|Home | About | Journals | Submit | Contact Us | Français|
Conceived and designed the experiments: HKP HW PF. Performed the experiments: HKP HW. Analyzed the data: HKP HW JZ PF. Contributed reagents/materials/analysis tools: HKP HW JZ SD. Wrote the paper: HKP HW PF.
Extremely high cancer incidence associated with patients with Fanconi anemia (FA) suggests the importance of the FA signaling pathway in the suppression of non-FA human tumor development. Indeed, we found that an impaired FA signaling pathway substantially contributes to the development of non-FA human tumors. However, the mechanisms underlying the function of the FA pathway remain less understood. Using RNA interfering approach in combining with cell proliferation and reporter assays, we showed that the function of FA signaling pathway is at least partly mediated through coupling with hRad6/hRad18 signaling (HHR6 pathway). We previously reported that FANCD2 monoubiquitination, a hallmark of the FA pathway activation, can be regulated by HHR6. Here we found that hRad18 can also regulate activation of the FA pathway. More importantly, we found that FANCD2 is capable of modulating activity of DNA translesion synthesis polymerase eta, an effector of HHR6 pathway. These results provide novel insights into how the FA pathway is intertwined with HHR6 pathway to maintain chromosomal stability and suppress the development of human cancer, representing an important conceptual advance in the field of FA cancer research.
Our genome is constantly bombarded by both exogenous and endogenous genotoxic stresses, eventually leading to DNA damage. Well-orchestrated cellular responses to DNA damage are absolutely required to maintain genome stability and thus prevent diseases . The coordinated responses will either eliminate damaged cells or repair the damage to ensure a normal cell growth . To date, homologous recombination, non-homologous end joining, nucleotide excision repair, base excision repair, translesion synthesis, and DNA-crosslink repair are known repair responses to DNA damage, among which DNA-crosslink repair is attributed to all other damage repair processes described . In response to DNA crosslinks, a mammalian DNA damage signaling pathway, called Fanconi Anemia (FA) pathway, is activated , . This signaling pathway is determined by similar symptoms displayed from at least 13 or 14 complementation groups of FA –, which is a rare human genetic disease featured with severe bone marrow failure, many congenital defects, and an extremely high cancer incidence , . Within the FA pathway, the multi-FA protein complex can act as an E3 ubiquitin ligase to monoubiquitinate FANCD2 and its paralog FANCI, and the monoubiquitinated FA proteins then function in concert with other known or unknown proteins to repair DNA damage and maintain chromosomal stability , . FANCD2 monoubiquitination thus appears to be a measure of the activation of this DNA-crosslink damage response pathway. Yeast or Bacteria can not survive through one single DNA-crosslink if not repaired . In humans, impaired FA signaling was recently identified to be an important factor in promoting the development of non-FA human cancer . However, it still remains unknown as to how the FA signaling pathway or FANCD2 protein functions.
We previously found that HHR6 (human homolog of yeast rad6), a major player in HHR6 pathway signaling upon DNA damage, can regulate FANCD2 monoubiquitination . In HHR6 pathway, also called postreplication repair (PRR), a set of complex DNA replication recovery/damage tolerance processes permit DNA synthesis over a damaged template. This damage response pathway is composed of two subpathways: error-prone translesion synthesis (TLS)  or an error-free system , involving downstream reinitiation followed by gap-filling through recombinational events. The key components in TLS are low fidelity DNA polymerases specialized in lesion bypass, which are evolutionally conserved. In human, there are at least four DNA polymerases belonging to Y superfamily (pol η, pol κ, pol ι and REV1) as well as a few from other polymerase families. The existence of these DNA polymerases with very low fidelity suggests that their participation in genome replication needs to be carefully regulated. But molecular events that regulate these TLS DNA polymerases in humans remain unclear. Studies in S. cerevisiae suggest that post-translational modification of proliferating cell nuclear antigen (PCNA), an accessory factor of replicative DNA polymerases, is important for the switch from the replicative polymerase to a TLS polymerase , –. PCNA monoubiquitination under the control of Rad6/Rad18 is believed to provide a platform for both error-prone and error-free TLS events possibly to occur. Here we found that hRad18 can also regulate FANCD2 activation, and that human TLS DNA polymerase eta (pol η) can be regulated by FANCD2, indicating the function of the FA pathway is at least partly mediated through coupling with the HHR6 pathway.
