novel mutations and polymorphisms in the fanconi anemia group c gene

HUMAN MUTATION 8:140-148 (1996)

RESEARCH ARTICLE

Novel Mutations and Pslymorphisms in the Fanconi Anemia Group C Gene Rachel A. Gibson, Neil V. Morgan, Laura H. Goldstein, Ian C. Pearson, Ian P. Kesterton, Nicola J. Foot, Stander Jansen, Charmaine Havenga, Thomas Pearson, Thomy J. de Ravel, Richard J. Cohn, Isabel M. Marques, Inderjeet Dokal, Irene Roberts, Judith Marsh, Sarah Ball, R. David Milner, Juan C. Llerena, Jr., Elena Samochatova, Sheila P. Mohan, Pushpa Vasudevan, Farkondeh Bijandi, Atieh Hajianpour, Manuela Murer-Orlando, and Christopher G. Mathew. Division of Medical and Molecular Genetics UMDS, Guy’s Hospital, London, UK (R.A. G., N. V.M., L. H. G., 1. C. P., I. P. K., N.J.F., F.B., M.M-O., C.G.M.), Departments of Human Genetics and of Paediatrics, University of the Orange Free State Medical School, Bloemfontein, South Afnca (S.J., C. H., T. P.), Departments of Human Genetics and Paedianics, South Afncan Institute of Medical Research and the University of the Witwatersrand, South Africa (T.J. dR., R.J. C., I. M. M. ), Department of Haematology, Royal Postgraduate Medical School, Hammersmith Hospital, London, UK (I.D., I. R.), Division of Huematology, St. George’s Hospital Medical School, London, UK (I.M., S.B.), King Faisal Specialist Hospital and Research Centre, Riyadh, S a d Arabia (R. D. M. ) , Centro di Genetica Medica, IFFIFimr, Rio de Janeiro, Brazil (J. C. L. ) , Institute of Paedianic H m t o b g y , Ministvy of Health of Russia, Moscow, Russia (E. S. ), Institute of Child Health and Hospital for Children, Madras, India (S. P. M., P. V.), Childrens’ Hospital, Sunset Boulevard, L o s Angeles, California (A. H.); Fax: 441 -71 -955-4644

Communicated by Bruce A.J. Ponder

Fanconi anemia (FA) is an autosomal recessive disorder associated with hypersensitivity to DNA cross-linking agents and bone marrow failure. At least four complementation groups have been defined, and the FA group C gene (FAC) has been cloned. We have screened 76 unrelated FA patients of diverse ethnic and geographic origins and from unknown complementation groups for mutations in the FAC gene either by chemical cleavage mismatch analysis or by single-strand conformational polymorphism (SSCP). Five mutations were detected in four patients (5.3%), including two novel mutations (W22X and L496R). Nine polymorphisms were detected, seven of which have not been described previously (663A-*G, L190F, IVS6+30C+T, I312V, V449M, Q465R, and 1974-A). Six of the nine polymorphisms occurred in patients or controls from the Tswana or Sotho chiefdoms of South Africa and were not found in 50 unrelated European controls. Restriction site assays were established for all 8 pathogenic mutations identified in the FAC gene to date and used to screen a total of 94 unrelated FA patients. This identified only one other group C patient, who was homozygous for the mutation IVS4 + 4A+T. This study indicates that the proportion of FA patients from complementation group C is generally likely to be less than 10%. Guidelines for the selection of FA patients for FAC mutation screening are proposed. o 1996 WiIeyLiss, Inc.

KEY WORDS: Fanconi anemia, Group C, Mutation, Polymorphism

INTRODUCTION Fanconi anemia (FA) is a rare autosomal reces-

sive disorder characterised by diverse congenital abnormalities, progressive bone marrow failure, and a predisposition to acute myeloid leukemia (Gordon-Smith and Rutherford, 1991). Cells from patients with FA have a high level of spontaneous chromosomal aberrations (Schroeder et al., 1964) and are hypersensitive to bifunctional alkylating agents such as mitomycin C, diepoxybutane, and cisp latin (Sasaki and Tonomura, 1973; Auer- bach et al., 1989). Demonstration of increased

chromosome breakage is used to confirm the clinical diagnosis of FA.

The disorder is genetically heterogeneous, with at least four complementation groups (A-D) having been identified (Strathdee et al., 1992a). The gene that is defective in Fanconi anemia Complemen-

Received June 16, 1995; accepted September 7, 1995. *To whom reprint requestsicorrespondence should be ad-

dressed.

0 1996 WILEY-LISS. INC.

MUTATIONS IN THE FAC GENE 141

tation Group C (FAC) has been cloned by functional complementation (Strathdee et al., 1992b; for nomenclature see Lehmann et al., 1994). FAC has been localised to chromosome 9q22.3 by in situ hybridisation (Strathdee et al., 1992a) and placed on the genetic linkage map within a 5-cM interval on chromosome 9q, flanked by three highly poly- morphic microsatellite markers (Gibson et al., 1994). Its location has recently been further refined by physical mapping data (Morris and Reis, 1994). The genomic structure of the coding sequence of the FAC gene has also been characterised (Gibson et al., 1993a), which allows mutation screening of genomic DNA.

A total of 6 pathogenic mutations have been identified in the FAC gene to date, of which four (80322, Q13X, R185X, R548X) would result in premature termination of translation of the FAC polypeptide (Strathdee et al., 199213; Gibson et al., 1993b; Murer-Orlando et al., 1993; Verlander et al,, 1994). The other two are a splice-site mutation in intron 4 (IVS4 + 4A+T), which is the predominant mutation in FA patients of Ash- kenazi Jewish origin (Whimey et al., 1993), and a missense mutation, L554P (Strathdee et al., 1992a), which has been shown by site-directed mutagenesis to abolish the functional comple- menting activity of the FAC protein (Gavish et al., 1993). The pathogenic status of a further missense mutation, D195V (Verlander et al., 1994), remains to be established.

Several polymorphisms have been described in the FAC gene. Whitney et al. (1993) detected two variants, G139E and 2045A+C, in the 3’UTR in a FA group A cell line. G139E and another variant, S26F, were found not to segregate with the FA phenotype in affected families (Verlander et al., 1994).

We have screened the complete coding region of the FAC gene for mutations in an ethnically and geographically diverse panel of 76 unrelated FA patients either by SSCP (Orita et al., 1989) or by chemical cleavage mismatch analysis (CCM) (Cotton et al., 1988; Montandon et al., 1989), in order to determine the proportion of patients who are from group C and to further investigate the spectrum and location of mutations in this gene. These patients and a further 18 unrelated FA patients (94 in all) were then screened for the 6 known pathogenic mutations, and for the new mutations identified in this study. We detected five mutations and nine polymorphisms in this panel of patients, including two mutations and seven polymorphisms that had not been described previously.

MATERIALS AND METHODS Patients

Patient material was sent to the Regional Ge- netics Centre at Guy’s Hospital from clinical departments of Paediatrics and Haematology in the United Kingdom, South Africa, Saudi Arabia, Brazil, India, and Russia. The clinical diagnosis of FA was based on hypoplastic anaemia with or without associated congenital abnormalities. A total of 94 unrelated FA patients were included in the screen for known FAC mutations, after in- formed consent had been obtained. sufficient material was available from 76 of these patients to screen the entire FAC coding sequence for mutations. Their ethnic or geographic origins were as follows: South African European, 24; South Afri- can black, 9; Indian subcontinent, 21; British, 13; Russian, 3; Middle Eastern, 2; Spanish, 1; Greek, 1; Turkish, 1; and Brazilian, 1. DNA samples were also obtained from 48 healthy individuals from the Tswana (33) or Sotho (15) chiefdoms of South Africa (both chiefdoms originate from the same Bantu-speaking family), in order to establish the population frequency of polymorphisms originally detected in Tswana FA patients in this study. The frequency of polymorphisms was determined in 100 chromosomes from 27 unrelated European FA patients and 23 controls of British descent.

Cytogenetic Analysis

The diagnosis of FA was confirmed in all cases by the demonstration of increased sensitivity of cultured peripheral blood lymphocytes or skin fibroblasts to the clastogenic effect of diepoxybutane (DEB) or mitomycin C (MMC) (Auerbach et al., 1989), as previously described (Gibson et al., 1994).

DNAiRNA Extraction Genomic DNA was extracted from whole blood

collected in EDTA or from cultured lymphoblasts, fibroblasts or chorionic villi cells using the salt/ chloroform method (Mullenbach et al., 1989). RNA was extracted from lymphocytes or from cultured fibroblasts, lymphoblasts or chorionic villi cells, using the acid guanidinium thiocyanate-phenol-chloroform extraction method (Chomczynski and Sacchi, 1987) and stored at -20°C as an eth- anol precipitate.

