Search

no abstract

Chemicals in the aquatic environment can be harmful to organisms and ecosystems. Knowledge on effect concentrations as well as on mechanisms and modes of interaction with biological molecules and signaling pathways is necessary to perform chemical risk assessment and identify toxic compounds. To this end, we developed criteria and a pipeline for harvesting and summarizing effect concentrations from the US ECOTOX database for the three aquatic species groups algae, crustaceans, and fish and researched the modes of action of more than 3,300 environmentally relevant chemicals in literature and databases. We provide a curated dataset ready to be used for risk assessment based on monitoring data and the first comprehensive collection and categorization of modes of action of environmental chemicals. Authorities, regulators, and scientists can use this data for the grouping of chemicals, the establishment of meaningful assessment groups, and the development of in vitro and in silico approaches for chemical testing and assessment.

Sophisticated approaches for the in-silico prediction of toxicity are required to support the risk assessment of chemicals. The number of chemicals on the global chemical market and the speed of chemical innovation stand in massive contrast to the capacity for regularizing chemical use. We recently proved our ready-to-use application deepFPlearn as a suitable approach for this task. Here, we present its extension deepFPlearn+ incorporating i) a graph neural network to feed our AI with a more sophisticated molecular structure representation and ii) alternative train-test splitting strategies that involve scaffold structures and the molecular weights of chemicals. We show that the GNNs outperform the previous model substantially and that our models can generalize on unseen data even with a more robust and challenging test set. Therefore, we highly recommend the application of deepFPlearn+ on the chemical inventory to prioritize chemicals for experimental testing or any chemical subset of interest in monitoring studies.

It is a daunting task to impute a large dataset that has majority of data missing from its tens of thousands of predictors, but still to preserve a known cluster structure in the imputed results. In this study, we propose a novel two-step approach for this task. First, we use simple linear classification models to derive the global and local features of cluster structures from a template ground truth (i.e. known gold data). Second, we integrate the cluster features extracted from each cluster into a neural network architecture and its training responses for guiding the imputation process. Since our neural network utilizes the global and local features of gold data in training the imputation network, we refer our neural network as GLIN (Global-Local Imputation Network). We test our imputation method on two high-dimensional datasets: a single cell dataset and a movie rating dataset, that have up to of 95% missing rates. Finally, we use four evaluation metrics: distance, correlation, data distribution, and predictability difference, to evaluate how well the cluster structures of the gold data are preserved in the imputed results.

The possible applications for human retinal organoids (HROs) derived from human induced pluripotent stem cells (hiPSC) rely on the robustness and transferability of the methodology for their generation. Standardized strategies and parameters to effectively assess, compare, and optimize organoid protocols are starting to be established, but are not yet complete. To advance this, we explored the efficiency and reliability of a differentiation method, called CYST protocol, that facilitates retina generation by forming neuroepithelial cysts from hiPSC clusters. Here, we tested seven different hiPSC lines which reproducibly generated HROs. Histological and ultrastructural analyses indicate that HRO differentiation and maturation are regulated. The different hiPSC lines appeared to be a larger source of variance than experimental rounds. Although previous reports have shown that HROs in several other protocols contain a rather low number of cones, HROs from the CYST protocol are consistently richer in cones and with a comparable ratio of cones, rods, and Müller glia. To provide further insight into HRO cell composition, we studied single cell RNA sequencing data and applied CaSTLe, a transfer learning approach. Additionally, we devised a potential strategy to systematically evaluate different organoid protocols side-by-side through parallel differentiation from the same hiPSC batches: In an explorative study, the CYST protocol was compared to a conceptually different protocol based on the formation of cell aggregates from single hiPSCs. Comparing four hiPSC lines showed that both protocols reproduced key characteristics of retinal epithelial structure and cell composition, but the CYST protocol provided a higher HRO yield. So far, our data suggest that CYST-derived HROs remained stable up to at least day 200, while single hiPSC-derived HROs showed spontaneous pathologic changes by day 200. Overall, our data provide insights into the efficiency, reproducibility, and stabil

