Editorial trend: adverse outcome pathway (AOP) and computational strategy - towards new perspectives in ecotoxicology.

The adverse outcome pathway (AOP) has been conceptualized in 2010 as an analytical construct to describe a sequential chain of causal links between key events, from a molecular initiating event leading to an adverse outcome (AO), considering several levels of biological organization. An AOP aims to identify and organize available knowledge about toxic effects of chemicals and drugs, either in ecotoxicology or toxicology, and it can be helpful in both basic and applied research and serve as a decision-making tool in support of regulatory risk assessment. The AOP concept has evolved since its introduction, and recent research in toxicology, based on integrative systems biology and artificial intelligence, gave it a new dimension. This innovative in silico strategy can help to decipher mechanisms of action and AOP and offers new perspectives in AOP development. However, to date, this strategy has not yet been applied to ecotoxicology. In this context, the main objective of this short article is to discuss the relevance and feasibility of transferring this strategy to ecotoxicology. One of the challenges to be discussed is the level of organisation that is relevant to address for the AO (population/community). This strategy also offers many advantages that could be fruitful in ecotoxicology and overcome the lack of time, such as the rapid identification of data available at a time t, or the identification of "data gaps". Finally, this article proposes a step forward with suggested priority topics in ecotoxicology that could benefit from this strategy.

Differences in chemical contaminants bioaccumulation and ecotoxicology biomarkers in Mytilus edulis and Mytilus galloprovincialis and their hybrids.

The Mytilus mussels are spread all over the world and many related species coexist in several areas and can produce hybrid offspring. Mussels have been used for decades in national and international programs to monitor chemical contamination in the environment. Differences in bioaccumulation and biotransformation abilities between species and their hybrids should be evaluated to assess the comparability of the results obtained within the international biomonitoring programs. The objective of this study was to characterize bioaccumulation abilities and biomarker responses in Mytilus edulis, Mytilus galloprovincialis and their hybrids via an in situ transplantation experimentation on their progenies. Four mussel groups (M. edulis, M. galloprovincialis and two hybrids batches) issued from genetically characterized parents were transplanted for one year in Charente Maritime (France) to ensure their exposure to identical sources of contamination. The bioaccumulation of several families of contaminants (trace metals, polycyclic aromatic hydrocarbons, polybrominated diphenyl ethers, polychlorinated biphenyls), the response of several biomarkers (DNA strand breaks level, lysosomal membrane stability, metallothionein content, acetylcholine esterase activity) and some physiological parameters (growth, mortality, gonadal development), were analyzed. Differences were observed between species, however they were contaminant-specific. Variations in contaminants levels were observed between progenies, with higher levels of Cu, PBDE, PCB in M. edulis, and higher levels of Cd, Hg, Zn in M galloprovincialis. This study demonstrated that variations in contaminant bioaccumulation and different biomarker responses exist between Mytilus species in the field. Data on species or the presence of hybrid individuals (or introgression) is an important additional parameter to add to biomonitoring programs databases.

Replicated anthropogenic hybridisations reveal parallel patterns of admixture in marine mussels.

Human-mediated transport creates secondary contacts between genetically differentiated lineages, bringing new opportunities for gene exchange. When similar introductions occur in different places, they provide informally replicated experiments for studying hybridisation. We here examined 4,279 Mytilus mussels, sampled in Europe and genotyped with 77 ancestry-informative markers. We identified a type of introduced mussels, called "dock mussels," associated with port habitats and displaying a particular genetic signal of admixture between M. edulis and the Mediterranean lineage of M. galloprovincialis. These mussels exhibit similarities in their ancestry compositions, regardless of the local native genetic backgrounds and the distance separating colonised ports. We observed fine-scale genetic shifts at the port entrance, at scales below natural dispersal distance. Such sharp clines do not fit with migration-selection tension zone models, and instead suggest habitat choice and early-stage adaptation to the port environment, possibly coupled with connectivity barriers. Variations in the spread and admixture patterns of dock mussels seem to be influenced by the local native genetic backgrounds encountered. We next examined departures from the average admixture rate at different loci, and compared human-mediated admixture events, to naturally admixed populations and experimental crosses. When the same M. galloprovincialis background was involved, positive correlations in the departures of loci across locations were found; but when different backgrounds were involved, no or negative correlations were observed. While some observed positive correlations might be best explained by a shared history and saltatory colonisation, others are likely produced by parallel selective events. Altogether, genome-wide effect of admixture seems repeatable and more dependent on genetic background than environmental context. Our results pave the way towards further genomic analyses of admixture, and monitoring of the spread of dock mussels both at large and at fine spacial scales.

