|
|
||||||||
J Thorac Cardiovasc Surg 2007;134:290-296
© 2007 The American Association for Thoracic Surgery
Surgery for Congenital Heart Disease |
a Department of Surgery/Division of Cardiovascular Surgery, Mayo Clinic College of Medicine, Rochester, Minn
b Departments of Medicine, Pediatrics, and Molecular Pharmacology and Experimental Therapeutics/Divisions of Cardiovascular Diseases and Pediatric Cardiology, Mayo Clinic College of Medicine, Rochester, Minn.
Received for publication October 24, 2006; revisions received January 22, 2007; accepted for publication February 15, 2007. * Address for reprints: Thoralf M. Sundt III, MD, Division of Cardiovascular Surgery, Mayo Clinic, 200 First St SW, Rochester, MN 55905. (Email: sundt.thoralf{at}mayo.edu).
| Abstract |
|---|
|
|
|---|
Methods: We performed a targeted mutational analysis of NOTCH1 using genomic DNA from 48 unrelated subjects with concomitant bicuspid aortic valve and thoracic aortic aneurysm using denaturing high-performance liquid chromatography and DNA sequencing. We focused on exons in which mutations associated with bicuspid aortic valve have been reported previously. Results were compared with control subjects with trileaflet aortic valves (n = 94), bicuspid aortic valves, and normal aortas (n = 22) and in subjects with tricuspid aortic valves and thoracic aortic aneurysms (n = 28).
Results: Four unique, nonsynonymous (3 novel) variants were identified in 5 (10.4%) of 48 patients with concomitant bicuspid aortic valves and thoracic aortic aneurysms compared with only 3 (2.1%) of 144 control subjects (P = .02). Of these, 2 novel missense mutations, A1343V and P1390T, were observed only in patients with bicuspid aortic valves and tricuspid aortic aneurysms.
Conclusions: This targeted analysis involving NOTCH1 exons previously implicated in familial and sporadic bicuspid aortic valve demonstrates overrepresentation of NOTCH1 missense variants among patients with bicuspid aortic valves and thoracic aortic aneurysms. Identification of aneurysm-predisposing susceptibility genes may lead to gene-directed surgical therapy of the ascending aorta for patients with bicuspid aortic valves.
| Introduction |
|---|
|
|
|---|
Both TAA disease and BAV disease are inheritable disorders in at least some cases. Familial clustering of BAV disease shown by Clementi,5
Cripe,6
and their colleagues has demonstrated heritability (h2) of 89% with autosomal dominant transmission and incomplete penetrance of the condition. In addition to Marfan syndrome and the more recently identified Loeys–Dietz syndrome,7
a genetic basis for some cases of familial TAA and dissection has been mapped by Milewicz, Basson, and their colleagues to 3 genetic loci: 5q13-14,8
11q23,9
and 3p24-25,10,11
the last of which has proven to be the transforming growth factor (TGF) ß-2 receptor.
Recently, an association between mutations in NOTCH1 and aortic valve disease has been described. NOTCH1 encodes for a transmembrane protein that activates a signaling pathway with an active role in cardiac embryogenesis, including aortic and pulmonary valve development as well as the development and maintenance of the aorta and other great vessels.12-15
Garg and colleagues16
reported NOTCH1 mutations in 2 pedigrees with an assortment of cardiovascular disease phenotypes, including BAV. Subsequently, Mohamed and coworkers17
reported 2 mutations in different exons of the same gene in patients with nonfamilial BAV; however, the status of the aorta was not described in the former and only briefly mentioned in the latter study. To date, no studies have identified genetic mutations common to both BAV and TAA phenotypes.
