Aminopeptidase A, which generates one of the main effector peptides of the brain renin-angiotensin system, angiotensin III, has a key role in central control of arterial blood pressure.

Overactivity of the brain renin-angiotensin system (RAS) has been implicated in the development and maintenance of hypertension in several experimental animal models. We have recently reported that, in the murine brain RAS, angiotensin II (AngII) is converted by aminopeptidase A (APA) into angiotensin III (AngIII),which is itself degraded by aminopeptidase N (APN), both peptides being equipotent to increase vasopressin release and arterial blood pressure when injected by the intracerebroventricular (i.c.v.) route. Because AngII is converted in vivo into AngIII, the exact nature of the active peptide is not precisely known. To delineate their respective roles in the central control of cardiovascular functions, specific and selective APA and APN inhibitors are needed to block the metabolic pathways of AngII and AngIII respectively. In the absence of such compounds for APA, we first explored the organization of the APA active site by site-directed mutagenesis. This led us to propose a molecular mechanism of action for APA similar to that proposed for the bacterial enzyme thermolysin deduced from X-ray diffraction studies. Secondly, we developed a specific and selective APA inhibitor, compound EC33 [(S)-3-amino-4-mercaptobutylsulphonic acid], as well as a potent and selective APN inhibitor, PC18 (2-amino-4-methylsulphonylbutane thiol). With these new tools we examined the respective roles of AngII and AngIII in the central control of arterial blood pressure. A central blockade of APA with the APA inhibitor EC33 suppressed the pressor effect of exogenous AngII, suggesting that brain AngII must be converted into AngIII to increase arterial blood pressure. Furthermore, EC33, injected alone i.c.v. but not intravenously, caused a dose-dependent decrease in arterial blood pressure by blocking the formation of brain AngIII but not systemic AngIII. This is corroborated by the fact that the selective APN inhibitor PC18 administered alone via the i.c.v. route increased arterial blood pressure. This pressor response was blocked by prior treatment with the angiotensin type 1 receptor antagonist losartan, showing that blocking the action of APN on AngIII metabolism leads to an increase in endogenous AngIII levels, resulting in arterial blood pressure increase through an interaction with angiotensin type 1 receptors. These results demonstrate that AngIII is a major effector peptide of the brain RAS, exerting a tonic stimulatory control over arterial blood pressure. Thus APA, the enzyme responsible for the formation of brain AngIII, represents a potential central therapeutic target that justifies the development of APA inhibitors, crossing the blood-brain barrier, as central anti-hypertensive agents.

Cloning and expression pattern of EPAS1 in the chicken embryo. Colocalization with tyrosine hydroxylase.

EPAS1 is a hypoxia-inducible transcription factor, highly expressed in vasculature and recently shown to be necessary for catecholamine production during embryogenesis. We report here the cloning and detailed expression pattern of this factor in the chicken embryo. We show that chicken EPAS1 presents an overall identity of 76% with the human sequence and that it is strongly expressed in the blood vessel wall, mostly in endothelial cells, but also in vascular smooth muscle cells. Moreover, we report non-vascular expression sites: liver, kidney, and, quite interestingly, cells of the sympathetic nervous system where EPAS1 is coexpressed with one of its putative target genes, the tyrosine hydroxylase. EPAS1 could therefore represent the link between the vascular system and the sympathetic nervous system, both sensitive to hypoxia.

PC18, a specific aminopeptidase N inhibitor, induces vasopressin release by increasing the half-life of brain angiotensin III.

Angiotensin III (AngIII), which is metabolized in vivo by aminopeptidase N (APN), was previously shown to be one of the main effector peptides of the brain renin-angiotensin system (RAS) in the control of vasopressin release. Recently, a potent APN inhibitor, PC18 (2-amino-4-methylsulfonyl butane thiol, methionine thiol), has been developed. In this study, we first checked the in vitro selectivity of PC18 towards APN, aminopeptidase A (APA) and aminopeptidase B (APB), three zinc metalloproteases with significant identity between their amino acid sequences. The Ki values of this compound on APN were found to be in the nanomolar range (Ki = 8.0 +/- 1.7 nM) but it was 2,150 and 125 times less active on APA and APB, respectively. Secondly, we evaluated in vivo the effect of brain APN inhibition with PC18 on the inactivation of brain AngIII and on vasopressin secretion in mice. For this purpose, mice received [3H]AngII intracerebroventricularly in the presence or absence of the APN inhibitor PC18 (30 microg). At different times after the injection, [3H]AngIII levels were evaluated from hypothalamus homogenates after separation by cation-exchange chromatography. PC18 induced an accumulation of [3H]AngIII, increasing its half-life 3.9 times as compared with control values. In addition, the effect of PC18 on vasopressin release was studied in mice. PC18 (10-100 microgram) was injected intracerebroventricularly, and plasma vasopressin levels were estimated by radioimmunoassay. PC18 increased vasopressin levels in a dose-dependent manner. The maximal increase in vasopressin release (+220%) is observed for a dose of PC18 of 100 microgram and was inhibited 75% by the coadministration of the AngII receptor antagonist (Sar1-Ala8)-AngII (0.5 microgram). These results indicate that in vivo, in the mouse brain, APN inhibition by PC18 increases the half-life of endogenous AngIII, resulting in an enhanced vasopressin release.

