G q-dependent signaling cascades stimulate water-seeking behavior

【摘要】  We used the mouse nephrin promoter to express a constitutively active G q [G L)] transgene in mice. As previously reported, the transgene was expressed in kidney, pancreas, and brain, and the kidney phenotype was characterized by albuminuria and reduced nephron mass. Additional studies revealed a second phenotype characterized by polyuria and polydipsia. The polyuric phenotype was not caused by abnormal glucose metabolism or hypercalcemia but was accompanied by reduced urinary concentrating ability. Additional studies found that 1 ) water restriction was associated with an appropriate increase in serum vasopressin levels in transgenic (TG) mice; 2 ) the urinary concentrating defect was not corrected by administration of desamino- D -arginine vasopressin (DDAVP); and 3 ) papillary length was similar in TG and non-TG mice. To examine the renal response to DDAVP at the molecular level, we monitored aquaporin 2 (AQP2) and vasopressin V2 receptor (V2R) mRNA levels in mouse kidney. Consistent with the known effects of vasopressin, administration of DDAVP caused a decrease in V2R mRNA levels and an increase in AQP2 mRNA levels in both TG and non-TG animals, suggesting an appropriate renal response to DDAVP in the TG mice. To determine whether the urine concentrating abnormality was the result of primary polydipsia, water intake by TG mice was restricted to the amount ingested by non-TG animals. After 5 days, urinary concentrating ability was similar in TG mice and non-TG littermate controls. These data are consistent with the notion that expression of the G L) transgene in the brain induced primary polydipsia in the TG mice.

【关键词】  polyuria vasopressin aquaporin

FLUID HOMEOSTASIS IS MAINTAINED by three coordinated processes including water acquisition, release of vasopressin (antidiuretic hormone or ADH), and renal concentrating mechanisms ( 1 ). In states of water depletion, the body releases ADH to decrease water loss by the kidney as well as stimulates water intake by enhancing thirst ( 1 ). The mechanisms that regulate these homeostatic processes are complex and redundant. For example, ADH release is stimulated not only by an increase in extracellular fluid osmolality but also by a decrease in circulating fluid volume ( 1 ). Volume depletion also stimulates ANG II generation both systemically as well as in the brain ( 1 ). In the systemic circulation, ANG II helps maintain intravascular volume by causing vasoconstriction and sodium retention ( 5, 7, 13 ). In the brain, ANG II stimulates synthesis and release of vasopressin and generates thirst ( 1, 12, 27, 28 ). The thirst response appears to be mediated, in part, by the generation and central release of ANG II and, in turn, stimulation of type 1 ANG II receptors (AT 1 receptors) in preoptic/hypothalamic regions of central nervous system ( 15, 16, 22 ). These redundant regulatory processes precisely regulate body water to maintain intravascular volume and serum osmolality.

AT 1 receptors belong to the large superfamily of G protein-coupled receptors (GPCRs) ( 25, 29, 30 ). In contrast to most mammals, rodents have two AT 1 receptor genes that are designated AT 1A and AT 1B ( 14, 29, 31 ). Both receptors couple to G proteins belonging to the Gq family ( 30 ). This receptor family has become an important target for the development of therapeutic agents because multiple G q-coupled receptor systems have been implicated in the pathogenesis of several common diseases including atherosclerosis, hypertension, and kidney diseases ( 12, 18, 20, 21, 30, 35, 39 ). Because of the association between G q-coupled receptors and renal disease, we investigated the role of G q-linked signaling cascades in glomerular disease processes by using the mouse nephrin promoter ( 11, 24 ) to target expression of a constitutively active Gq -subunit ( 17 ) to glomerular epithelial cells. As previously reported ( 39 ), we found that 1 ) the transgene was expressed in the kidney, pancreas, and brain; and 2 ) the kidney phenotype was characterized by albuminuria, reduced nephron mass, and enhanced susceptibility to glomerular injury. Additional studies revealed a second phenotype characterized by polyuria, polydipsia, and reduced urinary concentrating ability. The polyuric phenotype did not appear to be caused by either elevated serum glucose levels, hypercalcemia, abnormal ADH release, renal structural defects, or decreased tissue responsiveness to ADH. The reduced urinary concentrating ability was, however, corrected by restricting water intake of transgenic (TG) mice to the amount ingested by non-TG animals. These data are consistent with the notion that the urinary concentrating defect was due to primary polydipsia ( 11, 19 ). This condition results in high urine flow rates, a decrease in medullary interstitial tonicity, and a reduction in maximal urine concentrating ability ( 11, 19 ). Taken together, these data suggest that G q-dependent signaling cascades are likely to play an important role in regulating drinking behavior.