The fact that FANCD2 is activated during DNA synthesis or upon DNA damage suggests the existence of close interplays between FANCD2 and players in the processes of DNA synthesis and DNA damage repair. The hypersensitivity to DNA-crosslinking agents, such as UV-irradiation, inherent to cells derived from patients with FA or XPV (Xeroderma Pigmentosum Variant) , or to yeast deficient in Rad6 , , further indicates specific links among FANCD2, pol η, and human homologs of yeast Rad6 (HHR6). To uncover the mystery of FANCD2 function, we decided to reveal these potential signaling links. Indeed, we found that HHR6 can regulate FANCD2 activation/monoubiquitination . Here we questioned whether hRad18 can also regulate FANCD2 monoubiquitination upon DNA damage, becasue HHR6 cooperates with hRAD18 to modulate the function of proteins involved in DNA damage responses –. Using RNA interfering (RNAi) approach, we found cells carrying different levels of hRad18 protein displayed altered levels of monoubiquitinated FANCD2 following exposure to UV (Fig. 1A) or mitomycin C (MMC) (not shown), suggesting the involvement of hRad18 in the regulation of FANCD2 monoubiquitination. Moreover, FANCD2 focus formation, an additional measurement for FANCD2 activation/monoubiquitination was also found to be compromised in corresponding UV-treated cells carrying a low level of hRad18 expression (Fig. 1B). Taken together, upon DNA damage, FANCD2 activation is at least partly regulated by hRad18.
FANCD2 monoubiquitination is an essential event in the FA signaling transduction. Cells deficient in FANCD2 monoubiquitination, such as FA cells, are hypersensitive to interstrand DNA cross-linking (ICL) agents . To further verify hRad18 regulation of FANCD2, we tested whether the deficient monoubiquitination of FANCD2 triggered by a lower level of hRad18 expression also confers similar cell sensitivity to DNA crosslinking agents as the one triggered by deficient FA genes. Populations of U2OS or Hela cells, transfected with either non-specific RNAi oligos, specific RNAi oligos targeting hRad18 or FANCL, were treated with a series of different mitomycin C (MMC) concentrations (50 ng/ml–200 ng/ml) for 5 days to examine cell survival ability. Both cell lines showed a similar decline in survival rate with an exception about 5% more Hela cells survived at each drug dose tested respectively (Fig. 2A, and data not shown). Images of cell density for U2OS (Fig. 2B) or Hela cells (not shown) support their declined cell growth rates. Moreover, cell populations treated with RNAi oligos against either FANCL or hRad18 exhibited a similar reduction in cell growth following MMC treatment as compared to cells treated with control non-specific RNAi oligos. These results indicate that, although it remains unknown the manner, by which hRad18 regulates FANCD2 monoubiquitination, hRad18 and FA complex appear to be capable of sharing a common downstream target, FANCD2, in the responses of cells exposed to MMC tested here.
