Mutagenesis, Vol. 17, No. 6, 529-538,
November 2002
© 2002 UK Environmental Mutagen Society/Oxford University Press
The Fanconi anaemia genome stability and tumour suppressor network
Mutagenesis Group, Department of Genetics and Microbiology, Universitat Autònoma de Barcelona, 08193 Bellaterra, Barcelona, Spain
 |
Abstract |
|---|
 Top Abstract IntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
Fanconi anaemia (FA) is a rare autosomal recessive disease characterized by increased spontaneous and DNA crosslinker-induced chromosome instability, progressive pancytopenia and cancer susceptibility. An increasing number of genes are involved in FA, including the breast cancer susceptibility gene BRCA2. Five of the FA proteins (FANCA, FANCC, FANCE, FANCF and FANCG) assemble in a complex that is required for FANCD2 activation in response to DNA crosslinks. Active FANCD2 then interacts with BRCA1 and forms discrete nuclear foci. FANCD2 is independently phosphorylated by ATM (the protein whose gene is mutated in ataxia telangiectasia) in response to ionizing radiation. In addition, the FA proteins are interconnected with other nuclear and cytoplasmic factors all related to cellular responses to carcinogenic stress and to caretaker and gatekeeper functions. In this review, the most recently published data on the molecular biology of the FA pathway and its molecular crosstalk with ATM, BRCA1 and BRCA2, proteins involved in xenobiotic and reactive oxygen species metabolism, apoptosis, cell cycle control and telomere stability, are summarized. The currently available data indicate that FA is a central node in a complex nuclear and cytoplasmic network of tumour suppressor and genome stability pathways fully committed to prevent cancer.
|
Introduction |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
The integrity and stability of the genetic material is continuously being threatened by endogenous and exogenous factors such as chemical mutagens and radiation. To reduce the harmful effects of exposure to DNA-damaging agents, the human genome has evolved a complex network of genome stability pathways (Wood, 2001
). Germline mutations in these pathways increase cancer risk and are related to a number of cancer-prone syndromes, such as ataxia telangiectasia (AT), Nijmegen breakage syndrome (NBS), xeroderma pigmentosum (XP) and Fanconi anaemia (FA) (Hoeijmakers, 2001).
FA is a rare genetic disease characterized by increased spontaneous and mutagen-induced chromosome instability, a diverse assortment of congenital malformations, progressive pancytopenia and cancer susceptibility, especially acute myelogenous leukemia (AML), but also solid tumours, very often squamous cell carcinomas of the head and neck. Other cancers in FA are hepatic, gastrointestinal, brain and gynecological tumours, among others (Alter, 1996). According to the last up-date of the International Fanconi Anemia Register (IFAR) presented at the last International Fanconi Anemia Symposium in Portland (OR) by Dr A.D.Auerbach, 25.8% of FA patients develop the first malignancy at a median age of 14.5 years, 62% of which were haematological and the rest non-haematological tumours. The probability of developing cancer is age dependent and by 40 years of age it is higher than 50% for haematological cancers and 50% for solid tumours.
FA is usually a fatal disease, with a mean survival of 16 years (D’Andrea, 1996). Ninety-eight per cent of FA patients develop haematological abnormalities with a risk of death from haematological complications of 81% before 40 years of age (Auerbach et al., 1997). They present a number of haematological disorders, including bone marrow failure, aplastic anaemia, thrombocytopenia and pancytopenia. In addition, clonal cytogenetic abnormalities are observed in bone marrow cells in 67% of patients below 30 years of age (Auerbach et al., 1997).
The most frequent birth defects in FA are growth retardation and abnormalities in the skin, typically cafe-au-lait spots, in the upper extremities, especially in the thumbs and forearms, kidneys and gastroinstestinal system (Joenje and Patel, 2001). Thirty per cent of the patients, however, show no evidence of congenital abnormalities and, therefore, a diagnosis based on clinical features is extremely difficult (Strathdee and Buchwald, 1992). Therefore, the final confirmation of FA is actually performed by testing the chromosome-breaking ability of DNA crosslinking agents such as mitomycin C (MMC) and diepoxybutane (DEB) in the patients’ blood lymphocytes. The confirmation of a FA diagnosis by classical cytogenetic methods involves cell culturing, MMC or DEB treatment, microscopic analysis of metaphase chromosomes in blood lymphocytes and detection of chromosome breaks, usually of chromatid type, such as radial figures (see Figure 1). This technique also allows the detection of mosaic FA patients (Figure 2) with two mixed subpopulations of cells: FA cells and normal somatically reverted cells (Lo Ten Foe et al., 1997; Waisfisz et al., 1999).
|
|
Since several genes are involved in FA (see below), the subtyping is usually performed by cell fusion complementation studies. The cloning of the FA genes has recently allowed the subtyping of FA patients by retroviral complementation. Blood T cells and/or transformed lymphoblastoid cell lines (LCL) from FA patients are transduced with different retroviral vectors encoding FA gene cDNAs and then exposed to increasing concentrations of MMC to evaluate the potential genetic correction of the MMC hypersensitivity which characterizes FA cells (Casado et al., 2001
; Hanenberg et al., 2002). The complementation of FA phenotype by gene transfer has also been very useful in basic research studies aimed at unravelling the molecular biology and function of the FA proteins. Understanding the role of the FA proteins is essential not only to find a cure for the disease but also for the general population, since it is becoming clear that the FA pathway is a central node in a complex network of tumour suppressor and genome stability pathways.
