The Fanconi anemia pathway and ubiquitin

Fanconi anemia (FA) is a rare genetic disorder characterized by aplastic anemia, cancer/leukemia susceptibility and cellular hypersensitivity to DNA crosslinking agents, such as cisplatin. To date, 12 FA gene products have been identified, which cooperate in a common DNA damage-activated signaling pathway regulating DNA repair (the FA pathway). Eight FA proteins form a nuclear complex harboring E3 ubiquitin ligase activity (the FA core complex) that, in response to DNA damage, mediates the monoubiquitylation of the FA protein FANCD2. Monoubiquitylated FANCD2 colocalizes in nuclear foci with proteins involved in DNA repair, including BRCA1, FANCD1/BRCA2, FANCN/PALB2 and RAD51. All these factors are required for cellular resistance to DNA crosslinking agents. The inactivation of the FA pathway has also been observed in a wide variety of human cancers and is implicated in the sensitivity of cancer cells to DNA crosslinking agents. Drugs that inhibit the FA pathway may be useful chemosensitizers in the treatment of cancer. Publication history: Republished from Current BioData's Targeted Proteins database (TPdb; ).

The clinical course and the treatment of FA have been extensively reviewed elsewhere [3,4,16]. Clinically, FA is characterized by childhood onset aplastic anemia, increased cancer/leukemia susceptibility and developmental defects. Typically, FA patients develop bone marrow failure leading to aplastic anemia during the first decade of life and at least 20% develop malignancies. Most commonly, these include acute myelogenous leukemia and myelodysplastic syndrome, but also head and neck squamous cell carcinoma, gynecological squamous cell carcinoma, esophageal carcinoma, and liver, brain, skin and renal tumors [17][18][19]. FA subtypes FA-D1 and FA-N are associated with increased predisposition to medulloblastoma, Wilms' tumor and acute leukemia in early childhood, and are clinically different from the other FA The FA core complex and monoubiquitylation of FANCD2 Ubiquitin plays a crucial role in the regulation of the FA pathway. Eight FA proteins (FANCA, FANCB, FANCC, FANCE, FANCF, FANCG, FANCL and FANCM), a newly identified FANCM-interacting protein called FAAP24 (FANCA-associated polypeptide) [137] and an unidentified factor called FAAP100 form a nuclear protein complex (the FA core complex) required for monoubiquitylation of FANCD2 on lysine 561 [27,34,37,38]. One of the components of the FA core complex, FANCL, has a PHD (plant homeodomain) finger/RING finger domain exhibiting auto-ubiquitin ligase activity in vitro [38] [39][40][41]. FANCL associates through its PHD/RING finger domain with UBE2T, a ubiquitin conjugating enzyme (E2), which is also required for in vivo FANCD2 monoubiquitylation [42]. Taken together, the FA core complex is assumed to constitute a multi-subunit E3 ubiquitin ligase complex for FANCD2, in which FANCL is the catalytic E3 ubiquitin ligase subunit. The catalytic activity of this complex appears to be regulated through the inhibitory auto-monoubiquitylation of UBE2T, stimulated in the presence of FANCL [42]. However, as the direct monoubiquitylation of FANCD2 by the FA core complex has not been reconstituted so far in vitro, the existence of other E2 and E3 enzymes responsible for FANCD2 monoubiquitylation (and somehow controlled by the FA core complex, the E2 activity of UBE2T and the E3 activity of FANCL) cannot be ruled out formally.
Even though a direct DNA binding activity has been demonstrated for unmodified FANCD2 in vitro [43], monoubiquitylation of the protein is required for its translocation to chromatin in vivo and nuclear foci formation at the site of DNA damage, as well as for cellular resistance to DNA crosslinking agents [34,44]. FANCD2 monoubiquitylation and nuclear foci formation occur in response to DNA damaging agents (ionizing radiation, UV light irradiation, DNA crosslinking agents, hydroxyurea, etc.) and during S phase of the cell cycle even in the absence of exogenous DNA damage [34,45]. A DNA damage-activated signaling kinase, ATR, and a single-strand DNA binding protein complex, RPA, are required for DNA damage-inducible monoubiquitylation and foci formation of FANCD2, indicating an upstream role for these factors in the activation of the FA pathway [46]. The exact signal and activation cascade required for FANCD2 monoubiquitylation, however, remain elusive. The existence of a specific FANCD2 receptor responsible for recruitment of the monoubiquitylated protein to chromatin has also been proposed [44], but not demonstrated to date.
Roles for the FA core complex in DNA enzymatic processing The FA core complex is not simply the ubiquitin ligase complex for FANCD2. In addition to its requirement for FANCD2 monoubiquitylation, the FA core complex is required for the translocation into chromatin of monoubiquitylated FANCD2, and for cellular resistance to interstrand DNA crosslinks even if FANCD2 is localized in chromatin, as elegantly demonstrated in chicken DT40 cells using FANCD2-monoubiquitin and FANCD2-histone H2B chimeric proteins [47]. In addition, one component of the FA core complex, FANCM, harbors DNA helicase motifs, a degenerate nuclease motif and in vitro DNA-stimulated ATPase and translocase activities [27]. The newly identified FAAP24 protein, in complex with FANCM, has been shown to preferentially bind to ssDNA and branched DNA structures [137]. Speculatively, FANCM DNA translocase activity could play an important role in displacing the FA core complex along the DNA, allowing DNA damage recognition, or FAAP24 specificity for ssDNA structures may target FANCM and the FA core complex to abnormal, branched DNA structures. In summary, the FA core complex itself can interact with DNA and displays some important functions outside of FANCD2 monoubiquitylation. Furthermore, the FA core complex forms a larger complex with BLM, RPA and topoisomerase IIIα called BRAFT (for BLM, RPA, FA, and topoisomerase IIIα). Although the functional relevance of BRAFT is not clear, it harbors a DNA unwinding activity potentially relevant for DNA repair [37].
Negative regulation of the Fanconi anemia pathway by FANCD2 deubiquitylation FANCD2 monoubiquitylation is a regulated and reversible process. USP1, a deubiquitylating enzyme, removes ubiquitin from monoubiquitylated FANCD2, therefore negatively regulating the FA pathway [48]. Interestingly, USP1 also deubiquitylates monoubiquitylated PCNA (proliferating cell nuclear antigen), a DNA polymerase processivity factor [49]. PCNA monoubiquitylation leads to the switch from a replicative polymerase to a translesion synthesis (TLS) polymerase at the site of stalled replication forks [50] and participates in the bypass of DNA lesion. Monoubiquitylation of PCNA is dependent on RAD6 (an E2) and RAD18 (an E3) [51], but is independent of the FA core complex. Therefore, these two pathways (the FA pathway and TLS activation through PCNA monoubiquitylation) have different activation mechanisms, but share a common shutoff mechanism.
The Fanconi anemia pathway and DNA repair Interestingly, some TLS polymerases are also implicated in cellular resistance to interstrand DNA crosslinks [52]. For example, mutations in TLS polymerase-encoding REV3 or REV1 have been shown to be epistatic with FANCC mutations for crosslinker hypersensitivity in chicken DT40 cells [53]. Furthermore, REV1 and FANCD2 colocalize in nuclear foci [53], which could suggest a possible interaction between them. The possible interplay of FA proteins    [68][69][70].
Taken together, these evidences suggest that FANCD2 plays a role in regulating DNA repair (especially of interstrand DNA crosslinks) in chromatin in cooperation with BRCA1, FANCD1/BRCA2, FANCJ/BRIP1/BACH1, the FA core complex and other factors, but the precise mechanism by which this occurs remains unknown. The highly regulated process of FANCD2 monoubiquitylation could play a crucial role in orienting the repair of DNA damage to HR and/or in attracting TLS polymerases to sites of DNA damage.