In response to DNA damage, the signaling cascade initiated by HHR6-hRAD18 is known as HHR6 pathway/PRR , within which PCNA is a mediator of HHR6/hRad18 to regulate the function of DNA-translesion synthesis polymerases including pol η. We have found that FANCD2 can be regulated by HHR6  as well as by hRad18 (Fig. 1), suggesting FANCD2 may act as an additional mediator of HHR6/hRad18 to regulate downstream events of HHR6 pathway/PRR. Furthermore, the similar DNA damage sensitivity displayed by cells carrying deficient FANCD2 or pol η indicate a functional link between FANCD2 and pol η. We thereby anticipated a potential functional relation between FANCD2 and pol η, specifically between two protein focus formations given that both proteins are known to form foci in response to DNA damage , . Indeed, we found focus colocalization between FANCD2 and pol η in cells following UV exposure (Fig. 3A). To this point, we further asked whether FANCD2 plays regulatory roles for the formation of pol η foci. We created a cellular system expressing FANCD2 protein at different levels by transfecting U2OS cells with RNAi oligos targeting FANCD2 or non-specific for controls (Fig. 3B). We then re-transfected these cells with plasmid containing GFP-pol η, followed by analyzing the effects of FANCD2 protein expression levels on focus formation of pol η. We found that the percentage of green focus formation over total green cells is higher in cells transfected with control RNAi oligos, in which the levels of FANCD2 are normal, compared to cells with FANCD2 RNAi oligos leading to a lower level of FANCD2 protein (Fig. 3B and 3C). Therefore, FANCD2 is able to regulate the focus formation of pol η. Similar results were obtained in FA patient cells PD20 (FANCD2-/-), compared to relevant control cells carrying functional FANCD2 (data not shown). These results indicate that, in response to DNA damage, pol η may act as a functional downstream target of FANCD2 and thus, at least partly, mediate FANCD2 function in maintaining chromosomal stability.
HHR6 pathway/PPR is initiated upon the stalled replication forks resulting from many types of DNA lesions including the abasic one. Translesion synthesis DNA polymerases can synthesize DNA through lesion templates to prevent the collapse of the stalled replication-forks , . Among these lesion bypassing DNA polymerases, pol η can synthesize DNA along templates carrying damages including the abasic lesion by inserting A or G into the newly synthesized complementary strand , , . A well-defined abasic bypass reporter system  thus can provide a functional readout for FANCD2 regulation of pol η. We expected that the abasic-bypass reporter activity should be different in cells carrying different levels of FANCD2 protein expression, as suggested by data shown in Figure 3. To reveal this potential effect of FANCD2, we performed reporter assay of abasic translesion synthesis in vitro by using nuclear extracts (NEs) with deficient or proficient FANCD2. As illustrated in Figure 4A, the reporter mixture was mixed with NEs of FA patient cells carrying FANCD2-/- (PD20 cells) or FANCD2+/+ (complemented PD20 cells), respectively in the DNA synthesis buffer containing dNTPs, and incubated for 3 hours at room temperature. Subsequently we extracted plasmids and followed procedures described previously . We found that abasic bypass activity was dramatically decreased in NEs without FANCD2, compared to the NE carrying FANCD2 (Fig 4B). We also performed in vivo reporter assay by transfecting plasmids into U2OS cells carrying different levels of FANCD2 expression through RNAi approach, followed by similar procedures used for in vitro assay with an exception of extracting plasmids out of cells rather than from the reaction mixtures. The results were consistent with ones derived from the in vitro assay (data not shown). Clearly, FANCD2 plays an essential role in the regulation of, at least, bypassing abasic DNA lesion and appears to be a novel regulator of DNA translesion synthesis polymerase eta.
The similar DNA damage sensitivity between systems carrying either deficient HHR6 or improper FA signaling prompted us to explore a possible signaling link between HHR6 and FA pathways. Indeed, we found that not only HHR6  but also hRad18 are capable of regulating the activation of FANCD2 (Fig 1). More importantly, we found FANCD2 can regulate translesion synthesis (TLS) DNA polymerase eta (Fig. 3, ,4),4), which appears to be a first protein to be indentified acting as a mediator of FANCD2 function. This finding is consistent with the concept long-proposed in the field of FA study that FA proteins may play roles in TLS , , . Unlike many known TLS enzymes, pol η mostly synthesizes DNA accurately upon the lesion template , , as demonstrated by XPV resulting from a mutated pol η , . We believe chromosomal stability maintained by FANCD2/FA pathways is at least partly attributed to the function of pol η; pol η may as well be a potential candidate FA-like or FA-related gene for those unclassified FA patients. Furthermore, the regulation of pol η by FANCD2 may indicate a part of FANCD2 broader effects, rather than strictly functioning in the FA signaling. The majority of FA core complex genes are completely absent in many eukaryotic species that do contain orthologues of FANCD2 , . This difference also suggests the additional function that FANCD2 may have. In fact, FANCD2 has been suggested to participate in a separate signaling pathway that is activated by ATM in response to ionizing radiation , and it was recently also found to induce apoptosis . Collectively, our study provides a framework for our understanding of maintaining chromosomal stability through converging two DNA damage response pathways (Fig. 5), in which many questions await answers, such as how HHR6/hRad18 regulates the activation of FANCD2/the FA pathway, directly or indirectly? Or simply, Pol η is regulated by monoubiquitinated FANCD2, un-monoubiquitinated FANCD2, or both? We believe our ongoing and future studies, and those of others will provide refined mechanisms underlying the integration of HHR6 and FA pathways in maintaining genomic stability (Fig. 5).