DNA damage in dab (Limanda limanda) and haddock (Melanogrammus aeglefinus) from European seas.

Dab (Limanda limanda) and haddock (Melanogrammus aeglefinus) were collected from coastal and offshore locations in the Baltic (dab only), North Sea (haddock from one location only) and Iceland. Blood was analysed for DNA strand breaks using the comet assay and liver samples for DNA adduct concentrations. DNA strand breaks were at background levels in dab from the two Iceland locations and from the Dogger Bank. The highest levels were observed in dab from the Firth of Forth, Ekofisk and the German Bight. Hepatic DNA adducts in dab were highest at Ekofisk, in the Baltic and Dogger Bank, below detection limit in dab from Iceland and low in dab from the Firth of Forth and German Bight. There was large variation in DNA strand breaks between locations and individuals for haddock, particularly from Iceland. Adduct concentrations were elevated in haddock from both Iceland and the Firth of Forth. A general linear model (GLM) suggested that, in addition to location, the size of dab and its general condition contributed to explaining the observed variability in DNA strand breaks. A GLM for adducts in dab similarly allocated most of the variability to location, but with a possible contribution from CYP1A activity. There were no apparent differences between male and female dab for any of the methods. There was no obvious relationship between strand breaks and adducts in the same fish although dab from Ekofisk and Iceland had respectively high and low responses using both methods. The results from this large-scale study showed pollution-related genotoxicity for dab, that fish blood samples can be conserved prior to comet analyses and that there are clear species differences in genotoxic responses even when collected at the same location.

Assessment of contaminant concentrations in sediments, fish and mussels sampled from the North Atlantic and European regional seas within the ICON project.

Understanding the status of contaminants in the marine environment is a requirement of European Union Directives and the Regional Seas Conventions, so that measures to reduce pollution can be identified and their efficacy assessed. The international ICON workshop (Hylland et al., in this issue) was developed in order to test an integrated approach to assessing both contaminant concentrations and their effects. This paper describes and assesses the concentrations of trace metals, polycyclic aromatic hydrocarbons, and polychlorinated biphenyls in sediments, mussels, and fish collected from estuarine, coastal and offshore waters from Iceland to the Mediterranean Sea. For organic contaminants, concentrations progressively increased from Iceland, to the offshore North Sea, to the coastal seas, and were highest in estuaries. Metals had a more complex distribution, reflecting local anthropogenic inputs, natural sources and hydrological conditions. Use of internationally recognised assessment criteria indicated that at no site were concentrations of all contaminants at background and that concentrations of some contaminants were of significant concern in all areas, except the central North Sea.

Integrated monitoring of chemicals and their effects on four sentinel species, Limanda limanda, Platichthys flesus, Nucella lapillus and Mytilus sp., in Seine Bay: A key step towards applying biological effects to monitoring.

