Michael B. Black

A Comprehensive Statistical Analysis of Predicting In Vivo Hazard Using High-Throughput In Vitro Screening

Over the past 5 years, increased attention has been focused on using high-throughput in vitro screening for identifying chemical hazards and prioritizing chemicals for additional in vivo testing. The U.S. Environmental Protection Agencys ToxCast program has generated a significant amount of high-throughput screening data allowing a broad-based assessment of the utility of these assays for predicting in vivo responses. In this study, a comprehensive cross-validation model comparison was performed to evaluate the predictive performance of the more than 600 in vitro assays from the ToxCast phase I screening effort across 60 in vivo endpoints using 84 different statistical classification methods. The predictive performance of the in vitro assays was compared and combined with that from chemical structure descriptors. With the exception of chronic in vivo cholinesterase inhibition, the overall predictive power of both the in vitro assays and the chemical descriptors was relatively low. The predictive power of the in vitro assays was not significantly different from that of the chemical descriptors and aggregating the assays based on genes reduced predictive performance. Prefiltering the in vitro assay data outside the cross-validation loop, as done in some previous studies, significantly biased estimates of model performance. The results suggest that the current ToxCast phase I assays and chemicals have limited applicability for predicting in vivo chemical hazards using standard statistical classification methods. However, if viewed as a survey of potential molecular initiating events and interpreted as risk factors for toxicity, the assays may still be useful for chemical prioritization.

IVT-seq reveals extreme bias in RNA sequencing

BackgroundRNA-seq is a powerful technique for identifying and quantifying transcription and splicing events, both known and novel. However, given its recent development and the proliferation of library construction methods, understanding the bias it introduces is incomplete but critical to realizing its value.ResultsWe present a method, in vitro transcription sequencing (IVT-seq), for identifying and assessing the technical biases in RNA-seq library generation and sequencing at scale. We created a pool of over 1,000 in vitro transcribed RNAs from a full-length human cDNA library and sequenced them with polyA and total RNA-seq, the most common protocols. Because each cDNA is full length, and we show in vitro transcription is incredibly processive, each base in each transcript should be equivalently represented. However, with common RNA-seq applications and platforms, we find 50% of transcripts have more than two-fold and 10% have more than 10-fold differences in within-transcript sequence coverage. We also find greater than 6% of transcripts have regions of dramatically unpredictable sequencing coverage between samples, confounding accurate determination of their expression. We use a combination of experimental and computational approaches to show rRNA depletion is responsible for the most significant variability in coverage, and several sequence determinants also strongly influence representation.ConclusionsThese results show the utility of IVT-seq for promoting better understanding of bias introduced by RNA-seq. We find rRNA depletion is responsible for substantial, unappreciated biases in coverage introduced during library preparation. These biases suggest exon-level expression analysis may be inadvisable, and we recommend caution when interpreting RNA-seq results.

Relative Impact of Incorporating Pharmacokinetics on Predicting In Vivo Hazard and Mode of Action From High-Throughput In Vitro Toxicity Assays

The use of high-throughput in vitro assays has been proposed to play a significant role in the future of toxicity testing. In this study, rat hepatic metabolic clearance and plasma protein binding were measured for 59 ToxCast phase I chemicals. Computational in vitro-to-in vivo extrapolation was used to estimate the daily dose in a rat, called the oral equivalent dose, which would result in steady-state in vivo blood concentrations equivalent to the AC 50 or lowest effective concentration (LEC) across more than 600 ToxCast phase I in vitro assays. Statistical classification analysis was performed using either oral equivalent doses or unadjusted AC 50 /LEC values for the in vitro assays to predict the in vivo effects of the 59 chemicals. Adjusting the in vitro assays for pharmacokinetics did not improve the ability to predict in vivo effects as either a discrete (yes or no) response or a low effect level (LEL) on a continuous dose scale. Interestingly, a comparison of the in vitro assay with the lowest oral equivalent dose with the in vivo endpoint with the lowest LEL suggested that the lowest oral equivalent dose may provide a conservative estimate of the point of departure for a chemical in a dose-response assessment. Furthermore, comparing the oral equivalent doses for the in vitro assays with the in vivo dose range that resulted in adverse effects identified more coincident in vitro assays across chemicals than expected by chance, suggesting that the approach may also be used to identify potential molecular initiating events leading to adversity.