Mutation Analysis Total RNA was reverse transcribed, amplified

(RT-PCR), and subjected to CCM as described

142 GIBSON ET AL.

TABLE 1. Restriction Site Assays Developed for All Eight Known FAC Mutationsa

Artificial (A) or Size of products (bp) Restriction natural (N) ( N size of products in normal)

Mutation Primer sequence enzyme restriction site ( M size of product in mutant) Q13X F - 5‘ CAGTAGATCTITCTTGTGATGAT 3’ BclI A N 227 + 23

R - 5’ ACCACAAGTCCCGATTCTGGG 3’

R - 5’ ACCACAAGTCCCGATTCTGGG 3’

R - 5’ GTITCCAAAGTGGAAGCCTGAGCC 3‘

R - 5’ TITCAAAAGTGATAAATATTAAGTAC 3’

R:- 5’ CCACTTATTACCCTGACA 3’

R:- 5’ CTCTCCTTGACTAGGATGCTG 3’

R:- 5’ ACTTACTCCACAAATGCGTGG 3’

R - 5’ ACTTACTCCACAAATGCGTGG 3’

M. 250

M. 187 + 17 + 17 AG322 F - 5’ ACCATITCCTTCAGTGCTGG 3’ Bsp1286I A N 129 + 22

M- 151 IVS4 + 4A+T F - 5‘ CTCATATACmCAGCACTCAG 3’ ScaI A N: 108 + 23

M 131 R185x F - 5’ GTCCTTAATTATGCATGGCTC 3’- NlaIII N N 1 0 5 + 17

M 74 + 31 + 17 L496R F - 5’ CCTAGAAGTATGTCTGTCCTG 3‘ HhaI N N 303

M:218 + 85 R548X F - 5’ GTTATGGTCCGTCCCTGGAC 3’ AvaI N N 2 3 1 + 133

M 364 L554P F - 5’ GTTATGGTCCGTCCCTGGAC 3’ BbvI N N 2 6 0 + 104

M 364

w22x F - 5’ TGGATGCAGAAGCTTTCTGGATG 3’ Fokl A N 2 0 4 + 17

The base that was modified to create a restriction site is underlined.

previously (Gibson et al., 199313). Where no RNA was available, individual exons of the FAC gene were amplified from genomic DNA as previously described (Gibson et al., 1993a), except that 2.5 p1 of 0132P dCTP (3,000 Ci/mmol) was incorporated into the PCR product, and exon 6 was divided into two overlapping fragments of 171 bp and 216 bp, using primer F: 5 ’GTCCTTAATTATGCAT- GGCTC3 ’ with R: 5 ’ CTGAGGTTCACGTC- CATGACAG 3‘, and primer F: 5’AGTG. GCGTCCCTGTCACGAGTTTG3 ’ with R: 5 ’CAACACACCACAGCCTTCTAAG 3 ’, respectively. For SSCP analysis, samples were dena- tured and electrophoresed on a 6% nondenaturing polyacrylamide gel (1 : 100 bisacrylamide/acryla- mide) containing 5% glycerol, for 18 hr at 4 W and 4”C, in 1 X TBE (0.089 M Tris, 0.089 M boric acid, 0.02 M EDTA pH 8.0) buffer. FAC sequences that contained chemical cleavage sites or that showed reproducible SSCP bandshifts were characterised by direct sequencing of both DNA strands of the PCR product from two independent ampli- fications (Green et al., 1989; Winship, 1989).

PCR Assays for Known FAC Mutations

Restriction site assays were developed for all six known pathogenic mutations and for the new mutations identified in this study, to allow rapid, non- radioactive analysis of these mutations. Primers were synthesised on an Applied Biosystems 391 PCRMATE DNA synthesizer using cyanoethyl phos- phoramidite chemistry. PCR was carried out in 25-pl reactions with 250 ng of genomic DNA, 10 ng/pl of each primer, 0.5 mM of each dNTP, and 1.5 units of Taq polymerase in a buffer containing

6.7 mM MgC1, (Roberts et al., 1992), unless oth- erwise stated. After an initial denaturation at 94°C for 5 min, samples were amplified for 30 cycles of denaturation at 94°C for 1 min, annealing at 60°C for 1 min (50°C for IVS4 + +AT), and extension at 72°C for 1 min, followed by a final 5-min extension at 72°C. PCR products were then digested with 5-10 units of the appropriate restriction enzyme for a minimum of 2 hr at 37°C. (50°C for BclI). The products were analysed on a horizontal 7% polyacrylamide gel or in the case of L554P on a 5% agarose (Nusieve) gel. The PCR assays developed are described in Table 1. Four of the mutations (R185X, L496R, R548X, and L554P) al- tered a naturally occurring restriction site. Since the remaining 5 mutations did not alter a restriction site, artificial restriction sites were introduced into PCR products using primer-directed restriction site modification (Haliassos et al., 1989; Yau et al., 1993). Modified primers were designed from the sequence adjacent to the mutation using a computer program devised for this purpose (Davi- dow, 1992), such that a restriction site was generated from a wild-type, but not from a mutant allele. The restriction site in the wild-type allele therefore acts as a positive control for digestion.

RESULTS The coding sequence of the FAC gene was

screened for mutations in a total of 76 unrelated FA patients. RNA of adequate quality and quan- tity was available from 34 patients. This was screened by RT-PCR and chemical cleavage mismatch analysis, as this technique detects almost all mutations (Cotton et al., 1988; Montandon et al.,


TABLE 2. FAC Sequence Changes Detected in 76 FA Patients

Mutation (MI or

Nucleotide Amino acid polymorphism Method of No. of patients with Exon sequence change change (P) detection sequence change Reference 1 292C+T Q13X M CCM, SSCP 1 compound Murer-Orlando et al.

heterozygote (1993); Verlander et al. (1994)

1 3 2 0 h A W22X M SSCP 1 homozygote This paper 1 332C-T S26F P SSCP 1 heterozygote Verlander et al. (1994) 4 671 &A G139E P CCM, SSCP 1 homozygote, 1 Whitney et al. (1993);

6 808C-T R185X M CCM 1 homozygote Gibson et al. (1993b) 6 823C-T L190F P SSCP 1 heterozygote This paper Intron 6 lVS6 + 30 C+T None P SSCP 2 heterozygotes This paper 9 1189 A+G 1312V P SSCP 1 heterozygote This paper 13 1649A+G Q465R P SSCP 1 homozygote, 1 This paper

13 1749 T-G L496R M SSCP 2 homozygotes” This paper 14 1897C+T R548X M CCM 1 compound Murer-Orlando et al.

3’UTR 1 9 7 4 h A None P SSCP 2 heterozygotes This paper

heterozygote Verlander et al. (1994)

heterozygote

heterozygote (1993)

*brothers

1989). Only genomic DNA was available from the remaining 42 patients, and this was screened by SSCP analysis of each of the 14 exons that contain coding sequence. The mutations or variants detected by these methods are summarised in Ta- ble 2.

Mutations

Details of the first three mutations we detected in this study have already been published (Gibson et al., 199313; Murer-Orlando et al., 1993). Briefly, a Scottish patient was homozygous for a C+T transition that alters an arginine residue to a termination codon at amino acid residue 185 (R185X) (Gibson et al., 1993b). This information was used subsequently to provide the family with a rapid prenatal diagnostic test. The fetus was shown to be heterozygous for R185X, and therefore likely to be phenotypically normal. The DNA result was later confirmed by cytogenetic analysis. In a second family, a chorionic villus sample (CVS) from a foetus diagnosed as affected by cytogenetic analysis was screened for mutations and found to be a compound heterozygote for two nonsense mutations, Q13X and R548X (Murer-Orlando et al., 1993).

SSCP analysis of exon 1 in a patient who was the offspring of a consanguineous relationship identified a bandshift. Reamplification of this exon from genomic DNA and sequencing revealed (Fig. 1) that the patient was homozygous for a G+A transition at base 320, which changes a tryptophan residue to a termination codon at codon 22 and is

designated W22X. Both parents were found to be heterozygous for this sequence change (not shown). This patient is from Pakistan and appears to be the first patient not of European origin identified to date with a mutation in the FAC gene. She was 12 years old when the diagnosis of FA was made, and her deceased sister was diagnosed at the age of 10 years. Neither patient had any developmental abnormalities.

SSCP analysis of exon 13 in two brothers of Greek origin with FA identified a bandshift. Direct sequencing showed that they were both homozygous for a T+G transition at base 1749 of the FAC cDNA sequence (Fig. 2), which results in the substitution of leucine for arginine at codon 496, designated L496R. The complete FAC coding sequence and exodintron boundaries from these patients was amplified and analyzed by direct sequencing, and no other mutations were detected. L496R was not detected in 31 Greek controls, or in an additional 94 FA patients; this residue is conserved in the mouse Fac gene (Wevrick et al., 1993). The age at diagnosis of FA in these two patients was 12 years and 25 years, respectively, and neither had any developmental abnormalities.

