The relationship between developmental toxicity and aromatic-ring class profile of high-boiling petroleum substances

A QSAR study for predicting malformation in zebrafish embryo

BACKGROUND

Developmental toxicity tests are extremely expensive, require a large number of animals, and are time-consuming. It is necessary to develop a new approach to simplify the analysis of developmental endpoints. One of these endpoints is malformation, and one group of ongoing methods for simplifying is in silico models. In this study, we aim to develop a quantitative structure-activity relationship (QSAR) model and identify the best algorithm for predicting malformations, as well as the most important and effective physicochemical properties associated with malformation.

METHODS

The dataset was extracted from a reliable database called COMPTOX. Physicochemical properties (descriptors) were calculated using Mordred and RDKit chemoinformatics software. The data were cleaned, preprocessed, and then split into training and testing sets. Machine learning algorithms, such as gradient boosting model (GBM) and logistic regression (LR), as well as deep learning models, including multilayer perceptron (MLP) and neural networks (NNs) trained with train set data and different sets of descriptors. The models were then validated with test set and various statistical parameters, such as Matthew's correlation coefficient (MCC) and balanced accuracy (BAC) score, were used to compare the models.

RESULTS

A set of descriptors containing with 78% AUC was identified as the best set of descriptors. Gradient boosting was determined to be the best algorithm with 78% predictive power.

CONCLUSIONS

The descriptors that were the most effective for developing models directly impact the mechanism of malformation, and GBM is the best model due to its MCC and BAC.

Evaluating the in vitro developmental toxicity potency of a series of petroleum substance extracts using new approach methodologies (NAMs)

The present study evaluates the in vitro developmental toxicity and the possible underlying mode of action of DMSO extracts of a series of highly complex petroleum substances in the mouse embryonic stem cell test (mEST), the zebrafish embryotoxicity test (ZET) and the aryl hydrocarbon receptor reporter gene assay (AhR CALUX assay). Results show that two out of sixteen samples tested, both being poorly refined products that may contain a substantial amount of 3- to 7-ring polycyclic aromatic compounds (PACs), induced sustained AhR activation in the AhR CALUX assay, and concentration-dependent developmental toxicity in both mEST and ZET. The other samples tested, representing highly refined petroleum substances and petroleum-derived waxes (containing typically a very low amount or no PACs at all), were negative in all assays applied, pointing to their inability to induce developmental toxicity in vitro. The refining processes applied during the production of highly refined petroleum products, such as solvent extraction and hydrotreatment which focus on the removal of undesired constituents, including 3- to 7-ring PACs, abolish the in vitro developmental toxicity. In conclusion, the obtained results support the hypothesis that 3- to 7-ring PACs are the primary inducers of the developmental toxicity induced by some (i.e., poorly refined) petroleum substances and that the observed effect is partially AhR-mediated.

Integrative Chemical-Biological Grouping of Complex High Production Volume Substances from Lower Olefin Manufacturing Streams

Human cell-based test methods can be used to evaluate potential hazards of mixtures and products of petroleum refining ("unknown or variable composition, complex reaction products, or biological materials" substances, UVCBs). Analyses of bioactivity and detailed chemical characterization of petroleum UVCBs were used separately for grouping these substances; a combination of the approaches has not been undertaken. Therefore, we used a case example of representative high production volume categories of petroleum UVCBs, 25 lower olefin substances from low benzene naphtha and resin oils categories, to determine whether existing manufacturing-based category grouping can be supported. We collected two types of data: nontarget ion mobility spectrometry-mass spectrometry of both neat substances and their organic extracts and in vitro bioactivity of the organic extracts in five human cell types: umbilical vein endothelial cells and induced pluripotent stem cell-derived hepatocytes, endothelial cells, neurons, and cardiomyocytes. We found that while similarity in composition and bioactivity can be observed for some substances, existing categories are largely heterogeneous. Strong relationships between composition and bioactivity were observed, and individual constituents that determine these associations were identified. Overall, this study showed a promising approach that combines chemical composition and bioactivity data to better characterize the variability within manufacturing categories of petroleum UVCBs.

