The genomic organization of the Fanconi anemia group A (FAA) gene.
ABSTRACT Fanconi anemia (FA) is a genetically heterogenous disease involving at least five genes on the basis of complementation analysis (FAA to FAE). The FAA gene has been recently isolated by two independent approaches, positional and functional cloning. In the present study we describe the genomic structure of the FAA gene. The gene contains 43 exons spanning approximately 80 kb as determined by the alignment of four cosmids and the fine localization of the first and the last exons in restriction fragments of these clones. Exons range from 34 to 188 bp. All but three of the splice sites were consistent with the ag-gt rule. We also describe three alternative splicing events in cDNA clones that result in the loss of exon 37, a 23-bp deletion at the 5' end of exon 41, and a GCAG insertion at the 3' portion also in exon 41. Sequence analysis of the 5' region upstream of the putative transcription start site showed no obvious TATA and CAAT boxes, but did show a GC-rich region, typical of housekeeping genes. Knowledge of the structure of the FAA gene will provide an invaluable resource for the discovery of mutations in the gene that accounts for about 60-66% of FA patients.
- Genes & Development - GENE DEVELOP. 01/1991; 5(4):670-682.
- [show abstract] [hide abstract]
ABSTRACT: Fanconi anaemia (FA) is an autosomal recessive disease characterised by genetic heterogeneity, with at least five complementation groups (FA-A to FA-E). The FAC gene has been cloned and localised to 9q22.3. The most frequent defective gene, FAA, was recently mapped to chromosome 16q24.3, in a region of 10 cM between D16S498 and the telomere. Eleven FA-A and 16 unclassified Italian families were analysed by microsatellite markers. To define the localisation of the FAA locus further, microsatellites were analysed at 16q24. All the families were consistent with linkage, the highest lod score being observed with D16S1320. Evidence for common haplotypes was obtained in two genetic isolates from the Brenta basin and the Naples region. Autozygosity mapping and haplotype analysis suggest that the FAA locus is distal to D16S305.Human Genetics 02/1997; 99(1):93-7. · 4.63 Impact Factor
- [show abstract] [hide abstract]
ABSTRACT: Cell fusion studies using lymphoblastoid cell lines from Fanconi anaemia (FA) patients have identified five complementation groups (FA-A to FA-E) among European FA patients. In Italy, of the 45 FA families referred to the Italian Registry of Fanconi Anaemia (RIAF), 15 took part in a project for the identification of complementation groups. Since three immortalized lymphoblast lines were resistant to a cross-linking agent, we analysed only 12 patients by complementation analysis and found that 11 belong to complementation group A. Four and seven families came from two geographic clusters in the Veneto and Campania regions, respectively, which are thought to consist of aggregates of related families in reproductive isolation. The clinical characteristics of the patients showed both intra- and interfamilial heterogeneity, although overall the disease had a relatively mild course. Since the populations in both Veneto and Campania are likely to represent genetic isolates, our finding predicts linkage disequilibrium for markers flanking the FAA gene. DNAs from these FA families may thus be utilized for positional cloning of this gene through haplotype disequilibrium mapping.Human Genetics 06/1996; 97(5):599-603. · 4.63 Impact Factor
GE NOMICS 41, 309±314 (1997)
ARTICLE NO. GE974675
The Genomic Organization of the Fanconi
Anemia Group A (FAA) Gene
LEONARDA IANZANO,* MARIA D'APOLITO,* MARTA CENTRA,* MARIA SAVINO,* ORNA LEVRAN,²
ARLEEN D. AUERBACH,² ANNE-MARIE CLETON-JANSEN,³ NORMAN A. DOGGETT,§ JAN C. PRONK,Ø
ALEX J. TIPPING,? RACHEL A. GIBSON,? CHRISTOPHER G. MATHEW,? SCOTT A. WHITMORE,**
SINOULA APOSTOLOU,** DAVID F. CALLEN,** LEOPOLDO ZELANTE,* AND ANNA SAVOIA*,1
*Servizio di Genetica Medica, IRCCS-Ospedale CSS, I-71013 San Giovanni Rotondo, Italy; ² Laboratory of Human Genetics and
Hematology, The Rockefeller University, 1230 York Avenue, New York, New York 10021-6399; ³ Department of Pathology,