|
Fanconi anemia is genetically heterogeneous |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
FA is a very rare autosomal recessive genetic disease with a prevalence of 1–5 per million (Joenje and Patel, 2001
). The frequency of heterozygote carriers is estimated to be 1:200–1:300, with a higher frequency of up to ~1:100 in some ethnic groups, such as Ashkenazi Jews and South African Afrikaners. It is known that a common mutation in the FANCC gene is responsible for the majority of FA patients in Ashkenazi Jews (Whitney et al., 1993) and the majority of FA families among the Afrikaner population of South Africa have the same mutation in FANCA. Molecular and genealogical evidence obtained in this latter group confirmed the existence of a founder effect for FA in South Africa (Tipping et al., 2001). The frequency of FA patients and carriers in Spanish gypsies is also much higher than in the Spanish non-gypsy population. We are currently investigating an ancestral origin of the FA mutation in this ethnic group.
Cell fusion complementation studies revealed the existence of eight FA complementation groups (Joenje et al., 1997) and the presence of eight FA genes was initially anticipated: FANCA, FANCB, FANCC, FANCD1, FANCD2, FANCE, FANCF and FANCG. Howlett and co-workers have recently reported that cell lines derived from FA B and FA D1 patients have biallelic mutations in the breast cancer susceptibility gene BRCA2 and express truncated BRCA2 proteins (Howlett et al., 2002). This indicates that BRCA2 is actually a FA gene and, therefore, the current number of identified FA genes is seven.
Most of the genes (FANCA, FANCC, FANCE, FANCF and FANCG) were identified using the complementation cloning technique originally reported by Dr M.Buchwald’s group (Strathdee et al., 1992a). The first cloned FA gene was FANCC in 1992 (Strathdee et al., 1992b), followed by FANCA in 1996 (Fanconi Anaemia/Breast Cancer Consortium, 1996; Lo Ten Foe et al., 1996). Two years later, the same group led by Dr H.Joenje reported that FANCG was identical with XRCC9 (de Winter et al., 1998), a gene cloned by Liu and co-workers in 1997 and isolated by virtue of it complementing a Chinese hamster mutant cell line (UV40) hypersensitive to DNA crosslinkers (Liu et al., 1997). Dr Joenje’s group then reported the characterization of two new FA genes in 2000, FANCE (de Winter et al., 2000a) and FANCF (de Winter et al., 2000b). FANCD2 was identified by positional cloning in 2001 (Timmers et al., 2001) and the FA D1 subtype has been now reported to be caused by truncated forms of BRCA2 (Howlett et al., 2002). According to the last update of the IFAR mentioned above, 96% of FA patients belong to complementation groups A (60%), C (23%), and G (13%) in a population of 381 subtyped FA patients, most of them North American. The rest of the subtypes are very rare. For instance, only five FA D2 patients have been identified to date world wide (Taniguchi et al., 2002). We have also identified a cluster of FA D2 patients in Spain (unpublished results). Further information on FA genes and FA mutations can be obtained in an excellent recent review by Joenje and Patel (2001) and in the FA mutation database (www.fanconi.org).
There is some controversy about the genotype–phenotype relationships in FA. A recent study indicates that FA patients with mutations in the FANCG gene and patients homozygous for null mutations in FANCA are high risk groups with a poor haematological outcome (Tipping et al., 2001). However, the last update of the IFAR indicates that in FA C patients the age of onset of haematological problems is younger and overall survival is poorer when compared with groups A and G.
The molecular biology of FA is slowly being understood thanks to the cloning of FA genes, the availability of antibodies to FA proteins and cDNAs complementing the FA phenotypes. It was soon observed that FA proteins have nuclear and cytoplasmic functions and both of these roles probably cooperate in preventing chromosome instability and cancer. However, there is some controversy about the dual cytoplasmic and nuclear roles of the FA proteins (see the interesting letter to the editor and reply by A.D’Andrea and R.C.Cummings and M.Buchwald, respectively, in Nature Med., 7, 1259–1260).
|
All FA proteins participate in a common nuclear pathway |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
Most of the FA proteins (A, C, E, F and G) assemble in a multisubunit nuclear complex (the FA complex) which is required for the activation of FANCD2 via monoubiquitination at Lys561 (Garcia-Higuera et al., 2001
). Monoubiquitinated FANCD2 (FANCD2-L) is longer than the inactive FANCD2 (FANCD2-S), giving two distinguishable bands in a western blot when the FA pathway is fully functional or a single FANCD2-S band when one of the FA genes of the FA complex is mutated (see Figure 3). This allows a rapid screening and diagnosis of FA (Garcia-Higuera et al., 2001), since the great majority of patients have mutations in genes involved in the FA complex (see above). A recent report shows that FANCE binds to FANCC and FANCD2 and that FANCE is required for FANCC nuclear accumulation (Pace et al., 2002). This finding suggests that FANCE is the molecular link between FA complex assembly and FANCD2 downstream in the nucleus.
