Mutations in Transforming Growth Factor-; Receptor Type II Cause Familial Thoracic Aortic Aneurysms and Dissections
http://www.100md.com
《循环学杂志》
the Department of Internal Medicine and Institute of Molecular Medicine (H.P., V.T.F., J.C., S.N.H., D.M.M.)
the Department of Cardiothoracic and Vascular Surgery (A.L.E., H.J.S.)
the Structural Biology Research Center (C.S.R.), the University of Texas Health Science Center at Houston
the Department of Internal Medicine (E.S.), Cardiology Division, The Ohio State University, Columbus
the Department of Medical Genetic Services (P.F.G., C.Z.), Marshfield Clinic, Marshfield, Wis
the Department of Epidemiology (S.S.), the University of Texas M.D. Anderson Cancer Center, Houston, Tex
the Department of Pediatrics, the University of Iowa (M.C.W.), Iowa City.
Abstract
Background— A genetic predisposition for progressive enlargement of thoracic aortic aneurysms leading to type A dissection (TAAD) is inherited in an autosomal-dominant manner in up to 19% of patients, and a number of chromosomal loci have been identified for the condition. Having mapped a TAAD locus to 3p24–25, we sequenced the gene for transforming growth factor-; receptor type II (TGFBR2) to determine whether mutations in this gene resulted in familial TAAD.
Methods and Results— We sequenced all 8 coding exons of TGFBR2 by using genomic DNA from 80 unrelated familial TAAD cases. We found TGFBR2 mutations in 4 unrelated families with familial TAAD who did not have Marfan syndrome. Affected family members also had descending aortic disease and aneurysms of other arteries. Strikingly, all 4 mutations affected an arginine residue at position 460 in the intracellular domain, suggesting a mutation "hot spot" for familial TAAD. Despite identical mutations in the families, assessment of linked polymorphisms suggested that these families were not distantly related. Structural analysis of the TGFBR2 serine/threonine kinase domain revealed that R460 is strategically located within a highly conserved region of this domain and that the amino acid substitutions resulting from these mutations will interfere with the receptor’s ability to transduce signals.
Conclusion— Germline TGFBR2 mutations are responsible for the inherited predisposition to familial TAAD in 5% of these cases. Our results have broad implications for understanding the role of TGF-; signaling in the pathophysiology of TAAD.
Key Words: aneurysm ; aorta ; genetics ; dissection ; receptors, transforming growth factor beta
Introduction
Aneurysms and dissections are the major diseases affecting the aorta and are a leading cause of morbidity and mortality in the United States.1 The most common location for aortic aneurysms is in the infrarenal abdominal aorta, followed by the ascending thoracic aorta. Aortic dissections are classified according to the anatomic location of the initial tear in the aortic wall, with type A dissections initiating in the ascending thoracic aortic just above the aortic valve and type B dissections originating in the descending thoracic aortic just beyond the take-off of the subclavian artery. Without prophylactic surgical repair of the aorta, progressive enlargement of an ascending thoracic aortic aneurysm leads to a type A dissection; thus, thoracic aortic aneurysms and type A dissections (TAAD) are associated conditions.
TAAD is the major cardiovascular complication of Marfan syndrome (MFS), a pleiotropic disorder with involvement of the cardiovascular, ocular, and skeletal systems. MFS is inherited in an autosomal-dominant manner and is caused by mutations in the FBN1 gene on chromosome 15q.2 FBN1 encodes fibrillin-1, a large glycoprotein that is a component of extracellular matrix structures called microfibrils.3 A second locus for MFS, termed the MFS2 locus, was mapped to 3p24–25, and mutations in the transforming growth factor-; receptor type II (TGFBR2) have recently been described in patients with MFS.4
Clinical studies have indicated that up to 19% of nonsyndromic TAAD patients referred for surgery have affected first-degree relatives, supporting the hypothesis that genetic factors influence the formation of TAAD in individuals who do not have MFS.5,6 Aortic imaging of individuals at risk and analysis of the pedigrees determined that the condition is inherited primarily as an autosomal-dominant disorder with decreased penetrance and variable expression.7 Mapping studies in families with TAAD have identified 3 loci for the condition, TAAD1 on 5q13–14, TAAD2 on 3p24–25, and FAA1 on 11q.8–10 The TAAD2 locus encompasses the MFS2 locus, raising the possibility that these conditions are allelic.
The TGFBR2 gene falls within the 25-cM critical interval of the TAAD2 locus; therefore, we used DNA samples from affected members of 1 large family that were used to map TAAD2, along with samples from probands from 79 TAAD families that were not conclusively linked to another locus for sequencing the gene. Based on current diagnostic criteria, these families did not have MFS.11 TGFBR2 mutations were identified in 4 unrelated families, and all mutations affected arginine 460 in the cytoplasmic serine-threonine kinase domain of the receptor. The cardiovascular phenotype associated with TGFBR2 mutations in these families was predominantly ascending aortic disease; however, significant descending aortic disease and aneurysms of other vessels also occurred in affected family members.
Methods
Patient Groups
This study was approved by the Institutional Review Board at the University of Texas Health Science Center at Houston. Families with multiple members with ascending thoracic aortic aneurysms and/or aortic dissections were recruited for research. Clinical information and family histories were collected from medical records and interviews conducted by a research nurse or genetics counselor. Blood or buccal cell samples and medical records pertaining to cardiovascular disease were collected after consent was obtained. Family members had echocardiograms performed to assess heart structure and function and aortic root dimensions. Aortic diameters at the sinuses of Valsalva, the supra-aortic ridge, and the ascending aorta were measured from cross-sectional echocardiography images in the parasternal long-axis orientation and plotted against nomograms derived from normal individuals’ measurements. Individuals were considered affected if they had a true aneurysm, dissection of the thoracic aorta, or preaneurysmal dilation of the ascending aorta.12 Family members were examined for other skeletal features of MFS, including pectus deformities, highly arched palate, joint contractures, joint hypermobility, stria atrophica, kyphoscoliosis, and arachnodactyly, and ophthalmologic examinations were performed. Families with MFS were excluded from this study. The probands of the families with TGFBR2 mutations were all examined by one of the coauthors (E.S., P.G., M.W., or D.M.M.) and did not meet the current diagnostic criteria for MFS.11 DNA was obtained from 100 individuals without known aortic disease for use as controls.
Mutational Analysis
Genomic DNA was extracted from samples (peripheral blood or buccal cells) according to the manufacturer’s protocol with the PureGene genomic DNA isolation kit (Gentra Systems). Mutational analysis of the TGFBR2 gene (NM_003242) was performed by bidirectional direct sequencing of amplified genomic DNA fragments and intron-based, exon-specific primers (primer sequences and amplification conditions are available at http://www.uth.tmc.edu/schools/med/imed/med_gen/tgfbr2.htm). Sequencing was performed with Big Dye chemistry, and products were analyzed on the ABI 3100 genetic analyzer (Applied Biosystems). The visually inspected sequence was analyzed by BLAST alignment to identify alterations. All alterations were verified by a second independent amplification and sequencing reaction. DNA from family members of probands with mutations was sequenced to determine whether the mutation was present.