Comparison of PAC and MOAH for understanding the carcinogenic and developmental toxicity potential of mineral oils

The carcinogenicity and developmental toxicity of unrefined mineral oil is related to its 3-7 ring polycyclic aromatic compounds (PAC) content. Therefore, refining operations focus on the targeted removal PAC from mineral oil that may contain aromatics of low toxicological concern. There are thus, two types of aromatic substances in mineral oil: hazardous and non-hazardous. The first type consists of 3-7 ring PAC which may be naked (unsubstituted) or lowly alkylated. The second type or non-hazardous consists of 1-7 ring aromatics with high degree of alkylation or lack of bay or fjord regions. Although these are toxicologically different, they may both elute in the same fraction when using chromatography. To understand how these two aromatic types are related we have assessed the entire mineral oil refinement process by measuring total mineral oil aromatic hydrocarbons (MOAH) content by chromatography next to regulatory hazard tests which focus on 3-7 ring PAC. MOAH content is positively correlated to its molecular weight resulting in aromatic content bias for high viscosity substances. Hazard to 3-7 ring PAC is best controlled by the validated IP346 or modified Ames test. We explain the concept of high vs low alkylation by shortly reviewing new data on alkylated PAC.

Developmental toxicity testing of the fume condensate extracts of bitumen and oxidized asphalt in a series of in vitro alternative assays

The potential developmental toxicity and mode-of-action of fume condensate extracts of bitumen and oxidized asphalt were evaluated in the aryl hydrocarbon receptor (AhR) CALUX assay, the zebrafish embryotoxicity test (ZET), and the mouse embryonic stem cell test (mEST). In the AhR CALUX assay, both fume condensate extracts showed a concentration-dependent AhR induction following 6-h of exposure, but this activity was substantially reduced after 24-h, indicating a transient AhR activation. The main effect observed in the ZET was early embryonic lethality that occurred mostly in the 24 h-post-fertilization (hpf). This typically reflects non-specific toxicity rather than in vitro developmental toxicity of the fume condensate extracts tested since this effect was not seen as a result of the whole cumulative exposure period in the ZET (up to 96 hpf). No malformations were seen in any zebrafish embryo exposed to these fume condensate extracts, although some developed pericardial and/or yolk-sac edemas. Furthermore, both fume condensate extracts tested negative in the mEST. In conclusion, the results show that fume condensate extracts of bitumen and oxidized asphalt do not induce any in vitro developmental toxicity, which is in line with the results observed in the in vivo prenatal developmental toxicity studies performed with the same materials.

Prenatal developmental toxicity studies on fumes from oxidised asphalt (OA) in the rat

The prenatal developmental toxicity of the fumes of oxidised asphalt (OA) was tested by nose-only inhalation in the rat. The test material was generated by collecting fumes from the headspace of storage tanks filled with OA. The composition of these fumes was matched to fumes sampled at a workplace where the same OA was applied in a pour-and-roll operation, representing occupational exposure with high concentrations of fumes to not underestimate the possible hazard. In the main study, dams were exposed to 0, 53, 158 and 536 mg/m³ of fume (as total organic mass), for 6 h/day for 19 days p.c. The maternal NOAEC was 53 mg/m³ (lowest dose tested). In the high-dose group treatment-related effects on body weight gain were seen. In the mid- and high-dose groups treatment-related effects on food consumption, lung weights, and histopathological changes in lungs and the upper respiratory tract were observed. The NOAEC for prenatal developmental toxicity was 536 mg/m³ since no exposure-related effects were found in any of the exposure groups for any of the investigated reproductive endpoints. Furthermore, nose-only exposure to OA fumes in concentrations up to 536 mg/m³ from days 1-19 p.c. did not induce any significant fetal abnormalities.

