Field surveys reveal the presence of anti-androgens in an effluent-receiving river using stickleback-specific biomarkers

Chemical occurrence of pesticides and transformation products in two small lentic waterbodies at the head of agricultural watersheds and biological responses in caged Gasterosteus aculeatus

Recent monitoring campaigns have revealed the presence of mixtures of pesticides and their transformation products (TP) in headwater streams situated within agricultural catchments. These observations were attributed to the use of various agrochemicals in surrounding regions. The aim of this work was to compare the application of chemical and ecotoxicological tools for assessing environmental quality in relation to pesticide and TP contamination. It was achieved by deploying these methodologies in two small lentic water bodies located at the top of two agricultural catchments, each characterized by distinct agricultural practices (ALT: organic, CHA: conventional). Additionally, the results make it possible to assess the impact of contamination on fish caged in situ. Pesticides and TP were measured in water using active and passive samplers and suspended solid particles. Eighteen biomarkers (innate immune responses, oxidative stress, biotransformation, neurotoxicity, genotoxicity, and endocrine disruption) were measured in Gasterosteus aculeatus encaged in situ. More contaminants were detected in CHA, totaling 25 compared to 14 in ALT. Despite the absence of pesticide application in the ALT watershed for the past 14 years, 7 contaminants were quantified in 100 % of the water samples. Among these contaminants, 6 were TPs (notably atrazine-2-hydroxy, present at a concentration exceeding 300 ng·L-1), and 1 was a current pesticide, prosulfocarb, whose mobility should prompt more caution and new regulations to protect adjacent ecosystems and crops. Regarding the integrated biomarker response (IBRv2), caged fish was similarly impacted in ALT and CHA. Variations in biomarker responses were highlighted depending on the site, but the results did not reveal whether one site is of better quality than the other. This outcome was likely attributed to the occurrence of contaminant mixtures in both sites. The main conclusions revealed that chemical and biological tools complement each other to better assess the environmental quality of wetlands such as ponds.

Application of the Fpg-modified comet assay on three-spined stickleback in freshwater biomonitoring: toward a multi-biomarker approach of genotoxicity

Aquatic species are exposed to a wide spectrum of substances, which can compromise their genomic integrity by inducing DNA damage or oxidative stress. Genotoxicity biomarkers as DNA strand breaks and chromosomal damages developed on sentinel species have already proved to be relevant in aquatic biomonitoring. However, these biomarkers do not reflect DNA oxidative lesions, i.e., the 8-oxodG, recognized as pre-mutagenic lesion if not or mis-repaired in human biomonitoring. The relevance to include the measure of these lesions by using the Fpg-modified comet assay on erythrocytes of the three-spined stickleback was investigated. An optimization step of the Fpg-modified comet assay considering enzyme buffer impact, Fpg concentration, and incubation time has been performed. Then, this measure was integrated in a battery of genotoxicity and cytotoxicity biomarkers (considering DNA strand breaks, DNA content variation, and cell apoptosis/necrosis and density) and applied in a freshwater monitoring program on six stations of the Artois Picardie watershed (3-week caging of control fish). These biomarkers allowed to discriminate the stations regarding the genotoxic potential of water bodies and specifically by the measure of oxidative DNA lesions, which seem to be a promising tool in environmental genotoxicity risk assessment.

Spatial and temporal variations in anti-androgenic activity and environmental risk in a small river

The biological effects of multiple compounds have been widely investigated in aquatic environments. However, investigations of spatial and temporal variations in biological effects are rarely performed because they are time-consuming and labor-intensive. In this study, the variability of the anti-androgen, receptor-mediated activity of surface water samples was observed over 3 years using in vitro bioassays. Large-volume water samples were collected at one site upstream (Wer site) and two sites downstream (Sil and Nien sites) of a wastewater treatment plant (WWTP) outfall in the Holtemme River. Anti-AR activity was persistently present in all surface water samples over the three years. Large spatial variations in anti-androgenic activity were observed, with the lowest activity at the Wer site (mean concentration of 9.5 ± 7.2 μg flutamide equivalents/L) and the highest activity at the Sil site (mean concentration of 31.1 ± 12.0 μg flutamide equivalents/L) directly influenced by WWTP effluents. On the temporal scale, no distinct trend for anti-AR activity was observed among the seasons in all three years. The anti-androgenic activity at the upstream Wer site showed a decreasing trend from 2014 to 2016, indicating improved water quality. A novel bioanalytical-equivalent-based risk assessment method considering the frequency of risk occurrence was developed and then utilized to assess the environmental risk of anti-androgenic activity in the Holtemme River. The results revealed that the highest risk was present at the Sil site, while the risk was considerably reduced at the Nien site. The risk at the upstream Wer site was the lowest.

