- Email: [email protected]
Gain-of-function mutations in GATA6 lead to atrial fibrillation
Author’s Accepted Manuscript Gain of function mutations in GATA6 lead to atrial fibrillation Nathan R. Tucker, Saagar Mahida, Jiangchuan Ye, Elizabeth J. Abraham, Julie A. Mina, Victoria A. Parsons, Michael A. McLellan, Marisa A. Shea, Alan Hanley, Emelia J. Benjamin, David J. Milan, Honghuang Lin, Patrick T. Ellinor
PII: DOI: Reference:
www.elsevier.com/locate/buildenv
S1547-5271(16)30915-8 http://dx.doi.org/10.1016/j.hrthm.2016.10.014 HRTHM6894
To appear in: Heart Rhythm Cite this article as: Nathan R. Tucker, Saagar Mahida, Jiangchuan Ye, Elizabeth J. Abraham, Julie A. Mina, Victoria A. Parsons, Michael A. McLellan, Marisa A. Shea, Alan Hanley, Emelia J. Benjamin, David J. Milan, Honghuang Lin and Patrick T. Ellinor, Gain of function mutations in GATA6 lead to atrial fibrillation, Heart Rhythm, http://dx.doi.org/10.1016/j.hrthm.2016.10.014 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Gain of Function Mutations in GATA6 Lead to Atrial Fibrillation
Nathan R. Tucker,1¶ PhD, Saagar Mahida,1¶ MB, ChB, Jiangchuan Ye,1 MD, PhD, Elizabeth J. Abraham,1 BS, Julie A. Mina,1 BS, Victoria A. Parsons,1 BS, Michael A. McLellan,1 BS, Marisa A. Shea,2 BS, RN, Alan Hanley,1 MD, Emelia J. Benjamin,3-5 MD, MSc, David J. Milan,1,2 MD, Honghuang Lin,6PhD, and Patrick T. Ellinor,1,2* MD, PhD
1 Cardiovascular Research Center, Massachusetts General Hospital, Charlestown, MA, USA 2 Cardiac Arrhythmia Service, Massachusetts General Hospital, Boston, USA 3 National Heart, Lung, and Blood Institute's and Boston University's Framingham Heart Study, Framingham, Massachusetts, USA. 4 Preventive Medicine and Cardiovascular Medicine Sections, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts, USA. 5 Cardiology Division, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts, USA. 6 Computational Biomedicine Section, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts, USA.
* Corresponding Author Patrick T. Ellinor, MD, PhD Cardiovascular Research Center Massachusetts General Hospital 55 Fruit Street, Gray 109 Boston, Massachusetts, 02114 Email: [email protected]
¶ These authors contributed equally to this work
Running title: GATA6 mutations in AF Conflict statement: Dr. Ellinor is a Principal Investigator on a grant from Bayer Healthcare to the Broad Institute.
1
Abstract Background: The genetic basis of atrial fibrillation (AF) and congenital heart disease remains incompletely understood. Objective: We sought to determine the causative mutation in a family with AF, atrial septal and ventricular septal defects. Methods: We evaluated a pedigree with 16 family members, one with an atrial septal defect, one with a ventricular septal defect and three with AF; we performed whole exome sequencing in three affected family members. Given that early-onset AF was prominent in the family, we then screened individuals with early-onset AF, defined as an age of onset < 66 years, for mutations in GATA6. Variants were functionally characterized using reporter assays in a mammalian cell line. Results: Exome sequencing in three affected individuals identified a conserved mutation, R585L, in the transcription factor gene, GATA6. In the MGH AF Study the mean age of AF onset was 47.1 ± 10.9 years, 79% of the participants were male, and there was no evidence of structural heart disease. We identified three GATA6 variants (P91S, A177T, and A543G). Using wild-type and mutant GATA6 constructs driving NPPA promoter reporter, we found that three of the four variants had a marked upregulation of luciferase activity (R585L; 4.1 fold, p<0.0001; P91S; 2.5 fold, p=0.0002; A177T; 1.7 fold, p=0.03). Additionally, when co-overexpressed with GATA4 and MEF2C, GATA6 variants exhibited upregulation of the αMHC and NPPA activity. Conclusion: Overall, we found gain-of-function mutations in GATA6 in both a family with earlyonset AF and atrioventricular septal defects as well as in sporadic, early-onset AF.
Keywords: Atrial fibrillation, mutation, transcription factor, genetics
Introduction Atrial fibrillation (AF) is the most common cardiac arrhythmia and over the coming decades, the prevalence is projected to rise.1 Despite the widespread prevalence and consequences of AF, the molecular mechanisms underlying the arrhythmia remain incompletely understood. Thus, the identification of the genetic basis of AF is an important public health concern.
2
AF is heritable with a two-fold increased risk of the arrhythmia if one parent has AF.2–5 Over the past two decades, classical genetic studies in familial forms of AF have led to the identification of a series mutations in ion channel and non-ion channel genes.6 Among these is GATA6, a member of a family of six transcription factors that bind to a conserved consensus sequence.7 Loss-of-function variants in GATA6 have been previously reported to cause atrial fibrillation8–10 as well as congenital heart defects such as Tetralogy of Fallot and atrial septal defects (ASD).11,12
In the present study, we describe the characterization of a family with a complex phenotype of early-onset AF, ASD and ventricular septal defects (VSD) due to a novel gain-of-function variant in GATA6. Further, we present the results of candidate-gene screening for GATA6 in a large cohort with early-onset AF and the functional evaluation of the identified variants.
Methods Clinical evaluation of Family AF-772 and the MGH AF Study Sixteen members of the family underwent a standardized interview, physical examination, 12lead electrocardiogram and echocardiography. Affected individuals were identified as those with either AF on the electrocardiogram or patients with a documented ASD or VSD during echocardiographic analysis. All patients in the MGH AF Study13 developed AF at an early age (AF onset less than 66 years) and had structurally normal hearts. The mean age of onset in AF cases was 47.1 ± 10.9
3
years. Consistent with previous reports, the AF cases were predominantly male with a male to female ratio of approximately 4 to 1. 45% of AF cases had a family history of AF. Baseline characteristics of the AF cohort are presented in Supplemental Table 1. Written informed consent was obtained from all patients prior to performing the evaluations. The study was approved by the Institutional Review Board at Massachusetts General Hospital and complied with the Declaration of Helsinki.
Exome sequencing DNA was isolated from venous blood samples using the Puregene DNA Purification Kit (Gentra Systems, Inc.). Whole-exome sequencing was performed in three affected members of the family (22, 72 and 73). The exome regions were captured by the Agilent Sure Select Target Enrichment Assay (version 3). The enriched fragments were then pair-end sequenced by the Illumina HiSeq 2000 platform at Perkin Elmer.
Bioinformatic analyses BWA was used to align the short reads to the reference human genome (NCBI Build 37, hg19), which were then converted into BAM files by SAMtools. We also performed local realignment to minimize mismatching of bases across all the reads, and base quality recalibration to correct base qualities using the GATK software package. Duplicate reads were marked by Mark Duplicates using the Picard software package. Only the first read pair in each duplicate cluster was kept for downstream analysis. Single nucleotide variations (SNVs) were then called by the
4
Unified Genotype function using the GATK software package. SNVs with quality scores lower than 20 or strand bias higher than 60 were excluded.
Filtration and prioritization of the variants was performed according to previously published methods.14 First, since it is unlikely that common polymorphisms in the general population underlie the phenotype, all common variants (>0.01%) identified in publically available databases (dbSNP,15 1000 Genomes Project16 and Exome Variant Server) were removed from further analyses. Second, it is likely that Mendelian diseases are caused by variants that result amino acid substitution, therefore synonymous variants were excluded. Analysis of expression levels was performed using the GNF Expression Atlas 2 data from U133A and GNF1H chips in UCSC browser.