Prenatal developmental toxicity studies on fumes from bitumen in the rat

The prenatal developmental toxicity of bitumen fume was tested by nose-only inhalation in the rat. The fumes for exposure were collected from the headspace of a storage tank filled with a bitumen corresponding in composition to an anticipated worst-case occupational exposure. The composition of these fumes was compared to actual paving site fumes to ensure its representativeness for workplace exposures. In a dose-range-finding study male and female rats were exposed to 0, 103, 480 or 1043 mg/m3 of fume (as total organic mass), for 6 h/day during 20 days post conception (p.c.). Dose-related effects on body weight and lungs were observed in the mid- and high-dose groups. In the main study, dams were exposed to 0, 52, 151 and 482 mg/m3 of fume, for 6 h/day during 19 days p.c. The maternal NOAEL was 52 mg/m³. In the high-dose group treatment-related effects on body weight (gain), food consumption, lung weights, and histopathological changes in lungs and larynx were observed. In the mid-dose group only histopathological changes in the larynx and lungs were found. The NOAEL for prenatal developmental toxicity was 151 mg/m³ based on reduced fetal weight in the high-dose group (482 mg/m³). However, these changes are most likely a consequence of the maternal toxicity, in particular the reduction of maternal body weight gain by 26 % as compared to control. Nose-only exposure to bitumen fumes in concentrations up to 482 mg/m³ from days 1-19 p.c. did not induce any significant fetal anomalies.

Grouping of UVCB substances with new approach methodologies (NAMs) data

One of the most challenging areas in regulatory science is assessment of the substances known as UVCB (unknown or variable composition, complex reaction products and biological materials). Because the inherent complexity and variability of UVCBs present considerable challenges for establishing sufficient substance similarity based on chemical characteristics or other data, we hypothesized that new approach methodologies (NAMs), including in vitro test-derived biological activity signatures to characterize substance similarity, could be used to support grouping of UVCBs. We tested 141 petroleum substances as representative UVCBs in a compendium of 15 human cell types representing a variety of tissues. Petroleum substances were assayed in dilution series to derive point of departure estimates for each cell type and phenotype. Extensive quality control measures were taken to ensure that only high-confidence in vitro data were used to determine whether current groupings of these petroleum substances, based largely on the manufacturing process and physico-chemical properties, are justifiable. We found that bioactivity data-based groupings of petroleum substances were generally consistent with the manufacturing class-based categories. We also showed that these data, especially bioactivity from human induced pluripotent stem cell (iPSC)-derived and primary cells, can be used to rank substances in a manner highly concordant with their expected in vivo hazard potential based on their chemical compositional profile. Overall, this study demonstrates that NAMs can be used to inform groupings of UVCBs, to assist in identification of representative substances in each group for testing when needed, and to fill data gaps by read-across.

How similar is similar enough? A sufficient similarity case study with Ginkgo biloba extract

Botanical dietary supplements are complex mixtures that can be highly variable in composition and quality, making safety evaluation difficult. A key challenge is determining how diverse products in the marketplace relate to chemically and toxicologically characterized reference samples (i.e., how similar must a product be in order to be well-represented by the tested reference sample?). Ginkgo biloba extract (GBE) was used as a case study to develop and evaluate approaches for determining sufficient similarity. Multiple GBE extracts were evaluated for chemical and biological-response similarity. Chemical similarity was assessed using untargeted and targeted chemistry approaches. Biological similarity was evaluated using in vitro liver models and short-term rodent studies. Statistical and data visualization methods were then used to make decisions about the similarity of products to the reference sample. A majority of the 26 GBE samples tested (62%) were consistently determined to be sufficiently similar to the reference sample, while 27% were different from the reference GBE, and 12% were either similar or different depending on the method used. This case study demonstrated that approaches to evaluate sufficient similarity allow for critical evaluation of complex mixtures so that safety data from the tested reference can be applied to untested materials.