The International workshop on Integrated Assessment of CONtaminants impacts on the North sea (ICON) provided a framework to validate the application of chemical and biological assessment thresholds (BACs and EACs) in the Seine Bay in France. Bioassays (oyster larval anomalies, Corophium arenarium toxicity assay and DR Calux) for sediment and biomarkers: ethoxyresorufin-O-deethylase (EROD) activity, acetylcholinesterase (AChE) activity, lysosomal membrane stability (LMS), DNA strand breaks using the Comet assay, DNA adducts, micronucleus (MN), PAH metabolites, imposex, intersex and fish external pathologies were analysed in four marine sentinel species (Platichthys flesus, Limanda limanda, Mytilus sp. and Nucella lapilus). Polycyclic aromatic hydrocarbons (PAHs), polychlorinated biphenyls (PCBs) and heavy metals were analysed in biota and sediment. Results for sediment and four species in 2008-2009 made it possible to quantify the impact of contaminants using thresholds (Environmental Assessment Criteria/EAC2008: 70% and EAC2009: 60%) and effects (EAC2008: 50% and EAC2009: 40%) in the Seine estuary. The Seine estuary is ranked among Europe's most highly polluted sites.

Integrated chemical and biological assessment of contaminant impacts in selected European coastal and offshore marine areas.

This paper reports a full assessment of results from ICON, an international workshop on marine integrated contaminant monitoring, encompassing different matrices (sediment, fish, mussels, gastropods), areas (Iceland, North Sea, Baltic, Wadden Sea, Seine estuary and the western Mediterranean) and endpoints (chemical analyses, biological effects). ICON has demonstrated the use of a framework for integrated contaminant assessment on European coastal and offshore areas. The assessment showed that chemical contamination did not always correspond with biological effects, indicating that both are required. The framework can be used to develop assessments for EU directives. If a 95% target were to be used as a regional indicator of MSFD GES, Iceland and offshore North Sea would achieve the target using the ICON dataset, but inshore North Sea, Baltic and Spanish Mediterranean regions would fail.

Three simple biomarkers useful in conducting water quality assessments with bivalve mollusks.

While biomarkers are undeniably key tools in aquatic ecotoxicology to measure adverse effects linked to contamination events, their application is often inhibited by monetary constraints negating the possibility of having access to dedicated equipment, special wares, and/or expensive reagents. To offset this bottleneck, we propose three simple physiological biomarkers, quantifiable in bivalves, that are free of cost considerations and that can provide basic knowledge on animal health and water quality. Indeed, condition index (CI), growth index (GI), and SOS response (air-time survival) comprise measurements straightforward enough to perform by any laboratory or science body on the planet. Long-term monitoring or screening studies can be carried out with these biomarkers and they are able to provide robust information notably after exposure of bivalves to either singular or multiple agents of contamination. By highlighting examples of data generated in aquatic studies conducted in Eastern Canada under both laboratory and field situations with different species of marine and freshwater mollusks, we establish the suitability of these biomarkers for assessing environmental contamination. Their relationships with other biomarkers are also shown which further corroborate their value as reliable indicators of ecosystem health.

Integrated indicator framework and methodology for monitoring and assessment of hazardous substances and their effects in the marine environment.

Many maritime countries in Europe have implemented marine environmental monitoring programmes which include the measurement of chemical contaminants and related biological effects. How best to integrate data obtained in these two types of monitoring into meaningful assessments has been the subject of recent efforts by the International Council for Exploration of the Sea (ICES) Expert Groups. Work within these groups has concentrated on defining a core set of chemical and biological endpoints that can be used across maritime areas, defining confounding factors, supporting parameters and protocols for measurement. The framework comprised markers for concentrations of, exposure to and effects from, contaminants. Most importantly, assessment criteria for biological effect measurements have been set and the framework suggests how these measurements can be used in an integrated manner alongside contaminant measurements in biota, sediments and potentially water. Output from this process resulted in OSPAR Commission (www.ospar.org) guidelines that were adopted in 2012 on a trial basis for a period of 3 years. The developed assessment framework can furthermore provide a suitable approach for the assessment of Good Environmental Status (GES) for Descriptor 8 of the European Union (EU) Marine Strategy Framework Directive (MSFD).