|
The recently cloned FANCD2 gene is thought to be a key player in the FA pathway since it is the only FA gene conserved in evolution, with homologous genes found in distant species such as Drosophila (Timmers et al., 2001
; V.Castillo, O.Cabré, R.Marcos and J.Surrallés, submitted for publication; GenBank accession no. AJ459772). In addition, the FA protein complex is assembled in the absence of FANCD2, indicating that FANCD2 is downstream in the FA pathway. The ubiquitinated isoform of FANCD2 associates with the breast cancer susceptibility and double-strand break (DSB) repair protein BRCA1 in DNA damage-induced nuclear foci (Garcia-Higuera et al., 2001) or during S phase (unpublished observation). It is not known whether the FA complex has intrinsic ubiquitin ligase activity or whether it functions upstream of a ubiquitin ligase. Recent studies have shown that BRCA1 has ubiquitin ligase activity and it is therefore a good candidate in FANCD2 monoubiquitination (Grompe and D’Andrea, 2001). A second post-translational modification is also important in the response to DNA damage: ATM, the product of the gene mutated in the chromosome fragility syndrome AT, phosphorylates FANCD2 at Ser222 in response to ionizing radiation resulting in activation of a G1/S checkpoint (Taniguchi et al., 2002). FA complex-dependent monoubiquitination and ATM-dependent phosphorylation are two independent post-translational modifications of FANCD2. Thus, disruption of the FA complex or mutations in the downstream FA genes results in hypersensitivity to crosslinking agents whereas lack of ATM-dependent phosphorylation of FANCD2 leads to activation of a G1/S cell cycle checkpoint and radio-resistant DNA synthesis (Figure 4). Consistently, disruption of Ser222 of FANCD2 impairs the response to ionizing radiation whereas disruption of Lys561 results in hypersensitivity to MMC. However, the ATM-dependent response to ionizing radiations is not all mediated through the interaction with FANCD2, since FANCD2-deficient cells are only partially sensitive to ionizing radiation (Taniguchi et al., 2002) when compared with AT cells.
|
More recently, it has been shown that cells from FA B and FA D1 patients have biallelic mutations in BRCA2 and FA D1 cells express truncated BRCA2 protein. Therefore, BRCA2 is in fact a FA gene. Interestingly, wild-type BRCA2 cDNA functionally complements MMC hypersensitivity in FA D1 cells and BRCA1- and BRCA2-deficient cells are hypersensitive to MMC (Howlett et al., 2002
). This result is consistent with the fact that mice with truncating mutations in BRCA2 are not lethal and express a FA-like phenotype (Connor et al., 1997). |
DNA repair downstream of FANCD2 |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
The downstream function(s) of the FA proteins is still a mystery. Although the exact molecular defect is not known, FA cells are directly or indirectly deficient in DNA repair as they show abnormal rearrangements associated with V(D)J recombination (Smith et al., 1998
) and impaired fidelity in blunt DNA end joining (Escarceller et al., 1997, 1998). A deficient DNA end joining activity in extracts from FA fibroblasts has also been reported. This defect is observed only when the FA complex is not present and seems to be distinct from the DNA-PK/Ku-dependent non-homologous DNA end joinig (NHEJ) pathway (Lundberg et al., 2001).
Basic research on the role of BRCA1 and BRCA2 led to the finding that both proteins interact with the recombination protein Rad51 and are involved in DSB repair by homologous recombination (HR) (Patel et al., 1998; Scully, 2001), however, BRCA-deficient cells are not deficient in NHEJ (Patel et al., 1998). Interestingly, the base excision repair system in BRCA2-/- cells exhibits a reduced rate of DNA ligation in both the single nucleotide insertion and PCNA-dependent pathways, indicating a role for BRCA2 in the ligation of DNA strand breaks (Bogliolo et al., 2000). Cells deficient in HR repair, such as BRCA1- and BRCA2-deficient cells, are very sensitive to DNA crosslinking agents, indicating that HR is important in the processing of DNA crosslinks (Dronkert and Kanaar, 2001). All these observations lead to the conclusion that BRCA proteins are involved in HR repair of DSBs induced by ionizing radiation or MMC (Scully, 2001). Since BRCA1 is required for FANCD2 foci formation during S phase, it is conceivable that the FA pathway is involved in DSB repair by HR during S phase. Consistent with this, it has been reported that FA fibroblasts are hyper-recombinogenic (Thyagarajan and Campbell, 1997) and FANCD2 forms foci at meiotic recombination sites in mice (Garcia-Higuera et al., 2001).
FA knockout mice and patients with FA or FA carriers are not known to be at risk of breast cancer and BRCA1 mutations have not been associated with AML or oral cancers, suggesting that BRCA1 is not an integral part of the FA pathway (Joenje and Arwert, 2001). According to these authors, it is more likely that BRCA1 functions to facilitate the FA downstream reactions, possibly in combination with other BRCA1-associated DNA repair proteins. One of these downstream proteins could be BRCA2 or other BASC-associated proteins (Futaki and Liu, 2001).
In parallel with the above interactions, ATM not only phosphorylates FANCD2 but also an increasing number of proteins involved in DSB repair by both HR and NHEJ (Kastan and Lim, 2000). Intriguingly, some of the few FA D2 patients identified to date have AT features such as lymphoma and immunodeficiency (Taniguchi et al., 2002). It will be interesting to know which, if any, are the modulating roles of the FA proteins in the ATM-dependent phosphorylation cascade in response to DNA damage.
Finally, it has been reported that the FA cellular phenotype can be complemented by overexpression of thioredoxin cDNA with an added nuclear localization signal. This implies that thioredoxin is lacking in the nuclei of FA cells (Kontou et al., 2002). Since thioredoxin is known to be the reactive cofactor of ribonucleotide reductase, its deficiency reduces the availability of deoxyribonucleotides in DNA replication and repair processes, resulting in chromosomal instability (Kontou et al., 2002).
Further genetic and molecular biology studies are obviously required to dissect the exact DNA repair defect(s) in FA and the specific roles of each of the FA proteins.
|
Telomere dysfunction in Fanconi anaemia |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
FANCD2 foci have been observed on autosomal telomeres in mice diplonema, suggesting that the FA pathway might be involved in telomere length maintenance. Telomeres play important roles in genome stability and in maintaining the individuality of linear eukaryotic chromosomes. Telomeres are constituted of long arrays of TTAGGG sequences in association with a number of telomeric proteins. The telomeres protect the end of the chromosomes, preventing fusion (Blasco et al., 1997
). One of the major telomere-binding factors preventing chromosome end fusions is the TTAGGG repeat factor 2, TRF2 (Van Steensel et al., 1998). The telomeres of dividing cells are progressively shortened due to the impossibility of DNA polymerases replicating the terminal part of the lagging DNA strand, in addition to other chromosome-specific telomere shortening factors (Surrallés et al., 1999). Telomere integrity is also crucial for organism viability and, in particular, for normal functionality of the haematopoietic system (Lee et al., 1998; Herrera et al., 1999). Consistent with this, an apparent accelerated shortening of telomeres has been reported in FA patients (Ball et al., 1998; Leteurtre et al., 1999; Hanson et al., 2001; Callén et al., 2002a). However, the relationship between telomere shortening and haematological status in FA patients remains controversial (Callén et al., 2002b). Shortened telomeres have also been reported in similar haematopoietic syndromes, such as acquired aplastic anaemia (Boultwood et al., 1997), and myelodysplastic syndromes (Brummendorf et al., 2001) and in AT (Metcalfe et al., 1996). Consistent with impaired telomeres in FA, we recently reported a TRF2-independent increase in chromosome end fusions in FA (Figure 5; Callén et al., 2002a).
|
Resembling AT (Hande et al., 2001
), we observed the presence of extra-chromosomic telomeric DNA and a high frequency of excess telomeric signals per cell in FA, suggesting intensive breakage at telomeric sequences in FA cells (Callén et al., 2002a). This observation, together with the fact that telomere shortening in FA occurred concurrently in both chromosome arms at a similar magnitude (Figure 6), suggests that telomere erosion in FA is caused by a higher rate of breakage at TTAGGG sequences in vivo in differentiated cells, in addition to mere replicative shortening during lymphocyte proliferation (Callén et al., 2002a).
|
Consistent with telomere breakage as an alternative mechanism of telomere shortening, recent investigations have shown accumulation of breaks at telomeres after oxidative stress and subsequent induction of 8-oxo-7,8-dihydro-2'-deoxyguanosine (8-oxodG) at telomeric sequences in human fibroblasts in vitro (Oikawa and Kawanishi, 1999
; Von Zglinicki et al., 2000) and it is known that FA cells are highly impaired in their response to oxidative stress and present high levels of 8-oxodG (see below). Therefore, it is tempting to speculate that the telomere breakage in FA is causally related to the impaired response to oxidative stress and an accumulation of 8-oxodG in telomeric DNA. A possible explanation for the high frequency of end fusion in FA is that breakage at telomeres could result in loss of the protective end-capping structures (Callén et al., 2002a).
It is interesting that the Rad50/Mre11/NBS1 complex and other proteins involved in DSB repair, such as Ku80 and DNA-PK, are also implicated in telomere biology in normal mammalian cells (Lansdorp, 2000). NBS1, the product of the gene mutated in NBS, binds to telomeres in association with Rad50 and MRE11 only during S phase, suggesting a role of NBS1 in telomere replication or in protecting the telomeres during replication (Zhu et al., 2000). FANCD2 forms nuclear foci only in S phase in normal untreated cells, resembling NBS1. However, it is not known whether FANCD2 also binds to telomeres during S phase. Since ATM phosphorylates both FANCD2, BRCA1 and NBS1, the similarity of telomere abnormalities between AT and FA is very interesting. Thus, one of the functional consequences of the increasingly evident molecular links between AT, FA and NBS would be a defect in telomere maintenance.
