Familial aggregation studies indicate that up to 20% of

Transcription

1 Mapping a Locus for Familial Thoracic Aortic Aneurysms and Dissections (TAAD2) to 3p24 25 Sumera N. Hasham, PhD; Marcia C. Willing, MD, PhD; Dong-chuan Guo, PhD; Ann Muilenburg, MA; Rumin He, MD; Van T. Tran, MS; Steven E. Scherer, PhD; Sanjay S. Shete, PhD; Dianna M. Milewicz, MD, PhD Background Familial thoracic aortic aneurysms and dissections (TAAD) occur as part of known syndromes such as Marfan syndrome but can also be inherited in families in an autosomal dominant manner as an isolated condition. Previous studies have mapped genes causing nonsyndromic familial TAAD to 5q13 15 (TAAD1) and 11q23.2-q24 (FAA1). Further genetic heterogeneity for the condition was evident by the presence of TAAD in some families not linked to these known loci. Methods and Results A 4-generation family with dominant mode of inheritance of TAAD was studied. Affected status was determined by dilation of the ascending aorta, surgical repair of an aneurysm or dissection, or death as the result of aortic dissection. None of the family members evaluated met the diagnostic criteria for Marfan syndrome. After exclusion of known loci for familial TAAD, a genome-wide scan was carried out to map the defective gene causing the disease in the family. A locus was mapped to a 25-cM region on 3p24 25 with a maximum multipoint logarithm of the odds score of Conclusions A third locus for nonsyndromic TAAD was mapped to 3p24 25 and termed the TAAD2 locus. This locus overlaps a previously mapped second locus for Marfan syndrome, termed the MFS2 locus. Future characterization of the TAAD2 gene will determine if TAAD2 is allelic to MFS2. In addition, identification of the TAAD2 gene will improve the presymptomatic diagnosis of individuals with this life-threatening genetic syndrome and provide information concerning the pathogenesis of the disease. (Circulation. 2003;107: ) Key Words: aorta aneurysm genetics mapping Familial aggregation studies indicate that up to 20% of patients with thoracic aortic aneurysms and dissection (TAAD) who do not have Marfan syndrome (MFS) have a first-degree relative with the disease. 1,2 In the majority of families, TAAD is inherited in an autosomal dominant manner characterized by decreased penetrance and variable expressivity. 3 A locus for TAAD was mapped to the long arm of chromosome 5q13 14 by positional cloning and termed the TAAD1 locus. Nine of 15 TAAD families studied demonstrated evidence of linkage of the aortic disease to this locus. 4 A second locus associated primarily with thoracic aortic aneurysms, FAA1, has been identified at 11q, with one family linked to this locus. 5 Further genetic heterogeneity for TAAD is documented by the fact that 6 of 15 TAAD families are not linked to any known loci of aneurysm formation. We identified a large family with multiple members with TAAD but without the ocular or skeletal features of MFS. A genome-wide screen mapped the defective gene causing TAAD in this family to a 25 cm locus on 3p24 25, thus establishing this locus as the third locus responsible for autosomal dominant TAAD (TAAD2). Interestingly, this locus overlaps with a controversial locus for a MFS-like connective tissue disorder, the MFS2 locus Methods Family Characterization and Sample Collection The Institutional Review Board at the University of Texas Health Science Center at Houston approved this study. Family members were examined, and ophthalmologic examinations were conducted. Family members had echocardiograms at one of the two selected centers to assess heart structure and function. The diameters at the sinus of Valsalva, the supra-aortic ridge, and the aortic root were measured from cross-sectional echocardiography images in the parasternal long-axis orientation and plotted against nomograms derived from normal individuals measurements. 11 Individuals were identified as affected if they had an aneurysm or dissection of the ascending thoracic aorta; 16 affected individuals participated in the study. Individuals at risk for inheriting the condition with a normal echocardiogram were scored as unknown because of the decreased penetrance of TAAD. Spouses of family members were scored as Received January 9, 2003; revision received April 10, 2003; accepted April 25, From the Department of Internal Medicine, University of Texas Medical School at Houston (S.N.H., D.