Final Report Summary - DEVNERTOX (Toxic theats to the developing nervous system: in vivo and in vitro studies on the effects of mixture of neurotoxic substances potentially contaminating food)
The presence of contaminating substances in food is an issue that has received a lot of attention. Because of the sensitivity of the developing central nervous system to toxic insults, there is a particular concern about potential neurotoxic effects of substances that foetuses and children are exposed to, in utero, via breast milk or food. Methylmercury (MeHg) and polychlorinated biphenyls (PCBs) are persistent pollutants that enter the lowest stages of the food chain and then steadily become more concentrated at each stage. Although the developmental neurotoxic effects of these substances are well established, their mechanisms of action are not fully understood. In addition, few studies have investigated the effects of simultaneous exposures, which is the real-life situation for individuals consuming polluted fish, the main source of human exposure to MeHg and PCBs.
The project focused on generating experimental models to improve developmental neurotoxicity testing and risk assessment for neurotoxic food contaminants. Three compounds were studied alone and in combination, MeHg and two PCBs with different chemical properties; the non-dioxin-like di-ortho-substituted PCB 153 and the coplanar dioxin-like PCB 126.
In contrast to other fields of toxicology, there was still limited knowledge about the exact mechanism(s) of action of most environmental neurotoxicants, and this is one of the reasons why the identification of end points for neurotoxic effects had been progressing slowly. Moreover, the anatomical and functional complexity of the nervous system and the consequent multiplicity of adverse effects created additional difficulties.
DEVNERTOX studies suggested that a winning strategy to identify neurotoxic substances was to use in vitro methodologies to assess their toxic potential and characterise the mechanisms of actions. Changes in relevant endpoints were found in vitro at concentrations comparable to those critical for human neurodevelopment. Notably, similar effects on neurotransmission were found both in exposed animals and in cell systems exposed in vitro.
The complexity of the nervous system required the use of multiple cellular models that can represent different neural targets potentially affected in vivo. In this respect, neural stem cells seemed to be a promising model to investigate developmental neurotoxicity. Neural cells exposed to the selected toxic agents can undergo different types of cell death and, in turn, toxicants can trigger multiple death pathways that may cross-talk. The elucidation of the cell death pathways activated in different neural targets is highly relevant for our understanding of the mechanisms of action of toxic agents that can harm the nervous system.
The analysis of behaviour was a first-choice approach to detect subtle modifications in vivo, mainly because it offered the possibility to analyse the same animals at different stages of development. Impaired learning ability has been observed both in humans and animals prenatally exposed to PCBs or MeHg. In the present studies, learning and memory functions were found to be impaired in both rats and mice and by multiple tests. Studies on depressive-like behaviour indicate that the developmental exposure to MeHg could affect not only cognitive functions but also motivation-driven behaviour in mice. Neurochemical, hormonal and neurobehavioural alterations were found, sometimes with species differences. For example, toxicant exposure could impair a specific neurotransmission system in cultured cells as well as in rat brain, the rat results being correlated with decreased learning and memory.
In general, ageing did not to worsen behavioural effects. Gender differences were detected for a number of end points, including hormonal levels and some changes in behaviour and neurotransmission. Similar to the in vitro studies, effects of concomitant exposures to MeHg and either PCB differed between the different in vivo models.
To summarise, in vitro studies are important complementary approaches to studies in experimental animals. A battery of well characterised in vitro models, like those used in the DEVNERTOX project, allowed for a better mechanistic-based assessment of potentially neurotoxic substances. Neural stem cells are highly relevant to study neurodevelopmental effects and the project proposed that they can be used as sensitive in vitro models for the evaluation of toxic effects on both differentiation and cell survival. Results also showed that to compare in vivo and in vitro results it is important to measure actual levels in tissues and cells, since cellular levels can vary between different cell types exposed to the same concentrations in the cell medium. Due to differences in methodology and uncertainties in comparing in vitro and in vivo results such comparisons should be considered as "order of magnitude". The data on brain Hg levels in children were very limited and even less was known about human PCB 153 and 126 levels.
The lack of information that usually justifies experimental studies calls for the use of batteries of tests and multiple cell systems, which may lead to identification of the most sensitive cell type(s) and potential biomarkers. The most critical issue in risk assessment remains exposure characterisation, which is often based on rough surrogates (e.g. occurrence by intake calculations) of relevant target (cellular) doses experimentally determined, A biomonitoring approach using validated and relevant biomarkers seemed to be promising to characterise the risk of adverse neurodevelopmental effects and to calculate the margin of exposure resulting from food and other sources in target populations. Overall, DEVNERTOX studies suggested that such an approach is feasible. Further studies are necessary to address the issue of combined or complex exposures.
