Oliver A.H. Jones

Systems toxicology approaches for understanding the joint effects of environmental chemical mixtures

Environmental mixtures of chemicals constitute a prevalent issue in ecotoxicology and the development of new methods to reduce the uncertainties associated with their ecological risk assessment is a critical research need. Historically, a number of models have been explored to predict the potential combined effects of chemicals on species. These models, especially concentration addition and the independent action, have been applied to a number of mixtures. While often providing a good prediction of joint effect, there are cases where these models can have limitations: notably in cases where there are interactions for which they fail to adequately predict joint effects. To support the better mechanistic understanding of interactions in mixture toxicology a framework to support experimental studies to investigate the basis of observed interactions is proposed. The conceptual framework is derived from the extension of a three stage scheme which has previously been applied to understand chemical bioavailability. The framework considers that interactions in mixtures result from processes related to 1) the speciation, binding and transport of chemicals in the exposure medium (external exposure); 2) the adsorption, distribution, metabolism and excretion of chemicals within the organisms (toxicokinetics); 3) associations governing the binding and toxicity of the chemical(s) at the target site (toxicodynamics). The current state of the art in (eco)toxicology in relation to investigation of the mechanisms of interactions between chemicals is discussed with particular emphasis towards the multi-disciplinary tools and techniques within environmental chemistry; toxicology; biochemistry and systems biology that can be used to address such effects.

Environmental pollution and diabetes: a neglected association

Using cross-sectional data from the 1999–2002 US National Health and Examination Survey, Duk-Hee Lee and colleagues reported a strong correlation between insulin resistance and serum concen trations of persistent organic pollutants, especially for organochlorine com pounds. This result was a surprise for many people working in diabetes research, because most studies to date have focused on the eff ects of genetics and the westernisation of dietary habits and lifestyle, while ignoring the potential eff ect of xenobiotics. Nevertheless, as discussed by Porta, an increasing number of reports suggest that chronic dietary exposure to, and accumulation of, low concentrations of environmental pollutants within the body might also be associated with diabetogenesis. Ignoring the potential eff ects of xenobiotics therefore risks ignoring a potentially crucial component in the study of type 2 diabetes, obesity, and the metabolic syndrome. Lee and co-workers note two important fi ndings related to obesity. The fi rst was that the expected association between obesity and diabetes was absent in people with low concentrations of persistent organic pollutants in their blood. The second was that the association between obesity and diabetes became stronger as the concentrations of such pollutants in the blood increased. Interestingly, this report is not the fi rst time that the presence of persistent organic pollutants and other organochlorine compounds in blood have been associated with diabetes. Rylander and colleagues analysed the concentrations of 2,2’,4,4’,5,5’-hexachlorobiphenyl (CB-153) and 1,1dichloro-2,2-bis (p-chlorophenyl) -ethylene (p,p’-DDE, a breakdown product of DDT) in the serum of 196 men and 184 women in a fi shing community in Sweden. Both compounds showed a strong association with the occurrence of diabetes. Much the same results were seen in a population-based study in Belgium. Studies of US Air Force veterans of the Vietnam War exposed to Agent Orange also suggest an adverse relation between dioxin exposure and symptoms of diabetes. The potential of xenobiotics to disrupt glucose and lipid metabolism in mammals is a well-developed theory in toxicology. Indeed, many of the early toxicity responses in animal studies with a range of pollutants note glucosuria, dyslipidaemia, increased gluconeogenesis, and fatty liver. The versatility of high-throughput screening in metabolomics and metabonomics is an especially useful way of monitoring metabolic changes caused by disease or exposure to toxicants (eg, heavy metals) in animal models. Although most studies have tested acute exposure (ie, less than 2 weeks), in the few chronic exposure studies (ie, more than 3 months), glucosuria was recorded as long-term eff ect. This result suggests that diabetes could be exacerbated by chronic exposure to xenobiotics that disrupt normal metabolism of glucose and lipids. Although correlation does not necessarily prove the existence of a causal relation, biologically plausible ex plan ations exist for this association. For instance, dioxin-like compounds exert their eff ects through binding to the aryl-hydrocarbon receptor. This receptor is thought to antagonise peroxisome proliferator-activated receptors. One plausible hypothesis, there fore, is that the aryl-hydrocarbon receptor promotes dia beto genesis by antagonism of peroxisome proliferator-activated receptors. However, no data are avail able to support this See Correspondence page 302