Recent advances in the field of genomics introduce the possibility of genetic profiling as is currently used in pharmacogenomics. Identification of genes predictive of TAA in patients with BAV may foster "gene-directed" surgical therapy of the ascending aorta. We hypothesize that genetic variability among patients with BAV may explain the phenotypic variability in the subset of patients with BAV in whom TAA develops. We, therefore, performed a targeted mutational analysis of NOTCH1 from genomic DNA obtained from patients with BAV and TAA using denaturing high-performance liquid chromatography (DHPLC) and direct DNA sequencing. These results were compared with DNA from 188 reference alleles obtained from control subjects (n = 94) with normal, trileaflet aortic valves (TAV), as well as a limited number of subjects with BAV and normal aortas (n = 22) or TAV and TAA (n = 28).
| Materials and Methods |
|---|
|
|
|---|
|
Using polymerase chain reaction (PCR), DHPLC, and direct DNA sequencing, we performed a targeted mutational analysis of the 4 exons (11, 20, 25, and 29) of NOTCH1 previously implicated in BAV,16,17
as previously described.19
PCR amplification primers were designed with Oligo software (Molecular Biology Insights, Inc, Cascade, Colo). PCR primers are shown in Table 1. In brief, DHPLC is a sensitive method used to elucidate unknown gene mutations. It is based on thermal energy found in the formation and separation of double-stranded DNA fragments containing a mismatch in the base pairing between the "wild type" and "mutant," or heteroduplex, DNA strands. PCR-amplified DNA is injected onto a solid phase column that is heated to a specific temperature (individually optimized for each unique PCR product), which allows for partial denaturing of the DNA sequence of interest. A linear acetonitrile gradient based on the size of the PCR product is applied to the column to flush the DNA strands and send the PCR products through an ultraviolet detector, resulting in a chromatogram showing the samples elution profile. Since heteroduplex (mutant sequences) species are less thermodynamically stable than homoduplexes (normal sequences), these double-stranded complexes will begin to unravel at the elevated temperature and elute from the column sooner then their homoduplex or "wild-type" counterparts. As a result, "mutant" DNA yields a DHPLC profile that is different from the "wild type" profile, enabling the detection of mutation harboring samples whose precise genetic mutation is then established by direct DNA sequencing.
|
An in silico analysis was performed with PredictProtein (a Web-based tool found at www.predictprotein.org) to predict effect on secondary protein structure.20
Differences in the proportion of patients with DNA sequence variations among groups were statistically analyzed with the Fisher exact test. Patient demographic data were analyzed with the Student t test, Wilcoxon rank sum, or Fisher exact test where appropriate.
| Results |
|---|
|
|
|---|
|
Overall, 4 unique nonsynonymous (3 novel) variants were identified in 5 of 48 (10.4%) patients with BAV/TAA compared with only 3 of 144 (2.1%) patients with normal aortas (2/94 reference DNA with trileaflet valves, 1/28 TAV/TAA, and 0/22 with BAV/normal aortas; P = .02). Table 3 summarizes the clinical characteristics and phenotype of the NOTCH1 mutation-positive patients with BAV/TAA. Of the 5 patients (3 men, average age 52 ± 12 years, average aortic dimension 50 ± 5 mm) with NOTCH1 variants, 3 had fusion of the right and left coronary cusps, 3 presented with raphé, 4 underwent AVR, and 2 had evidence of aortic valve calcification. Only 1 of these 5 patients had a family history of either BAV or TAA (patient 1, Table 3). This patient, a 70-year-old woman with BAV and a 57-mm ascending aorta, had a family history of both BAV and TAA. We identified both polymorphisms, R1350L and P1377S, in this patient. No other patients with DNA sequence variants had family histories of either BAV or TAA.
|
|
Figure 3 is a topographic summary of DNA sequence variants from the targeted NOTCH1 analysis demonstrating both the missense mutations (A1343V and P1390T) and polymorphisms (R1350L and P1377S).
|
| Discussion |
|---|
|
|
|---|
We chose to study the potential association between NOTCH1 and patients with BAV/TAA because it has been implicated as a potential susceptibility gene for heritable cardiovascular disease. Garg and colleagues16
described NOTCH1 mutations in 2 pedigrees with complex, heterogeneous cardiac disease including BAV, although the status of the ascending aorta was not mentioned. Subsequently, studying nonfamilial cases, Mohamed and coworkers17
reported variants in 2 of 48 patients with BAV. Although not the focus of their study, the incidence of NOTCH1 sequence variations was 11% (2/18) among the individuals in their study with concomitant BAV/TAA. Given the size of this gene, we focused our attention on the 4 exons in which NOTCH1 mutations have been reported previously, and this restricted analysis of NOTCH1 may underestimate the relative contribution of genetic variation in NOTCH1 to the development of TAA.
NOTCH1 is a plausible candidate gene for BAV and TAA. Located at 9q34.3, NOTCH1 encodes for a transmembrane protein that functions as a transcriptional cofactor to activate the Notch signaling pathway, a signaling pathway found in developmental pathways of many organ systems including the human cardiovascular system.21,22
Specifically, NOTCH1 has a role in migration of cells from the cardiac cushions and cardiac jelly into the conotruncal cushions from which the aortic and pulmonary valves are formed.13-15
A role for NOTCH1 mutations in the pathogenesis of TAA is also plausible. In addition to its role in cardiogenesis and valvulogenesis, the Notch1 signaling pathway has a role in development and maintenance of the aorta and other vessels.13-15
Additionally, there is increasing evidence of interaction between the Notch1 pathway and TGFß.23
This is particularly exciting in light of recent reports of TGFß-2 receptor mutations and TAA in syndromal patients such as those with Loeys–Dietz and Ehlers–Danlos syndromes.7
TGFß has a known role in regulation of extracellular matrix of aortic tissue. On the basis of the interaction between the Notch1 pathway (valvulogenesis) and TGFß (aortic extracellular matrix regulation), we hypothesize that BAV patients with NOTCH1 mutations may be more susceptible to abnormal extracellular matrix regulation and, therefore, aneurysm formation. Although the association between TGFß receptor mutations and aortic aneurysm is in its infancy, we believe that exploring an association between NOTCH1 and TGFß in patients with nonsyndromal aortic aneurysms (including patients with BAV/TAA) is certainly intriguing and warranted.
There is increasing interest in the identification of the genetic determinants of aortic valve disease. In 2005, Garg and colleagues16
reported NOTCH1 truncation mutations (R1108X and H1505del) in two families with dominantly inherited familial aortic valve disease associated with aortic valve calcification. These deleterious mutations presumably cause haplo-insufficiency and a loss of function phenotype of Notch1 signaling. Among a 5-generation pedigree hosting the truncation mutation, 7 of 9 family members were affected and had calcific aortic stenosis in the setting of either BAV or TAV. Recently, Mohamed and colleagues17
identified NOTCH1 missense mutations in approximately 4% of sporadic cases of BAV.
Limitations of the present study include the heterogeneity in baseline demographics found among our cohorts, as well as the challenge of defining "abnormal" ascending aortic phenotype. BAV/TAA patients had a significantly greater proportion of males and subjects with hypercholesterolemia than did subjects with TAV/TAA. The overall impact this difference makes on interpretation is unknown but is a limitation of the present study. Additionally, defining significant aortic enlargement in patients with BAV remains problematic. One option is to use established nomograms indexed to body mass index, but these are imperfect because measurements may change with fluctuations in body mass. We used a liberal definition of ascending aortic abnormalities at greater than 4.0 cm. If anything, this may have diluted the observation of overrepresentation of DNA sequence variants found in the BAV/TAA cohort. Using a more conservative definition of aortic aneurysm of 4.5 cm or larger, the proportion of DNA variants in the BAV/TAA cohort would increase from 10.4% (5/48) to 12.5% (5/40).
Here, we report the identification of NOTCH1 amino acid altering variants in 10.4% (5/48) of patients with concomitant BAV/TAA, 0 of 22 BAV subjects without aneurysmal disease, and 1 of 28 patients with TAV/TAA. Although not found to be statistically significant owing to the limited size of our cohort, these data are congruent with recently reported data from Mohameds group.17
They too identified NOTCH1 missense mutations in 2 (11%) of 18 patients with sporadic BAV and concomitant aortic aneurysm and in none of 30 BAV patients without aneurysmal disease.
