Urine concentrating defect in prostaglandin EP 1 -deficient mice

【摘要】  We investigated the role of the prostaglandin E 2 (PGE 2 ) EP 1 receptor in modulating urine concentration as it is expressed along the renal collecting duct where arginine-vasopressin (AVP) exerts its anti-diuretic activity, and in the paraventricular and supraoptic nuclei of the hypothalamus where AVP is synthesized. The urine osmolality of EP 1 -null mice (EP 1 -/- ) failed to match levels achieved by wild-type (WT) counterparts upon water deprivation (WD) for 24 h. This difference was reflected by higher plasma osmolality in WD EP 1 -/- mice. Along the collecting duct, the induction and subapical to plasma membrane translocation of the aquaporin-2 water channel in WD EP 1 -/- mice appeared equivalent to that of WD WT mice as determined by quantitative RT-PCR and immunohistochemistry. However, medullary interstitial osmolalities dropped significantly in EP 1 -/- mice following WD. Furthermore, urinary AVP levels of WD EP 1 -/- mice were significantly lower than those of WD WT mice. This deficit could be traced back to a blunted induction of hypothalamic AVP mRNA expression in WD EP 1 -/- mice as determined by quantitative RT-PCR. Administration of the AVP mimetic [deamino-Cys 1, D -Arg 8 ]-vasopressin restored a significant proportion of the urine concentrating ability of WD EP 1 -/- mice. When mice were water loaded to suppress endogenous AVP production, urine osmolalities increased equally for WT and EP 1 -/- mice. These data suggest that PGE 2 modulates urine concentration by acting at EP 1 receptors, not in the collecting duct, but within the hypothalamus to promote AVP synthesis in response to acute WD.

【关键词】  vasopressin urine concentration prostaglandin E EP receptor

URINE CONCENTRATION IS A HIGHLY regulated process requiring the coordinated actions of a number of hormones at several tissue locales ( 2 ). The body responds to acute volume depletion, infection, or hyperosmolality by stimulating the production of antidiuretic hormone, arginine-vasopressin (AVP), within the hypothalamic paraventricular (PVN) and supraoptic (SON) neurons. AVP is then processed, stored, and secreted into the venous flow by the posterior pituitary to interact with renal V 2 receptors located in the principal cells of the collecting duct to promote aquaporin-2 (AQP2)-dependent water reabsorption ( 25 ). Many of these steps are subject to modulation by endocrine factors such as ANG II, catecholamines, as well as certain metabolites of the cyclooxygenase (COX) pathway, the prostaglandins (PGs) ( 2 ). Accordingly, renal prostaglandin E 2 (PGE 2 ) antagonizes the actions of AVP since NSAIDs transiently enhance urine concentration ( 1 ). Direct evidence for the underlying mechanism was provided by Hébert et al. ( 13 ) who demonstrated that PGE 2 attenuates vasopressin-stimulated osmotic water reabsorption in isolated microperfused rabbit collecting ducts. Subsequent cell culture studies showed that PGE 2 counteracts AVP action by enhancing AQP2 endocytotic uptake from the apical plasma membrane ( 33 ). On the other hand, recent studies showed that inhibition of COX by indomethacin reduced renal AQP2 expression in rats, implying that PGs may actually help maintain AQP2 levels ( 20 ).

In addition to the well-known renal effects of PGs on urine concentration, other evidence suggests that COX-derived products could influence this process by acting at central locales. Early studies showed that AVP release was stimulated by PGE 2 in the rat neurohypophysis ( 32 ). Ishikawa et al. ( 17 ) showed a reduced release of AVP in response to either hyperosmotic exposure or ANG II in guinea pig hypothalamoneurohypophyseal cultures following treatment with indomethacin. Finally, an intracerebroventricular (icv) injection of PGE 2 in rabbits increased the concentrations of plasma AVP ( 12 ). Thus, the accumulated evidence supports a role for PGs, including PGE 2 in mediating both central AVP release as well as more distal renal urine concentrating mechanisms. However, the specific receptors involved with such actions remain unidentified.

PGE 2 exerts its physiological influence by interacting with four E-prostanoid (EP) receptor subtypes designated EP 1, EP 2, EP 3, and EP 4 ( 10 ). These G protein-coupled receptors are distinguished by unique ligand binding profiles, signal transduction pathways, and exhibit widespread expression including the kidney, vasculature, and nervous system ( 6, 8, 9 ). With respect to PGE 2 s ability to modulate urine concentrating mechanisms, previous studies focused on the EP 3 receptor subtype as it is highly expressed in the principal cells of the collecting duct. Signaling through this subtype involves a G i -coupled mechanism which limits AVP-mediated cAMP production required for water transport ( 15 ). More recent studies have suggested that the EP 3 subtype carries out such putative diuretic effects by activating Rho kinase and F-actin polymerization ( 31 ). However, mice deficient for the EP 3 receptor can adequately concentrate their urine in response to water deprivation (WD) suggesting that this subtype may not be a prominent player ( 11 ). On the other hand, the EP 1 receptor is a reasonable candidate for modulating urine concentrating mechanisms since it is expressed along the AVP-responsive collecting duct and in the PVN and SON of the hypothalamus where AVP is synthesized ( 4, 23 ). EP 1 receptors can elicit phosphatidyl inositol turnover and intracellular Ca 2+ elevation through a G q protein. The requirement of the EP 1 receptor subtype in urine concentrating mechanisms has not been investigated.

In the present study, we propose a novel role for the EP 1 receptor subtype in the setting of body fluid control. Despite an abundant expression of EP 1 receptors along the collecting duct, we report that water-deprived mice lacking the EP 1 receptor inadequately concentrate their urine due to a defect in central AVP production.