Other roles for the Fanconi anemia proteins outside of DNA repair
Several FA proteins function outside of DNA crosslink repair. For example, the phosphorylation of FANCD2 at serine 222 by ATM kinase in response to ionizing radiation is required for the establishment of IR-induced intra-S phase checkpoint, but is not required for resistance to DNA crosslinking agents [74]. Additionally, FANCC (localized in both the cytoplasm and nucleus) is implicated in several cytoplasmic functions, such as JAK/STAT and apoptotic signaling [75][76][77][78][79][80].

Alterations in the Fanconi anemia pathway in human cancer in the general (non-FA) population
FA patients have an increased risk of developing leukemia and solid tumors, but alterations in the FA pathway have also been reported in a wide variety of human cancers in the general (non-FA) population [81][82][83][84][85][86][87][88][89][90][91][92][93][94] (reviewed in references [6,10]). Abnormalities in BRCA1 and BRCA2 in cancer have been reviewed elsewhere [95]. Methylation of FANCF, leading to its decreased expression, has been reported in ovarian cancer [81,82], breast cancer [84], non-small cell lung cancer [85], cervical cancer [86], testicular cancer [87], head and neck squamous cell carcinoma [85], and granulosa cell tumors of the ovary [83]. Inherited and somatic mutations of FANCC and FANCG have been detected in a subset of young onset pancreatic cancer [89,96]. Additionally, inherited mono-allelic mutations in BACH1/BRIP1/FANCJ and FANCN/PALB2 have been recently implicated in breast cancer predisposition (familial breast cancer) [97] [23,24], suggesting that these genes may be the low penetrance breast/ovarian cancer susceptibility genes researchers have been looking for to explain the non-BRCA1/non-BRCA2 cases. Because the integrity of the FA pathway is crucial for cellular resistance to interstrand DNA crosslinking agents (cisplatin, MMC, melphalan, etc.), tumors with defects in the pathway may be hypersensitive to these agents. In fact, in some cancer cell lines (human ovarian cancer cell lines (TOV-21G and 2008) [81], human myeloma cell lines (8226 and U266) [98] and a human pancreatic cancer cell line (PL11) [99]), the integrity of the FA pathway is a determinant of cisplatin (or melphalan) resistance in vitro or in vivo (mouse xenograft model) [81,[98][99][100].