METHODS

Creation of G L) TG mouse and study design. G L) TG mice were generated as described previously ( 39 ) and were maintained in the animal facilities of Duke University Medical Center. The mice were fed a standard rodent diet (LabDiet, PMI Nutrition International, St. Louis, MO) containing 0.40% sodium. The chemical composition of the diet can be obtained at www.labdiet.com. Unless otherwise indicated, mice were studied at 3-4 mo of age. For all studies, TG animals were matched for sex with non-TG littermates as controls. All animal procedures were approved by the Animal Care and Use Committee of Duke University.

Blood pressure measurements. Systolic blood pressure (BP) was measured in conscious mice by a noninvasive computerized tail-cuff method after 2 wk of training. The method has been validated previously and correlates with direct measurements of intra-arterial pressure ( 40 ). After the training period, BP was recorded daily for 5 days, and the value for each mouse represents the average of the daily measurements.

Measurement of water consumption and urine output. TG and non-TG mice were studied at 1, 2, 4, and 6 mo of age. For the studies, mice were placed in metabolic cages specifically designed for use with mice (Hatteras Instrument, Cary, NC). Water bottles, urine collection tubes, and body weights were measured immediately before the study and after the animals had been in the metabolic cages for 24 h. Water intake and urine output were defined as the difference in weight of the water bottles and urine collections tubes, respectively, at the beginning and end of the study period. Mice were permitted free access to water and food throughout the study.

Measurement of urine osmolality. Urine samples were collected by bladder massage immediately before the study. Mice were then either water restricted by removing their water bottles or given a subcutaneous (sc) administration of DDAVP (1.0 g/kg body wt). Urine samples were collected after either 8 h of water restriction or 4 h of DDAVP administration. Urine osmolality was measured using a vapor pressure osmometer (Wescor Instruments).

Measurement of serum sodium and serum calcium levels. Blood samples were collected after either 8 h of water restriction or 4 h of DDAVP administration using a retroorbital technique. Serum sodium was measured using a flame photometer according to the directions of the manufacturer (Analytical Instruments, Golden Valley, MN). Serum calcium levels were measured by IDEXX Laboratories (Westbrook, ME).

Measurement of serum creatinine levels. Mouse serum contains noncreatinine chromagens that interfere with the Jaffé alkaline picrate method for measuring creatinine levels ( 10 ). Therefore, we assessed serum creatinine concentrations using HPLC as described ( 10 ) with the following minor modifications. The column length was increased from 50 to 150 mm to provide baseline separation in all runs, including samples that were hemolyzed. Run time was increased to 20 min with an additional 2-min equilibration period between runs because some samples exhibited peaks downfield of 15 min. Flow rate was increased to 0.5 ml/min.

Measurement of plasma vasopressin levels. Plasma vasopressin levels were measured in a separate group of mice. For these studies, mice were water restricted for 8 h and then mice were killed by rapid decapitation without anesthesia and plasma was collected in tubes containing EDTA. Plasma vasopressin was measured by RIA (Peninsula Labs, Belmont, CA) according to the recommendations of the manufacturer.

