Leaders in Pharmaceutical Business Intelligence (LPBI) Group

Cardiovascular Genetics: Functional Characterization and Clinical Applications  @ 2013 Annual Conference of American Society of Human Genetics in Boston, 10/22-26, 2013

Reporter: Aviva Lev- Ari, PhD, RN

Sessions and Events 

The 63rd Annual Conference of American Society of Human Genetics in Boston, 10/22-26, 2013

http://www.ashg.org/cgi-bin/2013/ashg13SOE.pl 

PLATFORM ABSTRACTS

http://www.ashg.org/2013meeting/pdf/46025_Platform_bookmark%20for%20Web%20Final%20from%20AGS.pdf

We express a special interest in Session 58

Friday, October 25, 2013 Boston Convention Center 

2:00 PM–4:15 PM

Concurrent Platform (abstract-driven) Session E (54-62)

SESSION 58 – Cardiovascular Genetics: Functional Characterization and Clinical Applications

Room 205, Level 2, Convention Center

Human Syndromic Atrioventricular Septal Defect

367/2:00 A homozygous mutation in Smoothened, a member of the Sonic hedgehog (SHH)-GLI pathway is involved in human syndromic atrioventricular septal defect. W. S. Kerstjens-Frederikse, Y. Sribudiani, M. E. Baardman, L. M. A. Van Unen, R. Brouwer, M. van den Hout, C. Kockx, W. Van IJcken, A. J. Van Essen, P. A. Van Der Zwaag, G. J. Du Marchie Sarvaas, R. M. F. Berger, F. W. Verheijen, R. M. W. Hofstra.

A homozygous mutation in Smoothened, a member of the Sonic Hedgehog (SHH)-GLI pathway is involved in human syndromic atrioventricular septal defect.

W.S. Kerstjens-Frederikse1, Y. Sribudiani2, M.E. Baardman1, L.M.A. Van Unen2, R. Brouwer2, M. van den Hout2, C. Kockx2, W. Van IJcken2, A.J. Van Essen1, P.A. Van Der Zwaag1, G.J. Du Marchie

Sarvaas3, R.M.F. Berger3, F.W. Verheijen2, R.M.W. Hofstra2.

1) Dept Gen, Univ of Groningen, Univ Med Ctr Groningen, Netherlands;

2) Dept Gen, Erasmus Med Ctr, Rotterdam, Netherlands; 3) Dept Ped Cardiol, Univ of Groningen, Univ Med Ctr Groningen, Netherlands.

Introduction: Atrioventricular septal defect (AVSD) is a common congenital heart disease with a high impact on personal health. It is often accompanied by other congenital anomalies and in many of these syndromic AVSDs, defects in the sonic hedgehog (SHH)-GLI signalling pathway have been detected. SMO codes for the transmembrane protein smoothened (SMO), which is active in cells with a primary cilium and is located on the ciliary membrane. SMO is a key protein in the SHH-GLI signaling cascade.

Methods: Two probands, a twin boy and girl, presented with an AVSD, large fontanel, postaxial polydactyly and skin syndactyly of the second and third toes of both feet. The boy also had hypospadias. The parents were consanguineous and they had one healthy older child. Karyotyping was normal and Smith-Lemli-Opitz syndrome (SLOS) was excluded. Exome sequencing was performed and candidate variants were validated by Sanger sequencing.

Results: A novel homozygous missense mutation c.1725C>T (p.R575W) in SMO (7q32.3) was detected. Functional studies in fibroblasts of the patients showed normal expression of SMO protein but an abnormal localization of SMO, outside the cilia. Moreover we show severely reduced downstream GLI1 mRNA expression after stimulation with the SMO agonist purmorphamine. These results, together with the previously described association of SHH signalling defects with AVSD and SLOS, suggest that this SMO mutation is involved in syndromic AVSD in these patients.

Conclusion: We present the first reported smoothened mutation in humans, in two patients with an AVSD and a phenotype resembling Smith-Lemli-Opitz syndrome

Left Ventricular Noncompaction – Model in Zebrafish

368/2:15 Identification of PRDM16 as a disease gene for left ventricular non-compaction and the efficient generation of a personalized disease model in zebrafish. A.-K. Arndt, S. Schaefer, R. Siebert, S. A. Cook, H.-H. Kramer, S. Klaassen, C. A. MacRae.