Functional evaluation of HHATL activity The Sonic hedgehog (SHH) gene was received from plasmID at Harvard Medical School whereas HHATL and mutant K484N HHATL constructs were synthesized by Gen9 Bio (Cambridge, MA). COS7 cells were transfected using Lipofectamine LTX with Plus reagent (ThermoFisher). Western blotting was performed according to standard protocols. The antibodies used were anti-SHH N-terminal fragment (5E1; Developmental Studies Hybridoma Bank); anti-SHH (H-160; Santa Cruz Biotechnology Inc); anti-EGFP (JL-8; Fisher Scientific). Activity of wild-type and mutant HHATL was determined by evaluating the ratio of palmitoylated SHH to total SHH using the 5E1 and H160 antibodies, respectively, in three biological replicates.
5
Screening for GATA6 variants in the MGH AF Study A combination of high resolution melting and Sanger sequencing was used to perform candidate-gene screening. Exons 3, 5, and 6 of the GATA6 gene were screened using high resolution melting whereas the remaining coding exons (2, 4, and 7) were sequenced directly using Sanger sequencing. Light Scanner technology (Idaho Technologies Inc) was used to perform high resolution melting.
Luciferase assays The functional effect of the variant was analyzed by performing luciferase assays using coexpression of GATA6 variants in pCMV6 XL6 with three luciferase reporter constructs, NPPApromoter-luc (0.9kb), NPPB-promoter-luc (1.3kb), and αMHC-promoter-luc (0.7kb) in pGL4.10. NPPA, NPPB and αMHC are downstream target genes for GATA transcription factors and harbor GATA dependent promoters.17 CHO cells were transfected with reporter constructs and pRLSV40 according to manufacturer specifications using Lipofectamine LTX. Luciferase activities were measured at 24 hours using the dual luciferase reporter assay system (Promega). All experiments were performed in three independent biological replicates. For statistical analyses, two-way ANOVA was performed followed by multiple comparisons with Bonferroni correction.
6
Results Clinical characteristics of Family AF-772 We evaluated a total of 16 individuals from pedigree AF-772, a two-generation family with a complex phenotype of AF, ASD and VSD (Figure 1 and Table 1). The trait was inherited with an autosomal dominant pattern with five members of the pedigree being affected. Individual 7, the proband in the family, initially presented with shortness of breath, palpitations and two episodes of near syncope at age 51. Subsequent investigations revealed the presence of an ASD. She underwent surgery to correct an ostium primum ASD. Two of her descendants, 71 and 72, were diagnosed with AF at the ages of 14 and 16 years respectively. The proband´s sister, 2, had a single documented paroxysm of AF at age 49. The proband´s nephew, 22, had a VSD identified as an infant; there was no history of surgical intervention for the VSD. None of the other members of the family had a history of ASD, VSD or AF.
Exome sequencing to identify the causative variant Exome sequencing was performed in individuals 22, 72 and 73 using the Agilent Sure Select Target Enrichment Assay and the Illumina HiSeq 2000 platform. On average, 95.7% of short reads were mapped, reaching an average ~58-fold depth of coverage. 88.5% of targeted regions were covered by at least 10 unique reads (Supplemental Table 2).
An average of 18,825 variants was identified in the three affected individuals who underwent exome sequencing (Supplemental Table 3). The first stage of filtering of candidate variants
7
involved exclusion of synonymous variants and common variants reported in population databases (dbSNP,15 1000 Genomes Project,16 Exome Variant Server, and the EXac database18). Common variants were defined as variants with a minor allele frequency (MAF) of >5%. Rare variants were defined as variants with a MAF of 0.5-5%. Very rare variants were defined as variants with a MAF of. ≤0.5%. The number of novel or very rare missense variants (defined as singleton variants in public databases) for each individual varied between 385 and 548, nonsense variants varied between 7 and 17, and splice site variants varied between 0 and 7 (Supplemental Table 3). Subsequent filtering involved exclusion of variants that did not segregate with disease in individuals 22, 72 and 73. Using this strategy, the number of candidate variants was reduced to 13 novel or very rare missense variants, which were located in 11 genes (Supplemental Table 4). As individuals 72 and 73 developed AF at a young age, we analyzed the subset of variants shared by these individuals to identify potential variants that may modulate the effects of the disease-causing variant. We did not identify any compelling variants (Supplemental Table 5).
Prioritization of the 13 candidate variants was based on gene expression profiles, interspecies conservation, predicted effect on protein function, and involvement in cardiacrelated functional pathways. Expression data was available for 9 out of 11 candidate genes. Of these, four genes, (GATA6, TTN, HHATL and CCDC69) displayed high cardiac expression while the remaining 5 genes had either low or intermediate expression. Ten variants were located in residues that are conserved through evolution. Of note, one of the variants (R585L) was located in GATA6 with the highest Phylop conservation score of 4.94. (Supplemental Table 4).
8
After overall evaluation of these variant prioritization schemes, more comprehensive segregation analysis was performed for 5 of the 13 aforementioned candidate variants which exhibited cardiac expression, had a positive PhyloP conservation score, and had plausible biological links to cardiac disease following examination of literature. Segregation analysis was performed by direct Sanger sequencing of all 16 members of the family. The results from segregation analysis are included in Supplemental Table 6.
Among variants that underwent segregation analysis, R585L-GATA6 and K484N-HHATL represented the most compelling candidate variants for AF, ASD and VSD in pedigree AF-772. As HHATL is known to act as an antagonist to the HHAT-mediated addition of palmitate to the N-terminus of sonic hedgehog (SHH),19 we analyzed the HHATL variant function in vitro. We found no difference in the activity of mutant HHATL to inhibit the N-palmitoylation of SHH (Supplemental Figure 1). In contrast, a number of functional studies provide further evidence to suggest that the R585-GATA6 variant is instead the causative variant. GATA6, encoding a transcription factor that plays a critical role during cardiac development and morphogenesis, has previously been implicated in congenital heart defects, including ASD.
Five additional GATA6-R585L variant carriers, individuals 2, 3, 7, 9, and 21 were identified (Figure 1). Individual 7 had an ASD while individual 2 had paroxysmal AF. While individuals 3, 9, and 21 did not display any overt structural cardiac abnormalities or arrhythmias, individual 3 did demonstrate an abnormal P wave axis with P wave inversion in the lateral leads (I and AVL)
9
and evidence of right ventricular conduction delay (Supplemental Figure 2). Individuals 9 and 21 had normal baseline ECGs. There were no significant differences between ECG parameters (PR, QRS and QT interval) between mutation carriers and patients without the mutation (Table 1). We were unable to perform echocardiography with bubble contrast or Holter monitoring to exclude subtle septal defects and asymptomatic episodes of AF in individuals 3, 9 and 21. In the absence of this, the findings cannot be definitively attributed to incomplete penetrance.
Identification of GATA6 variants in a study with early-onset AF Given our findings in AF-772, we screened for GATA6 variants in 546 patients with AF from the MGH AF Study.13 We identified novel or very rare GATA6 variants in three out of the 546 earlyonset AF patients (Supplemental Table 7). One of the variants, A543G, was located in the Cterminus, while the other two variants, P91S and A177T, were located at the N-terminus of the peptide (Figure 2). None of these variants have previously been reported in the Exome Variant Server which contains sequence data from more than 6,500 individuals; however, the P91S variant was reported in a single individual among the ~2500 individuals sequenced in the 1000 Genomes Project.16 Similarly to R585L, the A177T variant was strongly conserved through evolution as assessed by the Phylop and GERP scores whereas the P91S and A543G residues were subject to more variability in mammals. The characteristics of the variants are summarized in Supplemental Table 8.
10
Functional evaluation of the GATA6 variants To identify the potential effects of the GATA6 variants on protein function, we evaluated the ability of exogenously expressed wild-type and mutant GATA6 to activate known GATA familytargeted promoters upstream of a luciferase reporter. Overexpression of wild-type GATA6 upregulated expression of luciferase in the promoters of the NPPA, NPPB, and αMHC genes, thus confirming GATA6 mediated activation of these promoters (Figure 3A). The R585L variant identified in family AF-772 exhibited a significant increase in luciferase activity with NPPA (4.1 fold, p<0.0001) and αMHC (1.7 fold, p=0.03) reporter constructs (Figure 3B); a slight, but nonsignificant, increase was seen with the NPPB construct.
The P91S variant resulted in an upregulation in luciferase activity with NPPA (2.5 fold, p=0.0002) and αMHC (1.7 fold, p=0.03) constructs, but not with an NPPB construct. The A177T variant resulted in upregulation of luciferase activity with NPPA (1.7 fold, p=0.03) but had no effect with αMHC and NPPB constructs. Finally, the A543G mutant displayed no net effect with any of the constructs compared to control. Of note, in comparison to the R585L variant, upregulation of luciferase activity associated with the P91S and A177T variants was less pronounced and more variable. This differential activity is not due to altered expression or localization, which is unaltered by inclusion of the variants (Supplemental Figure 3).
GATA6 has been reported to interact with a range of other transcription factors and co-factor proteins including TBX5, GATA4, NKX2.5, and MEF2C.17,22–25 To investigate synergistic effects with the GATA6 mutants, we co-overexpressed GATA6 constructs with TBX5, GATA4, NKX2-5, or
11
MEF2C. Of the four constructs, only NKX2.5 upregulated the expression of all three promoters independently of GATA6. Importantly, when overexpressed with GATA4 or MEF2C, all variants exhibited statistically significant upregulation of the NPPA and αMHC promoters as compared to control (Figure 4). Additionally, for all variants, co-expression with TBX5 demonstrated a significant upregulation of the αMHC promoter.
Discussion We performed whole exome sequencing in a large pedigree with a complex phenotype of earlyonset AF, ASD and VSD and identified a highly conserved mutation in the transcription factor gene, GATA6. We subsequently performed candidate-gene screening in a large cohort of patients with early-onset AF and identified three further GATA6 variants. All four variants exhibit a gain-of-function effect.
Loss-of-function mutations in GATA6 have been reported to underlie structural congenital cardiac defects such as ASD, VSD, Tetralogy of Fallot, and persistent truncus arteriosus.11,12,26 Interestingly, a gain-of-function GATA6 mutation has also been reported to cause Tetralogy of Fallot.11 In addition to the reports linking GATA6 mutations with structural cardiac defects, candidate-gene studies have identified four variants in the GATA6 gene (G469V, Y235S, Q206P, and Y265X) in patients with lone AF.8–10 Two of the variants (G469V and Y235S) have been defined as loss-of-function.
12
The findings of the present study further expand on the potential role of GATA6 variants in the pathogenesis of AF. While the frequency with which GATA6 variation contributes to AF at the population level remains unclear, the present study, when combined with previous work, suggest that GATA6 is an important target that should continue to be examined by whole exome and whole genome sequencing as larger cohorts are evaluated.
The specific variants found in the present study provide insight into the function of N- and Cterminal domains of the GATA6 protein. The R585L and A543G variants are located in highly conserved regions of the C-terminus of the peptide. Although the specific function of the Cterminus in the GATA transcription factors has remained unclear, truncation mutants of the GATA4 C-terminus resulted in an 80% reduction in transcriptional activity.27 The role of the Nterminal region which contains the P91S and A177T variants is similarly unclear. Previous deletion studies of the GATA family members suggest that this region contains critical domains for transcriptional activation of GATA4, GATA5 and GATA6, perhaps through transactivation domain activity.27 These findings are consistent with our results that indicate a critical role for the N- and C-terminal residues in the regulation of GATA6 transcriptional activation.
Of note, the R585L identified in Family AF-772 and the P91S and A543G variants found in individuals with early-onset AF, were identified in rare individuals in the 1000 Genomes Project and the ExAC database. Since the Exome Variant Server, the 1000 Genomes Projects, and the ExAC database contain data on ~6500, ~2500 and ~61000 individuals respectively, the estimated frequency would be 1 in ~18,000 alleles. The transmission of R585L in the family as
13
well as the clear gain-of-function effect strongly support R585L in the pathogenesis of the phenotype. However, our study highlights the challenges in using SNP repository data as a “control” population in the absence of clinical or functional data.
From a mechanistic perspective, GATA6 variants could influence susceptibility to AF in a number of ways. GATA6 is responsible for the maintenance of cardiac precursor cells in an undifferentiated proliferative state during cardiac development.28 GATA6 variants could cause abnormal differentiation of cardiomyocytes, a process which has been linked to an susceptibility to triggered arrhythmias.29 In addition to its role in cardiac development, GATA6 also regulates genes in the adult heart.30 Therefore, it is plausible that GATA6 variants cause altered expression of genes that maintain electrical stability in the atrium.
The GATA6 R585L variant is associated with disparate clinical phenotypes in pedigree 772. These observations include that of an ostium primum ASD, which is contrasted by the ostium secundum ASD in a previous report. It is notable that the previous report identified a loss-offunction variant, so it is possible that the direction of effect may impact the clinical phenotype. The finding of diverse clinical presentations for GATA6 variants are consistent with previous reports indicating that single transcription factor mutations result in multiple different phenotypes.31 During cardiac development, GATA transcription factors play a role in orchestrating complex regulatory networks involving multiple genes.32 It is plausible that the variable clinical phenotypes associated with GATA6 R585L are mediated by unidentified additional variants in genes involved in these regulatory pathways. At this stage, as proposed
14
mechanisms remain speculative, work in model systems would be important for better understanding pathways that may underlie the pathogenesis observed in this and other studies.
Conclusion In sum, we have found that gain of GATA6 function influences susceptibility to phenotypes ranging from early-onset AF to atrial and ventricular septal defects. Our findings further expand on the role of transcription factors in the pathogenesis of congenital heart disease and AF.
References 1. 2.
3. 4. 5.
6. 7.
8. 9. 10. 11.
Kannel WB, Benjamin EJ: Status of the epidemiology of atrial fibrillation. Med Clin North Am NIH Public Access, 2008; 92:17–40, ix. Fox CS, Parise H, D’Agostino RB, Lloyd-Jones DM, Vasan RS, Wang TJ, Levy D, Wolf PA, Benjamin EJ: Parental atrial fibrillation as a risk factor for atrial fibrillation in offspring. JAMA American Medical Association, 2004; 291:2851–2855. Arnar DO, Thorvaldsson S, Manolio TA, Thorgeirsson G, Kristjansson K, Hakonarson H, Stefansson K: Familial aggregation of atrial fibrillation in Iceland. Eur Heart J 2006; 27:708–712. Ellinor PT, Yoerger DM, Ruskin JN, MacRae CA: Familial aggregation in lone atrial fibrillation. Hum Genet 2005; 118:179–184. Darbar D, Herron KJ, Ballew JD, Jahangir A, Gersh BJ, Shen W-K, Hammill SC, Packer DL, Olson TM: Familial atrial fibrillation is a genetically heterogeneous disorder. J Am Coll Cardiol 2003; 41:2185–2192. Mahida S, Lubitz SA, Rienstra M, Milan DJ, Ellinor PT: Monogenic atrial fibrillation as pathophysiological paradigms. Cardiovasc Res 2011; 89:692–700. Molkentin JD: The Zinc Finger-containing Transcription Factors GATA-4, -5, and -6: UBIQUITOUSLY EXPRESSED REGULATORS OF TISSUE-SPECIFIC GENE EXPRESSION. J Biol Chem American Society for Biochemistry and Molecular Biology, 2000; 275:38949–38952. Yang Y-Q, Wang X-H, Tan H-W, Jiang W-F, Fang W-Y, Liu X: Prevalence and spectrum of GATA6 mutations associated with familial atrial fibrillation. Int J Cardiol 2012; 155:494–496. Yang Y-Q, Li L, Wang J, Zhang X-L, Li R-G, Xu Y-J, Tan H-W, Wang X-H, Jiang J-Q, Fang W-Y, Liu X: GATA6 loss-of-function mutation in atrial fibrillation. Eur J Med Genet 2012; 55:520–526. Li J, Liu W-D, Yang Z-L, Yang Y-Q: Novel GATA6 loss-of-function mutation responsible for familial atrial fibrillation. Int J Mol Med 2012; 30:783–790. Maitra M, Koenig SN, Srivastava D, Garg V: Identification of GATA6 sequence variants in patients with congenital heart defects. Pediatr Res 2010; 68:281–285.
15
12.
13. 14. 15. 16.
17.
18. 19.
20.
21.
22.
23.
24.
25. 26.
27.
28.
29.
Kodo K, Nishizawa T, Furutani M, Arai S, Yamamura E, Joo K, Takahashi T, Matsuoka R, Yamagishi H: GATA6 mutations cause human cardiac outflow tract defects by disrupting semaphorin-plexin signaling. Proc Natl Acad Sci U S A 2009; 106:13933–13938. Ellinor PT, Lunetta KL, Albert CM, et al.: Meta-analysis identifies six new susceptibility loci for atrial fibrillation. Nat Genet NIH Public Access, 2012; 44:670–675. Li M-X, Gui H-S, Kwan JSH, Bao S-Y, Sham PC: A comprehensive framework for prioritizing variants in exome sequencing studies of Mendelian diseases. Nucleic Acids Res 2012; 40:e53. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM, Sirotkin K: dbSNP: the NCBI database of genetic variation. Nucleic Acids Res Oxford University Press, 2001; 29:308–311. Abecasis GR, Altshuler D, Auton A, Brooks LD, Durbin RM, Gibbs RA, Hurles ME, McVean GA: A map of human genome variation from population-scale sequencing. Nature 2010; 467:1061– 1073. Charron F, Paradis P, Bronchain O, Nemer G, Nemer M: Cooperative interaction between GATA-4 and GATA-6 regulates myocardial gene expression. Mol Cell Biol American Society for Microbiology (ASM), 1999; 19:4355–4365. Lek M, Karczewski KJ, Minikel E V., et al.: Analysis of protein-coding genetic variation in 60,706 humans. Nature Nature Research, 2016; 536:285–291. Abe Y, Kita Y, Niikura T: Mammalian Gup1, a homolog of Saccharomyces cerevisiae glycerol uptake/transporter 1, acts as a negative regulator for N-terminal palmitoylation of Sonic hedgehog. FEBS J 2008; 275:318–331. De Franco E, Shaw-Smith C, Flanagan SE, Shepherd MH, International NDM Consortium, Hattersley AT, Ellard S: GATA6 mutations cause a broad phenotypic spectrum of diabetes from pancreatic agenesis to adult-onset diabetes without exocrine insufficiency. Diabetes 2013; 62:993–997. Lango Allen H, Flanagan SE, Shaw-Smith C, De Franco E, Akerman I, Caswell R, International Pancreatic Agenesis Consortium, Ferrer J, Hattersley AT, Ellard S: GATA6 haploinsufficiency causes pancreatic agenesis in humans. Nat Genet 2012; 44:20–22. Davis DL, Edwards A V, Juraszek AL, Phelps A, Wessels A, Burch JB: A GATA-6 gene heart-regionspecific enhancer provides a novel means to mark and probe a discrete component of the mouse cardiac conduction system. Mech Dev 2001; 108:105–119. Maitra M, Schluterman MK, Nichols HA, Richardson JA, Lo CW, Srivastava D, Garg V: Interaction of Gata4 and Gata6 with Tbx5 is critical for normal cardiac development. Dev Biol 2009; 326:368– 377. Molkentin JD, Antos C, Mercer B, Taigen T, Miano JM, Olson EN: Direct activation of a GATA6 cardiac enhancer by Nkx2.5: evidence for a reinforcing regulatory network of Nkx2.5 and GATA transcription factors in the developing heart. Dev Biol 2000; 217:301–309. Morin S, Charron F, Robitaille L, Nemer M: GATA-dependent recruitment of MEF2 proteins to target promoters. EMBO J European Molecular Biology Organization, 2000; 19:2046–2055. Wang J, Luo X-J, Xin Y-F, Liu Y, Liu Z-M, Wang Q, Li R-G, Fang W-Y, Wang X-Z, Yang Y-Q: Novel GATA6 mutations associated with congenital ventricular septal defect or tetralogy of fallot. DNA Cell Biol Mary Ann Liebert, Inc., 2012; 31:1610–1617. Morrisey EE, Ip HS, Tang Z, Parmacek MS: GATA-4 Activates Transcription Via Two Novel Domains That Are Conserved within the GATA-4/5/6 Subfamily. J Biol Chem American Society for Biochemistry and Molecular Biology, 1997; 272:8515–8524. Gove C, Walmsley M, Nijjar S, Bertwistle D, Guille M, Partington G, Bomford A, Patient R: Overexpression of GATA-6 in Xenopus embryos blocks differentiation of heart precursors. EMBO J European Molecular Biology Organization, 1997; 16:355–368. Zhang YM, Hartzell C, Narlow M, Dudley SC: Stem cell-derived cardiomyocytes demonstrate
16
30.
31. 32.
arrhythmic potential. Circulation 2002; 106:1294–1299. van Berlo JH, Elrod JW, van den Hoogenhof MMG, York AJ, Aronow BJ, Duncan SA, Molkentin JD: The Transcription Factor GATA-6 Regulates Pathological Cardiac Hypertrophy. Circ Res Lippincott Williams & Wilkins, 2010; 107:1032–1040. Seidman JG, Seidman C: Transcription factor haploinsufficiency: when half a loaf is not enough. J Clin Invest American Society for Clinical Investigation, 2002; 109:451–455. Brewer A, Pizzey J: GATA factors in vertebrate heart development and disease. Expert Rev Mol Med 2006; 8:1–20.
Figure legends Figure 1: Pedigree of family AF-772. AF cases are depicted in black, ASD cases are depicted in blue, VSD cases are depicted in red and individuals of unknown status are depicted in white. The presence of the R585L variant is indicated with a ‘+’ sign while ‘-‘ depicts genotype negative patients. Abbreviations, ASD, atrial septal defect; AF, atrial fibrillation; VSD, ventricular septal defect.
Figure 2: Location and conservation of GATA6 variants. A: Schematic representation of GATA4 and GATA6 proteins. The relative locations of GATA6 mutations are displayed as stars. B: Upper panel displays chromatogram traces generated from Sanger sequencing of PCR amplicons generated from each individual. Nucleotide changes and associated amino acid substitutions are detailed under the chromatogram. Lower panel illustrates the conservation of the relevant GATA6 peptide regions. Red outlines indicate the amino acid substituted by the observed mutation.
17
Figure 3: Analyses of GATA6 variant function by luciferase reporter assays. A: Relative luciferase activity recorded from CHO cells following cotransfection of reporter constructs with GATA6, GATA4, TBX5, MEF2c, NKX2.5, or empty vector. B: Relative luciferase activity recorded from CHO cells following transfection with reporter constructs and GATA6 variants. * p<0.05 when compared to wild-type GATA6. All data represent the averages of three independent biological replicates.
Figure 4: Functional analysis of potential synergistic actions of GATA6 variants with TBX5. Relative luciferase activity recorded from CHO cells following cotransfection of reporter constructs, GATA6 variants and either NKX2.5, TBX5, GATA4, or MEF2c. Color of the bar indicates the GATA6 variant as detailed by the key. * P value <0.05 when compared to wildtype GATA6.
18
Tables Table 1. Clinical Characteristics of AF-772 Family Members
ID
Sex
Age
Phenotype
PR
QRS
QT
1 2 21 22 3 4 43 5 6 61 62 7 71 72 73 8 9
F F M M F M F F M M F F M M M M M
67 66 33 31 65 64 27 62 60 28 25 58 28 25 22 55 52
AF VSD Abnormal ECG* ASD AF AF -
122 162 122 144 130 130 148 154 144 154 146 152 148 168
68 72 92 116 84 82 86 88 104 76 90 104 94 98
436 421 416 400 426 444 447 422 434 394 437 405 401 429
LVEF (%) 66 72 60 77 73 66 78 64 68 55 55 49 69
LA size (mm) 31 37 37 33 31 34 37 37 30 53 40 33 36
LVID (mm) 36 42 48 40 46 45 44 50 43 43 54 53 50
RV base (mm) 38 29 27 27 33 32 28 37 31 43
R585L variant N Y Y Y Y N N N N N N Y Y Y N Y
AF, atrial fibrillation; HR, heart rate; LVEF, left ventricular ejection fraction; LA, left atrium; RV, right ventricular *Abnormal ECG with P wave inversion in lateral leads and right ventricular conduction delay
19
Figure 1
20
Figure 2
21
Figure 3
22
Figure 4
23
PII: DOI: Reference:
www.elsevier.com/locate/buildenv
S1547-5271(16)30915-8 http://dx.doi.org/10.1016/j.hrthm.2016.10.014 HRTHM6894
To appear in: Heart Rhythm Cite this article as: Nathan R. Tucker, Saagar Mahida, Jiangchuan Ye, Elizabeth J. Abraham, Julie A. Mina, Victoria A. Parsons, Michael A. McLellan, Marisa A. Shea, Alan Hanley, Emelia J. Benjamin, David J. Milan, Honghuang Lin and Patrick T. Ellinor, Gain of function mutations in GATA6 lead to atrial fibrillation, Heart Rhythm, http://dx.doi.org/10.1016/j.hrthm.2016.10.014 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Gain of Function Mutations in GATA6 Lead to Atrial Fibrillation
Nathan R. Tucker,1¶ PhD, Saagar Mahida,1¶ MB, ChB, Jiangchuan Ye,1 MD, PhD, Elizabeth J. Abraham,1 BS, Julie A. Mina,1 BS, Victoria A. Parsons,1 BS, Michael A. McLellan,1 BS, Marisa A. Shea,2 BS, RN, Alan Hanley,1 MD, Emelia J. Benjamin,3-5 MD, MSc, David J. Milan,1,2 MD, Honghuang Lin,6PhD, and Patrick T. Ellinor,1,2* MD, PhD
1 Cardiovascular Research Center, Massachusetts General Hospital, Charlestown, MA, USA 2 Cardiac Arrhythmia Service, Massachusetts General Hospital, Boston, USA 3 National Heart, Lung, and Blood Institute's and Boston University's Framingham Heart Study, Framingham, Massachusetts, USA. 4 Preventive Medicine and Cardiovascular Medicine Sections, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts, USA. 5 Cardiology Division, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts, USA. 6 Computational Biomedicine Section, Department of Medicine, Boston University School of Medicine, Boston, Massachusetts, USA.
* Corresponding Author Patrick T. Ellinor, MD, PhD Cardiovascular Research Center Massachusetts General Hospital 55 Fruit Street, Gray 109 Boston, Massachusetts, 02114 Email: [email protected]
¶ These authors contributed equally to this work
Running title: GATA6 mutations in AF Conflict statement: Dr. Ellinor is a Principal Investigator on a grant from Bayer Healthcare to the Broad Institute.
1
Abstract Background: The genetic basis of atrial fibrillation (AF) and congenital heart disease remains incompletely understood. Objective: We sought to determine the causative mutation in a family with AF, atrial septal and ventricular septal defects. Methods: We evaluated a pedigree with 16 family members, one with an atrial septal defect, one with a ventricular septal defect and three with AF; we performed whole exome sequencing in three affected family members. Given that early-onset AF was prominent in the family, we then screened individuals with early-onset AF, defined as an age of onset < 66 years, for mutations in GATA6. Variants were functionally characterized using reporter assays in a mammalian cell line. Results: Exome sequencing in three affected individuals identified a conserved mutation, R585L, in the transcription factor gene, GATA6. In the MGH AF Study the mean age of AF onset was 47.1 ± 10.9 years, 79% of the participants were male, and there was no evidence of structural heart disease. We identified three GATA6 variants (P91S, A177T, and A543G). Using wild-type and mutant GATA6 constructs driving NPPA promoter reporter, we found that three of the four variants had a marked upregulation of luciferase activity (R585L; 4.1 fold, p<0.0001; P91S; 2.5 fold, p=0.0002; A177T; 1.7 fold, p=0.03). Additionally, when co-overexpressed with GATA4 and MEF2C, GATA6 variants exhibited upregulation of the αMHC and NPPA activity. Conclusion: Overall, we found gain-of-function mutations in GATA6 in both a family with earlyonset AF and atrioventricular septal defects as well as in sporadic, early-onset AF.
Keywords: Atrial fibrillation, mutation, transcription factor, genetics
Introduction Atrial fibrillation (AF) is the most common cardiac arrhythmia and over the coming decades, the prevalence is projected to rise.1 Despite the widespread prevalence and consequences of AF, the molecular mechanisms underlying the arrhythmia remain incompletely understood. Thus, the identification of the genetic basis of AF is an important public health concern.
2
AF is heritable with a two-fold increased risk of the arrhythmia if one parent has AF.2–5 Over the past two decades, classical genetic studies in familial forms of AF have led to the identification of a series mutations in ion channel and non-ion channel genes.6 Among these is GATA6, a member of a family of six transcription factors that bind to a conserved consensus sequence.7 Loss-of-function variants in GATA6 have been previously reported to cause atrial fibrillation8–10 as well as congenital heart defects such as Tetralogy of Fallot and atrial septal defects (ASD).11,12
In the present study, we describe the characterization of a family with a complex phenotype of early-onset AF, ASD and ventricular septal defects (VSD) due to a novel gain-of-function variant in GATA6. Further, we present the results of candidate-gene screening for GATA6 in a large cohort with early-onset AF and the functional evaluation of the identified variants.
Methods Clinical evaluation of Family AF-772 and the MGH AF Study Sixteen members of the family underwent a standardized interview, physical examination, 12lead electrocardiogram and echocardiography. Affected individuals were identified as those with either AF on the electrocardiogram or patients with a documented ASD or VSD during echocardiographic analysis. All patients in the MGH AF Study13 developed AF at an early age (AF onset less than 66 years) and had structurally normal hearts. The mean age of onset in AF cases was 47.1 ± 10.9
3
years. Consistent with previous reports, the AF cases were predominantly male with a male to female ratio of approximately 4 to 1. 45% of AF cases had a family history of AF. Baseline characteristics of the AF cohort are presented in Supplemental Table 1. Written informed consent was obtained from all patients prior to performing the evaluations. The study was approved by the Institutional Review Board at Massachusetts General Hospital and complied with the Declaration of Helsinki.
Exome sequencing DNA was isolated from venous blood samples using the Puregene DNA Purification Kit (Gentra Systems, Inc.). Whole-exome sequencing was performed in three affected members of the family (22, 72 and 73). The exome regions were captured by the Agilent Sure Select Target Enrichment Assay (version 3). The enriched fragments were then pair-end sequenced by the Illumina HiSeq 2000 platform at Perkin Elmer.
Bioinformatic analyses BWA was used to align the short reads to the reference human genome (NCBI Build 37, hg19), which were then converted into BAM files by SAMtools. We also performed local realignment to minimize mismatching of bases across all the reads, and base quality recalibration to correct base qualities using the GATK software package. Duplicate reads were marked by Mark Duplicates using the Picard software package. Only the first read pair in each duplicate cluster was kept for downstream analysis. Single nucleotide variations (SNVs) were then called by the
4
Unified Genotype function using the GATK software package. SNVs with quality scores lower than 20 or strand bias higher than 60 were excluded.
Filtration and prioritization of the variants was performed according to previously published methods.14 First, since it is unlikely that common polymorphisms in the general population underlie the phenotype, all common variants (>0.01%) identified in publically available databases (dbSNP,15 1000 Genomes Project16 and Exome Variant Server) were removed from further analyses. Second, it is likely that Mendelian diseases are caused by variants that result amino acid substitution, therefore synonymous variants were excluded. Analysis of expression levels was performed using the GNF Expression Atlas 2 data from U133A and GNF1H chips in UCSC browser.
Functional evaluation of HHATL activity The Sonic hedgehog (SHH) gene was received from plasmID at Harvard Medical School whereas HHATL and mutant K484N HHATL constructs were synthesized by Gen9 Bio (Cambridge, MA). COS7 cells were transfected using Lipofectamine LTX with Plus reagent (ThermoFisher). Western blotting was performed according to standard protocols. The antibodies used were anti-SHH N-terminal fragment (5E1; Developmental Studies Hybridoma Bank); anti-SHH (H-160; Santa Cruz Biotechnology Inc); anti-EGFP (JL-8; Fisher Scientific). Activity of wild-type and mutant HHATL was determined by evaluating the ratio of palmitoylated SHH to total SHH using the 5E1 and H160 antibodies, respectively, in three biological replicates.
5
Screening for GATA6 variants in the MGH AF Study A combination of high resolution melting and Sanger sequencing was used to perform candidate-gene screening. Exons 3, 5, and 6 of the GATA6 gene were screened using high resolution melting whereas the remaining coding exons (2, 4, and 7) were sequenced directly using Sanger sequencing. Light Scanner technology (Idaho Technologies Inc) was used to perform high resolution melting.
Luciferase assays The functional effect of the variant was analyzed by performing luciferase assays using coexpression of GATA6 variants in pCMV6 XL6 with three luciferase reporter constructs, NPPApromoter-luc (0.9kb), NPPB-promoter-luc (1.3kb), and αMHC-promoter-luc (0.7kb) in pGL4.10. NPPA, NPPB and αMHC are downstream target genes for GATA transcription factors and harbor GATA dependent promoters.17 CHO cells were transfected with reporter constructs and pRLSV40 according to manufacturer specifications using Lipofectamine LTX. Luciferase activities were measured at 24 hours using the dual luciferase reporter assay system (Promega). All experiments were performed in three independent biological replicates. For statistical analyses, two-way ANOVA was performed followed by multiple comparisons with Bonferroni correction.
6
Results Clinical characteristics of Family AF-772 We evaluated a total of 16 individuals from pedigree AF-772, a two-generation family with a complex phenotype of AF, ASD and VSD (Figure 1 and Table 1). The trait was inherited with an autosomal dominant pattern with five members of the pedigree being affected. Individual 7, the proband in the family, initially presented with shortness of breath, palpitations and two episodes of near syncope at age 51. Subsequent investigations revealed the presence of an ASD. She underwent surgery to correct an ostium primum ASD. Two of her descendants, 71 and 72, were diagnosed with AF at the ages of 14 and 16 years respectively. The proband´s sister, 2, had a single documented paroxysm of AF at age 49. The proband´s nephew, 22, had a VSD identified as an infant; there was no history of surgical intervention for the VSD. None of the other members of the family had a history of ASD, VSD or AF.
Exome sequencing to identify the causative variant Exome sequencing was performed in individuals 22, 72 and 73 using the Agilent Sure Select Target Enrichment Assay and the Illumina HiSeq 2000 platform. On average, 95.7% of short reads were mapped, reaching an average ~58-fold depth of coverage. 88.5% of targeted regions were covered by at least 10 unique reads (Supplemental Table 2).
An average of 18,825 variants was identified in the three affected individuals who underwent exome sequencing (Supplemental Table 3). The first stage of filtering of candidate variants
7
involved exclusion of synonymous variants and common variants reported in population databases (dbSNP,15 1000 Genomes Project,16 Exome Variant Server, and the EXac database18). Common variants were defined as variants with a minor allele frequency (MAF) of >5%. Rare variants were defined as variants with a MAF of 0.5-5%. Very rare variants were defined as variants with a MAF of. ≤0.5%. The number of novel or very rare missense variants (defined as singleton variants in public databases) for each individual varied between 385 and 548, nonsense variants varied between 7 and 17, and splice site variants varied between 0 and 7 (Supplemental Table 3). Subsequent filtering involved exclusion of variants that did not segregate with disease in individuals 22, 72 and 73. Using this strategy, the number of candidate variants was reduced to 13 novel or very rare missense variants, which were located in 11 genes (Supplemental Table 4). As individuals 72 and 73 developed AF at a young age, we analyzed the subset of variants shared by these individuals to identify potential variants that may modulate the effects of the disease-causing variant. We did not identify any compelling variants (Supplemental Table 5).
Prioritization of the 13 candidate variants was based on gene expression profiles, interspecies conservation, predicted effect on protein function, and involvement in cardiacrelated functional pathways. Expression data was available for 9 out of 11 candidate genes. Of these, four genes, (GATA6, TTN, HHATL and CCDC69) displayed high cardiac expression while the remaining 5 genes had either low or intermediate expression. Ten variants were located in residues that are conserved through evolution. Of note, one of the variants (R585L) was located in GATA6 with the highest Phylop conservation score of 4.94. (Supplemental Table 4).
8
After overall evaluation of these variant prioritization schemes, more comprehensive segregation analysis was performed for 5 of the 13 aforementioned candidate variants which exhibited cardiac expression, had a positive PhyloP conservation score, and had plausible biological links to cardiac disease following examination of literature. Segregation analysis was performed by direct Sanger sequencing of all 16 members of the family. The results from segregation analysis are included in Supplemental Table 6.
Among variants that underwent segregation analysis, R585L-GATA6 and K484N-HHATL represented the most compelling candidate variants for AF, ASD and VSD in pedigree AF-772. As HHATL is known to act as an antagonist to the HHAT-mediated addition of palmitate to the N-terminus of sonic hedgehog (SHH),19 we analyzed the HHATL variant function in vitro. We found no difference in the activity of mutant HHATL to inhibit the N-palmitoylation of SHH (Supplemental Figure 1). In contrast, a number of functional studies provide further evidence to suggest that the R585-GATA6 variant is instead the causative variant. GATA6, encoding a transcription factor that plays a critical role during cardiac development and morphogenesis, has previously been implicated in congenital heart defects, including ASD.
Five additional GATA6-R585L variant carriers, individuals 2, 3, 7, 9, and 21 were identified (Figure 1). Individual 7 had an ASD while individual 2 had paroxysmal AF. While individuals 3, 9, and 21 did not display any overt structural cardiac abnormalities or arrhythmias, individual 3 did demonstrate an abnormal P wave axis with P wave inversion in the lateral leads (I and AVL)
9
and evidence of right ventricular conduction delay (Supplemental Figure 2). Individuals 9 and 21 had normal baseline ECGs. There were no significant differences between ECG parameters (PR, QRS and QT interval) between mutation carriers and patients without the mutation (Table 1). We were unable to perform echocardiography with bubble contrast or Holter monitoring to exclude subtle septal defects and asymptomatic episodes of AF in individuals 3, 9 and 21. In the absence of this, the findings cannot be definitively attributed to incomplete penetrance.
Identification of GATA6 variants in a study with early-onset AF Given our findings in AF-772, we screened for GATA6 variants in 546 patients with AF from the MGH AF Study.13 We identified novel or very rare GATA6 variants in three out of the 546 earlyonset AF patients (Supplemental Table 7). One of the variants, A543G, was located in the Cterminus, while the other two variants, P91S and A177T, were located at the N-terminus of the peptide (Figure 2). None of these variants have previously been reported in the Exome Variant Server which contains sequence data from more than 6,500 individuals; however, the P91S variant was reported in a single individual among the ~2500 individuals sequenced in the 1000 Genomes Project.16 Similarly to R585L, the A177T variant was strongly conserved through evolution as assessed by the Phylop and GERP scores whereas the P91S and A543G residues were subject to more variability in mammals. The characteristics of the variants are summarized in Supplemental Table 8.
10
Functional evaluation of the GATA6 variants To identify the potential effects of the GATA6 variants on protein function, we evaluated the ability of exogenously expressed wild-type and mutant GATA6 to activate known GATA familytargeted promoters upstream of a luciferase reporter. Overexpression of wild-type GATA6 upregulated expression of luciferase in the promoters of the NPPA, NPPB, and αMHC genes, thus confirming GATA6 mediated activation of these promoters (Figure 3A). The R585L variant identified in family AF-772 exhibited a significant increase in luciferase activity with NPPA (4.1 fold, p<0.0001) and αMHC (1.7 fold, p=0.03) reporter constructs (Figure 3B); a slight, but nonsignificant, increase was seen with the NPPB construct.
The P91S variant resulted in an upregulation in luciferase activity with NPPA (2.5 fold, p=0.0002) and αMHC (1.7 fold, p=0.03) constructs, but not with an NPPB construct. The A177T variant resulted in upregulation of luciferase activity with NPPA (1.7 fold, p=0.03) but had no effect with αMHC and NPPB constructs. Finally, the A543G mutant displayed no net effect with any of the constructs compared to control. Of note, in comparison to the R585L variant, upregulation of luciferase activity associated with the P91S and A177T variants was less pronounced and more variable. This differential activity is not due to altered expression or localization, which is unaltered by inclusion of the variants (Supplemental Figure 3).
GATA6 has been reported to interact with a range of other transcription factors and co-factor proteins including TBX5, GATA4, NKX2.5, and MEF2C.17,22–25 To investigate synergistic effects with the GATA6 mutants, we co-overexpressed GATA6 constructs with TBX5, GATA4, NKX2-5, or
11
MEF2C. Of the four constructs, only NKX2.5 upregulated the expression of all three promoters independently of GATA6. Importantly, when overexpressed with GATA4 or MEF2C, all variants exhibited statistically significant upregulation of the NPPA and αMHC promoters as compared to control (Figure 4). Additionally, for all variants, co-expression with TBX5 demonstrated a significant upregulation of the αMHC promoter.
Discussion We performed whole exome sequencing in a large pedigree with a complex phenotype of earlyonset AF, ASD and VSD and identified a highly conserved mutation in the transcription factor gene, GATA6. We subsequently performed candidate-gene screening in a large cohort of patients with early-onset AF and identified three further GATA6 variants. All four variants exhibit a gain-of-function effect.
Loss-of-function mutations in GATA6 have been reported to underlie structural congenital cardiac defects such as ASD, VSD, Tetralogy of Fallot, and persistent truncus arteriosus.11,12,26 Interestingly, a gain-of-function GATA6 mutation has also been reported to cause Tetralogy of Fallot.11 In addition to the reports linking GATA6 mutations with structural cardiac defects, candidate-gene studies have identified four variants in the GATA6 gene (G469V, Y235S, Q206P, and Y265X) in patients with lone AF.8–10 Two of the variants (G469V and Y235S) have been defined as loss-of-function.
12
The findings of the present study further expand on the potential role of GATA6 variants in the pathogenesis of AF. While the frequency with which GATA6 variation contributes to AF at the population level remains unclear, the present study, when combined with previous work, suggest that GATA6 is an important target that should continue to be examined by whole exome and whole genome sequencing as larger cohorts are evaluated.
The specific variants found in the present study provide insight into the function of N- and Cterminal domains of the GATA6 protein. The R585L and A543G variants are located in highly conserved regions of the C-terminus of the peptide. Although the specific function of the Cterminus in the GATA transcription factors has remained unclear, truncation mutants of the GATA4 C-terminus resulted in an 80% reduction in transcriptional activity.27 The role of the Nterminal region which contains the P91S and A177T variants is similarly unclear. Previous deletion studies of the GATA family members suggest that this region contains critical domains for transcriptional activation of GATA4, GATA5 and GATA6, perhaps through transactivation domain activity.27 These findings are consistent with our results that indicate a critical role for the N- and C-terminal residues in the regulation of GATA6 transcriptional activation.
Of note, the R585L identified in Family AF-772 and the P91S and A543G variants found in individuals with early-onset AF, were identified in rare individuals in the 1000 Genomes Project and the ExAC database. Since the Exome Variant Server, the 1000 Genomes Projects, and the ExAC database contain data on ~6500, ~2500 and ~61000 individuals respectively, the estimated frequency would be 1 in ~18,000 alleles. The transmission of R585L in the family as
13
well as the clear gain-of-function effect strongly support R585L in the pathogenesis of the phenotype. However, our study highlights the challenges in using SNP repository data as a “control” population in the absence of clinical or functional data.
From a mechanistic perspective, GATA6 variants could influence susceptibility to AF in a number of ways. GATA6 is responsible for the maintenance of cardiac precursor cells in an undifferentiated proliferative state during cardiac development.28 GATA6 variants could cause abnormal differentiation of cardiomyocytes, a process which has been linked to an susceptibility to triggered arrhythmias.29 In addition to its role in cardiac development, GATA6 also regulates genes in the adult heart.30 Therefore, it is plausible that GATA6 variants cause altered expression of genes that maintain electrical stability in the atrium.
The GATA6 R585L variant is associated with disparate clinical phenotypes in pedigree 772. These observations include that of an ostium primum ASD, which is contrasted by the ostium secundum ASD in a previous report. It is notable that the previous report identified a loss-offunction variant, so it is possible that the direction of effect may impact the clinical phenotype. The finding of diverse clinical presentations for GATA6 variants are consistent with previous reports indicating that single transcription factor mutations result in multiple different phenotypes.31 During cardiac development, GATA transcription factors play a role in orchestrating complex regulatory networks involving multiple genes.32 It is plausible that the variable clinical phenotypes associated with GATA6 R585L are mediated by unidentified additional variants in genes involved in these regulatory pathways. At this stage, as proposed
14
mechanisms remain speculative, work in model systems would be important for better understanding pathways that may underlie the pathogenesis observed in this and other studies.
Conclusion In sum, we have found that gain of GATA6 function influences susceptibility to phenotypes ranging from early-onset AF to atrial and ventricular septal defects. Our findings further expand on the role of transcription factors in the pathogenesis of congenital heart disease and AF.
References 1. 2.
3. 4. 5.
6. 7.
8. 9. 10. 11.
Kannel WB, Benjamin EJ: Status of the epidemiology of atrial fibrillation. Med Clin North Am NIH Public Access, 2008; 92:17–40, ix. Fox CS, Parise H, D’Agostino RB, Lloyd-Jones DM, Vasan RS, Wang TJ, Levy D, Wolf PA, Benjamin EJ: Parental atrial fibrillation as a risk factor for atrial fibrillation in offspring. JAMA American Medical Association, 2004; 291:2851–2855. Arnar DO, Thorvaldsson S, Manolio TA, Thorgeirsson G, Kristjansson K, Hakonarson H, Stefansson K: Familial aggregation of atrial fibrillation in Iceland. Eur Heart J 2006; 27:708–712. Ellinor PT, Yoerger DM, Ruskin JN, MacRae CA: Familial aggregation in lone atrial fibrillation. Hum Genet 2005; 118:179–184. Darbar D, Herron KJ, Ballew JD, Jahangir A, Gersh BJ, Shen W-K, Hammill SC, Packer DL, Olson TM: Familial atrial fibrillation is a genetically heterogeneous disorder. J Am Coll Cardiol 2003; 41:2185–2192. Mahida S, Lubitz SA, Rienstra M, Milan DJ, Ellinor PT: Monogenic atrial fibrillation as pathophysiological paradigms. Cardiovasc Res 2011; 89:692–700. Molkentin JD: The Zinc Finger-containing Transcription Factors GATA-4, -5, and -6: UBIQUITOUSLY EXPRESSED REGULATORS OF TISSUE-SPECIFIC GENE EXPRESSION. J Biol Chem American Society for Biochemistry and Molecular Biology, 2000; 275:38949–38952. Yang Y-Q, Wang X-H, Tan H-W, Jiang W-F, Fang W-Y, Liu X: Prevalence and spectrum of GATA6 mutations associated with familial atrial fibrillation. Int J Cardiol 2012; 155:494–496. Yang Y-Q, Li L, Wang J, Zhang X-L, Li R-G, Xu Y-J, Tan H-W, Wang X-H, Jiang J-Q, Fang W-Y, Liu X: GATA6 loss-of-function mutation in atrial fibrillation. Eur J Med Genet 2012; 55:520–526. Li J, Liu W-D, Yang Z-L, Yang Y-Q: Novel GATA6 loss-of-function mutation responsible for familial atrial fibrillation. Int J Mol Med 2012; 30:783–790. Maitra M, Koenig SN, Srivastava D, Garg V: Identification of GATA6 sequence variants in patients with congenital heart defects. Pediatr Res 2010; 68:281–285.
15
12.
13. 14. 15. 16.
17.
18. 19.
20.
21.
22.
23.
24.
25. 26.
27.
28.
29.
Kodo K, Nishizawa T, Furutani M, Arai S, Yamamura E, Joo K, Takahashi T, Matsuoka R, Yamagishi H: GATA6 mutations cause human cardiac outflow tract defects by disrupting semaphorin-plexin signaling. Proc Natl Acad Sci U S A 2009; 106:13933–13938. Ellinor PT, Lunetta KL, Albert CM, et al.: Meta-analysis identifies six new susceptibility loci for atrial fibrillation. Nat Genet NIH Public Access, 2012; 44:670–675. Li M-X, Gui H-S, Kwan JSH, Bao S-Y, Sham PC: A comprehensive framework for prioritizing variants in exome sequencing studies of Mendelian diseases. Nucleic Acids Res 2012; 40:e53. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM, Sirotkin K: dbSNP: the NCBI database of genetic variation. Nucleic Acids Res Oxford University Press, 2001; 29:308–311. Abecasis GR, Altshuler D, Auton A, Brooks LD, Durbin RM, Gibbs RA, Hurles ME, McVean GA: A map of human genome variation from population-scale sequencing. Nature 2010; 467:1061– 1073. Charron F, Paradis P, Bronchain O, Nemer G, Nemer M: Cooperative interaction between GATA-4 and GATA-6 regulates myocardial gene expression. Mol Cell Biol American Society for Microbiology (ASM), 1999; 19:4355–4365. Lek M, Karczewski KJ, Minikel E V., et al.: Analysis of protein-coding genetic variation in 60,706 humans. Nature Nature Research, 2016; 536:285–291. Abe Y, Kita Y, Niikura T: Mammalian Gup1, a homolog of Saccharomyces cerevisiae glycerol uptake/transporter 1, acts as a negative regulator for N-terminal palmitoylation of Sonic hedgehog. FEBS J 2008; 275:318–331. De Franco E, Shaw-Smith C, Flanagan SE, Shepherd MH, International NDM Consortium, Hattersley AT, Ellard S: GATA6 mutations cause a broad phenotypic spectrum of diabetes from pancreatic agenesis to adult-onset diabetes without exocrine insufficiency. Diabetes 2013; 62:993–997. Lango Allen H, Flanagan SE, Shaw-Smith C, De Franco E, Akerman I, Caswell R, International Pancreatic Agenesis Consortium, Ferrer J, Hattersley AT, Ellard S: GATA6 haploinsufficiency causes pancreatic agenesis in humans. Nat Genet 2012; 44:20–22. Davis DL, Edwards A V, Juraszek AL, Phelps A, Wessels A, Burch JB: A GATA-6 gene heart-regionspecific enhancer provides a novel means to mark and probe a discrete component of the mouse cardiac conduction system. Mech Dev 2001; 108:105–119. Maitra M, Schluterman MK, Nichols HA, Richardson JA, Lo CW, Srivastava D, Garg V: Interaction of Gata4 and Gata6 with Tbx5 is critical for normal cardiac development. Dev Biol 2009; 326:368– 377. Molkentin JD, Antos C, Mercer B, Taigen T, Miano JM, Olson EN: Direct activation of a GATA6 cardiac enhancer by Nkx2.5: evidence for a reinforcing regulatory network of Nkx2.5 and GATA transcription factors in the developing heart. Dev Biol 2000; 217:301–309. Morin S, Charron F, Robitaille L, Nemer M: GATA-dependent recruitment of MEF2 proteins to target promoters. EMBO J European Molecular Biology Organization, 2000; 19:2046–2055. Wang J, Luo X-J, Xin Y-F, Liu Y, Liu Z-M, Wang Q, Li R-G, Fang W-Y, Wang X-Z, Yang Y-Q: Novel GATA6 mutations associated with congenital ventricular septal defect or tetralogy of fallot. DNA Cell Biol Mary Ann Liebert, Inc., 2012; 31:1610–1617. Morrisey EE, Ip HS, Tang Z, Parmacek MS: GATA-4 Activates Transcription Via Two Novel Domains That Are Conserved within the GATA-4/5/6 Subfamily. J Biol Chem American Society for Biochemistry and Molecular Biology, 1997; 272:8515–8524. Gove C, Walmsley M, Nijjar S, Bertwistle D, Guille M, Partington G, Bomford A, Patient R: Overexpression of GATA-6 in Xenopus embryos blocks differentiation of heart precursors. EMBO J European Molecular Biology Organization, 1997; 16:355–368. Zhang YM, Hartzell C, Narlow M, Dudley SC: Stem cell-derived cardiomyocytes demonstrate
16
30.
31. 32.
arrhythmic potential. Circulation 2002; 106:1294–1299. van Berlo JH, Elrod JW, van den Hoogenhof MMG, York AJ, Aronow BJ, Duncan SA, Molkentin JD: The Transcription Factor GATA-6 Regulates Pathological Cardiac Hypertrophy. Circ Res Lippincott Williams & Wilkins, 2010; 107:1032–1040. Seidman JG, Seidman C: Transcription factor haploinsufficiency: when half a loaf is not enough. J Clin Invest American Society for Clinical Investigation, 2002; 109:451–455. Brewer A, Pizzey J: GATA factors in vertebrate heart development and disease. Expert Rev Mol Med 2006; 8:1–20.
Figure legends Figure 1: Pedigree of family AF-772. AF cases are depicted in black, ASD cases are depicted in blue, VSD cases are depicted in red and individuals of unknown status are depicted in white. The presence of the R585L variant is indicated with a ‘+’ sign while ‘-‘ depicts genotype negative patients. Abbreviations, ASD, atrial septal defect; AF, atrial fibrillation; VSD, ventricular septal defect.
Figure 2: Location and conservation of GATA6 variants. A: Schematic representation of GATA4 and GATA6 proteins. The relative locations of GATA6 mutations are displayed as stars. B: Upper panel displays chromatogram traces generated from Sanger sequencing of PCR amplicons generated from each individual. Nucleotide changes and associated amino acid substitutions are detailed under the chromatogram. Lower panel illustrates the conservation of the relevant GATA6 peptide regions. Red outlines indicate the amino acid substituted by the observed mutation.
17
Figure 3: Analyses of GATA6 variant function by luciferase reporter assays. A: Relative luciferase activity recorded from CHO cells following cotransfection of reporter constructs with GATA6, GATA4, TBX5, MEF2c, NKX2.5, or empty vector. B: Relative luciferase activity recorded from CHO cells following transfection with reporter constructs and GATA6 variants. * p<0.05 when compared to wild-type GATA6. All data represent the averages of three independent biological replicates.
Figure 4: Functional analysis of potential synergistic actions of GATA6 variants with TBX5. Relative luciferase activity recorded from CHO cells following cotransfection of reporter constructs, GATA6 variants and either NKX2.5, TBX5, GATA4, or MEF2c. Color of the bar indicates the GATA6 variant as detailed by the key. * P value <0.05 when compared to wildtype GATA6.
18
Tables Table 1. Clinical Characteristics of AF-772 Family Members
ID
Sex
Age
Phenotype
PR
QRS
QT
1 2 21 22 3 4 43 5 6 61 62 7 71 72 73 8 9
F F M M F M F F M M F F M M M M M
67 66 33 31 65 64 27 62 60 28 25 58 28 25 22 55 52
AF VSD Abnormal ECG* ASD AF AF -
122 162 122 144 130 130 148 154 144 154 146 152 148 168
68 72 92 116 84 82 86 88 104 76 90 104 94 98
436 421 416 400 426 444 447 422 434 394 437 405 401 429
LVEF (%) 66 72 60 77 73 66 78 64 68 55 55 49 69
LA size (mm) 31 37 37 33 31 34 37 37 30 53 40 33 36
LVID (mm) 36 42 48 40 46 45 44 50 43 43 54 53 50
RV base (mm) 38 29 27 27 33 32 28 37 31 43
R585L variant N Y Y Y Y N N N N N N Y Y Y N Y
AF, atrial fibrillation; HR, heart rate; LVEF, left ventricular ejection fraction; LA, left atrium; RV, right ventricular *Abnormal ECG with P wave inversion in lateral leads and right ventricular conduction delay
19
Figure 1
20
Figure 2
21
Figure 3
22
Figure 4
23