Archives

  • 2022-09
  • 2022-08
  • 2022-07
  • 2022-06
  • 2022-05
  • 2022-04
  • 2021-03
  • 2020-08
  • 2020-07
  • 2020-03
  • 2019-11
  • 2019-10
  • 2019-09
  • 2019-08
  • 2019-07
  • br A br C br D br B br

    2022-09-01


    A
    C
    D
    B
    E
    (legend on next page)
    damage caused by reactive oxygen species (Viel et al., 2017), and this mechanism could plausibly mediate its manifestation as a consequence of in vitro cell culture.
    Activities of APOBEC-Associated Signatures 2 and 13 SBS2 and SBS13 have been attributed to APOBEC cytidine deaminase activity and are common in human cancer, although the factors responsible for their activation remain mysterious (Alexandrov et al., 2013a; Nik-Zainal et al., 2012a; Roberts et al., 2013). SBS2 and SBS13 show substitutions of Pimozide at TCN trinucleotides, with SBS2 predominantly characterized by C>T and SBS13 by C>G and C>A mutations. The C>T transi-tions may arise by replication of uracils generated by APOBEC cytidine deamination, whereas the C>G, C>A, and potentially additional C>T substitutions may be introduced by error-prone polymerases following uracil excision and generation of abasic sites by uracil-DNA glycosylase (UNG) (Helleday et al., 2014; Roberts and Gordenin, 2014).
    The rates of acquisition of SBS2 and SBS13 during in vitro cul-ture were highly variable between cancer cell lines, between different daughter lineages of the same cancer cell line, and over time in the same daughter lineages (Figure 3; Table S3). Despite the clear presence of these signatures in the stocks of the cervical (SW756 and SiHa) and the breast (AU565) cancer cell lines, there was no discernible continuing activity in their daughter lineages. Of the remaining eight cell lines with promi-nent APOBEC activity in stocks, continuing generation of SBS2 and SBS13 was detectable in three breast cancers (BT-474, MDA-MB-453, HCC38), two B cell lymphomas (BC-1, JSC-1), a squamous carcinoma of the tongue (CAL-27), a lung adenocar-cinoma (NCI-H650), and possibly a cutaneous squamous cell carcinoma (A388).
    To investigate further this variability of APOBEC-associated signature activity, we subjected two breast cancer (MDA-MB-453, BT-474) and two B cell lymphoma (BC-1, JSC-1) cell lines to further rounds of subcloning and the squamous carcinoma of the tongue cell line (CAL-27) to seven serial rounds of subclon-ing over short periods of time (Figure 4; signatures annotation in Figure 3 and Table S3). These experiments confirmed that sub-stantial differences in numbers of APOBEC-associated muta-tions often occur between daughter clones from the same parent. For example, the burden of APOBEC-associated signa-tures acquired across multiple clones (A4a–A4j) derived from a single clone (A3a) from the BC-1 cell line, following 20 days of in vitro culture, varied >100-fold (Figure 4B; Table S3).
    Fluctuation in numbers of APOBEC-associated mutations was also observed between different phases of individual cell line-ages. For example, daughter A1a from the BC-1 cell line ac-quired 3,157 mutations at cytosine bases in 82 days, the majority of which were SBS2 and SBS13 mutations (Figure 4B). During its
    succeeding round of propagation, which was for the same period of time, one of the two granddaughter clones acquired 21,246 of such mutations and the other 5,014. Indeed, in the daughter lineages of some cell lines, mutational activity ap-peared to cease completely before reactivating and then discon-tinuing again (although we cannot exclude the presence of extremely small numbers of mutations). This intermittent tempo-ral pattern of activity was most obvious during the serial propa-gation of CAL-27, in which evidence of APOBEC mutagenesis was clearly seen during one period of lineage A (clone A4a) and was absent from at least three others (Figure 4A). The varia-tion in activities of SBS2 and SBS13 was not obviously associ-ated with differences in cell-proliferation rates (Figure S4A) and was in marked contrast to other signatures acquired in vitro in the same set of clones (Figure 3; Table S3). The results indicate that SBS2 and SBS13 mutations can be generated in short, intense bursts of activity with long intervening periods of silence, a pattern that we have termed ‘‘episodic mutagenesis.’’