Identification of PRDM16 as a disease gene for left ventricular noncompaction

and the efficient generation of a personalized disease

model in zebrafish. A.-K. Arndt1,2, S. Schaefer3, R. Siebert4, S.A. Cook5,

H.-H. Kramer2, S. Klaassen6, C.A. MacRae1. 

1) Cardiovascular Division, Brigham and Women’s Hospital, Boston, MA;

2) Department of Congenital Heart Disease and Pediatric Cardiology, University Hospital of Schleswig- Holstein, Kiel, Germany,;

3) Max-Delbruck-Center for Molecular Medicine, Berlin, Germany; 4) Institute of Human Genetics, University Hospital Schleswig Holstein, Kiel, Germany;

5) National Heart Centre, Singapore;

6) Department of Pediatric Cardiology, Charité, Berlin, Germany.

Using our own data and publically available array comparative genomic hybridization data, we identified the transcription factor PRDM16(PR domain containing 16) as a causal gene for the cardiomyopathy associated with monosomy 1p36, and confirmed its role in individuals with non-syndromic left ventricular noncompaction cardiomyopathy (LVNC) and dilated cardiomyopathy (DCM). In a cohort of 75 non-syndromic patients with LVNC we detected 3 sporadic mutations, including 1 truncation mutant, 1 frameshift null mutation, and a single missense mutant. In addition, in a series of cardiac biopsies from 131 individuals with DCM, we found 5 individuals with 4 previously unreported non-synonymous variants in the coding region of PRDM16. None of the PRDM16 mutations identified were observed in over 6500 controls.

PRDM16 has not previously been associated with cardiovascular disease. Modeling of PRDM16 haploinsufficiency and a human truncation mutant in zebrafish resulted in impaired cardiomyocyte proliferation with associated physiologic defects in cardiac contractility and cell-cell coupling.

Using a phenotype-driven screening approach in the fish, we have identified 5 compounds that are able to rescue the physiologic defects associated with mutant or haploinsufficient PRDM16. Notably, all of the compounds had the capacity to restore cardiomyocyte proliferation and to prevent apoptosis in the model. Wildtype zebrafish also demonstrated a significant increase in cardiomyocyte numbers after treatment with the compounds suggesting a pro-proliferative effect of the compounds. In addition, the compounds also rescued the contractile and electrical defects observed in these disease models. These findings underline the importance of personalized disease models for specific pathways, to accelerate the exploration of disease biology and the development of innovative therapeutic approaches.

Genetics of Cerebral Small Vessel Disease

369/2:30 Mutation and copy number variation of FOXC1 causes cerebral small vessel disease. C. R. French, S. Seshadri, A. L. Destefano, M. Fornage, D. J. Emery, M. Hofker, J. Fu, A. J. Waskiewicz, O. J. Lehmann.

Mutation and copy number variation of FOXC1 causes cerebral small vessel disease. C.R. French1, S. Seshadri2, A.L Destefano3, M. Fornage4, D.J. Emery5, M. Hofker6, J. Fu6, A.J. Waskiewicz7, O.J. Lehmann1, 8.

1) Ophthalmology, University of Alberta, Edmonton, AB, Canada;

2) Department of Neurology, Boston University, Boston, MA, U. S. A;

3) School of Public Health, Boston University, Boston, MA, U. S. A;

4) Institute of Molecular Medicine and School of Public Health, University of Texas Health Sciences

Center, Houston, TX, U.S.A;

5) Department of Radiology, University of Alberta, Edmonton, AB, Canada;

6) Department of Medical Genetics, University Medical Center Groningen, Groningen, The Netherlands;

7) Department of Biological Sciences, University of Alberta, Edmonton, AB, Canada;

8) Department of Medical Genetics, University of Alberta, Edmonton, AB, Canada.

Cerebral small vessel disease (CSVD) represents a major risk factor for stroke and cognitive decline in the elderly. The ability to readily visualize its microangiopathic features by magnetic resonance imaging provides opportunities for using markers of CSVD to identify novel stroke associated pathways. Using targeted genome-wide association analysis we identified CSVD associated single nucleotide polymorphisms (SNPs) adjacent to the forkhead transcription factor FOXC1, and using eQTL analysis in two independent data sets, demonstrate that such SNP’s are associated with FOXC1 expression levels.

We further demonstrate, using magnetic resonance imaging, that patients with either FOXC1 mutation or copy number variation exhibit CSVD. These findings, present in patients as young as two years of age and observed with missense and nonsense mutations as well as FOXC1-encompassing segmental deletion and duplication, demonstrate FOXC1 dysfunction induces cerebral small vessel pathology. A causative role for FOXC1 in the development and maintenance of cerebral vasculature is supported by the cerebral hemorrhage generated by morpholino-induced suppression of FOXC1 orthologs in a zebrafish model system. Furthermore, in vivo imaging demonstrates profoundly impaired migration of neural crest cells and their subsequent association with nascent vasculature, a process required for the differentiation of perivascular mural cells. In addition, foxc1 inhibition reduces the expression of pdgfra, a gene critically required for vascular stability via its role in mural cell recruitment. Taken together, these data support a requirement for Foxc1 in stabilizing newly formed vasculature via recruitment of neural crest derived mural cells, and define a casual role for FOXC1 in cerebrovascular pathology.

Genetics & Brugada Syndrome

370/2:45 Genetic association of common variants with a rare cardiac disease, the Brugada syndrome, in a multi-centric study. C. Dina, J. Barc, Y. Mizusawa, C. A. Remme, J. B. Gourraud, F. Simonet, P. J. Schwartz, L. Crotti, P. Guicheney, A. Leenhardt, C. Antzelevitch, E. Schulze-Bahr, E. R. Behr, J. Tfelt-Hansen, S. Kaab, H. Watanabe, M. Horie, N. Makita, W. Shimizu, P. Froguel, B. Balkau, M. Gessler, D. Roden, V. M. Christoffels, H. Le Marec, A. A. Wilde, V. Probst, J. J. Schott, R. Redon, C. R. Bezzina.

Genetic association of common variants with a rare cardiac disease,

the Brugada Syndrome, in a multi-centric study. C. Dina1,2, J. Barc3, Y.

Mizusawa3, C.A. Remme3, J.B. Gourraud1,2, F. Simonet1, P.J. Schwartz4,

L. Crotti4, P. Guicheney5, A. Leenhardt6, C. Antzelevitch7, E. Schulze-Bahr8,

E.R. Behr9, J. Tfelt-Hansen10, S. Kaab11, H. Watanabe12, M. Horie13, N.

Makita14, W. Shimizu15, P. Froguel 16, B. Balkau17, M. Gessler18, D.

Roden19, V.M. Christoffels3, H. Le Marec1,2, A.A. Wilde3, V. Probst1,2, J.J.

Schott1,2, R. Redon1,2, C.R. Bezzina3.

1) Thorx Inst, INSERM UMR 1087, CNRS, Nantes, France;

2) CHU Nantes, l’institut du thorax, Nantes, France;

3) Heart Failure Research Center, Academic Medical Center, Amsterdam, Netherlands;

4) University of Pavia, Pavia, Italy;

5) InsermUMR956, UPMC, Paris, France;

6) Cardiology Unit, Hôpital Bichat, Assistance Publique- Hôpitaux de Paris, Nantes, France;

7) Department of Experimental Cardiology, Masonic Medical Research Laboratory, Utica, NY, United States;

8) Department of Cardiovascular Medicine, University Hospital, Münster, Germany;

9) Cardiovascular Sciences Research Centre, St George’s University, London, United Kingdom;

10) Laboratory of Molecular Cardiology, University of Copenhagen, Copenhagen, Denmark;

11) 1Department of Medicine I, Ludwig-Maximilians University, Munich, Germany;

12) Department of Cardiovascular Biology and Medicine, Niigata University Graduate School of Medical and Dental Sciences, Niigata, Japan;

13) Department of Cardiovascular and Respiratory Medicine, Shiga University of Medical Science, Otsu, Japan;

14) Department of Molecular Physiology, Graduate School of Biomedical Sciences, Nagasaki University, Nagasaki, Japan;

15) Division of Arrhythmia and Electrophysiology, Department of Cardiovascular Medicine, National Cerebral and Cardiovascular Center, Suita, Osaka, Japan;

16) CNRS UMR 8199, Pasteur Institute, Lille, France;

17) Inserm UMR 1018, Centre for research in Epidemiology and Population Health, Villejuif, France;

18) Theodor-Boveri-Institute, University of Wuerzburg, Wuerzburg, Germany;

19) Department of Medicine and Pharmacology, Vanderbilt University School of Medicine, Nashville, TN, United States.

The Brugada Syndrome (BrS) is considered as a rare Mendelian disorder with autosomal dominant transmission. BrS is associated with an increased risk of sudden cardiac death and specific electrocardiographic features consisting of ST-segment elevation in the right precordial leads. Loss-of-function mutations in SCN5A, encoding the pore-forming subunit of the cardiac sodium channel (Nav1.5), are identified in ~20% of patients. However, studies in families harbouring mutations in SCN5A have demonstrated low disease penetrance and in some instances absence of the familial SCN5A mutation in some affected members. These observations suggest a more complex inheritance model. To identify common genetic factors modulating disease risk, we conducted a genome-wide association study on 312 individuals with BrS and 1115 ancestry-matched controls. Two genomic regions displayed significant association. Both associations were replicated on two independent case/control sets from Europe (598/855) and Japan (208/1016) and a third locus emerged, all three with extremely significant p-values (1.10-14 down to 1.10-68). To our knowledge, this is the first time that several common variants are associated with a rare disease, with very high effect (Osdds-ratio) ranging from 1.58 to 2.55. While two loci displaying association hits had already been shown to influence ECG parameters in the general population, the third one encompasses a transcription factor which had never been related to cardiac arrhythmia. We showed that this factor regulates Nav1.5 channel expression in hearts of homozygous knockout embryos and influence cardiac conduction velocity in adult heterozygous mice. At last, we found that the cumulative effect of the 3 loci on disease susceptibility was unexpectedly large, indicating that common genetic variation may have a strong impact on predisposition to rare disease.

Mutations, Vasculopathy with Fever and Early Onset Strokes

371/3:00 Loss-of-function mutations in CECR1, encoding adenosine deaminase 2, cause systemic vasculopathy with fever and early onset strokes. Q. Zhou, A. Zavialov, M. Boehm, J. Chae, M. Hershfield, R. Sood, S. Burgess, A. Zavialov, D. Chin, C. Toro, R. Lee, M. Quezado, A. Ombrello, D. Stone, I. Aksentijevich, D. Kastner.

Loss-of-Function Mutations in CECR1, Encoding Adenosine Deaminase

2,Cause Systemic Vasculopathy with Fever and Early Onset

Strokes. Q. Zhou1, A. Zavialov2, M. Boehm3, J. Chae1, M. Hershfield4, R.

Sood5, S. Burgess6, A. Zavialov2, D. Chin1, C. Toro7, R. Lee8, M. Quezado9,

A. Ombrello1, D. Stone1, I. Aksentijevich1, D. Kastner1.

1) Inflammatory Disease Section, NHGRI, Bethesda, USA;

2) Turku Centre for Biotechnology, University of Turku, Turku, Finland;

3) Laboratory of Cardiovascular Regenerative Medicine, NHLBI, Bethesda, USA;

4) Department of Medicine, Duke University Medical Center, Durham, USA;

5) Zebrafish Core, NHGRI, Bethesda, USA;

6) Developmental Genomics Section, NHGRI, Bethesda, USA;

7) NIH Undiagnosed Diseases Program, NIH, Bethesda, USA;

8) Translational Surgical Pathology Section, NCI, Bethesda, USA;

9) General Surgical Pathology Section, NCI, Bethesda, USA.

We recently observed 5 unrelated patients with fevers, systemic inflammation, livedo reticularis, vasculopathy, and early-onset recurrent ischemic strokes. We performed exome sequencing on affected patients and their unaffected parents. The 5 patients shared 3 missense mutations in CECR1, encoding adenosine deaminase 2 (ADA2), with the genotypes A109D/ Y453C, Y453C/G47A, G47A/H112Q, R169Q/Y453C, and R169Q/28kb genomic deletion encompassing the 5’UTR and first exon of CECR1.

All mutations are either novel or present at low frequency (<0.001) in several large databases, consistent with the recessive inheritance. The Y453C mutation was present in 2/13004 alleles in an NHLBI database. Both alleles are found in 2 affected siblings who suffered from late-onset ischemic stroke, indicating that heterozygous mutations in ADA2 might be associated with susceptibility to adult stroke. Computer modeling based on the crystal structure of the human ADA2 suggests that CECR1 mutations either disrupt protein stability or impair ADA2 enzyme activity. All patients had at least a 10-fold reduction in serum and plasma concentrations of ADA2, and reduced ADA2-specific adenosine deaminase activity. Western blots showed a decrease in protein expression in supernatants of cultured patients’ cells. ADA2 is homologous to ADA1, which is mutated in some patients with SCID.

In contrast to ADA1, ADA2 is expressed predominantly in myeloid cells and is a secreted protein, and its affinity for adenosine is much less than ADA1. Animal models suggest that ADA2 is the prototype for a family of growth factors (ADGFs).Although there is no mouse homolog of CECR1, there are 2 zebrafish homologs, Cecr1a and Cecr1b. Using morpholino technology to knock down the expression of the ADA2 homologs, we observed intracranial hemorrhages in approximately 50% of the zebrafish embryos harboring the knockdown construct, relative to 3% in controls. Immunohistochemical studies of endothelial cells from patients’ skin biopsies demonstrate a diffuse systemic vasculopathy characterized by impaired endothelial integrity, endothelial cellular activation, and a perivascular infiltrate of CD8 T-cells and CD163-positive macrophages. ADA2 is not expressed in the endothelial cells. Our data suggest that ADA2 may be necessary for vascular integrity in the developing zebrafish as an endothelial cell-extrinsic growth factor, and that the near absence of functional ADA2 in patients may lead to strokes by a similar mechanism.

Genetics of Atherosclerotic Plaque in Patients with Chronic Coronary Artery Disease

372/3:15 Genetic influence on LpPLA2 activity at baseline as evaluated in the exome chip-enriched GWAS study among ~13600 patients with chronic coronary artery disease in the STABILITY (STabilisation of Atherosclerotic plaque By Initiation of darapLadIb TherapY) trial. L. Warren, L. Li, D. Fraser, J. Aponte, A. Yeo, R. Davies, C. Macphee, L. Hegg, L. Tarka, C. Held, R. Stewart, L. Wallentin, H. White, M. Nelson, D. Waterworth.

Genetic influence on LpPLA2 activity at baseline as evaluated in the exome chip-enrichedGWASstudy among ~13600 patients with chronic coronary artery disease in the STABILITY (STabilisation of Atherosclerotic plaque By Initiation of darapLadIb TherapY) trial.

L. Warren1, L. Li1, D. Fraser1, J. Aponte1, A. Yeo2, R. Davies3, C. Macphee3, L. Hegg3,

L. Tarka3, C. Held4, R. Stewart5, L. Wallentin4, H. White5, M. Nelson1, D.

Waterworth3.

1) GlaxoSmithKline, Res Triangle Park, NC;

2) GlaxoSmithKline, Stevenage, UK;

3) GlaxoSmithKline, Upper Merion, Pennsylvania, USA;

4) Uppsala Clinical Research Center, Department of Medical Sciences, Uppsala University, Uppsala, Sweden;

5) 5Green Lane Cardiovascular Service, Auckland Cty Hospital, Auckland, New Zealand.

STABILITY is an ongoing phase III cardiovascular outcomes study that compares the effects of darapladib enteric coated (EC) tablets, 160 mg versus placebo, when added to the standard of care, on the incidence of major adverse cardiovascular events (MACE) in subjects with chronic coronary heart disease (CHD). Blood samples for determination of the LpPLA2 activity level in plasma and for extraction of DNA was obtained at randomization. To identify genetic variants that may predict response to darapladib, we genotyped ~900K common and low frequency coding variations using Illumina OmniExpress GWAS plus exome chip in advance of study completion. Among the 15828 Intent-to-Treat recruited subjects, 13674 (86%) provided informed consent for genetic analysis. Our pharmacogenetic (PGx) analysis group is comprised of subjects from 39 countries on five continents, including 10139 Whites of European heritage, 1682 Asians of East Asian or Japanese heritage, 414 Asians of Central/South Asian heritage, 268 Blacks, 1027 Hispanics and 144 others. Here we report association analysis of baseline levels of LpPLA2 to support future PGx analysis of drug response post trial completion. Among the 911375 variants genotyped, 213540 (23%) were rare (MAF < 0.5%).

Our analyses were focused on the drug target, LpPLA2 enzyme activity measured at baseline. GWAS analysis of LpPLA2 activity adjusting for age, gender and top 20 principle component scores identified 58 variants surpassing GWAS-significant threshold (5e-08).

Genome-wide stepwise regression analyses identified multiple independent associations from PLA2G7, CELSR2, APOB, KIF6, and APOE, reflecting the dependency of LpPLA2 on LDL-cholesterol levels. Most notably, several low frequency and rare coding variants in PLA2G7 were identified to be strongly associated with LpPLA2 activity. They are V279F (MAF=1.0%, P= 1.7e-108), a previously known association, and four novel associations due to I1317N (MAF=0.05%, P=4.9e-8), Q287X (MAF=0.05%, P=1.6e-7), T278M (MAF=0.02%, P=7.6e-5) and L389S (MAF=0.04%, P=4.3e-4).

All these variants had enzyme activity lowering effects and each appeared to be specific to certain ethnicity. Our comprehensive PGx analyses of baseline data has already provided great insight into common and rare coding genetic variants associated with drug target and related traits and this knowledge will be invaluable in facilitating future PGx investigation of darapladib response.

Genetics of influence IL-18 regulation in patients with Acute Coronary Syndrome

373/3:30 Genome-wide association study identifies common and rare genetic variants in caspase-1-related genes that influence IL-18 regulation in patients with acute coronary syndrome. A. Johansson, N. Eriksson, E. Hagström, C. Varenhorst, A. Åkerblom, M. Bertilsson, T. Axelsson, B. J. Barratt, R. C. Becker, A. Himmelmann, S. James, H. A. Katus, G. Steg, R. F. Storey, A. Syvänen, L. Wallentin, A. Siegbahn.

Genome-wide association study identifies common and rare genetic

variants in caspase-1-related genes that influence IL-18 regulation in

patients with Acute Coronary Syndrome. A. Johansson1, 2, N. Eriksson1,

E. Hagström1,3, C. Varenhorst1,3, A. Åkerblom1,3, M. Bertilsson1, T. Axelsson4,

B.J. Barratt5, R.C. Becker6, A. Himmelmann7, S. James1,3, H.A.

Katus8, G. Steg9, R.F. Storey10, A. Syvänen4, L. Wallentin1,3, A. Siegbahn1,11.

1) Uppsala Clinical Research Center, Uppsala University, Sweden;

2) Department of Immunoloy, Genetics and Pathology, Uppsala University, Sweden;

3) Department of Medical Sciences, Cardiology, Uppsala University, Sweden;

4) Department of Medical Sciences, Molecular Medicine, Science for Life Laboratory, Uppsala University, Sweden;

5) AstraZeneca R&D, Alderley Park, Cheshire, UK;

6) Duke Clinical Research Institute, Duke University Medical Center, Durham, North Carolina, USA;

7) AstraZeneca Research and Development, Mölndal, Sweden;

8) Medizinishe Klinik, Universitätsklinikum Heidelberg, Heidelberg, Germany;

9) INSERM-Unité 698, Paris, France; Assistance Publique-Hôpitaux de Paris, Hôpital Bichat, Paris, France; Université Paris-Diderot, Sorbonne-Paris Cité, Paris, France;

10) Department of Cardiovascular Science, University of Sheffield, Sheffield, UK;

11) Department of Medical Sciences, Clinical Chemistry, Uppsala University, Sweden.

Interleukin 18 (IL-18) levels are increased in patients with acute coronary syndromes (ACS) and correlated with myocardial injury. We performed a genome-wide association study (GWAS) to identify genetic determinants of IL-18 levels in patients with ACS. In the PLATelet inhibition and patient Outcomes (PLATO) trial, enrolling a broad selection of ACS patients, baseline plasma IL-18 levels were measured in 16633 patients. Of these, 9340 were successfully genotyped using Illumina HumanOmni2.5 or HumanOmniExpressExome BeadChip and SNPs imputed using 1000 Genomes Phase I integrated variant set. Seven independent associations, in five chromosomal regions, were identified. The first region, with two independent (r2 = 0.11) association signals (rs34649619, p = 1.17*10−50 and rs360718, p = 2.03*10−12), is located within IL18. Both top SNPs are located in predicted promoter regions, and the insertion polymorphism rs34649619 (T/TA) disrupts a transcription factor binding site for FOXI1, FOXD3 and FOXA2. The second region, also represented by two independent (r2 = 0.003) association signals (rs385076, p = 6.99*10−72 and rs149451729, p = 3.79*10−16), is located in NLRC4. While rs385076 overlaps with a regulatory region, rs149451729 is a rare coding variant resulting in an amino acid substitution, predicted to be deleterious. The third region is located upstream of CARD16, CARD17, and CARD18 and one of the top SNPs (rs17103763, p = 6.19*10−9) has previously been associated with expression levels of CARD16. The two remaining chromosomal regions are located within GSFMF/MROH6 (rs2290414, p = 5.66*10−17) and RAD17 (rs17229943, p = 5.00*10−12).

While the latter genes have not been associated with IL-18 production previously, others are known to be involved in IL-18 release. NLRC4 is an inflammasome that activates the inflammatory cascade in the presence of bacterial molecules. It recruits and activates procaspase-1, which in its turn is responsible for the maturation of pro-IL-18. CARD16-18, also known as COP1, INCA and ICEBERG, encode caspase inhibitors, known to bind to and prevent procaspase-1 activation. Our results suggest that SNPs in IL18 and caspase-1-associated genes are important for IL-18 production. By combining the identified SNPs in a Mendelian randomization study, the causal effect of IL-18 on clinical endpoints could be further evaluated in a longitudinal study.

Thoracic Aortic Aneurysmal Genes

374/3:45 Prevalence and predictors of pneumothorax in patients with connective tissue disorders enrolled in the GenTAC (National Registry of Genetically Triggered Thoracic Aortic Aneurysms and Cardiovascular Conditions) Registry. J. P. Habashi, G. L. Oswald, K. W. Holmes, E. M. Reynolds, S. LeMaire, W. Ravekes, N. B. McDonnell, C. Maslen, R. V. Shohet, R. E. Pyeritz, R. Devereux, D. M. Milewicz, H. C. Dietz, GenTAC Registry Consortium.

Prevalence and Predictors of Pneumothorax in Patients with Connective Tissue Disorders Enrolled in the GenTAC (National Registry of Genetically Triggered Thoracic Aortic Aneurysms and Cardiovascular Conditions) Registry.

J.P. Habashi1, G.L. Oswald2, K.W. Holmes1,5, E.M.

Reynolds10, S. LeMaire3, W. Ravekes1, N.B. McDonnell4, C. Maslen5, R.V.

Shohet6, R.E. Pyeritz7, R. Devereux8, D.M. Milewicz9, H.C. Dietz2, GenTAC

Registry Consortium.

1) Dept Pediatric Cardiology, Johns Hopkins Univ, Baltimore, MD;

2) Dept. Medical Genetics, Johns Hopkins Univ, Baltimore, MD;

3) Baylor College of Medicine, Houston TX;

4) NIA at Harbor Hospital, Baltimore, MD;

5) Oregon Health & Science University, Portland, OR;

6) Queen’s Medical Center, Honolulu, HI;

7) The University of Pennsylvania, Philadelphia, PA; 8) Weill Cornell Medical College of Cornell University, New York NY;

9) University of Texas Medical School at Houston, Houston, TX;

10) University of Maryland, Baltimore, MD.

Spontaneous pneumothorax—described as escape of air into the pleural space surrounding the lung in the absence of traumatic injury—is a rare occurrence in the general population (0.1-0.5%), however is well recognized in Marfan syndrome (MFS)(4-5%). Associations between pneumothorax and other connective tissue disorders (CTDs) are less well recognized. We sought to examine potential associations of

• pneumothorax with MFS,
• vascular Ehlers-Danlos syndrome (vEDS) and other CTDs.

Phenotypic data were analyzed on all GenTAC patients with confirmed diagnoses of

• MFS,
• vEDS,
• Loeys-Dietz syndrome (LDS),
• bicuspid aortic valve with aortic enlargement (BAVe) or
• familial thoracic aortic aneurysm and dissection (FTAAD)

to assess the prevalence of pneumothorax and associated features (1918 total pts).

