Primary systemic carnitine deficiency is caused by mutations in a gene encoding sodium ion-dependent carnitine transporter.

Primary systemic carnitine deficiency (SCD; OMIM 212140) is an autosomal recessive disorder characterized by progressive cardiomyopathy, skeletal myopathy, hypoglycaemia and hyperammonaemia. SCD has also been linked to sudden infant death syndrome. Membrane-physiological studies have suggested a defect of the carnitine transport system in the plasma membrane in SCD patients and in the mouse model, juvenile visceral steatosis. Although the responsible loci have been mapped in both human and mouse, the underlying gene has not yet been identified. Recently, we cloned and analysed the function of a novel transporter protein termed OCTN2. Our observation that OCTN2 has the ability to transport carnitine in a sodium-dependent manner prompted us to search for mutations in the gene encoding OCTN2, SLC22A5. Initially, we analysed the mouse gene and found a missense mutation in Slc22a5 in jvs mice. Biochemical analysis revealed that this mutation abrogates carnitine transport. Subsequent analysis of the human gene identified four mutations in three SCD pedigrees. Affected individuals in one family were homozygous for the deletion of a 113-bp region containing the start codon. In the second pedigree, the affected individual was shown to be a compound heterozygote for two mutations that cause a frameshift and a premature stop codon, respectively. In an affected individual belonging to a third family, we found a homozygous splice-site mutation also resulting in a premature stop codon. These mutations provide the first evidence that loss of OCTN2 function causes SCD.

In vivo desensitization of glycogenolysis to Ca2+-mobilizing hormones in rat liver cells.

Rat hepatocytes contain several types of Ca2+-linked receptors, all of which stimulate glycogen breakdown by increasing cytosolic free Ca2+ concentration [( Ca2+]c). In vivo desensitization of this Ca2+ messenger system was studied in hepatocytes isolated from either pheochromocytoma (PHEO)-harboring and chronically norepinephrine (NE)-infused rats. Homologous desensitization for alpha 1-adrenergic receptor-mediated phosphorylase activation developed in the early stage of PHEO rats (3-4 wk after implantation), whereas, in the later stage of tumor development or in the NE-infused rats, phosphorylase responses to all Ca2+-mobilizing stimulations were subsensitive (heterologous desensitization). In the homologous desensitization, the [Ca2+]c response to alpha 1-adrenergic stimulation was selectively reduced. We found, using the phenoxybenzamine inactivation method, that there was a linear relationship between alpha 1 receptor density and the [Ca2+]c response; consequently, the blunted [Ca2+]c response to alpha 1-adrenergic stimulation could not be explained by the 34% downregulation of alpha 1 receptors seen in these rats. These results indicated that uncoupling at a step proximal to alpha 1 receptor-stimulated [Ca2+]c increase is also of primary importance in homologous desensitization of phosphorylase activation. On the other hand, heterologous desensitization also involved alteration(s) at steps distal to the rise in [Ca2+]c. Our data demonstrate that prolonged exposure to catecholamines results in desensitization of the [Ca2+]c mobilization pathway and may involve multiple mechanisms.

Alpha1-adrenoceptors are required for normal male sexual function.

BACKGROUND AND PURPOSE: Alpha(1)-adrenoceptor antagonists are extensively used in the treatment of hypertension and lower urinary tract symptoms associated with benign prostatic hyperplasia. Among the side effects, ejaculatory dysfunction occurs more frequently with drugs that are relatively selective for alpha(1A)-adrenoceptors compared with other drugs of this class. This suggests that alpha(1A)-adrenoceptors may contribute to ejaculation. However, this has not been studied at the molecular level. EXPERIMENTAL APPROACH: The physiological contribution of each alpha(1)-adrenoceptor subtype was characterized using alpha(1)-adrenoceptor subtype-selective knockout (KO) mice (alpha(1A)-, alpha(1B)- and alpha(1D)-AR KO mice) since the subtype-specific drugs available are only moderately selective. We analysed the role of alpha(1)-adrenoceptors in the blood pressure and vascular response as well as ejaculation by determining these variables in alpha(1)-adrenoceptor subtype-selective KO mice and in mice with all their alpha(1)-adrenoceptor subtypes deleted (alpha(1)-AR triple-KO mice). KEY RESULTS: The pregnancy rate was reduced by 50% in alpha(1A)-adrenoceptor KO mice, and this reduction was dramatically enhanced in alpha(1)-adrenoceptor triple-KO mice. Contractile tension of the vas deferens in response to noradrenaline was markedly decreased in alpha(1A)-adrenoceptor KO mice, and this contraction was completely abolished in alpha(1)-adrenoceptor triple-KO mice. This attenuation of contractility was also observed in the electrically stimulated vas deferens. CONCLUSIONS AND IMPLICATIONS: These results demonstrate that alpha(1)-adrenoceptors, particularly alpha(1A)-adrenoceptors, are required for normal contractility of the vas deferens and consequent sperm ejaculation as well as having a function in fertility.

Selective production of transgenic mice using green fluorescent protein as a marker.

The low efficiency of transgenic animal production by microinjection has been a serious problem especially for the production of transgenic livestock. We developed a method to selectively produce transgenic mice using green fluorescent protein (GFP) as a marker. Using this method, we obtained eight fetuses and four live-born mice derived from 55 GFP-positive blastocysts. PCR analysis showed 11 out of 12 mice (fetuses and newborn mice) were transgenic. Southern blot analysis showed that 8 out of 12 were transgenic. GFP expression was also observed in bovine blastocysts, suggesting that this method should contribute to the efficient production of transgenic livestock.

Body water balance and body temperature in vasopressin V1b receptor knockout mice.

In an attempt to determine whether there is a specific vasopressin receptor (V(1b)) subtype involved in the regulation of body water balance and temperature, vasopressin V(1b) receptor knockout mice were used. Daily drinking behavior and renal excretory function were examined in V(1b)-deficient (V(1b)(-/-)) and control (V(1b)(+/+)) mice under the basal and stress-induced condition. In addition, body temperature and locomotor activity were measured with a biotelemetry system. The baseline daily water intake and urine volume were larger in V(1b)(-/-) mice than in V(1b)(+/+) mice. V(1b)(-/-) mice (V(1b)(-/-)) had significantly higher locomotor activity than wild-type, whereas the body temperature and oxygen consumption were lower in V(1b)(-/-) than in the V(1b)(+/+) mice. Next, the V(1b)(-/-) and V(1b)(+/+) mice were subjected to water deprivation for 48 hr. Under this condition, their body temperature decreased with the time course, which was significantly larger for V(1b)(-/-) than for V(1b)(+/+) mice. Central vasopressin has been reported to elicit drinking behavior and antipyretic action, and the V(1b) receptor has been reported to be located in the kidney. Thus, the findings suggest that the V(1b) receptor may be, at least in part, involved in body water balance and body temperature regulation.

Effects of arachidonic acid on FFA4 receptor: Signaling, phosphorylation and internalization.

Arachidonic acid increased intracellular calcium, in cells expressing green fluorescent protein-tagged human FFA4 receptors, with an EC50 of ~40µM. This action was not blocked by cyclooxygenase or lipoxigenase inhibitors but it was inhibited by AH7614, a FFA4 antagonist. Arachidonic acid induced ERK activation accompanied by EGF receptor transactivation. However, EGF transactivation was not the major mechanism through which the fatty acid induced ERK phosphorylation, as evidenced by the inability of AG1478 to block it. Arachidonic acid increased FFA4 receptor phosphorylation that reached its maximum within 15min with an EC50 of ~30µM; inhibitors of protein kinase C partially diminish this effect and AH7614 blocked it. Arachidonic acid induced rapid and sustained Akt/PKB phosphorylation and FFA4 - ß-arrestin interaction. Confocal microscopy evidenced that FFA4 receptor activation and phosphorylation were associated to internalization. In conclusion, arachidonic acid is a bona fide FFA4 receptor agonist.

Cloning and characterization of the promoter of murine cytohesin-1 gene.

Mouse cytohesin-1 (CTH-1) is a guanine nucleotide exchange factor, specific for ADP ribosylation factors. The CTH-1 gene promoter was characterized by deletion mapping and mutational analysis. The region between -101 and -38 relative to the transcription start site showed the essential promoter activity when transfected into both NIH3T3 and COS7 cells. The nucleotides of the core promoter region contain three tandem GC boxes, which can offer potential binding sites for the ubiquitous transcription factor Sp-1 family. Mutational analysis revealed that these tandem GC boxes are indispensable for activation of the gene transcription.

Real-time optical monitoring of ligand-mediated internalization of alpha1b-adrenoceptor with green fluorescent protein.

The study of G protein-coupled receptor signal transduction and behavior in living cells is technically difficult because of a lack of useful biological reagents. We show here that a fully functional alphalb-adrenoceptor tagged with the green fluorescent protein (alphalbAR/GFP) can be used to determine the molecular mechanism of intemalization of alphalbAR/ GFP in living cells. In mouse alphaT3 cells, alpha1bAR/GFP demonstrates strong, diffuse fluorescence along the plasma membrane when observed by confocal laser scanning microscope. The fluorescent receptor binds agonist and antagonist and stimulates phosphatidylinositol/Ca2+ signaling in a similar fashion to the wild receptor. In addition, alpha1bAR/ GFP can be internalized within minutes when exposed to agonist, and the subcellular redistribution of this receptor can be determined by measurement of endogenous fluorescence. The phospholipase C inhibitor U73,122, the protein kinase C activator PMA, and inhibitor staurosporine, and the Ca2+-ATPase inhibitor thapsigargin were used to examine the mechanism of agonist-promoted alphalbAR/GFP redistribution. Agonist-promoted internalization of alphalbAR/GFP was closely linked to phospholipase C activation and was dependent on protein kinase C activation, but was independent of the increase in intracellular free Ca2+ concentration. This study demonstrated that real-time optical monitoring of the subcellular localization of alphalbAR (as well as other G protein-coupled receptors) in living cells is feasible, and that this may provide a valuable system for further study of the biochemical mechanism(s) of agonist-induced receptor endocytosis.

Chloroethylclonidine reveals that alpha (1 A)-adrenoceptors mediate contraction in aorta of alpha (1 D)-adrenoceptor knockout mice.

1 We have characterized the alpha(1)-adrenoceptor subtypes present in isolated aorta of the alpha(1D)-adrenoceptor knockout (KO) mice, by chloroethylclonidine (CEC)-induced alkylation and their protection by selective alpha(1)-adrenoceptor antagonists. 2 The alpha(1D)-adrenoceptor is involved in the contractile response to noradrenaline in wild type (WT) mouse aorta. 3 In WT mice 5-methylurapidil (5-MU, an alpha(1A)-adrenoceptor antagonist) or BMY 7378 (8-[2-[4-(2-methoxyphenyl)-1-piperazinyl] ethyl]-8-azaspiro[4.5] decane-7,9 dione, a selective alpha(1D)-adrenoceptor antagonist), protected the receptors from CEC-induced (alpha(1B/D)-adrenoceptor) alkylation, the combination of both antagonists resulted in complete protection, while AH11110A (1-[biphenyl-2-yloxy]-4-imino-4-piperidin-1-yl-butan-2-ol, an alpha(1B)-adrenoceptor antagonist) did not protect. 4 In aorta of KO mice there was a 19-fold rightward shift in noradrenaline effective concentration (EC(50)) compared with WT; while 5-MU alone or in combination with AH11110A protected alpha(1)-adrenoceptors to the same extent. 5 The data indicate that alpha(1A)-adrenoceptors mediate contraction and suggest their role in maintaining homeostasis in the alpha(1D)-adrenoceptors KO mice.

Desensitization of beta-adrenergic receptors by pheochromocytoma.

Prolonged stimulation of cells by beta-adrenergic receptor agonists may lead to diminished responsiveness of the cells to subsequent activation by catecholamines. This phenomenon has been termed desensitization; the mechanism(s) for desensitization may involve an apparent loss in the number of beta-adrenergic receptors or an alteration in receptor-effector coupling. We have examined the consequences of prolonged stimulation of beta-adrenergic receptors in an interesting rat model harboring pheochromocytoma. New England Deaconess Hospital rats with transplanted pheochromocytomas developed systolic hypertension and plasma norepinephrine concentrations approximately 40-fold greater than controls. beta-Adrenergic receptors were quantitated in several tissues from controls and rats with transplanted pheochromocytoma using the beta-adrenergic receptor antagonist [125I]iodocyanopindolol. Down-regulation of beta 1-receptors was found in heart tissue (22.8 vs. 13.6 fmol/mg protein; P less than 0.001) and adipocytes (29,400 vs. 2,800 sites/cell; P less than 0.001). Also, maximal isoproterenol-stimulated cAMP accumulation in isolated adipocytes was diminished in pheochromocytomic animals (13.1 vs. 4.9 pmol cAMP/10(5) cells/min; P less than 0.05). Interestingly, there was no change in beta-receptors in lung and mesenteric artery, which predominantly contain beta 2-receptors. Furthermore, the competition curves of isoproterenol in the heart membranes from control and pheochromocytomic rats in the absence and presence of guanylylimidodiphosphate indicated uncoupling of the beta-adrenergic receptors in pheochromocytomic animals. Rats with pheochromocytoma secreting large amounts of norepinephrine provide a valuable model system for studying the in vivo development of desensitization.

Spatial and temporal regulation of protein expression by bldA within a Streptomyces lividans colony.

The bldA gene encodes the only tRNA for the UUA codon that, although dispensable in genes important for primary vegetative growth of Streptomyces spp., is important in genes that serve a regulatory purpose in the differentiation. To investigate this role further, the spatial and temporal expression profiles of the bldA-regulated and unregulated genes within a Streptomyces colony were examined using modified genes for the green fluorescent protein (gfp) as an expression-tag. A comparative study, based on computer-assisted quantitative analysis of the GFP fluorescence, revealed that the presence of TTA codons in gfp results in a temporal delay of translation and, consequently, changed the spatial pattern of the GFP expression within a colony, especially during early differentiation. The delay of GFP expression was undetectable at 60 h post-inoculation. These results provide the first extensive evidence that the bldA does indeed play a significant regulatory role during colony differentiation.

Identification of the sequence responsible for the nuclear localization of human Cdc6.

The Cdc6 is the essential protein for the initiation of DNA replication. Cdc6 is localized in the G1 nucleus, and abnormal nuclear localization of this protein induces irregular initiation of DNA replication. We identified here that amino acids K57 and R58 in the human Cdc6 protein play an important role in the nuclear localization of the protein. The fundamental features of the mechanism regulating the localization of Cdc6 seem to be maintained in yeast, Xenopus, and human, since the amino acid sequence surrounding K57 and R58, (S/T)PXKR(L/I), is conserved in these species. Substitution of amino acid residue S54 with E and not Q blocked partially the nuclear localization of the protein, implying that the phosphorylation at S54 is involved in the regulating mechanism of the cell cycle-dependent localization of Cdc6.

Key amino acids of vasopressin V1a receptor responsible for the species difference in the affinity of OPC-21268.

A non-peptide, vasopressin V1a receptor-selective antagonist, OPC-21268, exhibited a markedly higher affinity for the rat V1a receptor (Ki = 380 nM) than for the human V1a receptor (Ki = 140 microM). To delineate the region responsible for the high affinity binding of OPC-21268 for the rat V1a receptor, we have constructed a series of chimeric human and rat V1a receptors, and examined the chimeric and point-mutated receptors by competitive radioligand binding analysis. The results showed that the transmembrane domain (TMD) VI-VII of the vasopressin V1a receptor, in particular the amino acid residue Ala-342 in TMD VII, is the major component conferring the rat-selective binding of OPC-21268 to the V1a receptor.

Flow cytometry analysis of alpha1-adrenoceptor subtypes.

To characterize the alpha1-adrenoceptor subtypes, we developed a flow cytometry method using the fluorescent ligand BODIPY-FL prazosin and the anti-peptide antibody against the alpha1b-adrenoceptor amino terminus (designated 1B-N1-C) as probes. Three alpha1-adrenoceptors (alpha1a, alpha1b and alpha1d) expressed in CHO cells were detected by BODIPY-FL prazosin; however, only alpha1b-adrenoceptor subtype was detected by the anti-peptide antibody 1B-N1-C. Furthermore, the flow cytometry analysis with 1B-N1-C specifically identified alpha1b-adrenoceptor in native cells of hamster DDT1-MF2 cells, rat hepatocytes and cardiomyocytes.

Cloning, functional expression and tissue distribution of human alpha 1c-adrenoceptor splice variants.

We report the cloning and characterization of two isoforms of human alpha 1c-adrenoceptor cDNA (alpha 1c-2, alpha 1c-3). These isoforms are generated by alternative splicing and differ from the clone we previously isolated (alpha 1c-1) in their length and sequences of the C-terminal domain. Tissue distribution of mRNAs showed that these variants co-express with alpha 1c-1 in the human heart, liver, cerebellum and cerebrum. Despite the structural differences, functional experiments in transfected CHO cells showed that the three isoforms have similar ligand binding properties, and all couple with phospholipase C/Ca2+ signaling pathway.

Age-related change in adrenergic regulation of glycogen phosphorylase in rat hepatocytes.

The effects of development on adrenergic regulation of glycogenolysis were studied in rat liver. In isolated hepatocytes, the activation of glycogen phosphorylase by alpha-adrenergic stimulation decreased moderately with advancing age. However, activation by beta-adrenergic receptors more markedly declined and almost disappeared in isolated hepatocytes from 6-month-old rats. Furthermore, the ability to glucagon to stimulate phosphorylase activity and cAMP generation was found to decrease with increasing age. The age-related changes in the pattern of stimulation of glycogen phosphorylase by catecholamines in isolated hepatocytes were accompanied by parallels changes in the numbers of alpha 1- and beta-adrenoceptors in membranes prepared from the isolated hepatocytes. A progressive decrease in the total number of alpha 1-receptors measured with [3H]-prazosin and beta-receptors measured with [125I]cyanopindolol (CYP) was found with increasing age. The loss of beta-adrenergic receptors was much more dramatic. Our results suggest that age-related alterations of hepatic glycogen phosphorylase activation by catecholamines may in part be explained by the changes in the expression adrenergic receptors.

Blockade of the renin-angiotensin system in heart failure in conscious dogs.

OBJECTIVE: To study the different cardiac and renal hemodynamic effects of an angiotensin converting enzyme inhibitor and an angiotensin II receptor antagonist in experimental heart failure in conscious dogs. DESIGN AND METHODS: We compared the effects of the angiotensin converting enzyme inhibitor, captopril, with those of the angiotensin II (Ang II) subtype-1 receptor antagonist, losartan, on hemodynamics and hormonal changes in congestive heart failure by rapid ventricular pacing on conscious dogs. Furthermore, we characterized the Ang II receptors in canine heart, using the canine cardiac membrane fraction in heart failure. RESULTS: Acute intravenous administration of captopril improved the cardiac output by 19% (P < 0.01) but losartan did not, although blockade of the renin-angiotensin system by losartan (1.1 mumol/kg) or captopril (0.69 mumol/kg) induced similar changes in the plasma renin activity, norepinephrine and arginine vasopressin, and a similar decrease in mean arterial pressure (-10 mmHg). Renal blood flow was increased by either losartan or captopril. In the binding study, losartan produced a single displacement curve (IC50 = 0.25 mumol/l), while the Ang II subtype-2 (AT2) receptor antagonist PD123319 did not, indicating that the predominant Ang II receptor is type-1 (AT1) in canine heart. Neither the ratio of AT1 to AT2 receptors nor the receptor density changed with the development of heart failure. CONCLUSIONS: The lack of effect of losartan on cardiac output may be the result of its inability to block non-AT1 receptor-mediated Ang II activities adequately. Captopril may improve cardiac output by means of mechanisms not mediated by Ang II, such as locally increasing bradykinin.

Identification of alpha 1C-adrenergic receptor mRNA in bovine retinal pigment epithelium.

PURPOSE: To examine the localization of a novel alpha 1-adrenergic receptor subtype of alpha 1C receptor in the eye, we compared the amount of alpha 1C receptor transcripts in bovine retinal pigment epithelium (RPE) and in neural retina by employing reverse transcription of mRNA and the polymerase chain reaction (RT-PCR) assay. METHODS: RT-PCR assay with specific primers for the alpha 1C-adrenergic receptor and for beta-actin showed linear relationships between input quantity of RNA and the amount of amplified PCR products for the alpha 1C-adrenergic receptor and also for beta-actin, when PCR was conducted for 24 and 15 cycles, respectively. RESULTS: The RT-PCR assay demonstrated that the spontaneous expression level of the alpha 1C-adrenergic receptor mRNA was much higher in bovine RPE than in neural retina; the alpha 1C-adrenergic receptor/beta-actin ratio from RPE was 0.33 +/- 0.15 (n = 4), whereas that from neural retina was virtually zero. CONCLUSIONS: The RT-PCR assay is a sensitive semiquantitative method for a low abundance mRNA in a limited number of cells. Using the alpha 1C receptor as a model, we demonstrated the usefulness of this assay by showing the uneven distribution of the alpha 1C receptor transcripts in bovine RPE cells and neural retina.

Selectivity of the imidazoline alpha-adrenoceptor agonists (oxymetazoline and cirazoline) for human cloned alpha 1-adrenoceptor subtypes.

1. To investigate the structure-activity relationships of alpha-adrenoceptor agonists for the alpha 1-adrenoceptor subtypes, we have compared the imidazoline class of compounds, oxymetazoline and cirazoline, with the phenethylamine, noradrenaline, in their affinities and also in their intrinsic activities in Chinese hamster ovary (CHO) cells stably expressing the cloned human alpha 1-adrenoceptor subtypes (alpha 1a-, alpha 1b-, and alpha 1d-subtypes). 2. Radioligand binding studies with [125I]-HEAT showed that cirazoline and oxymetazoline had higher affinities at alpha 1a-subtype than at alpha 1b- and alpha 1d-subtypes, while noradrenaline had higher affinity at the alpha 1d-subtype than at alpha 1a- and alpha 1b-subtypes. 3. In functional studies, cirazoline caused transients of cytosolic Ca2+ concentrations ([Ca2+]i response) in a concentration-dependent manner and developed a maximal response similar to that to noradrenaline in CHO cells expressing the alpha 1a-subtype, while it acted as a partial agonist at alpha 1b- and alpha 1d-adrenoceptors. Oxymetazoline, on the other hand, was a weak agonist at alpha 1a-adrenoceptors, and has no intrinsic activity at the other subtypes. 4. Using the phenoxybenzamine inactivation method, the relationships between receptor occupancy and noradrenaline-induced [Ca2+]i response for alpha 1a- and alpha 1d-subtypes were found to be linear, whereas it was moderately hyperbolic for the alpha 1b-subtype, indicating the absence of receptor reserves in CHO cells expressing alpha 1a- and alpha 1d-subtypes while there exists a small receptor reserve for CHO cells expressing the alpha 1b-subtype. 5 In summary, our data obtained in cells exclusively expressing a single receptor subtype support the idea that the relative role of agonist affinity and intrinsic activity may vary depending on the subtype of alphal-adrenoceptor.

Effects of castration on contraction and alpha(1)-adrenoceptor expression in rat prostate.

1. The prostate function is regulated by androgens and alpha-adrenergic activity. Clinically, antiandrogens and/or alpha(1)-adrenergic antagonists are commonly used to treat symptomatic prostatic hypertrophy. To elucidate the effects of androgen deprivation on prostate contractility via alpha(1)-adrenoceptor, the characteristics and expression of alpha(1)-adrenoceptors were examined in castrated rats. 2. Isolated prostate strips from intact and castrated rats were subjected to a phenylephrine stimulated contraction. Prazosin (10 nM), [(3)H]-prazosin and phenoxybenzamine (3 - 300 nM) were used for inhibition assay, receptor characterization and partial alkylation of alpha-adrenoceptor, respectively. The mRNA content of three subtypes of alpha-adrenoceptors was determined by reverse transcription combined with polymerase chain reaction (RT - PCR). 3. Contractile response to phenylephrine increased in castrated rats, which could be explained by a relative increase of the stromal component. A lowered contraction potency was also noted in castrated rats. Receptor binding assay indicated minimal changes in the affinity or density of alpha(1)-adrenoceptor. Escalating alkylation of the alpha(1)-adrenoceptor population resulted in a rightward shift in the contraction-response curves before depressing maximal contractile force, and the suppression was detected at lower doses in castrated rats. RT - PCR study confirmed the expression of three types of alpha(1)-adrenoceptor, alpha(1a), alpha(1b) and alpha(1d)-adrenoceptors, in intact rat prostate, and revealed that alpha(1a)-adrenoceptor, but not alpha(1b) or alpha(1d)-adrenoceptors, was down-regulated in castrates. 4. The results show that androgen deprivation suppressed alpha(1)-adrenergic contractility of rat prostate strips, and the suppression was associated with down-regulation of receptor reserve for the alpha(1a)-adreneroceptor population expressed in intact rat prostate.