Standard reporting requirements for biological samples in metabolomics experiments: Environmental context

Metabolomic technologies are increasingly being applied to study biological questions in a range of different settings from clinical through to environmental. As with other high-throughput technologies, such as those used in transcriptomics and proteomics, metabolomics continues to generate large volumes of complex data that necessitates computational management. Making sense of this wealth of information also requires access to sufficiently detailed and well annotated meta-data. Here we provide standard reporting requirements for describing biological samples, taken from an environmental context and involved in metabolomic experiments. It is our intention that these reporting requirements should guide and support the standardised annotation, dissemination and interpretation of environmental metabolomics meta-data.

Mixtures of similarly acting compounds in Daphnia magna: From gene to metabolite and beyond

Daphnia are an important and widely studied model species in ecological and toxicological studies throughout the world and an official (OECD) recommended test organism. Their small size, wide distribution and easy growth conditions make this organism ideal for functional genomics based studies, including metabolic profiling and transcriptomics. In this study we used an integrated systems approach in which transcriptomic, metabolomic and energetic responses of juvenile (4days old) daphnids were evaluated in response to exposure to two poly aromatic hydrocarbons (pyrene and fluoranthene) and binary mixtures thereof. In addition, these responses were linked to responses measured during chronic experiments (21days) assessing survival, growth and reproductive traits. Custom Daphnia magna microarrays were used to assess transcriptomic changes. Hierarchical cluster analysis did not result in a clear distinction between the single compounds suggesting similar molecular modes of action. Cluster analysis with both the single compounds and the binary mixture treatments resulted in a separation of treatments based on differences in toxic ratios rather than component differences. Changes in the metabolic profiles of the organisms were investigated using Nuclear Magnetic Resonance Spectroscopy and Gas and Liquid Chromatography Mass Spectrometry. These multivariate metabolomic datasets were analyzed with Principal Components Analysis and Partial Least Squares Discriminant Analysis. The major metabolite changes responsible for the differences observed indicated a disturbance in aminosugar metabolism in all cases. The study demonstrates the potential of omics to provide screening tools for monitoring of the freshwater environment--in invertebrate species--which is reasonably rapid, cost-effective and has the potential to greatly increase the amount of information obtained from aquatic toxicology testing.

Size-dependent effects of low level cadmium and zinc exposure on the metabolome of the Asian clam, Corbicula fluminea

The toxic effects of low level metal contamination in sediments are currently poorly understood. We exposed different sized Asian clams, Corbicula fluminea, to sediment spiked with environmentally relevant concentrations of either zinc, cadmium or a zinc-cadmium mixture for one week. This freshwater bivalve is well suited for sediment toxicity tests as it lives partly buried in the sediment and utilises sediment particles as a food resource. After one week, the whole tissue composition of low molecular weight metabolites was analysed by nuclear magnetic resonance spectroscopy (NMR) and gas chromatography-mass spectrometry (GC-MS). The condition index (ratio of tissue dry weight to volume inside the shell valves) was also measured. Small and large clams were clearly differentiated by their metabolic composition and the two size classes showed opposite responses to the mixture spiked sediment. No effects of zinc alone on the metabolome were found and cadmium only influenced the smaller size class. The main perturbations were seen in amino acid and energy metabolism, with small clams using amino acids as an energy resource and larger clams primarily drawing on their larger storage reserves of carbohydrates. Our study demonstrates that metabolomics is a useful technique to test for low level toxicity which does not manifest in mortality or condition index changes. The differing effects between the two size classes stress that it is important to consider age/size when conducting metabolomic and ecotoxicology assessments, since testing for the effects on only one size class makes it more difficult to extrapolate laboratory results to the natural environment.

A comparative metabolomic study of NHR-49 in Caenorhabditis elegans and PPAR-α in the mouse

Proton Nuclear Magnetic Resonance spectroscopy and Gas Chromatography Mass Spectrometry based metabolomics has been used in conjunction with multivariate statistics to examine the metabolic changes in Caenorhabditis elegans following the deletion of nuclear hormone receptor‐49 (nhr‐49). Deletion of the receptor produced profound changes in fatty acid metabolism, in particular an increase in the ratio of unsaturated to saturated fatty acids, a decrease in the concentration of glucose and increases in lactate and alanine. Given the proposed functional similarity between nhr‐49 and the mammalian peroxisome proliferator‐activated receptors (PPARs) these changes were compared with the metabolome of the PPAR‐α null mouse. The metabolomic approach demonstrated a number of similarities including the regulation of lipid synthesis, β‐oxidation of fatty acids and changes in glycolysis/gluconeogenesis.

