Journal of Human Hypertension (2007) 21, 387–392 & 2007 Nature Publishing Group All rights reserved 0950-9240/07 $30.00 www.nature.com/jhh ORIGINAL ARTICLE Intra-erythrocyte cation concentrations in relation to the C1797T b-adducin polymorphism in a general population

T Richart1, L Thijs1, T Kuznetsova1, V Tikhonoff1,2, L Zagato3, P Lijnen1, R Fagard1, J Wang4, G Bianchi3 and JA Staessen1 1Division of Hypertension and Cardiovascular Rehabilitation, Department of Cardiovascular Diseases, Studies Coordinating Centre, University of Leuven, Leuven, Belgium; 2Department of Clinical and Experimental Medicine, University of Padua, Padua, Italy; 3Division of Nephrology, Dialysis and Hypertension, University Vita Salute San Raffaele, Milan, Italy and 4Centre for Epidemiological Studies and Clinical Trials, Ruijin Hospital, Shanghai Institute of Hypertension, Shanghai, China

Genetic variability in the ADD1 (Gly460Trp) and ADD2 P ¼ 0.93). In men, iK, iMg and iNa did not differ according (C1797T) subunits of the cytoskeleton adducin to ADD1 genotypes. In men, iK (R 2 ¼ 0.128) increased plays a role in the pathogenesis of hypertension, with age and serum Na þ , but decreased with serum total possibly via changes in intracellular cation concentra- calcium and the daily intake of alcohol. iMg (R2 ¼ 0.087) tions. ADD2 1797CC homozygous men have decreased decreased with age, but increased with serum total erythrocyte count and hematocrit. We investigated calcium. After adjustment for these covariates (Pp0.04 possible association between intra-erythrocyte cations for all), findings in men for iK (CC versus T: 85.8 versus and the adducin polymorphisms. In 259 subjects (mean 87.3 mmol/l; P ¼ 0.14) and iMg (1.91 versus 1.82 mmol/l; age 47.7 years), we measured intra-erythrocyte Na þ P ¼ 0.03) remained consistent. In 136 women, none of [iNa], K þ [iK] and Mg2 þ [iMg], serum cations and the phenotype–genotype relations reached significance. adducin genotypes. Genotype frequencies (ADD1: GlyGly Changes in intra-erythrocyte cations in ADD2 1797CC 61.5%, Trp 38.5%; ADD2: CC 80.4%, T 19.6%) complied homozygous men might lead to osmotic fragility of with Hardy–Weinberg proportions. In men, ADD2 CC erythrocytes, but to what extent they reflect systemic homozygotes (n ¼ 100) compared to T-carriers (n ¼ 23) changes or are possibly involved in blood pressure had slightly lower iK (85.8 versus 87.5 mmol/l cells; P ¼ regulation remains unknown. 0.107), higher iMg (1.92 versus 1.80 mmol/l cells; P ¼ Journal of Human Hypertension (2007) 21, 387–392. 0.012), but similar iNa (6.86 versus 6.88 mmol/l cells; doi:10.1038/sj.jhh.1002154; published online 15 February 2007