How can we quantify impacts of contaminants in marine ecosystems? The ICON project.

An international workshop on marine integrated contaminant monitoring (ICON) was organised to test a framework on integrated environmental assessment and simultaneously assess the status of selected European marine areas. Biota and sediment were sampled in selected estuarine, inshore and offshore locations encompassing marine habitats from Iceland to the Spanish Mediterranean. The outcome of the ICON project is reported in this special issue as method-oriented papers addressing chemical analyses, PAH metabolites, oxidative stress, biotransformation, lysosomal membrane stability, genotoxicity, disease in fish, and sediment assessment, as well as papers assessing specific areas. This paper provides a background and introduction to the ICON project, by reviewing how effects of contaminants on marine organisms can be monitored and by describing strategies that have been employed to monitor and assess such effects. Through the ICON project we have demonstrated the use of an integrating framework and gleaned more knowledge than ever before in any single field campaign about the impacts contaminants may have in European marine areas.

Biomarkers of general stress in mussels as common indicators for marine biomonitoring programmes in Europe: The ICON experience.

This study investigated whether general stress biomarkers in mussels can be applied as common first-tier biomarkers in regional biomonitoring programmes in the North Sea (including Iceland) and western Mediterranean Sea. Stress on Stress (SoS) and lysosomal membrane stability (LMS) biomarkers were analysed in resident mussels (Mytilus sp.) from 8 coastal sites and in transplanted mussels (Mytilus galloprovincialis) from two Spanish Mediterranean coastal sites. The assessment of results, as input to pollution monitoring strategies, was performed jointly for LMS and SoS data from the two regions. Contaminant body burden of the mussels was compared with biomarker results. The results demonstrated that these two general and non-expensive stress biomarkers in mussel can be applied throughout European waters, providing a cost-effective and harmonised approach to screen contaminant-related biological effects within the framework of wide-scale pollution biomonitoring programmes, such as that proposed by the European Union, i.e. the Marine Strategy Framework Directive.

Comparative embryotoxicity and genotoxicity of the herbicide diuron and its metabolites in early life stages of Crassostrea gigas: Implication of reactive oxygen species production.

Herbicides are one of the major classes of pollutants contaminating coastal waters over the world. Among them, diuron (3-(3,4-dichlorophenyl)-1,1-dimethylurea) is a phenylurea herbicide frequently detected in oyster-producing area, known to be toxic for this important exploited non-target species. With the aim to investigate the mechanisms by which diuron displays its toxicity in oyster, the implication of both biotransformation and oxygen reactive species (ROS) production was studied considering embryotoxicity and genotoxicity as endpoints. Comparative embryotoxicity and genotoxicity of diuron and its main metabolites (DCPMU, DCPU and 3,4-DCA) were thus studied on oyster larvae by the embryo-larval bioassay on D larvae and the comet assay on trochophore larvae, respectively. Exposures were also performed in presence and absence of known ROS scavenger compounds - ascorbic acid and N-acetylcysteine, to evaluate the involvement of oxyradicals in the toxic responses. In the case of diuron, the production of ROS on exposed oyster larvae was also measured using 2',7'-dichlorodihydrofluorescein diacetate as a probe for flow cytometric analysis. The results we obtained showed the embryotoxicity and genotoxicity of diuron and its metabolites in early life stages of the Pacific oyster. For concentrations ranging from 0.05 to 0.5µgL(-1), diuron appeared significantly more embryotoxic than DCPMU and DCPU (p<0.001). Embryotoxicity decreased with diuron metabolism as follows: diuron≥DCPMU=DCPU, highlighting that biotransformation can constitute a true detoxication pathways in oyster larvae by decreasing the toxicity of the parent compound. In the opposite, no difference was observed between diuron and its metabolites concerning larval development when considering a lower and more environmentally realistic range of concentrations (0.002-0.050µgL(-1)). 3,4-DCA was the only compound that did not show any sign of embryotoxicity, even at concentrations up to 5µgL(-1). Concerning genotoxicity, no significant difference was observed between diuron and all of its metabolites including 3, 4 DCA with damages detected from the concentration of 0.05µgL(-1). As for diuron, the toxicity of the metabolites seems to be mediated in some part by ROS production as clearly demonstrated by the decrease in genotoxicity and developmental abnormalities in the presence of the oxidant scavenger, ascorbic acid.

