Adopting duplex sequencing technology for genetic toxicity testing: A proof-of-concept mutagenesis experiment with N-ethyl-N-nitrosourea (ENU)-exposed rats

Abstract

Duplex sequencing (DS) is an error-corrected next-generation sequencing method in which molecular barcodes informatically link PCR-copies back to their source DNA strands, enabling computational removal of errors in consensus sequences. The resulting background of less than one artifactual mutation per 10⁷ nucleotides allows for direct detection of somatic mutations. TwinStrand Biosciences, Inc. has developed a DS-based mutagenesis assay to sample the rat genome, which can be applied to genetic toxicity testing. To evaluate this assay for early detection of mutagenesis, a time-course study was conducted using male Hsd:Sprague Dawley SD rats (3 per group) administered a single dose of 40 mg/kg N-ethyl-N-nitrosourea (ENU) via gavage, with mutation frequency (MF) and spectrum analyzed in stomach, bone marrow, blood, and liver tissues at 3 h, 24 h, 7 d, and 28 d post-exposure. Significant increases in MF were observed in ENU-exposed rats as early as 24 h for stomach (site of contact) and bone marrow (a highly proliferative tissue) and at 7 d for liver and blood. The canonical, mutational signature of ENU was established by 7 d post-exposure in all four tissues. Interlaboratory analysis of a subset of samples from different tissues and time points demonstrated remarkable reproducibility for both MF and spectrum. These results demonstrate that MF and spectrum can be evaluated successfully by directly sequencing targeted regions of DNA obtained from various tissues⁠, a considerable advancement compared to currently used in vivo gene mutation assays.

Keywords: DuplexSeq™; EcNGS; Error-corrected NGS; Pig-a gene mutation assay; Rat-50 mutagenesis assay; Transgenic rodent assay.

Fig. 1

A – D. Mutation frequency over time for stomach (A), bone marrow (B), blood (C), and liver (D) tissue samples at exposure timepoints 3 h, 24 h, 7 d, and 28 h. Note that vehicle control rat tissues were sampled only at 28 d. Horizontal bar indicates the mean, vertical bars indicate lower and upper 95% confidence intervals. *P < 0.05.

Fig. 2

A – D. Mutation frequency at each target in the DuplexSeq Rat Mutagenesis Assay for stomach (A), bone marrow (B), blood (C), and liver (D) tissue sampled. *Genic targets.

Fig. 3.

Mutation frequency at each target in the DuplexSeq Rat Mutagenesis Assay for each tissue at the 28-d time point. *Genic targets.

Fig. 4.

Unique mutation counts and normalized proportions of the 6 canonical base substitution types for each sample. Note that vehicle control rat tissues were sampled only at 28 d.

Fig. 5

A, B. Comparison of mutation frequency for a subset of samples that underwent DNA extraction, library preparation, and sequencing at either TwinStrand Biosciences, Inc., or ILS, LLC. Error bars indicate lower and upper 95% Wilson confidence intervals. The number of unique mutations detected per sample is noted above each bar; * Indicates a control male rat liver DNA provided in the DuplexSeq Rat Mutagenesis Assay library preparation kit (A). Correlation analysis of TwinStrand versus DTT/ILS mutation frequencies shown in 5A; shaded area represents the 95% confidence interval of the regression line (B).

Fig. 6.

Cosine similarity analysis of trinucleotide sequence spectra obtained by TwinStrand Biosciences, Inc. versus ILS, LLC for the samples shown in Fig. 5A and Fig. 5B.