β2-Adrenergic Receptor Polymorphisms and Sudden Cardiac Death
A Signal to Follow
Sudden cardiac death (SCD) is defined as an unexpected death from a cardiac cause generally within 1 hour of the onset of symptoms. The great majority of SCD cases are due to ventricular fibrillation and subsequent hemodynamic collapse. Eighty percent of SCD is attributable to ischemic heart disease, with much of the remainder caused by cardiomyopathies/heart failure and intrinsic conduction system abnormalities.1 These predisposing conditions place patients at risk for SCD, and a trigger for the fatal arrhythmias may come from an imbalance of the parasympathetic and sympathetic nervous systems,2–5 the latter being mediated by cardiac β1- and β2-adrenergic receptors (β1AR, β2AR). Indeed, treatment with β-blockers of patients with heart failure and those who have suffered a myocardial infarction significantly reduces ventricular tachyarrhythmias and SCD. Typically, the pathophysiology of SCD is considered in terms of the underlying disease. However, there is evidence for genetic variability of the β1AR and β2AR genes that has functional consequence in transfected cells, endogenously expressing cells, and transgenic mice.6,7 Familial clustering of certain pathological arrhythmias and the unexplained variability in susceptibility among unrelated individuals to fatal arrhythmias raise the possibility of common polymorphisms such as those of the β1AR or β2AR being genetic risk factors for SCD.
Article p 1842
In this issue of Circulation, Sotoodehnia et al8 examined polymorphisms of the β2AR and potential associations with SCD. The study used individuals in the Cardiovascular Health Study (CHS), amounting to 4441 European Americans and 808 African Americans, with SCD occurring in 156 and 39 cases, respectively. In this population, homozygosity for Gln27 was associated with a hazard ratio (HR) of 1.56 (95% confidence interval [CI], 1.17 to 2.09) for SCD compared with those with 1 or 2 Glu27 alleles (Glu27 carriers). This association held in European Americans (HR, 1.62; 95% CI, 1.18 to 2.23) but not African Americans, for whom the HR was 1.23 but the confidence interval spanned unity (95% CI, 0.61 to 2.48). This may be due to the smaller sample size and the lower frequency of the Gln27 allele in African Americans and thus a lack of power. A replication study consisted of 155 SCD subjects from the Cardiac Arrest Blood Study (CABS) and 144 control subjects, all classified as European American. In that study, Gln27 homozygosity also was found to be associated with SCD, with an adjusted odds ratio of 1.83 (95% CI, 1.11 to 3.04). Of interest is whether Gln27 is uniquely associated with SCD or with ischemia or heart failure, which predisposes to SCD. In the present study,8 there was no apparent modification of the risk of SCD associated with Gln27 and a history of myocardial infarction, nor was there an association of Gln27 with nonfatal myocardial infarctions or non-SCD atherosclerotic deaths. This is somewhat puzzling in that this same group has recently reported an association between Gln27 and coronary events in the CHS cohort.9 Additionally, as introduced earlier, the most common underlying disease in SCD is ischemic heart disease. Thus, it remains unclear whether Gln27 is a unique, independent risk factor for SCD.
There was no association with the polymorphisms at amino acid 16. As discussed below, however, these polymorphisms of the β2AR occur in various combinations (haplotypes or, in the case of 2 variable positions, diplotypes), so an analysis of the common diplotypes (H1, H2 and H3) was performed. Because a diplotype can be homozygous or heterozygous, there are many possible diplotype pairs. However, because of linkage disequilibrium, some diplotypes are not observed, and in this case, there were 6 diplotype pairs. Any diplotype that had homozygous Gln27 was associated with SCD (in the CHS cohort). Several observations regarding the diplotype analysis (Figure 2 of Sotoodehnia et al8) are intriguing. First, the most predictive diplotype was homozygous H3, which differs from H2 by the presence of a Gly or Arg at amino acid 16 (Figure 1 of Sotoodehnia et al8). Although the study was underpowered to fully address this, it may well be that the specific diplotype Gly16/Gln27 is a more robust indicator of SCD risk than the others, indicating a role for the position 16 residue.
The plot is actually much thicker, though. The β2AR gene is intronless and located on chromosome 5q31-q32. Early studies of the coding region identified several nonsynonymous single nucleotide polymorphisms (SNPs) altering the encoded amino acid compared with a reference sequence. Interestingly, this reference sequence (the reported sequence from the first cloning of the gene) was not the most common. Nevertheless, it often has been called the “wild-type” sequence and the variations from it have been called the polymorphic sequence. To eliminate any confusion, the specific alleles (or combinations) are now used to indicate the receptor variant of interest. Two common SNPs occur at nucleotide 46 (codon 16), where the translated amino acids are either Arg or Gly, and at nucleotide 79 (codon 27), where Gln or Glu can be found. A rare variant at nucleotide 491 (codon 164) also has been identified, where Thr is most common and Ile is the minor allele. Early studies with cells transfected with cDNAs that mimicked these SNPs revealed that the Ile164 receptor is markedly defective in signaling to Gs/adenylyl cyclase.10 In contrast, the more common variants all had similar signaling to Gs/adenylyl cyclase. However, agonist-promoted downregulation of the Gly16/Glu27 receptor was enhanced ≈50% compared with Arg16/Glu27 when cells expressing equivalent levels of these receptors were treated for 24 hours with the agonist isoproterenol.11 The Gly16 phenotype was observed regardless of whether the position 27 amino acid was Glu or Gln. The mechanism for this altered regulation is not altogether clear but appears to occur after the receptor internalization process, at a point where the receptor undergoes protein degradation. These data indicated that the consequence of these variations was centered around long-term agonist-mediated downregulation and that, of the combinations, the position 16 variation seemed to play the dominant role. How, then, did Sotoodehnia et al8 find that variation at position 27 imparted risk for SCD?