Genotyping
Three microsatellite markers within and flanking TGFBR2 were selected from the UniSTS database for construction of regional haplotypes. Amplification was performed with fluorescence-tagged primers. Products were run on the ABI 3100 genetic analyzer and alleles assigned with Genescan software (Applied Biosystems). In addition, 3 polymorphisms, rs10537389, rs5847638, and rs11466521, located at a distance of 657 and 504 bp upstream and 256 bp downstream, respectively, from the mutation were genotyped by direct sequencing in individuals from the 4 families to ascertain whether the mutations were associated with the same alleles in families with identical mutations (Figure 2). To correctly establish the phase for these markers the following individuals were sequenced: (1) individuals III:5, III:6, and IV:7 in TAA035; (2) individuals III:2, III:3, and IV:2 in TAA150; (3) individuals V:18, V:20, and VI:10 in TAA067; and (4) individuals IV:4, V:5, and V:7 in TAA090 (Figures 1 and 2).
Protein Structure
The program MODELLER was used to construct a homology model of TGFBR2.13 Because the kinase domains of TGFBR1 and TGFBR2 share 41% sequence identity, we built the model on the basis of the crystal structure of TGFBR1. As of October 18, 2004, all TGFBR1 kinase structures deposited in the Protein Data Bank have been determined in the inactive conformation, with either a small-molecule inhibitor bound at the ATP-binding site or with the TGFBR2 phosphorylation sites blocked by another protein.14,15 The structure of a constitutively active (T204D) TGFBR1 kinase has been solved, but the coordinates are not yet available.15 Therefore, we used the highest-resolution structure (2.3 ;) of TGFBR1 bound to an inhibitor (Protein Data Bank identification, 1PY5) as our template. This strategy is valid because the C-terminal lobe (the region in which the TAAD mutations that we observed are located) of all known protein kinases does not undergo major conformational changes on phosphorylation, as unambiguously borne out in all known crystal structures.16 Nine models of the highest optimization level were built for human TGFBR2, and the one with the lowest objective function was selected for further analysis. Model quality was assessed with the programs PROSA II and PROFILE-3D.17,18
Statistical Analysis
We obtained confidence intervals by the exact binomial test provided with statistical software for exact inference (STATXACT 6.0, Cytel Software Cambridge).19
Results
Family TAA035, a large TAAD kindred whose data were used to map the TAAD2 locus to 3p24–25, was found to have the TGFBR2 missense mutation 1378CT in exon 5, resulting in the substitution of cysteine for an arginine at amino acid 460 (R460C) (Figure 1). The mutation segregated with the disease in this family, and this alteration was not found on 200 unrelated chromosomes from ethnically matched white controls. The clinical features of this family were previously reported.9
The same TGFBR2 mutation, 1378CT in exon 5, was present in the proband from family TAA150 (Figure 1, Table 1). The proband (IV:4) presented at 41 years of age with an ascending aneurysm and mitral valve prolapse. The proband was a member of a 4-generation family with autosomal-dominant inheritance of TAAD. Family members primarily presented with type A dissections. Individual III:2 underwent aortic imaging and was found to have an ascending aneurysm of 4.5 cm at the level of the sinuses of Valsalva at the age of 56 years. The mutation segregated with aortic disease in this family.
Another TGFBR2 mutation, 1379GA, also altering arginine 460 but to a histidine (R460H), was identified in 2 large TAAD kindreds, TAA090 and TAA067 (Figure 1, Table 1). This alteration segregated with the disease in these families and was not found on 200 unrelated chromosomes from ethnically matched white controls. The proband of family TAA090 presented with type B aortic dissection at the age of 43 years and was treated medically. During the next 12 years, she had 2 operations to repair her ascending aorta, aortic valve, and aortic arch. The proband’s family demonstrated dominant inheritance of aortic disease, with affected individuals presenting with both ascending and descending thoracic aortic disease. In addition, some affected individuals also had carotid and cerebral aneurysms and dissections, as well as enlargement of the pulmonary artery. In family TAA067, the proband (V:18) presented at age 42 years with a type A aortic dissection that was surgically repaired. Family TAA067 demonstrated autosomal-dominant inheritance of aortic disease, with the majority of individuals presenting with ascending thoracic aortic disease.
In families TAA035, TAA090, and TAA067, there are individuals older the age of 18 years who carry the TGFBR2 mutation leading to aortic disease in their family but who do not have documented aortic disease to date, demonstrating the decreased penetrance of the mutation described previously. All 5 of these individuals are female.
In addition to the 2 missense mutations in TGFBR2, 3 known single-nucleotide polymorphisms (SNPs) were identified in the patient population (Table 2). These SNPs occurred at frequencies comparable to those listed in the NCBI database. No additional disease-associated alterations were observed.
Three microsatellite markers and 3 polymorphisms within or near TGFBR2 were analyzed to determine whether the 1378CT mutation occurred on the same haplotype in families TAA035 and TAA150 (Figure 2). Likewise, these same microsatellites and polymorphisms were also analyzed in families TAA067 and TAA090. Analysis of the haplotype segregating with the mutation in the family indicated that the mutation occurred on a unique haplotype in each family, indicating that these families are not distantly related (Figure 2). These data also suggest that the mutation arose independently in families TAA035 and TAA150 and in families TAA067 and TAA090.
To identify the structural basis of the contribution of mutations at R460 to familial TAAD, we constructed a homology model of the TGFBR2 cytoplasmic domain by using the crystal structure of the TGFBR1 kinase domain as a template (Figure 3A).14,15 These 2 domains share a high degree (41%) of amino acid sequence identity and therefore have the same overall structure. Furthermore, all protein kinases, even those with low sequence similarity, have a typical canonical bilobal structure in which an N-terminal lobe rich in ;-strands is linked to a C-terminal lobe (C-lobe) dominated by -helixes.21 Our model reveals that R460 is located at the end of the F-helix in the C-lobe of TGFBR2 kinase and forms 3 key hydrogen bonds with the D-helix, 2 of which involve its guanidinium side chain (Figure 3A and 3B). This implies a key role for the R460 residue in maintaining the structural integrity of the TGFBR2 catalytic loop for effective signaling.
Discussion
We established that TGFBR2 mutations are a cause of thoracic aortic aneurysms and dissections by identifying mutations in 4 unrelated families. Our data suggest that mutations in TGFBR2 are responsible for 5% of familial TAAD (95% confidence interval, 1.4% to 12.3%). Consistent with this, our previous linkage study provided a similar frequency for familial disease linked to chromosome 3p24–25.9 Although the primary vascular disease in family members with TGFBR2 mutations was progressive enlargement of the ascending aorta, leading to type A dissection, affected individuals developed aneurysms and dissections involving the descending aorta and aneurysms of other arteries. Family TAA090 had 4 affected members who presented with type B dissections, 3 individuals with cerebral aneurysms, 2 individuals with carotid artery aneurysms, and 1 individual with a dilated pulmonary artery. Although cerebral aneurysms also aggregate in families, only individuals with aortic disease had cerebral aneurysms in family TAA090, suggesting that these vascular diseases were segregating together. Family TAA067 had an affected member with a renal artery aneurysm and a dilated descending thoracic aorta and another individual who died of a cerebral aneurysm. Family TAA035 had an affected member who presented with an abdominal aortic aneurysm and popliteal aneurysm.9
To our knowledge, the TGFBR2 mutations revealed by our study are unique and have not been reported in patients with cancer or MFS.4,22 Indeed, none of the affected members of the families with TGFBR2 mutations met the current diagnostic criteria of MFS.11 Family members in TAA035 were extensively evaluated, and those affected did not have significant skeletal or ocular features of the condition.9 Therefore, TGFBR2 mutations affecting R460 appear to lead to familial TAAD and not MFS, indicating an association between the genotype of R460 alterations and familial TAAD.