Temporal concordance between apical and transcriptional points of departure for chemical risk assessment.

The number of legacy chemicals without toxicity reference values combined with the rate of new chemical development is overwhelming the capacity of the traditional risk assessment paradigm. More efficient approaches are needed to quantitatively estimate chemical risks. In this study, rats were dosed orally with multiple doses of six chemicals for 5 days and 2, 4, and 13 weeks. Target organs were analyzed for traditional histological and organ weight changes and transcriptional changes using microarrays. Histological and organ weight changes in this study and the tumor incidences in the original cancer bioassays were analyzed using benchmark dose (BMD) methods to identify noncancer and cancer points of departure. The dose-response changes in gene expression were also analyzed using BMD methods and the responses grouped based on signaling pathways. A comparison of transcriptional BMD values for the most sensitive pathway with BMD values for the noncancer and cancer apical endpoints showed a high degree of correlation at all time points. When the analysis included data from an earlier study with eight additional chemicals, transcriptional BMD values for the most sensitive pathway were significantly correlated with noncancer (r = 0.827, p = 0.0031) and cancer-related (r = 0.940, p = 0.0002) BMD values at 13 weeks. The average ratio of apical-to-transcriptional BMD values was less than two, suggesting that for the current chemicals, transcriptional perturbation did not occur at significantly lower doses than apical responses. Based on our results, we propose a practical framework for application of transcriptomic data to chemical risk assessment.

Cross-species Comparisons of Transcriptomic Alterations in Human and Rat Primary Hepatocytes Exposed to 2,3,7,8-Tetrachlorodibenzo-p-dioxin

A toxicogenomics approach was used to qualitatively and quantitatively compare the gene expression changes in human and rat primary hepatocytes exposed to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). Hepatocytes from five individual rats and five individual humans were exposed for 24 h to 11 concentrations of TCDD ranging from 0.00001 to 100nM and a vehicle control. Gene expression changes were analyzed using whole-genome microarrays containing 13,002 orthologs. Significant changes in expression of individual orthologs at any concentration (fold change [FC] ± 1.5 and false discovery rate < 0.05) were higher in the rat (1547) compared with human hepatocytes (475). Only 158 differentially expressed orthologs were common between rats and humans. Enrichment analysis was performed on the differentially expressed orthologs in each species with 49 and 34 enriched human and rat pathways, respectively. Only 12 enriched pathways were shared between the two species. The results demonstrate significant cross-species differences in expression at both the gene and pathway level. Benchmark dose analysis of gene expression changes showed an average 18-fold cross-species difference in potency among differentially expressed orthologs with the rat more sensitive than the human. Similar cross-species differences in potency were observed for signaling pathways. Using the maximum FC in gene expression as a measure of efficacy, the human hepatocytes showed on average a 20% lower efficacy among the individual orthologs showing differential expression. The results provide evidence for divergent cross-species gene expression changes in response to TCDD and are consistent with epidemiological and clinical evidence showing humans to be less sensitive to TCDD-induced hepatotoxicity.

Comparison of Microarrays and RNA-Seq for Gene Expression Analyses of Dose-Response Experiments

