Fibrillin-1 (FBN1) Mutations in Patients With Thoracic Aortic Aneurysms
Background Mutations in the FBN1 gene are the cause of the Marfan syndrome, an autosomal dominant disorder with skeletal, ocular, and cardiovascular complications. Aneurysms or dissections of the ascending thoracic aorta are the major cardiovascular complications of the disorder. We tested the hypothesis that FBN1 mutations cause thoracic aortic aneurysms or dissections in patients who do not have the Marfan syndrome.
Methods and Results The FBN1 gene was screened for mutations by use of genomic DNA from two patients with thoracic aortic aneurysms who did not have the Marfan syndrome. Individual FBN1 exons were amplified with intron-based exon-specific primers; the DNA fragments were screened for mutations using single-stranded conformational polymorphism analysis; and aberrantly migrating bands were sequenced directly. We identified a missense mutation in one patient, D1155N in exon 27. Dermal fibroblasts from the affected individual were used to study the effect of the missense mutation D1155N on fibrillin-1 cellular processing. The mutation decreased the amount of fibrillin-1 deposited into the pericellular matrix. A second putative FBN1 mutation was identified in the second patient, P1837S in exon 44. Although this alteration was not observed in 234 chromosomes from unrelated individuals, the alteration may represent a rare polymorphism.
Conclusions Results of these studies support the hypothesis that FBN1 mutations cause thoracic aortic aneurysms in patients who do not have the Marfan syndrome. This information is important for understanding the pathogenesis of aortic aneurysms and identification of individuals at risk for developing thoracic aortic aneurysms or dissections.
Aortic aneurysms are estimated to be responsible for 1% to 2% of the deaths in industrialized societies, with death usually caused by rupture of an asymptomatic, undiagnosed aneurysm. Aortic aneurysms are classified as TAAs or AAAs, depending on the anatomic location of the aneurysm. The familial aggregation of AAAs has been shown in many studies, implicating a genetic component to the cause of these aneurysms. Familial forms of ascending aortic dilatation or dissection have been reported, suggesting that a genetic defect may be the cause of some TAAs.1 2
MFS is an autosomal dominant disorder characterized by pleiotropic manifestations involving primarily the cardiovascular, skeletal, and ocular systems.3 The major cardiovascular complications of the disorder are formation of an aneurysm and dissection of the thoracic aorta, primarily the ascending aorta. Although any of the features of MFS can occur as an isolated finding in an individual, it is the constellation of findings involving these systems and the autosomal dominant inheritance of these findings that provide the basis of the clinical diagnosis of MFS.4
MFS results from mutations in the FBN1 gene that encodes fibrillin-1, a protein found in elastic fiber–associated microfibrils. The fact that TAAs are a major manifestation of MFS suggests that mutations in the FBN1 gene may cause TAAs in the absence of other complications of MFS. A recently reported FBN1 missense mutation in a family with autosomal dominant inheritance of ascending aortic aneurysms and dissections supports this hypothesis.2 Affected individuals in the family also had mild skeletal and ocular features of MFS. Most affected individuals in this family did not require surgical intervention of their aortic disease until the sixth or seventh decade of life. In contrast, patients with classic MFS usually require surgery in their early 30s.5 Therefore, the identified FBN1 mutation resulted in a predisposition to form ascending aortic aneurysms or to dissect the aorta later in life in individuals with many features of the MFS.
To further characterize the role of FBN1 mutations in the origin of TAAs, we have screened for FBN1 mutations in TAA patients who do not have MFS. We have identified FBN1 alterations in two such individuals. In contrast to the previously described TAA family with an FBN1 mutation, these individuals had aortic complications in their third to fourth decades of life, with minimal skeletal and no ocular features of MFS. These results suggest that FBN1 mutations play a role in the cause of TAAs in the absence of features of MFS. This information is important not only for understanding the pathogenesis of aortic disease but also for identifying individuals at risk for developing aortic aneurysms.
Patient P062 was a 44-year-old white man (height, 193 cm; arm span, 197.5 cm; upper-to-lower segment ratio, 0.96) with minimal pectus excavatum, pes cavus with mild contractures of the toes, and a high arched palate. He did not have dolichostenomelia, arachnodactyly, myopia, or lens dislocation. He presented with 6- to 7-cm dilatation of his aortic root, mitral valve prolapse, and mitral annular calcifications. Three days later, the patient had an acute dissection of his ascending thoracic aorta and underwent successful surgical repair of his aorta. His medical history was significant for rheumatic fever at 5 years of age and inguinal hernia repair in his early 30s. There was no family history of MFS or aortic aneurysms, and his parents were healthy and in the sixth decade of life; his family declined to participate in these studies.
Patient P018 was a 39-year-old white man (height, 176.5 cm; arm span, 175 cm; upper-to-lower-segment ratio, 0.88) with a dilated aortic root treated with Inderal. Serial echocardiograms over several years showed aortic root measurements of 46 to 50 mm, mitral valve prolapse, mild mitral regurgitation, and mild aortic insufficiency. He had a high arched palate and mild thoracic scoliosis. He did not have arachnodactyly, dolichostenomelia, pectus deformity, myopia, or ectopia lentis. He had two herniorrhaphies at 32 and 33 years of age. His mother (172 cm tall) died at the age of 33 of an aortic aneurysm. There were no other family histories of aortic aneurysms and no other living family members.
For controls, blood was collected from patients with thoracic aortic aneurysms (n=30) or dissections (n=7) referred to one of the authors for either genetic evaluation (Dr Milewicz) or TAA surgery (Dr Coselli). Twenty-five TAA patients had aneurysms or dissections involving the ascending aorta; 12 patients had aneurysms or dissections originating in the descending thoracic aorta. Blood was also collected from 80 unrelated individuals without known aortic disease.
Mutation Identification and Conformation
Blood samples and skin biopsies were obtained from the patients after appropriate consent was signed. Fibroblasts were explanted from the skin biopsies by use of previously described techniques.6 Genomic DNA was obtained from white blood cells or dermal fibroblasts with standard techniques. Individual FBN1 exons were amplified from genomic DNA by use of intron-based, exon-specific primers.7 SSCP was performed with amplified DNA fragments.8 Any aberrantly migrating fragments were analyzed more than three times to confirm the abnormality. The amplified DNA fragments were purified and sequenced with both the sense and antisense amplification primers by use of ABI automated sequencer model 373A.
Total RNA was isolated from cultured dermal fibroblasts and reverse-transcribed by use of a primer complementary to nucleotides 3525 through 3540 of FBN1 with M-MuLV reverse transcriptase according to recommended conditions (Boehringer Mannheim Biochemicals). Subsequently, PCR amplification was done after the addition of the sense primer, corresponding to nucleotides 3292 through 3310, and Taq polymerase.
Dermal fibroblasts from patient P062 and an age-matched control were used to radiolabel synthesized proteins to study fibrillin-1 cellular processing as previously described.6 The identities of both profibrillin and fibrillin were determined, and the intensities of the profibrillin and fibrillin bands were quantified as described elsewhere by use of an electronic autoradiographer (InstantImager, Packard Instrument Co).6 The amount of fibrillin incorporated into the matrix by P062 cells was the average of four separate experiments.
Identification of FBN1 Mutations
The FBN1 gene was screened for mutations by use of genomic DNA from patients P018 and P062 by amplification of individual exons and analysis of the amplified fragments by SSCP. Of the 65 exons of FBN1, 34 were screened for mutations in these individuals.
Amplification of FBN1 exon 27 with patient P062 genomic DNA revealed an aberrantly migrating band (Fig 1A⇓). The amplified exon was purified and sequenced directly, revealing a G-to-A transition at position 3463 (Fig 1B⇓) that substituted an aspartic acid for an asparagine (D1155N). Analysis of genomic DNA from 80 unrelated control subjects and 37 TAA patients failed to reveal a similar SSCP alteration. Reverse transcription–PCR of exons 26 through 28 of the FBN1 cDNA by use of RNA from patient P062 fibroblast cell strain was used to confirm the mutation and verify that the mutation did not affect RNA splicing (Fig 1C⇓).
Genomic DNA from patient P018 revealed an aberrantly migrating band when exon 44 was amplified and screened for a mutation (Fig 2A⇓). Sequencing of exon 44 revealed a C-to-T transition at nucleotide 5509, altering a proline to a serine (P1837S; Fig 2B⇓). Screening of genomic DNA from 80 unrelated control subjects and 37 TAA patients failed to reveal this SSCP alteration in any other individual.
Fibrillin Protein Studies
Previously established protocols for the study of fibrillin-1 synthesis, secretion, and extracellular matrix incorporation were used to study the effect of the D1155N mutation on fibrillin with cells from P062. Although normal amounts of the fibrillin-1 precursor (profibrillin-1) were synthesized and secreted, cells from patient P062 deposited less fibrillin-1 into the extracellular matrix compared with control cells (66% of control values; Fig 3⇓).
Two novel FBN1 alterations have been identified in individuals with TAAs involving the ascending thoracic aorta. Patient P062 had no family history of MFS or aortic disease and had no ocular features of MFS. He was tall (193 cm) and had a mild chest wall deformity but did not have dolichostenomelia or arachnodactyly. He presented with a 6- to 7-cm ascending TAA that dissected a few days later. Patient P018 had a dilated ascending aorta at the age of 39. He was of average height; had mild scoliosis and a highly arched palate; but lacked arachnodactyly, dolichostenomelia, or eye problems. Patient P018's mother died of aortic rupture at 33 years of age but was not known to have MFS. Therefore, these FBN1 alterations occurred in individuals with significant disease involving their ascending aorta occurring relatively early in life without ocular features of MFS and minimal skeletal manifestations.
Mutations causing MFS are spread throughout the FBN1 gene.7 Most mutations are missense mutations that alter amino acids in 1 of 47 EGF-like domains. Of these EGF-like domains, 43 are predicted to bind calcium based on a conserved consensus sequence for calcium binding.9 The vast majority of mutations in MFS patients are predicted to abolish calcium binding to a cbEGF-like domain by either disrupting the tertiary structure of the domain or specifically altering the amino acids critical for calcium binding.7 The residues in cbEGF-like domains involved in calcium binding include the aspartic acid at position 2, glutamic acid at position 5, asparagine at position 19, and aromatic tyrosine or phenylalanine at position 24.10 11 The mutation identified in patient P062 altering an aspartic acid to an asparagine at position 2 in a cbEGF-like domain is predicted to decrease the affinity of calcium binding to that domain. All 43 of the cbEGF-like domains in fibrillin-1 have a consensus sequence in positions 2 through 6 of Asp-Ile/Val-Asp/Asn-Glu-Cys, indicating that the aspartic acid in position 2 is highly conserved in the fibrillin-1 protein.9 12 Mutating the aspartic acid in position 2 to a glutamic acid reduces the affinity of calcium binding for a cbEGF found in human factor IX, although it did not disrupt the secondary structure of the EGF-like domain by NMR spectrum analysis.13 Although many other proteins with cbEGF-like domains have an asparagine instead of an aspartic acid at position 2, this substitution is predicted to decrease the efficiency of calcium binding.10 11 14 Therefore, the milder clinical manifestations of this mutation may be the result of decreased affinity for calcium rather than lack of binding.
Fibrillin-1 protein and synthesis studies with fibroblasts obtained from patient P062 indicated that the cells made normal amounts of profibrillin-1 and secreted the profibrillin-1 efficiently, but the processed fibrillin-1 was poorly incorporated into the extracellular matrix. Fibrillin-1 processing studies of MFS cell strains have indicated that this processing abnormality is associated with mutations that alter the consensus amino acids required for calcium binding to cbEGF domains.15 Therefore, the decreased incorporation of fibrillin-1 into the pericellular matrix further substantiates the hypothesis that an FBN1 mutation affecting calcium binding to a cbEGF of fibrillin-1 is the underlying cause of TAA formation in this individual.