The thyroid hormones (THs) regulate various physiological mechanisms in mammals, such as cellular metabolism, cell structure, and membrane transport. The therapeutic drugs propylthiouracil (PTU) and phenytoin are known to induce hypothyroidism and decrease blood thyroid hormone levels. To analyze the impact of these two drugs on systemic metabolism, we focused on metabolic changes after treatment. Therefore, in a rat model, the metabolome of thyroid and liver tissue as well as from the blood plasma, after 2-week and 4-week administration of the drugs and after a following 2-week recovery phase, was investigated using targeted LC-MS/MS and GC-MS. Both drugs were tested at a low dose and a high dose. We observed decreases in THs plasma levels, and higher doses of the drugs were associated with a high decrease in TH levels. PTU administration had a more pronounced effect on TH levels than phenytoin. Both drugs had little or no influence on the metabolomes at low doses. Only PTU exhibited apparent metabolome alterations at high doses, especially concerning lipids. In plasma, acylcarnitines and triglycerides were detected at decreased levels than in the controls after 2- and 4-week exposure to the drug, while sphingomyelins and phosphatidylcholines were observed at increased levels. Interestingly, in the thyroid tissue, triglycerides were observed at increased concentrations in the 2-week exposure group to PTU, which was not observed in the 4-week exposure group and in the 4-week exposure group followed by the 2-week recovery group, suggesting an adaptation by the thyroid tissue. In the liver, no metabolites were found to have significantly changed. After the recovery phase, the thyroid, liver, and plasma metabolomic profiles showed little or no differences from the controls. In conclusion, although there were significant changes observed in several plasma metabolites in PTU/Phenytoin exposure groups, this study found that only PTU exposure led to adaptation-dependent ch

Many chemicals are present in our environment, and all living species are exposed to them. However, numerous chemicals pose risks, such as developing severe diseases, if they occur at the wrong time in the wrong place. For the majority of the chemicals, these risks are not known. Chemical risk assessment and subsequent regulation of use require efficient and systematic strategies. Lab-based methods—even if high throughput—are too slow to keep up with the pace of chemical innovation. Existing computational approaches are designed for specific chemical classes or sub-problems but not usable on a large scale. Further, the application range of these approaches is limited by the low amount of available labeled training data. We present the ready-to-use and stand-alone program deepFPlearn that predicts the association between chemical structures and effects on the gene/pathway level using a combined deep learning approach. deepFPlearn uses a deep autoencoder for feature reduction before training a deep feed-forward neural network to predict the target association. We received good prediction qualities and showed that our feature compression preserves relevant chemical structural information. Using a vast chemical inventory (unlabeled data) as input for the autoencoder did not reduce our prediction quality but allowed capturing a much more comprehensive range of chemical structures. We predict meaningful—experimentally verified—associations of chemicals and effects on unseen data. deepFPlearn classifies hundreds of thousands of chemicals in seconds. We provide deepFPlearn as an open-source and flexible tool that can be easily retrained and customized to different application settings at https://github.com/yigbt/deepFPlearn.

Human organoids could facilitate research of complex and currently incurable neuropathologies, such as age-related macular degeneration (AMD) which causes blindness. Here, we establish a human retinal organoid system reproducing several parameters of the human retina, including some within the macula, to model a complex combination of photoreceptor and glial pathologies. We show that combined application of TNF and HBEGF, factors associated with neuropathologies, is sufficient to induce photoreceptor degeneration, glial pathologies, dyslamination, and scar formation: These develop simultaneously and progressively as one complex phenotype. Histologic, transcriptome, live-imaging, and mechanistic studies reveal a previously unknown pathomechanism: Photoreceptor neurodegeneration via cell extrusion. This could be relevant for aging, AMD, and some inherited diseases. Pharmacological inhibitors of the mechanosensor PIEZO1, MAPK, and actomyosin each avert pathogenesis; a PIEZO1 activator induces photoreceptor extrusion. Our model offers mechanistic insights, hypotheses for neuropathologies, and it could be used to develop therapies to prevent vision loss or to regenerate the retina in patients suffering from AMD and other diseases.