Grouping of Petroleum Substances as Example UVCBs by Ion Mobility-Mass Spectrometry to Enable Chemical Composition-Based Read-Across

Substances of Unknown or Variable composition, Complex reaction products, and Biological materials (UVCBs), including many refined petroleum products, present a major challenge in regulatory submissions under the EU Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) and US High Production Volume regulatory regimes. The inherent complexity of these substances, as well as variability in composition obfuscates detailed chemical characterization of each individual substance and their grouping for human and environmental health evaluation through read-across. In this study, we applied ion mobility mass spectrometry in conjunction with cheminformatics-based data integration and visualization to derive substance-specific signatures based on the distribution and abundance of various heteroatom classes. We used petroleum substances from four petroleum substance manufacturing streams and evaluated their chemical composition similarity based on high-dimensional substance-specific quantitative parameters including m/z distribution, drift time, carbon number range, and associated double bond equivalents and hydrogen-to-carbon ratios. Data integration and visualization revealed group-specific similarities for petroleum substances. Observed differences within a product group were indicative of batch- or manufacturer-dependent variation. We demonstrate how high-resolution analytical chemistry approaches can be used effectively to support categorization of UVCBs based on their heteroatom composition and how such data can be used in regulatory decision-making.

A chemical-biological similarity-based grouping of complex substances as a prototype approach for evaluating chemical alternatives

Comparative assessment of potential human health impacts is a critical step in evaluating both chemical alternatives and existing products on the market. Most alternatives assessments are conducted on a chemical-by-chemical basis and it is seldom acknowledged that humans are exposed to complex products, not individual substances. Indeed, substances of Unknown or Variable composition, Complex reaction products, and Biological materials (UVCBs) are ubiquitous in commerce yet they present a major challenge for registration and health assessments. Here, we present a comprehensive experimental and computational approach to categorize UVCBs according to global similarities in their bioactivity using a suite of in vitro models. We used petroleum substances, an important group of UVCBs which are grouped for regulatory approval and read-across primarily on physico-chemical properties and the manufacturing process, and only partially based on toxicity data, as a case study. We exposed induced pluripotent stem cell-derived cardiomyocytes and hepatocytes to DMSO-soluble extracts of 21 petroleum substances from five product groups. Concentration-response data from high-content imaging in cardiomyocytes and hepatocytes, as well as targeted high-throughput transcriptomic analysis of the hepatocytes, revealed distinct groups of petroleum substances. Data integration showed that bioactivity profiling affords clustering of petroleum substances in a manner similar to the manufacturing process-based categories. Moreover, we observed a high degree of correlation between bioactivity profiles and physico-chemical properties, as well as improved groupings when chemical and biological data were combined. Altogether, we demonstrate how novel in vitro screening approaches can be effectively utilized in combination with physico-chemical characteristics to group complex substances and enable read-across. This approach allows for rapid and scientifically-informed evaluation of health impacts of both existing substances and their chemical alternatives.

