Zebrafish exposure to environmentally relevant concentration of depleted uranium impairs progeny development at the molecular and histological levels

Abstract

Uranium is an actinide naturally found in the environment. Anthropogenic activities lead to the release of increasing amounts of uranium and depleted uranium (DU) in the environment, posing potential risks to aquatic organisms due to radiological and chemical toxicity of this radionucleide. Although environmental contaminations with high levels of uranium have already been observed, chronic exposures of non-human species to levels close to the environmental quality standards remain scarcely characterized. The present study focused on the identification of the molecular pathways impacted by a chronic exposure of zebrafish to 20 μg/L of DU during 10 days. The transcriptomic effects were evaluated by the use of the mRNAseq analysis in three organs of adult zebrafish, the brain the testis and the ovaries, and two developmental stages of the adult fish progeny, two-cells embryo and four-days larvae. The results highlight generic effects on the cell adhesion process, but also specific transcriptomic responses depending on the organ or the developmental stage investigated. The analysis of the transgenerational effects of DU-exposure on the four-day zebrafish larvae demonstrate an induction of genes involved in oxidative response (cat, mpx, sod1 and sod2), a decrease of expression of the two hatching enzymes (he1a and he1b), the deregulation of the expression of gene coding for the ATPase complex and the induction of cellular stress. Electron microscopy analysis of skeletal muscles on the four-days larvae highlights significant histological impacts on the ultrastructure of both the mitochondria and the myofibres. In addition, the comparison with the transcriptomic data obtained for the acetylcholine esterase mutant reveals the induction of protein-chaperons in the skeletal muscles of the progeny of fish chronically exposed to DU, pointing towards long lasting effects of this chemical in the muscles. The results presented in this study support the hypothesis that a chronic parental exposure to an environmentally relevant concentration of DU could impair the progeny development with significant effects observed both at the molecular level and on the histological ultrastructure of organs. This study provides a comprehensive transcriptomic dataset useful for ecotoxicological studies on other fish species at the molecular level. It also provides a key DU responsive gene, egr1, which may be a candidate biomarker for monitoring aquatic pollution by heavy metals.

Fig 1. Differential gene expression analysis in the adult tissues following a 10 days chronic exposure to DU.

(A-C) MA-plot for the differential analysis in the adult brain (A), testis (B) and ovaries (C). Expression for each gene is shown on the x-axis as the log10 of the mean normalized expression, and the log2 of the fold change on y-axis. Significant adjusted p-values (FDR) < = 0.01 are highlighted in yellow to red, and non-significant changes in grey. (D) Venn diagram of genes mis-regulated in the brain, testis and ovaries. All genes significantly regulated (up- or down- regulated) were used to compare the three tissues. (E) Fold change of genes involved in regeneration and cell adhesion in DU exposed brain and testis compared to controls (all adjusted p-values (FDR) < 0.01; error-bars represent the standard deviation to the mean).

Fig 2. Heatmap of GO terms enriched in DU-exposed adult’s tissues and their progeny.

The p-values from Fisher’s exact-test are indicated, as well as the associated GO terms. The datasets of up- and down-regulated genes used for the pathways analysis are indicated on the top of the heatmap (relative to fold change as DU exposed/control). The set of genes down-regulated in both adult brain and testis is also indicated. Yellow: Fisher’s exact test p-values < 0.01; black: non-significant enrichment.

Fig 3. Differential expression analysis in two-cells stage embryos and 96 hpf larvae obtained from DU-exposed adult fish.

(A) MA-plot at the two-cells stage. (B) MA-plot at 96 hpf. (C) Five-ways diagram of all genes regulated in adult brain, testis, ovaries and in the progeny of DU-exposed fish at two-cells stage and 96 hpf. (D) Examples of expression changes in the progeny of DU-exposed fish of genes involved in DNA repair (rad51d, rad50, rad2), splicing (isy1, syf2), chromatin remodelling (hat1 and hdac5), lipid transport (apoeb, cetp), oxidative stress (duox) and cell adhesion (pcdh1gc6) (all adjusted p-values (FDR) < 0.01; error-bars represent the standard deviation to the mean).

Fig 4. Expression analysis of electron transport chain complex genes in testis and brain of DU exposed fish and in their progeny at two-cells stage and 96 hpf.

(A) Heatmap of adjusted p-values of genes involved in mitochondrial oxydo-reduction process (n = 101). The colour code displays the log10(adjusted p-value) from the differential expression analysis. (B) MA-plot focusing on the differential expression of the electron transport chain complex genes in the 96 hpf larvae (fold change as DU/C). Most of the genes are up-regulated in DU exposed larva.

Fig 5. Histological analysis of 96 hpf larvae from DU-exposed adult fish compared to controls.

(A-A’) Toluidine-blue staining of the larva at the level of the trunk. Vacuole-like structures (appearing as white dots) are abundant in DU-treated larvae compared to controls. (B-C) Mitochondrial morphology in skeletal muscles observed by transmission electronic microscopy of control larvae. The inner mitochondrial membranes are dense and well visible (black arrow head). (B’-C’) Alteration of mitochondrial morphology and absence or decreases density of inner mitochondrial membranes (*) in larvae from DU-exposed fish. (D-E) High-resolution transmission electronic microscopy of skeletal muscles in controls and (D’-E’) in the 96 hpf larvae obtained from DU-exposed fish. Disruption of myofibres (red asterisks) and swelling of Z-bands are visible in DU larvae. Z- and A-bands are indicated. Triads constituted of t-tubes and the two flanking cisterna of the sarcoplasmic reticulum are visible in the controls (black arrows) but are absent or deformed in larvae from DU-treated fish (red arrows).

Fig 6. Induction of protein chaperones expression following myofibre damage in the 96 hpf larvae from DU-exposed adult zebrafish.

(A) MA-plot focusing on the differential expression of the 169 genes involved in cellular stress (from the ache mutant, see Material and methods) and their regulation in 96 hpf larvae obtained from DU exposed. Significant adjusted p-values < 0.01 are indicated in red and yellow. (B) Log2 fold change of specific protein-chaperon in the skeletal muscles (all adjusted p-values (FDR) < 0.01; error-bars represent the standard deviation to the mean).