-c.G., R.H., V.T.T., D.M.); the Department of Pediatrics, University of Iowa, Iowa City (M.C.W., A.M.); Human Genome Sequencing Center, Baylor College of Medicine, Houston, Tex (S.E.S.); and the Department of Epidemiology, University of Texas M.D. Anderson Cancer Center, Houston, Tex (S.S.S.). Guest editor for this article was Marschall S. Runge, MD, University of North Carolina at Chapel Hill. Correspondence to Dianna M. Milewicz, MD, PhD, 6431 Fannin, MSB 1.614, Houston, TX Dianna.M.Milewicz@uth.tmc.edu 2003 American Heart Association, Inc. Circulation is available at DOI: /01.CIR

2 Hasham et al Mapping the TAAD2 Locus to 3p Figure 1. Pedigree of the TAA035 family and segregation of the 3p24 25 haplotype at TAAD2. Closed symbols indicate affected; open symbols, unknown diseased status in family members and normal in spouses. Plus symbols indicate samples used for genome-wide scan; minus symbols, family members included for linkage confirmation. Blackened bar indicates markers associated with the disease. unaffected. Family members were examined for other skeletal features of MFS, based on the Ghent nosology for MFS. 12 After obtaining informed consent, buccal cells, blood, autopsy samples, and/or skin biopsy specimens were collected from family members. Genomic DNA was isolated from 52 family members as described previously. 4 Genotyping Fluorescently tagged primers (382 pairs, ABI Prism Linkage mapping sets-md-10, Applied Biosystems) were used to amplify autosomal polymorphic sequences spaced at 10-cM intervals. Primers to analyze other polymorphic markers were designed according to the Cooperative Human Linkage Center 13 and the Genome Database. 14 The positions of the markers were based on public genome databases (Ensembl, 15 Golden Path, 16 and the University of South Hampton- School of Medicine 17 ). The amplified products were analyzed on an ABI Prism 3100 Genetic Analyzer. Genemapper 2 and Genescan software were used for allele assignments (Applied Biosystems). Multiple markers were used to study linkage to known loci of aneurysm formation, FBN1, MFS2, TAA1, and FAA1. Twenty-seven markers spaced on a 32-cM region were used to analyze the putative locus on 12q and 27 markers located on a 30-cM region on 3p were used for fine mapping and defining the critical interval for TAAD2. 18 Multiplexed amplification of DNA was done by using three primer sets for genome scan markers with Ampli Taq Gold (Applied Biosystems). The PCR amplifications for fine mapping and sequencing were carried out with the use of HotStart Taq DNA polymerase (Qiagen Inc). Sequencing Mutation analysis of candidate gene, FBLN2 (XM_003053), was performed by bidirectional sequence analysis of amplified genomic DNA fragments, using intron-based, exon-specific primers. Total RNA was harvested as previously described from fibroblasts and used for reverse transcription and amplification of the FBLN2 cdna. 4 The amplified DNA fragments were sequenced with the use of a Big-Dye Terminator sequencing kit and analyzed on ABI-3700 sequencer (Applied Biosystems). Statistical Analysis Two-point LOD score analysis was carried out for linkage studies at known loci of aneurysm formation and for the genome scan analysis. Multipoint analysis was used for the fine mapping at 12q and 3p loci. Pairwise and multipoint LOD scores were calculated with MLINK and LINKMAP programs of the computer software FASTLINK, version3.p. 19,20 Allele frequencies for each marker were obtained by using the founders in the pedigrees and penetrance values were used as reported earlier. 