Polymorphic markers flanking the TGFBR2 mutation indicate that the families are not related, providing evidence for a "hot spot" for mutations causing this disease. Previously identified TGFBR2 mutations in patients with MFS were more diverse than the mutations identified in families with TAAD, 3 missense mutations (L308P, S449F, and R537C) and a splicing error of exon 6. These TGFBR2 mutations in MFS patients also involved the serine-threonine kinase domain and have been determined to diminish receptor signaling induced by TGF-; when coexpressed with a TGF-; responsive promoter in an in vitro assay system. It is interesting to note that none of the TGFBR2 mutations identified in this study involve the extracellular ligand-binding domain of the receptor, leading to speculation that the pathogenesis of the aortic disease may be dependent on specific inactivation of the intracellular kinase domain but preservation of the extracellular TGF-;–binding domain.
TGF-; superfamily members signal through heteromeric complexes of type II and type I transmembrane serine-threonine kinase receptors.23 TGF-; induces the assembly of a heteromeric receptor complex of type I and type II receptors, within which the TGF;R2 transphosphorylates and activates the type I receptor. Codon 460, altered in the families, is an invariant arginine residue within the TGF-; receptor family. Its position is significant in light of our structural findings based on homology modeling of the TGFBR2 protein. The amino acid R460 is located at the end of the F-helix, which has conserved functions in other structurally well-studied protein kinases, such as the prototypical cAMP-dependent protein kinase A (PKA). In PKA, the F-helix functions as a signal transducer by conveying ATP binding at the active site to the substrate binding site.16 Although this helix does not directly interact with the substrate or ATP, it serves to maintain the structural integrity of the catalytic loop. On constructing the homology model of TGFBR2, we find that the interactions between the F-helix and the catalytic loop in TGFBR2 kinase are essentially identical to those observed in PKA (Figure 3B). Additionally, in the vast majority (>90%) of known protein kinase structures, the D-helix is either close to the substrate or makes direct contacts with it.24 Thus, mutations giving rise to histidine or cysteine substitutions at R460 will perturb F-helix–D-helix communications and diminish the signaling ability of TGFBR2. It is worth noting that somatic mutations that affect residues in the neighborhood of R460 have been reported in human cancers.22,25 Analogously, one of the TGFBR2 mutations found in MFS (Arg-537Cys located in the I-helix) also has a counterpart in human head and neck cancers (Arg-537Pro).4,22 Taken together, our structural findings suggest that alterations in the key C-lobe helixes (F and I) can dramatically perturb TGFBR2 signaling in response to TGF-;.26
The TGF-; signaling pathway generally has a negative effect on cell growth, and inactivation of this signaling pathway has been shown to contribute to tumorigenesis.27 TGFBR2 is inactivated by somatic mutations in human gastrointestinal tumors with microsatellite instability, along with a number of other cancers, including ovarian cancer, gliomas, and squamous cell cancer.22 The study of somatic TGFBR2 mutations in tumor cells has contributed to understanding the structure and function of the receptor. A number of somatic missense mutations occur in close proximity to R460, including M457T and T458A. Other missense mutations in the intracellular domain of TGFBR2 have been extensively studied, including P525L in exon 7.26 Studies of this mutant receptor in mink lung epithelial cells demonstrated that the mutant receptor bound ligand, autophosphorylated, formed a complex with TGF-; type I receptor, but failed to phosphorylate the associated TGF-; type receptor I. In contrast to TGFRBR2 mutations in cancer cells, the mutations in TAAD patients are heterozygous germline mutations; therefore, a wild type receptor will be expressed in cells in addition to the mutated receptor. Interestingly, there does not appear to be a strong predisposition to cancers in the TAAD families with TGFBR2 mutations. Although an affected individual in TAA090 died of an invasive carcinoma of the parotid gland at the age of 51 years (Figure 1, IV:1), there are no other known cancers in affected individuals despite the advanced age of some affected individuals. Therefore, mutations in TGFBR2 possibly contribute to human vascular disease and cancer through different pathways.
The bioavailability of active TGF-; ligand is tightly regulated and dependent on its release from a large latent complex to which TGF-; is noncovalently associated with its propeptide fragment, the latency-associated peptide, and covalently linked to latent TGF-;-binding protein.28 This complex associates with fibrillin-1–containing extracellular microfibrils, and mutations in fibrillin-1 lead to diminished amounts of microfibrils in the extracellular matrix.29 An accepted mouse model of MFS, fibrillin-1–deficient mice, demonstrated increased amounts of active TGF-; in tissues when compared with wild type mice, suggesting that diminished microfibrils, leading to upregulation of TGF-; signaling in tissue, may be responsible for manifestations of MFS.30 Furthermore, antagonism of active TGF-; prevented the pulmonary parenchymal abnormalities observed in these mice. It is difficult to reconcile a pathway of excessive TGF-; signaling as a cause of aortic disease in MFS when the TGFBR2 mutations identified in both MFS and familial TAAD are predicted to disrupt the serine-threonine kinase domain of the protein and result in diminished TGF-; signaling. Relevant to these facts is the surprising observation that fibroblast-specific expression of a kinase-deficient TGFBR2 in a transgenic mouse model leads to paradoxical activation of TGF-; signaling pathways.31 Further studies will delineate the effect of mutations in the TGFBR2 on TGF-; signaling and possibly lead to a common pathway of dysregulation of TGF-; signaling as a cause of aortic aneurysms and dissections.
Ascending aortic aneurysms are often asymptomatic, undiagnosed, and frequently lead to sudden and fatal type A dissections. In the vast majority of families with multiple affected members, the disease is inherited in an autosomal-dominant manner with decreased penetrance, primarily in women.7 In addition, there is variable expression of the disease within a family, with 1 member dying of a type A dissection in their 20s and another member dying in their 70s. These aspects of inheritance of the disorder make it difficult to determine who has not inherited the defective gene, and therefore, it is currently recommended that all individuals in a family at risk for inheriting the defective gene undergo imaging of their aorta routinely throughout their entire life. Although mutations in TGFBR2 are a rare cause of familial disease, the information provided to the family concerning risk for aortic disease is significant. Therefore, families with multiple members with TAAD should consider undergoing evaluation for TGFBR2 mutations. Individuals shown to carry TGFBR2 mutations should be advised to have surveillance of the thoracic aorta routinely and prophylactic repair of the aorta when the aortic diameter approaches 5.0 cm. Further studies are needed to determine whether more extensive aortic/arterial imaging may be warranted in families with TGFBR2 mutations.
Acknowledgments
The following sources provided funding for these studies: RO1 HL62594 (Dr Milewicz), MO1RR02558 (General Clinical Research Center), and the Texgen Research Foundation. Dr Raman is a Pew Scholar. Dr Milewicz is a Doris Duke Distinguished Clinical Scientist. We thank the patients and their families for participating in these studies.
References
Lilienfeld DE, Gunderson PD, Sprafka JM, Vargas C. Epidemiology of aortic aneurysms, I: mortality trends in the United States, 1951 to 1981. Arteriosclerosis. 1987; 7: 637–643.
Dietz HC, Pyeritz RE. Mutations in the human gene for fibrillin-1 (FBN1) in the Marfan syndrome and related disorders. Hum Mol Genet. 1995; 4: 1799–1809.
Handford PA, Downing AK, Reinhardt DP, Sakai LY. Fibrillin: from domain structure to supramolecular assembly. Matrix Biol. 2000; 19: 457–470.
Mizuguchi T, Collod-Beroud G, Akiyama T, Abifadel M, Harada N, Morisaki T, Allard D, Varret M, Claustres M, Morisaki H, Ihara M, Kinoshita A, Yoshiura K, Junien C, Kajii T, Jondeau G, Ohta T, Kishino T, Furukawa Y, Nakamura Y, Niikawa N, Boileau C, Matsumoto N. Heterozygous TGFBR2 mutations in Marfan syndrome. Nat Genet. 2004; 36: 855–860.
Biddinger A, Rocklin M, Coselli J, Milewicz DM. Familial thoracic aortic dilatations and dissections: a case control study. J Vasc Surg. 1997; 25: 506–511.
Coady MA, Davies RR, Roberts M, Goldstein LJ, Rogalski MJ, Rizzo JA Hammond GL, Kopf GS, Elefteriades JA. Familial patterns of thoracic aortic aneurysms. Arch Surg. 1999; 134: 361–367.
Milewicz DM, Chen H, Park ES, Petty EM, Zaghi H, Shashidhar G, Willing M, Patel V. Reduced penetrance and variable expressivity of familial thoracic aortic aneurysms/dissections. Am J Cardiol. 1998; 82: 474–479.
Guo D, Hasham S, Kuang SQ, Vaughan CJ, Boerwinkle E, Chen H, Abuelo D, Dietz HC, Basson CT, Shete SS, Milewicz DM. Familial thoracic aortic aneurysms and dissections: genetic heterogeneity with a major locus mapping to 5q13–14. Circulation. 2001; 103: 2461–2468.
Hasham SN, Willing MC, Guo DC, Muilenburg A, He R, Tran VT, Scherer SE, Shete SS, Milewicz DM. Mapping a locus for familial thoracic aortic aneurysms and dissections (TAAD2) to 3p24–25. Circulation. 2003; 107: 3184–3190.
Vaughan CJ, Casey M, He J, Veugelers M, Henderson K, Guo D, Campagna R, Roman MJ, Milewicz DM, Devereux RB, Basson CT. Identification of a chromosome 11q23.2-q24 locus for familial aortic aneurysm disease, a genetically heterogeneous disorder. Circulation. 2001; 103: 2469–2475.
De Paepe A, Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet. 1996; 62: 417–426.
Roman MJ, Devereux RB, Kramer-Fox R, O’Loughlin J. Two-dimensional echocardiographic aortic root dimensions in normal children and adults. Am J Cardiol. 1989; 64: 507–512.
Sali A, Blundell TL. Comparative protein modelling by satisfaction of spatial restraints. J Mol Biol. 1993; 234: 779–815.
Huse M, Chen YG, Massague J, Kuriyan J. Crystal structure of the cytoplasmic domain of the type I TGF; receptor in complex with FKBP12. Cell. 1999; 96: 425–436.
Sawyer JS, Anderson BD, Beight DW, Campbell RM, Jones ML, Herron DK, Lampe JW, McCowan JR, McMillen WT, Mort N, Parsons S, Smith EC, Vieth M, Weir LC, Yan L, Zhang F, Yingling JM. Synthesis and activity of new aryl- and heteroaryl-substituted pyrazole inhibitors of the transforming growth factor-; type I receptor kinase domain. J Med Chem. 2003; 46: 3953–3956.
Taylor SS, Yang J, Wu J, Haste NM, Radzio-Andzelm E, Anand G. PKA: a portrait of protein kinase dynamics. Biochim Biophys Acta. 2004; 1697: 259–269.
Sippl MJ. Recognition of errors in three-dimensional structures of proteins. Proteins. 1993; 17: 355–362.
Luthy R, Bowie JU, Eisenberg D. Assessment of protein models with three- dimensional profiles. Nature. 1992; 356: 83–85.
Lathrop GM, Lalouel JM, Julier C, Ott J. Multilocus linkage analysis in humans: detection of linkage and estimation of recombination. Am J Hum Genet. 1985; 37: 482–498.
Rotzer D, Roth M, Lutz M, Lindemann D, Sebald W, Knaus P. Type III TGF-; receptor-independent signalling of TGF-;2 via T;RII-B, an alternatively spliced TGF-; type II receptor. EMBO J. 2001; 20: 480–490.
Hardie G, Hanks S. The Protein Kinase Facts Book, Volume II. New York, NY: Academic Press; 1995.
Kim SJ, Im YH, Markowitz SD, Bang YJ. Molecular mechanisms of inactivation of TGF-; receptors during carcinogenesis. Cytokine Growth Factor Rev. 2000; 11: 159–168.
Shi Y, Massague J. Mechanisms of TGF-; signaling from cell membrane to the nucleus. Cell. 2003; 113: 685–700.
Akamine P, Madhusudan, Wu J, Xuong NH, Ten Eyck LF, Taylor SS. Dynamic features of cAMP-dependent protein kinase revealed by apoenzyme crystal structure. J Mol Biol. 2003; 327: 159–171.
Nerlich AG, Sauer U, Ruoss I, Hagedorn HG. High frequency of TGF-;-receptor-II mutations in microdissected tissue samples from laryngeal squamous cell carcinomas. Lab Invest. 2003; 83: 1241–1251.
Carcamo J, Zentella A, Massague J. Disruption of transforming growth factor-; signaling by a mutation that prevents transphosphorylation within the receptor complex. Mol Cell Biol. 1995; 15: 1573–1581.
Blobe GC, Schiemann WP, Lodish HF. Role of transforming growth factor-; in human disease. N Engl J Med. 2000; 342: 1350–1358.
Saharinen J, Hyytiainen M, Taipale J, Keski-Oja J. Latent transforming growth factor-; binding proteins (LTBPs)—structural extracellular matrix proteins for targeting TGF-; action. Cytokine Growth Factor Rev. 1999; 10: 99–117.
Milewicz DM, Pyeritz RE, Crawford ES, Byers PH. Marfan syndrome: defective synthesis, secretion, and extracellular matrix formation of fibrillin by cultured dermal fibroblasts. J Clin Invest. 1992; 89: 79–86.
Neptune ER, Frischmeyer PA, Arking DE, Myers L, Bunton TE, Gayraud B, Ramirez F, Sakai LY, Dietz HC. Dysregulation of TGF-; activation contributes to pathogenesis in Marfan syndrome. Nat Genet. 2003; 33: 407–411.
