Abstract
Gitelmańs syndrome (GS) is an inherited recessive disorder caused by homozygous or compound heterozygous loss of function mutations of the NaCl cotransporter (NCCT) gene encoding the kidney-expressed NCCT, the pharmacological target of thiazide diuretics. An observational study estimated the prevalence of GS to 19/1 000 000, in Sweden, suggesting that ∼1% of the population carries one mutant NCCT allele. As the phenotype of GS patients, who always carry two mutant alleles, is indistinguishable from that seen in patients treated with high-dose thiazide diuretics, we aimed at investigating whether subjects carrying one mutated NCCT allele have a phenotype resembling that of treatment with low-dose thiazide diuretics. We screened first-degree relatives of 18 of our patients with an established clinical end genetic diagnosis of GS for NCCT loss of function mutations and identified 35 healthy subjects carrying one mutant allele (GS-heterozygotes). Each GS-heterozygote was assigned a healthy control subject matched for age, BMI and sex. GS-heterozygotes had markedly lower blood pressure (systolic 103.3 ± 16.4 versus 123.2 ± 19.4 mmHg; diastolic 62.5 ± 10.5 versus 73.1 ± 9.4 mmHg; P < 0.001) than controls. There was no significant difference between the groups either in plasma concentration or urinary excretion rate of electrolytes, however, GS-heterozygotes had higher fasting plasma glucose concentration. Similar to patients being treated with low-dose thiazide diuretics, GS-heterozygotes have markedly lower blood pressure and slightly higher fasting plasma glucose compared with control subjects. Our findings suggest that GS-heterozygotes, the prevalence of which can be estimated to 1%, are partially protected from hypertension through partial genetic loss of function of the NCCT. However, as our study had a case–control design, it is important to underline that any potential effects on population blood pressure and risk of future cardiovascular disease need to be examined in prospective and population-based studies.
INTRODUCTION
Essential hypertension is a complex, polygenic disease leading to substantial cardiovascular structural damage, morbidity and mortality worldwide (1). No single gene has shown to be responsible by itself for essential hypertension, but several genes contribute to blood pressure maintenance and homeostasis (2). On the other hand, rare forms of Mendelian disorders because of mutations in genes coding for proteins in pathways that affect renal sodium reabsorption cause either severe hyper- or hypotension (3). Gitelman syndrome (GS), OMIM 263800, is an autosomal recessive disorder caused by homozygous or compound heterozygous mutations in the thiazide-sensitive NaCl-cotransporter (NCCT) gene (4). The gene is expressed in the apical cell membrane of the distal convolute tubule in the kidney. This renal segment is responsible for nearly 5% of the NaCl reabsorption in the kidney and the NCCT is also the molecular target of thiazide diuretics, one of the most effective and widely used antihypertensive agents (4). The genetically defective function of the NCCT in GS leads to hypotension, hypokalemia, hyperreninemic hyperaldosteronism, mild metabolic alkalosis, hypomagnesemia and hypocalciuria as the predominant features (4–6).
Research on the genetics of essential hypertension have focused most of the attention on common single nucleotide polymorphisms (SNPs) in genes implicated in obvious regulators of blood pressure homeostasis and the effect of each such SNP is expected to be small to moderate. Mutations in Mendelian recessive blood pressure disorders have much greater effects on blood pressure but are usually quite rare in the population. Gitelman's syndrome has an estimated prevalence of 19/1 000 000 in Sweden (7), suggesting a prevalence of heterozygous NCCT loss of function mutation carriers of ≈1% at the general population level. Since the clinical features of GS patients, who are homozygous or compound heterozygous for NCCT loss of function mutations, resemble that of patients treated by high-dose thiazide diuretic, we hypothesized that carriers of a single loss of function mutation of the NCCT gene (GS-heterozygotes) have a phenotype similar to patients treated by low-dose thiazide diuretics.
Thus the first aim of the present study was to assess the effect on blood pressure level of a single loss of function mutation in the NCCT gene. As thiazide treatment, apart from lowering blood pressure, is known to be associated with slight reductions of plasma potassium levels and elevations in fasting plasma glucose, the second aim of the study was to examine the effects of heterozygosity of NCCT loss of function mutations on electrolytes and glucometabolic parameters.
RESULTS
We found one NCCT loss of function mutation, corresponding to one of the two mutations present in the relative with GS, in 35 out of 36 potential heterozygous subjects who were screened. The single person who did not carry any GS mutation (a brother of a GS patient) and his matched control were excluded from the successive analyses. The seven different GS loss of function mutations of the NCCT are reported in Table 1. None of the control subjects carried any NCCT loss of function mutation.
Mutations detected in GS-heterozygotes
| Heterozygous | Family | GS mutations | Article |
|---|---|---|---|
| 1 | 1 | Gly741Arg (exon 18) | Simon et al. (4) |
| 2–8 | 2 | ||
| 9 | 1 | Gly439Ser (exon 10) | Mastroianni et al. (28) |
| 10–12 | 2 | ||
| 13 | 3 | ||
| 14 | 4 | c.2859+1G>T (intron 24) | Simon et al. (4) |
| 15–18 | 5 | ||
| 18–20 | 6 | ||
| 21–22 | 7 | ||
| 23–24 | 4 | Gly729Val (exon 18) | Cruz et al. (29) |
| 25–26 | 7 | ||
| 27–29 | 8 | Arg964Glu (exon 25) | Simon et al. (4) |
| 30–34 | 9 | Gly230Asp (exon 5) | Fava et al. (23) |
| 35 | 10 | c.1569-1G>A (intron 12) | Fava et al. (23) |
| 36 | 9 | No mutation found |
| Heterozygous | Family | GS mutations | Article |
|---|---|---|---|
| 1 | 1 | Gly741Arg (exon 18) | Simon et al. (4) |
| 2–8 | 2 | ||
| 9 | 1 | Gly439Ser (exon 10) | Mastroianni et al. (28) |
| 10–12 | 2 | ||
| 13 | 3 | ||
| 14 | 4 | c.2859+1G>T (intron 24) | Simon et al. (4) |
| 15–18 | 5 | ||
| 18–20 | 6 | ||
| 21–22 | 7 | ||
| 23–24 | 4 | Gly729Val (exon 18) | Cruz et al. (29) |
| 25–26 | 7 | ||
| 27–29 | 8 | Arg964Glu (exon 25) | Simon et al. (4) |
| 30–34 | 9 | Gly230Asp (exon 5) | Fava et al. (23) |
| 35 | 10 | c.1569-1G>A (intron 12) | Fava et al. (23) |
| 36 | 9 | No mutation found |
Heterozygous subjects are numbered consecutively from 1 to 35.
Mutations detected in GS-heterozygotes
| Heterozygous | Family | GS mutations | Article |
|---|---|---|---|
| 1 | 1 | Gly741Arg (exon 18) | Simon et al. (4) |
| 2–8 | 2 | ||
| 9 | 1 | Gly439Ser (exon 10) | Mastroianni et al. (28) |
| 10–12 | 2 | ||
| 13 | 3 | ||
| 14 | 4 | c.2859+1G>T (intron 24) | Simon et al. (4) |
| 15–18 | 5 | ||
| 18–20 | 6 | ||
| 21–22 | 7 | ||
| 23–24 | 4 | Gly729Val (exon 18) | Cruz et al. (29) |
| 25–26 | 7 | ||
| 27–29 | 8 | Arg964Glu (exon 25) | Simon et al. (4) |
| 30–34 | 9 | Gly230Asp (exon 5) | Fava et al. (23) |
| 35 | 10 | c.1569-1G>A (intron 12) | Fava et al. (23) |
| 36 | 9 | No mutation found |
| Heterozygous | Family | GS mutations | Article |
|---|---|---|---|
| 1 | 1 | Gly741Arg (exon 18) | Simon et al. (4) |
| 2–8 | 2 | ||
| 9 | 1 | Gly439Ser (exon 10) | Mastroianni et al. (28) |
| 10–12 | 2 | ||
| 13 | 3 | ||
| 14 | 4 | c.2859+1G>T (intron 24) | Simon et al. (4) |
| 15–18 | 5 | ||
| 18–20 | 6 | ||
| 21–22 | 7 | ||
| 23–24 | 4 | Gly729Val (exon 18) | Cruz et al. (29) |
| 25–26 | 7 | ||
| 27–29 | 8 | Arg964Glu (exon 25) | Simon et al. (4) |
| 30–34 | 9 | Gly230Asp (exon 5) | Fava et al. (23) |
| 35 | 10 | c.1569-1G>A (intron 12) | Fava et al. (23) |
| 36 | 9 | No mutation found |
Heterozygous subjects are numbered consecutively from 1 to 35.