Disease models, knockouts, assays
Generation of various knockout murine models for FA (Fanca [101,102], Fancc [103,104], Fancg [105,106], Fancd2 [107] [126] (Table 1). Drosophila and C. elegans models will therefore be useful for testing the functions of key proteins in the FA pathway. As only FANCD2, FANCL and FANCM are present in Drosophila, these proteins may constitute the minimal FA pathway machinery. In turn, C. elegans may enable a better understanding of the roles of FANCD1/BRCA2 and FANCJ in the context of the minimal machinery constituted by FANCD2 and FANCM. As with vertebrate mutants, Drosophila fancd2 and fancl mutants, as well as C. elegans fancd2 mutants, show hypersensitivity to DNA crosslinking agents [125,126]. Drosophila and C. elegans models may, therefore, be useful to dissect the roles and regulations of the FA pathway in a less complex network. The zebrafish model may prove a valuable tool with which to investigate the impact of FA pathway failure on development as fancd2-deficient zebrafish embryos develop similar defects to those found in children with FA, including shortened body length, microcephaly (small head) and microphthalmia (small eyes) [127]. The Xenopus model mostly constitutes a powerful tool with which to investigate the regulation of the FA pathway in vitro through cell-free assay systems using replicating egg extracts [124]. Such assays could also be utilized for screening drugs that modulate the FA pathway.

Disease targets and ligands
The FA pathway is required for cellular resistance to DNA crosslinking agents, including widely used anti-cancer drugs such as cisplatin, MMC and melphalan. As such, inhibition of the FA pathway will lead to sensitization of tumor cells to these drugs. Therefore, the FA pathway is an attractive target for developing small molecule inhibitors that may be clinically useful as chemosensitizers in the treatment of cancer. As described above, the FA pathway involves formation of a multi-subunit protein complex harboring E3 ligase activity, several enzymes (ubiquitin ligase and conjugating enzyme, deubiquitinating enzyme, kinase, ATPase/DNA translocase and ATPase/helicase) and many protein-protein or protein-DNA interactions. All of these components are potential targets for small molecule inhibitors of the FA pathway.
Drug development targeting the FA pathway is still in the early stages, and therefore not much information is available. A high-throughput cell-based screening assay for small molecule inhibitors of the FA pathway has, however, been developed by Alan D'Andrea (Dana-Farber Cancer Institute), Toshiyasu Taniguchi (Fred Hutchinson Cancer Research Center) and their colleagues [128]. In this screen, inhibition of DNA damage-induced FANCD2 nuclear foci formation was utilized as a readout. Continued screening is ongoing and a partial result focusing on one inhibitor, curcumin (diferuloylmethane), has been published [128]. Curcumin is a low-molecular-weight polyphenol derived from the rhizome Curcuma longa and is the principal ingredient in the spice turmeric [129]. Curcumin inhibits FANCD2 monoubiquitylation and nuclear foci formation, although its exact target in the FA pathway has not been identified [128]. However, curcumin sensitizes an ovarian cancer cell line to cisplatin in an FA pathway-dependent manner, suggesting that curcumin sensitization of cisplatin mostly occurs through FA pathway inhibition [128]. In order to establish curcumin as a useful cisplatin chemosensitizer, a detailed isobologram analysis of combinations of curcumin with cisplatin, in vivo studies using mouse models, and identification of the target of curcumin in the FA pathway are required.

New frontiers in drug discovery
Elucidating the mechanism of action of the candidate inhibitors identified in the above-mentioned screen [128] will be crucial for development of specific and effective inhibitors of the FA pathway, and for further understanding the regulation of the FA pathway. Precise analyses of the effects of each drug on individual steps in the FA pathway will be required to identify their specific targets. These analyses will include in vitro ATR kinase assay, in vitro UBE2T and FANCL autoubiquitylation assays, assessment of FA core complex formation and evaluation of nuclear foci formation of DNA repair proteins including BRCA1. A better understanding of the regulatory mechanisms, as well as the identification of new partners of the FA pathway, is also crucial for the identification and development of FA pathway inhibitors. Although multiple groups have been working on structural analysis of FA proteins [130,131], so far few FA protein crystal structures (BRCA2/FANCD1 [130] and FANCF [131]) have been reported, but such studies could provide useful information for drug design as well as for elucidation of the targets of the candidate inhibitors.

Competing interests
The authors declare that they have no competing interests.