In contrast, the alteration identified in patient P018 would not be predicted to disrupt calcium binding to the EGF-like domain. The transition alters a proline in the carboxyl-terminal region of the EGF-like domain, which is not involved in calcium binding. Although we cannot exclude the possibility that this alteration is a rare polymorphism, the fact that the change was not observed in 160 chromosomes from unrelated individuals, in addition to 74 chromosomes from individuals with TAAs, suggests that this alteration is a mutation that predisposed this individual to form TAAs. We cannot predict the effect of the missense mutation on the structure or function of the cbEGF-like domain, and dermal fibroblasts are not available from the patient to study the effect of the mutation on fibrillin-1 cellular processing. It is interesting to note that a FBN1 polymorphism (P1148A) also altering a proline in the same carboxyl-terminal region of another cbEGF-like domain has been proposed to be associated with aortic aneurysm formation.7
The results of this study suggest that FBN1 mutations play a role in the origin of TAAs. Establishing a role of FBN1 mutations in the pathogenesis of TAAs is important for both understanding the causes of this condition and identifying individuals at risk for developing aneurysms. Further studies are needed to establish the prevalence of FBN1 gene mutations in patients with TAAs, to determine what proportion of inherited TAAs are due to FBN1 mutations, and to ascertain why these FBN1 mutations are not associated with skeletal or ocular complications of the MFS.
Selected Abbreviations and Acronyms
|AAA||=||abdominal aortic aneurysm|
|EGF||=||epidermal growth factor|
|PCR||=||polymerase chain reaction|
|SSCP||=||single-stranded polymorphism analysis|
|TAA||=||thoracic aortic aneurysm|
This work was supported by an American Heart Association Grant-in-Aid (93014300) to Dr Milewicz and NIH grant M01-RR-02-558 (CRC grant). Dr Milewicz is a Pfizer Scholar. We gratefully acknowledge the patients for participating in this research protocol.
Guest Editor for this article was J. David Bristow, MD, Oregon Health Sciences University, Portland.
- Received May 23, 1996.
- Revision received September 5, 1996.
- Accepted September 24, 1996.
- Copyright © 1996 by American Heart Association
Pyeritz RE. The Marfan syndrome. In: Royce PM, Steinmann B, eds. Connective Tissue and Its Heritable Disorders. New York, NY: Wiley-Liss & Sons; 1993:437-468.
Beighton P, dePaepe A, Danks D, Finidori G, Gedde-Dahl T, Goodman R, Hall JG, Hollister DW, Horton W, McKusick VA, Opitz JM, Pope FM, Pyeritz RE, Rimoin DL, Sillence D, Spranger JW, Thompson E, Tsipouras P, Viljoen D, Winship I, Young I. International nosology of heritable disorders of connective tissue, Berlin, 1986. Am J Med Genet. 1988;29:581-594.
Finkbohner R, Johnston D, Crawford ES, Coselli J, Milewicz DM. The Marfan syndrome: long-term survival and complications following aortic aneurysm repair. Circulation. 1995;91:728-733.
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.
Selander-Sunnerhagen M, Ullner M, Persson E, Teleman O, Stenflo J, Drakenberg T. How an epidermal growth factor (EGF)-like domain binds calcium. J Biol Chem. 1992;267:19642-19649.
Pereira L, D'Alessio M, Ramirez F, Lynch J, Sykes B, Pangilinan T, Bonadio J. Genomic organization of the sequence coding for fibrillin, the defective gene product in Marfan syndrome. Hum Mol Genet. 1993;2:961-968.
Aoyama T, Tynan K, Dietz HC, Francke U, Furthmayr H. Missense mutations impair intracellular processing of fibrillin and microfibril assembly in Marfan syndrome. Hum Mol Genet. 1993;2:2135-2140.