Archives

  • 2018-07
  • 2019-04
  • 2019-05
  • 2019-06
  • 2019-07
  • 2019-08
  • 2019-09
  • 2019-10
  • 2019-11
  • 2019-12
  • 2020-01
  • 2020-02
  • 2020-03
  • 2020-04
  • 2020-05
  • 2020-06
  • 2020-07
  • 2020-08
  • 2020-09
  • 2020-10
  • 2020-11
  • 2020-12
  • 2021-01
  • 2021-02
  • 2021-03
  • 2021-04
  • 2021-05
  • 2021-06
  • 2021-07
  • 2021-08
  • 2021-09
  • 2021-10
  • 2021-11
  • 2021-12
  • 2022-01
  • 2022-02
  • 2022-03
  • 2022-04
  • 2022-05
  • 2022-06
  • 2022-07
  • 2022-08
  • 2022-09
  • 2022-10
  • 2022-11
  • 2022-12
  • 2023-01
  • 2023-02
  • 2023-03
  • 2023-04
  • 2023-05
  • 2023-06
  • 2023-07
  • 2023-08
  • 2023-09
  • 2023-10
  • 2023-11
  • 2023-12
  • 2024-01
  • 2024-02
  • 2024-03
  • 2024-04
  • Homologous recombination HR has important roles in the repai

    2019-11-05

    Homologous recombination (HR) has important roles in the repair of stalled or collapsed DNA replication forks, as well as of DNA double-strand breaks. Among the factors required for HR in human HPF are RAD51, BRCA1, and BRCA2 (Davies et al., 2001, Sung and Klein, 2006, Xia et al., 2001). BRCA1 and BRCA2 are required for the efficient association of RAD51 with DNA, as well as stabilization of RAD51 filaments on single-stranded DNA (ssDNA). In yeast, Rad52 largely performs this so-called Rad51 mediator function (Jensen et al., 2010, Mortensen et al., 1996, San Filippo et al., 2006). Yeast cells also possess an additional Rad52-like protein called Rad59, which plays a more limited and specialized role in certain sub-pathways of HR (Bai and Symington, 1996, Pannunzio et al., 2008, Signon et al., 2001). One of these sub-pathways is break-induced replication (BIR), a form of recombination-dependent DNA replication that can repair one-ended DNA breaks that can arise, for example, when a replication fork collapses upon encountering a break in the template DNA. A subclass of BIR includes microhomology-mediated BIR (MMBIR), which can lead to the formation of CNVs without RAD51-mediated strand invasion (Hastings et al., 2009, Ottaviani et al., 2014, Sakofsky et al., 2015). The principal biochemical function of Rad52 family members is to promote the annealing of complementary ssDNA molecules, which probably explains how Rad52/59 can catalyze some HR processes in the absence of Rad51-mediated DNA strand invasion (Gasior et al., 1998, Kagawa et al., 2001, Kagawa et al., 2002, Lao et al., 2008, McIlwraith and West, 2008). It is not clear from analysis of HPF sequence conservation whether human RAD52 is the functional homolog of yeast Rad52 or Rad59. Human RAD52 forms oligomers that have robust ssDNA annealing activity (Kagawa et al., 2001, Kagawa et al., 2002). Small molecule disruption of RAD52 oligomerization impairs some recombination activities of RAD52 (Chandramouly et al., 2015, Sullivan et al., 2016) and is synthetically lethal with BRCA2 deficiency (Chandramouly et al., 2015, Feng et al., 2011). Current models propose that RAD52 operates in a BRCA2-independent HR backup pathway to load RAD51 onto DNA (Lok and Powell, 2012). However, the precise role of RAD52 in human cells remains to be defined. The requirement for POLD3 in MiDAS suggests that the DNA synthesis might proceed via a BIR-like process (Costantino et al., 2014). We therefore investigated how HR impacts the mitotic DNA damage response in human cells following RS. We demonstrate that RAD52, but not RAD51 or BRCA2, is required for MiDAS. Consistent with RAD52 acting at an early step in MiDAS, RAD52 depletion disrupts the timely recruitment of MUS81-EME1 and POLD3 to CFSs in mitosis.
    Results
    Discussion MiDAS does not require RAD51 or BRCA2. Nevertheless, RAD51 and BRCA2 function earlier in the cell cycle to reduce reliance on the MiDAS pathway. This is consistent with the known functions of RAD51 in the re-start of collapsed replication forks from a one-ended DNA break in late S/G2 phase cells (Sirbu et al., 2011, Zellweger et al., 2015). We propose that human RAD52 plays a more specialized role in promoting RAD51-independent DNA repair in mitosis at a time when BRCA2 and RAD51 are apparently excluded from chromatin. This most likely involves the ssDNA annealing activity of RAD52. Consistent with this, a RAD51 binding-defective derivative of RAD52 is still competent for DNA annealing (Kagawa et al., 2001). Our data are consistent with a role for RAD52 in BIR. Conventional BIR in yeast requires both Rad51 and Rad52 for homology-mediated template switching. In contrast, the microhomology-driven branch of BIR, MMBIR, can, under certain circumstances, be Rad51 independent (Anand et al., 2013, Hastings et al., 2009, Ottaviani et al., 2014, Sakofsky et al., 2015). Moreover, in yeast, some Rad52-dependent events involving annealing of regions of homology, such as would occur during BIR (Anand et al., 2013, Hastings et al., 2009, Mott and Symington, 2011), create DNA substrates that can be resolved efficiently by MUS81-EME1 (Gaillard et al., 2003, Osman et al., 2003). Both BIR and MMBIR require Pol32, whereas HR-mediated gene conversion does not (Donnianni and Symington, 2013, Hicks et al., 2010, Lydeard et al., 2007, Lydeard et al., 2010, Mayle et al., 2015, Sakofsky et al., 2015). Based on these considerations, therefore, we propose that MiDAS most likely occurs via a RAD51-independent MMBIR process in human cells. Given that RAD51-independent MMBIR requires much less homology (1–6 nucleotides) for template switching (Hastings et al., 2009, Ottaviani et al., 2014), this process would be optimal during the narrow time window in early mitosis when MiDAS occurs, as it negates the need for extensive DNA end resection and Rad51-driven homology searching. In the context of a collapsed replication fork, where the sister chromatid is in very close proximity, an extensive homology search is unnecessary and could even be undesirable. As illustrated in the model presented in Figure 4H, RAD52-mediated DNA annealing from a collapsed replication fork into regions of micro-homology could fulfill the requirement for a rapid completion of DNA synthesis to take place in mitosis, albeit at the potential cost of increased mutagenesis and CNVs at CFSs.