Keywords: adducin; erythrocytes; polymorphism; genetics; osmotic

Introduction knockout mice show a phenotype characterised by mild anemia and compensated hemolysis.1,2 In The membrane-skeleton protein adducin stimulates humans, a common polymorphism (C1797T) occurs the assembly of the spectrin– network, which in exon 15 of the b-adducin . We have demon- determines the structural integrity of the cell strated previously in cross-sectional and prospec- membrane. Adducin is composed of either a- and tive studies that genetic variability in the subunits of b-ora- and g-subunits, which are encoded by the cytoskeleton protein adducin plays a role in the different . In erythrocytes, a- and b-adducin are pathogenesis of hypertension, possibly via altera- abundantly present, whereas most other tissues tion of the intracellular cation concentrations.3,4 predominantly express a-g-heterodimers. b-Adducin In addition, we have recently noticed that in men consuming alcohol, b-adducin CC homozygosity was associated with lower erythrocyte count, 5 Correspondence: Dr JA Staessen, Studies Coordinating Centre, haemoglobin level and haematocrit. The osmotic Laboratory of Hypertension, Hypertension and Cardiovascular stability of erythrocytes decreases with lower intra- Rehabilitation Unit, Department of Cardiovascular Diseases, cellular potassium concentration.6 We therefore Campus Gasthuisberg, Herestraat 49, box 702, Leuven, B-3000 measured the intra-erythrocyte concentration of Belgium. cations in a sub-sample of our previous study E-mail: [email protected] 5 Received 21 August 2006; revised 27 November 2006; accepted 10 population to test possible association with the December 2006; published online 15 February 2007 a- and b-adducin polymorphisms. Intra-erythrocyte cations and adducin T Richart et al 388 Methods iNa þ and 2.1270.12% for iK þ . The inter-assay variation determined in five blood samples was Study population 2.3771.03% for iNa þ and 3.2271.18% for iK þ . The Ethics Committee of the University of Leuven approved the protocol of the Flemish study on Environment, Genes and Health Outcomes Determination of genotypes (FLEMENGHO). All subjects gave informed consent. Genomic DNA was extracted from peripheral blood. From August 1985 to November 1990, we recruited a Allelic discrimination of the ADD1 460 glycine random sample of the households living in a (Gly)/460 tryptophan (Trp) and the ADD2 C1797 T geographically defined area of Northern Belgium.7 polymorphism was carried out using the 50-nuclease We stratified the sample by sex and age (20–39, assay9 on an ABI Prism 7700 apparatus (Perkin 40–59 and X60 years) to recruit equal numbers of Elmer, Foster City, CA, USA). The forward and participants in each of the six strata. From June 1996 reverse primers and the 460Gly and 460Trp probes to December 2000, we enrolled nuclear families, employed in the TAQMan2 assay were 50-CGTCCAC using the former participants (1985–1990) as index ACCTTAGTCTTCGACTT-30 50-GGAGAAGACAAGA persons. The participation rate among the 2310 TGGCTGAACTC-30,50-FAM-TTCCATTCTGCCCTTC subjects contacted was 66.1%. Our previous study5 CTCGGATAMRA-30 and 50-TET-TTCCATTCTGCCA included 1870 subjects with haematological pheno- TTCCTCGGAA-TAMRA-3,0 respectively (where FAM, types and the C1797T b-adducin genotype. Of these, TAMRA and TET are dyes used with the TAQMan 277 subjects were randomly selected and underwent system). For the ADD2 C1797T polymorphism, the additional measurements of their intra-erythrocyte forward and reverse primers and the 1797C and 1797T cation concentrations and the concentration of probes employed in the TAQMan2 assay were 50-AGG serum ionised calcium. We excluded 18 subjects, AACGAGAGCCAGGCTCT-30 50-TTCATCAAAACAC because of missing values. ACACCTACCAAT-30,50-VIC2-TTCTTCAGCGTTGC Trained nurses measured each participant’s blood CCTCCACATTAMRA2-30 and 50-FAM2-TCTTCAGT pressure (BP) five times consecutively by conven- GTTGCCCTCCACATCTG-TAMRA2-30 (where VIC2, tional sphygmomanometry after the subjects had TAMRA2 and FAM2 are dyes used with the rested for at least 5 min in the sitting position. These TAQMan system). Per 25 ml, the polymerase chain five readings were averaged for analysis. Hyper- reaction (PCR) fluid contained 50 ng of DNA, tension was defined as a BP of at least 140 mm Hg 200 nmol of primers, 50 nmol of FAM probe and systolic or 90 mm Hg diastolic, or as use of anti- 100 nmol of VIC probe. The amplification condi- hypertensive drugs. Body mass index (BMI) was tions were 501C for 2 min and 951C for 10 min, body weight in kilograms divided by height in followed by 40 cycles at 951C for 15 s and 621C for metres squared. We used a standardized and 1 min. The genotyping procedure was established validated8 questionnaire to collect information after we had confirmed the polymorphism (C1797T, on medical history, smoking habits, intake of starting from ATG; dbSNP number rs4984; URL: alcohol, use of medications and the menstrual cycle http://www.ncbi.nlm.nih.gov/SNP) by sequencing of women. From the type and number of alcoholic 17 individuals using the ABI Prism Big Dye beverages used each day, we calculated alcohol Terminator cycle sequencing ready reaction kit consumption in grams per day. (Applied Biosystems, Foster City, CA, USA). We tested the frequency of the polymorphism in a random population of 250 blood donors from North Italy, and found that the minor allele had a Measurement of phenotypes frequency of 15%. For quality control, 10% of the Venous blood samples were collected into ethylene- DNA samples in our study were genotyped in diamine-tetraacetic acid-containing tubes. Erythro- duplicate in a blinded fashion. Duplicate genotypes cyte count, haemoglobin, and haematocrit were were confirmatory in all samples. measured in whole-blood specimens with a SYS- MEX HST 403XE model automated cell counter (Systmex Corporation, Chuku, Kobe, Japan). Serum Statistical analysis ionised calcium and pH were determined using For database management and statistical analysis, an ICA2 ionised calcium analyser (Radiometer, we used SAS software, version 9.1.3 and JMP, Copenhagen, Denmark). version 6 (SAS Institute, Cary, NC, USA). Measure- Erythrocytes (1.5 ml) were washed three times ments with a skewed distribution were norma- with ice-cold 140 mM choline chloride. After centri- lised by logarithmic transformation. Comparisons fugation at 5500 r.p.m. at 41C for 3 min, the super- of means and proportions were performed using natant and the top layer of the cells were discarded Student’s t-test and Fisher’s exact test, respectively. by aspiration. The cells were lysed with double- We computed Pearson’s correlation coefficients distilled water and the Na þ ,Kþ and Mg2 þ concen- between the cation concentrations in erythrocytes trations were measured by atomic absorption and serum and compared these correlation co- spectrometry. The intra-assay (7s.e.) variation de- efficients between women and men using Fisher’s termined in 10 blood samples was 1.6370.31% for z-transformation.

