A distinct replication fork protection pathway connects Fanconi anemia tumor suppressors to RAD51-BRCA1/2

Abstract

Genes mutated in patients with Fanconi anemia (FA) interact with the DNA repair genes BRCA1 and BRCA2/FANCD1 to suppress tumorigenesis, but the molecular functions ascribed to them cannot fully explain all of their cellular roles. Here, we show a repair-independent requirement for FA genes, including FANCD2, and BRCA1 in protecting stalled replication forks from degradation. Fork protection is surprisingly rescued in FANCD2-deficient cells by elevated RAD51 levels or stabilized RAD51 filaments. Moreover, FANCD2-mediated fork protection is epistatic with RAD51 functions, revealing an unanticipated fork protection pathway that connects FA genes to RAD51 and the BRCA1/2 breast cancer suppressors. Collective results imply a unified molecular mechanism for repair-independent functions of FA, RAD51, and BRCA1/2 proteins in preventing genomic instability and suppressing tumorigenesis.

Figure 1. FA/BRCA Gene Network Protects Stalled DNA Replication Forks

(A) A graphical representation of the FA/BRCA gene network depicts FA core complex proteins (FANCA, B, C, E, F, G, L, M), which promote the monoubiquitination (Ub) of FANCD2 and FANCI. BRCA-related proteins (FANCD1/BRCA2, FANCN/PALB2, and FANCJ/BRIP1) and recently identified FANCO/RAD51C and FANCP/SLX4 are not required for FANCD2–FANCI monoubiquitination and act downstream or in parallel to canonical FA proteins. While BRCA1 is part of the FA/BRCA gene network, BRCA1 mutations have not been found in FA patients. BLM interacts with the gene network, but its loss causes a distinct syndrome. (B) Graphical sketch of experimental design of fork protection assay. Lengths of nascent replication tracts (labeled with IdU) are measured by DNA spreading after 5 hr of replication stalling with HU. Representative DNA fiber images are given. Scale bars (white) correspond to 4 μm. (C) Preformed IdU tract lengths measuring replication fork stability by DNA spreading in patient-derived FANCD2-defective PD20 cells, but not cells complemented with the wild-type protein, shorten with HU. Median IdU tract lengths are given in parentheses here and in subsequent figures. (D) Nascent tract length distribution curve measured by DNA spreading in patient-derived FANCA-defective GM6914 cells show nascent strand shortening with HU, unlike cells complemented with the wild-type protein. (E) Nascent tract length distribution curve measured by DNA spreading in PD20 cells expressing the FANCD2 K561R mutant defective for monoubiquitination show nascent strand shortening with HU. See also Figure S1.

Figure 2. FA Pathway Suppresses Genomic Instability upon Replication Stalling

(A) Chromosomal aberrations measured by metaphase chromosome spreads with FANCA-deficient and –complemented GM6914 cells with HU (± SD, n = 40). Representative images of chromosomal aberrations of metaphase chromosomes are given. Scale bars (gray) correspond to 2 μm. (B) Cell survival analysis of FANCA-defective, patient-derived GM6914 cells and cells complemented with FANCA upon continuous HU treatment (± SEM, n = 4). (C) Cell survival analysis of FANCA-defective cells and cells complemented with FANCA upon continuous gemcitabine (GEM) treatment (± SEM, n = 3). (D and E) Nascent tract length distribution curves measured by DNA spreading in patient-derived FANCA-defective GM6914 cells and cells complemented with the wild-type protein with gemcitabine [GEM, (D)] and camptothecin [CPT, (E)]. See also Figure S2.

Figure 3. Parallel and Downstream Functions of FA-Associated Proteins

(A) Replication fork stability analysis by DNA spreading of preformed IdU tracts in BRCA1-deficient mouse ES cells (mES Brca1^−/−) and mouse ES cells containing wild-type BRCA1 (mES Brca1^+/−) with HU. Median IdU tract lengths are given in parenthesis here and in subsequent graph panels. (B) Replication fork stability analysis by DNA spreading of preformed IdU tracts in BLM-depleted mouse ES cells with negative doxycycline (DOX) control of BLM expression (mES Blm^tet/tet +DOX) and BLM proficient ES cells (mES Blm^tet/tet). See inset, western blot for BLM expression. (C) Replication recovery analysis after fork stalling with HU as measured by DNA spreading of CldU replication tracts in BLM-depleted (mES Blm^tet/tet +DOX) and BLM proficient mouse ES cells (mES Blm^tet/tet). Median CldU tract lengths are given in parentheses here and in subsequent graph panel. (D) Replication recovery analysis (CldU tract length) after fork stalling is in FANCD2-defective and in FANCD2-complemented PD20 cells. See also Figure S3.

Figure 4. FA Genes Protect Against MRE11-Dependent Fork Degradation

(A) Replication fork stability analysis by DNA spreading of IdU replication tracts in FANCD2-deficient PD20 cells during various exposure times to HU. Inset; the rate of IdU tract length change is 0.87 μm/hr, estimated to be ~2.2 kb/hr. (B) Replication fork stability analysis by DNA spreading of IdU replication tracts in FANCD2-deficient PD20 with chemical inhibition of MRE11 nuclease by treatment with mirin, with and without HU. See also Figure S4.

Figure 5. Functional Interaction of FA and RAD51 Proteins at Stalled Forks

(A) Replication fork stability analysis by DNA spreading of IdU replication tracts in FANCD2-complemented PD20 cells expressing flag-tagged BRC4-peptide (see western blot, inset), which disrupts RAD51 binding to DNA. (B) Replication fork stability analysis by DNA spreading of IdU replication tracts in FANCD2-deficient PD20 cells expressing flag-tagged BRC4-peptide (see western blot, inset). (C) Replication fork stability analysis by DNA spreading of IdU replication tracts in FANCD2-deficient PD20 cells expressing mutant RAD51 K133R (see western blot, inset), which forms stable filaments, upon fork stalling with HU. (D) Replication fork stability analysis by DNA spreading of IdU replication tracts in FANCD2-deficient PD20 cells overexpressing wild-type (WT) RAD51 (see western blot, inset), which promotes filament assembly, upon fork stalling with HU.

Figure 6. Model of FA/BRCA Gene Network Functions in Replication Fork Stability

Nucleotide depletion, as caused by oncogene activation or chemotherapeutic agents, stalls replication forks. FA/BRCA proteins stabilize RAD51 at stalled replication forks to protect nascent strands from MRE11-dependent degradation. RAD51 filament stabilization in the absence of FA/BRCA proteins is sufficient for fork protection. This can be achieved by gain of function mutant RAD51 or overexpression of wild-type RAD51, as commonly seen in tumor cells. BLM-TopIIIα acts downstream in the restart of stalled forks. Protein colors: BRCA2, pink; FANCD2, green; ubiquitin, dark green; BRCA1, light pink; wild-type RAD51, dark blue; RAD51 K133R, steel gray; PCNA, blue (doughnut); MRE11 yellow (pacman). BLM-TopIIIα, scalpel. See also Figure S5.