|
Fanconi anaemia, chromatin remodelling and transcriptional regulation |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
Recent reports have uncovered a molecular crosstalk between the FA pathway and chromatin remodelling and transcriptional activity. Chromatin remodelling and transcription regulatory functions have been described for BRCA1 and BRCA2, specifically for the C-terminal domain of BRCA1 (Scully, 2001
). In addition, BRCA1 and BRCA2 are required for the transcription-coupled repair of 8-oxodG in human cells (Le Page et al., 2000) and high levels of 8-oxodG have been reported in FA (see below). The C-terminal 20 amino acids of FANCD2 contains a highly acidic HMG-like domain, suggesting a possible mechanism for its chromatin association (Garcia-Higuera et al., 2001). Other chromatin modifier factors interacting with FA proteins are BRG1, as a subunit of the SWI/SNF complex (Otsuki et al., 2001), and histone acetyltransferases (Reuter et al., 2000). Strikingly, BRCA2 itself is a histone acetyltransferase (Siddique et al., 1998) and BRCA1 is also associated with a human SWI/SNF-related complex, linking chromatin remodeling to breast cancer (Bochar et al., 2000).
It has been proposed that FANCC may be linked to a transcriptional repression pathway involved in chromatin remodelling through interaction with FAZF (Hoatlin et al., 1999). It is also known that the FA proteins bind to chromatin and nuclear matrix but are excluded from condensed mitotic chromosomes (Qiao et al., 2001). These findings would imply that the action of the FA proteins is modulated by chromatin remodelling and transcriptional factors (Callén et al., 2002c). We have recently analysed chromosome breakage in the largest constitutively heterochromatic region in the human genome, the 1q12 band, in lymphocytes from FA patients, carriers and healthy controls after treatment with MMC and DEB to conclude that, unlike the overall genome, the sensitivity of this band to the chromosome-breaking activity of crosslinking agents is independent of a functional FA pathway. Thus, the action of the FA pathway is unevenly distributed through the human genome (Callén et al., 2002c).
All these data taken together would be indicative of a role of transcriptional activity and chromatin remodelling in the FA pathway or vice versa. There is a body of evidence that transcription and chromatin remodelling play important modulating roles in other DNA repair systems, such as nucleotide excision repair (Surrallés et al., 2002). Since transcription takes place in the same substrate as repair, replication and recombination, it is therefore not surprising that these processes are physically and functionally connected (Aguilera, 2002). Future experiments will probably uncover what the functional significance of the interconnection between the FA pathway and chromatin remodelling and transcriptional factors is.
|
Oxidative stress and cytoplasmic roles of Fanconi anaemia proteins |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
Oxygen radicals generated during the reduction of O2 can attack DNA bases or deoxyribose residues to produce damaged bases or strand breaks. Oxygen radicals can also oxidize lipid and protein molecules and generate intermediates that can react with DNA forming adducts. All these kinds of lesions can lead to mutation or apoptosis (reviewed in Marnett, 2000
). Over the years many studies have implicated endogenous reactive oxygen species (ROS) in the pathogenesis of FA. An abnormal sensitivity to oxygen of FA cells is suggested by the effects of oxygen tension on the levels of chromosome aberration in FA lymphocytes (Joenje et al., 1981; Joenje and Oostra, 1983) and by the observation that FA fibroblasts grow better at lower oxygen tensions than FA fibroblasts cultivated at 20% oxygen (Saito et al., 1993).
Increased formation of 8-oxodG can be detected in FA lymphoblasts (Takeuchi and Marimoto, 1993) and in vivo there is a statistically significant increase in the levels of 8-oxodG in the DNA of lymphocytes from healthy heterozygotes and affected members of FA families (Degan et al., 1995). 8-oxodG is the most common and mutagenic of the various oxidative DNA lesions. If left unrepaired, the outcome of this lesion is G
T transversions, which are frequently found in tumour-relevant genes (Loft and Poulsen, 1996). It has also been demonstrated that MMC and DEB have a redox-dependent mechanism of toxicity (Belcourt et al., 1996; Clarke et al., 1997; Vlachodimitropoulos et al., 1997; Korkina et al., 2000; Pagano, 2000) and that overexpression of thioredoxin cDNA in FA fibroblasts can abolish the effects of the two crosslinking agents (Ruppitsch et al., 1998). Moreover, ROS scavenging agents such as vitamin E can reduce the percentage of spontaneous chromosome aberrations in FA lymphocytes (Pincheira et al., 2001). Taken together with the evidence that ROS can act also as signalling molecules, altering redox-sensitive kinases and transcription factors (Adler et al., 1999), and that ubiquitin-conjugating enzyme are also influenced by the redox status of the cell (Obin et al., 1998), these studies strongly point to an important role for oxidative damage in the FA phenotype.