All cell lines were obtained from the American Type Culture Collection (ATCC), with an exception of specially engineered ones. Mitomycin C was purchased from Sigma.
Equal numbers of U2OS or Hela cells were seeded in 60 mm dishes one day prior to transfection with control non-specific RNAi oligos or RNAi oligos against hRad18, or FANCL. Thirty-six hours posttransfection, cells were treated with, 50 ng/ml, 100 ng/ml, or 200 ng/ml MMC for 5 days. The number of surviving cells was scored, and cell numbers were plotted as cell survival curves (cell numbers of the samples treated with the drug were normalized to the cell numbers of the untreated control sample. Each drug dose was tested in triplicate). The shown images were RNAi oligo-transfected cells, which were treated with 50 ng/ml MMC for 3 days.
GP21 abasic plasmid (kan-resistance) and GP20-gap plasmid (cm-resistance) were constructed as described . The GP21 abasic primer: 5′-ACCGCAACGAAGTGATTC CCGTCGTGACTGXGAAAACCCTGGGCTACTTGAACCAGACCG -3′; GP-20 gap primer: 5′- ACCGCAACGAAGTGATTCCCGTCGTGACTGGGAAAACCCTGGG CTACTTGAACCAGACCG -3′; TLS (XmnI) primer: 5′- GGA ATC ACT TCG TTG -3′; and TLS (BstXI) primer: 5′- CTG GTT CAA GTA GCC -3′ (X is an abasic site). Nuclear Extracts (NE) were prepared from PD20 and PD20 + DANCD2 cells by using Kit#78833 (Thermo Scientific). The in vitro assay was set up by mixing GP20 (50 ng), GP21(50 ng), and NE (50 µg) in Tris-HCl buffer (40 mM, pH 7.5) including MgCl2 (5 mM), dithiothreitol (1 mM), bovine serum abumin (100 µg/ml), 10% glycerol, and dNTP (100 µM). The mixture was incubated at room temperature for 3 hours, followed by phenol chloroform extraction and precipitation of plasmid DNA. Subsequently, the extracted plasmid mixture was transformed into DH5α competent bacteria and equally spread on cm (chloramphenical) and kan (kanamycin) containing LB agar plates.
These techniques were performed essentially as described , , . HRad18 antibody was purchased from Santa Cruz and used at a dilution rate of 1500 for Western blotting. All RNAi ologos were purchased from Dhmarcom, targeting hRad18 (cttgctgtg tgactgtcac), FANCL (gacaagagctgtatgcact) , or FANCD2 (ccaggaagcaaccactttc).
We thank Dr. Zvi Livneh (Weizmann Institute of Science, Israel) for providing the abasic reporter system and working protocols. We also thank Dr. Iran Dikic (Goethe University Medical School, Germany) for providing us GFP-pol eta. We would also like to thank Dr. Fred Dudimah for proofreading the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
Funding: This study is partly supported by National Institutes of Health grant CA136532 to PF, and is also partly supported by the Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, and the Hormel Institute, University of Minnesota, Austin, Minnesota. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.