Measurement of urine sodium, creatinine, and glucose levels. Urine samples were collected by bladder massage, and glycosuria was assessed by dipstick according to the directions of the manufacturer (LW Scientific, Lawrenceville, GA). Because 24-h urine collections are contaminated by sodium-containing food, assessing urinary sodium excretion using this method is not reliable. Therefore, we measured urine sodium and urine creatinine levels in samples obtained by bladder massage. For these studies, urine creatinine concentration was measured by the Jaffé alkaline picrate method using a kit from Exocell (Philadelphia, PA). Urine sodium was measured by flame photometry according to the directions of the manufacturer (Analytical Instruments). Urine sodium excretion was expressed as a ratio of the urine sodium to urine creatinine concentration to control for differences in urine osmolality between groups.

Intraperitoneal glucose tolerance testing. After fasting for 6 h, the mice were given an intraperitoneal injection of 10% dextrose diluted with normal saline to 1 g glucose/kg body wt. Blood samples were collected from the tail veins immediately before the glucose injection and at 15, 30, 60, and 120 min afterward. Blood glucose levels were quantitated using the One Touch blood glucose meter (LifeScan) according to the directions of the manufacturer.

Papillary length measurements. After formalin fixation, kidneys were bisected and the central slice containing the renal papilla was obtained. The urinary collecting system was then carefully removed under a dissecting microscope. High-resolution quantitative computed tomography (Q-CT) was used to evaluate papillary length (µCT40; Scanco Medical, Basserdorf, Switzerland). The kidney slices containing the papilla were scanned at 55 kEv with a cone beam in high-resolution mode and a slice increment of 8 µm. Images from each group were generated at an identical threshold. Scanning was started when the papilla could first be visualized and extended proximally for 300 slices until the kidney slice was no longer visible. Morphometric analysis was performed by an investigator blinded to the treatment group. The distance between the kidney surface and the tip of the papilla was measured using the µCT40 computer software. The longest distance obtained by examining 20-30 central slices was defined as the papillary length.

Measurement of V2R and AQP2 mRNA levels. After euthanasia, the kidneys were harvested at baseline (water ad libitum) and either 2, 4, 12, or 24 h after administration of DDAVP (1.0 µg/kg sc). The kidneys were bisected on ice, and the central slice was obtained. The medulla was harvested and frozen in liquid nitrogen. Samples were stored at -80°C until total cellular RNA was prepared using TRIzol reagent (GIBCO) according to the recommendations of the manufacturer. The RT reaction was performed with Superscript reverse transcriptase (GIBCO) and oligo (dT) primers. Real-time quantitative PCR was performed using the iCycler iQ Real-Time PCR Detection System (Bio-Rad) and universal SYBR Green PCR Master Mix kit (Bio-Rad). For the studies, the thermal cycler was set at 50°C for 2 min and 95°C for 10 min before performance of 40 cycles of PCR with the cycler set at 95°C for 15 s and 60°C for 1 min, in a total volume 25 µl that included 300 nM of both primers. The amplification signals were normalized to endogenous cyclophilin A mRNA levels. The following sequences were used for the primers: V2R, AATGGCAGTGGGGTATTTGA and GTGCCACAAACACCATCAAG; AQP2, CCATGTCTCCTTCCTTCGAG and TTGTGGAGAGCATTGACAGC; and cyclophilin A, GGCCGATGACGAGCCC and TGTCTTTGGAACTTTGTCTGCAA. Data for V2R and AQP2 were expressed as relative expression by dividing the means for individual animals at each time point by the mean value for non-TG littermate controls at baseline.

Isolation of mouse glomeruli. Enriched glomerular preparations were prepared using previously described methods ( 39 ). Briefly, kidneys were harvested, bisected, and placed in ice-cold Tris-saline buffer (50 mM Tris·HCl, 154 mM NaCl, pH 7.4). The cortex was separated from the medulla, diced into small (1-2 mm 3 ) pieces, and then forced through a 200-mesh stainless steel sieve (75-µm pore diameter). The resulting suspension (30-40 ml) was forced through a 25-gauge needle and allowed to sediment by gravity for 50 min. At the end of 50 min, the upper portion of the suspension containing single cells and small tubular fragments was removed, and 30-40 ml of fresh Tris-saline buffer were added to the pellet. After vortexing, the suspension was allowed to settle by gravity for an additional 50 min on ice. The supernatant was then removed, and the pellet was used to prepare RNA with the TRIzol reagent (GIBCO) according to the recommendations of the manufacturer. Total cellular RNA was then frozen at -70°C. By light microscopy, the purity of the glomerular preparations ranged from 60 to 70%.