University of Leiden, 2300 RA Leiden, The Netherlands; §Center for Human Genome Studies, Los Alamos National Laboratory,
Los Alamos, New Mexico 87545;ØDepartment of Human Genetics, Free University, Amsterdam, 1081 HV Amsterdam,
The Netherlands; ?Division of Medical and Molecular Genetics UMDS, 8th Floor Guy's Tower, Guy's Hospital, London SE1 9RT,
United Kingdom; and **Department of Cytogenetics and Molecular Genetics, Adelaide Women's and Children's Hospital,
North Adelaide, South Australia 5006, Australia
Received November 26, 1996; accepted January 30, 1997
tients show spontaneous chromosome breakageand in-
creased levels of chromosomal aberrations induced by
cross-linking agents such as diepoxybutane and mito-
mycin C (Auerbach, 1993). FA is genetically heteroge-
neous with at least ® ve complementation groups (FA-
A toFA-E) (Strathdee et al., 1992a; J oenje et al., 1995),
each group presumably corresponding to a separate
disease gene (Buchwald, 1995). FAC, which maps to
chromosome 9q22.33, was the ® rst gene to be cloned
(Strathdee et al., 1992b). Recently, the FAA gene was
identi® ed by two independent approaches: positional
cloning (FAB Consortium, 1996) and functional cloning
(Lo Ten Foe et al., 1996). The FAA gene has an open
reading frame of 4365 bp encoding a protein of 1455
amino acids and is localized on chromosome 16q24.3
between D16S3121 and D16S303 (Pronk et al., 1995;
FAB Consortium, 1996). FAA and FAC have nosigni® -
cant homology, not only to any other proteins but also
toeach other, and may therefore represent elements of
a new pathway(s) involved in the prevention of DNA
damage or cell cycle regulation. The prevalence of the
FA-A subtypes has been estimated to be 60±66% as
determined by complementation analysis and homozy-
gosity mapping (Buchwald, 1995; Gschwend et al.,
1996), with a particularly high prevalence in Italy (Sa-
voia et al., 1996, 1997). Mutations in the FAA gene
have so far been described only in a few patients who,
in most cases, carry deletions of one or more exons
(FAB Consortium, 1996; Lo Ten Foe et al., 1996).
Identi® cation of mutations will be most rapidly ac-
complished by screening for mutations on genomic
DNA. Accordingly, knowledge of the structure of the
human FAA gene is important to future research in
FA. To date, only the boundaries of the exons involved
in some mRNA deletions and the position of trapped
exons have been described (FAB Consortium, 1996). In
F anconi anemia (F A) is a genetically heterogenous
disease involving at least ®ve genes on the basis of
complementation analysis (F AA to F AE ). T he F AA gene
has been recently isolated by two independent ap-
proaches, positional and functional cloning. In the
present study we describe the genomic structure of
the F AA gene. T he gene contains 43 exons spanning
approximately 80 kb as determined by the alignment
of four cosmids and the ®ne localization of the ®rst
and the last exons in restriction fragments of these
clones. E xons range from 34 to 188 bp. All but three of
the splice sites were consistent with the ag-gt rule. We
also describe three alternative splicing events in cDNA
clones that result in the loss of exon 37, a 23-bp deletion
at the 5? end of exon 41, and a GCAG insertion at the
3? portion also in exon 41. Sequence analysis of the 5?
region upstream of the putative transcription start
site showed no obvious T AT A and CAAT boxes, but did
show a GC-rich region, typical of housekeeping genes.
K nowledge of the structure of the F AA gene will pro-
vide an invaluable resource for the discovery of muta-
tions in the gene that accounts for about 60±66% of F A
? 1997 Academic Press
Fanconi anemia (FA) is an autosomal recessive dis-
ease characterized by progressive pancytopenia, con-
genital malformations, andan increased predisposition
to neoplasia (Butturini et al., 1994). Cells from FA pa-
Sequence data from this article have been deposited with the
EMBL Data Library under Accession Nos. Z83067±Z83095 and
1To whom correspondence should be addressed. Telephone: /39
882 410825. Fax: /39 882 411616. E-mail: email@example.com.