Journal of Human Hypertension Intra-erythrocyte cations and adducin T Richart et al 389 To examine the relationship of the intra-erythro- opposite was true for iK (Table 2). As expected, the cyte cation concentrations with serum and urinary serum concentrations of sodium, potassium and cations, anthropometric characteristics, and various magnesium and the 24-hour urinary excretion environmental and lifestyle factors, we used step- of sodium, potassium and calcium were lower in wise linear regression analysis with the P-value for women than men (Table 2). covariates to enter and stay in the model set at 0.15. Smoking was equally prevalent among women We considered as possible explanatory variables: and men. The median number of cigarettes smoked age, BMI, the serum concentrations of sodium, per day was 15 (interquartile range (IQR) 10–25). potassium, magnesium and ionised and total Compared to women, more men reported alcohol calcium, the 24-hour urinary excretion of sodium, intake of at least 5 g daily (Table 1). Among regular potassium, magnesium and calcium and their drinkers, the median daily alcohol consumption excretion ratios, current smoking (0, 1) and the was 20 g (IQR 10–23.2) in women and 20 (IQR quantity of alcohol consumed per day. Because 10–33.5) in men. Of the premenopausal women, of our previously reported sex-specific findings on 15 (17.9%) took oral contraceptives. None of the the association between hematological phenotypes 52 postmenopausal women was on hormone repla- and genetic polymorphisms, we analysed women cement therapy. and men separately.5,10 We performed retrospec- In men, we found positive and significant correla- tive power calculations, using the PROC POWER tions between iMg and serum total calcium (r ¼ 0.22, procedure. P ¼ 0.013) and between iK and serum Mg2 þ (r ¼ 0.27, P ¼ 0.029; Figure 1). In women, the corresponding correlation coefficients were –0.08 (P ¼ 0.32) and Results 0.09 (P ¼ 0.32), respectively. The P-values for the sex differences in the strength of these associations were Anthropometric characteristics and cation 0.14 and 0.013, respectively. measurements The 259 participants included 136 women (52.3%) and 32 hypertensive patients (12.3%) of whom only Phenotype–genotype associations one was taking antihypertensive drugs. Age ranged The frequencies of the a-adducin (ADD1) genotypes from 20 to 76 years. Table 1 lists the characteristics (GlyGly 61.5%, Trp 38.5%) and the b-adducin (ADD2) of the study population by gender. Systolic and dias- genotypes (CC 80.4%, T 19.6%) did not deviate from tolic BP, blood haemoglobin concentration, haema- Hardy–Weinberg equilibrium (P40.15). Sex distribu- tocrit and erythrocyte count were lower in women. tion, anthropometric characteristics and BP did not iNa was higher in men than women, whereas the differ according to the a-andb-adducin genotypes.

Table 1 Characteristic of the study participants Table 2 Intra-erythrocyte, serum and urinary cation concentra- tions

Characteristics Women Men P–value a w (n ¼ 136) (n ¼ 123) Characteristics Women Men P–value (n ¼ 136) (n ¼ 123) Age (years) 46.8713.3 48.7713.4 0.25 BMI (kg/m2)a 25.273.9 25.973.3 0.11 Intra-erythrocyte b 7 7 Sodium (mmol/l cells) 6.4271.01 6.8671.03 0.0005 SBP (mm Hg) 119.7 14.5 126.0 18.5 0.0042 7 7 DBP (mm Hg)b 71.478.5 74.578.7 0.0034 Potassium (mmol/l 87.60 4.32 86.12 4.58 0.0078 cells) Smokers (n%) 39 (28.7) 38 (30.7) 0.73 7 7 Alcohol intake X5g/ 12 (8.8) 37 (29.8) o0.0001 Magnesium (mmol/l 1.86 0.20 1.90 0.22 0.15 day (n%) cells) Contraceptive pill 15 (11) Serum use (n%) 7 7 Postmenopausal 52 (38) Sodium (mmol/l) 139.9 1.7 141.0 1.8 o0.0001 c Potassium (mmol/l) 4.0270.33 4.1270.31 0.0209 status (n%) 7 7 Haemoglobin 8.2870.71 9.3771.00 o0.0001 Magnesium (mmol/l) 0.80 0.07 0.83 0.07 0.0009 Total calcium (mmol/l) 2.3570.09 2.3770.10 0.10 (mmol/l) 7 7 Haematocrit (%) 38.372.9 42.772.7 o0.0001 Ionised calcium at pH 1.27 0.52 1.27 0.54 0.98 Serum erythrocyte 4.3070.38 4.8570.34 o0.0001 7.4 (mmol/l) count (1012 cells/l) Urine Sodium (mmol/24 h) 141754 175772 o0.0001 Abbreviations: BMI, body mass index, DBP, diastolic blood pressure; Potassium (mmol/24 h) 67722 78729 0.0008 SBP, systolic blood pressure. 7 7 7 Calcium (mmol/24 h) 4.29 2.45 5.10 2.51 0.0087 Plus-minus values are mean s.d. 7 7 P-values are for the differences between women and men. Magnesium (mmol/ 4.76 4.83 5.24 4.42 0.40 aBody mass index is weight in kilograms divided by the square of 24 h) 7 7 height in metres. Na/K 2.31 1.89 2.35 0.82 0.81 bOnly one participant was on anti-hypertensive drug treatment. cNone of the postmenopausal women was on hormonal replacement wP-values are for the differences between women and men. therapy. aPlus-minus values are mean7s.d.

Journal of Human Hypertension Intra-erythrocyte cations and adducin T Richart et al 390 CC homozygotes 105 2.5 T-allele carriers 2.3 95 2.1 85 1.9

1.7 75 r = 0.22 iK (mmol/L cells) r = 0.27 iMg (mmol/L cells) 1.5 P = 0.013 P = 0.029

2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 3.0 0.7 0.8 0.9 1.0 1.1 Serum total Ca2+ (mmol/L) Serum Mg2+ (mmol/L) Figure 1 Correlations between cations in erythrocytes and serum in 123 men. Open symbols denote b-adducin (ADD2) 1797CC homozygotes and closed symbols T allele carriers.

P = 0.11 P = 0.01 Discussion

97.5 2.45 The key finding of our study was that in men intra- 95.0 2.35 erythrocyte magnesium was on average 0.12 mmol/l 2.25 92.5 cells lower in b-adducin 1797T-allele carriers com- 2.15 90.0 pared to CC homozygotes, with an opposite trend in 2.05 the intra-erythrocyte potassium concentration. In 87.5 1.95 women, these associations with the ADD2 C1797T 85.0 1.85 polymorphism did not reach statistical significance. 82.5 1.75 In both sexes, none of the phenotype–genotype

