Human acute toxicity prediction of the first 50 MEIC chemicals by a battery of ecotoxicological tests and physicochemical properties

Global occurrence and aquatic hazards of antipsychotics in sewage influents, effluent discharges and surface waters

Despite increasing reports of pharmaceuticals in surface waters, aquatic hazard information remains limited for many contaminants, particularly for sublethal, chronic responses plausibly linked to molecular initiation events that are largely conserved across vertebrates. Here, we critically examined available refereed information on the occurrence of 67 antipsychotics in wastewater effluent and surface waters. Because the majority of sewage remains untreated around the world, we also examined occurrence in sewage influents. When sufficient information was available, we developed probabilistic environmental exposure distributions (EEDs) for each compound in each matrix by geographic region. We then performed probabilistic environmental hazard assessments (PEHAs) using therapeutic hazard values (THVs) of each compound, due to limited sublethal aquatic toxicology information for this class of pharmaceuticals. From these PEHAs, we determined predicted exceedances of the respective THVs for each chemical among matrices and regions, noting that THV values of antipsychotic contaminants are typically lower than other classes of human pharmaceuticals. Diverse exceedances were observed, and these aquatic hazards varied by compound, matrix and geographic region. In wastewater effluent discharges and surface waters, sulpiride was the most detected antipsychotic; however, percent exceedances of the THV were minimal (0.6%) for this medication. In contrast, we observed elevated aquatic hazards for chlorpromazine (30.5%), aripiprazole (37.5%), and perphenazine (68.7%) in effluent discharges, and for chlorprothixene (35.4%) and flupentixol (98.8%) in surface waters. Elevated aquatic hazards for relatively understudied antipsychotics were identified, which highlight important data gaps for future environmental chemistry and toxicology research.

The Antidepressants Amitriptyline and Paroxetine Induce Changes in the Structure and Functional Traits of Marine Nematodes

Increasing concentrations of the antidepressants amitriptyline and paroxetine were determined recently in marine habitats. However, their impact on marine biota is understudied, despite multiple undesirable effects they have on the environment. An important behavioral aspect that is increasingly measured following exposure to contaminants is the migration of fauna from contaminated areas. Hence, our aim was to better understand the migration pattern of marine meiobenthic fauna, but with a main focus on nematodes, following the exposure to both antidepressants, alone or in mixture. The experiment was carried out in microcosms, which comprised an uncontaminated upper and a lower contaminated compartment, where amitriptyline was added, alone or mixed with paroxetine, at concentrations of 0.4 and 40 µg L−1. The overall abundance of meiobenthic groups decreased significantly following exposure to amitriptyline in both compartments, a pattern augmented by the mixture with paroxetine. The migration of nematodes towards the upper compartments of microcosms was triggered by the level of contamination with antidepressants. As such, the species Terschellingia longicaudata showed no significant change in abundance, suggesting tolerance to both antidepressants. On the other hand, the abundances of nematode taxa Cyatholaimus prinzi, Calomicrolaimus sp., Calomicrolaimus honestus, Neochromadora sp., Chromadorina sp. and Chromadorina minor decreased significantly following the exposure to both antidepressants, even at low concentrations. At the end of the experiment, the dominant migratory nematodes belonged to deposit-feeders and omnivores-carnivores trophic guilds, with tail shapes of e/f types and body-sizes longer than 2 mm. Such functional traits increase their mobility in sediments and the chance to move away from contaminated habitats. Moreover, the sex ratio was imbalanced in the favor of males in contaminated lower compartments with mixtures of the lowest and highest concentrations of amitriptyline and paroxetine, suggesting that these drugs also affect the hormone system. In conclusion, the exposure to the antidepressants amitriptyline and paroxetine triggered important changes within nematode communities, as changes in taxonomic composition were a result of migration and survival of tolerant taxa, but equally acting on the hormone system and leading to unbalanced sex-ratio among the residents.

