The β2-adrenoceptor gene and hypertension: is it the promoter or the coding region or neither?

Our genetic make-up influences how susceptible we are to specific diseases as well as to their treatment and associated adverse events. In the current issue of the journal, Wu et al. [1] report on an association between single nucleotide polymorphisms (SNPs) in the gene of the β2-adrenoceptor (ADRB2) and the presence of arterial hypertension, which involves SNPs both in the promoter and the coding region of the gene but is specific for an ethnic group [1]. Many previous studies have investigated possible associations between SNPs in the coding region of the ADRB2, particularly for Arg16Gly and Gln27Glu, but have yielded conflicting results (Tables 1 and 2). Association studies with the rare Thr164Ile SNP have also yielded conflicting results [2,3], and little is known regarding a possible role of Arg19Cys in hypertension. Here, we discuss how the study by Wu et al. [1] may help to better understand a possible contribution of polymorphisms in the ADRB2 gene to the pathophysiology of hypertension.Table 1: Association studies between the Arg16Gly single nucleotide polymorphism in ADRB2 and hypertensionTable 2: Association studies between the Gln27Glu single nucleotide polymorphism in ADRB2 and hypertensionWithin the ADRB2 gene, nine SNPs have been identified in the coding region, of which five are synonymous and hence unlikely to be functionally relevant; another eight SNPs have been reported in the promoter region [4]. A mechanistic understanding of genotype–phenotype associations requires knowledge of the functional consequences of a given polymorphism. The linkage disequilibrium (LD) between various SNPs in ADRB2 makes it difficult to dissect the specific role of any given SNP, but several studies have addressed the functional consequences of SNPs in the coding region of the ADRB2 at the cellular, tissue and in-vivo level [4–6]. One approach has been to generate ADRB2 variants by site-directed mutagenesis with subsequent expression in suitable cell lines. Such studies found that the Arg16Gly and the Gln27Glu polymorphisms do not affect ligand binding to the receptor or its ability to promote adenylyl cyclase stimulation. However, the Gly16 variant undergoes enhanced agonist-promoted down-regulation, even when present in a heterozygous form only, whereas the Glu27 variant is largely resistant to such down-regulation. Because the Gly16 variant dominates functionally when both SNPs are concomitantly present [7], the desensitization-resistant Glu27 variant is expected to be functionally relevant only when found in homozygous Arg16 carriers, but this haplotype is rare, as it the Thr164Ile SNP [4]. Therefore, the Arg16Gly SNP appears to be more important than those at amino acid positions 27 and 164 at the level of the overall population. If the Gly16 variant is more prone to desensitization, it could be expected that its expression level, and hence responsiveness, is lower under physiological circumstances due to the tonic exposure to endogenous catecholamines; enhanced sympathetic activity in hypertensive subjects may further enhance such down-regulation. Although some studies support this hypothesis, the overall available data are not fully conclusive [4]. Although ADRB2 are expressed in various cardiovascular tissues, including prejunctional sympathetic nerve endings, it is assumed that those in vascular smooth muscle cells mediating vasodilatation are the physiologically most important with regard to blood pressure regulation. Hence, it would be expected that hypertension, if anything, is positively associated with a hypofunctional variant of ADRB2. This consideration yields the working hypothesis that among ADRB2 SNPs the Gly16 variant is the mechanistically most promising candidate for associations with hypertension. Despite a wealth of studies in various ethnic groups (Tables 1 and 2), no clear picture has emerged regarding an association of SNPs in ADRB2 and hypertension. Thus, in Caucasian populations, the majority of studies, including those in the largest populations, did not detect an association between genotype at amino acid position 16 and hypertension. A few studies did detect significant associations; some of them reported the Gly allele to be associated, whereas others unexpectedly found the Arg allele to be associated (Table 1). All studies carried out in black populations, except one conducted in individuals from the Caribbean, reported the lack of an association (Table 1). By contrast, associations were found in four out of seven Oriental populations, most of them Chinese; although three of them imply that Gly16 is positively associated with hypertension, one of them proposes Arg16 to be positively associated with hypertension (Table 1). With regard to the SNP in amino acid position 27 of the ADRB2, the vast majority of studies in Caucasians or Blacks did not detect associations with hypertension (Table 2). Although three out of seven studies in Oriental populations detected significant associations, these were ascribed to both the Gln and the Glu allele (Table 2). Thus, a large number of studies not detecting associations of alleles at amino acid positions 16 or 27 with hypertension, together with the positive studies reporting associations at either position for both possible alleles, indicates that the ‘positive’ studies largely reflect random scatter around the null hypothesis. A possible exception are some Oriental populations but, even in those, a possible random scatter around the null hypothesis needs to be considered. Scatter around the null hypothesis is therefore the most likely explanation for the inconsistent picture emerging from studies on ADRB2 and hypertension. Although this emerging picture may be somewhat disappointing, it is in agreement with the conflicting data from association studies between SNPs in the coding region of the ADRB2 and diseases such as asthma, heart failure and obesity [4]. Moreover, older studies demonstrating impaired β2-adrenoceptor function in hypertension found that such impairments can be corrected (e.g. upon a low-sodium diet) [8]. Similarly, altered expression of β2-adrenoceptors in tissues of animal models of hypertension was found to be correlated with the extent of blood pressure elevation, rather than being specific for models of hereditary hypertension [9]. Therefore, these functional data argue against a strong genetic component in the regulation of β2-adrenoceptor responsiveness in hypertension. The study by Wu et al. [1] expands upon previous findings by demonstrating that SNPs not only in the coding region, but also in the promoter of ADRB2 can be associated with hypertension [1]. Previous studies have identified SNPs in the promoter region of the ADRB2 gene and reported that, specifically, a SNP at nucleotide position –47 is in LD with those in the coding region [10–14]. Although such LD was reported in Caucasian, Black and Oriental ethnicities, none of these studies detected a significant association of promoter SNPs with hypertension [10–16]. Two explanations should be considered for these apparently contradictory results. First, most previous studies focused on the SNP at nucleotide position –47, whereas this SNP was not detected in the Chinese populations studied by Wu et al. [1]. Second, Wu et al. [1] specifically report the association between promoter SNPs and hypertension in one but not another ethnic group. Therefore, it is possible that certain SNPs occur predominantly in specific ethnic groups and hence their associations with hypertension are also specific for such groups. If SNPs in the promoter region of the ADRB2 gene are associated with hypertension, this could involve two, not necessarily mutually exclusive, mechanisms. First, it is possible that the promoter SNP affects transcription of the gene and/or stability of the resulting mRNA. Indeed, it has been shown that mutagenesis in the promoter of mouse ADRB2 can affect translation [17], but the effects of the various SNPs in the human ADRB2 promoter have not been assessed systematically. Second, it is possible that the promoter SNPs simply are indicators of those in the coding region, to which they are linked and which mechanistically are involved in an altered receptor function [4–6]. The present data are insufficient to answer this question. Possible approaches to address this would comprise systematic structure–activity relationship studies of the effects of SNPs on promoter function in the ADRB2 gene or volunteer studies in subjects who have the same genotype in the coding region but differ with respect to that of the promoter. Another possibility would be to study patient groups that are sufficiently large to allow the detection of the additional effects of promoter SNPs beyond those of the associated coding region SNPs using multiple regression analysis. Given the heterogeneity between previously reported studies on the role of coding region SNPs in the ADRB2 gene, a definitive answer is likely to require a combination of all three approaches. In conclusion, the overall evidence suggests that SNPs in the coding region of the ADRB2 are not associated with hypertension in a clinically meaningful manner in most ethnic groups, either because their effect is too weak and/or because the SNP is too rare to have major effects on the general population. The study by Wu et al. [1] raises the possibility that SNPs in the promoter region of the ADRB2 may be associated with hypertension but, based upon the LD between the SNPs in the promoter and coding region, these data have to be interpreted with caution until confirmed by other investigators. In a more general vein, these data highlight the importance of considering ethnicity in genotype–phenotype relationship studies because the possibility exists that such associations are more prominent in some ethnic groups compared to others. Epigenetic and gene–environment interactions might cause such differences [18], but it is likely that genetic effects of polymorphisms are usually identical across racial groups with strongly differing allele frequencies [19]. Acknowledgement I.N.H. is recipient of a grant from the Deutsche Forschungsgemeinschaft.