Using metabolic profiling to assess plant-pathogen interactions: an example using rice (Oryza sativa) and the blast pathogen Magnaporthe grisea

A metabolomics based approach has been used to study the infection of the Hwacheong rice cultivar (Oryza sativa L. cv. Hwacheong) with compatible (KJ201) and incompatible (KJ401) strains of the rice blast fungal pathogen Magnaporthe grisea. The metabolic response of the rice plants to each strain was assessed 0, 6, 12, 24, 36, and 48xa0h post inoculation. Nuclear Magnetic Resonance (NMR) spectroscopy and Gas and Liquid Chromatography Tandem Mass spectrometry (GC/LC-MS/MS) were used to study both aqueous and organic phase metabolites, collectively resulting in the identification of 93 compounds. Clear metabolic profiles were observed at each time point but there were no significant differences in the metabolic response elicited by each pathogen strain until 24xa0h post inoculation. The largest change was found to be in alanine, which was ~30% (±9%) higher in the leaves from the compatible, compared to the resistant, plants. Together with several other metabolites (malate, glutamine, proline, cinnamate and an unknown sugar) alanine exhibited a good correlation between time of fungal penetration into the leaf and the divergence of metabolite profiles in each interaction. The results indicate both that a wide range of metabolites can be identified in rice leaves and that metabolomics has potential for the study of biochemical changes in plant-pathogen interactions.

Omic analyses unravels global molecular changes in the brain and liver of a rat model for chronic sake (Japanese alcoholic beverage) intake

The effects of chronic administration of Sake (Japanese alcoholic beverage, Nihonshu) on brain and liver of female F334 (Fisher) rats were surveyed via global omic analyses using DNA microarray, 2‐DE, and proton nuclear magnetic resonance. Rats weaned at 4u2009wk of age were given free access to Sake (15% alcohol), instead of water. At 13 months of age, and 24u2009h after withdrawal of Sake supply, rats were sacrificed, and the whole brain and liver tissues dissected for analyses. In general, molecular changes in brain were found to be less than those in liver. Transcriptomics data revealed 36 and 9, and 80 and 62 up‐ and down‐regulated genes, in the brain and liver, respectively, with binding and catalytic activity gene categories the most prominently changed. Results suggested Sake‐induced fragility of brain and liver toxicity/damage, though no significant abnormalities in growth were seen. At protein level, a striking decrease was found in the expression of NADH dehydrogenase (ubiquinone) Fe‐S protein 1 in brain, suggesting attenuation of mitochondrial metabolism. In liver, results again suggested an attenuation of mitochondrial function and, in addition, glycoproteins with unknown function were induced at protein and gene levels, suggesting possible changes in glycoprotein binding in that organ. Metabolomic analysis of brain revealed significant increases in valine, arginine/ornithine, alanine, glutamine, and choline with decreases in isoleucine, N‐acetyl aspartate, taurine, glutamate, and gamma aminobutyric acid. Our results provide a detailed inventory of molecular components of both brain and liver after Sake intake, and may help to better understand effects of chronic Sake drinking.

Cellular acidosis in rodents exposed to cadmium is caused by adaptation of the tissue rather than an early effect of toxicity

Proton ((1)H) Nuclear Magnetic Resonance (NMR) spectroscopy was used to investigate the biochemical response of bank voles and wood mice (two wild rodent species frequently found on metal-contaminated sites) to chronic cadmium (Cd) insult. Similar effects, in terms of both metabolic changes (consistent with cellular acidosis) and induced metallothionin (MT) production were observed in all animals. These changes appeared to be an adaptation of the liver to toxic insult rather than onset of a toxic effect, and, in common with previous studies, were more marked in bank voles than wood mice. This may have reflected the greater Cd intake and assimilation of the former but was not explained by differences in concentrations of free (non MT-bound) Cd; concentrations of which were negligible in both voles and mice. Responses to Cd insult were detected in both species even though their bodies contained cadmium concentrations well below the World Health Organisation critical renal concentration of 200 mug/g dry mass.