4 Results Clinical Evaluations We studied a 4-generation family of Swiss-German heritage (Figure 1). The proband (IV:10) presented with a type I dissection, requiring surgical repair of the ascending aorta at the age of 14 years, and subsequently had a dissection of the transverse arch at the age of 24 years. She was diagnosed with Turner syndrome at 11 years of age and placed on 3 years of hormone treatment (oxandrolone). She did not have a bicuspid aortic valve or coarctation. Her father (III:8) presented at the age of 54 years with dissection of a5to6cmascending aortic aneurysm. Her brothers (IV:7 and IV:9) were diagnosed with aortic root aneurysms, and IV:7 had elective aortic root replacement at the age of 28 years. The proband s uncle (III:5) had an acute type I dissection at the age of 52 years. He died during attempted repair of a 7.2-cm aortic aneurysm at the age of 56 years. His older son (IV:5) had a Bentall composite graft repair of a progressively enlarging thoracic aortic aneurysm at the age of 22 years and had a history of bilateral inguinal hernias and a pectus excavatum. The younger son (IV:6) had a dilated sinus of Valsalva at the age of 11 years. The grandmother (II:2) had a dilated aortic root at the age of 72 years. She had a brother (II:3) and a sister (II:9) who had had surgical repair of aortic aneurysms, and a brother (II:7) who had died of aortic dissection at the age of 72 years. II:3 had two sons who had aortic dissections (III:11 and III:16) and a grandson who died of aortic dissection (IV:12) and another grandson who had prophylactic repair of an ascending aortic aneurysms (IV:11). Detailed clinical findings for 6 family members are shown in Table 1. Protusio acetabuli and dural ectasia were not noted in any family members. Affected status for genetic linkage studies was determined solely on the presence of an ascend-

3 3186 Circulation July 1, 2003 TABLE 1. Clinical Features of MFS in 6 Affected Family Members of TAA035 Skeletal Features Cardiovascular Features Id Age, y Sex HT, cm A/H US/LS SC Palate Arac Pectus MVP MR AR, cm SV, cm BSA, m 2 Diss III:8 54 M IV:5 22 M IV:6 11 M IV:7 28 M IV:9 30 M IV:10 14 F HT indicates height; A/H, arm span to height ratio; US/LS, upper segment to lower segment ratio; SC, scoliosis; Palate, highly arched palate; Arac, arachnodactyly; Pectus, pectus excavatum or carinatum; MVP, mitral valve prolapse; MR, mild mitral regurgitation; AR, aortic root diameter; SV, sinus of Valsalva diameter; BSA, body surface area; Diss, dissection; ( ), presence of a feature; (0), absence of a feature; and, data not available. ing aortic aneurysm or dissection (Table 2). Reduced penetrance and variable age of onset of the aortic disease in the family was indicated by II:5 having mild aortic dilatation, whereas his daughter (III:20) had a normal echocardiogram at age 54 years but her son (IV:18) had an enlarged aortic root. No family member had ectopia lentis. A second individual in the family with Turner syndrome (III:32) had bicuspid aortic valve and coarctation of the aorta but no aortic root dilation and was scored as unknown for the linkage analysis. One individual (III:18) was scored as unknown for the linkage analysis but was subsequently found to have a mildly enlarged sinus of Valsalva on echocardiogram (Table 2). Mapping of the TAAD2 Locus The haplotypes and LOD scores for markers at the FBN1, TAAD1, and FAA1 loci excluded the linkage of the TAAD phenotype in TAA035 family to these loci. For MFS2 locus, LOD scores of 1.05 for D3S1266, 0.59 for D3S3659 and 3.5 for D3S1286, were obtained. Haplotype construction at the MFS2 locus indicated that the segregation of the diseased haplotype in two unaffected individuals (III:2 and III:7), excluding linkage of the disease to this locus that was reported to be highly penetrant. 7 Therefore, we sought to map another locus for TAAD by using only the TAA035 family to avoid possible problems arising from further genetic heterogeneity for the condition. A genome-wide screen with 380 polymorphic markers was completed by using the original 12 DNA samples collected (Figure 1, plus symbols). LOD scores were obtained through the use of the MLINK program, assuming an autosomal dominant model with age-dependent penetrance. Eighteen markers demonstrated a suggestive LOD score 0.6 (P 0.05) (data not shown). Further analysis of the positive makers was done with samples from 16 additional family (Figure 1, minus symbols), and markers at two