was counted to assess the locomotor rate on OP50-seeded or unseeded plates for a 20 s duration. Data are expressed as the difference (Δ) in body bends per 20 s between worms in OP50-seeded plates and unseeded plates. 6-OHDA (6-hydroxydopamine)-treated N2 worms, whose DAergic neurons were injured, were used as positive control.

qRT-PCR

Wild-type C. elegans (N2) were incubated for 6 h at 20 °C and allowed to lay eggs, then the synchronized worms at the L1 larval stage were treated with or without 1 mM DU. Then, approximately 1000 worms were suspended in 1 mL M9 buffer, and the worms were washed with M9 buffer 3 times. Total RNAs from worms were prepared by using the TRIzol Reagent kit (Takara Bio, Kusatsu, Japan) according to the manufacturer’s instructions. The cDNA was generated with oligo(dT) primers (Promega, Madison, WI, USA) by using the reverse transcription system (Promega, Madison, WI, USA). The quantitative real-time PCR (qRT-PCR) was carried out using the SYBR Green real-time PCR master mix (Roche, Basel, Switzerland) on a TOptical Real-Time PCR system (Analytik Jena AG, Analytik Jena, Germany). The mRNA expression levels of genes were normalized to actin-1. The primer sequences for PCR are shown in Table .

Table 1

Primers Used in qRT-PCR

primers	sequences (5′-3′)
gcs-1 forward	aatcgattcctttggagacc
gcs-1 reverse	atgtttgcctcgacaatgtt
gst-4 forward	Cccattttacaagtcgatgg
gst-4 reverse	Cttcctctgcagtttttcca
gst-7 forward	Aggacaacagaatcccaaagg
gst-7 reverse	Agcaaatcccatcttcaccat
gst-10 forward	Gtctaccacgttttggatgc
gst-10 reverse	Actttgtcggcctttctctt
gst-20 forward	tttggagtcccgaactgag
gst-20 reverse	ttctagacagctcttcgcc
act-1 forward	tcggtatgggacagaaggac
act-1 reverse	catcccagttggtgacgata
sod-1 forward	ctcatggtggaccaaaatcc
sod-1 reverse	cgtgtcggtgagcttgatt
sod-2 forward	ggcatcaactgtcgctgttc
sod-2 reverse	gcacatgtggcaaccttcaa
sod-3 forward	tggctaaggatggtggagaa
sod-3 reverse	gccttgaaccgcaatagtgat
sod-4 forward	cagaccaatcaaaaacaactgg
sod-4 reverse	tgagacgcagacagtgagaca
sod-5 forward	cacacttcaatccatgcaaaa
sod-5 reverse	acgtttccaaggtcaccaac
ctl-1 forward	gaatgtgaagaattatttcgctga
ctl-1 reverse	aactcgattcctgggacgat
ctl-2 forward	caaggaactacttcgctgagg
ctl-2 reverse	aatgagtgtcggtgtacgagaa
ctl-3 forward	gaatgtgaagaattatttcgctga
ctl-3 reverse	aactcgattcctgggacgat

Statistical Analysis

Statistical analysis was performed using GraphPad Prism 5 software (GraphPad Software, La Jolla, CA, USA). Data were presented as mean ± SEM. For qRT-PCR assays, statistical analysis was done by Student’s t-test. For stress resistance and lifespan assays, statistics was analyzed by the log-rank test. For the α-synuclein and DAergic neuron fluorescence images, fluorescence intensities were quantified using ImageJ software, and the statistics was performed by Student’s t-test.