Relative to microarrays, RNA-seq has been reported to offer higher precision estimates of transcript abundance, a greater dynamic range, and detection of novel transcripts. However, previous comparisons of the 2 technologies have not covered dose-response experiments that are relevant to toxicology. Male F344 rats were exposed for 13 weeks to 5 doses of bromobenzene, and liver gene expression was measured using both microarrays and RNA-seq. Multiple normalization methods were evaluated for each technology, and gene expression changes were statistically analyzed using both analysis of variance and benchmark dose (BMD). Fold-change values were highly correlated between the 2 technologies, whereas the -log p values showed lower correlation. RNA-seq detected fewer statistically significant genes at lower doses, but more significant genes based on fold change except when a negative binomial transformation was applied. Overlap in genes significant by both p value and fold change was approximately 30%-40%. Random sampling of the RNA-seq data showed an equivalent number of differentially expressed genes compared with microarrays at ~5 million reads. Quantitative RT-PCR of differentially expressed genes uniquely identified by each technology showed a high degree of confirmation when both fold change and p value were considered. The mean dose-response expression of each gene was highly correlated between technologies, whereas estimates of sample variability and gene-based BMD values showed lower correlation. Differences in BMD estimates and statistical significance may be due, in part, to differences in the dynamic range of each technology and the degree to which normalization corrects genes at either end of the scale.

Evaluation of gene expression changes in human primary uroepithelial cells following 24-hr exposures to inorganic arsenic and its methylated metabolites.

Gene expression changes in primary human uroepithelial cells exposed to arsenite and its methylated metabolites were evaluated to identify cell signaling pathway perturbations potentially associated with bladder carcinogenicity. Cells were treated with mixtures of inorganic arsenic and its pentavalent or trivalent metabolites for 24 hr at total arsenic concentrations ranging from 0.06 μM to 18 μM. One series (five samples) was conducted with arsenite and pentavalent metabolites and a second (10 samples) with arsenite and trivalent metabolites. Similar gene expression responses were obtained for pentavalent or trivalent metabolites. A suite of eight gene changes was consistently identified across individuals that reflect effects on key signaling pathways: oxidative stress, protein folding, growth regulation, metallothionine regulation, DNA damage sensing, thioredoxin regulation, and immune response. No statistical significance of trend (NOSTASOT) analysis of these common genes identified lowest observed effect levels (LOELs) from 0.6 to 6.0 μM total arsenic and no observed effect levels (NOELs) from 0.18 to 1.8 μM total arsenic. For the trivalent arsenical mixture, benchmark doses (BMDs) ranged from 0.13 to 0.92 μM total arsenic; benchmark dose lower 95% confidence limits (BMDLs) ranged from 0.09 to 0.58 μM total arsenic. BMDs ranged from 0.53 to 2.7 μM and BMDLs from 0.35 to 1.7 μM for the pentavalent arsenical mixture. Both endpoints varied by a factor of 3 across individuals. Thisstudy is the first to examine gene expression response in primary uroepithelial cells from multiple individuals and to identify no effect levels for arsenical‐induced cell signaling perturbations in normal human cells exposed to a biologically plausible concentration range.

Cross-Species Transcriptomic Analysis of Mouse and Rat Lung Exposed to Chloroprene

β-Chloroprene (2-chloro-1,3-butadiene), a monomer used in the production of neoprene elastomers, is of regulatory interest due to the production of multiorgan tumors in mouse and rat cancer bioassays. A significant increase in female mouse lung tumors was observed at the lowest exposure concentration of 12.8 ppm, whereas a small, but not statistically significant increase was observed in female rats only at the highest exposure concentration of 80 ppm. The metabolism of chloroprene results in the generation of reactive epoxides, and the rate of overall chloroprene metabolism is highly species dependent. To identify potential key events in the mode of action of chloroprene lung tumorigenesis, dose-response and time-course gene expression microarray measurements were made in the lungs of female mice and female rats. The gene expression changes were analyzed using both a traditional ANOVA approach followed by pathway enrichment analysis and a pathway-based benchmark dose (BMD) analysis approach. Pathways related to glutathione biosynthesis and metabolism were the primary pathways consistent with cross-species differences in tumor incidence. Transcriptional BMD values for the pathway were more similar to differences in tumor response than were estimated target tissue dose surrogates based on the total amount of chloroprene metabolized per unit mass of lung tissue per day. The closer correspondence of the transcriptional changes with the tumor response is likely due to their reflection of the overall balance between metabolic activation and detoxication reactions, whereas the current tissue dose surrogate reflects only oxidative metabolism.

Combining transcriptomics and PBPK modeling indicates a primary role of hypoxia and altered circadian signaling in dichloromethane carcinogenicity in mouse lung and liver

ABSTRACT Dichloromethane (DCM) is a lung and liver carcinogen in mice at inhalation exposures ≥ 2000 ppm. The modes of action (MOA) of these responses have been attributed to formation of genotoxic, reactive metabolite(s). Here, we examined gene expression in lung and liver from female B6C3F1 mice exposed to 0, 100, 500, 2000, 3000 and 4000 ppm DCM for 90 days. We also simulated dose measures ‐ rates of DCM oxidation to carbon monoxide (CO) in lung and liver and expected blood carboxyhemoglobin (HbCO) time courses with a PBPK model inclusive of both conjugation and oxidation pathways. Expression of large numbers of genes was altered at 100 ppm with maximal changes in the numbers occurring by 500 or 2000 ppm. Most changes in genes common to the two tissues were related to cellular metabolism and circadian clock. At the lower concentrations, the changes in metabolism‐related genes were discordant – up in liver and down in lung. These processes included organelle biogenesis, TCA cycle, and respiratory electron transport. Changes in circadian cycle genes – primarily transcription factors ‐ showed strong concentration‐related response at higher concentrations (Arntl, Npas2, and Clock were down‐regulated; Cry2, Wee1, Bhlhe40, Per3, Nr1d1, Nr1d2 and Dbp) were up‐regulated with similar directionality in both tissues. Overall, persistently elevated HbCO from DCM oxidation appears to cause extended periods of hypoxia, leading to altered circadian coupling to cellular metabolism. The dose response for altered circadian processes correlates with the cancer outcome. We found no evidence of changes in genes indicative of responses to cytotoxic, DNA‐reactive metabolites.

Assessing molecular initiating events (MIEs), key events (KEs) and modulating factors (MFs) for styrene responses in mouse lungs using whole genome gene expression profiling following 1-day and multi-week exposures

ABSTRACT Styrene increased lung tumors in mice at chronic inhalation exposures of 20 ppm and greater. MIEs, KEs and MFs were examined using gene expression in three strains of male mice (the parental C57BL/6 strain, a CYP2F2(−/−) knock out and a CYP2F2(−/−) transgenic containing human CYP2F1, 2A13 and 2B6). Exposures were for 1‐day and 1, 4 and 26 weeks. After 1‐day exposures at 1, 5, 10, 20, 40 and 120 ppm significant increases in differentially expressed genes (DEGs) occurred only in parental strain lungs where there was already an increase in DEGs at 5 ppm and then many thousands of DEGs by 120 ppm. Enrichment for 1‐day and 1‐week exposures included cell cycle, mitotic M‐M/G1 phases, DNA‐synthesis and metabolism of lipids and lipoproteins pathways. The numbers of DEGs decreased steadily over time with no DEGs meeting both statistical significance and fold‐change criteria at 26 weeks. At 4 and 26 weeks, some key transcription factors (TFs) ‐ Nr1d1, Nr1d2, Dbp, Tef, Hlf, Per3, Per2 and Bhlhe40 ‐ were upregulated (|FC| > 1.5), while others ‐ Npas, Arntl, Nfil3, Nr4a1, Nr4a2, and Nr4a3 ‐ were down‐regulated. At all times, consistent changes in gene expression only occurred in the parental strain. Our results support a MIE for styrene of direct mitogenicity from mouse‐specific CYP2F2‐mediated metabolites activating Nr4a signaling. Longer‐term MFs include down‐regulation of Nr4a genes and shifts in both circadian clock TFs and other TFs, linking circadian clock to cellular metabolism. We found no gene expression changes indicative of cytotoxicity or activation of p53‐mediated DNA‐damage pathways. HighlightsStyrene response consistent with direct mitogenicity of Cyp2F2 and Nr4a signalingLonger term exposure show changes in circadian pathways.Changes in circadian pathways associated with Nr4a receptor family down‐regulationConsistent changes were seen only in wild type mice.No evidence of activation of p53‐mediated DNA‐damage or cell stress pathways