loci on 12q and 3p continued to show evidence of linkage to the phenotype (Table 3). Linkage analysis and haplotype construction with 12q markers in a 23 cm region between markers D12S83 and D12S1030 and samples from 31 TAA035 members excluded the 12q locus. An additional 26 makers located in a 30-cM region on 3p were used to confirm linkage of the phenotype to this locus using DNA from 51 family members (Figure 1). Pairwise analysis showed significant evidence of linkage to multiple markers (Table 4). Multipoint linkage analysis with all markers demonstrated 11 markers to have a positive LOD score above 3.0, with a maximum multipoint LOD score of 4.27 obtained with marker D3S2336 (Figure 2). Construction of the 3p haplotype with these 26 markers revealed recombinants between D3S3701 and D3S2301 in individual III:8 at the telomeric end and between D3S1619 and D3S1211 in individual III:5 at the centromeric end, defining a critical interval of 25 cm (Figure 1). Haplotype analysis indicated the segregation of the disease haplotype in 4 adults (III:2, III:20, III:29, and III:31; ages 54, 54, 44, and 38 years, respectively) who did not have any evidence of aortic disease, thus confirming decreased penetrance for the disorder. Fibulin-2 (FBLN2), an extracellular matrix protein known to interact with fibrillin-1, was located in the critical interval. 21 FBLN2 was sequenced by using both genomic DNA and fibroblast cdna using samples from two affected family members (IV:7, IV:9). Intron-based, exon-specific primers were designed to amplify exons 2 through 16 of FBLN2. Exon 1 was sequenced after amplification from fibroblast RNA by RT-PCR. Three known SNPs in FBLN2 exon 10, G836A, G845A, and G855A, were identified, but no diseasecausing alterations were found. Discussion The data presented in this study maps a locus for familial TAAD to 3p24 25 (TAAD2). The defective gene lies in a 25-cM region between D3S3701 and D3S1211. There are 184 transcripts, including hypothetical, predicted, and pseudogenes in the 25-cM interval, which are currently being analyzed. The clinical features in the family are consistent with a familial TAAD phenotype and include ascending thoracic aortic aneurysms and dissections inherited in an autosomal dominant manner associated with decreased penetrance and variable age of onset. The variable expression and decreased penetrance of the TAAD2 locus and other familial TAAD loci make it necessary to continue to monitor the aortic dimensions throughout an individual s lifetime if they are at risk for inheriting the condition. In addition, the decreased penetrance places the next generation at risk even if the parent is unaffected. The difficulties in monitoring at-risk family members emphasize the need for mapping the loci responsible for familial TAAD and characterizing the

4 Hasham et al Mapping the TAAD2 Locus to 3p TABLE 2. Cardiovascular Measurements of Family Members Obtained From Echocardiograms Id Dx Age, y Sex SV, cm SAR, cm AR, cm BSA, m 2 Diss Sur II:2 Aff 79 F II:3 Aff 71 M 0 II:5 Aff 76 M II:7 Aff 72 died M 0 II:9 Aff 71 F III:2 Unk 54 F III:3 Unk 44 M III:5 Aff 52 M III:7 Unk 50 M III:8 Aff 54 M III:10 Unk 41 F III:11 Aff 43 M III:13 Unk 37 F III:15 Unk 43 F III:16 Aff 38 died M 0 III:18 Unk 35 M III:20 Unk 54 F 3.3* 2.4* 2.1* 0 0 III:24 Unk 50 F 3.1* 3.1* 2.2* 0 0 III:26 Unk 45 M 3.4* 2.4* 2.3* 2.2* 0 0 III:29 Unk 44 F III:31 Unk 38 F III:32 Unk 41 F IV:5 Aff 21 M 5 0 IV:6 Aff 11 M IV:7 Aff 28 M IV:9 Aff 30 M IV:10 Aff 14 F IV:11 Aff 22 M IV:12 Aff 19 M IV:18 Aff 28 M IV:19 Unk 26 F IV:27 Unk 25 F IV:28 Unk 28 M 0 0 IV:30 Unk 19 M IV:31 Unk 16 F V:6 Unk 6 M Dx indicates family members affected with aortic dilation; aneurysm or dissection (Aff) and family members with aortic measurements in the normal range (Unk); AR, aortic root diameter; SV, sinus of Valsalva diameter; SAR, supra-aortic ridge diameter; BSA, body surface area; Diss, presence ( ) or absence (0) of aortic dissection; Sur, presence ( ) or absence (0) of surgical repair of the aorta; and, data not available or procedure not performed. *Sinus of Valsalva and/or aortic measurements obtained from MRA. Sinus of