With simple binary combinations of MeHg and PCB 153 and 126, partners were unable to demonstrate that sub-threshold exposure can result in effective cumulative doses. However, they found significant effects at doses close to those found in epidemiological studies. At these levels of exposure, they noted that even the sequence of asynchronous exposures to the same doses of the same toxicants may produce different effects, which calls for additional efforts to develop a conceptual framework where available human and animal data can be integrated with much cheaper and equally sound in vitro results.
The project focused on generating experimental models to improve developmental neurotoxicity testing and risk assessment for neurotoxic food contaminants. Three compounds were studied alone and in combination, MeHg and two PCBs with different chemical properties; the non-dioxin-like di-ortho-substituted PCB 153 and the coplanar dioxin-like PCB 126.
In contrast to other fields of toxicology, there was still limited knowledge about the exact mechanism(s) of action of most environmental neurotoxicants, and this is one of the reasons why the identification of end points for neurotoxic effects had been progressing slowly. Moreover, the anatomical and functional complexity of the nervous system and the consequent multiplicity of adverse effects created additional difficulties.
DEVNERTOX studies suggested that a winning strategy to identify neurotoxic substances was to use in vitro methodologies to assess their toxic potential and characterise the mechanisms of actions. Changes in relevant endpoints were found in vitro at concentrations comparable to those critical for human neurodevelopment. Notably, similar effects on neurotransmission were found both in exposed animals and in cell systems exposed in vitro.
The complexity of the nervous system required the use of multiple cellular models that can represent different neural targets potentially affected in vivo. In this respect, neural stem cells seemed to be a promising model to investigate developmental neurotoxicity. Neural cells exposed to the selected toxic agents can undergo different types of cell death and, in turn, toxicants can trigger multiple death pathways that may cross-talk. The elucidation of the cell death pathways activated in different neural targets is highly relevant for our understanding of the mechanisms of action of toxic agents that can harm the nervous system.
The analysis of behaviour was a first-choice approach to detect subtle modifications in vivo, mainly because it offered the possibility to analyse the same animals at different stages of development. Impaired learning ability has been observed both in humans and animals prenatally exposed to PCBs or MeHg. In the present studies, learning and memory functions were found to be impaired in both rats and mice and by multiple tests. Studies on depressive-like behaviour indicate that the developmental exposure to MeHg could affect not only cognitive functions but also motivation-driven behaviour in mice. Neurochemical, hormonal and neurobehavioural alterations were found, sometimes with species differences. For example, toxicant exposure could impair a specific neurotransmission system in cultured cells as well as in rat brain, the rat results being correlated with decreased learning and memory.
In general, ageing did not to worsen behavioural effects. Gender differences were detected for a number of end points, including hormonal levels and some changes in behaviour and neurotransmission. Similar to the in vitro studies, effects of concomitant exposures to MeHg and either PCB differed between the different in vivo models.
To summarise, in vitro studies are important complementary approaches to studies in experimental animals. A battery of well characterised in vitro models, like those used in the DEVNERTOX project, allowed for a better mechanistic-based assessment of potentially neurotoxic substances. Neural stem cells are highly relevant to study neurodevelopmental effects and the project proposed that they can be used as sensitive in vitro models for the evaluation of toxic effects on both differentiation and cell survival. Results also showed that to compare in vivo and in vitro results it is important to measure actual levels in tissues and cells, since cellular levels can vary between different cell types exposed to the same concentrations in the cell medium. Due to differences in methodology and uncertainties in comparing in vitro and in vivo results such comparisons should be considered as "order of magnitude". The data on brain Hg levels in children were very limited and even less was known about human PCB 153 and 126 levels.
The lack of information that usually justifies experimental studies calls for the use of batteries of tests and multiple cell systems, which may lead to identification of the most sensitive cell type(s) and potential biomarkers. The most critical issue in risk assessment remains exposure characterisation, which is often based on rough surrogates (e.g. occurrence by intake calculations) of relevant target (cellular) doses experimentally determined, A biomonitoring approach using validated and relevant biomarkers seemed to be promising to characterise the risk of adverse neurodevelopmental effects and to calculate the margin of exposure resulting from food and other sources in target populations. Overall, DEVNERTOX studies suggested that such an approach is feasible. Further studies are necessary to address the issue of combined or complex exposures.
With simple binary combinations of MeHg and PCB 153 and 126, partners were unable to demonstrate that sub-threshold exposure can result in effective cumulative doses. However, they found significant effects at doses close to those found in epidemiological studies. At these levels of exposure, they noted that even the sequence of asynchronous exposures to the same doses of the same toxicants may produce different effects, which calls for additional efforts to develop a conceptual framework where available human and animal data can be integrated with much cheaper and equally sound in vitro results.