Copyright ? 1997 by Academic Press
All rights of reproduction in any form reserved.
IANZANO ET AL.
addition, only three cosmids from a contig covering a
region of chromosome16q24.3 containing theFAA gene
have been reported (FAB Consortium, 1996; S. A.
Whitmore et al., in preparation). In this paper, we de-
scribethecompletestructureof theFAA geneincluding
the sequence of the 5? region. The gene contains 43
exons and each exon/intron boundary is de® ned. This
information will provide an excellent resource for fur-
ther discovery of mutations and for the development of
diagnostic tests in families carrying known mutations.
gene and overlapped c431F1, was also analyzed
Genomic organization. The exons and at least 60
bp of each ¯anking intron were obtained by sequencing
cosmid DNAs with speci® c oligonucleotides for theFAA
cDNA. Sequence analysis showed only an additional C
at position 5026 in the3? untranslated region (3?UTR),
when compared tothesequenceof theexons from geno-
mic clones to those of the cDNA (FAB Consortium,
1996; the A of the ATG of the initiator Met codon is
referred as nucleotide /1). The two variants 1501G/A
and 2151G/T, which distinguish the cDNAs obtained
by positional cloning and functional complementation,
were con® rmed on genomic DNA. They cause amino
acid substitutions G/S501 and M7/I717 and represent
polymorphisms as detected in 100 chromosomes (data
The gene consists of 43 exons ranging between 34
and 188 bp (Table 1). Splice junction boundaries agree
with published consensus sequences (Shapiroand Sen-
apathy, 1987) for donor and acceptor splicing sites. All
but three of the splice donor sites start with gt. In
the donor site of introns 16, 37, and 41 the normally
invariant gt dinucleotide is replaced by gc. Most of the
spliceacceptor sites endin cag (79%), whiletheremain-
der end in tag (17%) and aag (4%). Thesplicesites were
scored according to Shapiro and Senapathy (1987):
scores for donor sites vary from 68 to100; those for the
gc sites are 71 and 76. The acceptor sites score from
65 to 96 (Table 1).
Introns 4, 9, 12, 13, 16, 19, 24, 25, and 34 were
small, at 126, 372, 402, 390, 88, 92, 304, 141, and 142
bp, respectively. The genomic DNA containing exons
39 to 43 and the 3?UTR was entirely sequenced. The
3?UTR was not interrupted by any additional se-
quence, and the last four introns were 440, 188, 155,
and 173 bp in size. Sequence analysis with FASTA of
the regions ¯anking the exons revealed homologies
with Alu elements in introns 2, 6, 15, 26, 29, 31,
is homologous to yf14a03 and yh09a04, two overlap-
ping cDNA clones, and a directly selected cDNA, LL64
(FAB Consortium, 1996). Sequence analysis of these
ESTs revealed an out-of-frame deletion of 139 bp corre-
sponding to exon 37 of the FAA gene in yf14a03, a
deletion of the ® rst 23 bp of exon 41 in yh09A04, and,
in both cDNA clones, a GCAG insertion at the 3? end
also of exon 41. Sequence analysis of exon 41 and its
¯anking regions identi® ed cryptic sites for both ac-
ceptor and donor splice sites (Fig. 2). The 23-bp TCT-
TCTCTCCTACTTCCATGAAG deletion is likely due to
the recognition of an alternative 5? acceptor sequence
localized 23 bp from the start of exon 41. Similarly, the
GCAG insertion at the 3? end of exon 41 may be due
totheuseof thedonor siteAGgtgcaa (score65) alterna-
tive to the AGgcaggt site (score 71), which leads to the
correct cleavage of the intron. The 23-bp deletion or
MATERIALS AND METHODS
Isolation of genomic clones.
representing a 10-fold coverage of human chromosome 16 were
screened with cDNAs, ESTs, and STSs that mapped physically to
the FAA critical region. Cosmids were assembled into contigs based
on their restriction fragment pattern, and overlaps were con® rmed
by hybridization of individual fragments. Contigs were extended by
rescreening the high-density grids with cosmid end probes.
