Cellular and molecular consequences of defective Fanconi anemia proteins in replication-coupled DNA repair: mechanistic insights

Abstract

The Fanconi anemia (FA) molecular network consists of 15 "FANC" proteins, of which 13 are associated with mutations in patients with this cancer-prone chromosome instability disorder. Whereas historically the common phenotype associated with FA mutations is marked sensitivity to DNA interstrand crosslinking agents, the literature supports a more global role for FANC proteins in coping with diverse stresses encountered by replicative polymerases. We have attempted to reconcile and integrate numerous observations into a model in which FANC proteins coordinate the following physiological events during DNA crosslink repair: (a) activating a FANCM-ATR-dependent S-phase checkpoint, (b) mediating enzymatic replication-fork breakage and crosslink unhooking, (c) filling the resulting gap by translesion synthesis (TLS) by error-prone polymerase(s), and (d) restoring the resulting one-ended double-strand break by homologous recombination repair (HRR). The FANC core subcomplex (FANCA, B, C, E, F, G, L, FAAP100) promotes TLS for both crosslink and non-crosslink damage such as spontaneous oxidative base damage, UV-C photoproducts, and alkylated bases. TLS likely helps prevent stalled replication forks from breaking, thereby maintaining chromosome continuity. Diverse DNA damages and replication inhibitors result in monoubiquitination of the FANCD2-FANCI complex by the FANCL ubiquitin ligase activity of the core subcomplex upon its recruitment to chromatin by the FANCM-FAAP24 heterodimeric translocase. We speculate that this translocase activity acts as the primary damage sensor and helps remodel blocked replication forks to facilitate checkpoint activation and repair. Monoubiquitination of FANCD2-FANCI is needed for promoting HRR, in which the FANCD1/BRCA2 and FANCN/PALB2 proteins act at an early step. We conclude that the core subcomplex is required for both TLS and HRR occurring separately for non-crosslink damages and for both events during crosslink repair. The FANCJ/BRIP1/BACH1 helicase functions in association with BRCA1 and may remove structural barriers to replication, such as guanine quadruplex structures, and/or assist in crosslink unhooking.

Fig. 1

Heuristic model showing participation of the FANC proteins in the two components of ICL repair that occurs at blocked replication forks. Proteins written in bold font are known to be essential for the viability of proliferating vertebrate cells: ATR [268], Chk1 [269], Mre11 [270], RAD50 [271], BRCA1 [272], RPA [273], FANCD1/BRCA2 [274], and Rad51 [275]. (DT40 nbs1 null cells are viable but grow slowly [276]; in the mouse only nbs1 truncation mutants are viable [277].) Events may not occur exactly in the order shown. (1) ICLs can arise endogenously such as from the metabolism of ethanol to acetaldehyde [278] and are produced by many cancer chemotherapeutic agents. (2) The blocked replication fork with a possible requirement for RPA [279] activates the ATR kinase, which directly phosphorylates FANCG [214], the FANCI-FANCD2 complex, and Chk1 [19,183]; Chk1 subsequently phosphorylates FANCE [188]. Reversal of the blocked fork is proposed, based on the properties of the FANCM-FAAP24 translocase and BLM as discussed in the text. (3) FANCM-FAAP24 together with the FANC core subcomplex (A,B,C,E,F,G,L,FAAP100) (see Introduction) mediate the monoubiquitination of the FANCI-FANCD2 complex through the ubiquitin ligase activity of FANCL [15,17], an event that is necessary for replication-associated HRR to occur efficiently. The recently described Hes1 protein shares properties with core complex proteins and is also necessary for FANCI-FANCD2 monoubiquitination [219]. The chromatin remodeling protein Tip60 (a histone acetyltransferase) interacts with FANCD2, is epistatic with FANCC for MMC sensitivity, and is not required for FANCD2 monoubiquitination or focus formation [280]. (4) Repair of the ICL requires endonucleolytic cleavage on the 3′ side by the Mus81-Eme1 (and possibly Mus81-Eme2) complex [226,235]. (5) ICL release is enabled by endonucleolytic incision by XPF-ERCC1 on the 5′ side of the ICL, an event that is speculatively mediated by the helicase activity of FANCJ (also known as BRIP1 and BACH1). (6) The resulting gap provides a substrate for error-prone polymerases such as Pol ζ (Rev3-Rev7 complex) and the tightly associated Rev 1 [281]. The FANC core subcomplex is required for this mutagenic step and for Rev1 focus formation [51]. DNA synthesis likely produces a base substitution mutation (red circle). (7) The one-ended DSB is processed in preparation for HRR, an event that is thought to require the nuclease activity of the Mre11-Rad50-Nbs1 (MRN) complex. The NER machinery can excise the monoadduct when it becomes physically accessible. The FANCG^S7P putative ”replication restart” complex (FANCD1-FANCD2-FANCG-XRCC3) assembles to help initiate HRR [214]. (8) Rad51 nucleoprotein filament formation requires BRCA2 and PALB2 (FANCD1 and FANCN, respectively) and is followed by homologous pairing. (9) Strand exchange results in the formation of a D-loop, a substrate that can prime DNA synthesis, which perpetuates the mutation shown by the red circle. Since this synthesis may well be performed by a TLS polymerase (see text), another base substitution mutation (red hexagon) can arise with a much higher probability (i.e. 0.1–3%) than for a normal replicative polymerase (~10⁻⁹) [49,50]. (10) The crossover structure is processed by the nuclease activity of a resolvase, which results in a SCE if the crossover strands are incised as shown by the green arrows. Incision at the orange arrow would not cause SCE. (11–12) The replication fork is reestablished and replicative synthesis resumes.

Fig. 2

Models depicting the involvement of FANC proteins in TLS, resolving a structural barrier, and HRR. A. (1) A replication fork encounters a blocking lesion such as a chemically modified base other than a DNA crosslink (red trapezoid). (2) The FANC core complex recruits TLS polymerases. (3) The lesion is bypassed with likelihood of a base substitution (red circle). Then the replication fork independently encounters a structural barrier, such as a telomeric G4 or other secondary structural motif (red cross). (4) FANCJ helicase activity, acting alone or in concert with the core complex, alleviates this obstacle. B. (1) A replication fork has collapsed from enzymatic action [258] or spontaneous breakage as when encountering a nick or gap. (2) FANCM-FAAP24 senses the damage, and the active core complex monoubiquitinates FANCD2, which becomes associated with chromatin. (3) A putative “replication restart” FANC-HRR transition complex (same as shown in Fig. 1.7) containing FANCD1, FANCD2, FANCG, and XRCC3 forms [214]. We speculate this step is promoted by the deubiquitination of FANCD2-FANCI by Usp1 [282,283]. (4) HRR proceeds.