Strikingly, neither a clear impairment of the repair of 8-oxodG nor a particular susceptibility to oxidative damage-inducers such as H2O2 (Rey et al., 1993; Zhen et al., 1993; Lackinger et al., 1998; Will et al., 1998; Zunino et al., 2001) was observed in FA cell lines which, however, may have been made oxygen-resistant after the immortalization procedure. In fact, the loss of O2 sensitivity in transformed cells has been recognized as a general phenomenon, not confined to FA cell lines (Saito et al., 1995). Computational analysis of the FA complementation group A protein suggested that it contains a peroxidase domain and that FA proteins may be part of a general mechanism that protects cells from oxidative damage (Mian and Moser, 1998). This is also controversial, since recent findings cast doubts on the hypothesis that FANCA has peroxidase activity (Ren and Youssoufian, 2001).
At least one of the FA proteins that also has a cytoplasmic localization, FANCC, influences apoptotic pathways, cell cycle control and response to oxidative stress. Although FANCC protein can be found in the nucleus, an estimated 90% of FANCC resides in the cytoplasm (Yamashita et al., 1994; Youssoufian, 1994; Hoatlin et al., 1998, 1999). Moreover, there is evidence that FANCC binds to a number of cytosolic proteins, suggesting that FANCC, besides its role in the FA pathway of resistance to crosslinking agents, may play other roles in cell cycle control, protein transport, survival, signal transduction and regulation of detoxification (Pang et al., 2001a). Consistent with this, FANCC was shown to interact with the mitotic cyclin-dependent kinase cdc2 (Kupfer et al., 1997), the chaperones GRP94 (Hoshino et al., 1998) and HSP70 (Pang et al., 2001b), the signal transducer and activator of transcription STAT1 (Pang et al., 2000), FAZF, a member of the BTB/POZ family of transcriptional repressor proteins (Dai et al., 2002), NADPH:cytochrome P450 reductase (RED) (Kruyt et al., 1998) and glutathione S-transferase P1-1 (GSTP1) (Cumming et al., 2001).
The cytochromes P450 are a superfamily of haemoproteins that catalyse the oxidation of a number of endogenous and xenobiotic substances. P450-dependent metabolism requires two components, P450 and RED, with RED shuttling electrons from NADPH to P450. In FA C cells microsomal detoxification is abnormal, indicating a role for FANCC protein as a negative regulator of RED activity. An elevated activity of RED could lead to the generation of ROS, affecting cell viability (Kruyt et al., 1998).
FANCC also plays an important redox-regulator and anti-apoptotic role in haematopoiesis by ensuring the survival of haematopoietic progenitor cells through an interaction with GSTP1 (Cumming et al., 2001). GSTP1 is an enzyme that catalyses, by conjugation with GSH, the detoxification of xenobiotics and by-products of oxidative stress. FANCC exerts its function by preventing the formation of inactivating disulphide bonds within GSTP1, acting like a redox regulator of GSTP1. The role of FANCC as a regulator of proteins like RED and GSTP1, which are involved in the metabolism of xenobiotics and ROS, is consistent with the phenotype of FA (Cumming et al., 2001). These findings indicate that FANCC is multifunctional and that it probably has structurally separate functional domains for the nuclear damage function and for the cytoplasmic ROS-regulating and anti-apoptotic functions. In this regard, the study by Pang and co-workers showing how the crosslinker resistance function of FANCC depends on structural elements that differ from those required for the cytokine signalling functions of FANCC is particularly interesting (Pang et al., 2001a).
Recent findings have also indicated a novel interaction between FANCG and cytochrome P450 2E1 (CYP2E1) (Futaki et al., 2002). CYP2E1 is a member of the P450 superfamily and is associated with the production of reactive oxygen intermediates and the bioactivation of carcinogens. High constitutive levels of CYP2E1 generating high levels of ROS were found in a FA G lymphoblast cell line, whereas complementation of the FA G line with wild-type FANCG cDNA was associated with decreased CYP2E1. These findings suggested that the interaction of FANCG with CYP2E1 might alter redox metabolism and increase DNA oxidation.
All these data taken together may indicate a dual role for FANCA, FANCC and FANCG: protection of the cells against oxidative DNA damage in the cytoplasm and involvement in the nuclear response to DNA crosslinks (Futaki et al., 2002).
|
Fanconi anaemia proteins are involved in cell cycle regulation |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
The cellular responses to induced DNA damage include activation of cell cycle checkpoints that delay progression of cells through the cell cycle (Hartwell and Kastan, 1994
; Elledge, 1996). Ionizing radiation-induced checkpoints, for instance, are active at the transition from the G1 to the S phase, in the S phase and at the transition from the G2 phase to mitosis (G2/M). The mechanisms responsible for initiating these checkpoints facilitate the maintenance of the integrity of the genome, because they ensure that damaged DNA is neither replicated nor segregated into the daughter cells. The regulatory network of proteins involved in cell cycle checkpoints has been the focus of numerous studies, and in mammalian cells the protein ATM plays a central role in this network (Kastan and Lim, 2000; Abraham, 2001). In FA it is possible to observe accumulation of cells in the G2/M phase of the cell cycle, especially after MMC treatment (Dutrillaux et al., 1982; Kubbies et al., 1985). For instance, FA C lymphoblasts have both a spontaneous and MMC- or DEB-induced increase in the proportion of cells in the G2 fraction (Seyschab et al., 1993) and this behaviour can be corrected by the introduction of wild-type FANCC cDNA, resulting in a cell cycle phenotype similar to that of the wild-type (Walsh et al., 1994; Kruyt et al., 1996; Kupfer and D’Andrea, 1996).