Feminizing effects of ethinylestradiol in roach (Rutilus rutilus) populations with different estrogenic pollution exposure histories

Experimental exposures aimed at assessing the risks posed by estrogens in waste-water treatment work (WwTW) effluents to fish populations have rarely considered whether populations differ in their sensitivity to estrogenic compounds. This is despite evidence that selection at genes involved in the estrogen response has occurred in wild populations, and evidence that genotype can influence estrogen-response. In this study we compare the effects of a two-year exposure to a low measured concentration (1.3 ng/L) of ethinylestradiol (EE2) on the sexual development of roach (Rutilus rutilus) whose parental generation was sampled from two river stretches heavily contaminated with WwTW effluent and from two without any known WwTW effluent contamination. Exposure to EE2 significantly reduced the proportion of genetic males and induced a range of feminized phenotypes in males. Significantly, exposure also increased the proportion of genetic females with vitellogenic oocytes from 51 to 96%, raising the possibility that estrogen pollution could impact populations of annually spawning fish species through advancing female reproduction by at least a year. However, there was no evidence that river origin affected sensitivity to estrogens in either sex. Thus, we conclude that chronic exposure to low level EE2 has reproductive health outcomes for both male and female roach, but we find no evidence that the nature or magnitude of the response is affected by the population origin.

Melatonin and cortisol as components of the cutaneous stress response system in fish: Response to oxidative stress

The skin being a passive biological barrier that defends the organism against harmful external factors is also a site of action of the system responding to stress. It appears that melatonin (Mel) and its biologically active metabolite AFMK (N1-acetyl-N2-formyl-5-methoxykynuramine), both known as effective antioxidants, together with cortisol, set up a local (cutaneous) stress response system (CSRS) of fish, similar to that of mammals. Herein we comment on recent studies on CSRS in fish and show the response of three-spined stickleback skin to oxidative stress induced by potassium dichromate. Our study indicates that exposure of the three-spined stickleback to K₂Cr₂O₇ affects Mel and cortisol levels and pigment dispersion in melanophores in the skin. In our opinion, an increased concentration of Mel and cortisol in the skin may be the strategy to cope with oxidative stress, where both components act locally to prevent damage caused by active oxygen molecules. Furthermore, the pigment dispersion may be a valuable, easy-to-observe mark of oxidative stress, useful in the evaluation of fish welfare.

Preself-Feeding Medaka Fry Provides a Suitable Screening System for in Vivo Assessment of Thyroid Hormone-Disrupting Potential

Endocrine-disrupting chemicals are assessed based on their physiological potential and their potential associated adverse effects. However, suitable end points for detection of chemicals that interfere with the thyroid hormone (TH) system have not been established in nonmammals, with the exception of amphibian metamorphosis. The aims of the current study were to develop an in vivo screening system using preself-feeding medaka fry (Oryzias latipes) for the detection of TH-disrupting chemicals and elucidate the underlying molecular mechanism. 17α-Ethinylestradiol (EE2: <100 ng/L) did not induce mRNA expression of estrogen-responsive genes, vitellogenins (vtgs) mRNA. Meanwhile, coexposure with thyroxin (T4) induced an increase of vtg expression. TH-disrupting chemicals (thiourea (TU), perfluorooctanoic acid (PFOA), and tetrabromobisphenol A (TBBPA)) significantly suppressed EE2 (1,000 ng/L)-induced vtg1 expression, while T4 rescued their expression as well as that of thyroid hormone receptor α (tRα) and estrogen receptors (esrs). These results were supported by in silico analysis of the 5'-transcriptional regulatory region of these genes. Furthermore, the esr1 null mutant revealed that EE2-induced vtg1 expression requires mainly esr2a and esr2b in a TH-dependent manner in preself-feeding fry. Application of preself-feeding medaka fry as a screening system might help decipher the in vivo mechanisms of action of TH-disrupting molecules, while providing an alternative to the traditional animal model.