(As a consequence of the pair-wise matching, age (41.9 ± 16.9 versus 42.2 ± 16.5 years), sex (18 males/17 females) and BMI (23.9 ± 3.4 versus 23.6 ± 3.1 kg/m2) were similar, respectively, in GS-heterozygotes and controls (P > 0.05 for all).
GS-heterozygotes had highly significantly lower blood pressure for both systolic (103.3 ± 16.4 versus 123.2 ± 19.4 mmHg) and diastolic (62.5 ± 10.5 versus 73.1 ± 9.4 mmHg) compared with controls (Fig. 1). Furthermore, we found that GS-heterozygotes had higher glomerular filtration rate (GFR) and higher fasting plasma glucose concentration when compared with matched controls, whereas there was no significant difference in plasma concentration and urinary excretion of electrolytes also when the latter were expressed as a function of creatinine excretion (Table 2).
Systolic (A) and diastolic (B) blood pressure levels in GS-heterozygous subjects and sex/age/BMI matched controls.
Metabolic parameters, hormones, plasma and urinary electrolytes of the GS-heterozygous subjects and sex/age/BMI matched controls
| Variables | GS-heterozygotes | Controls | P-value |
|---|---|---|---|
| p-Glucose, mmol/l | 5.4 ± 0.8 | 4.8 ± 0.6 | <0.001 |
| p-Creatinine, μmol/l | 70.7 ± 12.8 | 79.5 ± 12.8 | 0.005 |
| s-Aldosterone, nmol/l | 0.28 ± 0.27 | 0.44 ± 0.38 | 0.004 |
| s-Renin, mU/l | 37.3 ± 27.3 | 26.1 ± 21.5 | 0.009 |
| s-Insulin, mU/l | 5.1 ± 2.5 | 5.6 ± 2.8 | 0.39 |
| HOMA indexa | 1.3 ± 0.7 | 1.2 ± 0.6 | 0.97 |
| GFR, ml/min | 81.5 ± 30.6 | 74.1 ± 31.8 | 0.03 |
| p-Sodium, mmol/l | 139.7 ± 2.8 | 139.9 ± 1.7 | 0.85 |
| p-Magnesium, mmol/l | 0.8 ± 0.07 | 0.9 ± 0.06 | 0.25 |
| p-Potassium, mmol/l | 3.8 ± 0.26 | 3.7 ± 0.27 | 0.44 |
| p-Calcium, mmol/l | 2.31 ± 0.08 | 2.36 ± 0.25 | 0.38 |
| p-Cholesterol, mmol/l | 4.9 ± 1.0 | 5.2 ± 1.2 | 0.29 |
| p-Triglycerides, mmol/l | 0.9 ± 0.5 | 1.2 ± 1.2 | 0.37 |
| u-Creatinine, μmol/24 h | 13.1 ± 5.3 | 13.4 ± 3.5 | 0.67 |
| u-Sodium, mmol/24 h | 164.8 ± 88.0 | 155.9 ± 53.9 | 0.52 |
| u-Sodium/u-creatinine | 12.7 ± 2.9 | 11.7 ± 4.2 | 0.37 |
| u-Potassium, mmol/24 h | 69.4 ± 29.1 | 69.7 ± 21.6 | 0.41 |
| u-Potassium/u-creatinine | 5.4 ± 1.6 | 5.2 ± 1.6 | 0.71 |
| u-Calcium, mmol/l | 4.9 ± 2.3 | 4.6 ± 2.2 | 0.44 |
| u-Calcium/u-creatinine | 0.4 ± 0.17 | 0.34 ± 0.13 | 0.12 |
| Variables | GS-heterozygotes | Controls | P-value |
|---|---|---|---|
| p-Glucose, mmol/l | 5.4 ± 0.8 | 4.8 ± 0.6 | <0.001 |
| p-Creatinine, μmol/l | 70.7 ± 12.8 | 79.5 ± 12.8 | 0.005 |
| s-Aldosterone, nmol/l | 0.28 ± 0.27 | 0.44 ± 0.38 | 0.004 |
| s-Renin, mU/l | 37.3 ± 27.3 | 26.1 ± 21.5 | 0.009 |
| s-Insulin, mU/l | 5.1 ± 2.5 | 5.6 ± 2.8 | 0.39 |
| HOMA indexa | 1.3 ± 0.7 | 1.2 ± 0.6 | 0.97 |
| GFR, ml/min | 81.5 ± 30.6 | 74.1 ± 31.8 | 0.03 |
| p-Sodium, mmol/l | 139.7 ± 2.8 | 139.9 ± 1.7 | 0.85 |
| p-Magnesium, mmol/l | 0.8 ± 0.07 | 0.9 ± 0.06 | 0.25 |
| p-Potassium, mmol/l | 3.8 ± 0.26 | 3.7 ± 0.27 | 0.44 |
| p-Calcium, mmol/l | 2.31 ± 0.08 | 2.36 ± 0.25 | 0.38 |
| p-Cholesterol, mmol/l | 4.9 ± 1.0 | 5.2 ± 1.2 | 0.29 |
| p-Triglycerides, mmol/l | 0.9 ± 0.5 | 1.2 ± 1.2 | 0.37 |
| u-Creatinine, μmol/24 h | 13.1 ± 5.3 | 13.4 ± 3.5 | 0.67 |
| u-Sodium, mmol/24 h | 164.8 ± 88.0 | 155.9 ± 53.9 | 0.52 |
| u-Sodium/u-creatinine | 12.7 ± 2.9 | 11.7 ± 4.2 | 0.37 |
| u-Potassium, mmol/24 h | 69.4 ± 29.1 | 69.7 ± 21.6 | 0.41 |
| u-Potassium/u-creatinine | 5.4 ± 1.6 | 5.2 ± 1.6 | 0.71 |
| u-Calcium, mmol/l | 4.9 ± 2.3 | 4.6 ± 2.2 | 0.44 |
| u-Calcium/u-creatinine | 0.4 ± 0.17 | 0.34 ± 0.13 | 0.12 |
P-value has been computed by t-test for independent samples. Data on electrolytes excretion have been presented as mmol/24 h and mmol/μmol creatinine.
BMI, body mass index; GFR, glomerular filtration rate.
aHOMA index has been computed as follows: p-glucose (mmol/l) × s-insulin (mU/l)/22.5.
Metabolic parameters, hormones, plasma and urinary electrolytes of the GS-heterozygous subjects and sex/age/BMI matched controls
| Variables | GS-heterozygotes | Controls | P-value |
|---|---|---|---|
| p-Glucose, mmol/l | 5.4 ± 0.8 | 4.8 ± 0.6 | <0.001 |
| p-Creatinine, μmol/l | 70.7 ± 12.8 | 79.5 ± 12.8 | 0.005 |
| s-Aldosterone, nmol/l | 0.28 ± 0.27 | 0.44 ± 0.38 | 0.004 |
| s-Renin, mU/l | 37.3 ± 27.3 | 26.1 ± 21.5 | 0.009 |
| s-Insulin, mU/l | 5.1 ± 2.5 | 5.6 ± 2.8 | 0.39 |
| HOMA indexa | 1.3 ± 0.7 | 1.2 ± 0.6 | 0.97 |
| GFR, ml/min | 81.5 ± 30.6 | 74.1 ± 31.8 | 0.03 |
| p-Sodium, mmol/l | 139.7 ± 2.8 | 139.9 ± 1.7 | 0.85 |
| p-Magnesium, mmol/l | 0.8 ± 0.07 | 0.9 ± 0.06 | 0.25 |
| p-Potassium, mmol/l | 3.8 ± 0.26 | 3.7 ± 0.27 | 0.44 |
| p-Calcium, mmol/l | 2.31 ± 0.08 | 2.36 ± 0.25 | 0.38 |
| p-Cholesterol, mmol/l | 4.9 ± 1.0 | 5.2 ± 1.2 | 0.29 |
| p-Triglycerides, mmol/l | 0.9 ± 0.5 | 1.2 ± 1.2 | 0.37 |
| u-Creatinine, μmol/24 h | 13.1 ± 5.3 | 13.4 ± 3.5 | 0.67 |
| u-Sodium, mmol/24 h | 164.8 ± 88.0 | 155.9 ± 53.9 | 0.52 |
| u-Sodium/u-creatinine | 12.7 ± 2.9 | 11.7 ± 4.2 | 0.37 |
| u-Potassium, mmol/24 h | 69.4 ± 29.1 | 69.7 ± 21.6 | 0.41 |
| u-Potassium/u-creatinine | 5.4 ± 1.6 | 5.2 ± 1.6 | 0.71 |
| u-Calcium, mmol/l | 4.9 ± 2.3 | 4.6 ± 2.2 | 0.44 |
| u-Calcium/u-creatinine | 0.4 ± 0.17 | 0.34 ± 0.13 | 0.12 |
| Variables | GS-heterozygotes | Controls | P-value |
|---|---|---|---|
| p-Glucose, mmol/l | 5.4 ± 0.8 | 4.8 ± 0.6 | <0.001 |
| p-Creatinine, μmol/l | 70.7 ± 12.8 | 79.5 ± 12.8 | 0.005 |
| s-Aldosterone, nmol/l | 0.28 ± 0.27 | 0.44 ± 0.38 | 0.004 |
| s-Renin, mU/l | 37.3 ± 27.3 | 26.1 ± 21.5 | 0.009 |
| s-Insulin, mU/l | 5.1 ± 2.5 | 5.6 ± 2.8 | 0.39 |
| HOMA indexa | 1.3 ± 0.7 | 1.2 ± 0.6 | 0.97 |
| GFR, ml/min | 81.5 ± 30.6 | 74.1 ± 31.8 | 0.03 |
| p-Sodium, mmol/l | 139.7 ± 2.8 | 139.9 ± 1.7 | 0.85 |
| p-Magnesium, mmol/l | 0.8 ± 0.07 | 0.9 ± 0.06 | 0.25 |
| p-Potassium, mmol/l | 3.8 ± 0.26 | 3.7 ± 0.27 | 0.44 |
| p-Calcium, mmol/l | 2.31 ± 0.08 | 2.36 ± 0.25 | 0.38 |
| p-Cholesterol, mmol/l | 4.9 ± 1.0 | 5.2 ± 1.2 | 0.29 |
| p-Triglycerides, mmol/l | 0.9 ± 0.5 | 1.2 ± 1.2 | 0.37 |
| u-Creatinine, μmol/24 h | 13.1 ± 5.3 | 13.4 ± 3.5 | 0.67 |
| u-Sodium, mmol/24 h | 164.8 ± 88.0 | 155.9 ± 53.9 | 0.52 |
| u-Sodium/u-creatinine | 12.7 ± 2.9 | 11.7 ± 4.2 | 0.37 |
| u-Potassium, mmol/24 h | 69.4 ± 29.1 | 69.7 ± 21.6 | 0.41 |
| u-Potassium/u-creatinine | 5.4 ± 1.6 | 5.2 ± 1.6 | 0.71 |
| u-Calcium, mmol/l | 4.9 ± 2.3 | 4.6 ± 2.2 | 0.44 |
| u-Calcium/u-creatinine | 0.4 ± 0.17 | 0.34 ± 0.13 | 0.12 |
P-value has been computed by t-test for independent samples. Data on electrolytes excretion have been presented as mmol/24 h and mmol/μmol creatinine.
BMI, body mass index; GFR, glomerular filtration rate.
aHOMA index has been computed as follows: p-glucose (mmol/l) × s-insulin (mU/l)/22.5.
The concentration of renin was higher in GS-heterozygous subjects, whereas the concentration of aldosterone was higher in healthy controls. There was a positive correlation between renin and aldosterone (r = 0.42, P < 0.001). Neither aldosterone nor renin or potassium correlated with glucose level in GS-heterozygotes (P > 0.05 for all).
None of the investigated subjects had diabetes, but nine GS-heterozygotes versus one control subject had impaired fasting glucose (IFG; P = 0.014). Among GS-heterozygous subjects, we did not find any difference in aldosterone level (0.29 ± 0.26 versus 0.27 ± 0.32 nmol/l; P = 0.84), renin level (36.9 ± 29.1 versus 38.3 ± 22.7 mU/l; P = 0.9), insulin (4.6 ± 2 versus 6.3 ± 3.3 mU/l; P = 0.08) or potassium (3.7 ± 0.3 versus 3.8 ± 0.2 mmol/l; P = 0.45) in subjects with IFG as compared with subjects without IFG. However, there was a strong inverse relation between potassium and fasting glucose level in GS-heterozygotes with IFG (rs = −0.78; P = 0.012; Fig. 2).
Bivariate correlation between plasma glucose and potassium level in GS-heterozygous subjects with impaired fasting glucose.
DISCUSSION
We found that GS-heterozygous subjects had significantly lower blood pressure and slightly higher fasting plasma glucose level when compared with matched control subjects without NCCT mutations.
This result is partially in agreement with a previous report from Cruz et al. (8), who studied one large Amish kindred. In this study there was no difference in blood pressure between carriers and non-carriers of NCCT loss of function mutations, but a higher urinary Na+/creatinine ratio in heterozygous subjects than that in wild-type subjects, interpreted as a self-selected increase salt intake in GS-heterozygotes to maintain normal level of blood pressure (8). Within the same kindred heterozygous children, but not adults, had significantly lower blood pressure than those of the wild-type relatives probably because of the fact that younger individuals have less opportunity to self-selected salt intake. At variance with the findings of that study, we did not find any difference in 24-h urinary Na+ excretion between GS-heterozygous subjects and controls. The urinary Na+/creatinine ratio in GS-heterozygotes and controls is comparable with that in wild-type subjects in the study of Cruz. Thus it is likely that the compensatory increase in salt intake suggested by these authors was not practiced by Swedish GS-heterozygous subjects resulting in significantly lower blood pressure levels.
In general, the pathophysiologic link between salt and blood pressure is predictable from the relationship between salt intake, renal tubular sodium reabsorption and vascular volume homeostasis. Increased net renal salt reabsorption necessitates increased water reabsorption to maintain plasma sodium concentration at steady state. The resulting increased intravascular volume augments venous blood return to the heart, thereby raising cardiac output and leading directly to elevated blood pressure. In consequence, reduced reabsorption of salt decreases blood volume and blood pressure (3).
Human GS-homozygous patients are characterized by low plasma level of potassium, magnesium and hypocalciuria. In line with the results reported by Cruz et al. (8), we found that GS-heterozygous subjects have similar plasma concentration of potassium and magnesium and urinay excretion of potassium; however, in contrast to the previous study we found no tendency towards hypocalciuria (Table 2).
There are several explanations for the development of hypokalaemia in GS. First of all hypokalemia is caused by secondary hyperaldosteronism (4) but studies in humans (9) and animals (10,11) suggest that plasma magnesium and dietary potassium can also contribute to potassium levels. In four out of six patients with GS the trans tubular K+ gradient (an index of potassium wasting through the kidney) was corrected after normalization of plasma magnesium (9), whereas in animal models of GS with targeted disruption of the NCCT, hypokalaemia was present only when a low potassium diet is administrated (10,11).
Other studies using the same NCCT−/− animal model have tried to elucidate the cause for hypocalciuria and hypomagnesemia typical of GS. It was demonstrated that a passive proximal reabsorption of Ca2+ together with Na+ is responsible for hypocalciuria (12,13), whereas the down regulation of the epithelial Mg2+ channel transient receptor potential channel subfamily M, member 6 (Trpm6) contribute to hypomagnesaemia (13).
In any case it is worth emphasizing that despite arguments for important roles of other electrolytes such as K+ and Ca2+ in blood pressure determination, for all the Mendelian traits studied, the single factor that consistently predicts the direction of blood pressure is net sodium balance. Thus, Mendelian hyper- and hypotensive syndromes can both feature either hypo- or hyperkalemia. Similar considerations apply to calcium handling. In all cases, however, the direction of blood pressure is predicted by the effect on net sodium balance (3).
Interestingly, we found a small but significant increase in fasting plasma glucose concentration in GS-heterozygous subjects compared with subjects without NCCT mutations. Thiazide diuretics, at low dose, has a small but significant glucose-elevating effect (14,15) and it is generally thought that the cause of this is a reduction in insulin sensitivity, which in turn is attributable to decreased plasma potassium concentration (16). In our study, the elevated glucose concentration in GS-heterozygous was not accompanied by reduced plasma potassium concentration arguing against hypokalaemia as the cause of increases in plasma glucose in our subjects as well as in patients treated with thiazide diuretics. Interestingly, however, potassium level was inversely correlated to glucose level in GS-heterozygous subjects with IFG suggesting that low potassium concentration may explain at least a part of the glucose elevations seen in GS-heterozygous subjects.
Other mechanisms has been claimed to explain impaired insulin sensitivity in thiazide-treated patients (17) and in particular an active role for the renin–angiotensin–aldosterone system (RAS) has been suggested (17,18). Moreover, the beneficial effect on glucose tolerance achieved by blocking the RAS in hypertensive patients (19) supports this hypothesis. In our sample plasma renin was higher and aldosterone lower in GS-heterozygotes when compared with matched controls. Nevertheless, the lack of difference in HOMA index between GS-heterozygous subjects and controls, and the lack of correlation between renin, aldosterone and fasting glucose in GS-heterozygous subjects as well as in the whole population argues against the hypothesis that elevated RAAS activity and insulin resistance would be the cause of elevated glucose in GS-heterozygous subjects.
Long-term cardiovascular consequences of the marked blood pressure reduction and the slight elevation in plasma glucose seen in GS-heterozygotes remain to be shown. Interestingly, a long-term follow-up of the Systolic Hypertension in Elderly People study showed that treatment with a low dose of thiazide diuretic (chlorthalidone) had beneficial effects on total and cardiovascular mortality. Although the presence of diabetes at baseline and diabetes that developed in the placebo-treated arm of the study were associated with an increased risk of cardiovascular events and total mortality, thiazide-induced diabetes had no effect on cardiovascular and total mortality (20). We therefore speculate that the markedly lower blood pressure seen in GS-heterozygous subjects is protective from future cardiovascular diseases, and that the effect of their higher level of plasma glucose is either harmless or negligible in relation to the markedly lower blood pressure when it comes to future cardiovascular risk.
In this study GS-heterozygotes had a slightly higher GFR with respect to controls. This finding is unexpected since GS patients as well as GS-heterozygotes are considered to have no alterations in kidney function (8), and also short-term treatment with thiazide diuretics is not associated with altered kidney function (21,22). It could be speculated that the low level of blood pressure may have better preserved kidney function in GS-heterozygotes.
Given the GS prevalence of 19 per million (7), ∼1% of the population can be expected to be GS-heterozygotes. Given the markedly lower blood pressure in GS-heterozygotes, NCCT loss of function mutations may have a significant effect even on population blood pressure.
However, as our study had a case–control design, it is important to underline that any potential effects on population blood pressure and risk of future cardiovascular disease need to be examined in prospective and population-based studies.
Moreover, some of GS-heterozygotes comes from the same family and the difference in blood pressure level, and the other variables between cases and controls could be partially explained by the within-family correlation of the analysed variables. On the other hand, the linear regression analysis we applied permits to take into account the familial clustering and adjust for this confounder adding safety to our results.
In conclusion, similar to patients being treated with low-dose thiazide diuretics, GS-heterozygotes have markedly lower blood pressure and slightly higher fasting plasma glucose compared with control subjects. Any potential effects on population blood pressure and risk of future cardiovascular disease need to be examined in prospective and population-based studies.
MATERIALS AND METHODS
Subjects and phenotyping
We collected 30 patients from Southern Sweden originating from 25 families with the clinical diagnosis of GS and found 24 different mutations in the NCCT gene (23). In 28 out of 30 GS patients (93%), we identified two mutations (homozygosity or compound heterozygosity) (23).
Thereafter, we collected 36 healthy first-degree relatives from 10 of the 18 families of GS patients potentially being GS-heterozygotes and matched them pair-wise (sex, age ± 3 years and BMI ± 1.5 kg/m2) with 36 healthy controls who all lacked a family history of GS and were participants or relatives of participants in either the Malmö Diet and Cancer study or the Malmö preventive Project, both of which are Swedish population-based materials (24,25). In the controls we sequenced all the exons and intronic boundaries of the entire NCCT gene to exclude sporadic GS mutations, whereas in potential GS-heterozygotes we sequenced only the NCCT gene exons and intronic boundaries in which the relative with GS carried mutations. All potential GS-heterozygotes (n = 36) and matched controls (n = 36) underwent a phenotyping programme in the morning after an overnight fast.