Using retinal organoid systems, organ-like 3D tissues, relies implicitly on their robustness. However, essential key parameters, particularly retinal growth and longer-term culture, are still insufficiently defined. Here, we hypothesize that a previously optimized protocol for high yield of evenly-sized mouse retinal organoids with low variability facilitates assessment of such parameters. We demonstrate that these organoids reliably complete retinogenesis, and can be maintained at least up to 60 days in culture. During this time, the organoids continue to mature on a molecular and (ultra)structural level: They develop photoreceptor outer segments and synapses, transiently maintain its cell composition for about 5–10 days after completing retinogenesis, and subsequently develop pathologic changes – mainly of the inner but also outer retina and reactive gliosis. To test whether this organoid system provides experimental access to the retina during and upon completion of development, we defined and stimulated organoid growth by activating sonic hedgehog signaling, which in patients and mice in vivo with a congenital defect leads to enlarged eyes. Here, a sonic hedgehog signaling activator increased retinal epithelia length in the organoid system when applied during but not after completion of development. This experimentally supports organoid maturation, stability, and experimental reproducibility in this organoid system, and provides a potential enlarged retina pathology model, as well as a protocol for producing larger organoids. Together, our study advances the understanding of retinal growth, maturation, and maintenance, and further optimizes the organoid system for future utilization.

The function of mucosal-associated invariant T (MAIT) highly depends on the mode of activation, either by recognition of bacterial metabolites via their T cell receptor (TCR) or in a TCR-independent manner via cytokines. The underlying molecular mechanisms are not entirely understood. To define the activation of MAIT cells on the molecular level, we applied a multi-omics approach with untargeted transcriptomics, proteomics and metabolomics. Transcriptomic analysis of E.coli- and TCR-activated MAIT cells showed a distinct transcriptional reprogramming, including altered pathways, transcription factors and effector molecules. We validated the consequences of this reprogramming on the phenotype by proteomics and metabolomics. Thus, and to distinguish between TCR-dependent and -independent activation, MAIT cells were stimulated with IL12/IL18, anti-CD3/CD28 or both. Only a combination of both led to full activation of MAIT cells, comparable to activation by E.coli. Using an integrated network-based approach, we identified key drivers of the distinct modes of activation, including cytokines and transcription factors, as well as negative feedback regulators like TWIST1 or LAG3.Taken together, we present novel insights into the biological function of MAIT cells, which may represent a basis for therapeutic approaches to target MAIT cells in pathological conditions.

Ideally, a Systems Toxicology (ST) approach is aimed at by (eco)toxicologists, i.e. a multidisciplinary area incorporating classical toxicological concepts with omics technologies, and the understanding of this through computational data sciences, chemistry, mathematics, and physics modelling. As outlined in several public reports (e.g. from ECHA-European Chemical Agency and EFSA-European Food Safety Authority), the way forward in the coming years in Europe is to integrate New Approach Methodologies (NAMs) (including omics technologies) into hazard and hence risk assessment (RA). Adverse Outcome Pathways (AOPs) describe a sequence of events in response to stress, from the molecular initiating event until an adverse outcome, which is relevant to RA or regulatory decision-making. AOPs are one of the facilitators to integrate mechanistic data into RA, but it is urgent to increase the inclusion of the vast mechanistic knowledge available, especially for the RA of novel smart and advanced materials (AdMa) with multifunctional characteristics. There are still many challenges to the routine usage of NAMs, e.g. omics-based information. Here, we summarise the current state of the art of ST, the benefits of human and environmental health cross knowledge and the available methods and output. The importance of this area has been highlighted for many years but is even more pressing in the context of AdMa. Furthermore, we outline the challenges and suggest recommendations for future implementation.

Organisms are exposed to ever changing complex mixtures of chemicals over the course of their lifetime. The need to more comprehensively describe this exposure and relate it to adverse health effects has led to formulation of the exposome concept in human toxicology. Whether this concept has utility in the context of environmental hazard and risk assessment has not been discussed in detail. In this paper, we propose – by analogy to the human exposome - to define the eco-exposome as the totality of the internal exposure (anthropogenic and natural chemicals, their biotransformation products or adducts, and endogenous signaling molecules that may be sensitive to an anthropogenic chemical exposure) over the life time of an ecologically relevant organism. We describe how targeted and non-targeted chemical analyses and bioassays can be employed to characterize this exposure and discuss how the adverse outcome pathway (AOP) concept could be used to link this exposure to adverse effects. Available methods, their limitations and/or requirement for improvements for practical application of the eco-exposome concept are discussed. Even though analysis of the eco-exposome can be resource-intensive and challenging, new approaches and technologies make this assessment increasingly feasible. Furthermore, an improved understanding of mechanistic relationships between external chemical exposure(s), internal chemical exposure(s), and biological effects could result in the development of proxies; that is, relatively simple chemical and biological measurements that could be used to complement internal exposure assessment or infer the internal exposure when it is difficult to measure.

no abstract