Identification of Unknown Antiandrogenic Compounds in Surface Waters by Effect-Directed Analysis (EDA) Using a Parallel Fractionation Approach

Among all the nuclear-receptor mediated endocrine disruptive effects, antiandrogenicity is frequently observed in aquatic environments and may pose a risk to aquatic organisms. Linking these effects to responsible chemicals is challenging and a great share of antiandrogenic activity detected in the environment has not been explained yet. To identify drivers of this effect at a hot spot of antiandrogenicity in the German river Holtemme, we applied effect-directed analysis (EDA) including a parallel fractionation approach, a downscaled luciferase reporter gene cell-based anti-AR-CALUX assay and LC-HRMS/MS nontarget screening. We identified and confirmed the highly potent antiandrogen 4-methyl-7-diethylaminocoumarin (C47) and two derivatives in the active fractions. The relative potency of C47 to the reference compound flutamide was over 5.2, whereas the derivatives were less potent. C47 was detected at a concentration of 13.7 μg/L, equal to 71.4 μg flutamide equivalents per liter (FEq/L) in the nonconcentrated water extract that was posing an antiandrogenic activity equal to 45.5 (±13.7 SD) FEq/L. Thus, C47 was quantitatively confirmed as the major cause of the measured effect in vitro. Finally, the antiandrogenic activity of C47 and one derivate was confirmed in vivo in spiggin-gfp Medaka. An endocrine disrupting effect of C47 was observed already at the concentration equal to the concentration in the nonconcentrated water extract, underlining the high risk posed by this compound to the aquatic ecosystem. This is of some concern since C47 is used in a number of consumer products indicating environmental as well as human exposure.

Estrogenic and anti-androgenic effects of the herbicide tebuthiuron in male Nile tilapia (Oreochromis niloticus)

Tebuthiuron is a phenylurea herbicide widely used in agriculture that can reach the aquatic environments, possibly posing negative effects to the aquatic biota. Phenylurea herbicides, such as diuron, are known to cause estrogenic and anti-androgenic effects in fish, but no such effects were yet reported for tebuthiuron exposure. Thus, the aim of this study was to evaluate if tebuthiuron, at environmentally relevant concentrations (100 and 200ng/L) and after 25days of exposure have estrogenic and/or anti-androgenic effects on male of Nile tilapia (Oreochromis niloticus), through the evaluation of plasmatic testosterone (T) and estradiol (E₂) levels, brain aromatase (CYP19) levels (western-blot), and by evaluating the histology of the testicles. When compared to the control group, plasmatic T levels decreased about 76% in the animals exposed to 200ng/L of tebuthiuron, while E₂ levels increased about 94%, which could be related to a significant increase (77%) in CYP19A1 levels, an enzyme that catalyzes the conversion of androgens into estrogens. Histological analyses of the testicles also demonstrated that tebuthiuron at both tested concentrations caused a decrease in the diameter of the seminiferous tubules and in the diameter of the lumen. Therefore, the gonadosomatic index (GSI) was reduced by 36% % in the animals exposed 200ng/L to tebuthiuron. Indeed, the relative frequency of spermatocytes and spermatids increased respectively 73% (200ng/L) and 61% (100ng/L) in the tebuthiuron exposed animals, possibly due to the impairment of sperm release into the lumen, that was decreased 93% (200ng/L) in the treated animals compared to the control. These results confirm that tebuthiuron causes estrogenic and anti-androgenic effects in Nile tilapias at environmentally relevant concentrations.

An integrative approach to assess ecological risks of surface water contamination for fish populations

Contamination of aquatic ecosystems is considered as one of the main threats to global freshwater biodiversity. Within the European Water Framework Directive (EU-WFD) a particular attention is dedicated to assess ecological risks of surface water contamination and mitigation of chemical pressures on aquatic ecosystems. In this work, we evaluated ecological risks of surface water contamination for fish populations in four EU-WFD rivers through an integrative approach investigating three Lines of Evidence (chemical contamination, biomarker responses as early warning signals of contamination impacting individuals and ecological analyses as an indicator of fish community disturbances). This work illustrates through 4 case studies the complementary role of biomarkers, chemical and ecological analyses which, used in combination, provide fundamental information to understand impacts of chemical pressures that can affect fish population dynamics. We discuss the limitations of this approach and future improvements needed within the EU-WFD to assess ecological risk of river contamination for fish populations.