Measurement of transgene mRNA levels. Total cellular RNA was prepared using TRIzol reagent (GIBCO) according to the recommendations of the manufacturer. The RT reaction was performed with Superscript reverse transcriptase (GIBCO) and oligo (dT) primers. Real-time quantitative PCR was performed using the iCycler iQ Real-Time PCR Detection System (Bio-Rad) and universal SYBR Green PCR Master Mix kit (Bio-Rad). For the studies, the thermal cycler was set at 50°C for 2 min and 95°C for 10 min before performance of 40 cycles of PCR with the cycler set at 95°C for 15 s and 60°C for 1 min, in a total volume 25 µl that included 300 nM of both primers. The amplification signals were normalized to endogenous cyclophilin A mRNA levels. The following sequences were used for the primers: G L) transgene, GATGCTCAAGGCCCTTCATA and GGAATCCAGGAGTGCTACGA; and cyclophilin A, GGCCGATGACGAGCCC and TGTCTTTGGAACTTTGTCTGCAA. Transgene mRNA levels were also expressed as relative expression by dividing values for individual animals by the mean value derived from the enriched glomerular preparations.

Paired water-drinking studies. Individual TG and non-TG mice were housed in separate cages for 5 days. During this time period, water intake in the TG group was restricted to the volume of water ingested by non-TG littermate controls. Body weight and food intake were measured daily for each animal. At the end of the 5-day period, mice were injected subcutaneously with 1.0 µg/kg DDAVP. Urine was collected by bladder massage 4 h after the DDAVP injection, and urine osmolality was measured using a vapor pressure osmometer (Wescor Instruments).

Measurement of urinary albumin excretion. Urine was collected in metabolic cages specifically designed for use in mice (Hatteras Instrument, Cary, NC). Albuminuria was evaluated using an enzyme immunoassay kit from Exocell.

Statistical analysis. Data are presented as means ± SE. For comparison of continuous variables between two groups, statistical significance was assessed by a Student's t -test using the Instat computer program (GraphPad software).

RESULTS

G L) TG mice are polyuric and polydipsic. We previously reported that the G L) transgene was expressed in kidneys, pancreas, and brain by semiquantitative RT-PCR ( 39 ). Figure 1 shows transgene mRNA levels by quantitative RT-PCR using primer pairs specific for the transgene. The highest levels of transgene expression were found in enriched glomerular preparations followed by hypothalamus, cerebellum, cortex of the kidney, cortex of the brain, and pancreas. Transgene mRNA was not detectable in either non-TG mice or in the heart, lung, liver, spleen, or skeletal muscle from TG mice (data not shown). The kidney phenotype of these animals was characterized by albuminuria, reduced nephron mass, and enhanced susceptibility to podocyte injury ( 39 ). Additional studies revealed a second phenotype, characterized by polyuria and polydipsia. Figure 2 shows water intake and urine output in TG and non-TG mice at 1, 2, 4, and 6 mo of age. As shown in Fig. 2 A, water intake was significantly increased in TG animals compared with non-TG littermate controls at each time point. Figure 2 B shows that the polydipsia was associated with enhanced urine output in TG mice compared with non-TG controls at all time points studied. Table 1 shows that the polydipsia and polyuria in TG animals were not accompanied by alterations in body weight. As shown in Table 2, baseline laboratory studies found no detectable glycosuria and no significant difference in either serum calcium levels, serum creatinine levels, or urinary sodium excretion. To determine whether TG mice had a subtle abnormality in glucose metabolism, we also performed intraperitoneal glucose tolerance testing. As shown in Table 3, blood glucose levels were similar in TG and non-TG animals at baseline and at each time point studied. Moreover, the area under the intraperitoneal glucose tolerance testing curve was similar in both groups [19,605 ± 1,263 (non-TG) vs. 17,903 ± 1,897 (TG) mg·min·dl -1; P = not significant (NS)]. Because we had previously reported that TG mice were albuminuric ( 39 ), we also determined whether polyuria altered urinary albumin excretion. For these studies, non-TG mice were allowed access to 10% dextrose in water, which induced a brisk polyuric response [1.77 ± 0.18 (baseline) vs. 17.99 ± 2.97 (dextrose) ml/24 h; P = 0.0008]. The increase in urine output, however, was not associated with an increase in urinary albumin excretion [36.8 ± 9.6 (baseline) vs. 43.0 ± 11.8 µg/24 h (dextrose); P = NS]. These data indicate that the presence of the G L) transgene induces polyuria and polydipsia, which were not associated with alterations in blood glucose levels, body weight, or urinary albumin excretion.