The answer may lie outside the coding region. Resequencing efforts12 of the 5′ flanking region showed multiple SNPs in the promoter, 5′-leader cistron (also called the βAR upstream protein, a 19–amino acid protein that alters β2AR translation), the 5′ untranslated region, and the coding region up to nucleotide 523 (Table). When the genotypes were phased (ie, arranged on the basis of parental chromosomes), 12 haplotypes were identified. As expected, not all the possible combinations (8192) of these SNPs were found in the population because of the relatively young age of the human species and thus linkage disequilibrium between many SNP pairs. Examination of the haplotypes shows that there are 4 that have allele frequencies >5% in at least 1 of the 4 populations (Table, haplotypes 1, 2, 4 and 6). Comparing US whites and African Americans, we find that haplotype 1 is observed much more commonly in African Americans, whereas haplotype 2 is more common in whites. Haplotypes 4 and 6 are prevalent in both populations. When whole-gene transfections were carried out with haplotypes 2 and 4, steady-state β2AR protein and mRNA levels were higher for haplotype 2.12 Haplotype 2 corresponds to Gly16/Glu27, whereas haplotype 4 (in whites) is Arg16/Gln27. So, the receptor that appears to undergo enhanced downregulation (Gly16/Glu27) may have higher “baseline” expression levels compared with Arg16/Gln27. Thus, it is not clear whether a β2AR with Gln27 will impose enhanced cardiomyocyte signaling under the dynamic conditions of an intact human in which a genetically imposed higher expression could be offset by increased downregulation if catecholamines are elevated. A comprehensive study of potentially altered transcription factor binding sites, expression patterns, and agonist regulation of the full haplotypes in a cell type of interest (smooth muscle, cardiomyocyte, etc) has not been published.
Examination of the Table also reveals that the Gly16/Glu27 is unique to haplotype 2; ie, no additional SNP sites are required for genotyping to identify this haplotype. This is also true for Gly16/Gln27 and haplotype 4 in whites, except for a minor partitioning with the rare haplotypes 7, 10, and 11. However, in a mixed population of whites and African Americans, Arg16/Gln27 is found with significant frequency in haplotypes 1 and 4. These 2 haplotypes differ at SNP sites −1023 and −654 of the promoter. It would have been interesting if these SCD studies would have genotyped at additional positions so that haplotypes 1 and 4 could have been identified. This is especially so because haplotype 1 occurs in ≈25% of African Americans. Although the expression pattern of haplotype 1 is not known, it may result in an improved predictive power for one or more haplotypes. Of course, when there is grouping by more haplotypes, then the sample sizes in some groups will be smaller. Nevertheless, if the effect is strong, then significance will be apparent. One never knows unless the experiment is done. It is perfectly acceptable to start with a larger set of haplotypes/genotypes and coalesce the groups as the analysis moves forward. Nevertheless, the diplotype identified by Sotoodehnia et al8 that imparts a higher risk of SCD in whites is a high-expression haplotype. This is consistent with the notion that a genetic mechanism that increases overall β2AR responsiveness in the heart could predispose to fatal arrhythmias and SCD.
The reported increased risk of SCD with Gln27 is somewhat small, and this “signal” needs follow-up. Given that cardiac adrenergic drive is a composite of multiple receptors (and prereceptor and postreceptor elements), it is likely that there will be small contributions from several other genes. Of particular interest would be the other adrenergic receptor genes, many of which are highly polymorphic, and those of the parasympathetic nervous system.
This work was supported by National Institutes of Health grant HL077101 SCCOR in Heart Failure.
Dr Liggett serves as a consultant to and on the Advisory Board for CardioDx.
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association.
Myerburg RJ, Castellanos A. Cardiac arrest and sudden cardiac death. In: Zipes DP, Libby P, Bonow RO, Braunwald E, eds. Braunwald’s Heart Disease. 7th ed. Philadelphia, Pa: Elsevier Saunders; 2005: 873–875.
Shimizu W, Antzelevitch C. Cellular basis for the ECG features of the LQT1 form of the long-QT syndrome: effects of β-adrenergic agonists and antagonists and sodium channel blockers on transmural dispersion of repolarization and torsade de pointes. Circulation. 1998; 98: 2314–2322.
Tomaselli GF, Zipes DP. What causes sudden death in heart failure? Circ Res. 2004; 95: 754–763.
Sotoodehnia N, Siscovick DS, Vatta M, Psaty BM, Tracy RP, Towbin JA, Lemaitre RN, Rea TD, Durda JP, Chang JM, Lumley TS, Kuller LH, Burke GL, Heckbert SR. β2-Adrenergic receptor genetic variants and risk of sudden cardiac death. Circulation. 2006; 113: 1842–1848.
Heckbert SR, Hindorff LA, Edwards KL, Psaty BM, Lumley T, Siscovick DS, Tang Z, Durda JP, Kronmal RA, Tracy RP. β2-Adrenergic receptor polymorphisms and risk of incident cardiovascular events in the elderly. Circulation. 2003; 107: 2021–2024.
Green SA, Cole G, Jacinto M, Innis M, Liggett SB. A polymorphism of the human β2-adrenergic receptor within the fourth transmembrane domain alters ligand binding and functional properties of the receptor. J Biol Chem. 1993; 268: 23116–23121.
Drysdale CM, McGraw DW, Stack CB, Stephens JC, Judson RS, Nandabalan K, Arnold K, Ruano G, Liggett SB. Complex promoter and coding region β2-adrenergic receptor haplotypes alter receptor expression and predict in vivo responsiveness. Proc Natl Acad Sci U S A. 2000; 97: 10483–10488.