METHODS

Construction of EP 1 targeting vector and generation of EP 1 -/- mice using the "Hit & Run" strategy. The strategy for generating EP 1 gene-targeted mice is described in detail in another concurrent manuscript. Briefly, mouse genomic clones containing the EP 1 gene locus were obtained by screening a 129SVJ/ola genomic P1 library (Genome Systems, St. Louis, MO) with a 233-bp murine EP 1 cDNA fragment from the coding region of exon 2. A 5,067-bp fragment containing all three exons of the EP 1 gene was subcloned into pBluescript (Stratagene) at Hin dIII and Xba I sites. An 895-bp fragment was then amplified from this clone, and using PCR-based mutagenesis its sequence was altered. The mutation introduces a stop codon in the coding region of the EP 1 receptor [C795T changes an in-frame CGA (R242) to TGA] thereby halting translation at the beginning of the third cytoplasmic loop of the EP 1 receptor, resulting in a nonfunctional receptor. The 5,067-kb mutant EP 1 fragment was then subcloned into the plasmid pKOScrambler NTKV (Lexigen, Woodlands, TX) which contains neo and TK selection cassettes.

Targeting the murine EP 1 gene was accomplished by two steps using the strategy for "Hit & Run." First, an EP 1 gene targeting vector was linearized by Bst EII at a unique site between exon 1 and 2, before electroporation of TL-1 embryonic stem (ES) cells. The linearized mutant gene inserts along with the neo and TK gene were introduced into ES cells, allowing homologous recombination via a single cross-over event which results in a direct repeat tandem duplication of the EP 1 gene. Insertion events were selected by G418 resistance and resistant clones were isolated and confirmed by Southern blotting. Since the direct repeat of the EP 1 gene is unstable, spontaneous deletion of one allele arising via single, reciprocal intrachromosomal recombination occurs between the direct repeats, leading to deletion of the selectable cassettes, the plasmid backbone, and one repeat of the targeted loci. Targeted ES cells bearing the mutant allele were injected into blastocysts and introduced into a foster mother according to standard procedures carried out at the Vanderbilt University Transgenic Core. Founder lines were obtained and mice were backcrossed to congeneity with C57BL/6 mice for 10 generations.

WD and water-loading experiments. For determination of drinking water intake and urine volume, adult mice (8-12 wk old) were transferred to metabolic cages (Nalge Nunc International) and were given free access to food and water for 3 days to acclimate them. Subsequently, urine volume and drinking water consumption were monitored for 5 consecutive days. For WD experiments, mice were deprived of access to drinking water by removing the water bottles and administered either vehicle (100 µl sc normal saline) or [deamino-Cys 1, D -Arg 8 ]-vasopressin (DDAVP; 1.0 µg/kg sc, 100 µl) 24 h before death. In each experiment, urine spot samples were obtained at 4 and 24 h following removal of water bottles. Body weights were recorded before and after WD. For some experiments, a small volume of blood was drawn from the right hindpaw, collected in heparinized tubes, and centrifuged at 1,500 g to isolate plasma. For water-loading experiments, mice were administered sterile water (0.5 ml/10 g ip) and concurrently injected with either vehicle (100 µl sc normal saline) or DDAVP (1.0 µg/kg sc, 100 µl). Spot urine samples were obtained at 0, 6, and 24 h following water loading. Urine and plasma osmolality were determined by freezing point depression (Advanced Instruments model 3MO plus). AVP peptide levels were assayed by ELISA (R&D Systems) according to the manufacturer's instructions by employing a 1:20 dilution of urine obtained from either control or 24-h WD mice. Results were corrected for urine creatinine (Cr) content determined by picric acid assay (Sigma).

Medullary osmolality determination. Mice were either given free access to water (control) or water deprived for 24 h. Kidneys were processed according to a method developed by Schmidt-Nielsen et al. ( 28a ) as modified by Fenton et al. ( 10a ). The kidneys were excised and the inner medullas were rapidly dissected, blotted onto Whatman 3M filter paper, and placed into preweighed microfuge tubes. Samples were weighed immediately and subsequently dried in an oven over a desiccant at 60°C for 8 h. After reweighing, 25 µl of deionized, distilled water were added to each tube and immersed in a boiling water bath for 3 min. Tubes were then centrifuged briefly and incubated at 4°C for 24 h to allow for diffusion. After centrifugation for 1 min at 8,000 g, the supernatant was analyzed for osmolality by freezing point depression and corrected for the original water content of the sample.

Quantitative RT-PCR. Groups of WT and EP 1 -/- mice either given free access to water or water deprived for 24 h were subsequently euthanized by cervical dislocation. Brains and kidneys were rapidly dissected. Brains were placed in a mouse brain matrix (Harvard Apparatus, Holliston, MA) and 2-mm coronal slices were collected from the rostral to caudal extent of the forebrain. The hypothalamic region surrounding the ventral portion of the third ventricle was excised from two consecutive brain slices and tissue was placed in a 1.5-ml Eppendorf tube containing 200 µl of RNeasy (GIBCO) for future real-time RT-PCR analysis.

Total RNA was extracted from the tissues and subjected to DNase treatment (Qiagen, Valencia, CA). AQP2 mRNA levels were determined by real-time RT-PCR using a two-step RT-PCR Syber Green detection approach. PreproAVP (vasopressin-neurophysin II GI:202341) mRNA levels were assayed using a Taq Man One-Step RT-PCR Master Mix Reagents (Applied Biosystems, Foster City, CA). Reactions were carried out using 50 ng of total RNA under the following conditions: 48°C for 30 min, 95°C for 10 min, and 40 cycles of 95°C for 15 s and 60°C for 1 min, using an ABI Prism 7000 Sequence Detection System. Primers and Taq Man probe for preproAVP: forward primer 5'-cgc tct ccg ctt gtt tcc t-3', reverse primer 5'-tgg gca gtt ctg gaa gta gca-3' and probe 6FAM-cct gct ggc ctt ctc ctc cgc-TAMRA. Primers for AQP2: forward primer 5'-atc ttt gcc tcc acc gat ga-3', reverse primer 5'-gcc ggt gaa ata gat ccc aag-3'. Relative values were derived by interpolation from a standard curve and were normalized to GAPDH mRNA levels in each sample as determined by either a Taq Man Rodent GAPDH Control Reagent kit or with specific mouse GAPDH primers when SYBR green detection was employed (Applied Biosystems).

Immunohistochemistry of tissue sections. Kidneys for immunohistochemical analysis were fixed in 4% paraformaldehyde, dehydrated through a gradient of alcohols, and embedded in paraffin wax. Embedded tissues were sectioned at 7 µm and mounted onto glass slides. Slides were rehydrated and washed three times in PBS and blocked with 5% normal goat serum (Vector Laboratories, Burlingame, CA) in PBS for 1 h at RT. Following three washings in PBS, the sections were incubated with the anti-rabbit AQP2 primary antibody (Santa Cruz Biotechnology) 1:1,000 for 1 h at RT, washed three times, and incubated with the secondary antibody (anti-rabbit) for 1 h at RT. Sections were washed three times in PBS and processed for Vector ABC/3,3'-diaminobenzidine staining according to the manufacturer's instructions (Vector Laboratories). Sections were visualized by high-power light microscopy.

Statistics. The values are presented as means ± SE. Statistical comparisons between two groups were performed using the unpaired Student's t -test. Comparisons between three or more groups were carried out using an ANOVA, followed by a Newman-Keuls posttest. P < 0.05 was considered statistically significant.

RESULTS

Urinary concentrating ability of EP 1 -/- mice. Overall, adult EP 1 -/- mice (8-12 wk of age) backcrossed for 10 generations onto a congenic C57Bl/6J background were healthy, fertile, and similar in size and weight to their WT counterparts (females: WT, 20.1 ± 1.0 g; EP 1 -/-, 20.0 ± 0.3 g, not significant; males: WT, 23.5 ± 0.4 g; EP 1 -/-, 24.2 ± 0.5 g, not significant). However, as shown in Fig. 1, daily urine output (WT, 0.05 ± 0.01 ml·g body wt -1 ·day -1; EP 1 -/-, 0.07 ± 0.01, P < 0.03) and water consumption (WT, 0.24 ± 0.01 ml·g body wt -1 ·day -1; EP 1 -/-, 0.31 ± 0.03, not significant) were slightly greater in EP 1 -/- mice compared with WT mice. Although a lower baseline urine osmolality was observed for the EP 1 -/- mice (WT, 2,166 ± 153 mosmol/kgH 2 O vs. EP 1 -/-, 1,954 ± 139 mosmol/kgH 2 O, P = 0.3), it was not statistically significant. To assess the near-maximal urine concentrating ability of EP 1 -/- mice, healthy littermates were water deprived (WD) for 24 h. As shown in Fig. 2 A, urine osmolality rose significantly for WT mice, but not for EP 1 -/- mice, as early as 4 h following the initiation of WD (WD WT, 3,328 ± 99 mosmol/kgH 2 O, vs. WD EP 1 -/-, 2,000 ± 180 mosmol/kgH 2 O, * P < 0.01). Furthermore, urine osmolalities for WD EP 1 -/- did not reach levels achieved by WD WT mice even after 24 h of WD ( Fig. 2 A, WD WT: 3,880 ± 170 mosmol/kgH 2 O vs. WD EP 1 -/- : 2,789 ± 170 mosmol/kgH 2 O, * P < 0.05). Consistent with a urine concentrating defect, EP 1 -/- plasma osmolality rose significantly higher than that of WT mice following WD (WD WT, 291 ± 7 mosmol/kgH 2 O vs. WD EP 1 -/-, 321 ± 4 mosmol/kgH 2 O, ** P < 0.01). Over this short time period, body weights decreased equally among WD WT and WD EP 1 -/- mice (WD WT: -9.5 ± 0.5% body wt vs. WD EP 1 -/- : -8.0 ± 0.5% body wt).

Fig. 1. Baseline urine output, water consumption, and body wts for wild-type (WT) and EP 1 -/- mice. Male and female mice (8-12 wk old) were housed in metabolic diuresis cages for 3 days before measurements. Mice were allowed free access to food and water throughout the study. Urine output (ml·25 g body wt -1 ·day -1; A ), water consumption (ml·25 g body wt -1 ·day -1; B ), and body wt (g; C ) were measured daily. The values presented are the means derived from 5 consecutive days using 6 mice per group ( n = 6). * P < 0.03 vs. WT. The trends in urine output and water consumption were similar in both male and female EP 1 -/- mice.

Fig. 2. Urine concentrating abilities of WT and EP 1 -/- mice. A : water deprivation (WD) was initiated by removal of water bottles from cages of WT and EP 1 -/- mice (ages 8-12 wk). Spot urine samples were obtained at 0-, 4-, and 24-h time points. Urine osmolality was assayed by freezing point depression. The experiment was repeated 3 times and the data are means ± SE of n = 13 mice in total. * P < 0.01 vs. WD EP 1 -/- 4 h. ** P < 0.05 vs. WD EP 1 -/- 24 h. B : plasma osmolality was measured in mice before and after 24 h WD. The data are means ± SE, n = 7 mice. * P < 0.01 vs. WT control. ** P < 0.01 vs. WT WD and EP 1 -/- control.

Medullary interstitial osmolality in WD EP 1 -/- mice. Urine concentrating mechanisms rely, in part, on the maintenance of an osmolar gradient across the tubular lumen and into the medullary interstitium. To identify the mechanism underlying the urine concentrating defect, renal medullary osmolalities were determined in WD animals. As shown in Fig. 3, renal medullary osmolality dropped significantly in the EP 1 -/- mice compared with WT counterparts following WD (EP 1 -/- -466 mosmol/kgH 2 O), indicating a possible defect in the regulation of the medullary osmolality gradient. The decline observed for the WD WT mice ( -135 mosmol/kgH 2 O) was not statistically significant compared with WT controls.

Fig. 3. Renal inner medullary osmolalities in water-deprived WT and EP 1 -/- mice. Water bottles were removed from a subset of cages of WT and EP 1 -/- mice (ages 8-12 wk) and after 24 h the animals were killed. The kidneys were removed and processed for inner medullary osmolality determination as described in METHODS. The data are means ± SE of n = 8 mice. * P < 0.05 vs. EP 1 -/- control.

AQP2 expression in WD EP 1 -/- mice. Urine concentration also depends, in part, on the expression, translocation, and activity of AQP2 water channels in the principal cells of the collecting duct. Such transcriptional and posttranslational control of AQP2 can be subject, in part, to AVP-mediated signaling cascades ( 24 ) which are, in turn, thought to be modulated by EP receptor signaling ( 13, 15 ). Other studies carried out in primary cultures of rat IMCD cells show that signal transduction via the EP 3 receptor opposes AVP-induced AQP2 translocation to the plasma membrane ( 31 ). We investigated whether the absence of the EP 1 receptor affects AQP2 expression and translocation in response to WD. As shown in Fig. 4 A, baseline AQP2 mRNA and protein expression were similar for WT and EP 1 -/- mice permitted free access to water (WT, 1.0 ± 0.1 relative units; EP 1 -/-, 0.9 ± 0.1 relative units, not significant). WD for 24 h induced a significant yet equivalent upregulation of renal AQP2 in WT and EP 1 -/- mice (WD WT, 2.1 ± 0.3 relative units; WD EP 1 -/-, 1.8 ± 0.3 relative units, P < 0.05 vs. respective water-fed controls; not significant vs. WD counterparts). Immunohistochemical analyses also revealed similar expression of AQP2 protein in WT and EP 1 -/- mice along the collecting duct from cortex to papilla ( Fig. 4 B, top; cortex shown). WD conferred a noticeable shift of AQP2 to the apical membranes in many of the principal cells from WT and EP 1 -/- mice ( Fig. 4 B, middle ). Last, the combination of WD and DDAVP administration markedly induced AQP2 translocation to apical membranes along the collecting duct in WT and EP 1 -/- mice ( Fig. 4 B, bottom ).

Fig. 4. Renal AQP2 in water-deprived WT and EP 1 -/- mice. Water bottles were removed from the cages of WT and EP 1 -/- mice (ages 8-12 wk) and after 24 h the animals were killed along with a group of control animals given free access to water. Another group of WT and EP 1 -/- mice were administered DDAVP (1.0 µg/kg sc) at the time of water bottle removal. The kidneys were removed and processed for either RNA isolation or immunohistochemistry as described in METHODS. A : real-time RT-PCR of renal AQP2 mRNA. DNase-treated mRNA was reverse-transcribed and subsequently assayed for AQP2 content using a primer set in conjunction with SYBR green-based Quantitect PCR detection kit (Qiagen). Data are normalized to GAPDH content and relative values are derived from standard curves constructed for each target. The data are means ± SE of n = 5 mice. * P < 0.05 vs. respective nonwater-deprived controls. B : AQP2 immunohistochemistry of renal sections ( x 400 magnification). Whole kidney sections (7-µm thickness) derived from control, water-deprived, and DDAVP-treated + water-deprived mice were subjected to immunohistochemical detection with a goat anti-AQP2 polyclonal antibody (1:1,000 dilution). Representative images of cortex AQP2 labeling obtained from groups of at least 5 mice are shown. Similar results were seen in outer/inner medullary regions. Arrowheads denote substantial apical membrane localization of AQP2.

Reduced AVP levels in WD EP 1 -/- mice. AVP is generated in the magnocellular neurons of the hypothalamic PVN and SON. Recent studies localized the EP 1 receptor to the PVN and SON ( 23 ). We therefore hypothesized that AVP production might be compromised in the absence of the EP 1 receptor. Accordingly, preproAVP mRNA levels were measured in microdissected hypothalamic regions from WD mice by quantitative RT-PCR. As shown in Fig. 5 A, a Taq Man primer/probe set directed against preproAVP mRNA detected significant induction of AVP mRNA expression in hypothalamic tissue of WT mice following WD for 24 h (WT, 0.4 ± 0.2 relative units vs. WD WT, 1.7 ± 0.4 relative units, P < 0.01). However, this effect was blunted in WD EP 1 -/- mice (EP 1 -/-, 0.3 ± 0.2 relative units vs. WD EP 1 -/-, 0.6 ± 0.1 relative units, not significant). Urinary AVP peptide levels were also determined for control and WD WT and EP 1 -/- mice by ELISA. As shown in Fig. 5 B, WD for 24 h yielded a substantial increase of urinary AVP in WT mice (WT, 4.9 ± 0.2 pg/mg Cr vs. WD WT, 16.1 ± 1.5 pg/mg Cr, P < 0.001). In contrast, this manipulation produced a more modest response by EP 1 -/- mice that was statistically significant from WT mice (EP 1 -/-, 4.2 ± 0.3 pg/mg Cr vs. WD EP 1 -/-, 11.2 ± 0.9 pg/mg Cr, P < 0.001; P < 0.01 vs. WD WT). Administration of a supraphysiological dose of DDAVP (1.0 µg/kg sc), a highly selective V 2 receptor agonist, restored a significant proportion of the concentrating ability in WD EP 1 -/- mice, with the change in urine osmolality increasing from +464 ± 243 mosmol/kgH 2 O to +1,624 ± 140 mosmol/kgH 2 O ( Fig. 6 A ). However, these mice were slightly less responsive than the DDAVP-treated WD WT mice (WD WT, +1,520 ± 122 mosmol/kgH 2 O vs. WD WT + DDAVP, +2,464 ± 146 mosmol/kgH 2 O). To better understand whether the defect in urine concentration possessed a renal component, mice were water loaded to suppress endogenous AVP levels. Mice were simultaneously administered water (0.5 ml/10 g ip) with either saline or DDAVP (1.0 µg/kg sc). As shown in Fig. 6 B, urine osmolalities measured 6 h post water loading revealed that WT and EP 1 -/- mice concentrated their urine similarly in response to DDAVP (water-loaded WT + DDAVP increasing from 1,420 ± 253 to 2,540 ± 232 mosmol/kgH 2 O, vs. water-loaded EP 1 -/- rising from 1,772 ± 311 to 3,036 ± 174 mosmol/kgH 2 O). Water loading without DDAVP administration yielded a diluted urine, as expected (water-loaded WT, 815 ± 62 mosmol/kgH 2 O, vs. water-loaded EP 1 -/-, 396 ± 101 mosmol/kgH 2 O). Urine osmolalities for all groups of mice returned to baseline values after 24 h.

Fig. 5. AVP synthesis in water-deprived WT and EP 1 -/- mice. A : hypothalamic AVP mRNA induction in WT and EP 1 -/- water-deprived mice. Water bottles were removed from the cages of WT and EP 1 -/- mice (ages 8-12 wk) and after 24 h the animals were killed along with a group of control animals with free access to water. Hypothalamic regions were excised and the total RNA was extracted as described in METHODS. A Taq Man primer/probe set directed against preproAVP was used to quantify AVP induction. Data are normalized to GAPDH content and relative values were derived from standard curves constructed for each target. The data are means ± SE of n = 5 mice. * P < 0.01 vs. WT control. B : urinary AVP content in water-deprived WT and EP 1 -/- mice. Mice were subjected to water deprivation for 24 h and urine AVP content was assayed by ELISA. Values presented are corrected for urine creatinine content. ** P < 0.001 vs. WT control. * P < 0.01 vs. EP 1 -/- control and water-deprived WT.

Fig. 6. DDAVP responsiveness of water-deprived and water-loaded WT and EP 1 -/- mice. A : DDAVP partially restores urinary concentrating ability of EP 1 -/- water-deprived mice. Mice were administered DDAVP (1.0 µg/kg sc) or saline at the time of water bottle removal and the urine osmolalities were determined at 0 and 24 h. The data are presented as the change ( ) in urine osmolality over 24 h and are means ± SE of n = 5 mice. * P < 0.01 vs. WD WT. ** P < 0.001 vs. WD EP 1 -/-. B : DDAVP equally increases urine osmolality of WT and EP 1 -/- water-loaded mice. Mice were administered DDAVP (1.0 µg/kg sc) or saline at the time of water loading (0.5 ml/10 g body wt ip) and the urine osmolalities were determined at 0, 6, and 24 h. The data are means ± SE of n = 7 mice. * P < 0.01 vs. each respective 0-h control.

DISCUSSION

Several lines of evidence suggest that PGE 2 contributes to homeostatic volume control by modulating urine concentrating mechanisms. However, the specific EP receptor(s) and their functional locales accounting for these effects have not been completely identified. In the present study, we investigated whether the PGE 2 EP 1 receptor subtype participates in urine concentration as it is expressed in the AVP-producing hypothalamus and along the AVP-responsive collecting duct. Our data reveal a defect in central AVP synthesis for mice lacking the EP 1 subtype following acute WD. The blunted hypothalamic AVP production in EP 1 -null animals suggests that this receptor is a pivotal player in the response to hyperosmolality. Moreover, our findings suggest that the EP 1 receptor acts upstream of AVP production, likely along paraventricular and/or supraoptic inputs within the hypothalamus. AVP is initially produced as a larger preproAVP protein in the hypothalamus. Following proteolytic processing by neurophysin, the mature AVP peptide is stored in the posterior pituitary where it awaits appropriate physiological cues before being secreted into the venous flow. Studies by Arima et al. ( 3 ) showed that pituitary stores of AVP are quickly depleted following WD thereby requiring induction of mRNA synthesis to sustain downstream urine concentrating mechanisms. Such stores likely account for the modest rise in urinary AVP in EP 1 -/- mice observed in the present study. However, we suspect that while these quantities are rapidly exhausted to similar extents in WT and EP 1 -/- mice, they are replenished at a slower rate in the latter animals. The net result is that the EP 1 -null kidney receives a suboptimal AVP quantity, thereby limiting urine concentration.

The intrarenal role of the EP 1 receptor is only partially characterized. In the kidney, this G q -coupled subtype is found in the mesangial cells and podocytes, where it may modulate certain maladaptive aspects of diabetic nephropathy such as glomerular hypertrophy, matrix expansion, proteinuria, and TGF- production ( 22 ). Along the nephron, the EP 1 receptor is abundantly expressed in the principal cells of the collecting duct, where it is thought to offset the buffering influence of EP 3 receptors upon the anti-diuretic actions of AVP ( 31 ). In vitro studies showed that PGE 2 can inhibit sodium transport in the isolated, perfused rabbit collecting duct by a mechanism coupled to the release of Ca 2+ from intracellular stores, consistent with EP 1 -mediated signal transduction ( 14 ). In contrast, work by Coffman and colleagues ( 29 ) supports an anti-natriuretic role for the EP 1 receptor. They observed a reduction in systolic blood pressure and a rise in plasma renin activity in EP 1 knockout mice fed a low-sodium diet. However, our data argue against a substantial role for this subtype in distal collecting duct Na + handling since EP 1 -/- mice exhibit appropriate anti-natriuresis or natriuresis when placed on either sodium-deficient or high-sodium diets, respectively (data not shown). This result is consistent with findings obtained in rats where EP 1 receptor levels were unaffected by dietary sodium manipulation ( 18 ). However, our interpretation must be tempered by the possibility that other receptor pathways may offer redundant mechanisms to compensate for loss of the EP 1 receptor.

PGE 2 tempers the actions of AVP in isolated, perfused rabbit collecting duct, yet when added alone, this eicosanoid promotes transepithelial water flow ( 15 ). Subsequent work suggested that PGE 2 exerts such anti-diuretic effects via cAMP-coupled EP 4 receptors ( 28 ), whereas PGE 2 s ability to reduce AVP-induced cAMP production implies EP1/3 signaling ( 26 ). However, the mechanism underlying these effects may be more complex, as demonstrated by the recent work of Klussman and colleagues ( 31 ). Their data suggest that EP 1 receptors antagonize the inhibitory effects of the EP 3 receptor on AQP2 translocation in rat inner medullary collecting duct cells through effects on Rho GTPase and the actin cytoskeleton. According to such a model, the urinary concentrating defect observed presently in EP 1 -/- mice would arise due to an unmasking of the EP 3 subtype, yielding significant attenuation of AVP-mediated water reabsorption. However, in vivo evidence for direct involvement of EP 3 receptors remains elusive. Coffman and colleagues showed that EP 3 -/- mice do not "over concentrate" their urine in response to WD, although inhibition of endogenous PGE 2 production by indomethacin yields an elevation of urinary osmolality in WT but not in EP 3 -/- mice ( 11 ). Such findings suggest that the EP 3 receptor may underlie increased urine concentration encountered with rodents and humans following NSAID administration ( 5 ). Our in vivo data with EP 1 -/- mice suggest that collecting duct EP 1 receptors may influence only a minor portion of AVP-induced water flux in the context of WD. Much of the urine concentrating ability of WD EP 1 -null animals was restored by DDAVP administration. Furthermore, WD combined with DDAVP induced a qualitatively similar AQP2 translocation from a subapical to apical plasma membrane along the collecting duct in WT and EP 1 -/- mice. When endogenous AVP levels were suppressed by acute water loading, WT and EP 1 -/- mice were equally responsive to DDAVP-induced urine concentration, once again suggesting that AQP2-based urine concentrating mechanisms are intact despite the absence of functional EP 1 receptors along the collecting duct. However, the acute reduction of medullary osmolality observed in EP 1 -/- mice following WD may also have contributed to the urine concentrating defect. Whether reduced AVP levels in response to WD might account for this drop in medullary osmolality in EP 1 -/- mice is unclear, and warrants further investigation. Furthermore, it was surprising that WD failed to increase medullary osmolality in WT mice. Although values trended toward a reduction in medullary osmolality, they did not achieve statistical significance. It is probable that the mouse can sustain its renal medullary osmolality for the 24-h period of WD and that an increase would likely follow at later time points. On the other hand, the result may be due to methodological limitations. Inasmuch as our data suggest that collecting duct EP 1 receptors play but a minor role in the modulation of AVP action within this nephron segment, we cannot rule out possibility that adaptive mechanisms in the collecting duct may have arisen in the absence of the EP 1 receptor.

Our work therefore suggests that the EP 1 receptor subtype largely contributes to urine concentrating mechanisms; not through its actions along the collecting duct, but through its expression in the hypothalamus. This hypothesis is consistent with recent work by two groups who localized the EP 1 receptor to the hypothalamus. In situ hybridization studies detected EP 1 receptor mRNA in the supraoptic nuclei ( 4 ) while others employed an EP 1 antiserum to verify expression in the central nucleus of amygdala and at synapses of corticotrophin-releasing hormone containing neurons of the PVN ( 23 ). These structures each contribute to the synthesis of AVP in response to a variety of physiological and pathophysiological stimuli. In fact, the involvement of PGs in mediating AVP synthesis was suggested by a number of reports. For example, icv administration of NSAIDs to rats can blunt AVP synthesis in response to hypertonic saline ( 7 ), acute WD, or injection of bacterial endotoxin ( 21 ). Furthermore, icv injection of PGE 2 in rats increased plasma AVP levels in some studies ( 12, 27 ) while not in others ( 21 ). Until now, the identity of the specific prostanoid receptor underlying these effects remained unclear. We suggest that such NSAID-induced abrogation of PGE 2 synthesis suppresses EP 1 receptor activation in the PVN/SON, thereby limiting AVP production and restricting renal urine concentrating mechanisms. Accordingly, our data show that AVP mRNA induction is blunted in EP 1 -/- mice in response to acute WD, while baseline production and circulating levels of AVP are sufficient to avoid a central diabetes insipidus phenotype in these mice. This finding is likewise consistent with studies in rats showing that while indomethacin treatment blunts AVP production in response to icv infusion of hypertonic saline, it does not affect baseline AVP levels ( 17 ).

A role for centrally expressed EP receptors was recently suggested by others. Adrenocorticotropic hormone (ACTH) is produced in the PVN, is stored in the anterior pituitary, and is released in response to inflammatory stress. Narumyia and colleagues ( 23 ) demonstrated that endotoxin-treated EP 1 -/- mice exhibited an impaired ACTH response coupled with reduced c- fos induction. One hour after receiving LPS injections, EP 1 -/- mice were unable to significantly elevate plasma ACTH levels in contrast to their WT counterparts.

Taken together, it would therefore appear that centrally produced PGE 2 acting via the EP 1 receptor subtype in the hypothalamus plays a significant role in mediating the body's response to various physiological and pathophysiological stresses. The involvement of the EP 1 subtype in urine concentrating mechanisms is of particular interest since recent studies are driving the development of EP 1 -selective ligands as potential anti-cancer agents ( 16, 19 ). Furthermore, such reagents have proved therapeutic for rats with glomerulopathies induced by diabetes or hypertension ( 22, 30 ). In light of the current study, the application of EP 1 receptor-based therapeutics may therefore require careful consideration of the potential effects on urine concentrating mechanisms.

NOTE ADDED IN PROOF

On p. F869 in METHODS, under the subheading Medullary osmolality determination, a protocol for determining renal medullary osmolality contains text which originally appeared in a 2004 article by R. A. Fenton et al. in the Proceedings of the National Academy of Sciences. Regrettably, this article was not referenced in our Articles in Press manuscript, and a reference that was listed [Schmidt-Nielsen et al. ( 4 )], corresponding to a 1983 article by B. Schmidt-Nielsen et al. in the American Journal of Physiol-Renal Physiology, was likewise absent from our reference list. With sincere apologies to the editors of the American Journal of Physiology-Renal Physiology, the Proceedings of the National Academy of Sciences, and, importantly, to the authors of these articles, we have amended and properly referenced the section in METHODS, which will appear in the final print version of the article.

GRANTS

C. R. J. Kennedy is a Canadian Institutes of Health Research (CIHR) New Investigator Awardee. This work is supported by operating grants from the CIHR (MOP44013 ) and NIH (DK37097).

【参考文献】
  Anderson RJ, Berl T, McDonald KD, and Schrier RW. Evidence for an in vivo antagonism between vasopressin and prostaglandin in the mammalian kidney. J Clin Invest 56: 420-426, 1975.

Antunes-Rodrigues J, de Castro M, Elias LL, Valenca MM, and McCann SM. Neuroendocrine control of body fluid metabolism. Physiol Rev 84: 169-208, 2004.

Arima H, Kondo K, Kakiya S, Nagasaki H, Yokoi H, Yambe Y, Murase T, Iwasaki Y, and Oiso Y. Rapid and sensitive vasopressin heteronuclear RNA responses to changes in plasma osmolality. J Neuroendocrinol 11: 337-341, 1999.

Batshake B, Nilsson C, and Sundelin J. Molecular characterization of the mouse prostanoid EP1 receptor gene. Eur J Biochem 231: 809-814, 1995.

Berl T, Raz A, Wald H, Horowitz J, and Czaczkes W. Prostaglandin synthesis inhibition and the action of vasopressin: studies in man and rat. Am J Physiol Renal Fluid Electrolyte Physiol 232: F529-F537, 1977.

Breyer RM, Bagdassarian CK, Myers SA, and Breyer MD. Prostanoid receptors: subtypes and signaling. Annu Rev Pharmacol Toxicol 41: 661-690, 2001.

Brooks DP, Share L, and Crofton JT. Role of brain prostaglandins in the control of vasopressin secretion in the conscious rat. Endocrinology 118: 1716-1722, 1986.

Coleman RA, Grix SP, Head SA, Loutit JB, Mallett A, and Sheldrick RLG. A novel inhibitory receptor in piglet saphenous vein. Prostaglandins 47: 151-168, 1994.

Coleman RA, Kennedy I, Humphrey PPA, Bunce K, and Lumley P. Prostanoids and their receptors. In: Comprehensive Medical Chemistry, edited by Emmet JC. Oxford: Pergammon Press, 1990, p. 643-714.

Coleman RA, Smith WL, and Narumiya S. International Union of Pharmacology classification of prostanoid receptors: properties, distribution, and structure of the receptors and their subtypes. Pharmacol Rev 46: 205-229, 1994.

Fenton RA, Chou CL, Stewart GS, Smith CP, and Knepper MA. Urinary concentrating defect in mice with selective deletion of phloretin-sensitive urea transporters in the renal collecting duct. Proc Natl Acad Sci USA 101: 7469-7474, 2004.

Fleming EF, Athirakul K, Oliverio MI, Key M, Goulet J, Koller BH, and Coffman TM. Urinary concentrating function in mice lacking EP 3 receptors for prostaglandin E 2. Am J Physiol Renal Physiol 275: F955-F961, 1998.

Hashimoto H, Noto T, and Nakajima T. A study on the release mechanism of vasopressin and oxytocin. Neuropeptides 12: 199-206, 1988.

Hebert RL, Jacobson HR, and Breyer MD. PGE 2 inhibits AVP-induced water flow in cortical collecting ducts by protein kinase C activation. Am J Physiol Renal Fluid Electrolyte Physiol 259: F318-F325, 1990.

Hebert RL, Jacobson HR, and Breyer MD. Prostaglandin E2 inhibits sodium transport in rabbit cortical collecting duct by increasing intracellular calcium. J Clin Invest 87: 1992-1998, 1991.

Hebert RL, Jacobson HR, Fredin D, and Breyer MD. Evidence that separate PGE 2 receptors modulate water and sodium transport in rabbit cortical collecting duct. Am J Physiol Renal Fluid Electrolyte Physiol 265: F643-F650, 1993.

Hull MA, Ko SC, and Hawcroft G. Prostaglandin EP receptors: targets for treatment and prevention of colorectal cancer? Mol Cancer Ther 3: 1031-1039, 2004.

Ishikawa S, Saito T, and Yoshida S. The effect of prostaglandins on the release of arginine vasopressin from the guinea pig hypothalamo-neurohypophyseal complex in organ culture. Endocrinology 108: 193-198, 1981.

Jensen BL, Mann B, Skott O, and Kurtz A. Differential regulation of renal prostaglandin receptor mRNAs by dietary salt intake in the rat. Kidney Int 56: 528-537, 1999.

Kawamori T, Kitamura T, Watanabe K, Uchiya N, Maruyama T, Narumiya S, Sugimura T, and Wakabayashi K. Prostaglandin E receptor subtype EP(1) deficiency inhibits colon cancer development. Carcinogenesis 26: 353-357, 2005.

Kim SW, Kim JW, Choi KC, Ma SK, Oh Y, Jung JY, Kim J, and Lee J. Indomethacin enhances shuttling of aquaporin-2 despite decreased abundance in rat kidney. J Am Soc Nephrol 15: 2998-3005, 2004.

Knigge U, Kjaer A, Kristoffersen U, Madsen K, Toftegaard C, Jorgensen H, and Warberg J. Histamine and prostaglandin interaction in regulation of oxytocin and vasopressin secretion. J Neuroendocrinol 15: 940-945, 2003.

Makino H, Tanaka I, Mukoyama M, Sugawara A, Mori K, Muro S, Suganami T, Yahata K, Ishibashi R, Ohuchida S, Maruyama T, Narumiya S, and Nakao K. Prevention of diabetic nephropathy in rats by prostaglandin E receptor EP1-selective antagonist. J Am Soc Nephrol 13: 1757-1765, 2002.

Matsuoka Y, Furuyashiki T, Bito H, Ushikubi F, Tanaka Y, Kobayashi T, Muro S, Satoh N, Kayahara T, Higashi M, Mizoguchi A, Shichi H, Fukuda Y, Nakao K, and Narumiya S. Impaired adrenocorticotropic hormone response to bacterial endotoxin in mice deficient in prostaglandin E receptor EP1 and EP3 subtypes. Proc Natl Acad Sci USA 100: 4132-4137, 2003.

Murillo-Carretero MI, Ilundain AA, and Echevarria M. Regulation of aquaporin mRNA expression in rat kidney by water intake. J Am Soc Nephrol 10: 696-703, 1999.

Nielsen S, Frokiaer J, Marples D, Kwon TH, Agre P, and Knepper MA. Aquaporins in the kidney: from molecules to medicine. Physiol Rev 82: 205-244, 2002.

Noland TD, Carter CE, Jacobson HR, and Breyer MD. PGE 2 regulates cAMP production in cultured rabbit CCD cells: evidence for dual inhibitory mechanisms. Am J Physiol Cell Physiol 263: C1208-C1215, 1992.

Puurunen J and Leppaluoto J. Centrally administered PGE2 inhibits gastric secretion in the rat by releasing vasopressin. Eur J Pharmacol 104: 145-150, 1984.

Sakairi Y, Jacobson HR, Noland TD, and Breyer MD. Luminal prostaglandin E receptors regulate salt and water transport in rabbit cortical collecting duct. Am J Physiol Renal Fluid Electrolyte Physiol 269: F257-F265, 1995.

Schmidt-Nielsen B, Graves B, and Roth J. Water removal and solute additions determining increases in renal medullary osmolality. Am J Physiol Renal Physiol 244: F472-F482, 1983.

Stock JL, Shinjo K, Burkhardt J, Roach M, Taniguchi K, Ishikawa T, Kim HS, Flannery PJ, Coffman TM, McNeish JD, and Audoly LP. The prostaglandin E2 EP1 receptor mediates pain perception and regulates blood pressure. J Clin Invest 107: 325-331, 2001.

Suganami T, Mori K, Tanaka I, Mukoyama M, Sugawara A, Makino H, Muro S, Yahata K, Ohuchida S, Maruyama T, Narumiya S, and Nakao K. Role of prostaglandin E receptor EP1 subtype in the development of renal injury in genetically hypertensive rats. Hypertension 42: 1183-1190, 2003.

Tamma G, Wiesner B, Furkert J, Hahm D, Oksche A, Schaefer M, Valenti G, Rosenthal W, and Klussmann E. The prostaglandin E2 analog sulprostone antagonizes vasopressin-induced antidiuresis through activation of Rho. J Cell Sci 116: 3285-3294, 2003.

Vilhardt H and Hedqvist P. A possible role of prostaglandin E2 in the regulation of vasopressin secretion in rats. Life Sci 9: 825-830, 1970.

Zelenina M, Christensen BM, Palmer J, Nairn AC, Nielsen S, and Aperia A. Prostaglandin E 2 interaction with AVP: effects on AQP2 phosphorylation and distribution. Am J Physiol Renal Physiol 278: F388-F394, 2000.

作者单位：1 Kidney Research Centre, Division of Nephrology, Department of Medicine, Ottawa Hospital, Ottawa Health Research Institute, 2 Department of Cellular and Molecular Medicine, University of Ottawa, Ottawa, Ontario, Canada; and 3 Division of Nephrology, Vanderbilt University Medical Center, Nashville,