Parental exposure to the herbicide diuron results in oxidative DNA damage to germinal cells of the Pacific oyster Crassostrea gigas.

Chemical pollution by pesticides has been identified as a possible contributing factor to the massive mortality outbreaks observed in Crassostrea gigas for several years. A previous study demonstrated the vertical transmission of DNA damage by subjecting oyster genitors to the herbicide diuron at environmental concentrations during gametogenesis. This trans-generational effect occurs through damage to genitor-exposed gametes, as measured by the comet-assay. The presence of DNA damage in gametes could be linked to the formation of DNA damage in other germ cells. In order to explore this question, the levels and cell distribution of the oxidized base lesion 8-oxodGuo were studied in the gonads of exposed genitors. High-performance liquid chromatography coupled with UV and electrochemical detection analysis showed an increase in 8-oxodGuo levels in both male and female gonads after exposure to diuron. Immunohistochemistry analysis showed the presence of 8-oxodGuo at all stages of male germ cells, from early to mature stages. Conversely, the oxidized base was only present in early germ cell stages in female gonads. These results indicate that male and female genitors underwent oxidative stress following exposure to diuron, resulting in DNA oxidation in both early germ cells and gametes, such as spermatozoa, which could explain the transmission of diuron-induced DNA damage to offspring. Furthermore, immunostaining of early germ cells seems indicates that damages caused by exposure to diuron on germ line not only affect the current sexual cycle but also could affect future gametogenesis.

Pesticides and Ostreid Herpesvirus 1 Infection in the Pacific Oyster, Crassostrea gigas.

Since 2008, mass mortality outbreaks have been reported in all French regions producing Pacific oysters, and in several Member States of the European Union. These mass mortality events of Pacific oysters are related to OsHV-1 infection. They occur during spring and summer periods leaving suspect the quality of the marine environment and the role of seasonal use of pesticides associated with the arrival of freshwater in oyster rearing areas. Pesticides have been also detected in French coastal waters, especially in areas of oyster production. Using PMA real-time PCR we showed that a mixture of 14 pesticides has no effect on the integrity of virus capsids from viral suspension in the conditions tested. A contact of oysters with this pesticide mixture was related to higher mortality rates among experimentally infected animals in comparison with control ones (no previous pesticide exposure before experimental infection). We therefore suggest that pesticides at realistic concentration can exert adverse effects on Pacific oysters and causes an increased susceptibility to the viral infection in experimental conditions.

Parental exposure to environmental concentrations of diuron leads to aneuploidy in embryos of the Pacific oyster, as evidenced by fluorescent in situ hybridization.

Changes in normal chromosome numbers (i.e. aneuploidy) due to abnormal chromosome segregation may arise either spontaneously or as a result of chemical/radiation exposure, particularly during cell division. Coastal ecosystems are continuously subjected to various contaminants originating from urban, industrial and agricultural activities. Genotoxicity is common to several families of major environmental pollutants, including pesticides, which therefore represent a potential important environmental hazard for marine organisms. A previous study demonstrated the vertical transmission of DNA damage by subjecting oyster genitors to short-term exposure to the herbicide diuron at environmental concentrations during gametogenesis. In this paper, Fluorescent in situ hybridization (FISH) was used to further characterize diuron-induced DNA damage at the chromosomal level. rDNA genes (5S and 18-5.8-28S), previously mapped onto Crassostrea gigas chromosomes 4, 5 and 10, were used as probes on the interphase nuclei of embryo preparations. Our results conclusively show higher aneuploidy (hypo- or hyperdiploidy) level in embryos from diuron-exposed genitors, with damage to the three studied chromosomal regions. This study suggests that sexually developing oysters are vulnerable to diuron exposure, incurring a negative impact on reproductive success and oyster recruitment.