After 10 min of supine rest in a quiet room, blood pressure was measured three times in the right arm at heart level in supine position using an ISO-STABIL 5 device (Speidel & Keller, Jungingen, Germany) with a tricuff manchett (AJ Medical, Lidingö, Sweden) and the mean of the three blood pressure measurements was recorded. Korotkoff sounds corresponding to ‘phase I’ was used to define the systolic and ‘phase V’ the diastolic blood pressure.
BMI was calculated after height and weight were recorded in light clothing without shoes. Electrolytes, lipid fractions and glucose were measured in fasting plasma samples and urinary electrolyte excretion was measured from 24-hour urine collections.
The diagnosis of IFG and diabetes has been based on WHO criteria (26). Plasma aldosterone concentration was measured by RIA diagnostic kits (Abbott Laboratories). The serum insulin concentrations were measured with a specific ELISA (DAKO). Renin was measured as direct active renin using the Nichols Diagnostics chemiluminescent immune assay on the automated Nichols Advantage System (San Juan, CA, USA).
All study participants had given written informed consent, and the ethics committee of the Medical Faculty of Lund University approved the study.
Mutation screening
Total genomic DNA was extracted from venous blood by standard methods (27). Sequencing of the 26 exons and exon–intron boundaries of the NCCT gene was performed with polymerase chain reaction (PCR) with primers published by Simon et al. (4) except for exon 17, for which primers published by Mastroianni et al. (28), were used. Exon 1 was amplified in two fragments (exons 1A and 1B). PCRs were performed with 50 ng genomic DNA in a total volume of 20 µl containing 10 pmol of each primer, 2 nmol dNTPs and 0.5 U Taq polymerase (Pharmacia Biotech) in either the PCR buffer recommended by the manufacturer (Pharmacia Biotech) (exons 1B, 2, 4, 6, 12, 14 to 15, 17 to 18, 20 to 21 and 24) or in 1x (NH4)2SO4-buffer (16 mmol/l (NH4)2SO4; 67 mmol/l Tris pH 8.8; 0.01% Tween) (exons 1A, 3, 5, 7–11, 13, 16, 19, 22 to 23 and 25 to 26). Reactions were performed with 1.5% formamide in either 1.5 mmol/l MgCl2 (exons 1–10, 12–17 and 19–26) or 3.0 mmol/l MgCl2 (exons 11 and 18). PCR conditions were as follows: initial denaturation at 94°C for 5 min, followed by 30 cycles of denaturation (94°C for 30 s), annealing (60°C for exons 8, 10 and 15; 62°C for exons 1A, 3, 7, 16, 17, 19, 22 and 25, 26; 64°C for exons 1B, 2, 5, 13–14, 18 and 23, 24 and 66°C for exons 4, 6, 9, 11, 12, 20 and 21; for 30 s), and extension (72° for 30 s), with the final extension at 72° for 10 min.
The PCR products were sequenced bidirectionally with the Thermo Sequenase II dye terminator cycle sequencing kit (Pharmacia Biotech) in an ABI PRISM 3100 automated DNA Sequencer (Perkin Elmer). Every mutation detected was confirmed in a new PCR followed by DNA sequencing reaction.
STATISTICS
All data were analysed with STATA statistical software (StataCorp, College Station, TX, USA). Frequency differences were analysed by χ2-test and Fisher exact test when appropriate. Continuous variables are presented as mean ± standard deviation (SD). Significance of differences in all the continuous variables was tested by using a linear regression model with genotype as the explanatory variable and using the robust option to model the familial clustering. Variables were not normally distributed where log transformed.
The Spearman coefficient (rs) was calculated to test for correlation between variables. All tests were two-sided and throughout P < 0.05 was considered statistically significant.
Conflict of Interest statement. No author has any financial interests or connections, direct or indirect, or other situations that might raise the question of bias in the work reported or the conclusions, implications, or opinions stated—including pertinent commercial or other sources of funding for the authors or for the associated departments, personal relationships or direct academic competition.
FUNDING
This study was supported by grants from the Swedish Medical Research Council, the Swedish Heart and Lung Foundation, the Medical Faculty of Lund University, Malmö University Hospital, the Albert Påhlsson Research Foundation, the Crafoord Foundation, the Ernhold Lundströms Research Foundation, the Region Skane, Hulda and Conrad Mossfelt Foundation and King Gustaf V and Queen Victoria Foundation and the Lennart Hanssons Memorial Fund.