Of 695 patients with Ghent criteria-confirmed MFS, 73 had experienced a spontaneous pneumothorax (prevalence 10.5%), higher than reported in the literature. The frequency of pneumothorax in vEDS patients (16/107, 15%) was similar to the frequency in the MFS group. The prevalences of pneumothorax in LDS (4/73, 5.5%), FTAAD (13/237, 5.5%), and BAVe (19/ 806, 2.4%) were significantly less than that for MFS and vEDS (p<0.001), yet greater than reported for the general population. In MFS patients with a pneumothorax, there was a three-fold increase in reported skeletal features of pectus carinatum, pectus excavatum, scoliosis and/or kyphosis compared to those without pneumothorax. Similarly, in vEDS, there was a four-fold increase in pectus carinatum, scoliosis and kyphosis in those patients with a pneumothorax compared to those without pneumothorax. In a subset of patients with self-reported data (n=846), smoking was not associated with increased prevalence of pneumothorax. Gender was not a predictor of pneumothorax in any of the diagnostic categories analyzed despite literature reports of increased prevalence in males. In patients enrolled in the GenTAC registry with a diagnosis of MFS, vEDS, BAVe, FTAAD or LDS, the prevalence of pneumothorax was significantly increased in all CTDs analyzed as compared to the general population. The prevalence of pneumothorax was significantly higher in patients with MFS or vEDS than in the other CTDs.

These data suggest that skeletal features may be a predictor for pneumothorax. Patients presenting with a spontaneous pneumothorax should be evaluated for several potential CTDs; such an evaluation could unmask an undiagnosed aortic aneurysm.

375/4:00 Surprising clinical lessons from targeted next-generation sequencing of thoracic aortic aneurysmal genes. B. Loeys, D. Proost, G. Vandeweyer, S. Salemink, M. Kempers, G. Oswald, H. Dietz, G. Mortier, L. Van Laer.

Surprising clinical lessons from targeted next generation sequencing of thoracic aortic aneurysmal genes. B. Loeys1,2, D. Proost1, G. Vandeweyer1, S. Salemink2, M. Kempers2, G. Oswald3, H. Dietz3, G. Mortier1, L. Van Laer1.

1) Center for Medical Genetics, University of Antwerp/ Antwerp University Hospital, Antwerp, Belgium;

2) Department of Genetics, Radboud University Medical Center, Nijmegen, The Netherlands;

3) Mc Kusick Nathans Institute for Genetic Medicine, Johns Hopkins University Hospital, Baltimore, USA.

Thoracic aortic aneurysm/dissection (TAA), an important cause of death in the industrialized world, is genetically heterogeneous and at least 14 causative genes have been identified, accounting for both syndromic and non-syndromic forms. The diagnosis is not always straightforward because a considerable clinical overlap exists between patients with mutations in different genes, and mutations in the same gene cause a wide phenotypic variability. Molecular confirmation of the diagnosis is becoming increasingly important for gene-tailored patient management but consecutive, conventional molecular TAA gene screening is expensive and labor-intensive. To shorten the turn-around-time, to increase mutation-uptake and to reduce the overall cost of molecular testing, we developed a TAA gene panel for next generation sequencing (NGS) of 14 TAA genes (ACTA2, COL3A1, EFEMP2, FBN1, FLNA, MYH11, MYLK, NOTCH1, SKI, SLC2A10, SMAD3, TGFB2, TGFBR1 and TGFBR2). We obtained enrichment with Haloplex technology and performed 2×150 bp paired-end runs on a Miseq sequencer in a series of 57 consecutive TAA patients, both syndromic and non-syndromic.

The sensitivity and false positive rate were previously shown to be 100% and 3%, respectively. Applying our NGS approach, we identified a causal mutation in 16 patients (28%). This uptake is really high as on average one molecular study per patient (range 0-6) was performed prior to inclusion in this study. One mutation was found in each of the 6 following genes: ACTA2, COL3A1, TGFBR1, MYLK, SMAD3, SLC2A10 (homozygous); two mutations inNOTCH1and eight in FBN1. An additional 6 variants of unknown significance were identified: 2 in FLNA, 2 in NOTCH1, 1 in FBN1 and 1 heterozygous in EFEMP2. All variants were confirmed by Sanger sequencing.

Remarkably, from the eight FBN1 positive patients, three patients had previously been tested FBN1 negative by certified labs, indicating that the sensitivity of Sanger sequencing is not 100%. Interestingly, in two FBN1 mutation positive patients

• the clinical diagnosis of Marfan syndrome was unsuspected. Similarly,
• the clinical diagnosis of vascular Ehlers-Danlos syndrome (COL3A1) had not been made. Finally,
• the ACTA2 mutation was identified postmortem from paraffin-embedded extracted DNA.

We conclude that our NGS approach for TAA genetic testing overcomes the intrinsic hurdles of Sanger sequencing and becomes a powerful tool in the elaboration of clinical phenotypes assigned to different genes.

Read Full Post »