Polymorphisms

A further seven sequence changes in the FAC gene were detected in the 76 FA patients, all of which are likely to be polymorphisms rather than pathogenic mutations. One patient was homozygous for the variant G139E (Whitney et al., 1993); another was heterozygous for S26F (Verlander et

144 GIBSON ET AL.

FIGURE I. Sequencing gel of part of exon 1 in a control sample (C) and in a FA patient (P). Arrow, position of the G+A transition at nucleotide 320 of the published cDNA sequence. This results in the replacement of a tryptophan residue (TGG) by a stop (TAG) at codon 22 (W22X).

FIGURE 2. Sequencing gel of part of exon 13 in a FA patient and a control. The arrow indicates the position of a T+G transition at base 1749, which would replace a leucine residue with an arginine at codon 496.

al., 1994). Both variants have been shown to segregate independently of the FA phenotype in families (Verlander et al., 1994) and are therefore polymorphisms.

The other five variants detected in our panel of

patients have not been described previously. A patient who was the offspring of a consanguineous relationship was heterozygous for a substitution of leucine for phenylalanine at codon 190, designated L190F. Since such patients would be identical by descent at the site of the causative mutation, this variant is likely to be a polymorphism. One patient was heterozygous for a C+T transition 30 bp downstream of the exon 6/intron 6 boundary (IVS6 + 30C+T), a region not known to be involved in RNA splicing. A second patient from a consanguineous relationship was heterozygous for a substitution of valine for isoleucine at codon 3 12, designated I312V, suggesting that this too is a polymorphism. In a family in which no DNA was available from the patient, one of the parents (an obligate heterozygote) was found to be homozygous for a glutamine-to-arginine substitution at codon 465, designated Q465R. Since the parent is not affected with FA, this variant must also be a polymorphism. Finally, two patients were found to be heterozygous for a G-A transition at nucleotide 1974 in the 3’ untranslated region, which is unlikely to be of functional significance.

The panel of patients screened for FAC gene mutations included four FA patients from the Tswana chiefdom of South Africa. Four of the seven polymorphisms were detected in one or more of the Tswana patients (G139E, IVS6 + 30C+T, Q465R, 1974-A). None of these four variants was detected in 100 chromosomes from European individuals (Table 3). The frequency of these variants was therefore investigated by SSCP or restriction site analysis in 48 normal individuals (33 Tswana and 15 Sotho). All four polymorphisms were detected in this sample, at frequencies of 1-7% (Table 3). The SSCP analysis revealed a further two polymorphisms. A silent A+G transition was detected at nucleotide 663 (Q136Q), and a substitution of methionine for valine at amino acid residue 449. Neither of these changes was observed in the European sample. Figure 3 shows all the sequence changes identified in the FAC gene to date.

Testing for Known FAC Mutations

A total of 94 unrelated patients (including the 76 in whom the complete coding region was screened) were tested for the 6 known FAC mutations, and for the two novel mutations detected in this study, by restriction site analysis (see Meth- ods). Both aborted twin foetuses from an Ash- kenazi Jewish couple were found to be homozygous for the IVS4+4A+T mutation. No mutations


663AIG 1312V V449M IVS6+30C>T I L190F I S26F

TABLE 3. FAC Sequence Changes Identified in the Tswana and Sotho Populations of South Africa

Nucleotide sequence Amino acid Tswana European

Exon change change chromosomes chromosomes Reference

1974GIA

~~ ~

4 663 A+G None 8/96 0/100 This paper 4 671 G+A G139E 5/96 0/100 Whitney et al. (1993); Verlander et al.

lntron 6 IVS6 + 30 C-tT None 6/96 0/100 This paper 13 1600 G+A V449M 3/96 0/100 This paper 13 1649 A+G Q465R 1/96 0/100 This paper 3'UTR 1974 G+A None 7/96 0/100 This paper

(1994)

SEQUENCE CHANGES IDENTIFIED IN THE FAC GENE

w 2 2 x R548X

Q13X 1 dG322 IVS4+4A>T R 1 8 5 X L496R

were detected in the remaining patients, apart from the four in whom the Ql3X, W22X, R185X, L496R, and R548X mutations were originally identified.

DISCUSSION

O n l y six pathogenic mutations have previously been reported in the FAC gene (Strathdee et al., 199213; Gibson et al., 1993b; Murer-Orlando et al., 1993; Whitney et al., 1993; Verlander et al., 1994), all of which have been found in patients of European ancestry. The patient in whom the W22X mutation was detected is from Pakistan and is therefore the first patient not of European origin to be identified with a mutation in the FAC gene. Thus, FAC mutations are not limited to a partic- ular geographical region or population.

Mutations that would produce premature termination of translation are generally accepted as very likely to be pathogenic. However, the status of missense mutations is often less clear. In this study,

one missense mutation (L496R) was detected. Its credentials as a pathogenic mutation are based on the following data: (1) both affected brothers were homozygous for this sequence change; (2) it was not detected in 62 ethnically matched chromosomes or in 82 other European chromosomes; (3) the substitution of arginine for leucine is a non- conservative change from a neutral, hydrophobic amino acid to a basic charged residue; (4) no other sequence changes were detected on sequencing of the entire coding region and exodintron boundaries in these patients; and (5) this amino acid is conserved in the mouse Fac gene. Formal proof that L496R is pathogenic could be obtained by site-directed mutagenesis of wild-type cDNA, followed by testing for functional complementation after transfection into an FA group C cell line, as was demonstrated for the L554P mutation (Gavish et al., 1993).

A reliable judgment as to the phenotypic consequences of the two novel mutations we have de-

146 GIBSON ET AL.

scribed is not possible on the basis of findings from only two patients with each mutation. We note that no developmental abnormalities were found in either of the siblings with W22X or with L496R and that one of the patients with L496R only pre- sented at the age of 25 years.

The fact that only 67% of the amino acid sequence of the human and murine FAC protein are identical and that the mouse Fac gene can fully restore resistance to cross-linking agents in human group C cells (Wevrick et al., 1993) suggests that a substantial number of polymorphisms can be tol- erated within the FAC coding sequence without any phenotypic consequences. This is reflected in our study in which nine polymorphisms (seven of which were previously undescribed) were detected. Six of these polymorphisms cause amino acid substitutions: three were conservative substitutions (L190F, I312V, and V449M), but three were non- conservative (S26F, G139E, and Q465R) and were not found in 100 European control chromosomes. This underlines the need for caution in consideration of the possible pathogenic status of rare sequence variants (Joenje et al., 1994). Five of the nine polymorphisms (663A+G, G139E, IVS6 + 30C+T, V449M, and 1974-A) were found to occur at a relatively high frequency in the Tswana or Sotho populations of South Africa. These polymorphisms may be useful markers for the study of the origins and migration of the populations of Southern Africa.

The detection of FAC gene mutations in 4 of 76 (5.3%) of FA patients in our panel contrasts with the 14.4% reported by Verlander et al. (1994) in 174 patients. However, that panel included 16 Jewish patients, of whom 12 were homozygous for the IVS4 + 4A+T mutation, which is very common in Ashkenazi Jews (Whimey et al., 1994). Pathogenic mutations were detected in 12 of 158 (7.6%) of non-Jewish patients in that panel, consistent with our data. All mutation studies reported to date may underestimate the proportion of FA patients in group C, as some mutations may be located in the promoter region of the FAC gene, and most screening methods do not detect all mutations. If the mutation resulted in instability of the mutant mRNA, it might not be detected by protocols that involved cDNA amplification by RT-PCR. Also, the detection rate of the SSCP technique could be improved by the use of more than one set of conditions for electrophoresis of the single-stranded PCR products. However, cur- rent mutation data are consistent with the 8% es- timate obtained by linkage analysis in 36 FA fam-

ilies (Gibson et al., 1994). It is possible that this figure could vary substantially in different ethnic groups. The sample size of most of the different groups in our panel is too small to permit population-specific estimates of the frequency of group C. However, no mutations were detected in 24 families from the Afrikaner population of South Af- rica, who have the highest reported incidence of FA in the world (1 in 22,000). This result is consistent with the results of a linkage study in 13 Afrikaner FA families, which generated a LOD score of -8.4 at the FAC locus (Gibson et al., 1994). Thus if a founder mutation is present in this population (Rosendorff et al., 1987), it is unlikely to be in the FAC gene.