Advancing Biological Wastewater Treatment: Extended Anaerobic Conditions Enhance the Removal of Endocrine and Dioxin-like Activities

Conventional activated sludge treatment of wastewater does not completely remove micropollutants. Here, extending anaerobic conditions may enhance biodegradation. To explore this, we combined iron-reducing or substrate-limiting and aerobic pilot-scale reactors directly at a wastewater treatment plant. To assess the removal of endocrine disrupting chemicals (EDCs) as group of micropollutants that adversely affects wildlife, we applied a bioanalytical approach. We used in vitro bioassays covering seven receptor-mediated mechanisms of action, including (anti)androgenicity, (anti)estrogenicity, retinoid-like, and dioxin-like activity. Untreated wastewater induced antiandrogenic, estrogenic, antiestrogenic, and retinoid-like activity. Full-scale as well as reactor-scale activated sludge treatment effectively removes the observed effects. Nevertheless, high antiandrogenic and minor dioxin-like and estrogenic effects persisted in the treated effluent that may still be environmentally relevant. The anaerobic post-treatment under substrate-limiting conditions resulted in an additional removal of endocrine activities by 17-40%. The anaerobic pre-treatment under iron-reducing conditions significantly enhanced the removal of the residual effects by 40-75%. In conclusion, this study demonstrates that a further optimization of biological wastewater treatment is possible. Here, implementing iron-reducing anaerobic conditions preceding aerobic treatment appears promising to improve the removal of receptor-mediated toxicity.

Oestrogenic pollutants promote the growth of a parasite in male sticklebacks

Aquatic environments are especially susceptible to anthropogenic chemical pollution. Yet although knowledge on the biological effects of pollutants on aquatic organisms is increasing, far less is known about how ecologically-important interspecific interactions are affected by chemicals. In particular, the consequences of anthropogenic pollution for the interaction of hosts and parasites are poorly understood. Here, we examine how exposure to 17β-oestradiol (E2)-a natural oestrogen and a model endocrine disrupting chemical (EDC) -affects infection susceptibility and emergent infection phenotypes in an experimental host-parasite system; three spined sticklebacks (Gasterosteus aculeatus) infected with the common, debilitating cestode Schistocephalus solidus. We exposed individual sticklebacks to a 0ngl(-1) (control), 10ngl(-1) or 100ngl(-1) E2 treatment before feeding them infective stages of S. solidus. E2 exposure significantly elevated vitellogenin (VTG) levels-a biomarker of exposure to xenoestrogens-in both female and male fish, and reduced their body condition. Susceptibility to parasite infection was unaffected by EDC exposure; however, E2 treatment and fish sex interacted significantly to determine the growth rate of parasites, which grew quickest in male hosts held under the higher (100ngl(-1)) E2 treatment. Tissue VTG levels and parasite mass correlated positively across the whole sample of experimentally infected fish, but separate regressions run on the male and female datasets demonstrated a significant relationship only among male fish. Hence, among males-but not females-elevated VTG levels elicited by E2 exposure led to more rapid parasite growth. We outline plausible physiological mechanisms that could explain these results. Our results demonstrate that oestrogenic pollutants can alter host-parasite interactions by promoting parasite growth, and that male hosts may be disproportionately affected. Because ecologically-relevant effects of infection on host antipredator responses, growth, energetics and reproductive development all depend on parasite mass in this host-parasite system, our results indicate that EDCs can mediate the ecological consequences of infections. We therefore consider the implications of our results for the ecology of hosts and parasites in polluted environments.

Biomarker responses to estrogen and androgen exposure in the brook stickleback (Culaea inconstans): A new bioindicator species for endocrine disrupting compounds

Small-bodied freshwater fish are commonly used in regulatory testing for endocrine disrupting compounds (EDCs) but most lack a sensitive and quantifiable androgen-specific biomarker. Brook stickleback (Culaea inconstans) are a North American freshwater fish whose males produce an androgen-regulated glycoprotein in the kidney called spiggin. Although spiggin induction in females has been used as an androgen-specific biomarker of exposure in other stickleback species it has not been characterized in brook stickleback. Therefore, our objective was to develop a bioassay using brook stickleback to measure estrogenic and androgenic responses and establish the sensitivity of traditional and novel biomarkers of exposure. We first developed and optimized a qPCR assay to measure spiggin and vitellogenin transcript levels in kidney and liver tissue, respectively. Basal levels were differentially expressed in mature wild-caught male and female brook stickleback. To determine their sensitivity to EDCs, fish were exposed to nominal concentrations of 1, 10 and 100ng/L of 17α-methyltestosterone (MT) or 17α-ethinylestradiol (EE2) for 21days (sampled at 7 and 21days) under semi-static renewal conditions. MT and EE2 exposure induced spiggin and vitellogenin transcripts in female kidneys and male livers, respectively. Exposure to EE2 also increased hepatosomatic index in both sexes and decreased gonadosomatic index in females. Histopathological alterations were observed in the kidney of EE2-exposed fish and an increase in kidney epithelium cell height occurred in MT-exposed females. Given the sensitivity of these endpoints, the brook stickleback is a promising new freshwater fish model for EDC evaluation and a potential bioindicator for EDCs in North American freshwater environments.