iK+ (mmol/L cells) 80.0 1.65 relations with the a-adducin Gly460Trp polymor- iMg++ (mmol/L cells) 1.55 phism reached significance. In men, there was also 77.5 no interaction between the ADD1 and ADD2 CC TCCTgenotypes in relation to the intra-erythrocyte cation concentrations. Figure 2 Intra-erythrocyte cation concentrations in 123 men by 11 b-adducin (ADD2) genotype. Horizontal lines in the scatterplot Investigations in rats, interventions in never- represent means. treated hypertensive patients and epidemiological studies12–14 have shown that the Trp point mutation in the a-adducin subunit is associated with a pheno- In men, BAD CC homozygotes (n ¼ 100) compared type characterised by higher activity of the sodium to T-carriers (n ¼ 23) had slightly lower iK (85.8 pump, hence increased tubular sodium reabsorption versus 87.5 mmol/l cells, P ¼ 0.107), higher iMg in the kidney and ultimately hypertension. Altera- (1.92 versus 1.80 mmol/l, P ¼ 0.012), but similar tions in the erythrocyte membrane mirror those iNa (6.86 versus 6.88, P ¼ 0.93; Figure 2). In men in renal tubular cells. Glorioso and co-workers15 iK, iMg and iNa did not differ according to the investigated erythrocytes from 268 never-treated a-adducin genotypes or the interaction between the North Sardinian patients with essential hyperten- a- and b-adducin polymorphisms. In 136 women, sion (34.7% women). They noticed that at Vmax, the none of the phenotype–genotype relations reached Na,K-pump, the Na,K,Cl-cotransport and the Li, significance (Table 3, P40.49). A sensitivity analy- Na-countertransport were faster in a-adducin Trp sis, from which we excluded hypertensive patients, allele carriers than GlyGly homozygotes.15 The produced results similar to those reported above, in current study included only 12.3% hypertensive both men and women. Retrospective power calcula- patients and had a sample size, which was insuffi- tions suggested that 120 subjects provided approxi- cient to confirm the well-known associations of BP mately 70 and 60% power to demonstrate significant with the a-adducin Gly460Trp polymorphism. (Po0.05) differences on one- and two-sided tests, We have demonstrated previously that in men respectively. consuming alcohol b-adducin 1797CC homozygos- In stepwise regression analysis involving men, ity was associated with lower erythrocyte count, iK (R2 ¼ 0.128) increased with age (P ¼ 0.029) and haemoglobin level and haematocrit.5 We hypo- serum sodium (P ¼ 0.0667), but decreased with thesised that male CC homozygotes compared with serum total calcium (P ¼ 0.0456) and the amount of T allele carriers might have a more fragile erythro- alcohol consumed per day (P ¼ 0.0163). iMg (R2 ¼ cyte membrane and that alcohol intake might unveil 0.087) decreased with age (P ¼ 0.0272), but increa- this innate greater fragility of erythrocytes.5 In other sed with serum total calcium (P ¼ 0.0217). After words, the b-adducin CC genotype might invigorate adjustment for these covariates, findings in men the structural and functional haematological distur- for iK (CC versus T, 85.8 versus 87.3 mmol/l, bances associated with excessive alcohol intake. P ¼ 0.14) and iMg (1.91 versus 1.82 mmol/l, Higher intra-erythrocyte potassium concentration P ¼ 0.03) remained consistent. increases the hydration and osmotic stability of

Journal of Human Hypertension Intra-erythrocyte cations and adducin T Richart et al 391 Table 3 Intra-erythrocyte electrolyte concentrations by genotype (unadjusted)

a-Adducin allele b-Adducin allele

Characteristica GlyGly Trp P-valuew CC T P-valuew

Women Sodium (mmol/l cells) 6.4570.11 6.3770.13 0.66 6.4071.04 6.4970.86 0.91 Potassium (mmol/l cells) 87.6070.49 87.5970.57 0.99 87.5374.50 87.8873.59 0.93 Magnesium (mmol/l cells) 1.8770.02 1.8570.21 0.49 1.8670.20 1.8870.22 0.73

Men Sodium (mmol/l cells) 6.8271.01 6.9571.07 0.49 6.8671.04 6.8871.00 0.93 Potassium (mmol/l cells) 85.9470.51 86.4570.70 0.56 85.8074.33 87.4775.41 0.11 Magnesium (mmol/l cells) 1.9070.22 1.9070.22 0.85 1.9270.22 1.8070.19 0.01