Denton CP, Zheng B, Evans LA, Shi-wen X, Ong VH, Fisher I, Lazaridis K, Abraham DJ, Black CM, de Crombrugghe B. Fibroblast-specific expression of a kinase-deficient type II transforming growth factor-; (TGF;) receptor leads to paradoxical activation of TGF-; signaling pathways with fibrosis in transgenic mice. J Biol Chem. 2003; 278: 25109–25119.(Hariyadarshi Pannu, PhD; )
the Department of Cardiothoracic and Vascular Surgery (A.L.E., H.J.S.)
the Structural Biology Research Center (C.S.R.), the University of Texas Health Science Center at Houston
the Department of Internal Medicine (E.S.), Cardiology Division, The Ohio State University, Columbus
the Department of Medical Genetic Services (P.F.G., C.Z.), Marshfield Clinic, Marshfield, Wis
the Department of Epidemiology (S.S.), the University of Texas M.D. Anderson Cancer Center, Houston, Tex
the Department of Pediatrics, the University of Iowa (M.C.W.), Iowa City.
Abstract
Background— A genetic predisposition for progressive enlargement of thoracic aortic aneurysms leading to type A dissection (TAAD) is inherited in an autosomal-dominant manner in up to 19% of patients, and a number of chromosomal loci have been identified for the condition. Having mapped a TAAD locus to 3p24–25, we sequenced the gene for transforming growth factor-; receptor type II (TGFBR2) to determine whether mutations in this gene resulted in familial TAAD.
Methods and Results— We sequenced all 8 coding exons of TGFBR2 by using genomic DNA from 80 unrelated familial TAAD cases. We found TGFBR2 mutations in 4 unrelated families with familial TAAD who did not have Marfan syndrome. Affected family members also had descending aortic disease and aneurysms of other arteries. Strikingly, all 4 mutations affected an arginine residue at position 460 in the intracellular domain, suggesting a mutation "hot spot" for familial TAAD. Despite identical mutations in the families, assessment of linked polymorphisms suggested that these families were not distantly related. Structural analysis of the TGFBR2 serine/threonine kinase domain revealed that R460 is strategically located within a highly conserved region of this domain and that the amino acid substitutions resulting from these mutations will interfere with the receptor’s ability to transduce signals.
Conclusion— Germline TGFBR2 mutations are responsible for the inherited predisposition to familial TAAD in 5% of these cases. Our results have broad implications for understanding the role of TGF-; signaling in the pathophysiology of TAAD.
Key Words: aneurysm ; aorta ; genetics ; dissection ; receptors, transforming growth factor beta
Introduction
Aneurysms and dissections are the major diseases affecting the aorta and are a leading cause of morbidity and mortality in the United States.1 The most common location for aortic aneurysms is in the infrarenal abdominal aorta, followed by the ascending thoracic aorta. Aortic dissections are classified according to the anatomic location of the initial tear in the aortic wall, with type A dissections initiating in the ascending thoracic aortic just above the aortic valve and type B dissections originating in the descending thoracic aortic just beyond the take-off of the subclavian artery. Without prophylactic surgical repair of the aorta, progressive enlargement of an ascending thoracic aortic aneurysm leads to a type A dissection; thus, thoracic aortic aneurysms and type A dissections (TAAD) are associated conditions.
TAAD is the major cardiovascular complication of Marfan syndrome (MFS), a pleiotropic disorder with involvement of the cardiovascular, ocular, and skeletal systems. MFS is inherited in an autosomal-dominant manner and is caused by mutations in the FBN1 gene on chromosome 15q.2 FBN1 encodes fibrillin-1, a large glycoprotein that is a component of extracellular matrix structures called microfibrils.3 A second locus for MFS, termed the MFS2 locus, was mapped to 3p24–25, and mutations in the transforming growth factor-; receptor type II (TGFBR2) have recently been described in patients with MFS.4
Clinical studies have indicated that up to 19% of nonsyndromic TAAD patients referred for surgery have affected first-degree relatives, supporting the hypothesis that genetic factors influence the formation of TAAD in individuals who do not have MFS.5,6 Aortic imaging of individuals at risk and analysis of the pedigrees determined that the condition is inherited primarily as an autosomal-dominant disorder with decreased penetrance and variable expression.7 Mapping studies in families with TAAD have identified 3 loci for the condition, TAAD1 on 5q13–14, TAAD2 on 3p24–25, and FAA1 on 11q.8–10 The TAAD2 locus encompasses the MFS2 locus, raising the possibility that these conditions are allelic.
The TGFBR2 gene falls within the 25-cM critical interval of the TAAD2 locus; therefore, we used DNA samples from affected members of 1 large family that were used to map TAAD2, along with samples from probands from 79 TAAD families that were not conclusively linked to another locus for sequencing the gene. Based on current diagnostic criteria, these families did not have MFS.11 TGFBR2 mutations were identified in 4 unrelated families, and all mutations affected arginine 460 in the cytoplasmic serine-threonine kinase domain of the receptor. The cardiovascular phenotype associated with TGFBR2 mutations in these families was predominantly ascending aortic disease; however, significant descending aortic disease and aneurysms of other vessels also occurred in affected family members.
Methods
Patient Groups
This study was approved by the Institutional Review Board at the University of Texas Health Science Center at Houston. Families with multiple members with ascending thoracic aortic aneurysms and/or aortic dissections were recruited for research. Clinical information and family histories were collected from medical records and interviews conducted by a research nurse or genetics counselor. Blood or buccal cell samples and medical records pertaining to cardiovascular disease were collected after consent was obtained. Family members had echocardiograms performed to assess heart structure and function and aortic root dimensions. Aortic diameters at the sinuses of Valsalva, the supra-aortic ridge, and the ascending aorta were measured from cross-sectional echocardiography images in the parasternal long-axis orientation and plotted against nomograms derived from normal individuals’ measurements. Individuals were considered affected if they had a true aneurysm, dissection of the thoracic aorta, or preaneurysmal dilation of the ascending aorta.12 Family members were examined for other skeletal features of MFS, including pectus deformities, highly arched palate, joint contractures, joint hypermobility, stria atrophica, kyphoscoliosis, and arachnodactyly, and ophthalmologic examinations were performed. Families with MFS were excluded from this study. The probands of the families with TGFBR2 mutations were all examined by one of the coauthors (E.S., P.G., M.W., or D.M.M.) and did not meet the current diagnostic criteria for MFS.11 DNA was obtained from 100 individuals without known aortic disease for use as controls.
Mutational Analysis
Genomic DNA was extracted from samples (peripheral blood or buccal cells) according to the manufacturer’s protocol with the PureGene genomic DNA isolation kit (Gentra Systems). Mutational analysis of the TGFBR2 gene (NM_003242) was performed by bidirectional direct sequencing of amplified genomic DNA fragments and intron-based, exon-specific primers (primer sequences and amplification conditions are available at http://www.uth.tmc.edu/schools/med/imed/med_gen/tgfbr2.htm). Sequencing was performed with Big Dye chemistry, and products were analyzed on the ABI 3100 genetic analyzer (Applied Biosystems). The visually inspected sequence was analyzed by BLAST alignment to identify alterations. All alterations were verified by a second independent amplification and sequencing reaction. DNA from family members of probands with mutations was sequenced to determine whether the mutation was present.