Sexual disruption in zebrafish (Danio rerio) exposed to mixtures of 17α-ethinylestradiol and 17β-trenbolone

Environmental estrogens and androgens can be present simultaneously in aquatic environments and thereby interact to disturb multiple physiological systems in organisms. Studies on interaction effects in fish of androgenic and estrogenic chemicals are limited. Therefore, the aim of the present study was to evaluate feminization and masculinization effects in zebrafish (Danio rerio) exposed to combinations of two synthetic steroid hormones detected in environmental waters: the androgen 17β-trenbolone (Tb) and the oestrogen 17α-ethinylestradiol (EE2). Juvenile zebrafish were exposed between days 20 and 60 post-hatch to different binary mixtures of Tb (1, 10, and 50 ng/L) and EE2 (2 and 5 ng/L). The endpoints studied were whole-body homogenate vitellogenin concentration at 40 days post-hatch, and sex ratio including gonad maturation at 60 days post-hatch. The feminizing potency of 5 ng/L of EE2, alone as well as in combination with Tb, was clear in the present study, with exposures resulting in almost all-female populations and females being sexually immature. Masculinization effects with male-biased sex ratios were observed when fish were exposed to 2 ng/L of EE2 in combination with Tb concentrations. Intersex fish were observed after exposure to mixtures of 2 ng/L EE2 with 50 ng/L Tb. Sexual maturity generally increased among males at increasing concentrations of Tb. The results of the present study show that exposure to environmentally relevant mixtures of an oestrogen and androgen affects the process of gonad differentiation in zebrafish and lead to sexual disruption.

Distribution of selected antiandrogens and pharmaceuticals in a highly impacted watershed

Endocrine disruption and high occurrences of intersex have been observed in wild fish associated with municipal wastewater treatment plant (WWTP) effluents in urbanized reaches of rivers around the globe. These reproductive effects have often been attributed to the presence of estrogen receptor agonists in effluents. However, recent studies have isolated a number of androgen receptor antagonists (antiandrogens) that may also contribute to the endocrine disruption observed at sites that are influenced by WWTP outfalls. This study aimed to characterize the spatial and temporal distribution of antiandrogenic personal care products (triclosan, chlorophene, dichlorophene, oxybenzone, 1-naphthol, and 2-naphthol), along with a herbicide (atrazine) and representative pharmaceuticals (carbamazepine, ibuprofen, naproxen, and venlafaxine) in the Grand River watershed in southern Ontario. Surface water sampling of 30 sites associated with six municipal WWTP outfalls was conducted during a summer low flow. Monthly samples were also collected immediately upstream and downstream of a major WWTP from August to November 2012. Atrazine was consistently found in all surface water sampling locations. Many of the target pharmaceuticals and triclosan were detected in WWTP effluents, especially those that did not nitrify. Under low flow conditions, the concentrations of triclosan and several pharmaceuticals increased directly downstream of the WWTPs then decreased rapidly with distance downstream. Chlorophene was either found at trace levels or below detection limits in the effluents while dichlorophene, oxybenzone, 1-naphthol, and 2-naphthol were not detected in any samples. Chlorophene was detected in surface water during the low flow summer period and once during the monthly sampling from August to November. However, the primary source of chlorophene did not appear to be associated with WWTP effluent. This study documents the spatial and temporal occurrence of several antiandrogens and pharmaceuticals in a highly impacted Canadian watershed. It supports previous observations that there is a diversity of contaminants in wastewater effluents and other sources that have the potential to alter endocrine function in wild fish.