Abbreviations: Gly, glycine; Trp, tryptophan. wP-values are for the significance of differences between genotypes. aPlus-minus values are mean7s.d. circulating erythrocytes.6 We therefore investigated haemoglobin level and haematocrit are influenced in a subsample of our previous study5 possible by menstrual blood loss and by the use of oral association between the intra-erythrocyte cation contraceptives or hormone replacement therapy. concentrations and the b-adducin C1797T poly- Fluctuations in female sex hormones owing to morphism. In the current study, male b-adducin menopausal status or menstrual cycle may mask CC homozygotes compared with T allele carriers haematological phenotypes associated with genetic had a slightly, albeit not significantly lower mutation of the b-adducin gene. intra-erythrocyte potassium concentration. These The present study has to be interpreted within the findings are in line with the above hypothesis.5 context of its limitations. We investigated a 13.9% The intra-erythrocyte potassium concentration is subsample of our previous study population.5 Our regulated via the K,Cl-cotransport and the Ca2 þ study lacked the statistical power to demonstrate activated K-channel, also known as the Gardos- the effects of the a-adducin Trp allele on BP and channel. Opening of the channel increases K þ efflux the intra-erythrocyte sodium concentration and the through the membrane by three orders of magni- effects of alcohol intake per se on erythrocyte tude.16,17 Magnesium inhibits the opening of the count, haemoglobin concentration and haematocrit.5 Gardos channel. Intra-erythrocyte magnesium repre- Although the P-value for the genotypic difference sents only 0.5% of the total body content of in intra-erythrocyte magnesium associated with the magnesium. Intra-erythrocyte magnesium and po- b-adducin polymorphism was 0.01, we cannot tassium are tightly regulated.18 This may explain exclude with certainty a type 1 error. Our present why, in the current study, we did not find any findings are therefore only hypothesis-generating association between the intra-erythrocyte concentra- and are currently being tested in a larger group of tions of magnesium and potassium. However, the never-treated hypertensive patients. positive correlation between the intra-erythrocyte potassium concentration and serum magnesium is consistent with the notion that magnesium inhibits the Gardos channel19 and might increase the What is known about the topic intra-erythrocyte potassium concentration. Another K Genetic variability in the ADD1 (Gly460Trp) and ADD2 (C1797T) subunits of the cytoskeleton protein adducin plays a potential mechanism affecting cytosolic magnesium role in the pathogenesis of hypertension, possibly via changes is the energy-requiring extrusion, which might in intracellular cation concentrations.1,3–5,13 be coupled to the movement of other cations. To K ADD2 1797CC homozygous men have decreased erythrocyte 5 what extent such ion exchange might underlie count and haematocrit. K Intra-erythrocyte cation concentrations affect the osmotic the positive correlation between intra-erythrocyte stability of membrane.6,16,17 magnesium and serum total calcium remains to be elucidated. Irrespective of the mechanism involved, What this study adds the difference among men in the intra-erythrocyte K ADD2 1797CC homozygous men exhibit slightly lower magnesium concentration between CC homozygotes concentrations of intra-erythrocyte potassium, higher magne- sium but similar sodium compared with T-allele carriers. and T-allele carriers is unlikely to contribute to the K In 136 women, none of the phenotype–genotype relations hypothesised higher innate fragility of erythrocytes reached significance. of CC homozygotes. K Changes in intra-erythrocyte cations in ADD2 1797CC Several factors might explain the absence of any homozygous men might lead to osmotic fragility of erythro- cytes, but to what extent they reflect systemic changes or are phenotype–genotype association in women in our possibly involved in blood pressure regulation remains 5 previous as well as in our current study. Haemato- unknown. logical phenotypes such as erythrocyte count,