Production and properties of a recombinant soluble form of aminopeptidase A.

Aminopeptidase A is a homodimeric membrane-bound zinc metallopeptidase anchored at the plasma membrane by a 22-amino-acid hydrophobic segment. The anchor segment separates a small N-terminal cytoplasmic domain from a large ectodomain that contains the active site. Site-directed mutagenesis was performed to investigate the role of the cytoplasmic domain of aminopeptidase A in membrane anchoring and routing of the enzyme. Expression in COS-7 cells of a mutant lacking the N-terminal cytoplasmic domain resulted in the efficient secretion of a catalytically active enzyme in the medium. The soluble mutated aminopeptidase A, purified from the medium of a stable cell line, exhibited similar biochemical features to those of the wild-type enzyme. Pulse/chase metabolic labeling experiments revealed that the soluble form is generated intracellularly at an early stage of biosynthesis, suggesting that the signal peptide/membrane anchor domain of aminopeptidase A is removed in the endoplasmic reticulum through the action of the signal peptidase.

Genotype-phenotype analysis of a newly discovered family with Liddle's syndrome.

OBJECTIVE: To investigate the clinical, biologic, and molecular abnormalities in a family with Liddle's syndrome and analyze the short- and long-term efficacies of amiloride treatment. PATIENTS: The pedigree consisted of one affected mother and four children, of whom three suffered from early-onset and moderate-to-severe hypertension. METHODS: In addition to the biochemical and hormonal measurements, genetic analysis of the carboxy terminus of the beta subunit of the epithelial sodium channel (beta ENaC) was conducted through single-strand conformation analysis and direct sequencing. The functional properties of the mutation were analyzed using the Xenopus expression system and compared with one mutation affecting the proline-rich sequence of the beta ENaC. RESULTS: Mild hypokalemia and suppressed levels of plasma renin and aldosterone were observed in all affected subjects. Administration of 10 mg/day amiloride for 2 months normalized the blood pressure and plasma potassium levels of all of the affected subjects, whereas their plasma and urinary aldosterone levels remained surprisingly low. A similar pattern was observed after 11 years of follow-up, but a fivefold increase in plasma aldosterone was observed under treatment with 20 mg/day amiloride for 2 weeks. Genetic analysis of the beta ENaC revealed a deletion of 32 nucleotides that had modified the open reading frame and introduced a stop codon at position 582. Expression of this beta 579del32 mutant caused a 3.7 +/- 0.3-fold increase in the amiloride-sensitive sodium current, without modification of the unitary properties of the channel. A similar increase was elicited by one mutation affecting the carboxy terminus of the beta ENaC. CONCLUSIONS: This new mutation leading to Liddle's syndrome highlights the importance of the carboxy terminus of the beta ENaC in the activity of the epithelial sodium channel. Small doses of amiloride are able to control the blood pressure on a long-term basis in this monogenic form of hypertension.

A sensitive reverse transcriptase polymerase chain reaction assay for measuring the effects of dehydration and gestation on rat amounts of vasopressin and ocytocin mRNAs.

This study describes a competitive reverse transcriptase polymerase chain reaction (RT-PCR) method for assaying the amounts of vasopressin (AVP) and ocytocin (OT) mRNAs in the rat hypothalamus and uterus. Despite the low concentrations of these mRNAs, the RT-PCR method readily measured both AVP and OT mRNAs in the same sample. A common internal standard for both reactions was designed to quantify the reaction. Both AVP and OT mRNAs were readily quantified in a 75 ng sample of total RNA from the hypothalamus. Water deprivation stimulated AVP mRNA production 3-fold and OT mRNA production 1.7-fold in the hypothalamus. Gestation only influenced the amount of OT mRNA in the hypothalamus (3-fold increase) and uterus (38-fold increase). The amount of AVP mRNA in the hypothalamus remained unchanged and no AVP mRNA was detected in the uteri of either non-pregnant or pregnant rats. This competitive RT-PCR is a powerful tool that provides rapid and precise assays of AVP and OT mRNAs.

The amino acid substitutions Ser288Gly and Val320Ala convert the cortisol producing enzyme, CYP11B1, into an aldosterone producing enzyme.

Transfection studies with cDNAs encoding hybrids between the highly similar cytochrome P450 enzymes, CYP11B1 (steroid 11 beta-hydroxylase) and CYP11B2 (aldosterone synthase) have identified which amino acids determine the different activities of the enzymes.

Molecular cloning, expression and tissue distribution of a chicken angiotensin II receptor.

A cDNA encoding a chicken angiotensin II receptor from adrenal gland was isolated to serve as a molecular tool to study the role of AngII in avian embryonic development. This cDNA, sharing a high homology with another avian receptor (turkey), encodes a protein of 359 amino acids with 75% sequence identity with the mammalian type 1 receptor. Transient expression has revealed pharmacological properties distinct from mammalian receptors and a functional coupling leading to the increase in inositol phosphate production. The AngII receptor mRNA is expressed in classical target organs for AngII (adrenal gland, heart, kidney) and, interestingly, in endothelial cells where it may mediate the peculiar vasorelaxation effect of AngII in the chicken.

Molecular characterization of the mineralocorticoid receptor in pseudohypoaldosteronism.

Pseudohypoaldosteronism (PHA) is characterized by salt-wasting and failure to thrive in the newborn, accompanied by high urinary levels of sodium despite hyponatremia, hyperkalemia and metabolic acidosis, elevation of plasma renin activity, and high plasma aldosterone levels. PHA patients are resistant to mineralocorticoid administration, but their symptoms ameliorate after a period of sodium supplementation, which can be discontinued in older subjects. Binding studies performed on mononuclear leukocytes of the family members affected by the disease have shown the absence of binding of [3H]aldosterone to the mineralocorticoid receptor (MR) in mononuclear leukocytes in two siblings and a marked reduction in another sibling and the father, suggesting either the absence of MR or a defect in the ligand binding domain of the MR in these patients. Molecular analysis of the MR in the members of this family did not reveal any major rearrangement or deletion of the MR gene. In addition, no mutation was found in the entire MR coding sequence by RT-PCR and direct sequencing of MR mRNA, and the semiquantitative RT-PCR analysis of the MR mRNA of one affected patient failed to show any quantitative abnormality in MR expression. These results do not exclude a molecular abnormality present in the MR gene being responsible for PHA. However, they indicate that in this family PHA is not related to a modification of the MR primary structure or to a major abnormality in MR expression.

Identification of two active site residues in human angiotensin I-converting enzyme.

Angiotensin I-converting enzyme (ACE) contains two zinc-dependent catalytic domains (N and C domains) each bearing the motif HEXXH where the two histidines form two of the three amino acid zinc ligands. Sequence alignment of each ACE domain with other zinc metalloproteases, indicates a glutamate residues which putatively constitutes the third zinc ligand and an aspartate residue which may form an indirect zinc interaction. We investigated the functional roles of the glutamate and aspartate residues in the ACE C domain (Glu987 and Asp991) using a cDNA encoding an inactive N domain. We mutated Glu987 to aspartate (E987D) or valine (E987V) and Asp991 to glutamate (D991E) or alanine (D991A). Catalytically active mutants (E987D, D991E and D991A) exhibited similar Km values for hippuryl-His-Leu compared to non-mutated C domain. E987D displayed a 300-fold decrease in kcat and a 25-fold reduction in sensitivity to the ACE inhibitor trandolaprilat, whose binding is zinc-dependent. E987V was catalytically inactive and did not bind [3H]trandolaprilat. D991E and D991A exhibited a 3.8- and 22-fold decrease in kcat, respectively, and the Ki' values for trandolaprilat were increased 8- and 29-fold. These results provide strong evidence that Glu987 constitutes the third zinc ligand in the ACE C domain and suggest a role for Asp991 in positioning the C domain active site zinc ion.

Synthetic cDNA encoding the rat AT1a receptor: a useful tool for structure-function relationship analysis.

To carry out systematic structure-function studies of the rat angiotensin II receptors by site directed mutagenesis, or production of chimeric receptors, we have produced a synthetic cDNA coding for the AT1a receptor. The synthetic cDNA is 1101 base pairs long, and contains 49 unique restriction sites that are on the average 23 base pairs apart, allowing replacement of specific restriction fragments by synthetic counterparts containing the desired modified sequence. The total cDNA was assembled in the expression vector pECE. After stable expression in Chinese Hamster Ovary cells, the protein encoded by this synthetic cDNA presents a pharmacological profile and a signal transduction mechanism indistinguishable from the wild type rat AT1a receptor.

Arginine vasopressin gene expression in chronic cardiac failure in rats.

Arginine vasopressin (AVP) is known to be increased in patients and experimental animals with chronic cardiac failure (CCF). The importance of an increase in biosynthesis of AVP in the hypothalamus has, however, not heretofore been investigated and is the purpose of the present study. CCF secondary to infarction of myocardial tissue was induced by ligation of the left anterior descending coronary artery and sham operated animals served as controls. Four weeks later hypothalamic AVP mRNA was determined by solution hybridization using sense and anti-sense strand RNA. The blood pressure was lower in CCF than sham animals (131.2 +/- 3.1 vs. 112.8 +/- 4.0 mm Hg, P less than 0.05) and the total heart, and right and left ventricle weights were significantly higher in CCF rats. Plasma AVP was higher in CCF (sham 6.78 +/- 0.30; CCF 11.46 +/- 0.64 pg/ml, P less than 0.001) and plasma atrial natriuretic peptide was also higher in CCF than sham animals (205 +/- 36 vs. 554 +/- 56 pg/ml, P less than 0.001). The AVP mRNA in hypothalamus was significantly higher in CCF than sham animals (55.5 +/- 3.7 vs. 95.9 +/- 4.0 pg/micrograms total RNA, P less than 0.001). There was no difference in beta-actin mRNA in the hypothalamus of sham and CCF rats, indicating that the AVP-mRNA increase was specific in CCF. These results therefore demonstrate that increased AVP biosynthesis in the hypothalamus, in addition to release of the hormone from the posterior pituitary, may occur in CCF.

Cytoplasmic and nuclear uptake of aldosterone in toad bladder: a mathematical modeling approach.

The mechanism of aldosterone uptake in the epithelial cells of toad bladder was studied using mathematical modeling. Two complementary approaches were used. The first involved analysis of cytosolic aldosterone binding at steady state according to models defined by the sum of independent noninteractive binding sites. The best model describing the experimental data corresponded to two specific binding sites with mean dissociation constant values of 0.20 and 60 nM for types 1 and 2, respectively. The second approach was based on the analysis of cytoplasmic and nuclear aldosterone uptake kinetics at 25 and 0 degrees C in intact bladder. Two models (A and B) were studied. They both implied the existence of two types of aldosterone binding sites as precursors of the corresponding chromatin bound complexes. In model A, nuclear translocation of the two types of receptors was assumed to obey first-order kinetics. In model B, the translocation process for type 1 sites involved a time lag leading to delayed binding to chromatin. Both models were found to fit the experimental data satisfactorily. The fit obtained for model B appeared to be better at low aldosterone concentrations.

Antihypertensive effect of fractionated sublingual administration of nifedipine in moderate essential hypertension.

The magnitude and duration of the antihypertensive effect of nifedipine were studied in 7 cases of moderate essential hypertension. In a double-blind crossover study, nifedipine 10 mg or a placebo were administered sublingually 4 times a day for 2 days, and the results were compared. Each dose of nifedipine reduced systolic and diastolic blood pressure by 14% both in the supine and upright positions. The antihypertensive action lasted for about 3 h and it was not cumulative. The reduction in blood pressure was associated with a temporary increase in heart rate. Administration of nifedipine 10 mg did not significantly raise plasma renin activity or plasma aldosterone. The drug was well tolerated and no side effects were detected.

Binding of glycyrrhetinic acid to kidney mineralocorticoid and glucocorticoid receptors.

Whether the mineralocorticoid effect of glycyrrhetinic acid is mediated by the adrenal glands or is due to a direct action on the renal tubule remains controversial. The affinity of glycyrrhetinic acid for mineralocorticoid receptors has been studied by several types of competition experiments. When rat kidney slices were incubated with 2 times 10- minus 9 M [3H]aldosterone, glycyrrhetinic acid (2 times 10- minus 5 M) was able to compete with aldosterone for the cytosolic receptor and to decrease the formation of a chromatin-[3Hi1 aldosterone-receptor complex. In cytosol, in vitro, 6 times 10- minus 4 M glycyrrhetinic acid was able to inhibit aldosterone binding by 70 percent, whereas the same dose produced only a 20 percent inhibition of dexamethasone binding. The apparent KDiss of glycyrrhetinic acid for the mineralocorticoid receptor was 2 times 10- minus 6 M. That glycyrrhetinic acid appeared to compete mainly with mineralocorticoid receptors was confirmed by sedimentation in the sucrose gradients: [3H]Aldosterone specifically bound to an 8 S peak was displaced by 5 times 10- minus 5 M glycyrrhetinic acid, whereas the [3H]dexamethasone peak was not affected by this compound. Glycyrrhizic acid showed no significant affinity for mineralocorticoid or glucocorticoid kidney receptor sites. Glycyrrhetinic acid and glycyrrhizic acid had no affinity for rat cortisol binding globulin. Glycyrrhetinic acid has a low but definite affinity for mineralocorticoid receptors and thus appears to have a direct mineralocorticoid action. The low affinity of this compound for mineralocorticoid receptors is in good agreement with the very high doses required to exhibit its biological activity.