Molecular Cell, Vol. 7, 241–248, February, 2001, Copyright 2001 by Cell Press
Positional Cloning of a Novel
Fanconi Anemia Gene, FANCD2
a molecular complex with primarily nuclear localization
(Kupfer et al., 1997; Garcia-Higuera et al., 1999; Waisfisz
et al., 1999a). FANCC also localizes to the cytoplasm,
indicating a potential role in that compartment (Yous-
soufian, 1994, 1996). No functional domains are appar-
ent in the protein sequences of the FA genes cloned to
date, and the biochemical function(s) of the FA pathway
remains unknown. Interestingly, database searches re-
veal no strong homologs of the FANCA, C, E, F, and
G proteins in nonvertebrate species. FANCF has weak
homology of unknown significance to an E. coli RNA
binding protein (de Winter et al., 2000b). The formation
of the nuclear complex among the FANCA, FANCC, and
FANCG proteins is abnormal in all complementation
groups studied except in cell lines from the rare group
D (Yamashita et al., 1998; Garcia-Higuera et al., 1999;
Garcia-Higuera and D’Andrea, 1999; Waisfisz et al.,
1999a). Therefore, FANCD may define a function of the
FA pathway either upstream, downstream, or indepen-
dent of the action of the multiprotein FA complex. Here,
we report thepositional cloning of a genemutated in the
cloned FA genes, FANCD2 has highly conserved homo-
vation of the FA pathway in lower organisms.
Cynthia Timmers,*# Toshiyasu Taniguchi,§#
James Hejna,* Carol Reifsteck,* Lora Lucas,*
Donald Bruun,* Matthew Thayer,†Barbara Cox,*
Susan Olson,* Alan D. D’Andrea,§Robb Moses,*
and Markus Grompe*‡
*Department of Molecular and Medical Genetics
‡Department of Pediatrics
Oregon Health Sciences University
Portland, Oregon 97201
§Dana-Farber Cancer Institute
Harvard Medical School
Boston, Massachusetts 02115
Fanconi anemia (FA) is a genetic disease with birth
defects, bone marrow failure, and cancer susceptibil-
ity. To date, genes for five of the seven known comple-
mentation groups have been cloned. Complementa-
tion group D is heterogeneous, consisting of two
distinct genes, FANCD1 and FANCD2. Here we report
the positional cloning of FANCD2. The gene consists
of 44 exons, encodes a novel 1451 amino acid nuclear
protein, and has two protein isoforms. Similar to other
FA proteins, the FANCD2 protein has no known func-
tional domains, but unlike other known FA genes,
FANCD2 is highly conserved in A. thaliana, C. elegans,
and Drosophila. Retroviral transduction of the cloned
FANCD2 cDNA into FA-D2 cells resulted in functional
complementation of MMC sensitivity.
mentation group D because lymphoblasts from this pa-
tient failed to complement HSC62 (Strathdee et al.,
1992a), the reference cell line for group D (Jakobs et al.,
1996). In previous publications, we therefore referred to
the gene mutated in PD20 as FANCD (Whitney et al.,
1995; Jakobs et al., 1996, 1997; Hejna et al., 2000). How-
ever, new results described in this paper suggest that
the gene mutated in HSC62 (FANCD1) and the gene
mutated in PD20 (FANCD2) are different.
Fanconi anemia (FA) is an autosomal recessive disorder
characterized by progressive bone marrow failure, can-
cer predisposition, and multiple developmental defects
(Fanconi, 1967; Alter, 1993). Cells from FA patients dis-
play a characteristic hypersensitivity to agents that pro-
duce interstrand DNA cross-links, such as mitomycin C
(MMC) or diepoxybutane (DEB). Therefore, the func-
tion(s) of the FA proteins pertain to multiple important
erogeneous, and to date at least seven complementa-
tion groups (FANCA–G) have been identified by cell
fusion techniques (Joenje et al. 1997, 2000). This obser-
vation has resulted in the hypothesis that the FA genes
define a multicomponent pathway involved in cellular
responses to DNA cross-links (Garcia-Higuera et al.,
1999). Five of the FA genes (FANCA, C, E, F, and G)
have been cloned (Strathdee et al., 1992b; The Fanconi
Anaemia/Breast Cancer Consortium, 1996; Lo Ten Foe
et al., 1996; de Winter et al., 1998, 2000a, 2000b), and
the FANCA, C, and G proteins have been shown to form
Mutation Analysis of Positional cDNA Candidates
We previously reported the localization of the gene mu-
tated in PD20 to chromosome 3p using microcell-medi-
ated chromosome transfer into FA cell line PD20 (Whit-
ney et al., 1995). Detailed analysis of five microcell
hybrids that contained small overlapping deletions en-
compassing the locus narrowed the candidate region
for the FANCD2 gene to ?200 kb (Hejna et al., 2000).