In vivo effects of metaldehyde on Pacific oyster, Crassostrea gigas: comparing hemocyte parameters in two oyster families.

Pollutants via run-off into the ocean represent a potential threat to marine organisms, especially bivalves such as oysters living in coastal environments. These organisms filter large volumes of seawater and may accumulate contaminants within their tissues. Pesticide contamination in water could have a direct or indirect toxic action on tissues or cells and could induce alteration of immune system. Bivalve immunity is mainly supported by hemocytes and participates directly by phagocytosis to eliminate pathogens. Some studies have shown that pesticides can reduce immune defences and/or modify genomes in vertebrates and invertebrates. Metaldehyde is used to kill slugs, snails and other terrestrial gastropods. Although metaldehyde has been detected in surface waters, its effects on marine bivalves including the Pacific oyster, Crassostrea gigas, have never been studied. Given the mode of action of this molecule and its targets (molluscs), it could be potentially more toxic to oysters than other pesticides (herbicides, fungicides, insecticides, etc.). Effects of metaldehyde on oyster hemocyte parameters were thus monitored through in vivo experiments based on a short-term exposure. In this work, metaldehyde at 0.1 µg/L, which corresponds to an average concentration detected in the environment, modulated hemocyte activities of Pacific oysters after an in vivo short-term contact. Individuals belonging to two families showed different behaviours for some hemocyte activities after contamination by metaldehyde. These results suggested that effects of pollutants on oysters may differ from an individual to another in relation to genetic diversity. Finally, it appears essential to take an interest in the effects of metaldehyde on a wide variety of aquatic invertebrates including those that have a significant economic impact.

Effects of increasing temperatures on biomarker responses and accumulation of hazardous substances in rope mussels (Mytilus galloprovincialis) from Bizerte lagoon.

This study examined the influence of increasing temperatures in spring and summer on biochemical biomarkers in Mytilus galloprovincialis mussels sampled from Bizerte lagoon (northern Tunisia). Spatial and seasonal variations in a battery of seven biomarkers were analyzed in relation to environmental parameters (temperature, salinity, and pH), physiological status (condition and gonad indexes), stress on stress (SoS), and chemical contaminant levels (heavy metals, polycyclic aromatic hydrocarbons (PAHs), and PCBs) in digestive glands. Integrated biological response (IBR) was calculated using seven biomarkers (acetylcholinesterase (AChE), benzo[a]pyrene hydroxylase (BPH), multixenobiotic resistance (MXR), glutathione S-transferase (GST), catalase (CAT), malondialdehyde (MDA), and metallothioneins (MT). Seasonal variations in biological response were determined during a critical period between spring and summer at two sites, where chemical contamination varies by a factor of 2 for heavy metals and a factor 2.5 for PAHs. The analysis of a battery of biomarkers was combined with the measurement of physiological parameters at both sites, in order to quantify a maximum range of metabolic regulation with a temperature increase of 11 °C between May and August. According to our results, the MT, MDA, CAT, and AChE biomarkers showed the highest amplitude during the 11 °C rise, while the BPH, GST, and MXR biomarkers showed the lowest amplitude. Metabolic amplitude measured with the IBR at Menzel Abdelrahmen-the most severely contaminated station-revealed the highest metabolic stress in Bizerte lagoon in August, when temperatures were highest 29.1 °C. This high metabolic rate was quantified for each biomarker in the North African lagoon area and confirmed in August, when the highest IBR index values were obtained at the least contaminated site 2 (IBR = 9.6) and the most contaminated site 1 (IBR = 19.6). The combined effects of chemical contamination and increased salinity and temperatures in summer appear to induce a highest metabolic adaptation response and can therefore be used to determine thresholds of effectiveness and facilitate the interpretation of monitoring biomarkers. This approach, applied during substantial temperature increases at two sites with differing chemical contamination, is a first step toward determining an environmental assessment criteria (EAC) threshold in a North African lagoon.