Comparative toxicity, phytochemistry, and use of 53 Philippine medicinal plants

The study compares the toxicity of 53 selected medicinal plants commonly used in the Philippines to treat various diseases. It uses as a benchmark Vitex negundo L., which was approved by the Philippine Food and Drug Administration as an herbal drug for cough and asthma after passing clinical trials for safety and efficacy. The methods were chosen for their simplicity and accessibility even for resource-limited laboratories. Extracts (95 % ethanol) of the medicinal parts of the plants were (1) chemically profiled using qualitative phytochemical tests that detect the presence of key classes of bioactive compounds; and (2) evaluated for toxicity using the brine shrimp (Artemia sp.) lethality assay (BSLA). General phytochemical screening revealed the presence of tannins in 50 plant extracts, alkaloids in 43, glycosides in 33, flavonoids in 31, steroids in 21, triterpenoids in 20, anthraquinones in 10, and saponins in 8. Extracts from eight plants had LC50 values lower than the potassium dichromate control (approximately 12 μg/mL) and were considered highly toxic; extracts from 21 plants had LC50 values between 12 μg/mL and 100 μg/mL and were considered moderately toxic; extracts from 19 plant extracts, including Vitex negundo and some common vegetables, had LC50 values between 100 μg/mL and 500 μg/mL, and were considered mildly toxic and likely to have reasonable safety margins; five plant extracts, including common vegetables, had LC50 values above 500 μg/mL and were considered essentially nontoxic. No apparent correlation could be found between toxicity and chemical diversity or a specific class of phytochemicals present. Our findings may serve as a guide for herbal drug and nutraceutical development, especially in prioritizing plants for more detailed safety studies.

Prediction of Human Lethal Doses and Concentrations of MEIC Chemicals from Rodent LD50 Values: An Attempt to Make Some Reparation

The prediction of human toxicities from animal toxicity tests is often poor, and is now discouraged and in some cases banned, especially those involving the LD₅₀ test. However, there is a vast number of historical LD₅₀ data in both public and in-house repositories that are being put to little use. This study examined the correlations between human lethality (doses and concentrations) of 36 MEIC chemicals and the median values of a large number of mouse and rat LD₅₀ values obtained for four different routes of administration. The best correlations were found with mouse and rat intraperitoneal LD₅₀ values (r² = 0.838 and 0.810 for human lethal dose, and r² = 0.753 and 0.785 for human lethal concentration). The results show that excellent prediction of human lethal dose and concentration can be made, for this series of chemicals at least, by using uncurated rodent LD₅₀ values, thus offering some reparation for the millions of rodent lives sacrificed in LD₅₀ testing.