Three candidate ESTs were localized in or near this
FANCD2 critical region (Hejna et al., 2000). Using 5? and
3? RACE to obtain full-length cDNAs, the genes were
sequenced andthe expression patternof eachwas ana-
lyzed by Northern blot. EST SGC34603 had ubiquitous
1992b; The Fanconi Anaemia/Breast Cancer Cosortium,
1996; Lo Ten Foe et al., 1996; de Winter et al., 1998,
2000b). Open reading frames were found for TIGR-
A004X28, AA609512, and SGC34603 and were 234, 531,
?To whom correspondence should be addressed (e-mail: grompem@
#These authors contributed equally to this work.
whereas only 3% (1/31) of control cDNA clones dis-
a frameshift and predicts a severely truncated protein
only 180 amino acids in length. The second alteration
was a paternally inherited missense change at position
1236 (R1236H). The segregation of the mutations in the
PD20 core family is depicted in Figure 3. These findings
suggested that SGC34603 is the FANCD2 gene.
The Protein Encoded by FANCD2 Is Absent in PD20
Tofurther confirmtheidentityof SGC34603asFANCD2,
blot analysis was performed (Figure 4). The specificity
of the antibody was shown by transient expression of
FANCD2 in PD20 cells (Figure 4c). In wild-type cells, this
antibody detected two bands (155 and 162 kDa) that we
call FANCD2-S and -L (best seen in Figure 4c). FANCD2
protein levels were markedly diminished in all MMC-
sensitive cell lines from patient PD20 (Figure 4a, lanes
cells from other complementation groups. Furthermore,
PD20 cells corrected by microcell-mediated transfer of
chromosome 3 also made normal amounts of protein
(Figure 4a, lane 3).
Figure 1. Northern Blot of FANCD2
Blots were probed with a full-length FANCD2 cDNA and exposed
for 24 hr. The four lanes on the far right are from human fetal tissues,
and the others are from human adults. The size markers are given
in kbp at the left.
and 4413 bp in length, respectively. All three were ana-
lyzed for mutations in PD20 cells by sequencing cloned
detected in TIGR-A004X28 and AA609512, five se-
quence changes were found in SGC34603 (Table 1).
Next, we determined the structure of the SGC34603
gene by using cDNA sequencing primers on BAC 177N7
from the critical region. Forty-four exons were discov-
ered, with the start codon localized in exon 2. The open
reading frame predicts a protein of 1451 amino acids
(Figure 2). Based on the genomic sequence information,
PCR primer pairs were designed, the exons containing
putative mutations were amplified, and allele-specific
assays were developed to screen the PD20 family as
well as 568 control chromosomes. Three of the alleles
were common polymorphisms; however, two changes
were not found in the controls and, thus, represented
potential mutations (Table 1). The first was a maternally
inherited A→G change at nt 376. In addition to changing
an amino acid (S126G), this alteration was associated
with missplicing and insertion of 13 bp from intron 5
into the mRNA. Forty-three of forty-three (100%) inde-
pendently cloned RT–PCR products with the maternal
mutation contained this insertion (data not shown),
Functional Complementation of FA-D2 Cells
with the FANCD2 cDNA
We next assessed the ability of the cloned FANCD2
cDNA to complement the MMC sensitivity of FA-D2
cells. The full-length FANCD2 cDNA was subcloned into
the retroviral expression vector, pMMP-puro, as pre-
viously described (Pulsipher et al., 1998). Retrovially
transduced PD20 cells expressed both isoforms of the
FANCD2 protein, FANCD2-S and FANCD2-L (Figure 4c).
corrected the MMC and DEB sensitivity of the cells (Ta-
ble 2, bottom). Furthermore, in an MMC kill curve assay,
retrovirally corrected PD20 cells were as resistant to
MMC as cells complemented by an intact human chro-
mosome 3 (data not shown). These results further sug-
gest that the cloned FANCD2 cDNA encodes the
FANCD2-S protein, which can be posttranslationally
modified to the FANCD2-L isoform.
Analysis of a Phenotypically Reverted PD20 Clone
We next generated additional evidence demonstrating
that the sequence variations in PD20 cells were not
functionally neutral polymorphisms. Toward this end,
we performed a molecular analysis of a revertant
lymphoblast clone (PD20-cl.1) from patient PD20 that
was no longer sensitive to MMC. Phenotypic reversion
and somatic mosaicism are frequent findings in FA and
have been associated with intragenic events such as
mitotic recombination or compensatory frameshifts (Lo
Ten Foe et al., 1997; Waisfisz et al., 1999b). Indeed,
?60% of maternally derived SGC34603 cDNAs had a
novel splice variant inserting 36 bp of intron 5 sequence
rather than the usually observed 13 bp (Figure 5). The
appearance of this in-frame splice variant correlated
with a de novo base change at position IVS5?6 from G
to A (Figure 5), and restoration of the correct reading
frame was confirmed by Western blot analysis (Figure
Table 1. FANCD2 Sequence Alterations
deletion of exon 17
aPD20 is heterozygous.
bVU008 is heterozygous.
Cloning of the FANCD2 Gene
Figure 2. Amino Acid Sequence of Human FANCD2 and Alignment with Fly and Plant Homologs
The human (top), fly (middle), and plant (bottom) FANCD2 protein sequences are shown. Hypothetical proteins from several nonvertebrate
eurkaryotes showed highly significant alignment scores using the BEAUTY algorithm (Worley et al., 1995). Black boxes indicate amino acid
identity, and gray boxes indicate similarity. The best alignment scores were observed with hypothetical proteins in Drosophila (p ? 8.4 ?
10?58, GenBank accession number AAF55806) and A. thaliana (p ? 9.4 ? 10?45, GenBank accession number B71413).
4a, lane 5). In contrast to all MMC-sensitive fibroblasts
and lymphoblasts from patient PD20, PD20-cl.1 pro-
duced readilydetectable amountsof FANCD2protein of
slightlyhigher molecularweightthanthe normalprotein.
Analysis of Cell Lines from Other “FANCD” Patients
The antibody was also used to screen additional FA
patient cell lines, including the reference cell line for FA
group D, HSC 62 (Strathdee et al., 1992a), and two other
Figure 3. Allele-Specific Assays for Mutation
Analysis of Two FANCD2 Families
The family pedigrees (a and d) and (b), (c),
(e), and (f) are vertically aligned such that the
corresponding mutationanalysis isbelow the
individual in question. (a)–(c) depict the PD20
family, and (d)–(f) depict the VU008 family. (b)
and (e) show the segregation of the maternal
mutations as detected by the creation of a
new MspI site (PD20) or DdeI site (VU008).
The paternally inherited mutations in both
gonucleotide hybridization (c and f).
Figure 4. WesternBlotAnalysisoftheFANCD2
Protein in Human Fanconi Anemia Cell Lines
Whole-cell lysates were generated from the
indicated fibroblast and lymphoblast lines.
Protein lysates (70 ?g) were probed directly
by immunoblotting with the anti-FANCD2
antiserum. The FANCD2 proteins (155 kDa
and 162 kDa) are indicated by arrows. Other
bands in the immunoblot are nonspecific.
(a) Cell lines tested included wild-type cells
(lanes 1 and 7), PD20 fibroblasts (lane 2),
PD20 lymphoblasts (lane 4), revertant MMC-
resistant PD20 lymphoblasts (lanes 5 and 6),
broblasts (lane 3). Several other FA group D
cell lines were analyzed, including HSC62
(lane 8) and VU008 (lane 9). FA-A cells were
HSC72 (lane 10), FA-C cells were PD4 (lane
11), and FA-G cells were EUFA316 (lane 12).
(b) Identification of a third FANCD2 patient.
FANCD2 protein was readily detectable in
(c) Specificity of the antibody. PD20i cells
transiently transfected with a FANCD2 ex-
pression vector displayed both isoforms of
the FANCD2 protein (lane 4) in contrast to
empty vector controls (lane 3) and untrans-
fected PD20i cells (lane 2). In wild-type cells,
the endogenous FANCD2 protein (two iso-
body (lane 1).
cell lines identified as group D by the European Fanconi
Anemia Registry (EUFAR). VU008 did not express the
FANCD2 protein (Figure 4a, lane 9) and was found to
be a compound heterozygote with a missense and non-
sense mutation, both in exon 12. Neither mutation was
found on 370 control chromosomes (Table 1; Figure
3). The missense mutation appears to destabilize the
FANCD2 protein, as there is no detectable FANCD2 pro-
Table 2. Chromosome Breakage Analysis of Hybrids and Retrovirally Transduced Cells
Cell Line/HybridsDEB (ng/ml) MMC (ng/ml)Percentage of Cells with Radials Phenotype
VU423i ? chr. 3, clone 1
VU423i ? chr. 3, clone 2
VU423i ? chr. 3, clone 3
PD20i ? empty vector002
0 PD20i ? FANCD2 vector00
Groups of experiments are separated by line spaces. S, cross-linker sensitive; R, cross-linker resistant; i, immortal fibroblast line; p, primary
aCell viability at this concentration was too low to score for radial formation, indicating the exquisite sensitivity of primary fibroblasts to
interstrand DNA cross-links.
Cloning of the FANCD2 Gene
FANCD2 has an open reading frame of 4353 bp encod-
weight of 166 kDa. Database searches revealed no
known functional domains or motifs, with the exception
of a HMG-like domain at the carboxyl terminus (Bach-
varovand Moss,1991).Unlikepreviously clonedFApro-
teins, hypothetical proteins from several nonvertebrate
eurkaryotes showed highly significant alignment scores
C. elegans. The Drosophila homolog has 28% amino
acid identity and 50% similarity to FANCD2 (Figure 2).