Pacific oyster (Crassostrea gigas) hemocyte are not affected by a mixture of pesticides in short-term in vitro assays.

Pesticides are frequently detected in estuaries among the pollutants found in estuarine and coastal areas and may have major ecological consequences. They could endanger organism growth, reproduction, or survival. In the context of high-mortality outbreaks affecting Pacific oysters, Crassostrea gigas, in France since 2008, it appears of importance to determine the putative effects of pesticides on oyster susceptibility to infectious agents. Massive mortality outbreaks reported in this species, mainly in spring and summer, may suggest an important role played by the seasonal use of pesticides and freshwater input in estuarine areas where oyster farms are frequently located. To understand the impact of some pesticides detected in French waters, their effects on Pacific oyster hemocytes were studied through short-term in vitro experiments. Bivalve immunity is mainly supported by hemocytes eliminating pathogens by phagocytosis and producing compounds including lysosomal enzymes and antimicrobial molecules. In this study, oyster hemocytes were incubated with a mixture of 14 pesticides and metaldehyde alone, a molecule used to eliminate land mollusks. Hemocyte parameters including dead/alive cells, nonspecific esterase activities, intracytoplasmic calcium, lysosome number and activity, and phagocytosis were monitored by flow cytometry. No significant effect of pesticides tested at different concentrations was reported on oyster hemocytes maintained in vitro for short-term periods in the present study. It could be assumed that these oyster cells were resistant to pesticide exposure in tested conditions and developing in vivo assays appears as necessary to better understand the effects of pollutants on immune system in mollusks.

Multigenerational exposure of the microalga Tetraselmis suecica to diuron leads to spontaneous long-term strain adaptation.

To investigate the ability of microalgae to develop stable, long-term resistance to herbicides, the marine microalga Tetraselmis suecica was exposed to the herbicide diuron (5 µg/L) for a 43-generation exposure period followed by a 12-generation depuration phase. During the first 25 generations, diuron-exposed cultures showed doubling times ranging from 1.95 to 2.6 days, which was 2 to 2.5-fold longer than control cultures. Between generations 25 and 38, during diuron exposure, two out of the three exposed cultures exhibited a spontaneous drop in doubling time. These results provided evidence of culture adaptation to diuron. To assess persistence of the diuron adaptation observed on growth performance, one of the adapted cultures (D3) was maintained for 12 months in unexposed conditions and then tested by a second, short-term exposure to diuron 5 µg/L, in parallel with a control culture (C1) for six generations. Flow cytometry analyses were used to monitor cell density, viability, morphology, relative chlorophyll content and intracellular reactive oxygen species (ROS) level. Under these conditions, diuron induced a strong increase of doubling time in exposed-C1 cultures (2.5-fold longer than unexposed-C1 cultures), but no significant increase occurred in exposed D3-cultures compared with unexposed D3- and unexposed C1-cultures, showing the persistence of adaptation in the previously-exposed strain D3. Intracellular ROS level showed the same trend. Significant differences were observed between these strains, with weaker effects of diuron on strain D3 compared with strain C1: forward scatter (FSC), representing relative cell size, decreased in exposed cultures (67.8% and 95% of the controls for C1 and D3, respectively), whereas FL3 as relative chlorophyll content increased in exposed cultures (115.6% and 108.6% of the controls for C1 and D3, respectively). Results of second exposure to diuron revealed that the adaptation of strain D3 had persisted after 12 months of depuration, as no growth impairment was observed. This study demonstrates the possible appearance of stable diuron resistance in microalgae in cases of strong, multigenerational chronic exposure to this herbicide in polluted environments.