The detection of FAC mutations in less than 10% of FA patients, other than those with Ash- kenazi Jewish ancestry, raises the question of whether the rather arduous and costly process of further mutation screening should be deferred until genes for other FA complementation groups have been cloned. The identification of FAC mutations is useful, since it can provide rapid and qualitative prenatal diagnosis in FA families belonging to complementation group C (Murer-Orlando et al., 1993), and we have diagnosed FA in twin foetuses aborted because of multiple congenital abnormalities, by the detection of homozygosity for the IVS4+4A+T mutation (Cox et al., in press). In addition, patients identified as belonging to complementation group C would be potential candi- dates for gene therapy (Walsh et al., 1994). We have resolved this question by developing a policy of selective screening, as follows. Clearly, any individual known to be from group C by complementation analysis should be screened, although this information is available for only a small number of FA patients. In FA patients not classified by complementation analysis, patients are screened for the 8 known FAC mutations using the simple PCR assays we have devised (Table 1). This is particu- larly useful for the IVS4 + 4A+T mutation in Jew- ish patients, but other less striking examples of population-specific mutations may emerge. For in- stance, we detected the R185X mutation in two unrelated Scottish parents of an FA patient; this mutation has now been found in a North Ameri- can family with Scottish ancestry (Verlander et al., 1994). Finally, families where DNA samples have been obtained from two or more affected individuals, or where there is a patient from a consanguineous relationship, are typed with highly poly- morphic markers flanking the FAC gene (Gibson et al., 1994), and the data used to calculate the


posterior probability (q) that the family is from group C, from the formula

wi = O1lOzi/alOzi + 1 - O1

where 01 is the prior probability that the patient is from group C, and Zi is the LOD score for FAC gene markers in the ith family (Ott, 1991). If we set 01 = 0.10, then for a family with a LOD score of + 0.6, mi = 0.31. We restrict mutation screening of the full FAC coding sequence to families with a LOD score equal to or greater than 0.6, since the chance of detecting a mutation is then approximately 30% or greater.

The low frequency of FAC gene mutations in Fanconi anaemia patients underlines the impor- tance of cloning the gene that specifies the major complementation group, FA(A). The establish- ment of a European collaboration for FA research (EUFAR), which has classified a significant number of families with multiple affected members as complementation group A (Gluckman and Joenje, 1995), has made linkage mapping and positional cloning of the FAA gene a realistic possibility.

ACKNOWLEDGMENTS This work was supported by the Medical Re-

search Council (UK), the Generation Trust, Fan- coni Anaemia Breakthrough (UK), and the South African MRC. We thank the FA patients and their families who have provided blood samples for this research.

REFERENCES Auerbach AD, Ghosh R, Pollio PC, Zhang M (1989) Diepoxy-

butane test for prenatal and postnatal diagnosis of Fanconi’s anemia. In Schroeder-Kurth TM, Auerbach AD, Obe G (eds): Fanconi Anemia: Clinical, Cytogenetic and Experimental As- pects. Berlin: Springer-Verlag, pp 71-82.

Chomczynski P, Sacchi N (1987) Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156-159.

Cotton RGH, Rodrigues NR, Campbell RD (1988) Reactivity of cytosine and thymine in single base pair mismatches with hy- droxylamine and osmium tetroxide and its application to the study of mutations. Proc Natl Acad Sci USA 85:4397-4401.

Cox PM, Gibson RA, Morgan N, Brueton LA (1996) VACTERL with hydrocephalus in twins due to Fanconi’s anaemia (FA): Mutation in the FAC gene. Am J Med Genet, in press.

Davidow LS (1992) Selecting PCR designed mismatch primers to create diagnostic restriction sites. Comp Appl Biosci 8:193- 194.

Gavish H, dos Santos CC, Buchwald M (1993) A Leu,,,-to-Pro substitution completely abolishes the functional complement- ing activity of the Fanconi anemia (FACC) protein. Hum Mol Genet 2:123-126.

Gibson RA, Buchwald M, Roberts RG, Mathew CG (1993a) Characterisation of the exon structure of the Fanconi anaemia group C gene by vectorette FCR. Hum Mol Genet 2:35-38.

Gibson RA, Hajianpour AK, Murer-Orlando M, Buchwald M, Mathew CG (1993b) k nonsense mutation and exon skipping in the Fanconi Anaemia group C gene. Hum Mol Genet

Gibson RA, Ford D, Jansen S, Savoia A, Havenga C, Milner RD, de Ravel TJ, Cohn RJ, Ball SE, Roberts I, Llerena JC, Vore- chovsky I, Pearson T, Birjandi F, Hussein SS, Murer-Orlando M, Easton DF and Mathew CG (1994) Genetic mapping of the FAC gene and linkage analysis in Fanconi anaemia families. J Med Genet 31:868-871.

Gluckman E, Joenje H (1995) Constitutional Aplastic Anaemia: A new concerted action for European Fanconi Anaemia Re- search (EUFAR). Eur J Cancer 31A:133.

Gordon-Smith EC, Rutherford TR (1991) Fanconi Anemia: Con- stitutional aplastic anemia. Semin Hematol 28:104-112.