Genotyping
Three microsatellite markers within and flanking TGFBR2 were selected from the UniSTS database for construction of regional haplotypes. Amplification was performed with fluorescence-tagged primers. Products were run on the ABI 3100 genetic analyzer and alleles assigned with Genescan software (Applied Biosystems). In addition, 3 polymorphisms, rs10537389, rs5847638, and rs11466521, located at a distance of 657 and 504 bp upstream and 256 bp downstream, respectively, from the mutation were genotyped by direct sequencing in individuals from the 4 families to ascertain whether the mutations were associated with the same alleles in families with identical mutations (Figure 2). To correctly establish the phase for these markers the following individuals were sequenced: (1) individuals III:5, III:6, and IV:7 in TAA035; (2) individuals III:2, III:3, and IV:2 in TAA150; (3) individuals V:18, V:20, and VI:10 in TAA067; and (4) individuals IV:4, V:5, and V:7 in TAA090 (Figures 1 and 2).
Protein Structure
The program MODELLER was used to construct a homology model of TGFBR2.13 Because the kinase domains of TGFBR1 and TGFBR2 share 41% sequence identity, we built the model on the basis of the crystal structure of TGFBR1. As of October 18, 2004, all TGFBR1 kinase structures deposited in the Protein Data Bank have been determined in the inactive conformation, with either a small-molecule inhibitor bound at the ATP-binding site or with the TGFBR2 phosphorylation sites blocked by another protein.14,15 The structure of a constitutively active (T204D) TGFBR1 kinase has been solved, but the coordinates are not yet available.15 Therefore, we used the highest-resolution structure (2.3 ;) of TGFBR1 bound to an inhibitor (Protein Data Bank identification, 1PY5) as our template. This strategy is valid because the C-terminal lobe (the region in which the TAAD mutations that we observed are located) of all known protein kinases does not undergo major conformational changes on phosphorylation, as unambiguously borne out in all known crystal structures.16 Nine models of the highest optimization level were built for human TGFBR2, and the one with the lowest objective function was selected for further analysis. Model quality was assessed with the programs PROSA II and PROFILE-3D.17,18
Statistical Analysis
We obtained confidence intervals by the exact binomial test provided with statistical software for exact inference (STATXACT 6.0, Cytel Software Cambridge).19
Results
Family TAA035, a large TAAD kindred whose data were used to map the TAAD2 locus to 3p24–25, was found to have the TGFBR2 missense mutation 1378CT in exon 5, resulting in the substitution of cysteine for an arginine at amino acid 460 (R460C) (Figure 1). The mutation segregated with the disease in this family, and this alteration was not found on 200 unrelated chromosomes from ethnically matched white controls. The clinical features of this family were previously reported.9
The same TGFBR2 mutation, 1378CT in exon 5, was present in the proband from family TAA150 (Figure 1, Table 1). The proband (IV:4) presented at 41 years of age with an ascending aneurysm and mitral valve prolapse. The proband was a member of a 4-generation family with autosomal-dominant inheritance of TAAD. Family members primarily presented with type A dissections. Individual III:2 underwent aortic imaging and was found to have an ascending aneurysm of 4.5 cm at the level of the sinuses of Valsalva at the age of 56 years. The mutation segregated with aortic disease in this family.
Another TGFBR2 mutation, 1379GA, also altering arginine 460 but to a histidine (R460H), was identified in 2 large TAAD kindreds, TAA090 and TAA067 (Figure 1, Table 1). This alteration segregated with the disease in these families and was not found on 200 unrelated chromosomes from ethnically matched white controls. The proband of family TAA090 presented with type B aortic dissection at the age of 43 years and was treated medically. During the next 12 years, she had 2 operations to repair her ascending aorta, aortic valve, and aortic arch. The proband’s family demonstrated dominant inheritance of aortic disease, with affected individuals presenting with both ascending and descending thoracic aortic disease. In addition, some affected individuals also had carotid and cerebral aneurysms and dissections, as well as enlargement of the pulmonary artery. In family TAA067, the proband (V:18) presented at age 42 years with a type A aortic dissection that was surgically repaired. Family TAA067 demonstrated autosomal-dominant inheritance of aortic disease, with the majority of individuals presenting with ascending thoracic aortic disease.
In families TAA035, TAA090, and TAA067, there are individuals older the age of 18 years who carry the TGFBR2 mutation leading to aortic disease in their family but who do not have documented aortic disease to date, demonstrating the decreased penetrance of the mutation described previously. All 5 of these individuals are female.
In addition to the 2 missense mutations in TGFBR2, 3 known single-nucleotide polymorphisms (SNPs) were identified in the patient population (Table 2). These SNPs occurred at frequencies comparable to those listed in the NCBI database. No additional disease-associated alterations were observed.
Three microsatellite markers and 3 polymorphisms within or near TGFBR2 were analyzed to determine whether the 1378CT mutation occurred on the same haplotype in families TAA035 and TAA150 (Figure 2). Likewise, these same microsatellites and polymorphisms were also analyzed in families TAA067 and TAA090. Analysis of the haplotype segregating with the mutation in the family indicated that the mutation occurred on a unique haplotype in each family, indicating that these families are not distantly related (Figure 2). These data also suggest that the mutation arose independently in families TAA035 and TAA150 and in families TAA067 and TAA090.
To identify the structural basis of the contribution of mutations at R460 to familial TAAD, we constructed a homology model of the TGFBR2 cytoplasmic domain by using the crystal structure of the TGFBR1 kinase domain as a template (Figure 3A).14,15 These 2 domains share a high degree (41%) of amino acid sequence identity and therefore have the same overall structure. Furthermore, all protein kinases, even those with low sequence similarity, have a typical canonical bilobal structure in which an N-terminal lobe rich in ;-strands is linked to a C-terminal lobe (C-lobe) dominated by -helixes.21 Our model reveals that R460 is located at the end of the F-helix in the C-lobe of TGFBR2 kinase and forms 3 key hydrogen bonds with the D-helix, 2 of which involve its guanidinium side chain (Figure 3A and 3B). This implies a key role for the R460 residue in maintaining the structural integrity of the TGFBR2 catalytic loop for effective signaling.
Discussion
We established that TGFBR2 mutations are a cause of thoracic aortic aneurysms and dissections by identifying mutations in 4 unrelated families. Our data suggest that mutations in TGFBR2 are responsible for 5% of familial TAAD (95% confidence interval, 1.4% to 12.3%). Consistent with this, our previous linkage study provided a similar frequency for familial disease linked to chromosome 3p24–25.9 Although the primary vascular disease in family members with TGFBR2 mutations was progressive enlargement of the ascending aorta, leading to type A dissection, affected individuals developed aneurysms and dissections involving the descending aorta and aneurysms of other arteries. Family TAA090 had 4 affected members who presented with type B dissections, 3 individuals with cerebral aneurysms, 2 individuals with carotid artery aneurysms, and 1 individual with a dilated pulmonary artery. Although cerebral aneurysms also aggregate in families, only individuals with aortic disease had cerebral aneurysms in family TAA090, suggesting that these vascular diseases were segregating together. Family TAA067 had an affected member with a renal artery aneurysm and a dilated descending thoracic aorta and another individual who died of a cerebral aneurysm. Family TAA035 had an affected member who presented with an abdominal aortic aneurysm and popliteal aneurysm.9
To our knowledge, the TGFBR2 mutations revealed by our study are unique and have not been reported in patients with cancer or MFS.4,22 Indeed, none of the affected members of the families with TGFBR2 mutations met the current diagnostic criteria of MFS.11 Family members in TAA035 were extensively evaluated, and those affected did not have significant skeletal or ocular features of the condition.9 Therefore, TGFBR2 mutations affecting R460 appear to lead to familial TAAD and not MFS, indicating an association between the genotype of R460 alterations and familial TAAD.
Polymorphic markers flanking the TGFBR2 mutation indicate that the families are not related, providing evidence for a "hot spot" for mutations causing this disease. Previously identified TGFBR2 mutations in patients with MFS were more diverse than the mutations identified in families with TAAD, 3 missense mutations (L308P, S449F, and R537C) and a splicing error of exon 6. These TGFBR2 mutations in MFS patients also involved the serine-threonine kinase domain and have been determined to diminish receptor signaling induced by TGF-; when coexpressed with a TGF-; responsive promoter in an in vitro assay system. It is interesting to note that none of the TGFBR2 mutations identified in this study involve the extracellular ligand-binding domain of the receptor, leading to speculation that the pathogenesis of the aortic disease may be dependent on specific inactivation of the intracellular kinase domain but preservation of the extracellular TGF-;–binding domain.
TGF-; superfamily members signal through heteromeric complexes of type II and type I transmembrane serine-threonine kinase receptors.23 TGF-; induces the assembly of a heteromeric receptor complex of type I and type II receptors, within which the TGF;R2 transphosphorylates and activates the type I receptor. Codon 460, altered in the families, is an invariant arginine residue within the TGF-; receptor family. Its position is significant in light of our structural findings based on homology modeling of the TGFBR2 protein. The amino acid R460 is located at the end of the F-helix, which has conserved functions in other structurally well-studied protein kinases, such as the prototypical cAMP-dependent protein kinase A (PKA). In PKA, the F-helix functions as a signal transducer by conveying ATP binding at the active site to the substrate binding site.16 Although this helix does not directly interact with the substrate or ATP, it serves to maintain the structural integrity of the catalytic loop. On constructing the homology model of TGFBR2, we find that the interactions between the F-helix and the catalytic loop in TGFBR2 kinase are essentially identical to those observed in PKA (Figure 3B). Additionally, in the vast majority (>90%) of known protein kinase structures, the D-helix is either close to the substrate or makes direct contacts with it.24 Thus, mutations giving rise to histidine or cysteine substitutions at R460 will perturb F-helix–D-helix communications and diminish the signaling ability of TGFBR2. It is worth noting that somatic mutations that affect residues in the neighborhood of R460 have been reported in human cancers.22,25 Analogously, one of the TGFBR2 mutations found in MFS (Arg-537Cys located in the I-helix) also has a counterpart in human head and neck cancers (Arg-537Pro).4,22 Taken together, our structural findings suggest that alterations in the key C-lobe helixes (F and I) can dramatically perturb TGFBR2 signaling in response to TGF-;.26
The TGF-; signaling pathway generally has a negative effect on cell growth, and inactivation of this signaling pathway has been shown to contribute to tumorigenesis.27 TGFBR2 is inactivated by somatic mutations in human gastrointestinal tumors with microsatellite instability, along with a number of other cancers, including ovarian cancer, gliomas, and squamous cell cancer.22 The study of somatic TGFBR2 mutations in tumor cells has contributed to understanding the structure and function of the receptor. A number of somatic missense mutations occur in close proximity to R460, including M457T and T458A. Other missense mutations in the intracellular domain of TGFBR2 have been extensively studied, including P525L in exon 7.26 Studies of this mutant receptor in mink lung epithelial cells demonstrated that the mutant receptor bound ligand, autophosphorylated, formed a complex with TGF-; type I receptor, but failed to phosphorylate the associated TGF-; type receptor I. In contrast to TGFRBR2 mutations in cancer cells, the mutations in TAAD patients are heterozygous germline mutations; therefore, a wild type receptor will be expressed in cells in addition to the mutated receptor. Interestingly, there does not appear to be a strong predisposition to cancers in the TAAD families with TGFBR2 mutations. Although an affected individual in TAA090 died of an invasive carcinoma of the parotid gland at the age of 51 years (Figure 1, IV:1), there are no other known cancers in affected individuals despite the advanced age of some affected individuals. Therefore, mutations in TGFBR2 possibly contribute to human vascular disease and cancer through different pathways.
The bioavailability of active TGF-; ligand is tightly regulated and dependent on its release from a large latent complex to which TGF-; is noncovalently associated with its propeptide fragment, the latency-associated peptide, and covalently linked to latent TGF-;-binding protein.28 This complex associates with fibrillin-1–containing extracellular microfibrils, and mutations in fibrillin-1 lead to diminished amounts of microfibrils in the extracellular matrix.29 An accepted mouse model of MFS, fibrillin-1–deficient mice, demonstrated increased amounts of active TGF-; in tissues when compared with wild type mice, suggesting that diminished microfibrils, leading to upregulation of TGF-; signaling in tissue, may be responsible for manifestations of MFS.30 Furthermore, antagonism of active TGF-; prevented the pulmonary parenchymal abnormalities observed in these mice. It is difficult to reconcile a pathway of excessive TGF-; signaling as a cause of aortic disease in MFS when the TGFBR2 mutations identified in both MFS and familial TAAD are predicted to disrupt the serine-threonine kinase domain of the protein and result in diminished TGF-; signaling. Relevant to these facts is the surprising observation that fibroblast-specific expression of a kinase-deficient TGFBR2 in a transgenic mouse model leads to paradoxical activation of TGF-; signaling pathways.31 Further studies will delineate the effect of mutations in the TGFBR2 on TGF-; signaling and possibly lead to a common pathway of dysregulation of TGF-; signaling as a cause of aortic aneurysms and dissections.
Ascending aortic aneurysms are often asymptomatic, undiagnosed, and frequently lead to sudden and fatal type A dissections. In the vast majority of families with multiple affected members, the disease is inherited in an autosomal-dominant manner with decreased penetrance, primarily in women.7 In addition, there is variable expression of the disease within a family, with 1 member dying of a type A dissection in their 20s and another member dying in their 70s. These aspects of inheritance of the disorder make it difficult to determine who has not inherited the defective gene, and therefore, it is currently recommended that all individuals in a family at risk for inheriting the defective gene undergo imaging of their aorta routinely throughout their entire life. Although mutations in TGFBR2 are a rare cause of familial disease, the information provided to the family concerning risk for aortic disease is significant. Therefore, families with multiple members with TAAD should consider undergoing evaluation for TGFBR2 mutations. Individuals shown to carry TGFBR2 mutations should be advised to have surveillance of the thoracic aorta routinely and prophylactic repair of the aorta when the aortic diameter approaches 5.0 cm. Further studies are needed to determine whether more extensive aortic/arterial imaging may be warranted in families with TGFBR2 mutations.
Acknowledgments
The following sources provided funding for these studies: RO1 HL62594 (Dr Milewicz), MO1RR02558 (General Clinical Research Center), and the Texgen Research Foundation. Dr Raman is a Pew Scholar. Dr Milewicz is a Doris Duke Distinguished Clinical Scientist. We thank the patients and their families for participating in these studies.
References
Lilienfeld DE, Gunderson PD, Sprafka JM, Vargas C. Epidemiology of aortic aneurysms, I: mortality trends in the United States, 1951 to 1981. Arteriosclerosis. 1987; 7: 637–643.
Dietz HC, Pyeritz RE. Mutations in the human gene for fibrillin-1 (FBN1) in the Marfan syndrome and related disorders. Hum Mol Genet. 1995; 4: 1799–1809.
Handford PA, Downing AK, Reinhardt DP, Sakai LY. Fibrillin: from domain structure to supramolecular assembly. Matrix Biol. 2000; 19: 457–470.
Mizuguchi T, Collod-Beroud G, Akiyama T, Abifadel M, Harada N, Morisaki T, Allard D, Varret M, Claustres M, Morisaki H, Ihara M, Kinoshita A, Yoshiura K, Junien C, Kajii T, Jondeau G, Ohta T, Kishino T, Furukawa Y, Nakamura Y, Niikawa N, Boileau C, Matsumoto N. Heterozygous TGFBR2 mutations in Marfan syndrome. Nat Genet. 2004; 36: 855–860.
Biddinger A, Rocklin M, Coselli J, Milewicz DM. Familial thoracic aortic dilatations and dissections: a case control study. J Vasc Surg. 1997; 25: 506–511.
Coady MA, Davies RR, Roberts M, Goldstein LJ, Rogalski MJ, Rizzo JA Hammond GL, Kopf GS, Elefteriades JA. Familial patterns of thoracic aortic aneurysms. Arch Surg. 1999; 134: 361–367.
Milewicz DM, Chen H, Park ES, Petty EM, Zaghi H, Shashidhar G, Willing M, Patel V. Reduced penetrance and variable expressivity of familial thoracic aortic aneurysms/dissections. Am J Cardiol. 1998; 82: 474–479.
Guo D, Hasham S, Kuang SQ, Vaughan CJ, Boerwinkle E, Chen H, Abuelo D, Dietz HC, Basson CT, Shete SS, Milewicz DM. Familial thoracic aortic aneurysms and dissections: genetic heterogeneity with a major locus mapping to 5q13–14. Circulation. 2001; 103: 2461–2468.
Hasham SN, Willing MC, Guo DC, Muilenburg A, He R, Tran VT, Scherer SE, Shete SS, Milewicz DM. Mapping a locus for familial thoracic aortic aneurysms and dissections (TAAD2) to 3p24–25. Circulation. 2003; 107: 3184–3190.
Vaughan CJ, Casey M, He J, Veugelers M, Henderson K, Guo D, Campagna R, Roman MJ, Milewicz DM, Devereux RB, Basson CT. Identification of a chromosome 11q23.2-q24 locus for familial aortic aneurysm disease, a genetically heterogeneous disorder. Circulation. 2001; 103: 2469–2475.
De Paepe A, Devereux RB, Dietz HC, Hennekam RC, Pyeritz RE. Revised diagnostic criteria for the Marfan syndrome. Am J Med Genet. 1996; 62: 417–426.
Roman MJ, Devereux RB, Kramer-Fox R, O’Loughlin J. Two-dimensional echocardiographic aortic root dimensions in normal children and adults. Am J Cardiol. 1989; 64: 507–512.
Sali A, Blundell TL. Comparative protein modelling by satisfaction of spatial restraints. J Mol Biol. 1993; 234: 779–815.
Huse M, Chen YG, Massague J, Kuriyan J. Crystal structure of the cytoplasmic domain of the type I TGF; receptor in complex with FKBP12. Cell. 1999; 96: 425–436.
Sawyer JS, Anderson BD, Beight DW, Campbell RM, Jones ML, Herron DK, Lampe JW, McCowan JR, McMillen WT, Mort N, Parsons S, Smith EC, Vieth M, Weir LC, Yan L, Zhang F, Yingling JM. Synthesis and activity of new aryl- and heteroaryl-substituted pyrazole inhibitors of the transforming growth factor-; type I receptor kinase domain. J Med Chem. 2003; 46: 3953–3956.
Taylor SS, Yang J, Wu J, Haste NM, Radzio-Andzelm E, Anand G. PKA: a portrait of protein kinase dynamics. Biochim Biophys Acta. 2004; 1697: 259–269.
Sippl MJ. Recognition of errors in three-dimensional structures of proteins. Proteins. 1993; 17: 355–362.
Luthy R, Bowie JU, Eisenberg D. Assessment of protein models with three- dimensional profiles. Nature. 1992; 356: 83–85.
Lathrop GM, Lalouel JM, Julier C, Ott J. Multilocus linkage analysis in humans: detection of linkage and estimation of recombination. Am J Hum Genet. 1985; 37: 482–498.
Rotzer D, Roth M, Lutz M, Lindemann D, Sebald W, Knaus P. Type III TGF-; receptor-independent signalling of TGF-;2 via T;RII-B, an alternatively spliced TGF-; type II receptor. EMBO J. 2001; 20: 480–490.
Hardie G, Hanks S. The Protein Kinase Facts Book, Volume II. New York, NY: Academic Press; 1995.
Kim SJ, Im YH, Markowitz SD, Bang YJ. Molecular mechanisms of inactivation of TGF-; receptors during carcinogenesis. Cytokine Growth Factor Rev. 2000; 11: 159–168.
Shi Y, Massague J. Mechanisms of TGF-; signaling from cell membrane to the nucleus. Cell. 2003; 113: 685–700.
Akamine P, Madhusudan, Wu J, Xuong NH, Ten Eyck LF, Taylor SS. Dynamic features of cAMP-dependent protein kinase revealed by apoenzyme crystal structure. J Mol Biol. 2003; 327: 159–171.
Nerlich AG, Sauer U, Ruoss I, Hagedorn HG. High frequency of TGF-;-receptor-II mutations in microdissected tissue samples from laryngeal squamous cell carcinomas. Lab Invest. 2003; 83: 1241–1251.
Carcamo J, Zentella A, Massague J. Disruption of transforming growth factor-; signaling by a mutation that prevents transphosphorylation within the receptor complex. Mol Cell Biol. 1995; 15: 1573–1581.
Blobe GC, Schiemann WP, Lodish HF. Role of transforming growth factor-; in human disease. N Engl J Med. 2000; 342: 1350–1358.
Saharinen J, Hyytiainen M, Taipale J, Keski-Oja J. Latent transforming growth factor-; binding proteins (LTBPs)—structural extracellular matrix proteins for targeting TGF-; action. Cytokine Growth Factor Rev. 1999; 10: 99–117.
Milewicz DM, Pyeritz RE, Crawford ES, Byers PH. Marfan syndrome: defective synthesis, secretion, and extracellular matrix formation of fibrillin by cultured dermal fibroblasts. J Clin Invest. 1992; 89: 79–86.
Neptune ER, Frischmeyer PA, Arking DE, Myers L, Bunton TE, Gayraud B, Ramirez F, Sakai LY, Dietz HC. Dysregulation of TGF-; activation contributes to pathogenesis in Marfan syndrome. Nat Genet. 2003; 33: 407–411.
Denton CP, Zheng B, Evans LA, Shi-wen X, Ong VH, Fisher I, Lazaridis K, Abraham DJ, Black CM, de Crombrugghe B. Fibroblast-specific expression of a kinase-deficient type II transforming growth factor-; (TGF;) receptor leads to paradoxical activation of TGF-; signaling pathways with fibrosis in transgenic mice. J Biol Chem. 2003; 278: 25109–25119.(Hariyadarshi Pannu, PhD; )