Subchronic and Developmental Toxicity of Aromatic Extracts

Aromatic extracts (AEs; distillate AEs [DAEs] and residual AEs [RAEs]) are complex, highly viscous liquid petroleum streams with variable compositions derived by extraction of aromatic compounds from distillate and residual petroleum fractions from a vacuum distillation tower, respectively. The DAEs generally contain significant amounts of polycyclic aromatic compounds (PACs) and are carcinogenic. The RAEs typically contain lower concentrations of biologically active PACs. The PACs in refinery streams can cause effects in repeated-dose and developmental toxicity studies. In a 13-week dermal study, light paraffinic DAE had several dose-related effects involving multiple organs; no-observed-effect level was <5 mg/kg/d, with no overt toxicity. Predicted dose-responses at 10% (PDR10s), modeled doses causing a 10% effect on sensitive end points based on PAC content, ranged from 25 to 78 mg/kg/d for untested paraffinic DAEs. The no observed adverse effect level (NOAEL) for developmental toxicity for light paraffinic DAE was 5 mg/kg/d. Statistically significant developmental effects at higher doses were associated with maternal effects. The PDR10s for developmental toxicity of paraffinic DAEs ranged from 7 to >2000 mg/kg/d, reflecting differences due to variation in PACs. The NOAELs for RAEs were 500 mg/kg for 90-day studies and 2000 mg/kg for developmental toxicity. Reproductive toxicity is not considered to be a sensitive end point for AEs based on the toxicity tests with DAEs, RAEs, and other PAC-containing petroleum substances. In vivo micronucleus tests on heavy paraffinic DAE, RAEs, and a range of other petroleum substances have been negative. The exception to this general trend was a marginally positive response with light paraffinic DAE. Most DAEs are considered unlikely to produce chromosomal effects in vivo.

Acute, Subchronic, and Developmental Toxicological Properties of Lubricating Oil Base Stocks

Lubricating oil base stocks (LOBs) are substances used in the manufacture of finished lubricants and greases. They are produced from residue remaining after atmospheric distillation of crude oil that is subsequently fractionated by vacuum distillation and additional refining steps. Initial LOB streams that have been produced by vacuum distillation but not further refined may contain polycyclic aromatic compounds (PACs) and may present carcinogenic hazards. In modern refineries, LOBs are further refined by multistep processes including solvent extraction and/or hydrogen treatment to reduce the levels of PACs and other undesirable constituents. Thus, mildly (insufficiently) refined LOBs are potentially more hazardous than more severely (sufficiently) refined LOBs. This article discusses the evaluation of LOBs using statistical models based on content of PACs; these models indicate that insufficiently refined LOBs (potentially carcinogenic LOBs) can also produce systemic and developmental effects with repeated dermal exposure. Experimental data were also obtained in ten 13-week dermal studies in rats, eight 4-week dermal studies in rabbits, and seven dermal developmental toxicity studies with sufficiently refined LOBs (noncarcinogenic and commonly marketed) in which no observed adverse effect levels for systemic toxicity and developmental toxicity were 1000 to 2000 mg/kg/d with dermal exposures, typically the highest dose tested. Results in both oral and inhalation developmental toxicity studies were similar. This absence of toxicologically relevant findings was consistent with lower PAC content of sufficiently refined LOBs. Based on data on reproductive organs with repeated dosing and parameters in developmental toxicity studies, sufficiently refined LOBs are likely to have little, if any, effect on reproductive parameters.

The mammalian toxicological hazards of petroleum-derived substances: an overview of the petroleum industry response to the high production volume challenge program

Petroleum-derived substances are complex and composed of aliphatic (normal-, iso-, and cycloparaffins), olefinic, and/or aromatic constituents. Approximately 400 of these complex substances were evaluated as part of the US Environmental Protection Agency voluntary High Production Volume (HPV) Challenge program. The substances were separated into 13 groups (categories), and all available data were assessed. Toxicology testing was conducted as necessary to fully address the end points encompassed by the HPV initiative. In a broad sense, volatile hydrocarbons may cause acute central nervous system effects, and those that are liquids at room temperature pose aspiration hazards if taken into the lungs as liquids and may also cause skin irritation. Higher boiling substances may contain polycyclic aromatic constituents (PACs) that can be mutagenic and carcinogenic and may also cause developmental effects. Substances containing PACs can also cause target organ and developmental effects. The effects of aliphatic constituents include liver enlargement and/or renal effects in male rats via an α-2u-globulin-mediated process and, in some cases, small but statistically significant reductions in hematological parameters. Crude oils may contain other constituents, particularly sulfur- and nitrogen-containing compounds, which are removed during refining. Aside from these more generic considerations, some specific petroleum substances may contain unusually toxic constituents including benzene, 1,3-butadiene, and/or n-hexane, which should also be taken into account if present at toxicologically relevant levels.

The toxicological effects of heavy fuel oil category substances

Heavy fuel oil (HFO) category substances are used to manufacture HFO, a product used in industrial boilers and marine diesel engines. Commercial HFOs and blending stream components are substances of complex and variable composition, composed of C20 to >C50 hydrocarbons, although lower molecular weight material may be added to reduce viscosity and improve flow characteristics. An HFO blending stream (catalytically cracked clarified oil [CCCO]) was tested for target organ and developmental toxicity in rats following repeated dermal administration at doses of 5, 25, or 50 mg/kg/d. In the repeated dose study, there was evidence of increased liver weights, reduced thymus weights, and reductions in hematological parameters with an overall no observed adverse effect level (NOAEL) of 5 mg/kg/d. In the developmental toxicity test, there were significant reductions in fetal survival, significant increases in resorption frequency, and significantly reduced fetal weights with an overall NOAEL of 5 mg/kg/d. These target organ and developmental effects are associated with the types and levels of aromatic constituents in these substances. Among HFO blending streams, CCCOs have the highest levels of aromatics and, because they produce the characteristic toxicological effects at the lowest levels, are considered as "reasonable worst-case examples" for this group of substances. Other HFO category members with lower levels of aromatics produce similar effects but have higher NOAELs. The potential for target organ and developmental effects of other HFO category members can be predicted from information on the types and levels of the aromatic constituents present in these substances.

Characterization of the Noncancer Hazards of Gas Oils

Gas oils, used to manufacture diesel fuel and residential heating oil, are complex hydrocarbon substances with carbon numbers of C9-C30 and boiling ranges of approximately 150°C to 450°C. Target organ (liver enlargement, reduced thymus weights, and reductions in hematological parameters) and developmental (reduced fetal viability, increased resorption frequency, and reduced fetal weights) effects are associated with aromatic constituents present in some gas oils. Two types of gas oils were tested for repeated-dose and developmental toxicity following repeated dermal administration. A blend of commercial diesel fuels containing 26% aromatics, primarily single-ring compounds, did not cause either target organ or developmental effects at levels up to 600 mg/kg/d. “Cracked” gas oils containing higher levels of aromatic constituents were also tested. Because of limited sample availability, 2 cracked gas oil samples were tested, one for systemic effects and the other for developmental toxicity. The sample tested in the repeated-dose toxicity study (81% aromatics including approximately 10% 3-ring compounds) produced increased liver weights, reduced thymus weights, and reductions in hematological parameters. The overall no observed adverse effect level (NOAEL) was 100 mg/kg/d. The sample tested for developmental toxicity (65% aromatics including approximately 5% 3-ring compounds) resulted in significant reductions in fetal survival, significant increases in resorption frequency, and significant reductions in fetal weights with an overall NOAEL of 100 mg/kg/d. In summary, gas oils may or may not cause target organ and/or developmental effects depending on the levels and types of aromatic constituents that they contain.

Toxicological Assessment of Green Petroleum Coke

Green petroleum coke is primarily inorganic carbon with some entrained volatile hydrocarbon material. As part of the petroleum industry response to the high production volume challenge program, the potential for reproductive effects was assessed in a subchronic toxicity/reproductive toxicity screening test in rats (OECD 421). The repeated-dose portion of the study provided evidence for dust accumulation and inflammatory responses in rats exposed to 100 and 300 mg/m3 but there were no effects at 30 mg/m3. In the reproductive toxicity screen, the frequency of successful matings was reduced in the high exposure group (300 mg/m3) and was not significantly different from control values but was outside the historical experience of the laboratory. The postnatal observations (external macroscopic examination, body weight, and survival) did not indicate any treatment-related differences. Additional tests conducted to assess the potential hazards to aquatic (fish, invertebrates, and algae) and soil dwelling organisms (earthworms and vascular plants) showed few effects at the maximum loading rates of 1000 mg coke/L in aquatic studies and 1000 mg coke/kg soil in terrestrial studies. The only statistically significant finding was an inhibition of algal growth measured as either biomass or growth rate.

The relationship between repeat-dose toxicity and aromatic-ring class profile of high-boiling petroleum substances

A study was undertaken within the context of the U.S. EPA HPV Chemical Challenge Program to (1) characterize relationships between PAC content and repeat-dose toxicities of high-boiling petroleum substances (HBPS) and (2) develop statistical models that could be used to predict the repeat-dose toxicity of similar untested substances. The study evaluated 47 repeat-dose dermal toxicity and 157 chemical compositional studies. The four most sensitive endpoints of repeat-dose toxicity were platelet count, hemoglobin concentration, relative liver weight and thymus weight. Predictive models were developed for the dose-response relationships between the wt.% concentration of each of seven ring classes of aromatic compounds (the "ARC profile") and specific effects, with high correlations (r=0.91-0.94) between the observed and model-predicted data. The development of the mathematical models used to generate the results reported in this study is described by Nicolich et al. (2013). Model-generated dose-response curves permit the prediction of either the effect at a given dose or the dose that causes a given effect. The models generate values that are consistent with other standard measures. The models, using compositional data, can be used for predicting the repeat-dose toxicity of untested HBPS.

Assessing the mammalian toxicity of high-boiling petroleum substances under the rubric of the HPV program

In 1998, the US EPA announced the HPV Challenge Program, a voluntary chemical data collection effort. The Petroleum HPV Testing Group (PHPVTG(1)) volunteered to provide data on approximately 110 high boiling petroleum substances (HBPS), i.e. substances with final boiling points ≥ approximately 650°F (343°C). These HBPS are substances of unknown and variable composition (UVCBs) that are composed of numerous individual constituents. Toxicity studies have shown that some HBPS can produce systemic (repeat-dose) and developmental effects, and some are mutagenic under in vitro conditions. The papers in this supplement show that these effects are related to the profiles of aromatic constituents in these substances. Further, it is shown that the effects on selected repeat-dose and developmental toxicity endpoints and mutagenic activity in bacterial assays can be predicted from compositional information using models based on the aromatic-ring class profile, "ARC profile" as defined by gas chromatographic separation of the DMSO-soluble fraction of the starting materials. This chromatographic method and the predictive models provide an efficient means of characterizing for screening purposes the potential for repeat-dose, developmental effects and bacterial mutagenicity of HBPS and can reduce the number of animal tests that would be required if these tests were conducted on all 110 HBPS.

The development of statistical models to determine the relationship between aromatic-ring class profile and repeat-dose and developmental toxicities of high-boiling petroleum substances

The repeat-dose and developmental toxicities of certain petroleum refinery streams are related to their polycyclic aromatic compound (PAC) content (Feuston et al., 1994). Building on this foundation, and working within the context of the US EPA High Production Volume (HPV) Chemical Challenge Program, we: (1) characterized relationships between PAC content and repeat-dose and developmental toxicities of high boiling petroleum substances (HBPS), and (2) developed statistical models that can be used to predict critical effects of similar untested substances. Data from 39 dermal toxicity studies of HBPS were used to develop statistical models to predict the dose-response relationships between the weight percent concentration of each of their 1-7 aromatic ring classes and 4 repeat-dose and 3 developmental endpoints (absolute thymus weight, hemoglobin count, platelet count, liver to body weight, live fetus count, fetal weight, and percent resorptions). The correlations between the observed and model-predicted values are >0.90. The predictive ability of the models was tested via a series of evaluation or corroboration methods. As is shown in the paper, using only compositional data of untested HBPS, the models can be used to predict the effect at a given dose or the dose that causes an effect of a stipulated magnitude.