MEIC Evaluation of Acute Systemic Toxicity

The Multicenter Evaluation of In Vitro Cytotoxicity (MEIC) programme was set up to evaluate the relevance for human acute toxicity of in vitro cytotoxicity tests. At the end of the project in 1996, 29 laboratories had tested all 50 reference chemicals in 61 cytotoxicity assays. Five previous articles have presented the in vitro data and the human database to be used in the evaluation. This article presents three important parts of the final evaluation: a) a comparison of rat and mouse oral LD50 with human acute lethal doses for all 50 chemicals; b) a display of the correlations between IC50 (concentration causing 50% inhibition) values from all 61 assays and three independent sets of human acute lethal blood concentrations, i.e. clinical lethal concentrations, forensic lethal concentrations, and peak concentrations; and c) a series of comparisons between average IC50 values from ten human cell line 24-hour assays and human lethal blood concentrations. In the latter comparisons, results from correlations were linked with known human toxicity data for the chemicals, to provide an understanding of correlative results. This correlative/mechanistic approach had the double purpose of assessing the relevance of the in vitro cytotoxicities, and of testing a series of hypotheses connected with the basal cytotoxicity concept. The results of the studies were as follows. Rat LD50 predictions of human lethal dosage were only relatively good (R2 = 0.61), while mouse LD50s gave a somewhat better prediction (R2 = 0.65). Comparisons performed between IC50 values from the 61 assays and the human lethal peak concentrations demonstrated that human ceil line tests gave the best average results (R2 = 0.64), while mammalian and fish cell tests correlated less well (R2 = 0.52–0.58), followed by non-fish ecotoxicological tests (R2 = 0.36). Most of the 61 assays underpredicted human toxicity for digoxin, malathion, carbon tetrachloride and atropine sulphate. In the correlative/mechanistic study, the 50 chemicals were first separated into three groups: A = fast-acting chemicals with a restricted passage across the blood–brain barrier; B = slow-acting chemicals with a restricted passage across the blood–brain barrier; and C = chemicals which cross the blood–brain barrier freely, while inducing a non-specific excitation/depression of the central nervous system (CNS). The IC50 values for chemicals in group C were divided by a factor of ten to compensate for a hypothetical extra vulnerability of the CNS to cytotoxicity. Finally, the average human cell line IC50 values (24-hour IC50 for groups A and C, and after 48-hour for group B) were compared with relevant human lethal blood concentrations (peak concentrations for groups A and C, and 48-hour concentrations for group B). As a result, in vitro toxicity and in vivo toxicity correlated very well for all groups (R2 = 0.98, 0.82 and 0.85, respectively). No clear overprediction of human toxicity was made by the human cell tests. The human cell line tests underpredicted human toxicity for only four of the 50 chemicals. These outlier chemicals were digoxin, malathion, nicotine and atropine sulphate, all of which have a lethal action in man through interaction with specific target sites not usually found in cell lines. Potassium cyanide has a cellular human lethal action which cannot be measured by standard anaerobic cell lines. The good prediction of the human lethal whole-blood concentration of this chemical was not conclusive, i.e. was probably a “false good correlation”. Another two chemicals in group C resulted in “false good correlations”, i.e. paracetamol and paraquat. The comparisons thus indicated that human cell line cytotoxicities are relevant for the human acute lethal action for 43 of the 50 chemicals. The results strongly support the basal cytotoxicity concept, and further point to the non-specific CNS depression being the obligatory reaction of humans to cytotoxic concentrations of chemicals, provided that the chemicals are able to pass the blood–brain barrier.

The use of freshwater planarians in environmental toxicology studies: Advantages and potential

Regarding the humane use of animals in scientific research, invertebrates are often recommended in toxicological studies. "Freshwater planarians" refers to numerous free-living freshwater members of the Class "Turbellaria" of the phylum Platyhelminthes. This group of invertebrates has received extensive attention from biologists for many years because of their unique biological characteristics, such as the primitive form of the central nervous system and notable capability to regenerate tissues. Using freshwater planarians as test animals in chemical toxicity studies has grown in popularity since the 1960s. Results from various toxicological experiments have collectively suggested that freshwater planarians can serve as not only alternative models for chemical toxicity screenings in laboratories but also as potential bioindicators for the quality of freshwater environments. However, thus far, no standardized battery of tests for conducting toxicological studies that includes freshwater planarians has been proposed. This paper comprehensively reviews the toxicological information obtained from chemically exposed planarians and proposes practical factors for consideration in toxicity experiments with freshwater planarians as test organisms.

Micropollutants in source separated wastewater streams and recovered resources of source separated sanitation

The quality of anaerobic sludge and struvite from black water treatment system, aerobic sludge from grey water treatment system and effluents of both systems was assessed for organic micropollutant content in order to ensure safety when reusing these products. Use of anaerobic black water sludge and struvite as soil amendments is recommended based on the low micropollutant content. Aerobic grey water sludge is recommended for disposal, because of the relatively high micropollutant concentrations, exceeding those in sewage sludge. Effluents of black and grey water treatment systems require post-treatment prior to reuse, because the measured micropollutant concentrations in the effluents are above ecotoxicological thresholds.

