Article

• The EMBO Journal (2008) 27, 179 - 187
• doi:10.1038/sj.emboj.7601934

Published online: 22 November 2007

• Subject Categories:

  • Membranes and Transport | Cellular Metabolism

Invalidation of TASK1 potassium channels disrupts adrenal gland zonation and mineralocorticoid homeostasis

Dirk Heitzmann^1,2, Renaud Derand³, Stefan Jungbauer¹, Sascha Bandulik¹, Christina Sterner¹, Frank Schweda¹, Abeer El Wakil³, Enzo Lalli³, Nicolas Guy³, Raymond Mengual⁴, Markus Reichold¹, Ines Tegtmeier¹, Saïd Bendahhou³, Celso E Gomez-Sanchez⁵, M Isabel Aller⁶, William Wisden⁷, Achim Weber⁸, Florian Lesage³, Richard Warth^1,9 and Jacques Barhanin^3,9

1. Institute of Physiology, University of Regensburg, Regensburg, Germany
2. Clinic and Policlinic for Internal Medicine II, University of Regensburg, Regensburg, Germany
3. Institut de Pharmacologie Moléculaire et Cellulaire, CNRS and Université de Nice Sophia Antipolis, Valbonne Sophia Antipolis, France
4. Centre Hospitalo-Universitaire de Nice, Nice, France
5. Division of Endocrinology, GV Montgomery VA Medical Center, Jackson, MS, USA
6. Instituto de Neurociencias de Alicante, Consejo Superior de Investigaciones Cientificas-Universidad Miguel Hernández, San Juan de Alicante, Spain
7. Institute of Medical Sciences, University of Aberdeen, Aberdeen, UK
8. Department of Pathology, University Hospital Zurich, Zurich, Switzerland
9. These authors contributed equally to this work

Correspondence to:

Richard Warth, Institute of Physiology, University of Regensburg, Universitaetsstrasse 31, NWF III—VKL, Regensburg 93053, Germany. Tel.: +49 941 943 2894; Fax: +49 941 943 2896; E-mail: richard.warth@vkl.uni-regensburg.de

Received 25 May 2007; Accepted 2 November 2007

• Keywords:

  • aldosterone,
  • arterial hypertension,
  • KCNK3,
  • KCNK9,
  • TASK-1

Introduction

The mineralocorticoid aldosterone is the major regulator of extracellular fluid and salt balance thereby controlling arterial blood pressure. The major target organ of aldosterone is the kidney, where it promotes Na⁺ retention and K⁺ secretion in the aldosterone-sensitive distal nephron. Dysregulation of aldosterone secretion leading to hyperaldosteronism is causal for about 3% of the cases of arterial hypertension (Nussberger, 2003). Moreover, aldosterone contributes to several aspects of cardiac fibrosis, cardiovascular dysfunction as well as progressive renal disease (Ibrahim and Hostetter, 2003; Remuzzi et al, 2005). The importance of aldosterone as clinical risk factor has been highlighted by recent clinical trials (Aldosterone Evaluation Study, (RALES); EPlerenone Heart failure and SUrvival Study (EPHESUS)) demonstrating the beneficial effects of mineralocorticoid receptor antagonists in patients with heart failure (Chai and Danser, 2006).

The mechanisms regulating aldosterone secretion in glomerulosa cells of the adrenal cortex have been studied for already half a century. The most important physiological stimuli for aldosterone secretion are angiotensin II, high plasma K⁺, and, for a minor part, ACTH. Binding of angiotensin II to AT1 receptors stimulates phospholipase C, which in turn triggers InsP₃-dependent Ca²⁺ release from intracellular stores. The immediate phase of aldosterone secretion is followed by sustained secretion, which is dependent on membrane depolarization and activation of T- and L-type Ca²⁺ channels. High plasmatic K⁺ concentrations stimulate exclusively the sustained phase of aldosterone secretion via influx of Ca²⁺ through voltage-dependent Ca²⁺ channels (Lotshaw, 2001). In this respect, the glomerulosa cell appears to be a unique sensor of extracellular K⁺ (Spat and Hunyady, 2004): increases of the extracellular K⁺ concentration by approximately 1 mmol/l are sufficient to double aldosterone secretion; maximal secretion occurs at 8 mmol/l of extracellular K⁺ (Lotshaw, 2001). The basis for this unique sensitivity for plasma potassium concentration is a very high background K⁺ conductance, which makes the membrane voltage strictly follow the K⁺ equilibrium potential: at low extracellular K⁺, the membrane voltage is hyperpolarized; upon relatively small increases in extracellular K⁺, membrane depolarization between -80 and -70 mV suffices to activate T-type Ca²⁺ channels in rat glomerulosa cells (Szabadkai et al, 1999; Lotshaw, 2001).

The relative contributions of molecularly identified K⁺ channels to membrane depolarization and stimulation of aldosterone secretion have not been fully elucidated. Studies from genetically modified mice have pointed to a role of voltage-dependent KCNE1/KCNQ1 channels for the regulation of aldosterone secretion (Arrighi et al, 2001). Moreover, Ca²⁺-activated MaxiK channels have been proposed as K⁺ channels limiting aldosterone secretion (Sausbier et al, 2005). These K⁺ channels are not constitutively open at rest and therefore cannot underlie the background conductance necessary to confer the exquisite extracellular K⁺ sensitivity to glomerulosa cells. Electrophysiological recordings of glomerulosa cells have emphasized the importance for leak-type K⁺ channels of the 2P domain (K2P) family (Czirjak et al, 2000). The human K2P channel family comprises 15 different members, which are characterized by their typical structure of four transmembrane domains and a tandem of two pore-forming loops (Lesage and Lazdunski, 2000). These channels exhibit little voltage dependence, remain open at negative membrane potentials, and can be modulated by a variety of physical and chemical stimuli such as external pH, membrane stretch, protein kinase A- or protein kinase C-dependent phoshorylation, and PiP₂ membrane depletion. Recently, task1 (KCNK3) and task3 (KCNK9), two K2P channels, were reported to be the dominant background channels in rat glomerulosa cells (Czirjak et al, 2000; Czirjak and Enyedi, 2002). In bovine glomerulosa cells, trek1, another K2P channel, is thought to be a major channel setting the membrane potential (Enyeart et al, 2004). The situation appears confusing mainly because the channels are poorly characterized beyond electrophysiological methods. However, a pivotal role of TASK1 is strongly suggested by the unusual high abundance of its messenger RNA in adrenal glands of several species including human and mouse (http://symatlas.gnf.org/SymAtl
as/).

In this study, we have investigated the specific contribution of task1 K⁺ channels to the control of aldosterone secretion using the task1 knockout (task1^-/-) mouse as a model. We provide evidence that task1^-/- mice have a strongly impaired mineralocorticoid homeostasis resulting in salt retention, arterial hypertension associated with low plasma renin activity. Surprisingly, this phenotype was restricted to the female gender. This pathology was caused by abnormal zonation of the adrenal cortex in female mice. In young male mice, the same zonation defect was observed, which however regressed after puberty. Age- and gender-dependent task3 expression might rescue the adrenal gland abnormalities.

Plasma aldosterone, renin, and K⁺ levels

On normal salt diet, adult task1^-/- mice had a higher plasma aldosterone concentration than wild-type and heterozygous mice (task1^+/+ 39473 ng/l, n=52; task1^+/- 23544 ng/l, n=22; task1^-/- 2361452 ng/l, n=55). This hyperaldosteronism was restricted to the female gender as males displayed normal aldosterone values (Figure 1). To assess the response upon physiological regulators of aldosterone secretion, the salt intake was modified by varying the salt content of the diet. The response to high and low K⁺ intake as well as low Na⁺ was similar in heterozygous and wild-type mice of either sex. By contrast, female task1^-/- mice exhibited a total loss of the physiological control of aldosterone secretion; plasma aldosterone concentration was neither stimulated by high K⁺ and low Na⁺ diet nor was it decreased by low K⁺ diet (Figure 1A). In female task1^-/- mice, invariable high plasma concentrations of aldosterone were paralleled by reduced plasma renin activity (Figure 2A), which is indicative for primary hyperaldosteronism. Aldosterone stimulates K⁺ secretion and Na⁺ reabsorption in its target tissues. In female mice, plasma K⁺ concentration was reduced (Figure 2B); plasma Na⁺ concentration was not different between the genotypes. However, aldosterone-regulated Na⁺ transport through the epithelial Na⁺ channels (ENaC) was increased in the distal colon mucosa of female task1^-/- mice (Figure 2C).

Figure 1.

Aldosterone levels in task1^-/- mice. Plasma aldosterone levels of adult female (A) and male (B) mice on various salt diets are shown. In female task1^+/+ (+/+) and task1^+/- (+/-) mice, aldosterone concentrations are not different from each other under the three diet conditions (P=0.679), whereas they are considerably higher (P<0.0001) in task1^-/- mice (-/-). In males, aldosterone levels varied with the diet independently from the genotype (P>0.1). The numbers of female mice per genotype were 11–27 (+/+), 13–15 (+/-), and 17–32 (-/-). Corresponding numbers for males were 10–27 (+/+), 3–9 (+/-), and 9–23 (-/-). NK: normal K⁺ (0.75%) diet; HK: high-K⁺ (3%) diet; LK: low-K⁺ (0.05%) diet; LNa: low-Na⁺ (<0.005%) diet. ^*Indicates statistically different from normal K⁺ diet (NK) or between the groups as indicated. NS: not significantly different.

View full figure (60 KB)

Figure 2.

Primary hyperaldosteronism-associated symptoms under normal diet condition. (A) Female task1^-/- mice showed a strongly decreased plasma renin activity, which was paralleled by reduced renin-specific immunostaining in female task1^-/- kidney (right picture) compared with wild type (left picture) (n=20). In males, renin activities were similar for both genotypes (n=15). (B) Hypokalemia in female task1^-/- mice (n=14); normokalemia in male mice (n=15). (C) Ussing chamber experiments of distal colonic mucosa. As a measure of ENaC-dependent Na⁺ reabsorption, amiloride-sensitive equivalent short circuit current (I_sc) was increased in female task1^-/- mice (left hand side; n=14). An original experiment showing the effect of luminal amiloride (10 M) on transepithelial voltage (V_te) is depicted on the right hand side. ^*Indicates statistically different from task1^+/+ mice.

View full figure (61 KB)

Blood pressure and heart rate

Since enhanced salt retention in hyperaldosteronism can cause arterial hypertension, we assessed blood pressure by tail cuff measurements. Systolic blood pressure values of male task1^-/- were not significantly different from those of male wild-type mice. In female task1^-/- mice, however, systolic arterial blood pressure was 15 mmHg higher than in wild-type mice (Figure 3A). This increase in blood pressure was paralleled by a significant bradycardia, suggesting that the elevated blood pressure was not caused by a higher sympathetic nerve tone. In male mice, no difference in the heart rate was observed (Figure 3B).

Figure 3.

Hyperaldosteronism and hypertension in task1^-/- mice. (A) In tail cuff measurements, female task1^-/- mice displayed an increased systolic blood pressure compared with task1^+/+ mice (n=20). Male task1^-/- only tended to have a higher blood pressure (P=0.067; n=20). (B) Heart rate was lower in female task1^-/- compared with task1^+/+ mice (n=20). In male mice, the heart rate was not different between the genotypes (P=0.20; n=20). (C) Inhibition of the mineralocorticoid receptor by canrenoate normalized systolic blood pressure of female task1^-/- mice (n=12, paired experiments). ^*Indicates statistically different.

View full figure (59 KB)

To further evaluate the contribution of increased aldosterone concentration in female task1^-/- mice as cause for the arterial hypertension, the effect of the aldosterone receptor blocker canrenoate was tested in another set of experiments. Under control conditions, female task1^-/- mice displayed higher systolic blood pressure values than wild-type mice. Next, canrenoate was added to the drinking water (canrenoate 1.25 g/l) for 10 days. From days 6–10, arterial blood pressure was determined. Canrenoate led to a decrease of systolic blood pressure in task1^-/- but not in wild-type mice, thereby diminishing the difference in blood pressure between the genotypes (Figure 3C). These data confirm that the arterial hypertension of female task1^-/- mice is caused by aldosterone-induced expansion of the extracellular volume.

Localization of aldosterone-producing cells in the adrenal cortex

The high aldosterone plasma concentrations of female task1^-/- mice pointed to an increased secretion by aldosterone-producing cells in the zona glomerulosa of the adrenal cortex. Interestingly, immunofluorescence experiments using an antibody directed against the aldosterone synthase (CYP11B2) disclosed a sex-dependent effect of the task1 knockout on adrenocortical zonation. In wild-type mice of either sex, the aldosterone synthase-specific staining was localized exclusively in the glomerulosa cells of the outer layer of the adrenal cortex. In adrenal glands of female task1^-/- mice, however, the zona glomerulosa was virtually absent and aldosterone synthase-positive cells were observed in deeper zones of the adrenal cortex (Figure 4, right upper panels). In contrast, adult male task1^-/- mice did not show abnormal zonation patterns (Figure 4, right lower panels). High K⁺ diet is known to drastically increase aldosterone synthase expression in glomerulosa cells. Thus, we checked whether high K⁺ diet could unmask responsive glomerulosa-like cells in adrenal glands from female task1^-/- mice. Aldosterone synthase-specific staining was strongly increased after K⁺-rich diet in female task1^+/+ mice. In female task1^-/- mice, no such increase could be observed (Figure 5A). Clearly, as observed for aldosterone itself, the aldosterone synthase expression is independent of the salt diet in female task1^-/- mice. Western blot analysis of aldosterone synthase (Figure 5B) and real-time PCR experiments (data not shown) further supported this conclusion.

Figure 4.

Mislocalization of aldosterone synthase in adrenal glands of task1^-/- mice is sex-dependent. In adult task1^+/+ mice of either sex and in male task1^-/- mice, aldosterone synthase staining is restricted to the zona glomerulosa cells of the outer adrenal cortex. In female task1^-/- mice (right upper picture), however, the regular aldosterone synthase staining is disrupted and broadened to the inner parts of adrenal cortex. The right part of each panel shows higher magnification pictures of the adrenal cortex (aldosterone synthase in green, differential interference contrast in gray scale, nuclear staining with HOE33342 in blue).

View full figure (377 KB)

Figure 5.

Effect of K⁺-rich diet on aldosterone synthase expression. (A) In female task1^+/+ mice (left side), high-K⁺ diet strongly increased aldosterone synthase-specific staining in glomerulosa cells. In female task1^-/- mice (right side), high-K⁺ diet did not change localization and abundance of aldosterone synthase. (B) Western blot analysis of aldosterone synthase (cyp11b2) protein abundance in adrenal glands at normal K⁺ diet (NK) and high-K⁺ diet (HK). The upper trace shows a typical blot (the bands of higher molecular weight are considered non-specific), and the lower trace shows the quantification of aldosterone synthase protein expression (normalized to beta-actin; task1^+/+, n=8; task1^-/-, n=8). ^*Indicates statistically different from normal K⁺ diet (NK).

View full figure (117 KB)

The mislocalization of aldosterone synthase in the zona fasciculata raised the question whether aldosterone secretion could be under the control of ACTH instead of angiotensin II. Dexamethasone (6 mg/kg, s.c.) was administered to female mice for 3 days to decrease plasma ACTH concentrations. This treatment virtually suppressed aldosterone in female task1^+/+ and task1^-/- mice (Figure 6). Therefore, hyperaldosteronism of female task1^-/- mice was fully glucocorticoid-remediable.

Figure 6.

Glucocorticoid-remediable hyperaldosteronism in female task1^-/- mice. Plasma aldosterone concentration has been measured in task1^+/+ and task1^-/- mice before (con) and after (dexa) administration of dexamethason (n=6). ^*Indicates statistically different from task1^+/+ mice.

View full figure (7 KB)

Corticosterone synthesis in adrenal glands

Next we investigated whether the replacement of the typical zona fasciculata by aldosterone-producing cells in female task1^-/- mice was paralleled by an impairment of corticosterone synthesis. To this end, expression of the corticosterone-producing enzyme 11beta-hydroxylase (CYP11B1) in adrenal glands was determined by semiquantitative real-time PCR. No difference in 11beta-hydroxylase expression was observed between task1^-/- and task1^+/+, but female mice showed a threefold higher expression level than male mice (data not shown), which is in agreement with previously published data (Bielohuby et al, 2007). Additionally, plasma corticosterone concentration was similar in both genotypes (female task1^+/+ 39135 g/l, n=20; female task1^-/- 34126 g/l, n=41).

Dynamics of adrenal cortex zonation

The following series of experiments was aimed at elucidating possible mechanisms underlying the gender differences of adrenal cortex zonation in task1^-/- mice. Obviously, mechanisms compensating for the absence of TASK1 were present in adult male task1^-/- but absent in female task1^-/- mice. Adrenal glands from mice of both sexes were compared before the age of puberty. Immunofluorescence of adrenal glands of mice at postnatal day 18 disclosed a normal zonation pattern of wild-type mice of either sex but abnormal zonation in male and female task1^-/- mice (Figure 7A).

Figure 7.

Age and sex dependence of adrenocortical zonation. (A) In 18-day-old task1^+/+ mice of either sex, aldosterone synthase-specific staining is restricted to glomerulosa cells of outer adrenal cortex (left pictures). In task1^-/- mice of the same age, aberrant aldosterone synthase-specific staining is found in the mid-region of adrenal cortex (right pictures). The staining in the adrenal medulla is considered non-specific. (B) Effects of castration, estradiol (estr.), and testosterone (test.) treatment on aldosterone synthase localization. Male mice were castrated at the age of 5 weeks, followed by treatment with or without estradiol benzoate (4 g/g/day, s.c., for 5 weeks). In male task1^+/+, castration (with or without estradiol treatment) did not affect adrenocortical zonation. Sham-operated male task1^-/- showed normal zonation, too (upper panel). In male task1^-/-, castration prevented normal zonation. Estradiol appeared to reduce aldosterone synthase expression without having clear effects on zonation patterns. In female task1^-/- mice, treatment with testosterone for 3 weeks induced normal zonation, highlighting the importance of androgens for adrenocortical rezonation in task1^-/- mice (lower panel).

View full figure (315 KB)

To elucidate the possible role of sex hormones on adrenocortical zonation of task1^-/- mice in more detail, aldosterone synthase localization was determined in male mice after castration. Castration was performed at the age of 5 weeks (when male task1^-/- mice are not yet zonated) followed by an estrogen treatment. At the age of 10 weeks, castrated task1^+/+ mice showed normal zonation irrespective of the treatment with or without estrogen (Figure 7B). Sham-operated task1^-/- mice also showed normal zonation. However, castration prevented normal zonation in task1^-/- mice. Treatment with estrogens did not enforce abnormal aldosterone synthase expression; if at all, high-dose estrogen treatment reduced aldosterone synthase expression without clearly restoring normal zonation pattern. These data suggested that the compensatory mechanisms allowing restoration of adrenocortical zonation in male task1^-/- mice were androgen-dependent. To further evaluate this hypothesis, 4-week-old female task1^-/- mice were treated with testosterone, and adrenocortical zonation was analyzed by aldosterone synthase-specific immunofluorescence. In fact, aldosterone synthase localization in female task1^-/- mice was redirected to the zona glomerulosa after testosterone treatment (Figure 7B).

Electrical properties of adrenocortical cells

To evaluate the contribution of task1 channels for the whole-cell conductance, primary cultured adrenocortical cells of adult male task1^-/- and task1^+/+ mice were examined by the patch-clamp technique. Cells from task1^-/- mice exhibited a more depolarized membrane voltage at resting conditions. In contrast to task1^+/+ cells, task1^-/- cells displayed a non-significant change of whole-cell conductance upon acid extracellular pH. However, the membrane voltage, which is more sensitive to small changes of K⁺ conductance, depolarized significantly at pH 6, suggesting the presence of other acid-sensitive K⁺ channels (e.g., homomeric task3 channels) in task1^-/- cells (Figure 8A and B). Angiotensin II inhibited the whole-cell conductance and depolarized the membrane of task1^+/+ and task1^-/- cells. Interestingly, the inhibition of the whole-cell conductance by angiotensin II was more pronounced than by acidic extracellular pH, indicating that angiotensin II—besides TASK-like channels—inhibits acid-insensitive channels, too. The task1-related K⁺ channel task3 is also expressed in adrenal glands. Therefore, it is possible that task3 inhibition underlies the acid-induced depolarization observed in task1^-/- cells. Actually, real-time PCR experiments disclosed a higher task3 gene expression in adult male mice as compared with young mice. In females, task3 expression did not appear to be age-dependent (Figure 8C).

Figure 8.

Adrenocortical TASK channel function and gene expression. Whole-cell current measurements in primary cultured mouse adrenocortical cells: (A) Typical traces of a task1^+/+ (upper traces) and a task1^-/- cell (lower traces) under control conditions, at low extracellular pH, and after stimulation with angiotensin II (1 nM). Cells have been clamped from -95 to +30 mV (25 mV steps). (B) Summary of similar whole-cell measurements. For better comparison, whole-cell currents (measured at a voltage step from -95 to +30 mV) have been normalized to cell capacitances (pA/pF; upper diagram). Membrane voltages have been measured in current clamp zero mode (lower diagram). ^*Indicates statistical difference from control; and ^#indicates statistical difference between task1^+/+ and task1^-/-. (C) Age- and sex-dependent task3 mRNA expression. Real-time PCR measurements revealed an increase in adrenal task3 mRNA abundance in adult male mice compared with young and female mice (young mice (2–4 weeks of age), n=48; adult mice (4–6 month of age) n=14). Task3 expression was not affected by the task1 gene disruption.

View full figure (95 KB)

Hyperaldosteronism and hypertension

The disruption of the task1 K⁺ channel gene allowed the identification of the physiological functions of this leak channel in determining mineralocorticoid homeostasis. Female task1^-/- mice exhibited hyperaldosteronism accompanied by marked arterial hypertension. The significantly lower heart rate of female task1^-/- mice compared with wild-type mice foreclosed the possibility that augmentation of the sympathetic nerve tone is causal for the arterial hypertension observed. Blockade of the mineralocorticoid receptor by canrenoate normalized high blood pressure in female task1^-/- mice without affecting it in wild-type mice. Thus, hyperaldosteronism-induced volume expansion seems to be the decisive factor for arterial hypertension related to task1 gene disruption. Under physiological conditions, the most important stimuli for aldosterone secretion are activation of the renin–angiotensin pathway and hyperkalemia. Female task1^-/- mice showed low plasma renin activity and low plasma K⁺ concentration, clearly indicating a defective regulation of mineralocorticoid secretion. Surprisingly, the disturbances in mineralocorticoid homeostasis were restricted to the female gender. Male task1^-/- mice were perfectly normal with regard to the plasma aldosterone levels and displayed normal regulation of aldosterone secretion. Clearly, in male animals the task1 channel is not an essential factor for adrenocortical function and its absence can be compensated.

Task1/task3 channel function

TASK1 belongs to a subgroup of 2P-domain K⁺ channels comprising TASK1, TASK3, and TASK5 (KCNK15). In heterologous expression systems, TASK5 was found to be inactive, whereas TASK1 and TASK3 were able to form functional homomeric channels (Karschin et al, 2001). Nevertheless, there is some evidence that TASK1 and TASK3 might form functional heterodimers in vitro and in vivo (Czirjak and Enyedi, 2001; Berg et al, 2004; Kang et al, 2004; Lotshaw, 2006). One might speculate that task3 is the channel that compensates for task1 gene disruption in adult mice. Accordingly, all previous studies on task1^-/- mice published so far have reported a moderate phenotype. The absence of severe neurological symptoms in task1^-/- mice has been interpreted as either compensation by task3 channels in neurons in which the task1 and task3 coexpress, or that task1 serves as a modulatory subunit for task3 (Aller et al, 2005; Linden et al, 2006; Meuth et al, 2006). We observed a male-specific increase in TASK3 expression in adult mice that may account for the restoration of the normal adrenal gland function in male mice. In patch-clamp experiments, the membrane voltage of primary cultured adrenocortical cells of task1^-/- mice was more depolarized compared with cells of task1^+/+ mice. In contrast to task1^+/+ cells, acidic extracellular pH had no significant effect on the whole-cell conductance of task1^-/- cells in which acid-sensitive task1 were absent. However, acidic pH still induced a significant depolarization in task1^-/- cells, likely because even small conductance changes (e.g., by inhibition of homomeric task3 channels) can induce significant changes of the membrane voltage. In cells of both genotypes, the effect of angiotensin II was stronger than the effect of acidic pH, suggesting that angiotensin II probably leads to inhibition of 'TASK-like' and 'non-TASK-like' channels, which are insensitive to extracellular acidification. Creation of task1/task3 double knockout mice will be necessary to definitively appreciate the contribution of task3 to the membrane conductance and its eventual role for the late zonation process observed in male task1^-/- mice.

Task1 determines zonation of adrenal cortex

Classically, primary hyperaldosteronism is caused by adrenal gland hyperplasia or aldosterone-producing adenoma (Conn, 1955; Carroll et al, 1996; Ganguly, 1998), and one might expect such pathohistological findings in female task1^-/- mice. Examination of the morphology of paraffin-embedded adrenal glands of adult wild-type and task1^-/- mice allowed excluding any adenoma-like changes. Surprisingly, aldosterone synthase-specific immunofluorescence staining disclosed a dramatic change in functional zonation of adrenal glands: The localization of aldosterone-producing cells was shifted from the subcapsular region to the internal cortex, which is the typical site of corticosterone-producing cells. Nevertheless, corticosterone plasma concentrations and 11beta-hydroxylase gene expression were not modified in task1^-/- as compared with wild-type mice. These data indicate that despite the spectacular zonation defect, corticosterone production is not affected. To our knowledge no other example showing such dramatic changes in the process of adrenal cortex zonation has been described so far.

Adrenal zonation is a dynamic process, which is dependent on age, sex, and species (Tanaka and Matsuzawa, 1995; Parker et al, 1997; Wotus et al, 1998). Apart from steroidogenic enzyme expression, a few markers have been identified that are expressed in different zones of the rodent adrenal cortex (Okamoto et al, 1997; Mitani et al, 1999). They include nuclear receptors of the NR4 family (Nurr1, NGFI-B) that have been associated with zone-specific transcriptional regulation of the aldosterone synthase and 3beta-hydroxysteroid dehydrogenase type II (HSD3B2) genes (Bassett et al, 2004a, 2004b, 2004c; Lu et al, 2004). However, the molecular mechanisms controlling adrenocortical zonation remain mostly unknown. The task1 knockout mouse with its 'all or none' modified zonation appears to be a promising model to unveil these mechanisms.

Sex-dependent rescue of adrenal abnormalities

Gender difference usually suggests that sex hormones are involved in the respective physiological function. Therefore, adrenal gland morphology and zonation patterns were studied in animals before puberty. In fact, at 18 days of age, male task1^-/- mice exhibited the same severely compromised zonation of adrenal glands as females. At this developmental stage, wild-type mice of both sex showed already normal zonation patterns. Taken together, these observations indicate that task1 is a K⁺ channel essential for the determination of adrenal gland zonation and aldosterone secretion in male and female mice before puberty. After puberty, unknown factors, possibly task3 K⁺ channels, substitute for the absence of task1 and promote compensatory mechanisms in male task1^-/- mice. To further evaluate the contribution of male sex hormones as factors triggering the restoration of the zonation process, we have investigated the effect of castration and treatment with estrogen. In male task1^+/+ mice, castration did not affect adrenocortical zonation. However, castration abrogated the ability of male task1^-/- mice for developing normal zonation, suggesting a critical role of androgens for driving the compensatory mechanisms in male task1^-/- mice. The role of androgens for the compensation of the task1 gene disruption was corroborated by the observation that female task1^-/- mice display normal aldosterone synthase distribution after testosterone treatment. The fact that regular zonation and biochemical properties of the adrenal cortex seem to be under the strict control of the expression of a K⁺ channel is certainly one of the most striking results from this study.

Clinical implications

Inactivation of task1 leads to inappropriate high plasma aldosterone levels and, as a consequence, to fluid retention and arterial hypertension associated with low plasma renin activity. This hyperaldosteronism is fully glucocorticoid-suppressible. In humans, glucocorticoid-suppressible aldosteronism is due to a gene crossover recombination. This results in an 11beta-hydroxylase/aldosterone synthase chimeric gene that is expressed in the zona fasciculata and regulated by ACTH (Lifton et al, 1992; Pascoe et al, 1995). However, cases of glucocorticoid-suppressible aldosteronism that do not have a chimeric gene have been described in which no clear explanation for the abnormality was found (Gordon et al, 1995; Mulatero et al, 1998; Fardella et al, 2001). An interesting possibility is that abnormalities in the expression of TASK1, which is highly expressed in human adrenal gland (http://www.ncbi.nlm.nih.gov/Un
iGene/ESTProfileViewer.cgi?ug
list=Hs.645288), might be responsible for a fasciculata expression of the aldosterone synthase in those patients with glucocorticoid-suppressible aldosteronism who do not have a chimeric gene.

In conclusion, physiological task1 activity is required for normal development of adrenocortical zonation. After puberty, the dysregulation of aldosterone secretion and the adrenocortical zonation defect can be rescued in male task1^-/- mice implying androgen-driven upregulation of compensatory mechanisms, for example, increased expression of other K⁺ channels such as task3. At present, there are only very few cases of other K⁺ channels directly linked to a developmental disorder, for example, KCNJ2 (Kir2.1) whose mutations can cause Andersen Syndrome (OMIM 170390). However, in this syndrome the developmental changes are small and result in only discrete modifications of the adult organs. The task1^-/- mouse is the first example for a severe but reversible disorganization of the adrenal cortex and, thus, it represents a unique model to study adrenocortical zonation. In future studies, the availability of this mouse model might greatly facilitate the identification and characterization of cellular factors underlying differentiation and development of adrenal cortex.

The task1^-/- knockout mouse

Task1^-/- knockout mice have been generated as described previously (Aller et al, 2005). The animals used in this study were backcrossed for seven generations into the C57Bl/6J genetic background. Mice were chronically maintained on a normal diet (chow, R03T-25; SAFE, 0.75% K⁺, 0.27% Na⁺). For the experiments, mice were fed high- (3%) or low- (0.05%) K⁺ diet (INRA, France), or low-Na⁺ diet (<0.005% Na⁺), for 2 weeks before the measurements. The animals had free access to food and water. The experimental protocols were approved by the local councils for animal care and were conducted according to the German and French law for animal care.

Blood analysis measurements

Blood was collected into heparin-treated capillary tubes from mice tails after small lateral incision. Samples were centrifuged and plasma was frozen and kept at -20°C. Aldosterone concentrations were determined in unextracted plasma using a solid-phase ¹²⁵I RIA kit (Immunotech, Marseille, France) with a very low cross-reactivity with corticosterone. Plasma Na⁺ and K⁺ concentrations were determined using an AVL 9180 Electrolyte analyzer (Roche, Switzerland). Corticosterone concentrations were measured using an ¹²⁵I RIA kit from MP Biomedicals, LLC (USA).

Blood pressure measurements

Blood pressure measurements were carried out by tail cuff manometry (TSE, Germany). Mice were conditioned to the measurement procedure during four consecutive days. Before measurement, the animals were allowed to stay up for at least 5 min in the measurement device, which was warmed to 37°C. Each day, blood pressure was determined as mean values of 10 cycles, with each cycle consisting of four individual measurements. For inhibition of the mineralocorticoid receptor, canrenoate (1.25 g/l plus sucrose 10%) was added to the drinking water for 10 days (blood pressure measurements were performed during the last 5 days). Before canrenoate treatment, a control period (5 days) was performed during which only sucrose (10%) had been added to the drinking water.

Plasma renin activity

Blood samples were taken from the tail vein. After an incubation of the mouse plasma (1.5 h, 37°C) with plasma of bilaterally nephrectomized male rats as renin substrate, angiotensin I (ng/ml/h) was generated. Afterwards angiotensin I (ng/ml/h) was measured by ¹²⁵I RIA (Byk & DiaSorin Diagnostics, Germany) to determine plasma renin activity.

Ussing chamber

Female wild-type and task1^-/- mice were killed by cervical dislocation and the distal colon was removed. After separation from the muscle layer, colonic mucosa was transferred into a mini Ussing chamber. Experimental solutions were warmed to 37°C by water jackets. The control solution contained (in mmol/l) HEPES 5, NaCl 140, Na-acetate 5, KH₂PO₄ 0.4, K₂HPO₄ 1.6, D-glucose 5, MgCl₂ 1, CaCl₂ 1.3, pH 7.4). Indomethacin (10 mol/l) was added to prevent generation of endogenous prostaglandins. Data were collected and analyzed by a PowerLab setup (AD Instruments, Australia).

Immunofluorescence

Mice were anesthetized with isoflurane. After incision of the vena cava inferior, mice were perfused with 3% paraformaldehyde-containing PBS through a polyethylene catheter inserted into the abdominal aorta. After removing and freezing adrenal glands and kidneys, cryosectioning (10 m) was performed on Polysine slides (Kindler, Freiburg, Germany). For unmasking epitopes, the sections were incubated in 0.1% SDS for 5 min. After washing with PBS and blocking with 5% bovine serum albumin (10 min), a polyclonal aldosterone synthase antibody raised in rabbit (dilution 1:50; Wotus et al, 1998) or a polyclonal renin antibody raised in chicken (dilution 1:300; Kurtz et al, 2007) was applied. Cy3 goat anti-rabbit (dilution 1:500; Dianova, Hamburg, Germany) or TRITC donkey anti-chicken (dilution 1:500; Dianova, Hamburg, Germany) was applied as secondary antibodies. The sections were examined with a filter wheel-based imaging system (Universal Imaging Corporation, Dowingtown, PA, USA) mounted on an inverted microscope (Axiovert 200 M; Zeiss, Jena, Germany; excitation 525–550 nm, emission 575–635 nm) and with confocal microscope (LSM 510 Meta; Zeiss, Jena, Germany; excitation 543 nm, emission 560–615 nm).

Hormonal treatment and castration

Five-week-old male mice were anesthetized with isoflurane and castrated via scrotal incision. For estrogen treatment, desexed mice received daily injections of estradiol benzoate (4 g/g/day, s.c.). At the age of 10 weeks, mice were killed and used for immunofluorescence. To test the effect of androgens in females, female task1^-/- mice (4 weeks old) were injected twice a week with 1 mg testosterone (Androtardyl, Schering) dissolved in 50 l sesame oil for 3 weeks (Wunderlich et al, 2005).

Real-time PCR

Adrenal gland total RNA was isolated using RNeasy mini kit (Qiagen, Hilden, Germany) and reverse transcribed into cDNA using M-MLV reverse transcriptase (Promega, Mannheim, Germany). TASK3-specific (sense primer: CAC TGT CAT CAC AAC TAT CGG; antisense primer: CAG CGT AGA ACA TAC AGA AGG) real-time PCR was performed using SYBR green (Qiagen, Hilden, Germany) and beta-actin (sense primer: CCA CCG ATC CAC ACA GAG TAC TT; antisense primer: GAC AGG ATG CAG AAG GAG ATT ACT G) as reference gene in a LightCycler machine (Roche, Basel, Switzerland).

Western blot analysis

Adrenal glands from mice were rapidly frozen and stored at -80°C until used. Tissue was pulverized with mortar and pestle in liquid nitrogen and homogenized in RIPA (0.1% SDS, 1% Igepal, 0.5% sodiumdeoxycholate, with protease inhibitor cocktail in PBS, pH 7.4) using a 27-gauge needle. After incubation for 15 min on ice, cellular debris were removed by centrifugation at 10 000 g at 4°C for 10 min. Protein concentration of supernatant was determined by Bradford based protein assay (Bio-Rad). Equal amounts of protein were mixed with reducing Laemmli sample buffer (Bio-Rad) and separated by electrophoresis through 8% SDS–PAGE, transferred to a nitrocellulose membrane, rinsed with PBS, and blocked with 5% nonfat milk and 0.1% Tween in PBS for 1 h. Membranes were incubated overnight with the aldosterone synthase antibody (dilution 1:500; raised in rabbit; Wotus et al, 1998) in Tween–PBS with 0.1% BSA and 0.02% NaN₃, washed, and incubated with HRP-conjugated secondary antibody (dilution 1:7500) for 1 h and again washed. Images were developed with Western Blotting Luminol Reagent (Santa Cruz) and visualized on Medical X-ray film (Fotochemische Werke GmbH, Berlin).

Primary cell culture and patch clamp

During isoflurane anesthesia adult male task1^-/- and task1^+/+ mice were perfused with a collagenase-containing Ringer-type solution (collagenase (type CLS II) 1 mg/ml; Biochrom, Berlin, Germany). Adrenal cortex was harvested, cut into small pieces, and digested for another 15 min at 37°C. Single cells and cell clusters were seeded on culture dishes (Falcon, Germany) and used for patch-clamp experiments 4–24 h after seeding. Whole-cell recordings were performed using an EPC-10 amplifier (Heka, Germany). The patch pipette solution contained (in mM) 95 K-gluconate, 30 KCl, 4.8 Na₂HPO₄, 1.2 NaH₂PO₄, 5 glucose, 2.38 MgCl₂, 0.726 CaCl₂, 1 EGTA, 3 ATP, pH 7.2. The extracellular Ringer-type solution contained (in mM) 145 NaCl, 0.4 KH₂PO₄, 1.6 K₂HPO₄, 5 glucose, 1 MgCl₂, 1.3 CaCl₂, 5 HEPES, pH 7.4. Differences in whole-cell currents induced by a voltage step from -95 to +30 mV were normalized to cell capacitance (pA/pF) as a measure of cell surface area. Leak subtraction was not performed. All experiments were performed at 37°C.

Statistics

Data are shown as mean valuess.e.m. from 'n' observations. Paired as well as unpaired Student's t-test were used as appropriate. A P-value of <0.05 was accepted to indicate statistical significance.

We thank M Larroque for expert assistance and P Meneton for helpful discussions and scientific advice. The study was supported by the Deutsche Forschungsgemeinschaft (SFB699 to RW), the Centre National de la Recherche scientifique (JB), and by the European Section of Aldosterone Council (11AD5B to JB).

Aller MI, Veale EL, Linden AM, Sandu C, Schwaninger M, Evans LJ, Korpi ER, Mathie A, Wisden W, Brickley SG (2005) Modifying the subunit composition of TASK channels alters the modulation of a leak conductance in cerebellar granule neurons. J Neurosci 25: 11455–11467 | Article | PubMed | ISI | ChemPort |

Arrighi I, Bloch-Faure M, Grahammer F, Bleich M, Warth R, Mengual R, Drici MD, Barhanin J, Meneton P (2001) Altered potassium balance and aldosterone secretion in a mouse model of human congenital long QT syndrome. Proc Natl Acad Sci USA 98: 8792–8797 | Article | PubMed | ChemPort |

Bassett MH, Suzuki T, Sasano H, De Vries CJ, Jimenez PT, Carr BR, Rainey WE (2004a) The orphan nuclear receptor NGFIB regulates transcription of 3beta-hydroxysteroid dehydrogenase. Implications for the control of adrenal functional zonation. J Biol Chem 279: 37622–37630 | Article | ChemPort |

Bassett MH, Suzuki T, Sasano H, White PC, Rainey WE (2004b) The orphan nuclear receptors NURR1 and NGFIB regulate adrenal aldosterone production. Mol Endocrinol 18: 279–290 | Article | ChemPort |

Bassett MH, White PC, Rainey WE (2004c) The regulation of aldosterone synthase expression. Mol Cell Endocrinol 217: 67–74 | Article | PubMed | ChemPort |

Berg AP, Talley EM, Manger JP, Bayliss DA (2004) Motoneurons express heteromeric TWIK-related acid-sensitive K+ (TASK) channels containing TASK-1 (KCNK3) and TASK-3 (KCNK9) subunits. J Neurosci 24: 6693–6702 | Article | PubMed | ISI | ChemPort |

Bielohuby M, Herbach N, Wanke R, Maser-Gluth C, Beuschlein F, Wolf E, Hoeflich A (2007) Growth analysis of the mouse adrenal gland from weaning to adulthood: time- and gender-dependent alterations of cell size and number in the cortical compartment. Am J Physiol Endocrinol Metab 293: E139–E146 | Article | PubMed | ChemPort |

Carroll J, Dluhy R, Fallo F, Pistorello M, Bradwin G, Gomez-Sanchez CE, Mortensen R (1996) Aldosterone-producing adenomas do not contain glucocorticoid-remediable aldosteronism chimeric gene duplications. J Clin Endocrinol Metab 81: 4310–4312 | Article | PubMed | ChemPort |

Chai W, Danser AH (2006) Why are mineralocorticoid receptor antagonists cardioprotective? Naunyn Schmiedebergs Arch Pharmacol 374: 153–162 | Article | PubMed | ChemPort |

Conn JW (1955) Part I. Painting background. Part II. Primary aldosteronism, a new clinical syndrome. J Lab Clin Med 45: 3–17 | PubMed | ISI | ChemPort |

Czirjak G, Enyedi P (2001) Formation of functional heterodimers between the TASK-1 and TASK-3 two pore domain potassium channel subunits. J Biol Chem 277: 5426–5432 | Article | PubMed | ChemPort |

Czirjak G, Enyedi P (2002) TASK-3 dominates the background potassium conductance in rat adrenal glomerulosa cells. Mol Endocrinol 16: 621–629 | Article | PubMed | ISI | ChemPort |

Czirjak G, Fischer T, Spat A, Lesage F, Enyedi P (2000) TASK (TWIK-related acid-sensitive K+ channel) is expressed in glomerulosa cells of rat adrenal cortex and inhibited by angiotensin II. Mol Endocrinol 14: 863–874 | Article | PubMed | ISI | ChemPort |

Enyeart JA, Danthi SJ, Enyeart JJ (2004) TREK-1 K+ channels couple angiotensin II receptors to membrane depolarization and aldosterone secretion in bovine adrenal glomerulosa cells. Am J Physiol 287: E1154–E1165 | ChemPort |

Fardella CE, Pinto M, Mosso L, Gomez-Sanchez C, Jalil J, Montero J (2001) Genetic study of patients with dexamethasone-suppressible aldosteronism without the chimeric CYP11B1/CYP11B2 gene. J Clin Endocrinol Metab 86: 4805–4807 | Article | PubMed | ChemPort |

Ganguly A (1998) Primary aldosteronism. N Engl J Med 339: 1828–1834 | Article | PubMed | ISI | ChemPort |

Gordon RD, Stowasser M, Klemm SA, Tunny TJ (1995) Primary aldosteronism—some genetic, morphological, and biochemical aspects of subtypes. Steroids 60: 35–41 | Article | PubMed | ISI | ChemPort |

Ibrahim HN, Hostetter TH (2003) Aldosterone in renal disease. Curr Opin Nephrol Hypertens 12: 159–164 | Article | PubMed | ChemPort |

Kang D, Han J, Talley EM, Bayliss DA, Kim D (2004) Functional expression of TASK-1/TASK-3 heteromers in cerebellar granule cells. J Physiol 554: 64–77 | Article | PubMed | ISI | ChemPort |

Karschin C, Wischmeyer E, Preisig-Muller R, Rajan S, Derst C, Grzeschik KH, Daut J, Karschin A (2001) Expression pattern in brain of TASK-1, TASK-3, and a tandem pore domain K(+) channel subunit, TASK-5, associated with the central auditory nervous system. Mol Cell Neurosci 18: 632–648 | Article | PubMed | ChemPort |

Kurtz L, Schweda F, de Wit C, Kriz W, Witzgall R, Warth R, Sauter A, Kurtz A, Wagner C (2007) Lack of connexin 40 causes displacement of renin-producing cells from afferent arterioles to the extraglomerular mesangium. J Am Soc Nephrol 18: 1103–1111 | Article | PubMed | ChemPort |

Lesage F, Lazdunski M (2000) Molecular and functional properties of two-pore-domain potassium channels. Am J Physiol 279: F793–F801 | ISI | ChemPort |

Lifton RP, Dluhy RG, Powers M, Rich GM, Gutkin M, Fallo F, Gill Jr JR, Feld L, Ganguly A, Laidlaw JC, Murnaghan DJ, Kaufman C, Stockigt JR, Ulick S, Lalouel J-M (1992) Hereditary hypertension caused by chimaeric gene duplications and ectopic expression of aldosterone synthase. Nat Genet 2: 66–74 | Article | PubMed | ISI | ChemPort |

Linden AM, Aller MI, Leppa E, Vekovischeva O, Aitta-Aho T, Veale EL, Mathie A, Rosenberg P, Wisden W, Korpi ER (2006) The in vivo contributions of TASK-1-containing channels to the actions of inhalation anesthetics, the alpha(2) adrenergic sedative dexmedetomidine, and cannabinoid agonists. J Pharmacol Exp Ther 317: 615–626 | Article | PubMed | ISI | ChemPort |

Lotshaw DP (2001) Role of membrane depolarization and T-type Ca2+ channels in angiotensin II and K+ stimulated aldosterone secretion. Mol Cell Endocrinol 175: 157–171 | Article | PubMed | ChemPort |

Lotshaw DP (2006) Biophysical and pharmacological characteristics of native two-pore domain TASK channels in rat adrenal glomerulosa cells. J Membr Biol 210: 51–70 | Article | PubMed | ChemPort |

Lu L, Suzuki T, Yoshikawa Y, Murakami O, Miki Y, Moriya T, Bassett MH, Rainey WE, Hayashi Y, Sasano H (2004) Nur-related factor 1 and nerve growth factor-induced clone B in human adrenal cortex and its disorders. J Clin Endocrinol Metab 89: 4113–4118 | Article | PubMed | ChemPort |

Meuth SG, Aller MI, Munsch T, Schuhmacher T, Seidenbecher T, Meuth P, Kleinschnitz C, Pape HC, Wiendl H, Wisden W, Budde T (2006) The contribution of TWIK-related acid-sensitive K+-containing channels to the function of dorsal lateral geniculate thalamocortical relay neurons. Mol Pharmacol 69: 1468–1476 | Article | PubMed | ISI | ChemPort |

Mitani F, Mukai K, Miyamoto H, Suematsu M, Ishimura Y (1999) Development of functional zonation in the rat adrenal cortex. Endocrinology 140: 3342–3353 | Article | PubMed | ChemPort |

Mulatero P, Veglio F, Pilon C, Rabbia F, Zocchi C, Limone P, Boscaro M, Sonino N, Fallo F (1998) Diagnosis of glucocorticoid-remediable aldosteronism in primary aldosteronism: aldosterone response to dexamethasone and long polymerase chain reaction for chimeric gene. J Clin Endocrinol Metab 83: 2573–2575 | Article | PubMed | ChemPort |

Nussberger J (2003) Investigating mineralocorticoid hypertension. J Hypertens Suppl 21: S25–S30 | PubMed | ChemPort |

Okamoto M, Takemori H, Halder SK, Hatano O (1997) Zona glomerulosa-specific factor: cloning and function. Steroids 62: 73–76 | Article | PubMed | ChemPort |

Parker Jr CR, Mixon RL, Brissie RM, Grizzle WE (1997) Aging alters zonation in the adrenal cortex of men. J Clin Endocrinol Metab 82: 3898–3901 | Article | PubMed | ChemPort |

Pascoe L, Jeunemaitre X, Lebrethon MC, Curnow KM, Gomez-Sanchez CE, Gasc JM, Saez JM, Corvol P (1995) Glucocorticoid-suppressible hyperaldosteronism and adrenal tumors occurring in a single French pedigree. J Clin Invest 96: 2236–2246 | Article | PubMed | ISI | ChemPort |

Remuzzi G, Perico N, Macia M, Ruggenenti P (2005) The role of renin–angiotensin–aldosterone system in the progression of chronic kidney disease. Kidney Int Suppl 68: S57–S65 | Article

Sausbier M, Arntz C, Bucurenciu I, Zhao H, Zhou XB, Sausbier U, Feil S, Kamm S, Essin K, Sailer CA, Abdullah U, Krippeit-Drews P, Feil R, Hofmann F, Knaus HG, Kenyon C, Shipston MJ, Storm JF, Neuhuber W, Korth M et al (2005) Elevated blood pressure linked to primary hyperaldosteronism and impaired vasodilation in BK channel-deficient mice. Circulation 112: 60–68 | Article | PubMed | ISI | ChemPort |

Spat A, Hunyady L (2004) Control of aldosterone secretion: a model for convergence in cellular signaling pathways. Physiol Rev 84: 489–539 | Article | PubMed | ISI | ChemPort |

Szabadkai G, Varnai P, Enyedi P (1999) Selective inhibition of potassium-stimulated rat adrenal glomerulosa cells by ruthenium red. Biochem Pharmacol 57: 209–218 | Article | PubMed | ChemPort |

Tanaka S, Matsuzawa A (1995) Comparison of adrenocortical zonation in C57BL/6J and DDD mice. Exp Anim 44: 285–291 | Article | PubMed | ChemPort |

Wotus C, Levay-Young BK, Rogers LM, Gomez-Sanchez CE, Engeland WC (1998) Development of adrenal zonation in fetal rats defined by expression of aldosterone synthase and 11beta-hydroxylase. Endocrinology 139: 4397–4403 | Article | PubMed | ChemPort |

Wunderlich F, Dkhil MA, Mehnert LI, Braun JV, El Khadragy M, Borsch E, Hermsen D, Benten WP, Pfeffer K, Mossmann H, Krucken J (2005) Testosterone responsiveness of spleen and liver in female lymphotoxin beta receptor-deficient mice resistant to blood-stage malaria. Microbes Infect 7: 399–409 | Article | PubMed | ChemPort |

Main navigation

• Search EMBO Journal

  Search

• Advanced search

• The EMBO Journal

  • Home
  • Aims and scope
  • Current issue
  • Advance online publication
  • Web focuses
  • Archive:
  • Browse by issue
  • Browse by subject
  • Browse by category
  • Free online sample issue
  • Press releases

• Authors and Referees

  • Editorial process
  • Guide for authors
  • Submit an article
  • Guide for referees
  • Editors and Editorial Board
  • Contact editorial office

• Customer Service

  • Subscribe
  • Order sample copy
  • Purchase articles
  • Reprints and permissions
  • Contact NPG
  • Advertising