Cycle sequencing. Sequencing of exons and exon/intron bound-
aries, and in some cases of whole introns, was carried out on comids
c444B9, c431F1, c352A12, and c361H2. DNA was denatured in 2 M
NaOH and 2 mM EDTA for 30 min at 37?C. Sequence analysis was
performed with 10 mg of cosmid DNA using a ¯uorescence-labeled
dideoxy-nucleotide termination method (Dye terminator) in an auto-
mated DNA sequencer 373A (ABI). The exon/intron boundaries were
determined using speci® c oligonucleotides derived from the cDNA
sequence. The nucleotide sequences of exons and their ¯anking re-
gions have been submitted tothe EMBL Data Bank under Accession
Nos. Z83067 to Z83095 and Z83151.
Restriction endonuclease analysis of FAA clones.
ization of the cosmids was ® rst carried out by restriction endonucle-
ase digestion analysis with XbaI, XbaI±NotI, EcoRI, EcoRI±XbaI,
BamHI, and BamHI±XbaI, followed by agarose gel electrophoresis.
The sizes of the digested fragments were determined before their
transfer to membranes. Southern blots were hybridized with six dif-
ferent probes representing various portions of the full-length FAA
cDNA sequence: (I) 18/709, (II) 597/1359, (III) 1226/2014, (IV) 1901/
2830, (V) 2779/3840, and (VI) 3752/4484 (in this paper, the A of the
ATG of the initiator Met codon is nucleotide /1). Tocharacterize the
5? region of the FAA gene a BamHI fragment of 5.8 kb from cosmid
c444B9 containing exon 1 was subcloned into pBluescript and se-
High-density arrayed cosmid grids
The 3? region of the gene
length was obtained by screening a gridded chromo-
some16 cosmid library (Stallings et al., 1992; Longmire
et al., 1993). Cosmids c431F1, c352A12, and c361H2
have already been described as genomic clones con-
taining ESTs and trapped exons of the FAA gene (FAB
Consortium, 1996). In particular, cosmid c431F1 con-
tained the trapped exons ET6.1, ET6.6, and ET6.4,
which correspond to exon 7, exon 14, and exons 18 to
20, respectively. From the overlapping region of
c352A12 and c431F1 were trapped ET2 and ET1,
known to be exon 21 and exon 22. The shared portion
of c352A12 and c361H2 contained ET19 and yh09a04,
which are exons 31±32 and 41±43, respectively. To
characterize the genomic structure of FAA, an addi-
tional cosmid c444B9, which included the 5? end of the
A cosmid contig of about 650 kb in
STRUCTURE OF THE FAA GENE
theGCAG insertion aloneor thepresenceof both create
a frameshift and a downstream stop codon.
Localization of the ® rst and last exons.
double digestions of cosmids with restriction enzymes,
such as NotI±XbaI, XbaI, XbaI±EcoRI, EcoRI, EcoRI±
BamHI, BamHI, and BamHI±XbaI, were performed.
Fragments containing the exons were identi® ed by hy-
bridization with six different probes representing vari-
ous portions of the cDNA sequence (see Materials and
Methods). Data from restriction analysis and hybrid-
ization were utilized to generate a partial restriction
map of theFAA gene(Fig. 1). Exon 1 was identi® ed in a
0.7-kb EcoRI±BamHI fragment of cosmid c444B9 and
exons 39 to 43 in a 4.0-kb XbaI fragment of both
c353A12 and c361H2 cosmids. Thus, the human FAA
gene spans a region of 80 kb, as estimated from restric-
tion analysis of the genomic clones.
Characterization of the 5? region.
ment of 5.8 kb containing the 0.7-kb EcoRI±BamHI
fragment with exon 1 of the FAA gene was subcloned
and sequenced. We have determined the sequence of
750 bp encompassing exon 1. Theanalysis of the region
upstream of the putative transcription start site failed
to locate any TATA and CAAT boxes, but identi® ed a
highly GC-rich element (Fig. 3). The GC content was
72% for the region between 0516 (EcoRI restriction
site) and the putative transcription start site. Inspec-
tion of the sequence with Proscan (Version 1.7) in the
5? region of the FAA gene revealed a consensus se-
quence for transcription factor EGR-1 (Caoet al., 1990)
overlapping toa sequencefor AP-2 (Williams andTjian,
1991), two potential Sp1 sites (Schmidt et al., 1989), a
GCF-binding sequence (Kageyama and Pastan, 1989),
and a CTF/NF-1 recognition element (Gounari et al.,
1990). Comparison of the 5? region of the FAA gene to
that of the FAC (Savoia et al., 1995) did not show any
homology. However, both promoters share Alu repeats
within 1000 bp from the respective putative transcrip-
tion start sites.
A BamHI frag-
The present study describes the structural organiza-
tion of the FAA gene. The gene consists of 43 exons and
spans a region of approximately 80 kb. The published
cDNA sequences werecon® rmedon genomicDNA, with
the exception of an extra C in the 3?UTR. The strategy
of direct cycle sequencing of cosmids using speci® c
cDNA primers proved to be an effective way of de-
termining the exon structure of the gene, avoiding se-
quential hybridization and subcloning of small restric-
tion fragments. Sequence ¯anking splice junctions
have been determined for all 43 FAA exons. The 5?
donor and3? acceptor sites at thesplicejunctions corre-
lated well with published consensus sequences (Sha-
piroand Senapathy, 1987). All but three of the 5? donor
sites started with the dinucleotide gt. The nonconsen-
sus 5? splice junction sequence, gc, was found in the
F IG. 1.
Contig of cosmids c444B9, c431F1, c352A12, and c361H2 containing the FAA gene. The localizations of exon 1 and exons 39 to 43 in
fragments BamHI, 5.8 kb, and XbaI, 4.0 kb, respectively, are indicated. The symbols for restriction endonucleases are as follows: EcoRI, E; BamHI,
B; and XbaI, X.
IANZANO ET AL.
T ABL E 1
Alignment of the Boundaries of E xons and Introns
ExonSize5?-Donor siteScore3?-Acceptor siteScoreExon
TTC tga / 3? UTR
donor site of exons 16, 37, and 41 (Table 1). In vitro
mutagenesis studies have shown that, among the dif-
ferent substitutions of the gt at the donor signal, gc is
the only one that allows an accurate removal of the
intron, although more slowly. The cleavage is deter-
mined by the5? splice region as a wholevia thecomple-
mentarity to the 5? end of snRNA (Aebi et al., 1987).
Since only a few of the thousands of splice sequences
F IG. 2.
by capital letters. Consensus sequences for normal (underlined) and cryptic (boxed) splice sites are indicated. Scores for splice sites are in
Cryptic splice sites at both 5? and 3? ends of exon 41. Intron sequences are represented by lowercase letters and cDNA sequences
STRUCTURE OF THE FAA GENE
F IG. 3.
Nucleotide position /1 is assigned to the A of the ATG of the initiator Met codon. Exon 1 is represented by capital letters. Putative
recognition sites for Sp1, EGR-1, AP-2, GCF, and CTF/NF-1 are boxed. The sequence upstream of the asterisk is homologous to an Alu
Nucleotide sequence of the 0.7-kb EcoRI±BamHI fragment containing exon 1 and the 5? ¯anking region of the FAA gene.
that have been identi® ed have nonconsensus gc donor
sites (J ackson, 1991), it was surprising toidentify three
of them in the FAA gene. In the partial cDNA clones,
yf14a03 and yh09a04, both the skipping of exon 37 and
the alternative splicing at the 3? end of exon 41 involve
a gc site. Although the scores for these donor sites were
76 and71, respectively, it is likely that a de® cient corre-
spondence to the overall 5? splice region accounts for
the alternative splicing observed at these exons. Other
factors, such as unstable secondary structures of DNA,
might also be responsible for defective cleavage of in-
trons 37 and 41 (Balvay et al., 1993). Since the alterna-
tive splicing can be correlated to the presence of either
nonconsensus donor sitegc or cryptic signals, thealter-
native cleavages are not likely artifacts. Consistent
with this, the GCAG insertion was in fact found in
two independent ESTs. The question of whether the
different transcripts have a biological function may
lead to insights into the role of the FAA protein.
The putative regulatory regions are relatively GC
rich and do not have classical TATA and CAAT boxes.
Thus, the human FAA promoter has structural fea-
tures in common with those of several housekeeping
genes and is consistent with the wide expression pat-
tern of the mRNA in human tissues (FAB Consortium,
1996; LoTen Foe et al., 1996). Nosigni® cant homology
of the putative regulatory sequence of the FAA gene to
the FAC promoter has been found (Savoia et al., 1995).
The 5? regions of both genes share a GC-rich sequence,
and there are Alu repeats upstream of the putative
transcription start site. However, a more precise char-
acterization of thetranscription start site(s) anda func-
tional analysis of thepromoter by a reporter geneassay
and identi® cation of the trans-acting factors will pro-
vide advances in the knowledgeof the expression of the
In FA at least ® ve genes are involved in the patho-
genesis of the disease, but only FAA and FAC have
IANZANO ET AL.
Rooimans, M. A., Schroeder-Kurth, T., Wegner, R.-D., Gille,
J . J . P., Buchwald, M., and Arwert, F. (1995). Classi® cation of Fan-
coni anemia patients by complementation analysis: Evidence for
a ® fth genetic subtype. Blood 86: 2156±2160.
Kageyama, R., and Pastan, I. (1989). Molecular cloning and charac-
terization of a human DNA binding factor that represses transcrip-
tion. Cell 59: 815±825.
Longmire, J . L., Brown, N. C., Meincke, L. J ., Campbell, M. L., Al-
bright, K. L., Fawcett, J . J ., Campbell, E. W., Moyzis, R. K., Hilde-
brand, C. E., Evans, G. A., and Deaven, L. L. (1993). Construction
and characterization of partial digest DNA libraries made from
¯ow-sorted human chromosome 16. Genet. Anal. Tech. Appl. 10:
Lo Ten Foe, J ., Rooimans, A. A., Bosnoyan-Collins, L., Alon, N.,
Wijker, M., Parker, L., Lightfoot, J ., Carreau, M., Callen, D. F.,
Savoia, A., Cheng, N. C., van Berkel, C. G. M., Strunk, M. H. P.,
Gille, J . J . P., Pals, G., Kruyt, F. A. E., Pronk, J . C., Arwert, F.,
Buchwald, M., and J oenje, H. (1996). Expression cloning of a cDNA
for the major Fanconi anemia gene, FAA. Nature Genet. 14: 320±
Pronk, J . C., Gibson, R. A., Savoia, A., Wijker, M., Morgan, N. V.,
Melchionda, S., Ford, D., Temtamy, S., Ortega, J . J ., J ansen, S.,
Havenga, C., Cohn, R. J ., de Ravel, T. J ., Roberts, I., Westerveld,
A., Easton, D. F., J oenje, H., Mathew, C. G., and Arwert, F. (1995).
Localization of the Fanconi anemia complementation group A gene
tochromosome 16q24.3 by linkage analysis and allelic association.
Nature Genet. 11: 338±340.
Savoia, A., Centra, M., Ianzano, L., de Cillis, G. P., Zelante, L., and
Buchwald, M. (1995). Characterization of the 5? region of the Fan-
coni anaemia group C (FACC) gene. Hum. Mol. Genet. 4: 1321±
Savoia, A., Zatterale, A., Del Principe, D., and J oenje, H. (1996).
Fanconi anaemia in Italy: High prevalence of complementation
group A in two geographic clusters. Hum. Genet. 97: 599±603.
Savoia, A., Piemontese, M. R., Savino, M., Zatterale, A., Pronk, J .,
Arwert, F., J oenje, H., Ramenghi, U., Dagna-Bricarelli, F.,
Dallapiccola, B., and Zelante, L. (1997). Linkage analysis of Fan-
coni anaemia in Italy and mapping of the complementation group
A gene. Hum. Genet. 99: 93±97.
Schmidt, M. C., Zhou, Q., and Berk, A. J . (1989). Sp1 activates tran-
scription without enhancing DNA-binding activity of the TATA
box factor. Mol. Cell. Biol. 9: 3299±3307.
Shapiro, M. B., and Senapathy, P. (1987). RNA splice junctions of
different classes of eukaryotes: Sequence statistics and functional
implications in geneexpression. Nucleic Acids Res. 15: 7155±7174.