It has also been demonstrated that oxygen tension can modulate the proliferation of diploid normal fibroblasts, inducing growth arrest in G2/M (Balin et al., 1978). Such a phenomenon in FA cells has also been correlated with their anomalous sensitivity to oxygen. In fact, such an effect is remarkably reduced when the FA cells are grown in low oxygen concentrations (Schindler and Hoehn, 1988; Hoehn et al., 1989; Seyschab et al., 1993).
Studies with caffeine, a drug that induces the cells to enter mitosis irrespective of DNA damage, showed that the accumulation in G2/M in FA is completely abolished after caffeine treatment, suggesting that the presence of DNA damage is the primary cause of the cell cycle alterations in FA (Pincheira et al., 1988; Seyschab et al., 1994). Therefore, it has been proposed that there is a defect in FA occurring earlier in the cell cycle so that many lesions reach G2 phase (Kupfer and D’Andrea, 1996). Indeed, recent results indicate that in contrast to normal cells, FA cells could be inefficient in arresting S phase cell cycle progression in response to lesions induced by crosslinking agents (Sala-Trepat et al., 2000) and that DNA damage-resistant DNA synthesis, which has been known for many years as a hallmark of AT cells following exposure to ionizing radiation (Scott et al., 1994), also occurs in FA cells following treatment with a DNA crosslinking agent (Centurion et al., 2000).
G2/M accumulation is enhanced in cells lacking the ionizing radiation-induced S phase checkpoint (ATM, NBS-1-/- and BRCA-1-/-) (Xu et al., 2002). Since the product of FANCD2 interacts with BRCA1 (see above), a role at least for FANCD2 in the control of cell cycle checkpoints is possible. Dr A.D’Andrea’s group has recently shown that FANCD2 cells have a defect in the ionizing radiation-induced S phase checkpoint and show radio-resistant DNA synthesis similar to AT fibroblasts (Taniguchi et al., 2002). After ionizing radiation, the product of the ATM gene phosphorylates the FANCD2 protein and this modification is required to activate an ionizing radiation-induced S phase checkpoint, but not for resistance to MMC, as explained before. In contrast, FA C cells have a normal S phase checkpoint, indicating that only FANCD2 has a role in the S phase checkpoint and that the monoubiquitination of FANCD2 is not necessary for regulation of the cell cycle after DNA damage (Taniguchi et al., 2002).
|
Abnormal apoptosis in Fanconi anaemia |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
The involvement of the FA genes in the control of apoptosis would explain one of the more important clinical symptoms of this pathology: a deficiency of haematopoiesis. In fact, the induction and repression of apoptosis under the control of several cytokines play a key role in the maturation of the progenitor cells of the bone marrow. It has been generally observed that FA lymphoblasts spontaneously enter apoptosis in vitro more frequently than normal cells (Ridet et al., 1997
). Several laboratories have studied the induction of p53-mediated apoptosis in FA cell lines after exposure to MMC and ionizing radiation, but the results are somewhat controversial. Rosselli et al.(1995) observed defective induction of p53, but other studies reported normal p53 induction in FA cells (Kruyt et al., 1996; Ridet et al., 1997).
An alternative line of evidence on the possible role of FANCC in the regulation of apoptosis comes from studies with growth factor-dependent cell lines. These studies have shown that constitutive expression of the FANCC cDNA protects, temporarily, mouse and human cell lines from apoptosis caused by growth factor withdrawal (Cumming et al., 1996). Moreover, mutations in the FANCC gene lead to an enhancement of interferon-
(IFN-)-induced apoptosis in haematopoietic cells at doses that have no effect on normal haematopoietic stem cells (Rathbun et al., 1997) and haematopoietic progenitors from FANCC transgenic mice were up to 10-fold less sensitive to the cytolytic effect of fas ligation (Wang et al., 1998). Furthermore, tumour necrosis factor 
(TNF-) is overexpressed in FA patients (Rosselli et al., 1994) and haematopoietic stem cells from FANCC-/- knockout mice and from FA C patients are hypersensitive to IFN- and TNF-. This hypersensitivity results, in part, from the capacity of these cytokines to prime the fas pathway to apoptosis and activation of the caspase 8 and caspase 3 family (Whitney et al., 1996; Rathbun et al., 1997, 2000; Haneline et al., 1998). Thus, the fas pathway seems to be up-regulated, at least in FA C cells. High constitutive expression of the IFN--inducible genes, the IFN-stimulated gene factor 3 subunit (ISGF3 ), IFN regulatory factor 1 (IRF-1) and the cyclin-dependent kinase inhibitor p21 (WAF1), were found in FA C mutant lymphoblastoid cell lines, low density bone marrow cells and murine embryonic fibroblasts (Fagerlie et al., 2001).
