Environmental metabolomics is a powerful approach to investigate the response of organisms to contaminant exposure at a molecular scale. However, metabolomic responses to realistic environmental conditions can be hindered by factors intrinsic to the environment and the organism. Hence, a well-designed experimental exposure associated with adequate statistical analysis could be helpful to better characterize and relate the observed variability to its different origins. In the current study, we applied a multifactorial experiment combined to Analysis of variance Multiblock Orthogonal Partial Least Squares (AMOPLS), to assess the metabolic response of wild marine mussels, Mytilus galloprovincialis, exposed to a wastewater treatment plant effluent, considering gender as an experimental factor. First, the total observed variability was decomposed to highlight the contribution of each effect related to the experimental factors. Both the exposure and the interaction gender × exposure had a statistically significant impact on the observed metabolic alteration. Then, these metabolic patterns were further characterized by analyzing the individual variable contributions to each effect. A main change in glycerophospholipid levels was highlighted in both males and females as a common response, possibly caused by oxidative stress, which could lead to reproductive disorders, whereas metabolic alterations in some polar lipids and kynurenine pathway were rather gender-specific. This may indicate a disturbance in the energy metabolism and immune system only in males. Finally, AMOPLS is a useful tool facilitating the interpretation of complex metabolomic data and is expected to have a broad application in the field of ecotoxicology.

Scientists often set ambitious targets using environmental metabolomics to address challenging ecotoxicological issues. This promising approach has a high potential to elucidate the mechanisms of action (MeOAs) of contaminants (in hazard assessments) and to develop biomarkers (in environmental biomonitoring). However, metabolomics fingerprints often involve a complex mixture of molecular effects that are hard to link to a specific MeOA (if detected in the analytical conditions used). Given these promises and limitations, here we propose an updated review on the achievements of this approach. Metabolomics-based studies conducted on the effects of pharmaceutical active compounds in aquatic organisms provide a relevant means to review the achievements of this approach, as prior knowledge about the MeOA of these molecules could help overcome some shortcomings. This review highlighted that current metabolomics advances have enabled more accurate MeOA assessment, especially when combined with other omics approaches. The combination of metabolomics with other measured biological endpoints has also turned out to be an efficient way to link molecular effects to (sub)-individual adverse outcomes, thereby paving the way to the construction of adverse outcome pathways (AOPs). Here, we also discuss the importance of determining MeOA as a key strategy in the identification of MeOA-specific biomarkers for biomonitoring. We have put forward some recommendations to take full advantage of environmental metabolomics and thus help fulfil these promises.

There is growing evidence of the presence of pharmaceuticals in natural waters and their accumulation in aquatic organisms. While their mode of action on non-target organisms is still not clearly understood, their effects warrant assessment. The present study assessed the metabolome of the Mediterranean mussel (Mytilus galloprovincialis) exposed to a 10 µg/L nominal concentration of the antidepressant venlafaxine (VLF) at 3 time-points (1, 3, and 7 days). Over the exposure period, we observed up- or down-modulations of 113 metabolites, belonging to several metabolisms, e.g., amino acids (phenylalanine, tyrosine, tryptophan, etc.), purine and pyrimidine metabolisms (adenosine, cyclic AMP, thymidine, etc.), and several other metabolites involved in diverse functions. Serotonin showed the same time-course modulation pattern in both male and female mussels, which was consistent with its mode of action in humans, i.e., after a slight decrease on the first day of exposure, its levels increased at day 7 in exposed mussels. We found that the modulation pattern of impacted metabolites was not constant over time and it was gender-specific, as male and female mussels responded differently to VLF exposure.