The FANCD2 homologs also have no currently known
function or domains, and no functional studies have
been carried out in the respective species. No proteins
similar to FANCD2 were found in E. coli or S. cerevisiae.
The functional significance of the HMG-like domain is
logs do not contain a similar domain.
Although Fanconi anemia is a rare disease, the pleiotro-
of their wild-type function for diverse cellular processes
including genome stability, apoptosis, cell cycle con-
trol, and resistance to DNA cross-links (D’Andrea and
Grompe, 1997). At the organismal level, FA proteins are
involved in maintenance of hematopoietic and gonadal
of many different structures, including the skeleton and
urogenital systems (Alter, 1993; Whitney et al., 1996).
However, despite the interest of many laboratories in
this disorder and the cloning of several FA genes, little
progress has been made in the understanding of the
biochemical function(s) of the pathway. Similar to other
FA genes that have been cloned, the protein sequence
of the novel gene identified here does not shed light
onto the function of the pathway.
FANCD2 has some unusual properties that may pro-
vide some functional insights. First, FANCD2 has highly
conserved homologs in organisms that are readily ame-
nable to genetic studies, such as the identification of
suppressors. Because no homologs of FANCA, C, E,
and G have been identified in nonvertebrates, such ex-
periments have not been possible with the FA genes
described to date. In light of the functional connection
between BRCA1 and FANCD2 (Garcia-Higuera et al.,
2001 [this issue of Molecular Cell]), it is noteworthy that
BRCA1 does not appear to have a highly conserved
homolog in fruit flies. Perhaps the Drosophila FANCD2
homolog will be useful in the genetic dissection of this
DNA damage–response pathway. Second, as shown in
the accompagning paper, FANCD2 colocalizes and in-
teracts with the breast cancer protein, BRCA1 (Garcia-
Higuera et al., 2001), suggesting a connection between
the pathway and BRCA1-mediated DNA repair.
Our studies also show that Fanconi anemia group D
is genetically heterogenous, consisting of at least two
genes, FANCD1 and FANCD2. This conclusion is based
on the absence of mutations in some group D patients,
full complementation between different group D cell
lines in whole-cell fusions, and the failure of chromo-
Figure 5. Molecular Basis for the Reversion of PD20 Lymphoblasts
samples (right lane) yielded a single band of 114 bp, whereas PD20
of 13 bp of intronic sequence into the maternal allele. Reverted,
MMC-resistant lymphoblasts (middle lane) from PD20 revealed a
third, in-frame splice variant of 114 ? 36 bp.
(b) Schematic representation of splicing at the FANCD exon 5/intron
5 boundary. In wild-type cDNA, 100% of splice events occur at the
proper exon/intron boundary, whereas the maternal A→G mutation
(indicated by arrow) leads to aberrant splicing, also in 100%. In the
reverted cells, all cDNAs with the maternal mutation also had a
second sequence change (fat arrow) and showed a mixed splicing
pattern with insertion of either 13 bp (40% of mRNA) or 36 bp (60%
tein in lysates from VU008 cells. A third patient, PD733,
to complement PD20 in whole-cell fusions (data not
shown). RT–PCR showed the absence of exon 17 caus-
ing an internal deletion of the protein (Table 1). The
genomic mutations in this patient have not yet been
of FANCD2 protein in cell lysates derived from these
patients substantiates the identity of FANCD2 as an FA
In contrast, readily detectable amounts of both iso-
forms of the FANCD2 protein were found in HSC 62
(Figure 4a, lane 8) and VU423 (data not shown). cDNA
and genomic DNA from both cell lines were extensively
phenotype (Table 2, top). Microcell-mediated chromo-
some transfer of an intact human chromosome 3 into
breakage phenotype (Table 2, middle). Taken together,
is distinct from FANCD2, and that it is not located on
with primers MG979 5?-ACTGGACTGTGCCTACCCACTATG-3? and
MG984 5?-CCTGTGTGAGGATGAGCTCT-3?. Primers MG818 5?-AGA
GGTAGGGAAGGAAGCTAC-3? and MG813 5?-CCAAAGTCCACTT
CTTGAAG-3? were used for exon 37. Wild-type (5?-TTCTCCCGA
AGCTCAG-3? for R302W and 5?-TTTCTTCCGTGTGATGA-3? for
R1236H) and mutant (5?-TTCTCCCAAAGCTGAG-3? for R302W and
5?-TTTCTTCCATGTGATGA-3? for R1236H) oligonucleotides were
end-labeled with [?-32P]ATP and hybridized to dot-blotted target
PCR products as previously described (Wu et al., 1989). The VU008
nonsense mutation (Q320X) in exon 12 created a novel DdeI site.