Stallings, R. L., Doggett, N. A., Callen, D., Apostolou, S., Chen, L. Z.,
Nancarrow, J . K., Whitmore, S. A., Harris, P., Michison, H., Breun-
ing, M., Sarich, J ., Fickett, J ., Cinkosky, M., Torney, D. C., Hilde-
brand, C. E., and Moyzis, R. K. (1992). Evaluation of a cosmid con-
tig physical map of human chromosome 16. Genomics 13: 1031±
Strathdee, C. A., Duncan, A. M. V., and Buchwald, M. (1992a). Evi-
dence for at least four Fanconi anaemia genes including FACC on
chromosome 9. Nature Genet. 1: 196±198.
Strathdee, C. A., Gavish, H., Shannon, W. R., and Buchwald, M.
(1992b). Cloning of cDNAs for Fanconi's anaemia by functional
complementation. Nature 356: 763±767.
Williams, T., and Tjian, R. (1991). Analysis of the DNA-binding and
activation properties of the human transcription factor AP-2.
Genes Dev. 5: 670±682.
been identi® ed thus far. However, both represent ap-
proximately 70±75% of all FA patients, with FAA ac-
counting for 60±66% (Buchwald, 1995; Gschwend et
al., 1996). To date, seven different mutations of the
FAA gene have been identi® ed in patients. Five in-
volved deletions of the mRNA, and another, a splice
site mutation, caused the utilization of a downstream
cryptic splice site (FAB Consortium, 1996; Lo Ten Foe
et al., 1996). In all these cases, knowledge of the struc-
ture of the FAA gene will allow identi® cation of the
molecular defect at the genomic level.
This study was supported by grants from the Italian Ministry of
Health, Telethon±Italy (E.364), the Italian Association for Cancer
Research, the Medical Research Council (UK), the Fanconi Anaemia
Research Fund, the National Institutes of Health (U.S.A.) Grant
32987, theDutch Cancer Society, and theNational Health and Medi-
cal Research Council of Australia.
Aebi, M., Hornig, H., and Weissmann, C. (1987). 5? cleavage site in
eukaryotic pre-mRNA splicing is determined by the overall 5?
splice region, not by the conserved 5? GU. Cell 50: 237±246.
Auerbach, A. D. (1993). Fanconi anemia diagnosis and the diepoxy-
butane (DEB) test. Exp. Hematol. 21: 731±733.
Balvay, L., Libri, D., and Fiszman, M. Y. (1993). Pre-mRNA second-
ary structure and the regulation of splicing. Bioessays 15: 165±
Buchwald, M. (1995). Complementation groups: One or more per
gene? Nature Genet. 11: 228±230.
Butturini, A., Gale, R. P., Verlander, P. C., Adler-Brecher, B., Gillio,
A., and Auerbach, A. D. (1994). Hematologic abnormalities in Fan-
coni anemia: An international Fanconi anemia study. Blood 84:
Cao, X. M., Koski, R. A., Gashler, A., McKiernan, M., Morris, C. F.,
Gaffney, R., Hay, R. V., and Sukhatme, V. P. (1990). Identi® cation
and characterization of the Egr-1 gene product, a DNA-binding
zinc ® nger protein induced by differentiation and growth signals.
Mol. Cell Biol. 10: 1931±1939.
FAB Consortium. (1996). Positional cloning of the Fanconi anaemia
group A gene. Nature Genet. 14: 324±328.
Gounari, F., De Francesco, R., Schmitt, J ., van der Vliet, P., Cortese,
R., and Stunnenberg, H. (1990). Amino-terminal domain of NF1
binds to DNA as a dimer and activates adenovirus DNA replica-
tion. EMBO J . 9: 559±566.
Gschwend, M., Levran, O., Kruglyak, L., Ranade, K., Verlander,
P. C., Shen, S., Faure, S., Weissenbach, J ., Altay, C., Lander, E. S.,
Auerbach, A. D., and Botstein, D. (1996). A locus for Fanconi ane-
mia on 16q determined by homozygosity mapping. Am. J . Hum.
Genet. 59: 377±384.
J ackson, I. J . (1991). A reappraisal of non-consensus mRNA splice
sites. Nucleic Acids Res. 19: 3797±3798.
J oenje, H., Lo Ten Foe, J . R., Oostra, A. B., van Berkel, C. G. M.,