Fate of pharmaceuticals in full-scale source separated sanitation system

Removal of 14 pharmaceuticals and 3 of their transformation products was studied in a full-scale source separated sanitation system with separate collection and treatment of black water and grey water. Black water is treated in an up-flow anaerobic sludge blanket (UASB) reactor followed by oxygen-limited autotrophic nitrification-denitrification in a rotating biological contactor and struvite precipitation. Grey water is treated in an aerobic activated sludge process. Concentration of 10 pharmaceuticals and 2 transformation products in black water ranged between low μg/l to low mg/l. Additionally, 5 pharmaceuticals were also present in grey water in low μg/l range. Pharmaceutical influent loads were distributed over two streams, i.e. diclofenac was present for 70% in grey water, while the other compounds were predominantly associated to black water. Removal in the UASB reactor fed with black water exceeded 70% for 9 pharmaceuticals out of the 12 detected, with only two pharmaceuticals removed by sorption to sludge. Ibuprofen and the transformation product of naproxen, desmethylnaproxen, were removed in the rotating biological contactor. In contrast, only paracetamol removal exceeded 90% in the grey water treatment system while removal of other 7 pharmaceuticals was below 40% or even negative. The efficiency of pharmaceutical removal in the source separated sanitation system was compared with removal in the conventional sewage treatment plants. Furthermore, effluent concentrations of black water and grey water treatment systems were compared with predicted no-effect concentrations to assess toxicity of the effluent. Concentrations of diclofenac, ibuprofen and oxazepam in both effluents were higher than predicted no-effect concentrations, indicating the necessity of post-treatment. Ciprofloxacin, metoprolol and propranolol were found in UASB sludge in μg/g range, while pharmaceutical concentrations in struvite did not exceed the detection limits.

Methods for assigning confidence to toxicity data with multiple values--Identifying experimental outliers

The assessment of data quality is a crucial element in many disciplines such as predictive toxicology and risk assessment. Currently, the reliability of toxicity data is assessed on the basis of testing information alone (adherence to Good Laboratory Practice (GLP), detailed testing protocols, etc.). Common practice is to take one toxicity data point per compound - usually the one with the apparently highest reliability. All other toxicity data points (for the same experiment and compound) from other sources are neglected. To show the benefits of incorporating the "less reliable" data, a simple, independent, statistical approach to assess data quality and reliability on a mathematical basis was developed. A large data set of toxicity values to Aliivibrio fischeri was assessed. The data set contained 1813 data points for 1227 different compounds, including 203 identified as non-polar narcotic. Log KOW values were calculated and non-polar narcosis quantitative structure-activity relationship (QSAR) models were built. A statistical approach to data quality assessment, which is based on data outlier omission and confidence scoring, improved the linear QSARs. The results indicate that a beneficial method for using large data sets containing multiple data values per compound and highly variable study data has been developed. Furthermore this statistical approach can help to develop novel QSARs and support risk assessment by obtaining more reliable values for biological endpoints.

Risk assessment on the presence of pharmaceuticals in sediments, soils and waters of the Pego-Oliva Marshlands (Valencia, eastern Spain)

This study is focused on the occurrence of 17 pharmaceuticals in waters (34 samples), sediments (16 samples) and soils (23 samples, at two different depths) in a typical Mediterranean coastal wetland (Pego-Oliva marsh, Spain). Soil and sediment samples were extracted by pressurized liquid extraction (PLE). Aqueous extracts from PLE and water samples were concentrated by solid-phase extraction (SPE) and determined by liquid-chromatography tandem mass spectrometry (LC-MS/MS). Pharmaceuticals were detected in concentrations up to 112 ng/L in water samples, up to 15.1 ng/g sediments and up to 8.4 ng/g in soil. In surface waters, ibuprofen and codeine were the compounds more frequently detected (up to 59 ng/L and 63 ng/L, respectively). Ground and tap water samples analyzed were also contaminated with pharmaceuticals. The 94% of sediments and the 80% of agricultural soils were polluted (mostly by carbamazepine and acetaminophen). Diffusion of codeine and fluoroquinolones to deeper soil horizons was observed. Possible relationships between variables were established by Pearson correlations and principal components analysis (PCA). An environmental risk assessment based on the available long-term data was performed. Results showed actual risk for the lowest trophic level, and for fishes, due to the presence of fluoroquinolones and ibuprofen. Nevertheless, the presence of pharmaceuticals in the environment is not limited only to an ecological problem since contamination also affects drinking water, being a potential risk to human health.

In vitro biotransformation of amitriptyline and imipramine with rat hepatic S9 fraction: evaluation of the toxicity with Spirotox and Thamnotoxkit F Tests

Pharmaceutical products, as well as their related metabolites, end up in the aquatic environment after use. Little is known about the effects and the hazard of exposure to drugs for aquatic organisms. This study was designed to assess the ecotoxicity of amitriptyline (AMI), imipramine (IMI), and their metabolites. The tested drugs were very toxic to the protozoan Spirostomum ambiguum and the crustacean Thamnocephalus platyurus with the LC50 values around 1 mg l(-1). Moreover, simple additivity occurs between the drugs and their N-desmethyl metabolites. Tested compounds were incubated with S9 rat hepatocyte fraction at 37 degrees C for 4 hours. Unchanged drugs and metabolites were determined using high-pressure liquid chromatography-photodiode array detector. AMI and IMI were biotransformed almost completely. Three AMI and IMI metabolites were detected: desmethyl-, didesmethyl-, and N-oxide. The toxicity of the deproteinated reaction mixtures (TU) was compared to the toxicity equivalency units (TEU) calculated based on the concentrations of the drugs and their LC50 values. It has been demonstrated that the toxicity of mixture of metabolites to Spirotox and Thamnotoxkit F is higher than the predicted value calculated from the concentrations of the drugs and their N-desmethylated derivatives in the sample. The results indicate that the harmfulness of the drug metabolites should be taken into consideration in the ecotoxicological studies.

Interaction effects of multi-metal solutions (As, Cr, Cu, Ni, Pb and Zn) on life history traits in the rotifer Plationus patulus

While many studies have determined toxicity of single heavy metals, few have addressed the combined effects of multiple elements at different concentrations. In order to estimate how metal mixtures affect population dynamics of freshwater rotifers, we exposed Plationus patulus to environmentally relevant concentrations and combinations of arsenic and five heavy metals (Cr, Cu, Ni, Pb and Zn) for 8 days. The life-history parameters, intrinsic rate of increase (r), relative cumulative reproduction (RCR), mortality ratio, mictic ratio, fertilization ratio, and number of deformed offspring were used as endpoints. Exposures were repeated 4 times and considered as blocks in the experimental design. Differences between treatments were evaluated through ANOVA and LSD mean analysis. The combined presence of As, Cu, Ni, Pb, and Zn in different concentrations produced a variety of changes in the population dynamics of P. patulus. However, only Cr produced a significant element*environment interaction. The presence of Cr leads to significant decreases in the intrinsic rate of increase and increased mortality ratio in multi-metal environments. The population growth (as r) decreased from 0.25 to -0.16 due to Cr (10 microg/L) interaction with the remaining elements, and values 0.16-0.12 were observed in treatments containing high levels (50 microg/L) of Cr. Mortality ratio increased from 0.066 to 0.822 due to Cr (10 and 50 microg/L) interactions with the remaining elements at high concentrations. Responses of this basal consumer to metal mixtures will provide insights into how aquatic food webs and their constituent organisms can be altered by anthropogenic inputs like heavy metal contamination.

Towards a new age of virtual ADME/TOX and multidimensional drug discovery

With the continual pressure to ensure follow-up molecules to billion dollar blockbuster drugs, there is a hurdle in profitability and growth for pharmaceutical companies in the next decades. With each success and failure we increasingly appreciate that a key to the success of synthesized molecules through the research and development process is the possession of drug-like properties. These properties include an adequate bioactivity as well as adequate solubility, an ability to cross critical membranes (intestinal and sometimes blood-brain barrier), reasonable metabolic stability and of course safety in humans. Dependent on the therapeutic area being investigated it might also be desirable to avoid certain enzymes or transporters to circumvent potential drug-drug interactions. It may also be important to limit the induction of these same proteins that can result in further toxicities. We have clearly moved the assessment of in vitro absorption, distribution, metabolism, excretion and toxicity (ADME/TOX) parameters much earlier in the discovery organization than a decade ago with the inclusion of higher throughput systems. We are also now faced with huge amounts of ADME/TOX data for each molecule that need interpretation and also provide a valuable resource for generating predictive computational models for future drug discovery. The present review aims to show what tools exist today for visualizing and modeling ADME/TOX data, what tools need to be developed, and how both the present and future tools are valuable for virtual filtering using ADME/TOX and bioactivity properties in parallel as a viable addition to present practices.

Towards a new age of virtual ADME/TOX and multidimensional drug discovery

With the continual pressure to ensure follow-up molecules to billion dollar blockbuster drugs, there is a hurdle in profitability and growth for pharmaceutical companies in the next decades. With each success and failure we increasingly appreciate that a key to the success of synthesized molecules through the research and development process is the possession of drug-like properties. These properties include an adequate bioactivity as well as adequate solubility, an ability to cross critical membranes (intestinal and sometimes blood-brain barrier), reasonable metabolic stability and of course safety in humans. Dependent on the therapeutic area being investigated it might also be desirable to avoid certain enzymes or transporters to circumvent potential drug-drug interactions. It may also be important to limit the induction of these same proteins that can result in further toxicities. We have clearly moved the assessment of in vitro absorption, distribution, metabolism, excretion and toxicity (ADME/TOX) parameters much earlier in the discovery organization than a decade ago with the inclusion of higher throughput systems. We are also now faced with huge amounts of ADME/TOX data for each molecule that need interpretation and also provide a valuable resource for generating predictive computational models for future drug discovery. The present review aims to show what tools exist today for visualizing and modeling ADME/TOX data, what tools need to be developed, and how both the present and future tools are valuable for virtual filtering using ADME/TOX and bioactivity properties in parallel as a viable addition to present practices.

Evaluation of the Cytotoxic Effects of MEIC Chemicals 31–50 on Primary Culture of Rat Hepatocytes and Hepatic and Non-hepatic Cell Lines

The cytotoxicities of 20 chemicals (numbers 31–50) from the Multicenter Evaluation of In Vitro Cytotoxicity (MEIC) programme were assessed with a primary culture of rat hepatocytes and with two hepatic cell lines (Hep G2 and FaO) and one non-hepatic cell line (3T3). The cytotoxicities of the chemicals were evaluated by using the MTT test after the cells had been exposed to the chemicals for 24 hours. For a better evaluation of results, dose–response curves were mathematically linearised and cytotoxicity was expressed as IC50 values and IC10 values (the concentration causing 50% and 10% loss of cell viability, respectively). We found that all the compounds showed similar acute basal cytotoxicity in all four cellular systems (regardless of whether the cells were, or were not, metabolically competent or were or were not of human origin). When these results were used to predicit human toxicity in terms of a mathematical parameter (prediction error [PE]), we found that all four systems gave similar predictions of human toxicity. The best cytotoxicity parameter included in the PE calculation was the IC50/10, because of an underestimation of human toxicity by in vitro systems. However, when PEs were calculated for rodent toxicity, better results were obtained. Data from the literature obtained by using other experimental models for predicting human toxicity were analysed according to the same criteria. We conclude that cellular systems are better predictive tools for human toxicity than are prokaryotic cells or whole-organism models.

Cytotoxicity of MEIC Chemicals to Rainbow Trout R1 Cell Line and Multivariate Comparison with Ecotoxicity Tests

The cytotoxic effects of the first ten chemicals from the Multicenter Evaluation of In Vitro Cytotoxicity (MEIC) programme on the fibroblast-like R1 cell line, derived from rainbow trout (Oncorhynchus mykiss), were measured by using three endpoints: cell attachment (24-hour exposure, crystal violet protein stain); cell viability (24-hour exposure, neutral red uptake cytotoxicity assay); and cell growth (144-hour exposure, crystal violet protein stain). The results were compared with published MEIC toxicity data from fish and mammalian cells, and from ecotoxicity tests with bacteria, invertebrates and plants by using multivariate statistical techniques. For eight of the ten compounds under consideration, a high degree of correlation in toxicity ranking between the various bioassays was observed, irrespective of the test systems or endpoints utilised. This similarity might be explained by the basal cytotoxicity concept. The 20% dissimilarity of the results, however, indicates the influence of test system-specific or end-point-specific sensitivities.

Modelling of human acute toxicity from physicochemical properties and non-vertebrate acute toxicity of the 38 organic chemicals of the MEIC priority list by PLS regression and neural network

Linear and non-linear modelling of human acute toxicity (as human lethal concentrations; HLCs) of the 38 organic chemicals from the 50 priority compounds of the Multicentre Evaluation of In Vitro Cytotoxicity (MEIC) programme was investigated. The models obtained were derived either from a set of 23 physicochemical properties of the compounds or from their acute toxicities to five aquatic non-vertebrates together with the physicochemical properties. For the linear type, modelling was performed using a partial least square projection to latent structures (PLS) regression method; for the non-linear models, both PLS regression and neural network were utilized. A neural network using a combination of backpropagation and cascade-correlation algorithms was applied in this study. The results generally reveal a slightly better predictive performance of the models obtained from PLS regression than those obtained from neural networks. However, the model composed of physicochemical properties (PC-model) from the trained neural network using a back propagation algorithm with pruning technique proved superior to that trained with a combination of backpropagation and cascade-correlation algorithms after leave-one-out cross-validation. The predictive power of the PC-models, whether linear or non-linear, was comparable with that of the corresponding models consisting of both structural descriptors and the ecotoxicological tests (ECOPC-models), except for the battery (ECOPC-model) from the neural network. The composition of the 'best' PLS and neural network models points to the importance of the combination of physicochemical properties reflecting lipophilicity, size, volume, intermolecular binding forces and electronic properties of the molecule. All the aquatic non-vertebrate tests are shown to be essential in explaining human acute toxicity. However, the degree of contribution differed, with the crustacean (Artemia salina) and the bacterial (Microtox) bioassays being more important to the linear and non-linear PLS models, whereas the crustacean (Artemia salina and Streptocephalus proboscideus) tests, and the rotifer (Brachionus calyciflorus) assay were important to the neural network models. The organochlorine (lindane) and bipyridinium (paraquat) pesticides were common outliers in all the models. Moreover, the latter two compounds and the organophosphate (malathion) pesticide were also common outliers in all ECOPC-models. Other types of pesticides, however, fit the models. The predicted HLCs of a number of non-pesticides, including some chlorinated compounds, also deviated from the observed HLCs by more than one order of magnitude.

QSAR models for predicting the acute toxicity of selected organic chemicals with diverse structures to aquatic non-vertebrates and humans

The linear and non-linear relationships of acute toxicity (as determined on five aquatic non-vertebrates and humans) to molecular structure have been investigated on 38 structurally-diverse chemicals. The compounds selected are the organic chemicals from the 50 priority chemicals prescribed by the Multicentre Evaluation of In Vitro Cytotoxicity (MEIC) programme. The models used for the evaluations are the best combination of physico-chemical properties that could be obtained so far for each organism, using the partial least squares projection to latent structures (PLS) regression method and backpropagated neural networks (BPN). Non-linear models, whether derived from PLS regression or backpropagated neural networks, appear to be better than linear models for describing the relationship between acute toxicity and molecular structure. BPN models, in turn, outperform non-linear models obtained from PLS regression. The predictive power of BPN models for the crustacean test species are better than the model for humans (based on human lethal concentration). The physico-chemical properties found to be important to predict both human acute toxicity and the toxicity to aquatic non-vertebrates are the n-octanol water partition coefficient (Pow) and heat of formation (HF). Aside from the two former properties, the contribution of parameters that reflect size and electronic properties of the molecule to the model is also high, but the type of physico-chemical properties differs from one model to another. In all of the best BPN models, some of the principal component analysis (PCA) scores of the 13C-NMR spectrum, with electron withdrawing/accepting capacity (LUMO, HOMO and IP) are molecular size/volume (VDW or MS1) parameters are relevant. The chemical deviating from the QSAR models include non-pesticides as well as some of the pesticides tested. The latter type of chemical fits in a number of the QSAR models. Outliers for one species may be different from those of other test organisms.