The wild-type PCR product digests into a 117 and 71 bp product,
whereas the mutant allele yields three fragments of 56, 61, and 71
bp in length. PCR in all of the above assays was performed with 50
ng of genomic DNA for 37 cycles of 94?C for 25 s, 50?C for 25 s,
and 72?C for 35 s.
some 3 to correct these cells. Thus, the number of con-
firmed complementation groups has once again in-
creased from seven to eight (Joenje et al., 2000). It is of
interest to note that the original reference cell line for
tions in FANCD2 (PD20, VU202 [the sibling of VU008])
have some common properties that distinguish them
from other FA complementation groups. In these cells,
the nuclear complex between FANCA, G, and C forms
normally, whereas the complex does not form in groups
A, B, C, E, F, and G (Yamashita et al., 1998; Garcia-
Waisfisz et al., 1999a). Indeed, work described in the
accompanying paper by Garcia-Higuera et al. shows
that the FANCA, B, C, E, F, and G proteins are required
for posttranslational modification of FANCD2 in re-
sponse to DNA damage and cell cycle status (Garcia-
Higuera et al., 2001). In contrast, cells from group D1
patients are the only FA cells that display normal modifi-
cation of FANCD2, suggesting that FANCD1 acts in par-
allel with or downstream from FANCD2.
Generation of an Anti-FANCD2 Antiserum
A rabbit polyclonal antiserum against FANCD2 was generated using
a GST-FANCD2 (N-terminal) fusion protein as an antigen source. A
5? fragment was amplified by polymerase chain reaction (PCR) from
the full-length FANCD2 cDNA with the primers DF4 EcoRI (5?-AGCC
TCgaattcGTTTCCAAAAGAAGACTGTCA-3?) and DR816 Xh (5?- GGT
product of 841 bp encoding the amino-terminal 272 amino acids of
the FANCD2 polypeptide was digested with EcoRI/XhoI and sub-
cloned into the EcoRI/XhoI sites of the plasmid pGEX4T-1 (Phar-
macia). A GST-FANCD2 (N-terminal) fusion protein of the expected
size (54 kDa) was expressed in E. coli strain DH5?, purified over
glutathione S-Sepharose, and used to immunize a New Zealand
White rabbit. An FANCD2-specific immune antiserum was affinity
purified over an AminoLink Plus column (Pierce) loaded with GST
protein and over an AminoLink Plus column loaded with the GST-
FANCD2 (N-terminal) fusion protein.
Human adult and fetal multitissue mRNA blots were purchased from
Clontech (Palo Alto, CA). Blots were probed with32P-labeled DNA
from EST clone SGC34603. Standard hybridization and washing
the blot with an actin cDNA probe (data not shown).
Cells were lysed with 1? sample buffer (50 mM Tris-HCl [pH 6.8],
86 mM 2-mercaptoethanol, 2% sodium dodecyl sulfate [SDS]),
boiled for 5 min, and subjected to 7.5% polyacrylamide SDS gel
electrophoresis. After electrophoresis, proteins were transferred to
a nitrocellulose filter in transfer buffer (Tris 25 mM, glycine 200 mM)
at 400 mA at 4?C for 4 hr. The filter was blocked for 1 hr in 5%
nonfat milk in TBS (50 mM Tris-HCl [pH 8.0], 150 mM NaCl) and was
incubated in primary antibody (1:1000 dilution) in TBS plus 0.1%
(v/v) Tween 20 (TBS-T) overnight at 4?C . After extensive washing
in TBS-T, the filter was incubated in horseradish peroxidase anti-
for 1 hr at room temperature. After extensive washing in TBS-T,
enzyme-linked chemiluminescence (Amersham) was performed.
Total cellular RNA was reverse transcribed using a commercial kit
(GIBCO–BRL). The 5? end section of FANCD2 was amplified from
the resulting patient and control cDNA with a nested PCR protocol.
The first round was performed with primers MG471 5?-AATC
GAAAACTACGGGCG-3? and MG457 5?-GAGAACACATGAATGAA
CGC-3?. The PCR product from this round was diluted 1:50 for a
subsequent round using primers MG492 5?-GGCGACGGCTTCTCG
GAAGTAATTTAAG-3? and MG472 5?-AGCGGCAGGAGGTTTATG-3?.
The PCR conditions were as follows: 94?C for 3 min, 25 cycles
of 94?C for 45 s, 50?C for 45 s, 72?C for 3 min, and 5 min of 72?C at
the end. The 3? portion of the gene was amplified as described
above but with primers MG474 5?-TGGCGGCAGACAGAAGTG-3?
and MG475 5?-TGGCGGCAGACAGAAGTG-3?. The second round of
PCR was performed with MG491 5?-AGAGAGCCAACCTGAGCG
ATG-3? and MG476 5?-GTGCCAGACTCTGGTGGG-3?. The PCR
and sequenced using internal primers.
Cell Lines and Transfections
PD20i is an immortalized and PD733 a primary FA fibroblast cell
line generated by the Oregon Health Sciences Fanconi Anemia Cell
Repository (Jakobs et al., 1996). PD20 lymphoblasts were derived
from bone marrow samples. VU008 is a lymphoblast and VU423 a
fibroblast line generated by the European Fanconi Anemia Registry
(EUFAR). VU423i was an immortalized line derived by transfection
with SV40 T-antigen (Jakobs et al., 1996) and telomerase (Bodnar
et al., 1998). The other FA cell lines have been previously described
(Strathdee et al., 1992a; Joenje et al., 1997). Human fibroblasts were
cultured in ?MEM and 20% fetal calf serum. Transformed lympho-
blasts were cultured in RPMI 1640 supplemented with 15% heat-
inactivated fetal calf serum.
Allele-specific assays were performed in the PD20 family and 290
control samples (?580 chromosomes) The PD20 family is of mixed
Northern European descent, and VU008 is a Dutch family (Whitney
et al., 1995). Control DNA samples were from unrelated individuals
in CEPH families (n ? 95), samples from unrelated North American
families with either ectodermal dysplasia(n ? 95) or Fanconi anemia
a novel MspI restriction site. For genomic DNA, the assay involved
amplifying genomic DNA using the primers MG792 5?-AGGAGA
CACCCTTCCTATCC-3? located in exon 4 and MG803 5?- GAAGTTG
GCAAAACAGACTG-3?, which is in intron 5. The size of the PCR
product was 340 bp, yielding two fragments of 283 bp and 57 bp
upon MspI digestion if the mutation was present. For analysis of the
reverted cDNA clones, PCR was performed using primers MG924
5?-TGTCTTGTGAGCGTCTGCAGG-3? and MG753 5?-AGGTTTTGA
TAATGGCAGGC-3?. The paternal exon 37 mutation (R1236H) in
PD20 and exon 12 missense mutation (R302W) in VU008 were
tested by allele-specific oligonucleotide (ASO) hybridization (Wu
et al., 1989). For the exon 12 assay, genomic DNA was amplified
Whole-Cell and Microcell Fusions
For the whole-cell fusion experiments, a PD20 cell line (PD20i) resis-
tant to hygromycin B and deleted for the HPRT locus was used
(Jakobs et al., 1997). Controls included PD24 (primary fibroblasts
from affected sibling of PD20) and PD319i (Jakobs et al., 1997)
(immortal fibroblasts from a non-A, C, D, or G FA patient). Cells
(2.5 ? 105) from each cell line were mixed in a T25 flask and allowed
to recover for 24 hr. The cells were washed with serum-free medium
and then fused with 50% PEG for 1 min (Mercer and Schlegel, 1979).
After removal of the PEG, the cells were washed three times with
Cloning of the FANCD2 Gene
serum-free medium and allowed to recover overnight in complete
medium without selection. The next day, cells were split 1:10 into
selective medium containing 400 ?g/ml hygromycin B (Roche Mo-
lecular) and 1? HAT. After the selection was complete, hybrids were
passaged once and then analyzed as described below. Microcell-
mediated transfer of human chromosome 3 into VU423 cells was
performed as previously described by us (Whitney et al., 1995).
Garcia-Higuera, I., and D’Andrea, A.D. (1999). Regulated binding
of the Fanconi anemia proteins, FANCA and FANCC. Blood 93,
Garcia-Higuera, I., Kuang, Y., Naf, D., Wasik, J., and D’Andrea, A.D.
(1999). Fanconi anemia proteins FANCA, FANCC, and FANCG/
XRCC9 interact in a functional nuclear complex. Mol. Cell. Biol. 19,
Garcia-Higuera, I., Taniguchi, T., Ganesan, S., Meyn, S., Timmers,
C., Hejna, J., Grompe, M., and D’Andrea, A.D. (2001). Interaction of
the Fanconi anemia proteins and BRCA1 in a common pathway.
Mol. Cell 7, this issue, 249–262.
Hejna, J.A., Timmers, C.D., Reifsteck, C., Bruun, D.A., Lucas, L.W.,
Jakobs, P.M., Toth-Fejel, S., Unsworth, N., Clemens, S.L., Garcia,
D.K., et al. (2000). Localizationof the Fanconi anemia complementa-
tion group D gene to a 200-kb region on chromosome 3p25.3. Am.