Taken together, these independent investigations suggest a significant overrepresentation of NOTCH1 missense mutations in BAV/TAA (7 of 66 compared with 0 of 52; P = .01). These observations may offer insight into predicting which patients with BAV are at risk for the development of TAA. To be sure, these studies warrant further investigation into the role of NOTCH1 missense variants in the pathogenesis of aneurysmal associated valve disease. Whether these NOTCH1 missense variants lead to a "gain" or "loss" of function phenotype will require further functional characterization.
Exploration of additional NOTCH1 exons should be included in future investigations into the role of NOTCH1 in the extracellular matrix of the ascending aorta.
| Footnotes |
|---|
* Clinician Investigator Program, Mayo School of Graduate Medical Education, Mayo Clinic College of Medicine. ![]()
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. R. Moreno, L. Astudillo, S. Elmariah, K. R. Purushothaman, M. Purushothaman, P. A. Lento, S. K. Sharma, V. Fuster, and D. H. Adams Increased macrophage infiltration and neovascularization in congenital bicuspid aortic valve stenosis J. Thorac. Cardiovasc. Surg., October 1, 2011; 142(4): 895 - 901. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. I. Bostrom, N. M. Rajamannan, and D. A. Towler The Regulation of Valvular and Vascular Sclerosis by Osteogenic Morphogens Circ. Res., August 19, 2011; 109(5): 564 - 577. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. S. Kerstjens-Frederikse, G. J. Du Marchie Sarvaas, J. S. Ruiter, P. C. Van Den Akker, A. M. Temmerman, J. P. Van Melle, R. M. W. Hofstra, and R. M. F. Berger Left ventricular outflow tract obstruction: should cardiac screening be offered to first-degree relatives? Heart, August 1, 2011; 97(15): 1228 - 1232. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. L. McBride and V. Garg Heredity of bicuspid aortic valve: is family screening indicated? Heart, August 1, 2011; 97(15): 1193 - 1195. [Full Text] [PDF] |
||||
![]() |
C. C. Badiu, W. Eichinger, S. Bleiziffer, G. Hermes, I. Hettich, M. Krane, R. Bauernschmitt, and R. Lange Should root replacement with aortic valve-sparing be offered to patients with bicuspid valves or severe aortic regurgitation? Eur J Cardiothorac Surg, November 1, 2010; 38(5): 515 - 522. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Jafar-Nejad, J. Leonardi, and R. Fernandez-Valdivia Role of glycans and glycosyltransferases in the regulation of Notch signaling Glycobiology, August 1, 2010; 20(8): 931 - 949. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. H. Anderson, R. P. Thompson, and C. B. Kern Development of Aortic Valves With 2 and 3 Leaflets J. Am. Coll. Cardiol., December 8, 2009; 54(24): 2319 - 2320. [Full Text] [PDF] |
||||
![]() |
K. L. McBride, M. F. Riley, G. A. Zender, S. M. Fitzgerald-Butt, J. A. Towbin, J. W. Belmont, and S. E. Cole NOTCH1 mutations in individuals with left ventricular outflow tract malformations reduce ligand-induced signaling Hum. Mol. Genet., September 15, 2008; 17(18): 2886 - 2893. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. D. Pearson, R. Devereux, B. Loeys, C. Maslen, D. Milewicz, R. Pyeritz, F. Ramirez, D. Rifkin, L. Sakai, L. Svensson, et al. Report of the National Heart, Lung, and Blood Institute and National Marfan Foundation Working Group on Research in Marfan Syndrome and Related Disorders Circulation, August 12, 2008; 118(7): 785 - 791. [Full Text] [PDF] |
||||
![]() |
P. W. M. Fedak Bicuspid aortic valve syndrome: heterogeneous but predictable? Eur. Heart J., February 2, 2008; 29(4): 432 - 433. [Full Text] [PDF] |
||||
![]() |
A. Della Corte, C. Quarto, C. Bancone, C. Castaldo, F. Di Meglio, D. Nurzynska, L. S. De Santo, M. De Feo, M. Scardone, S. Montagnani, et al. Spatiotemporal patterns of smooth muscle cell changes in ascending aortic dilatation with bicuspid and tricuspid aortic valve stenosis: Focus on cell matrix signaling J. Thorac. Cardiovasc. Surg., January 1, 2008; 135(1): 8 - 18. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |