β2-Adrenergic Receptor Gene Polymorphism, Age, and Cardiovascular Phenotypes
Previous studies suggest that variants of the β2-adrenergic receptor (ADRB2) may differently affect functional responses to adrenergic stimulation, thereby possibly modulating cardiovascular and metabolic phenotypes. We examined the hypothesis that G/R16 and Q/E27 polymorphism of ADRB2, or their haplotypes, may modulate blood pressure, cardiovascular structure, and function or metabolic cardiovascular risk factors in the general population. We examined a random sample of the general population (n=571; age, 35 to 64 years). Neither clinic nor 24-hour ambulatory blood pressure was significantly associated with ADRB2 genotypes in the overall population. Cardiac structure and function were also not influenced by ADRB2 polymorphism. After adjustment for potential confounders, association of the R16 allele with higher systolic blood pressure was observed in the subgroup of younger people (below age of 50 years). Haplotype analysis showed that higher blood pressure values were more specifically associated with the presence of R16-Q27. Younger people carrying the R16-Q27 haplotype also showed a trend toward lower heart rate, higher BMI, lower glycemia, and higher trygliceridemia, which is consistent with the hypothesis of a genetic predisposition to reduced cardiovascular and metabolic response to ADRB2 stimulation. This study does not provide evidence of a major role of ADRB2 gene variability in blood pressure modulation. However, association of ADRB2 polymorphism with cardiovascular and metabolic effects can be observed in younger subjects, before the development of age-related decline of ADRB2-mediated activity. Our study emphasizes the necessity of taking into account (patho)-physiological changes related to aging (in this case, decreased efficiency of ADRB2 signaling) when analyzing phenotypic effects of genetic variants.
In the attempt to dissect genetic factors influencing complex cardiovascular phenotypes, such as blood pressure (BP), several studies have focused on adrenergic receptors, and particularly on the β2 subtype, as a potential “candidate” gene.1 There are many reasons for considering the β2-adrenergic receptor (ADRB2) and its genetic variability a potential relevant contributor to BP regulation in humans. In fact, besides its key role as a mediator of the vasodilatory response to adrenergic agonists in the vasculature, this receptor may also modulate cardiac function (and perhaps structure), renal excretion of salt and water, as well as adipose tissue turnover, with additional effects on BP regulation. On the other hand, this receptor is involved in the homeostasis of several metabolic parameters, including plasma glucose and lipids.
Many different diallelic polymorphic sites have been detected in the ADRB2 sequence; in particular, 2 variants of ADRB2 present both evidence of functional relevance and high degree of heterozygosity, which are the basis for a possible pathophysiological role in the modulation of “complex” traits. The first polymorphism determines a glycine (G) to arginine (R) change at amino acid position 16 (G/R 16), whereas the second one consists of a glutamine (Q) to glutamate (E) change at codon 27 (Q/E 27). Previous in vivo and in vitro studies suggest that these variants of ADRB2 may differently affect functional responses to adrenergic stimulation, thereby possibly modulating cardiovascular and metabolic phenotypes2–9; moreover, recent data underscore the need to analyze the combined effect of single nucleotide polymorphisms within the same gene, that is, to examine the transcriptional and functional characteristics of specific haplotypes,4,5 and to examine these effects in the human vascular bed.6–9
The present study was undertaken to examine the main hypothesis that G/R 16 and Q/E 27 polymorphism of ADRB2, or haplotype combinations thereof, may modulate BP in the general population. To this purpose, we measured both clinic and 24-hour ambulatory BP, the latter being considered a more sensitive and reliable approach to detect small intergenotype differences. In addition, we also examined possible effects on other cardiovascular parameters (cardiac structure and function, as evaluated by echocardiography) and metabolic phenotypes possibly influenced by ADRB2 activity.
We examined a sample of subjects 35 to 64 years of age selected from the general population of Vobarno, a small town in northern Italy of approximately 8000. The participants were randomly extracted from the electoral rolls. Almost 80% of the invited people accepted to participate to the study project and gave their informed consent; in 571 unrelated subjects, we were able to obtain both genotypic and phenotypic information.
The protocol was approved by the ethics committee of our institution. After overnight fasting, each subject was admitted in the morning hours (8:00 to 9:00 am) to the outpatient clinic. Blood samples were taken for biochemical assessment (including plasma glucose, triglycerides, cholesterol and uric acid) and for genomic DNA extraction.
A careful medical history was collected by a physician, with special attention to family and personal history of hypertension, ischemic heart disease, stroke, diabetes mellitus, and dyslipidemia.
Clinic BP and 24-hour ambulatory BP were measured according to the recommendations of JNC VI10 and of the Italian Society of Hypertension,11 respectively. The diagnosis of hypertension was based on office BP values (systolic BP≥140 or diastolic BP≥90 mm Hg) or on current antihypertensive treatment. We considered normotensive persons those individuals with BP <140/90 mm Hg, not taking any drug treatment potentially interfering with BP.
The assessment of cardiac structure and function was performed by high-quality echocardiographs (HP Sonos 1500 and 5500), equipped with 2.5-MHz transducers with M-mode, 2D, and Doppler capabilities. Further details about the echographic procedures have been published elsewhere.12
Genomic DNA was extracted from peripheral blood samples by standard (phenol/chloroform) technique. A single polymerase chain reaction (PCR), giving an amplified product of 173 bp, was performed (sense, 5′-CTTCTTGCTGGCACGCAAT-3′; antisense, 5′-GTTCGAGCGTCTGCAGACGG-3′); the 25-μL PCR reaction contained 100 ng of genomic DNA, 0.5 mmol/L of each deoxynucleoside triphosphate, 2,5 μL of PCR 10×buffer (100 mmol/L Tris-HCl, 15 mmol/L MgCl2, 500 mmol/L KCl, pH 8,3), 0.5 pmol of each primer, and 0.05 U of Taq DNA polymerase. PCR was started with denaturation at 94°C for 3 minutes, followed by 35 cycles of denaturation (94°C, 30 seconds), annealing (58°C, 30 seconds), and extension (72°C, 45 seconds), with final extension at 72° for 7 minutes.
For the G/R16 polymorphism, the amplified product was digested at 60°C for 1 hour with 2 U of Bsr DI. Fragments were resolved on 12% polyacrylamide gel with 1× TBE buffer and visualized under ultraviolet illumination after staining with ethidium bromide. This digestion produces fragments of 108, 43, and 22 bp for the G allele and 130 and 43 bp for the R allele.
For the Q/E27 polymorphism, the amplified product was digested at 37°C for 1 hour with 1 U of Fnu 4H I and examined as above, resulting in fragments of 123 and 50 bp for the Q allele and 173 bp for the E allele. Restriction fragment pattern was assessed on polyacrylamide gel by 2 independent readers.
On the basis of the total sample size, the previously known standard deviation of 24-hour systolic and diastolic BP12 and the presumable genotype distribution in white populations,13 the present study had 80% power to detect, with a 0.05 2-sided significance level, differences in systolic/diastolic ambulatory BP means of 3.4/2.5 mm Hg (R16 recessive), 2.4/1.8 mm Hg (R16 dominant), 3.5/2.6 mm Hg (E27 recessive), and 2.4/1.8 mm Hg (E27 dominant).
The association of ADRB2 polymorphisms with categorical variables was tested by χ2 test statistic, whereas the association with continuous variables was tested by ANOVA, considering all the possible models of inheritance (codominant, dominant, and recessive). In addition, the influence of potential confounders (such as age, gender, BMI, and others, according to the phenotype under investigation) was assessed by multivariate analysis (SPSS). When considering quantitative phenotypes possibly affected by drug treatment, the analysis was limited to subjects who were not taking any drug. No probability value adjustment for multiple comparisons, such as the Bonferroni adjustment or others, was applied. Values are given as mean±SEM.
Genotype and Haplotype Distribution
In the overall population (n=571), allele frequency (0.62 and 0.38 for G16 and R16; 0.63 and 0.37 for Q27 and E27, respectively) and genotype distribution (G/G16=0.38, G/R16=0.48, R/R16=0.14; Q/Q27=0.38, Q/E27=0.49, E/E27=0.13) were similar to those previously reported in whites and in agreement with Hardy-Weinberg law expectations.
When genotype distribution of the 2 polymorphisms was examined (Table 1), there was clear evidence of linkage disequilibrium, 3 combinations (G/R16-E/E27, R/R16-Q/E27, and R/R16-E/E27) being not as frequent as it could be expected on the basis of relative frequencies of G/R16 and Q/E27 genotypes. Such a finding suggests that the R16-E27 haplotype (ie, adenine at base position 46 and guanine at position 79) is rare in this population sample, which is consistent with previous observations reporting a relative frequency of <0.01 for the R16-E27 haplotype. Although haplotype could not be unequivocally assessed in subjects heterozygotes for both polymorphisms (G/R16-Q/E27), in the remaining subjects, the relative haplotype frequency was 0.36 for G16-Q27, 0.32 for R16-Q27, 0.31 for G16-E27, and 0.01 for R16-E27, respectively. A comparison of ADRB2 nucleotide sequence in different animal species suggests that G16-E27 is the common ancestral form, from which G16-Q27 and R16-Q27 variants are subsequently derived in humans (data not shown). Genotype and haplotype distribution were similar when subjects were stratified according to age (median) or gender.
Demographic and anthropometric parameters, according to G/R16 and Q/E27 genotype, are summarized in Table 2. No significant difference of age, gender distribution, height, weight, and BMI was observed among G/R16 and Q/E27 genotypes.
A positive family history for ischemic heart disease (53% versus 37%, P=0.01), diabetes mellitus (39% versus 29%, P=0.09), and dyslipidemia (28% versus 18%, P=0.03) was more frequent in R16 homozygotes. On the other hand, the prevalence of parental hypertension was not influenced by either G/R16 or Q/E27 genotype.
Prevalence of hypertension was similar among G/R16 and Q/E27 genotypes (Table 3). When considering BP as a quantitative trait in the overall sample, no statistically significant differences (either 24-hour or office measurements) related to ADRB2 polymorphism were detectable. However, after exclusion of subjects receiving pharmacological treatment potentially interfering with BP values and adjustment by gender, age, and BMI, we noticed a trend toward association of higher ambulatory systolic BP with allele R16 and allele Q27 (Table 3). Noticeably, when the whole population was stratified according to median age (49 years), this association was determined by BP differences in younger (35 to 49 years of age) rather than in older subjects (50 to 65 years), (P=0.016 for G/R16, P=0.08 for Q/E27). In the subgroup of younger people, prevalence of hypertension was slightly higher in carriers of the R16 allele (49.4 versus 41%, P=0.07). Similar results were obtained when daytime (6 am to 10 pm) and nighttime (10 pm to 6 am) ambulatory BP measurements were considered separately. Considering the possibility that the 2 polymorphisms may interact and manifest their effect only in specific haplotypes, we compared subjects homozygotes for each haplotype, that is to say, carriers of 2 copies of G16-E27, G16-Q27, or R16-Q27, respectively (there was only one subject homozygote for R16-E27). Twenty-four-hour systolic BP was significantly higher (P=0.03) in homozygotes for the haplotype R16-Q27 than in G16-E27 homozygotes (Figure 1), with intermediate values for the G16-Q27 homozygotes (not shown).
We also considered other BP variability parameters potentially influenced by adrenergic activity, such as standard deviation of all ambulatory measurements, BP difference between daytime and nighttime measurements, difference between office and ambulatory monitored values, or prevalence of white-coat hypertension (subjects who appear hypertensive at office measurements but not at 24-hour ambulatory monitoring), but none of these indexes was associated to β2-AR polymorphism.
Diastolic BP differences among genotypes, though concordant with those observed for systolic values, never reached conventional statistical significance.
Heart rate did not differ significantly among genotypes of either ADRB2 polymorphism, both at office and at 24-hour ambulatory measurements (Table 3). However, a slight trend toward lower heart rate in carriers or R16 was observed, and it resulted statistically significant for office blood pressure in younger people (G/G16=70.6±1.0; G/R16=68.8±0.87; R/R16=67.2±1.8 beats/min; mean±SEM, P=0.05). No statistically significant difference was observed in heart rate variability indexes (standard deviation of ambulatory measurements, average daytime versus average nighttime difference, office versus 24-hour ambulatory difference; data not shown).
We did not find any significant heart rate difference according to ADRB2 haplotype despite a trend toward lower heart rate in R16-Q27 subjects (data not shown).
Cardiac Structure and Function
No difference in left ventricular (LV) geometry or calculated LV mass, as well as in indexes of LV systolic and diastolic function, was observed among ADRB2 genotypes, either in raw data (Table 4) or after adjustment for confounders (data not shown). Cardiac index was slightly but not significantly lower in R16-Q27 (2.98±0.11 L/min/m2) as compared with G16/E27 homozygotes (3.22±0.15 L/min/m2, P=0.18).
Fasting plasma glucose was significantly associated with G/R16, higher levels being found in subjects with at least one copy of R16 (G/G16=92.1±0.81; G/R16=95.1±1.1; R/R16=94.4±1.5 mg/dL; mean±SEM, P<0.05). In addition, prevalence of diabetes mellitus type 2 was significantly higher in carriers of R16 (4.3% versus 1.4%, P=0.053). The Q/E27 genotype did not influence plasma glucose.
No significant differences of plasma cholesterol, triglycerides, or uric acid levels were detected among ADRB2 genotypes. A multivariate analysis of biochemical phenotypes, with adjustment by gender, age, and BMI, confirmed (P=0.02) the association of G/R16 polymorphism with glycemia.
Haplotype analysis showed that subjects homozygous for R16-Q27 had significantly higher levels of plasma glucose than G16-Q27 (94.41±1.5 versus 89.8±1.5, P=0.05) but similar to G16-E27 homozygotes (93.8±1.8, NS). Noticeably, a subgroup analysis revealed an opposite trend in younger versus older subjects (cut-point: 50 years of age), as illustrated in Figure 2. In fact, carriers of R16-Q27 (possibly associated to reduced β2-AR mediated activity) had slightly lower levels of plasma glucose and higher levels of plasma triglycerides (as well as lower BMI) at younger age, whereas the opposite was found in older subjects.
The simplest conclusion that could be drawn from the results of the present investigation is that 2 common genetic variants of ADRB2, the G/R16 and Q/E27 polymorphisms, have no major influence on blood pressure in a white population. However, some findings of this study deserve more in-depth consideration. In fact, we have found some evidence that after careful adjustment by common confounders, there is a trend for association of R16 and Q27 alleles to higher BP levels, particularly in younger subjects. Moreover, haplotype combination of these 2 alleles is also clearly associated to higher BP levels throughout the day.
One of the confounding factors we have taken into account is chronic pharmacological treatment, which may interfere with blood pressure regulation; this is an often underestimated aspect in genetic epidemiology studies examining blood pressure as a quantitative trait. As a matter of fact, 20% of the subjects participating to the Vobarno project were regularly taking some kind of drug, mostly antihypertensives or oral contraceptives, and it is therefore appropriate to perform quantitative data analysis by excluding pharmacologically treated subjects. Considering arterial hypertension as a dichotomous variable, according to current guidelines criteria, permits inclusion of all subjects in the analysis: Also in this case, there is a trend toward association of the R16 allele with hypertension in younger people.
A similar association was found in previous studies; the R16 variant was significantly more frequent in offspring of hypertensive parents than in offspring of normotensive parents in the Bergen Blood Pressure Study,14 and a study in German twins showed that R16 is associated with higher systolic blood pressure levels.15 Moreover, a recent study in Scandinavians reported an increased risk of hypertension in subjects with type 2 diabetes, homozygous carriers of R16, and higher systolic blood pressure levels in subjects with at least one R16 allele than in discordant siblings (homozygous for the G16 allele).16 This study also found association of the Q27 allele with the hypertensive condition, a result confirmed in a large case-control study in the Japanese population.17
Other studies do not confirm the association of R16 and/or Q27 with hypertension18–23 or appear somehow contradictory.6,24–26 Some reports provide evidence of linkage of blood pressure at the ADRB2 locus26,27 but no association with ADRB2 diallelic variants,27 suggesting linkage disequilibrium with other unidentified genetic variants. However, genetic background, ethnicity, and pleiotropy (with distinct effects on glucose metabolism and body weight regulation) may originate these heterogeneous findings. The decision to examine a sample of general population instead of a case-control or a family-based study may also be relevant, since there is some evidence that association of ADRB2 polymorphisms with blood pressure is mostly found in normotensive subjects.27
Further support to the results of our study comes from functional studies in humans. Dishy et al9 recently reported enhanced desensitization of ADRB2 (ie, decline of dorsal hand vein vasodilation in response to prolonged isoproterenol local infusion) in R16-Q27 homozygotes. The desensitization appeared to be related to R16 rather than to Q27; on the other hand, G16-E27 homozygotes had a significantly higher maximal vasodilatory response to isoproterenol. A very similar finding of increased dorsal hand vein vasodilation in response to isoproterenol in G16-E27 homozygotes was obtained by another group of investigators.8 Both studies are consistent with the hypothesis of a reduced β2-AR mediated vasodilation in carriers of the R16-Q27 haplotype, possibly resulting in increased blood pressure values, such as those observed in the present study. There is also evidence that ADRB2-mediated responses are similarly affected in lung.5
A functional role of ADRB2 polymorphism has also been observed in in vitro studies from Liggett and coworkers, suggesting a reduced desensitization in vivo for the R16 and Q27 alleles.2,3,28 In addition, it has been shown that ADRB2 mRNA levels and receptor density in HEK293 cells transfected with the G16-E27 haplotype is ≈50% greater than in those transfected with the R16-Q27 haplotype.5 Altogether, a large body of evidence provides the rational basis for considering ADRB2 polymorphism functionally relevant, the R16-Q27 haplotype being associated with reduced ADRB2-mediated activity.