Although most of the known effects of IFN- are thought to be transduced through signal transducer and activator of transcription 1 (STAT1) activation (Schindler and Darnell, 1995), STAT1 signalling is paradoxically suppressed in FA C cells (Pang et al., 2000). Several studies have tried to determine the role and partners of FANCC in the regulation of this apoptotic pathway. The IFN-inducible, double-stranded RNA-dependent protein kinase (PKR) is inducible by both IFN- and TNF and influences fas activity and activates caspases 3 and 8. PKR is excessively activated in FA C cells. This aberrant activation state is suppressed by expression of the normal FANCC gene (Pang et al., 2001c). In ‘pull-down’ and co-immunoprecipitation experiments FANCC protein was shown to interact and act in concert with the molecular chaperone Hsp70 to prevent apoptosis in haematopoietic cells exposed to IFN- and TNF- (Pang et al., 2001b). The FAZF protein member of the BTB/POZ protein family also interacts with FANCC. FAZF is expressed in primary haematopoietic CD34+ progenitor cells, increases during early proliferation and is down-regulated during terminal differentiation in both erythroid and myeloid lineages. Overexpression of FAZF modulates a G1 phase cell cycle arrest followed by increased apoptosis (Dai et al., 2002). Thus, at least FANCC seems to have not only an indirect role in apoptosis, interacting with proteins involved in the maintenance of the redox state of the cell (Kruyt et al., 1998; Cumming et al., 2001), but also a more direct role in the fas-mediated apoptotic pathway.
Recent studies suggest that FANCA and FANCG could also be involved in the regulation of apoptosis. Treatment of normal HeLa fibroblasts with TNF- induces FANCG protein expression and phosphorylation. FANCA is induced concurrently with FANCG and the FANCA/FANCG complex is increased in the nucleus following TNF- treatment (Futaki et al., 2001). Also, the peculiar metabolic regulation in FA cells could explain both the defective apoptosis and susceptibility to oxidative stress of FA cells. The ability of FA cells to sustain metabolic insults interfering with energy production and balance may be linked with the pathological manifestations of the disease, including the susceptibility to AML (Bogliolo et al., 2002).
The involvement of the various FA proteins in the apoptotic pathway could explain the defects in haemopoiesis that are common to all FA patients, regardless of their complementation group.
|
Concluding remarks |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
The main conclusion that arises from the present review is that the FA pathway has become a central node in a cytoplasmic and nuclear network of genome stability and tumour suppressor pathways (Figure 7
). FA proteins most probably play crucial roles in DNA repair, telomere stability, chromatin remodelling, transcriptional regulation, detoxification of ROS, cell cycle control and apoptosis. Resolving the complexity of these pathways, the molecular basis of the FA phenotype and the role of the FA pathway in cancer is a huge scientific challenge for the years to come. FA mouse models will offer a versatile system for dissecting the FA pathways in vivo (Wong and Buchwald, 2002). These studies will hopefully clarify the complicated physiology of FA and provide insights into the mechanism of bone marrow failure and cancer in FA patients. In addition, studies of gene therapy in mouse models will provide important information which could lead to a cure for this life-threatening disease (Gush et al., 2000; Río et al., 2002).
|
|
Acknowledgments |
|---|
Fanconi Anemia research in our Group is supported in part by the Fanconi Anemia Research Fund Inc. (Oregon, USA), the Spanish Ministry of Health and Consumption (FIS, 99/1214), and the Commission of the European Union (Euratom, F1S5-1999-00071). M.B. is supported by a long-term post-doctoral Marie Curie fellowship awarded by the Commission of the European Union. E.C. is supported by a pre-doctoral fellowship awarded by the Universitat Autònoma de Barcelona (UAB). J.S. is supported by a ‘Ramón y Cajal’ project entitled ‘Genome stability and DNA repair’ co-financed by the Spanish Ministry of Science and Technology and the UAB.
|
Notes |
|---|
1 To whom correspondence should be addressed. Tel: +34 93 581 18 30; Fax: +34 93 581 23 87; Email: jordi.surralles{at}uab.es
|
References |
|---|
|
TopAbstractIntroductionFanconi anemia is genetically...All FA proteins participate...DNA repair downstream of...Telomere dysfunction in Fanconi...Fanconi anaemia, chromatin...Oxidative stress and cytoplasmic...Fanconi anaemia proteins are...Abnormal apoptosis in Fanconi...Concluding remarksReferences |
|---|
-
Abraham,R.T. (2001) Cell cycle checkpoint signaling through the ATM and ATR kinases. Genes Dev., 15, 2177–2196.
Adler,V., Yin,Z., Tew,K.D. and Ronai,Z. (1999) Role of redox potential and reactive oxygen species in stress signaling. Oncogene, 18, 6104–6111.[ISI][Medline]
Aguilera,A. (2002) The connection between transcription and genomic instability. EMBO J., 21, 195–201.[ISI][Medline]
Alter,B.P. (1996) Fanconi’s anemia and malignancies. Am. J. Hematol., 53, 99–110.[ISI][Medline]
Auerbach,A.D., Buchwald,M. and Joenje,H. (1997) Fanconi anaemia. In Vogelstein,B. and Kinzler,K.W. (eds), The Genetic Basis of Human Cancer. McGraw-Hill, New York, NY.