Critical Analysis of Strand-Biased Somatic Mutation Signatures in TP53 versus Ig Genes, in Genome-Wide Data and the Etiology of Cancer

Figure 1

Explanation for TP53 strand-biased mutations in the “Endogenous Patterns” outlined in the text based on the reverse transcriptase model of Ig SHM, namely, transcription-coupled DNA and RNA deamination followed by reverse transcription, adapted and modified from Figure 5 in Steele, 2009, [3] following Steele and Lindley, 2010, [4]. Shown are some hypothesised DNA and RNA intermediates highlighted for the generation of the main strand-biased mutation signatures involving A-to-G, G-to-A, G-to-T, and G-to-C. Black lines are DNA strands, red lines are mRNA, and blue lines are cDNA strands copied off mRNA by a cellular reverse transcriptase such as DNA polymerase η. Steps (a) through (d) show various mutated DNA and RNA intermediates and substrate complexes for both deamination reactions, 8oxoG modifications in RNA, Wu and Li., 2008 [5] and cDNA synthesis (it is not known if 8oxoG sites are preferred in unpaired loops or dsRNA regions). In overview, mutations are first introduced at the DNA level by AID/APOBEC family-mediated C-to-U deaminations and then uracil DNA glycosylase (UNG)-generated abasic sites in the TS (which can further mature into single strand nicks via the action of AP endonuclease (APE1). These template sites are transcribed into mRNA by RNA Pol II generating G-to-A and G-to-C modifications, respectively, in the mRNA, Kuraoka et al, 2003 [6] which on reverse transcription, integration, and DNA replication result in G-to-A and G-to-C mutations in the NTS. Separately adenosine-to-inosine (A-to-I) RNA editing events at WA targets in the nascent ds mRNA stem loops may be copied back into DNA by reverse transcription via Pol-η, Franklin et al., 2004 [7]. Also shown are 8oxoG modifications in mRNA which on reverse transcription, integration and DNA replication would result in strand-biased G-to-T transversions on the NTS. The strand invasion (?) and integration of newly synthesised cDNA TS (?) are hypothesized necessary steps. In more detail (a) RNA Pol II introduces mutations in mRNA as it copies the AID/APOBEC lesions in TS DNA [6], concurrently A-to-I RNA edited sites appear in RNA stem(-loops) forming in nascent mRNA near the transcription bubble [Steele et al., 2006 [8] or 8oxoG modifications via reactive oxygen species [5]. (b) Formation of RT-priming substrate (DNA polymerase-η) by annealing of nicked TS strand with an exposed 3′-OH end. This could arise due to excision at a previous AID-mediated abasic site or an excision introduced by endonuclease activity associated with the MSH2-MSH6 heterodimer engaging a U:G mispaired lesion. (c) Extension of new TS by cDNA synthesis from the 3′-OH end copying the already base modified mRNA template (with I base pairing preferentially, like G, with C; and 8oxoG mispairing with A). (d) Then an unknown and indeterminant number of steps involving strand invasion (?), heteroduplex formation and/or resolution of heteroduplex (?), full length copying of newly synthesized transcribed strand (?) cDNA.