A key issue of the present study is the fact that the association of R16-Q27 haplotype with increased blood pressure is limited to younger people. This is consistent with several findings suggesting that sympathetic nervous system hyperactivity is mainly observed in younger hypertensive persons, in which increased peripheral sympathetic nerve activity and higher levels of circulating catecholamines have been found. In addition, an age-related decline in ADRB2 responsiveness is well documented and may contribute to obscure the influence of ADRB2 polymorphism on blood pressure regulation at older age.29 Noteworthy, heart rate and cardiac index were slightly reduced in younger subjects carrying the R16 variant, possibly the result of reduced ADRB2-mediated activity on the heart.
No significant differences were observed on other echocardiographically assessed indexes of LV systolic or diastolic function. Only 2 groups of researchers have previously assessed hemodynamic parameters in relation to G/R16 ADRB2 polymorphism by means of impedance cardiography; they did not find any significant difference among genotypes of either stroke volume index or cardiac index in white6,18 or in black18 individuals. We also did not observe any variation of LV geometry or calculated LV mass associated to ADRB2 polymorphism. A similar negative finding (no association of LV mass index to ADRB2 polymorphism) was reported in a black South African population.22 On the other hand, Busjahn et al15 reported a modest but significant association of increased interventricular septum thickness to the R16 allele in a group of young subjects, but it probably reflected the corresponding increase of blood pressure also found in R16 carriers. Overall, there is no convincing evidence in the literature supporting a role for ADRB2 in promoting cardiovascular hypertrophy. Interestingly, transgenic mice overexpressing ADRB2 in the heart present enhanced cardiac contractility but no ventricular hypertrophy, which is observed instead in transgenic mice overexpressing the α1B-adrenergic receptor.30
A few studies have explored the association of ADRB2 polymorphisms with type 2 diabetes mellitus, obesity, or various metabolic parameters (plasma glucose, insulin, triglycerides, fatty acids, cholesterol), with variable results.20,31–33 Our findings concerning metabolic phenotypes can be interpreted under different perspectives. In the whole population, higher fasting plasma glucose and prevalence of diabetes (as well as more frequent positive family history for diabetes) were associated with the R16 allele; on the other hand, results are quite different in younger and older subjects. We are aware of the risk of a post hoc subgroup analysis, but at the same time we must underscore the coherence of trends observed in younger people carrying the R16-Q27 haplotype (higher BP, lower HR, higher BMI, lower glycemia, higher trygliceridemia) with the hypothesis of a genetic predisposition to reduced cardiovascular and metabolic response to ADRB2 stimulation, which is obscured at older age, when ADRB2 activity is decreased.
These results do not provide evidence of a major role of ADRB2 polymorphism in the modulation of blood pressure in the general population, which is the main a priori end point of the present study. However, they also indicate that association of the ADRB2 polymorphism is associated with cardiovascular and metabolic effects in younger subjects, before the development of an age-related decline of ADRB2-mediated activity. Our study emphasizes the need of taking into account (patho)-physiological changes related to aging (in this case, decreased efficiency of ADRB2 signaling) when analyzing phenotypic effects of genetic variants.
This study was supported in part by COFIN grants from the Italian Ministry of University and Scientific and Technologic Research. We thank Dr Alessandra Panarotto, who provided excellent technical assistance.
- Received October 9, 2002.
- Revision received October 29, 2002.
- Accepted December 4, 2002.
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.
Gratze G, Fortin J, Labugger R, Binder A, Kotanko P, Timmermann B, Luft FC, Hoehe MR, Skrabal F. β2-adrenergic receptor variants affect resting blood pressure and agonist induced vasodilation in young adult Caucasians. Hypertension. 1999; 33: 1425–1430.
Cockroft JR, Gazis AG, Cross DJ, Wheatley A, Dewar J, Hall IP, Noon JP. β2-Adrenoceptor polymorphism determines vascular reactivity in humans. Hypertension. 2000; 36: 371–375.
G. Parati S, Bosi M, Castellano M, Di Rienzo G, Germanò S, Lattuada P, Mormino L, Mos S, Omboni P, Palatini A, Ravogli D, Rizzoni P, Verdecchia M, Zito. Guidelines for 24-h non-invasive ambulatory blood pressure monitoring: report from a working group of the Italian Society of Hypertension. High Blood Pressure. 1995; 4: 168–174.
M. L. Muiesan G, Pasini M, Salvetti S, Calebich R, Zulli M, Castellano D, Rizzoni G, Bettoni A, Cinelli E, Porteri V, Corsetti E, Agabiti Rosei. Cardiac and vascular structural changes: prevalence and relation to ambulatory blood pressure in a middle-aged general population in northern Italy: the Vobarno Study. Hypertension. 1996; 27: 1046–1052.
The CANVAS ADRB2-ADRB2A46-13/07/99; The CANVAS ADRB2-ADRB2A79-13/07/99. Available at http://genecanvas.idf.inserm.fr.
Timmermann B, Mo R, Luft FC, Gerdts E, Busjahn A, Omvik P, Li GH, Schuster H, Wienker TF, Hoehe MR, Lund-Johansen P. β-2 Adrenoceptor genetic variation is associated with genetic predisposition to essential hypertension: the Bergen Blood Pressure Study. Kidney Int. 1998; 53: 1455–1460.
Busjahn A, Li G-H, Faulhaber H-D, Rosenthal M, Becker A, Jeschke E, Schuster H, Timmermann B, Hoehe MR, Luft FC. β2-Adrenergic receptor gene variations, blood pressure, and heart size in normal twins. Hypertension. 2000; 35: 555–560.
Bengtsson K, Orho-Melander M, Melander O, Lindblad U, Ranstam J, Rastam L, Groop L. β2-adrenergic receptor gene variation and hypertension in subjects with type 2 diabetes. Hypertension. 2001; 37: 1303–1308.
Kato N, Sugiyama T, Morita K, Kurihara H, Sato T, Yamori Y, Yazaki Y. Association analysis of β2-adrenergic receptor polymorphisms with hypertension in Japanese. Hypertension. 2001; 37: 286–292.
Herrmann V, Büscher R, Go MM, Ring KM, Hofer JK, Kailasam MT, O’Connor DT, Parmer RJ, Insel PA. β2-adrenergic receptor polymorphisms at codon 16, cardiovascular phenotypes and essential hypertension in white and African Americans. Am J Hypertens. 2000; 13: 1021–1026.
Herrmann S-M, Nicaud V, Tiret L, Evans A, Kee F, Ruidavets JB, Arveiler D, Luc G, Morrison C, Hoehe MR, Cambien MP. Polymorphisms of the β2-adrenoceptor (ADRB2) gene and essential hypertension: the ECTIM and PEGASE studies. J Hypertens. 2002; 18: 167–172.
Kotanko P, Binder A, Tasker J, DeFreitas P, Kamdar S, Clark AJL, Skrabal F, Caulfield M. Essential hypertension in African Caribbeans associates with a variant of the β2-adrenoceptor. Hypertension. 1997; 30: 773–776.
Ranade K, Shue WHH, Hung YJ, Hsuing CA, Chiang FT, Pesich R, Hebert J, Olivier M, Chen YDI, Pratt R, Olshen R, Curb D, Botstein D, Risch N, Cox DR. The glycine allele of a glycine/arginine polymorphism in the b2-adrenergic receptor gene is associated with essential hypertension in a population of Chinese origin. Am J Hypertens. 2001; 14: 1196–1200.
Bray MS, Krushkal J, Li L, Ferrell R, Kardia S, Sing CF, Turner ST, Boerwinkle E. Positional genomic analysis identifies the β2-adrenergic receptor gene as a susceptibility locus for human hypertension. Circulation. 2000; 101: r9–r15.
Tomaszewski M, Brain NJR, Charchar FJ, Wang WYS, Lacka B, Padmanabahn S, Clark JS, Anderson NH, Edwards HV, Zukowska-Szczechowska E, Grzeszczak W, Dominiczak AF. Essential hypertension and β2-adrenergic receptor gene: linkage and association analysis. Hypertension. 2002; 40: 286–291.
Julius S, Valentini M, Palatini P. Overweight and hypertension: a 2-way street? Hypertension. 2000; 35: 807–813.