Valsalva and/or aortic root measurements prior to surgical repair. Patient underwent elective prophylactic surgical repair of the aorta but aortic measurements are not available. genes at these loci to conclusively identify individuals at risk for TAAD in these families. Interestingly, TAAD2 maps onto MFS2, a previously mapped 9-cM locus between D3S1293 and D3S1283 identified by using a large French family. The phenotype of the affected members in the French family was similar to MFS, with complications in the skeletal and cardiovascular systems but no ocular features. The characterization of the clinical phenotype and the mapping of the locus were controversial and discounted by many investigators Although these 2 loci could result from alterations in different genes, the similarities of the cardiovascular phenotype would suggest that the same gene is responsible for both syndromes. Subsequent to the mapping of the MFS2 locus, there have

5 3188 Circulation July 1, 2003 TABLE 3. Two-Point LOD Score Obtained With Positive Markers in the Genome Scan With 31 Samples From TAA035 LOD score at Theta Locus Position, cm D1S D1S D1S D2S D3S D3S D4S D4S D5S D6S D9S D12S D12S D12S D14S D14S D15S D19S D20S been no other families identified that have been linked to this locus, either with MFS or familial TAAD. 4,5 Two individuals in TAA035 (IV:10 and III:32) had Turner syndrome, which is associated with TAAD, aortic coarctation, and bicuspid aortic valve. 22 One individual (IV:10) had an aortic dissection at the age of 14 years, followed by another dissection at 24 years of age. Review of the literature revealed no reports of a woman with Turner syndrome with aortic dissection at such a young age unless coarctation of the aorta was present Since she has the affected 3p haplotype associated with disease in the family, the early onset of her aortic disease may be due to inheriting two genetic conditions that predispose her to this disease. A second woman (III:32) with Turner syndrome in the family had both bicuspid aortic valve and coarctation of the aorta but did not have aortic dilation or dissection at the age of 43 years, consistent with her not inheriting the affected 3p haplotype. One individual (III:18, 35 years of age) who did not inherit the affected haplotype at the TAAD2 locus was subsequently found to have a mildly enlarged sinus of Valsalva on echocardiogram. The nomograms established by Roman et al 11 are based on age and body surface area. The age grouping used in this study are birth to 18 years of age, 18 to 40 years Figure 2. Three-point linkage analysis of marker data at the TAAD2 locus on chromosome 3p. y-axis, LOD score; x-axis, position of the markers on chromosome 3p, based on their relative chromosomal distance. Position of the markers is given in centimorgans (cm).

6 Hasham et al Mapping the TAAD2 Locus to 3p TABLE 4. Two-Point LOD Score Obtained With Positive Markers on 3p24 25 in TAA035 LOD Score at Locus Position, cm D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S D3S of age, and 40 years of age. If individual III:18 were a few years older, his measurements would fall within the normal range for the age group 40 years of age. Therefore, the nomograms used may be less accurate for individuals who fall near the age divisions. Alternatively, this individual may have aortic disease that is etiologically distinct from the aortic disease in other family members. This has been observed previously in a family with an FBN1 mutation who had an individual with aortic dissection who did not carry the mutant allele. 