Green PM, Bentley DR, Mibashan RS, Nilsson IM, Giannelli F (1989) Molecular pathology of haemophilia B. EMBO J

Haliassos A, Chomel JC, Tesson L, Baudis M, Kruh J, Kaplan JC, Kitzis A (1989) Modification of enzymatically amplified DNA for the detection of point mutations. Nucleic Acids Res 17: 3606.

Joenje H, Lo Ten Foe JR, Arwert F, ten Kate LP (1994) Comple- mentation testing and genetic classification of Fanconi’s anaemia. Lancet 343:862.

Lehmann AR, Bootsma D, Clarkson SG, Cleaver JE, McAlpine PJ, Tanaka K, Thompson LH, Wood RD (1994) Nomencla- ture of human DNA repair genes. Mutat Res 31541-42.

Montandon A], Green PM, Giannelli F, Bentley DR (1989) Di- rect detection of point mutations by mismatch analysis: Appli- cation to haemophilia B. Nucleic Acids Res 9:3347-3358.

Morris DJ, Reis A (1994) A YAC contig spanning the nevoid basal cell carcinoma syndrome, Fanconi anaemia group C, and xe- roderma pigmentosum group A loci on chromosome 9q. Ge- nomics 23:23-29.

Mullenbach R, Lagoda PJL, Welter C (1989) An efficient salt- chloroform extraction of DNA from blood and tissues. Trends Genet 5:391.

Murer-Orlando M, Llerena JC Jr, Bijandi F, Gibson RA, Mathew CG (1993) FACC gene mutations and early prenatal diagnosis of Fanconi’s anaemia. Lancet 342:686.

Orita M, Sasaki Y, Siya T, Hayashi K (1989) Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 5:874-879.

Ott J (1991) Analysis of Human Genetic Linkage. rev. ed. Balti- more: Johns Hopkins University Press.

Roberts RG, Coffey AG, Bobrow M, Bentley DR (1992) Deter- mination of the exon structure of the distal portion of the dystrophin gene by vectorette PCR. Genomics 13:942-950.

Rosendorff J, Bernstein R, Macdougall L, Jenkins T (1987) Fan- coni Anemia: Another disease of unusually high prevalence in the Afrikaans population of South Africa. Am J Med Genet

Sasaki MS, Tonomura A (1973) A high susceptibility of Fanconi’s anemia to chromosome breakage by DNA cross-linking agents. Cancer Res 33:1829-1833.

Schroeder TM, Anschiiu F, Knopp A (1964) Spontane chromo- somenaberrationen bei familiarer Panmyelopathie. Humange- netik 1:194-196.

Strathdee CA, Duncan AMV, Buchwald M (1992a) Evidence for at least four Fanconi anaemia genes including FACC on chromosome 9. Nature Genet 1:196-198.

Strathdee CA, Gavish H, Shannon WR, Buchwald M (1992b) Cloning of cDNAs for Fanconi’s anaemia by functional complementation. Nature 356:763-767.

2:797-799.

8~1067-1072.

27:793-797.

148 GIBSON ET AL.

Verlander PC, Lin JD, Udono MU, Zhang Q, Gibson RA, Mathew CG, Auerbach AD (1994) Mutation analysis of the Fanconi anemia gene FACC. Am J Hum Genet 54:595-601.

Walsh CE, Grompe M, Vanin E, Buchwald M, Young NS, Nien- huis AW, Liu JM (1994) A functionally active retrovirus vec- tor for gene therapy in Fanconi anemia group C. Blood 84: 453-459.

Wevrick R, Clarke CA, Buchwald M (1993) Cloning and analysis of the murine Fanconi anemia group C cDNA. Hum Mol Genet 2655-462.

Whitney MA, Saito H, Jakobs PM, Gibson RA, Moses RE, Grompe M (1993) A common mutation in the FACC gene

causes Fanconi anaemia in Ashkenazi jews. Nature Genet

Witney MA, Jakobs P, Kaback M, Moses RE, Grompe M (1994) The Ashkenazi Jewish Fanconi anemia mutation: Incidence among patients and carrier frequency in the at-risk population. Hum Mutat 3:339-341.

Winship PR (1989) An improved method for directly amplifying PCR amplified material using dimethylsulphoxide. Nucleic Acids Res 17:1266.

Yau SC, Roberts RG, Bobrow M, Mathew CG (1993) Direct diagnosis of carriers of point mutations in hchenne muscular dystrophy. Lancet 341:273-275.

4:202-205.

novel mutations and polymorphisms in the fanconi anemia group c gene

Documents