J. Hum. Genet. 66, 1540–1551.
Jakobs, P.M., Sahaayaruban, P., Saito, H., Reifsteck, C., Olson, S.,
Joenje, H., Moses, R.E., and Grompe, M. (1996). Immortalization
of four new Fanconi anemia fibroblast cell lines by an improved
procedure. Somat. Cell. Mol. Genet. 22, 151–157.
and Grompe, M. (1997). Complementation group assignments in
Fanconi anemia fibroblast cell lines from North America. Somat.
Cell. Mol. Genet. 23, 1–7.
Rooimans, M.A., Ebell, W., van Weel, M., Pronk, J.C., Buchwald, M.,
and Arwert, F. (1997). Evidence for at least eight Fanconi anemia
genes. Am. J. Hum. Genet. 61, 940–944.
Joenje,H., Levitus,M.,Waisfisz, Q.,D’Andrea,A., Garcia-Higuera,I.,
Pearson, T., van Berkel, C.G., Rooimans, M.A., Morgan, N., Mathew,
C.G., and Arwert, F. (2000). Complementation analysis in Fanconi
anemia: assignment of the reference FA-H patient to group A. Am.
J. Hum. Genet. 67, 759–762.
Kupfer, G.M., Naf, D., Suliman, A., Pulsipher, M., and D’Andrea, A.D.
(1997). The Fanconi anaemia proteins, FAA and FAC, interact to
form a nuclear complex. Nat. Genet. 17, 487–490.
Lo Ten Foe, J.R., Rooimans, M.A., Bosnoyan-Collins, L., Alon, N.,
Wijker, M., Parker, L., Lightfoot, J., Carreau, M., Callen, D.F., Savoia,
A., et al. (1996). Expression cloning of a cDNA for the major Fanconi
anaemia gene, FAA. Nat. Genet. 14, 320–323.
Lo Ten Foe, J.R., Kwee, M.L., Rooimans, M.A., Oostra, A.B., Veer-
man, A.J., van Weel, M., Pauli,R.M., Shahidi, N.T., Dokal, I., Roberts,
I., et al. (1997). Somatic mosaicism in Fanconi anemia: molecular
basis and clinical significance. Eur. J. Hum. Genet. 5, 137–148.
Mercer, W.E., and Schlegel, R.A. (1979). Phytohemagglutinin en-
hancement of cell fusion reduces polyethylene glycol cytotoxicity.
Exp. Cell Res. 120, 417–421.
Naf, D., Kupfer, G.M., Suliman, A., Lambert, K., and D’Andrea, A.D.
(1998). Functional activity of the Fanconi anemia protein FAA re-
quires FAC binding and nuclear localization. Mol. Cell. Biol. 18,
Pulsipher, M., Kupfer, G.M., Naf, D., Suliman, A., Lee, J.S., Jakobs,
P., Grompe,M., Joenje,H., Sieff,C., Guinan, E.,et al.(1998). Subtyp-
ing analysis of Fanconi anemia by immunoblotting and retroviral
gene transfer. Mol. Med. 4, 468–479.
Strathdee, C.A., Duncan, A.M., and Buchwald, M. (1992a). Evidence
for at least four Fanconi anaemia genes including FACC on chromo-
some 9. Nat. Genet. 1, 196–198.
Strathdee, C.A., Gavish, H., Shannon, W.R., and Buchwald, M.
(1992b). Cloning of cDNAs for Fanconi’s anemia by functional com-
plementation. Nature 356, 763–767.
Waisfisz, Q., de Winter, J.P., Kruyt, F.A., de Groot, J., van der Weel,
L., Dijkmans, L.M., Zhi, Y., Arwert, F., Scheper, R.J., Youssoufian,
H., et al. (1999a). A physical complex of the Fanconi anemia proteins
FANCG/XRCC9 and FANCA. Proc Natl Acad Sci U S A 96, 10320–
Waisfisz, Q., Morgan, N.V., Savino, M., de Winter, J.P., van Berkel,
C.G., Hoatlin, M.E., Ianzano, L., Gibson, R.A., Arwert, F., Savoia, A.,
et al. (1999b). Spontaneous functional correction of homozygous
Retroviral Transduction of FA-D2 Cells
and Complementation Analysis
puro (Pulsipher et al., 1998). Retroviral supernatants were used to
were analyzed for MMC sensitivity by the crystal violet assay (Naf
et al., 1998).
Chromosome Breakage Analysis
Core Lab at OHSU (Portland, OR). For the analysis (Cohen et al.,
1982), cells were plated into T25flasks, allowed to recover, and then
treated with 300 ng/ml of DEB for two days. After treatment, the
KCl and 3:1 methanol:acetic acid. Slides were stained with Wright’s
stain, and 50–100 metaphases were scored for radials.
This work was supported by NHLBI program project grant
1PO1HL48546 (M. G., S. O., and R. E. M.) and R01HL52725 (A. D’A.).
and the European Fanconi Anemia Registry for making cell lines
VU008 and VU423 available. We also thank the Fanconi Anemia
Research Fund for their support.