26 In summary, we have mapped a second locus for TAAD to 3p24 25 and named the locus TAAD2. The genetic heterogeneity of familial TAAD is an obstacle in mapping loci causing the disease and families such as TAA035 become critical in definitively establishing the loci responsible for this condition. We have previously identified a major locus for familial TAAD on 5q (TAAD1), and another minor locus was mapped to 11q (FAA1). Eighteen TAAD families described previously failed to show linkage to 3p24 25, indicating that TAAD2 is a minor locus for TAAD. 4,5 Since some families are not linked to any of the identified loci for this condition, there is at least one more unmapped locus for familial TAAD. Future characterization of the defective genes at these loci will aid in the clinical treatment of the disease in these families and provide insight into the cause of this condition. Acknowledgments This work was supported by RO1 HL62594 (D.M.M.), and MO1RR02558 (General Clinical Research Center). Sumera N. Hasham is a Schissler Fellow. Dr Milewicz is a Doris Duke Distinguished Clinical Scientist. We thank the families for participating in this study. References 1. Biddinger A, Rocklin M, Coselli J, et al. Familial thoracic aortic dilatations and dissections: a case control study. J Vasc Surg. 1997;25: Coady MA, Davies RR, Roberts M, et al. Familial patterns of thoracic aortic aneurysms. Arch Surg. 1999;134: Milewicz DM, Chen H, Park ES, et al. Reduced penetrance and variable expressivity of familial thoracic aortic aneurysms/dissections. Am J Cardiol. 1998;82: Guo D, Hasham S, Kuang SQ, et al. Familial thoracic aortic aneurysms and dissections: genetic heterogeneity with a major locus mapping to 5q Circulation. 2001;103: Vaughan CJ, Casey M, He J, et al. Identification of a chromosome 11q23.2-q24 locus for familial aortic aneurysm disease, a genetically heterogeneous disorder. Circulation. 2001;103: Kainulainen K, Steinmann B, Collins F, et al. Marfan syndrome: no evidence for heterogeneity in different populations, and more precise mapping of the gene. Am J Hum Genet. 1991;49: Boileau C, Jondeau G, Babron MC, et al. Autosomal dominant Marfan-like connective-tissue disorder with aortic dilation and skeletal anomalies not linked to the fibrillin genes. Am J Hum Genet. 1993;53: Collod G, Babron MC, Jondeau G, et al. A second locus for Marfan syndrome maps to chromosome 3p24.2-p25. Nat Genet. 1994;8:

7 3190 Circulation July 1, Gilchrist DM. Marfan syndrome or Marfan-like connective-tissue disorder. Am J Hum Genet. 1994;54: Dietz H, Francke U, Furthmayr H, et al. The question of heterogeneity in Marfan syndrome. Nat Genet. 1995;9: Roman MJ, Devereux RB, Kramer-Fox R, et al. Two-dimensional echocardiographic aortic root dimensions in normal children and adults. Am J Cardiol. 1989;64: De Paepe A, Devereux RB, Dietz HC, et al. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet. 1996;62: The Cooperative Human Linkage Center. Available at: gai.nci.nih.gov/chlc/. Accessed May 30, The Genome Database: An International Collaboration in Support of the Human Genome Project. Available at: Accessed May 30, The Wellcome Trust Sanger Institute. Project Ensembl. Available at: Accessed May 30, UCSC Genome Bioinformatics. Available at: Accessed May 30, University of South Hampton-School of Medicine. Available at: ftp:// cedar.genetics.soton.ac.uk/pub/chrom3/map.html. Accessed May 30, The University of Texas Health Science Center at Houston. Department of Internal Medicine. Division of Medical Genetics. Available at: Accessed May 30, Cottingham RW Jr, Idury RM, Schaffer AA. Faster sequential genetic linkage computations. Am J Hum Genet. 1993;53: Lathrop GM, Lalouel JM, Julier C, et al. Multilocus linkage analysis in humans: detection of linkage and estimation of recombination. Am J Hum Genet. 1985;37: Reinhardt DP, Sasaki T, Dzamba BJ, et al. Fibrillin-1 and fibulin-2 interact and are colocalized in some tissues. J Biol Chem. 1996;271: Sybert VP. Cardiovascular malformations and complications in Turner syndrome. Pediatrics. 1998;101:E Akimoto N, Shimizu T, Ishikawa M, et al. The surgical treatment of aortic dissection in a patient with Turner s syndrome: report of a case. Surg Today. 1994;24: Bordeleau L, Cwinn A, Turek M, et al. Aortic dissection and Turner s syndrome: case report and review of the literature. J Emerg Med. 1998; 16: Elsheikh M, Casadei B, Conway GS, et al. Hypertension is a major risk factor for aortic root dilatation in women with Turner s syndrome. Clin Endocrinol (Oxf). 2001;54: Francke U, Berg MA, Tynan K, et al. A Gly1127Ser mutation in an EGF-like domain of the fibrillin-1 gene is a risk factor for ascending aortic aneurysm and dissection. Am J Hum Genet. 1995;56: