| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Division of Nephrology and Hypertension, Inselspital, University of Berne (A.O., B.D., P.A., T.Z., V.P., B.M.F., F.J.F., P.F.), 3010 Berne, Switzerland; and Pediatric Service, Hospital Nacional Prof. A. Posadas (M.N.D., H.R.), Buenos Aires, Argentina
Address all correspondence and requests for reprints to: Paolo Ferrari, M.D., Division of Nephrology and Hypertension, Inselspital, University of Berne, Freiburgstrasse 10, 3010 Berne, Switzerland. E-mail: paolo.ferrari{at}insel.ch
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Here, we describe two siblings, 1 and 2 yr old, who were
diagnosed with hypokalemic hypertension and low plasma aldosterone and
renin levels, indicating mineralocorticoid hypertension. Analysis of
urinary steroid metabolites showed a markedly impaired metabolism of
cortisol, with (tetrahydrocortisol +
5
-tetrahydrocortisol)/tetrahydrocortisone ratios of 40–60,
and nearly absent urinary free cortisone. Although phenotypically
normal, the heterozygous parents showed a disturbed cortisol
metabolism.
Genetic analysis of the HSD11B2 gene from the AME patients revealed the homozygous deletion of six nucleotides in exon 2 with the resultant loss of amino acids Leu114 and Glu115, representing the first alteration found in the cofactor-binding domain. The deletion mutant, expressed in HEK-293 cells, showed an approximately 20-fold lower maximum velocity but increased apparent affinity for cortisol and corticosterone. In contrast, two additionally constructed substitutions, Glu115 to Gln or Lys, showed increased maximal velocity and apparent affinity for 11ß-hydroxyglucocorticoids. Functional analysis of wild-type and mutant proteins indicated that a disturbed conformation of the cofactor-binding domain, but not the missing negative charge of Glu115, led to the observed decreased activity of the deletion mutant. Considered together, these findings provide evidence for a role of Glu115 in determining cofactor-binding specificity of 11ßHSD2 and emphasize the importance of structure-function analysis to elucidate the molecular mechanism of AME.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Two isoforms of 11ßHSD have been identified and characterized. The isoform 11ßHSD1 (4) is NADP dependent, catalyzes both reduction and dehydrogenation of glucocorticoids, is expressed in most tissues, and at least in vitro regulates access of glucocorticoids to the glucocorticoid receptor (4, 5, 6). Although the physiological function of 11ßHSD1 in vivo is ill defined, this enzyme was excluded as the cause of AME (7). The 11ßHSD2 enzyme, which preferably uses the cofactor NAD, is predominantly found in mineralocorticoid target tissues such as kidney, colon, and salivary glands as well as in the placenta and some fetal tissues (2, 3). As in vitro the mineralocorticoid receptor has a similar affinity as aldosterone and cortisol, and in vivo cortisol is found in concentrations 100-1000 times higher than those of aldosterone, a mechanism of selectivity for aldosterone in mineralocorticoid target organs is needed (8, 9). This selectivity is provided by 11ßHSD2, a unidirectional enzyme, catalyzing the oxidation of biologically active 11ß-hydroxyglucocorticoids (cortisol and corticosterone) into their inactive 11-keto forms (cortisone and 11-dehydrocorticosterone), thereby rendering protection of the nonselective mineralocorticoid receptor from occupation by cortisol (10, 11).
Mutations in the HSD11B2 gene generating a compromised 11ßHSD2 enzyme activity lead to an overstimulation of the mineralocorticoid receptor by cortisol, thus causing sodium retention, hypokalemia, and high blood pressure. To date, about 60 cases of AME syndrome have been reported in the literature, and approximately 20 different mutations located in exons 3–5 of the HSD11B2 gene resulting in a functional impairment have been described (12).
In this study we describe a deletion mutation causing AME and two additional missense mutations in exon 2 of the HSD11B2 gene not found in patients.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
A 22-month-old girl (J.L.) was admitted to the hospital with poor weight gain (height and weight less than third percentile) and severe hypertension (160/120 mm Hg). Analysis of blood and urine revealed the presence of metabolic alkalosis (pH 7.44; serum bicarbonate, 33.9 mmol/L), hypokalemia (2.7 mmol/L) with hyperkaliuria (75 mmol/day), low PRA (PRA 1.1 ng/mL·h), and serum aldosterone levels (60 pg/mL). Additional features were hypercalciuria (>3.2 mmol/day) and nephrocalcinosis. An echocardiographic examination showed mild left ventricular hypertrophy. Her creatinine clearance was normal. The patient was previously investigated at 11 months of age because of dehydration, poor weight gain, and hypokalemia. In that occasion the diagnosis of Bartter’s syndrome was made because blood pressure was not recorded. As currently the suspicion of a form of monogenic mineralocorticoid hypertension was high, the girl’s 6-month-old brother (N.L.) was also investigated. The boy also presented with growth retardation (height and weight less than third percentile) and hypertension (139/57 mm Hg). Laboratory findings showed severe hypokalemia (2.5 mmol/L) with hyperkaliuria (53 mmol/day), metabolic alkalosis (pH 7.46; serum bicarbonate, 31 mmol/L), very low PRA (0.07 ng/mL·h), and serum aldosterone (56 pg/mL). Hypercalciuria (>2.1 mmol/day), nephrocalcinosis, and left ventricular hypertrophy were also present. His serum creatinine was normal.
Both patients were treated with dexamethasone (0.25 mg/kg·day) for 2
weeks and after a period of 7 days of washout they received
spironolactone (2 mg/kg·day) for 2 weeks. Compared with baseline,
systolic and diastolic blood pressures decreased, on the average, by
19% and 23%, respectively, during dexamethasone and by 19% and 23%,
respectively, during spironolactone. Plasma potassium increased, on the
average, by 32% during dexamethasone and 28% during spironolactone.
Salt intake averaged 1.2 g/day (20 mmol/day sodium) throughout the
metabolic studies. Analysis of the urinary steroid profile by gas
chromatography/mass spectrometry was compatible with AME (Table 1
) (13, 14, 15), and the
diagnosis was confirmed by genetic analysis as described below.
|
The parents of the two patients were first degree cousins with no history of essential hypertension, who remained normotensive with no change in serum electrolytes after receiving a sodium load for 7 days (300 mmol/day).
Urinary steroid profile
Urine samples were analyzed by gas chromatography-mass
spectrometry using a method similar to that described by Shackleton
(13). Sample preparation consisted of preextraction,
enzymatic hydrolysis, extraction from the hydrolysis mixture,
derivatization, and gel filtration. To 1.5 mL urine, 2.5 µg
medroxyprogesterone were added as a recovery standard. The sample was
extracted on a Sep-Pak C18 column, dried,
reconstituted in 0.1 mol/L acetate buffer (pH 4.6), and hydrolyzed with
powdered Helix pomatia enzyme
(Sigma; 12.5 mg) and 12.5 µL ß-
glucuronidase/arylsulfatase liquid enzyme (Roche Molecular Biochemicals, Indianapolis, IN). The resulting free steroids
were extracted once more, followed by addition of an internal standard
mixture [5-androstane-3,17-diol, stigmasterol, cholesteryl
butyrate (2.5 µg each), and 3ß,5ß-tetrahydroaldosterone (0.15
µg)] and derivatization to form the methyloxime-trimethylsilyl
ethers. Samples were analyzed on a Hewlett-Packard Co. gas
chromatograph 6890 (Palo Alto, CA) equipped with a mass selective
detector 5973 and an autoinjector 7683 by selective ion monitoring. The
derivatized samples were analyzed during a temperature-programmed run
(210–265 C) over a 35-min period. A steroid mixture containing a known
amount of all steroid metabolites to be measured was analyzed on a
regular basis to act as a calibration standard (regular updating of
responses and retention times). In each case the ion peak abundance was
quantified against that of stigmasterol internal standard.
Analysis of genomic DNA
Genomic DNA was extracted from peripheral blood leukocytes. The exons and intron-exon boundaries of the HSD11B2 gene were amplified by PCR, and PCR products were analyzed as previously described (16). DNA was visualized by silver staining, and the variant product derived from exon 2 of the patient was verified by sequencing.
Constructs for expression
For expression of the HSD11B2 wild-type gene, a complementary DNA construct in the expression plasmid pcDNA3, containing an engineered Kozak consensus sequence 5' to the initiator ATG codon and a FLAG epitope tag for facilitated detection at the 3'-end, was used (17). Attachment of the FLAG epitope to the 3'-end did not affect the activity of the protein. Mutations were introduced by site-directed mutagenesis using the Quick Change mutagenesis kit from Stratagene (La Jolla, CA). All constructs were verified by sequencing.
Assay for 11ßHSD2 activity
HEK-293 cells, devoid of endogenous 11ßHSD2 activity, were cultured and transfected according to the calcium phosphate precipitation method as described previously (17). Twelve hours later, the medium was replaced to remove the Ca2+ phosphate precipitate, and 48 h posttransfection the medium was replaced by charcoal-treated DMEM. The cells were harvested 72 h posttransfection, washed once, and resuspended in prewarmed (37 C) charcoal-treated DMEM. No cofactor was added in measurements of whole cell activities. Oxidative activities of 11ßHSD2 constructs were measured using lysates of cells subjected to a freeze/thaw cycle and resuspended in a buffer containing 20 mmol/L Tris-HCl (pH 7.4), 250 mmol/L sucrose, 1 mmol/L ethyleneglycol-bis-(ß-aminoethyl ether)-N,N,N',N'-tetraacetic acid, 1 mmol/L ethylenediamine tetraacetate, 1 mmol/L MgCl2, and 400 µmol/L NAD or NADP, respectively. Reactions to determine the rate of conversion of 11ß- hydroxy- into 11-keto glucocorticoids was started by mixing cells or cell extracts corresponding to 0.5–10 µg total proteins with reaction mixture. The assay was performed in a total volume of 20 µL containing 400 µmol/L cofactor, 30 nCi tritiated steroid, and unlabeled steroid at different concentrations ranging from 5 nmol/L to 1 µmol/L. Steroids were analyzed by thin layer chromatography. The expression level of different transfections was determined semiquantitatively by immunoblotting using an antibody specific against the FLAG epitope as described previously (17). Enzyme kinetics were analyzed by the Eadie-Hofstee linear transformation of the Michaelis-Menten equation. Km and maximum velocity (Vmax) values were calculated by unweighted linear regression analysis with mean values of at least three independent transfections. Only conversion rates between 10–60% were considered for calculation. Significance was tested by ANOVA.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Both affected children displayed typical signs and features of
AME. PRA, serum aldosterone, and the urinary metabolite
tetrahydroaldosterone (Table 1
) were significantly decreased,
indicating a suppression of the renin-aldosterone axis. Evidence of an
impaired metabolism of cortisol to cortisone was given by the
diminished amounts of total urinary cortisone (E), urinary free
cortisone, and the metabolites tetrahydrocortisone (THE),
20-dihydrocortisone, 20ß-dihydrocortisone, and cortolones (Table 1) (13, 14, 15). The defect in the conversion of cortisol to
cortisone was reflected by the significantly increased ratios of total
and free, unconjugated cortisol to cortisone and the ratios of their
corresponding metabolites (dihydrocortisol/dihydrocortisone,
(THF+5THF)/THE, cortols/cortolones). The THF/5THF ratio was
decreased in both patients, suggesting a shift from 5ß-reductase
toward 5-reductase metabolism (18).
The disturbed steroid metabolism together with the increased blood pressure, metabolic alkalosis, hypokalemia with hyperkaliuria, and the observed growth retardation led to the diagnosis of AME. Treatment of the patients with spironolactone and amiloride normalized blood pressure and electrolytes and improved growth progression of the patients.
Both heterozygous parents were normotensive, and sodium loading did not
lead to a significant change in their serum electrolytes. However, the
observed low concentrations of urinary tetrahydroaldosterone and the
decreased urinary total and free cortisone clearly revealed a disturbed
steroid metabolism in both parents (Table 1). Whereas the
(THF+5THF)/THE ratio in the father was normal, the ratio in the
mother was slightly above the upper range compared with that in normal
women. The THF/5THF ratio was at the upper range in the father, but
was significantly elevated in the mother.
Genotypic studies
Single strand conformation polymorphism analysis of DNA fragments amplified from genomic DNA of the two affected children revealed a band shift in the PCR product of exon 2 that was not present in fragments from normal individuals. Both parents showed a double band demonstrating heterozygosity. Sequence analysis of the HSD11B2 gene from the two patients yielded a homozygous deletion of six nucleotides in exon 2, resulting in the loss of amino acids Leu114 and Glu115. The father and mother, who were first degree cousins, were heterozygous for the same mutation. The mutation was absent in control individuals and was not detected in a mutational screening of individuals with essential hypertension (16).
In vitro expression analysis of mutations
Transfected HEK-293 cells were tested for their ability to convert
physiological concentrations of tritiated cortisol or corticosterone to
cortisone or 11-dehydrocorticosterone, respectively. Intact cells
efficiently converted cortisol with a Km of 461
nmol/L and a Vmax of 0.93 nmol/L·h/mg of total
protein and converted cortisosterone with a Km of
56 nmol/L and a Vmax of 0.77 nmol/L·h/mg total
protein. Expression of the 11ßHSD2 mutant 
L114,E115 resulted in an
enzyme with a catalytic efficiency
(Vmax/Km) of 9.5% compared
with the wild-type enzyme using cortisol and 3.9% with corticosterone
as substrate. Kinetic parameters in cell lysates could not be
calculated for the deletion mutant L114,E115 due to its very low
activity. Immunoblotting identified the presence of an immunoreactive
mutated protein of identical molecular weight as the normal enzyme and
revealed approximately equal levels of expression of mutant and
wild-type enzymes (Fig. 1).
|
L114,E115, the two mutant enzymes, E115Q and E115K,
converted cortisol with Vmax values that were
increased to 135% and 158% of wild-type maximal activity,
respectively. Mutant enzymes E115Q and E115K displayed
Km values for the substrate corticosterone lower
than that of wild-type 11ßHSD2, but Vmax values
were clearly increased. In cell lysates the cofactor preference of
11ßHSD2 for the wild-type and mutant enzymes was measured using
corticosterone as a substrate in the presence of either NAD or NADP.
The wild-type enzyme converted corticosterone with a
Km of 4.4 nmol/L and a Vmax
of 1.34 nmol/L·h/mg total protein when NAD was used as a cofactor
(Table 2). A comparable
Vmax of 1.57 nmol/L·h/mg total protein, but a
significantly increased Km of 159 nmol/L was
obtained when the cofactor NADP was used (Table 2). Both deletion
mutants, E115Q and E115K, were stably expressed to a similar extent as
the wild-type enzyme (Fig. 1) and converted corticosterone in the
presence of NAD with a Km comparable to that of
wild-type enzyme and a slightly increased Vmax
(Table 2). In contrast, the two mutant enzymes showed a higher apparent
affinity for corticosterone, and Vmax was
significantly enhanced to 135% and 155% of wild-type activity for
mutants E115Q and E115K in the presence of the cofactor NADP (Table 2).
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
The heterozygous parents were normotensive with normal serum
electrolytes but low aldosterone and a mild defect in peripheral
cortisol to cortisone metabolism. Previously, Li et al.
(22) reported that the heterozygous father of an AME
patient with the mutation A328V had a slightly increased
(THF+5THF)/THE ratio, whereas Morineau et al.
(23) found normal steroid metabolites in the heterozygous
mother and brother of an unrelated family with the A328V mutation.
Heterozygous individuals have one wild-type and one mutated copy of the
HSD11B2 gene. As 11ßHSD2 functions as a dimer (24), the
observed slight reduction in cortisol to cortisone metabolism in
heterozygous parents could result from either haplotype insufficiency
or a dominant negative effect of the mutation. Evidence for the latter
was suggested previously in a study in vitro using
coexpression of wild-type and R337C mutant plasmids of 11ßHSD2
(25). Nevertheless, it is difficult to extrapolate from
the in vitro data the effect on enzymatic activity of
11ßHSD2 in vivo in heterozygous individuals, as
differences in the genetic background, environment, or compensatory
mechanisms for the reduced 11ßHSD2 protein might modulate a direct
relationship between in vitro and in vivo
findings. Methodological limitations in the measurement of the urinary
steroid profile should also be considered for the contrasting reports
mentioned (22, 23).
A three-dimensional structural model of 11ßHSD2 based on the known
crystal structures of other members of the short-chain dehydrogenase
reductase family indicates a critical role of the deleted
Glu115 in cofactor binding (19). To
elucidate whether the loss of this negative charge is responsible for
the loss of function, or whether the deletion of both
Leu114 and Glu115 residues
resulted in a disturbed cofactor binding, we compared L114,E115 with
two additional mutants, E115Q and E115K. As attempts to purify the
membrane-anchored 11ßHSD2 have failed to date, direct cofactor
binding studies were not possible. Therefore, the effects of mutations
in Glu115 on cofactor binding and specificity
were investigated indirectly. Experiments with cell lysates using
saturating cofactor concentrations showed that the deletion mutant
L114,E115 was inactive on cell lysis, without any evidence for
protein degradation, suggesting an irreversible conformational change
leading to an inactive protein. Both mutant E115Q and E115K remained
stable in cell lysates and exhibited a decrease in apparent
Km and an increase in Vmax
for corticosterone when the cofactor NADP was used. In contrast, when
the cofactor NAD was used, the activities of E115Q and E115K were
similar to wild-type activity, with a tendency to increased
Vmax for the mutants. Thus, the loss of the
negative charge at position 115 favors conversion of 11ß-hydroxy to
11-keto substrate, and the deletion of Leu114 and
Glu115 disturbs the conformation of the
cofactor-binding site, making the electron transfer from the substrate
to the cofactor less efficient. The kinetic data obtained from whole
cell measurements are highly variable between different groups, with
Km values for cortisol between 60 nmol/L
(24) and 620 nmol/L (20). The observed
differences in apparent Km values may be
explained in part by differences in the experimental procedures and
cell lines used for transfection experiments.
The analysis of other members of the short-chain dehydrogenase
reductase family of proteins demonstrates a critical role of charged
amino acids between ß-strand B and -helix C of the conserved
ßß structure (Fig. 2)
(19, 26, 27, 28). Enzymes preferring the cofactor NAD, such as
11ßHSD2, 17ßHSD2, 11-cis-retinol dehydrogenase, and
alcohol dehydrogenase, contain a negatively charged amino acid residue
at this position that causes a repulsion of the negatively charged
2'-phosphate group of NADP. In contrast, enzymes preferring NADP, such
as 11ßHSD1, 17ßHSD1, and retinol dehydrogenase, have a positively
charged residue between ß-strand B and -helix C that stabilizes
the negatively charged 2'-phosphate group of NADP and by that shields
any repulsive effect of negatively charged residues in the
neighborhood. The observed effects of the substitutions of
Glu115 by Gln or Lys of 11ßHSD2 may be
explained by the loss of repulsion of the 2'-phosphate group of NADP.
However, residue Asp91 of 11ßHSD2, which is
located in analogous position to Ser111 of
17ßHSD1, a residue shown to interact with the 2'-phosphate of NADP
(28), is expected to lead to a repulsion of the
2'-phosphate group, even in the absence of
Glu115.
|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received May 30, 2000.
Revised August 21, 2000.
Revised October 5, 2000.
Accepted December 4, 2000.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  A. G. Atanasov, I. D. Ignatova, L. G. Nashev, B. Dick, P. Ferrari, F. J. Frey, and A. Odermatt Impaired Protein Stability of 11beta-Hydroxysteroid Dehydrogenase Type 2: A Novel Mechanism of Apparent Mineralocorticoid Excess J. Am. Soc. Nephrol., April 1, 2007; 18(4): 1262 - 1270. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Morineau, V. Sulmont, R. Salomon, B. Fiquet-Kempf, X. Jeunemaitre, J. Nicod, and P. Ferrari Apparent Mineralocorticoid Excess: Report of Six New Cases and Extensive Personal Experience J. Am. Soc. Nephrol., November 1, 2006; 17(11): 3176 - 3184. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Paterson, J. R. Seckl, and J. J. Mullins Genetic manipulation of 11{beta}-hydroxysteroid dehydrogenases in mice Am J Physiol Regulatory Integrative Comp Physiol, September 1, 2005; 289(3): R642 - R652. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Draper and P. M Stewart 11{beta}-Hydroxysteroid dehydrogenase and the pre-receptor regulation of corticosteroid hormone action J. Endocrinol., August 1, 2005; 186(2): 251 - 271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. A. Carvajal, A. A. Gonzalez, D. G. Romero, A. Gonzalez, L. M. Mosso, E. T. Lagos, M. d. P. Hevia, M. P. Rosati, T. O. Perez-Acle, C. E. Gomez-Sanchez, et al. Two Homozygous Mutations in the 11{beta}-Hydroxysteroid Dehydrogenase Type 2 Gene in a Case of Apparent Mineralocorticoid Excess J. Clin. Endocrinol. Metab., June 1, 2003; 88(6): 2501 - 2507. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Vogt, B. M. Frey, and F. J. Frey Reply Nephrol. Dial. Transplant., March 1, 2003; 18(3): 620 - 621. [Full Text] [PDF]
|
|
|
|
|
|
M. G. Mohaupt and F. J. Frey Mineralocorticoid receptor malfunction: further insights from rare forms of hypertension Nephrol. Dial. Transplant., April 1, 2002; 17(4): 539 - 542. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Odermatt, P. Arnold, and F. J. Frey The Intracellular Localization of the Mineralocorticoid Receptor Is Regulated by 11beta -Hydroxysteroid Dehydrogenase Type 2 J. Biol. Chem., July 20, 2001; 276(30): 28484 - 28492. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