Journal of Human Hypertension Intra-erythrocyte cations and adducin T Richart et al 392 Acknowledgements generated by maximal Gardos-channel activation in normal and sickle red blood cells. Blood 2005; 105: The Flemish Study on Environment, Genes and 361–367. Health Outcomes (FLEMENGHO) is part of the 7 Staessen AJ, Roels H, Emelianov D, Vangronsveld J, European Project on Genes in Hypertension Kuznetsova T, Thijs L et al. Environmental exposure (EPOGH), which is endorsed by the European to cadmium, forearm bone density, and risk of Council for Cardiovascular Research and the Euro- fractures: prospective population study. Lancet 1999; pean Society of Hypertension. FLEMENGHO would 353: 1140–1144. 8 Staessen AJ, Fagard R, Amery A. Life style as a not have been possible without the voluntary colla- determinant of blood pressure in the general popula- boration of participants and their general practi- tion. Am J Hypertens 1994; 7: 685–694. tioners. The municipality Hechtel-Eksel (Belgium) 9 Livak KJ, Flood SJ, Marmaro J, Giusti W, Deetz K. gave logistic support. Research included in the Oligonucleotides with fluorescent dyes at opposite present study was partially funded by the European ends provide a quenched probe system useful for Union (Grants IC15-CT98-0329-EPOGH and QLGI- detecting PCR product and nucleic acid hybridization. CT-2000-01137-EURNETGEN), the Fonds voor We- PCR Methods Appl 1995; 4: 357–362. tenschappelijk Onderzoek Vlaanderen, Ministry of 10 Brand E, Wang JG, Herrmann SM, Staessen JA. An the Flemish Community, Brussels, Belgium (Grants epidemiological study of blood pressure and metabolic b G.0424.03 and G.0575.06), the Katholieke Universi- phenotypes in relation to the G 3 C825T polymor- phism. J Hypertens 2003; 21: 729–737. teit Leuven, Leuven, Belgium (Grants OT/99/28, 11 Tripodi G, Casari G, Tisminetzky S, Bianchi G, OT/00/25 and OT/05/49). We gratefully acknowl- Devescovi G, Muro A et al. Characterisation and edge the technical assistance of Tamara Coenen, chromosomal localisation of the rat a- and b-adducin- Sandra Covens, Marie-Jeanne Jehoul, Lieve Lomme- encoding genes. Gene 1995; 166: 307–311. len, Yvette Piccart, Hanne Truyens, Katrien Staessen 12 Glorioso N, Filigheddu F, Troffa C, Soro A, Parpaglia and Renilde Wolfs (Studies Coordinating Centre, PP, Tsikoudakis A et al. Interaction of the a1-Na, Leuven, Belgium). K-ATPase and Na,K,2Cl-cotransporter genes in human essential hypertension. Hypertension 2001; 38: 204–209. References 13 Castejon AM, Alfieri AB, Hoffmann IS, Rathinavelu A, Cubeddu LX. Alpha-adducin polymorphism, salt 1 Muro AF, Marro ML, Gajovic S, Porro F, Luzzatto L, sensitivity, nitric oxide excretion, and cardiovascular Baralle FE. Mild spherocytic hereditary elliptocytosis risk factors in normotensive Hispanics. Am J Hyper- and altered levels of a- and g-adducins in b-adducin- tens 2003; 16: 1018–1024. deficient mice. Blood 2000; 95: 3978–3985. 14 Barlassina C, Norton GR, Samani NJ, Woodwiss AJ, 2 Porro F, Costessi L, Marro M, Baralle F, Muro A. The Candy GC, Radevski I et al. Adducin polymorphism in erythrocyte skeletons of beta-adducin deficient mice hypertensives of South African ancestry. Am J Hyper- have altered levels of tropomyosin, tropomodulin and tens 2000; 13: 719–723. EcapZ. FEBS Lett 2004; 576: 36–40. 15 Glorioso N, Filligheddu F, Cusi D, Troffa C, Conti M, 3 Castejon AM, Alfieri AB, Hoffmann IS, Rathinavelu A, Natalizio M et al. adducin 460Trp allele is associated Cubeddu LX. Alpha-adducin polymorphism, salt sen- with erythrocyte Na transport in North Sardinian sitivity, nitric oxide excretion, and cardiovascular risk primary hypertensives. Hypertension 2002; 39(part factors in normotensive Hispanics. Am J Hypertens 2): 357–362. 2003; 16: 1018–1024. 16 Gardos G. The function of calcium in the potassium 4 Tikhonoff V, Kuznetsova T, Stolarz K, Bianchi G, permeability of human erythrocytes. Biochim Biophys Casiglia E, Kawecka-Jaszcz K et al. B-adducin poly- Acta 1958; 30: 653–654. morphisms, blood pressure, and sodium excretion in 17 Lew V, Ferreira H. Calcium Transport and the Proper- three European population. Am J Hypertens 2003; 20: ties of a Calcium-Activated Potassium Channel in Red 840–846. Cell Membranes 1978. Academic Press: London, UK. 5 Wang JG, Barlassina C, Bianchi G, Fagard R, Zagato L, 18 Resnick LM, Gupta RK, Laragh JH. Intracellular free Staessen JA. Haematological phenotypes in relation to magnesium in eythrocytes of essential hypertension: the C1797T b-adducin polymorphism in a Caucasian Relation to blood pressure and serum divalent cations. population. Clin Sci 2003; 104: 369–376. Proc Natl Acad Sci USA 1984; 81: 6511–6515. 6 Lew V, Tiffert T, Etzion Z, Perdomo D, Daw N, 19 Simons TJ. Calcium-dependent potassium exchange in MacDonald L et al. Distribution of dehydration rates human red cell ghosts. J Physiol 1976; 256: 227–244.

Journal of Human Hypertension