Environmental concentrations of anti-androgenic pharmaceuticals do not impact sexual disruption in fish alone or in combination with steroid oestrogens

Sexual disruption in wild fish has been linked to the contamination of river systems with steroid oestrogens, including the pharmaceutical 17α-ethinylestradiol, originating from domestic wastewaters. As analytical chemistry has advanced, more compounds derived from the human use of pharmaceuticals have been identified in the environment and questions have arisen as to whether these additional pharmaceuticals may also impact sexual disruption in fish. Indeed, pharmaceutical anti-androgens have been shown to induce such effects under laboratory conditions. These are of particular interest since anti-androgenic biological activity has been identified in the aquatic environment and is potentially implicated in sexual disruption alone and in combination with steroid oestrogens. Consequently, predictive modelling was employed to determine the concentrations of two anti-androgenic human pharmaceuticals, bicalutamide and cyproterone acetate, in UK sewage effluents and river catchments and their combined impacts on sexual disruption were then assessed in two fish models. Crucially, fish were also exposed to the anti-androgens in combination with steroid oestrogens to determine whether they had any additional impact on oestrogen induced feminisation. Modelling predicted that the anti-androgenic pharmaceuticals were likely to be widespread in UK river catchments. However, their concentrations were not sufficient to induce significant responses in plasma vitellogenin concentrations, secondary sexual characteristics or gross indices in male fathead minnow or intersex in Japanese medaka alone or in combination with steroid oestrogens. However, environmentally relevant mixtures of oestrone, 17β-oestradiol and 17α-ethinylestradiol did induce vitellogenin and intersex, supporting their role in sexual disruption in wild fish populations. Unexpectedly, a male dominated sex ratio (100% in controls) was induced in medaka and the potential cause and implications are briefly discussed, highlighting the potential of non-chemical modes of action on this endpoint.

Anti-androgens act jointly in suppressing spiggin concentrations in androgen-primed female three-spined sticklebacks - prediction of combined effects by concentration addition

Increasing attention is being directed at the role played by anti-androgenic chemicals in endocrine disruption of wildlife within the aquatic environment. The co-occurrence of multiple contaminants with anti-androgenic activity highlights a need for the predictive assessment of combined effects, but information about anti-androgen mixture effects on wildlife is lacking. This study evaluated the suitability of the androgenised female stickleback screen (AFSS), in which inhibition of androgen-induced spiggin production provides a quantitative assessment of anti-androgenic activity, for predicting the effect of a four component mixture of anti-androgens. The anti-androgenic activity of four known anti-androgens (vinclozolin, fenitrothion, flutamide, linuron) was evaluated from individual concentration-response data and used to design a mixture containing each chemical at equipotent concentrations. Across a 100-fold concentration range, a concentration addition approach was used to predict the response of fish to the mixture. Two studies were conducted independently at each of two laboratories. By using a novel method to adjust for differences between nominal and measured concentrations, good agreement was obtained between the actual outcome of the mixture exposure and the predicted outcome. This demonstrated for the first time that androgen receptor antagonists act in concert in an additive fashion in fish and that existing mixture methodology is effective in predicting the outcome, based on concentration-response data for individual chemicals. The sensitivity range of the AFSS assay lies within the range of anti-androgenicity reported in rivers across many locations internationally. The approach taken in our study lays the foundations for understanding how androgen receptor antagonists work together in fish and is essential in informing risk assessment methods for complex anti-androgenic mixtures in the aquatic environment.

Parental exposure to methyl methane sulfonate of three-spined stickleback: contribution of DNA damage in male and female germ cells to further development impairment in progeny

Data regarding the link between DNA integrity of germ cells and the quality of progeny in fish exposed to genotoxicant are scarce although such information is of value to understand genotoxic effects of contaminants in aquatic fauna. This work aimed at studying the consequences of a parental exposure during the breeding season on offspring quality in three-spined stickleback. After in vivo exposure of adult fish to methyl methane sulfonate, a model alkylating compound, a clear increase in DNA damage was observed in erythrocytes of both genders, here used as a biomarker of exposure. MMS exposure significantly affected sperm DNA integrity but neither female fecundity nor fertilization success. In order to understand the contribution of each sex to potential deleterious effects in progeny due to parental exposure, mating of males and females exposed or not to MMS, was carried out. Exposure of both males and females or of males alone led to a significant increase in both mortality during embryo-larval stages and abnormality rate at hatching that appeared to be sensitive stages. Thus, in accordance with recent studies carried out in other freshwater fish species, such development defects in progeny were clearly driven by male genome, known to be devoid of DNA repair capacity in spermatozoa. The next step will be to investigate the link between DNA damage in stickleback sperm and reproductive impairment in natural populations exposed to complex mixture of genotoxicants.