1,3-Butadiene-Induced Adenine DNA Adducts Are Genotoxic but Only Weakly Mutagenic When Replicated in Escherichia coli of Various Repair and Replication Backgrounds

Abstract

The adverse effects of the human carcinogen 1,3-butadiene (BD) are believed to be mediated by its DNA-reactive metabolites such as 3,4-epoxybut-1-ene (EB) and 1,2,3,4-diepoxybutane (DEB). The specific DNA adducts responsible for toxic and mutagenic effects of BD, however, have yet to be identified. Recent in vitro polymerase bypass studies of BD-induced adenine (BD-dA) adducts show that DEB-induced N⁶,N⁶-DHB-dA (DHB = 2,3-dihydroxybutan-1,4-diyl) and 1,N⁶-γ-HMHP-dA (HMHP = 2-hydroxy-3-hydroxymethylpropan-1,3-diyl) adducts block replicative DNA polymerases but are bypassed by human polymerases η and κ, leading to point mutations and deletions. In contrast, EB-induced N⁶-HB-dA (HB = 2-hydroxy-3-buten-1-yl) does not block DNA synthesis and is nonmutagenic. In the present study, we employed a newly established in vivo lesion-induced mutagenesis/genotoxicity assay via next-generation sequencing to evaluate the in vivo biological consequences of S-N⁶-HB-dA, R,R-N⁶,N⁶-DHB-dA, S,S-N⁶,N⁶-DHB-dA, and R,S-1,N⁶-γ-HMHP-dA. In addition, the effects of AlkB-mediated direct reversal repair, MutM and MutY catalyzed base excision repair, and DinB translesion synthesis on the BD-dA adducts in bacterial cells were investigated. BD-dA adducts showed the expected inhibition of DNA replication in vivo but were not substantively mutagenic in any of the genetic environments investigated. This result is in contrast with previous in vitro observations and opens the possibility that E. coli repair and bypass systems other than the ones studied here are able to minimize the mutagenic properties of BD-dA adducts.

Figure 1

Metabolic activation of 1,3-butadiene and some adenine adducts it produces. A, 1,3-butadiene (BD) is metabolized by CYP2E1 to produce 3,4-epoxy-1-butene (EB). EB can react with adenine bases on DNA to form the N⁶-HB-dA adduct. It can also undergo further oxidation by CYP2E1 to become 1,2,3,4-diepoxbutene (DEB) or be hydrolyzed to 1,2-dihydroxy-3-butene by epoxide hydrolase (EH). DEB can react with adenine to form a variety of adducts, including N⁶,N⁶-DHB-dA, 1,N⁶-γ-HMHP-dA and 1,N⁶-α-HMHP-dA. Under physiological conditions, 1,N⁶-γ-HMHP-dA and 1,N⁶-α-HMHP-dA can interconvert with the equilibrium shifting towards the latter. All of these adducts have stereoisomers. The asterisks denote stereocenters. Note that the adducts shown here are not an exclusive list of adducts that can be generated by BD. B, the structures of the four stereospecific BD-induced adenine adducts investigated in this current study.

Figure 2

Overall procedure of the in vivo lesion-induced mutagenesis assay with next-generation sequencing. Single-stranded M13 lesion-containing and control genomes, each with a unique lesion barcode (line segments with different shades of green), were mixed together at a known ratio and introduced into E. coli host cells with a specific repair and replication capability. In part one of the study, the host cells had various AlkB, DinB and SOS-induction statuses. In the second part, host cells had different MutM and MutY statuses. Following in vivo replication, sequencing libraries were prepared from the amplified progeny genome. During this step, the progeny DNA from each repair/replication background received another unique barcode, designating the different backgrounds (line segments with different shades of red or blue). N represents the base at the site originally contained the lesion in the progeny DNA. The resulting DNA was pooled and sequenced. Lesion mutagenicity and genotoxicity under the various repair/replication conditions were obtained by analyzing the sequencing data based on the two sets of barcodes.

Figure 3

Lesion bypass efficiency of the four BD-induced adenine adducts. Lower bypass efficiency represents stronger lesion genotoxicity. A, bypass results in cells with different AlkB, DinB and SOS-induction statuses. B, bypass results in cells with different MutM and MutY statuses. Each graphed value corresponds to the mean of three independent biological replicates; error bars represent one SD.

Figure 4

Mutation frequency and specificity of the four BD-induced adenine adducts in cells with different AlkB, DinB and SOS-induction statuses. A, G, T and C represent the possible bases at the lesion site (N in Fig. 2) after in vivo replication and Δ denotes the occurrence of deletions at the lesion site. Each graphed value corresponds to the mean of three independent biological replicates; error bars represent one SD. To better show the low percentage events, the G, T, C and Δ results were zoomed-in and shown in the insets.

Figure 5

Mutation frequency and specificity of the four BD-induced adenine adducts in cells with different MutM and MutY statuses. A, G, T and C represent the possible bases at the lesion site (N in Fig. 2) after in vivo replication and Δ denotes the occurrence of deletions at the lesion site. Each graphed value corresponds to the mean of three independent biological replicates; error bars represent one SD. To better show the low percentage events, the G, T, C and Δ results were zoomed-in and shown in the insets.