Received August 14, 2000; revised December 11, 2000.
tol. 85, 9–14.
tion factor xUBF contains 5 tandemly repeated HMG homology
boxes. Nucleic Acids Res. 19, 2331–2335.
Bodnar, A.G., Ouellette, M., Frolkis, M., Holt, S.E., Chiu, C.P., Morin,
G.B., Harley, C.B., Shay, J.W., Lichtsteiner, S., and Wright, W.E.
(1998). Extension of life-span by introduction of telomerase into
normal human cells. Science 279, 349–352.
Cohen, M.M., Simpson, S.J., Honig, G.R., Maurer, H.S., Nicklas,
J.W., and Martin, A.O. (1982). The identification of Fanconi anemia
genotypes by clastogenic stress. Am. J. Hum. Genet. 34, 794–810.
D’Andrea, A.D., and Grompe, M. (1997). Molecular biology of Fan-
de Winter, J.P., Waisfisz, Q., Rooimans, M.A., van Berkel, C.G.,
Bosnoyan-Collins, L., Alon, N., Carreau, M., Bender, O., Demuth, I.,
Schindler, D., et al. (1998). The Fanconi anaemia group G gene
FANCG is identical with XRCC9. Nat. Genet. 20, 281–283.
de Winter, J.P., Leveille, F., van Berkel, C.G., Rooimans, M.A., van
Der Weel, L., Steltenpool, J., Demuth, I., Morgan, N.V., Alon, N.,
Bosnoyan-Collins,L., etal.(2000a). Isolationofa cDNArepresenting
the Fanconi anemia complementation group E gene. Am. J. Hum.
Genet. 67, 1306–1308.
de Winter, J.P., Rooimans, M.A., van Der Weel, L., van Berkel, C.G.,
Alon, N., Bosnoyan-Collins, L., de Groot, J., Zhi, Y., Waisfisz, Q.,
Pronk, J.C., et al. (2000b). The Fanconi anaemia gene FANCF en-
codes a novel protein with homologyto ROM. Nat. Genet. 24, 15–16.
Fanconi, G. (1967). Familial constitutional panmyelocytopathy, Fan-
coni’s anemia (F.A.) I. Clinical aspects. Semin. Hematol. 4, 233–240.
The Fanconi Anaemia/Breast Cancer Consortium. (1996). Positional
cloning of the Fanconi anaemia group A gene. Nat. Genet. 14,
Molecular Cell Download full-text
Fanconi anaemia alleles reveals novel mechanistic basis for reverse
mosaicism. Nat. Genet. 22, 379–383.
Whitney, M., Thayer, M., Reifsteck, C., Olson, S., Smith, L., Jakobs,
cell mediated chromosome transfer maps the Fanconi anaemia
group D gene to chromosome 3p. Nat. Genet. 11, 341–343.
Whitney, M.A., Royle, G., Low, M.J., Kelly, M.A., Axthelm, M.K.,
Reifsteck, C., Olson, S., Braun, R.E., Heinrich, M.C., Rathbun, R.K.,
gamma-interferon in mice with a targeted disruption of the Fanconi
anemia C gene. Blood 88, 49–58.
Worley, K.C., Wiese, B.A., and Smith, R.F. (1995). BEAUTY: an en-
hanced BLAST-based search tool that integrates multiple biological
information resources into sequence similarity search results. Ge-
nome Res 5, 173–184.
Wu, D.Y., Nozari, G., Schold, M., Conner, B.J., and Wallace, R.B.
(1989). Direct analysis of single nucleotide variation in human DNA
and RNA using in situ dot hybridization. DNA 8, 135–142.
Yamashita, T., Kupfer, G.M., Naf, D., Suliman, A., Joenje, H., Asano,
S., andD’Andrea, A.D.(1998). TheFanconi anemiapathway requires
FAA phosphorylation and FAA/FAC nuclear accumulation. Proc.
Natl. Acad. Sci. USA 95, 13085–13090.
Youssoufian, H. (1994). Localization of Fanconi anemia C protein to
the cytoplasm of mammalian cells. Proc. Natl. Acad. Sci. USA 91,
Youssoufian, H. (1996). Cytoplasmic localization of FAC is essential
for the correction of a prerepair defect in Fanconi anemia group C
cells. J. Clin. Invest. 97, 2003–2010.
GenBank Accession Numbers
The GenBank accession numbers for the candidate cDNAs are
G26488 for EST TIGR-A004X28, G26135 for the EST SGC34603/
FANCD2, and AA609512 for the third EST. The accession numbers
for the human FANCD2 exons/gene structure are AF273222–
for the FANCD2 homologs are AAF55806 (Drosophila), CAB10276
and AAD23659 (A. thaliana), and CAB63365 (C. elegans).