Fig. 1. Quantitative assessment of G L) transgene expression. Previous studies found that the transgene was expressed in the kidney, pancreas, and portions of the brain ( 39 ). To quantitate transgene mRNA, quantitative RT-PCR was performed using total cellular RNA prepared from the indicated tissues, as described in METHODS, using primer pairs specific for the transgene. The transgene was expressed at highest levels in mouse glomeruli, followed by hypothalamus, cerebellum, the cortex of the kidney, cortex of the brain, and pancreas. Values are means ± SE; n = 4 TG and 4 non-TG mice. P < 0.025 or * P < 0.01 vs. glomeruli. P < 0.01 vs. hypothalamus.

Fig. 2. Effect of the transgene on water intake and urine output. A and B : daily water intake and daily urine output, respectively, in TG and non-TG mice at 1, 2, 4, and 6 mo of age. Both water intake and urine output were significantly increased in TG mice compared with non-TG littermate controls at each time point studied. Values are means ± SE; n = 24 TG and 24 non-TG mice. * P < 0.01, P < 0.05, P < 0.0001 vs. non-TG.

Table 1. Body weight by age

Table 2. Baseline laboratory studies

Table 3. Intraperitoneal glucose tolerance test

Urine concentrating ability in G L) TG mice. To determine the effect of the transgene on urine concentrating ability, we measured both urine osmolality and serum ADH levels before and after water restriction. As shown in Fig. 3 A, urine osmolality was significantly lower in TG mice compared with non-TG controls before water restriction [1,938 ± 87 (non-TG) vs. 915 ± 176 mosmol/kgH 2 O (TG); P < 0.01]. After 8 h of water restriction, urine osmolality increased significantly in both TG mice [915 ± 176 (baseline) vs. 1,820 ± 356 mosmol/kgH 2 O (water restriction); P < 0.001] and in non-TG littermate controls [1,938 ± 87 (baseline) vs. 3,891 ± 157 mosmol/kgH 2 O (water restriction); P < 0.001]. Urine osmolality, however, remained significantly lower in TG mice compared with non-TG animals after the water restriction period [3,891 ± 157 (non-TG) vs. 1,820 ± 356 mosmol/kgH 2 O (TG); P < 0.001]. Serum sodium levels increased in both TG mice [135 ± 2.1 (baseline) vs. 152 ± 2.1 mM (water restriction); P < 0.005] and controls [138 ± 1.9 (baseline) vs. 149 ± 1.4 mM (water restriction); P < 0.01] following water restriction. At the end of the water restriction period, serum sodium levels were similar in TG mice and non-TG animals [149 ± 1.4 (non-TG) vs. 152 ± 2.1 mM (TG); P = NS]. To determine whether ADH was secreted appropriately in response to dehydration, we measured serum vasopressin levels after 8 h of water restriction. As shown in Fig. 3 B, serum ADH levels increased significantly in both TG mice (113 ± 21 to 400 ± 41 pg/ml; P < 0.005) and non-TG controls (142 ± 32 to 283 ± 34 pg/ml; P < 0.001) after 8 h of water restriction. These data suggest that TG mice have reduced urine concentrating ability and that this urine concentrating defect is not the result of reduced release of vasopressin.

Fig. 3. Effect of water restriction on urine osmolality and plasma antidiuretic hormone (ADH) levels. A : urine osmolality levels before and after water restriction. Urine osmolality was significantly reduced in TG mice compared with non-TG controls before water restriction. After 8 h of water restriction, urine osmolality was significantly increased in both groups but remained significantly lower in TG mice compared with non-TG animals. B : serum ADH levels before and after water restriction. ADH levels tended to be lower in TG mice compared with non-TG mice before water restriction. After 8 h of water restriction, serum ADH levels increased significantly in both groups. Values are means ± SE; n = 8 TG and 8 non-TG controls. * P < 0.01, P < 0.001 vs. non-TG. P < 0.005, P < 0.001 vs. before water restriction.

To further investigate the ability of ADH to enhance water conservation by the kidney, mice were given DDAVP and urine osmolality was measured immediately before and 4 h after treatment. As shown in Fig. 4, urine osmolality was significantly lower in TG mice compared with non-TG animals at baseline [2,120 ± 163 (non-TG) vs. 1,157 ± 302 mosmol/kgH 2 O (TG); P < 0.01]. Treatment with DDAVP significantly increased urine osmolality in both groups. Similar to the water restriction studies, however, urine osmolality remained significantly lower in the TG mice compared with non-TG controls after treatment with DDAVP [3,922 ± 184 (non-TG) vs. 2,491 ± 416 mosmol/kgH 2 O (TG); P < 0.005]. These data suggest that a pharmacological dosage of DDAVP does not correct the urine concentrating defect.

Fig. 4. Effect of DDAVP on urine osmolality. Urine osmolality was measured immediately before and 4 h after administration of DDAVP (1.0 µg/kg sc). Before DDAVP, urine osmolality was significantly reduced in TG mice compared with non-TG controls. Urine osmolality increased significantly in both groups after DDAVP but remained significantly lower in TG mice compared with non-TG animals. Values are means ± SE; n = 9 TG and non-TG controls. * P < 0.01, P < 0.005 vs. non-TG. P < 0.025, P < 0.005 vs. before DDAVP injection.

We next determined whether kidneys of TG mice responded appropriately to ADH at the molecular level. These studies were based on previous investigations ( 38 ) in rodents which suggested that AQP2 and V2R mRNA levels were upregulated and downregulated, respectively, after treatment with ADH. For these studies, total cellular mRNA was collected from the kidney medulla immediately before and at the indicated times after administration of DDAVP. Levels of V2R and AQP2 mRNA were measured using quantitative RT-PCR. As shown in Fig. 5 A, V2R mRNA levels were significantly higher in TG mice compared with non-TG animals at baseline. Treatment with DDAVP significantly reduced V2R mRNA levels in both groups at the 2-h time point. In contrast to previous reports in rats ( 38 ), V2R mRNA levels were significantly increased compared with baseline in both TG mice and non-TG controls at the 12-h time point. At the 24-h time point, V2R mRNA levels were similar to the expression pattern at baseline. Figure 5 B shows results for AQP2. In contrast to V2R mRNA levels, AQP2 mRNA levels were similar in TG and non-TG mice at baseline. After treatment with DDAVP, AQP2 mRNA levels increased significantly in both groups, with the highest levels observed at the 12-h time point. Taken together, these data suggest that kidneys from TG mice respond appropriately to ADH at the molecular level. While this response is qualitatively similar to the pattern observed in non-TG controls, V2R mRNA levels were increased in TG mice compared with non-TG littermate controls at baseline.

Fig. 5. Effect of DDAVP on V2R and AQP2 mRNA levels in renal medulla. For these studies, V2R and AQP2 mRNA levels were assessed by quantitative RT-PCR before and at the indicated times after administration of DDAVP as described in METHODS. A : results for V2R mRNA levels. At baseline, V2R mRNA levels were significantly increased in TG mice compared with non-TG animals. Treatment with DDAVP significantly reduced V2R mRNA levels in both groups at the 2-h time point and significantly increased V2R mRNA levels at the 12-h time point. By 24 h, V2R mRNA levels were similar to the expression pattern at baseline. B : results for AQP2 mRNA levels. AQP2 mRNA levels were similar in TG and non-TG mice at baseline. After treatment with DDAVP, AQP2 mRNA levels increased significantly in both groups. By 24 h, AQP2 mRNA levels had returned to baseline in both groups. Values are means ± SE; n = 13 non-TG and 13 TG mice studied at baseline, n = 14 non-TG and 14 TG mice studied at 2 h, and n = 4 non-TG and 4 TG mice studied at the 4-, 12-, and 24-h time points. P < 0.025 vs. non-TG. P < 0.025, * P < 0.01, P < 0.005, ** P < 0.0001 vs. baseline.

Papillary length is an important determinant of urine concentrating ability ( 2, 32 ). We, therefore, determined whether a reduction in papillary length contributed to the reduction in urine concentrating ability observed in our TG animals. For these studies, we determined papillary length in both TG mice and non-TG controls by high-resolution Q-CT scan as described in METHODS. Papillary length was measured between the kidney surface and the tip of the papilla, as indicated in Fig. 6. There was no significant difference in papillary length between TG mice and their non-TG littermates [6.7 ± 0.2 (non-TG) vs. 6.6 ± 0.1 mm (TG); P = NS]. These data indicate that structural abnormalities are not likely the cause of the urine concentrating defect.

Fig. 6. Measurement of papillary length. A high-resolution quantitative computed tomography scan was used to determine papillary length as described in METHODS. The papillary length measurements were made between the kidney surface and the tip of the papilla using µCT40 computer software as indicated.

Urine concentrating ability after paired water drinking. Previous studies suggest that high urine flow rates decrease medullary interstitial tonicity and, in turn, cause a reduction in maximal urine concentrating ability ( 11, 19 ). To investigate this possibility, water intake by TG mice was restricted to the amount ingested by non-TG controls for 5 days. As shown in Table 4, both TG and non-TG mice lost weight during the paired water-drinking period. The weight loss was significantly greater in TG mice compared with non-TG controls. Figure 7 shows urine concentrating ability in response to DDAVP following the period of paired water drinking. Urine osmolality was similar in TG mice and non-TG littermate controls before DDAVP [2,236 ± 210 (non-TG) vs. 2,196 ± 253 mosmol/kgH 2 O (TG); P = NS]. Four hours after treatment with dDAVP, urine osmolality increased to a similar extent in TG and non-TG animals [3,218 ± 289 (non-TG) vs. 3,068 ± 261 mosmol/kgH 2 O (TG); P = NS]. These studies are consistent with the notion that the urine concentrating defect is corrected by decreasing urine flow rates, which, in turn, may restore medullary tonicity.

Table 4. Paired water-drinking study

Fig. 7. Effect of paired water drinking on urine concentrating ability. Paired water drinking was performed as described in METHODS. After 5 days of paired water consumption, urine osmolality was measured immediately before and 4 h after administration of DDAVP (1.0 µg/kg sc). At baseline, urine osmolality was similar in both groups. After DDAVP, urine osmolality increased significantly in both groups to similar levels. Values are means ± SE; n = 20 non-TG and 20 TG mice. * P < 0.01 vs. before ADH.

DISCUSSION

Fluid homeostasis is mediated by coordinated and redundant physiological processes that include water intake, release of ADH, and renal concentrating mechanisms ( 1 ). We found that G L) TG mice were both polyuric and polydipsic and that these abnormalities were associated with reduced urine concentrating ability after either water restriction or DDAVP administration. The renal concentrating abnormality did not appear to be the result of abnormal glucose metabolism, hypercalcemia, or renal structural defects, and kidneys from TG mice appeared to respond appropriately to ADH at the molecular level. The inability of TG mice to maximally concentrate their urine was, however, corrected by paired water drinking for 5 days. These data are compatible with the idea that the reduction in urine concentrating ability is the result of high urine flow rates, which may decrease medullary interstitial tonicity and reduce maximal urine concentrating ability ( 11, 19 ). Taken together, these data suggest that primary polydipsia is the cause of the polyuria in our TG mice and, in turn, causes the renal concentrating defect in the TG animals. G q-dependent signaling cascades, therefore, are likely to play a key role in regulating drinking behavior.

Regulation of thirst has been an intense area of study by neurophysiologists. As little as a 1-2% increase in effective osmotic pressure stimulates osmoreceptors in the brain, which activate neuronal mechanisms that generate thirst ( 1, 22 ). Studies over the last several decades have localized the osmoreceptive neurons to the hypothalamic regions of the brain ( 21, 22 ). These regions are localized to preoptic/hypothalamic areas in the anteroventral third ventricular wall, including the organum vasculosum of the lamina terminalis (OVLT) and the median preoptic nucleus (MnPO) ( 4, 21, 22 ). An important aspect of the regulatory response is that ANG II-dependent signaling cascades play a crucial role in the neuronal pathways stimulating thirst ( 1, 11, 22 ). The MnPO is rich in AT 1 receptors, is not accessible to systemic ANG II, and is likely to be the site of the angiotensinergic synapses that respond to osmotic and hormonal stimuli to generate a thirst response ( 15, 22 ). In this regard, studies by Sigmund and co-workers ( 8 ) have found that centrally administered ANG II stimulates drinking behavior predominantly through the AT 1B receptor in mice, with a lesser contribution of the AT 1A receptor. Both these AT 1 receptor systems are coupled to the G proteins belonging to the Gq family ( 30 ). Consistent with an important role for G q-dependent pathways in generating thirst, the present studies suggest that 1 ) the G L) transgene was highly expressed in the hypothalamus; and 2 ) the phenotype of TG mice was characterized by polydipsia and polyuria, which did not appear to be the result of either central or nephrogenic diabetes insipidus ( 33 ). In this scenario, the constitutively active G L) obviates the need for AT 1 receptor-induced G q activation and, in turn, the G L) transgene directly stimulates water-seeking behavior.

In addition to the paired water-drinking studies, there is other evidence that the polyuria did not result from diabetes insipidus. First, both TG and non-TG mice responded appropriately to DDAVP at the molecular level ( 38 ), with a decrease in V2R mRNA levels and an increase in AQP2 mRNA levels. Second, V2R mRNA levels were higher in TG mice compared with non-TG mice at baseline. In this regard, the level of expression of multiple hormone receptor systems has been shown to be regulated by the ambient concentration of ligand ( 9 ). Thus in the continuous presence of ligand, receptor expression is downregulated ( 9 ). With regard to the V2R, both chronic and acute changes in ADH levels have been shown to regulate V2R mRNA levels ( 34, 37, 38 ). In this scenario, V2Rs of non-TG mice are chronically exposed to higher concentrations of ADH than TG animals because control mice need to generate a urine that is more concentrated than TG animals to remain in water balance. In this setting, V2R mRNA levels would be downregulated in non-TG mice compared with TG mice at baseline. Consistent with this hypothesis, serum ADH levels tended to be higher in non-TG mice compared with TG animals ( Fig. 3 B ), although the difference was modest. However, measurements of ADH, even if carefully collected, tend to be easily affected by external factors. For example, both strong emotions and stress stimulate ADH release ( 36 ). Even under the most carefully controlled circumstances, it is likely that these factors play a role in serum ADH levels measurements in mice. Taken together, these findings provide additional evidence that the polyuric phenotype of our TG mice was caused by primary polydipsia.

We had previously found that kidneys of TG mice appeared structurally normal by routine light microscopy, although nephron mass was modestly reduced in TG mice compared with sex- and age-matched controls ( 39 ). In humans, a decrease in nephron mass has little effect on water metabolism until either there is a large reduction in renal function or the medullary architecture is disrupted ( 1, 33 ). In this regard, serum creatinine levels were similar in TG and non-TG animals, suggesting that the residual nephrons in the TG mice had compensated for the reduced nephron mass. Moreover, medullary architecture was normal by light microscopic examination ( 39 ). Taken together, these data suggest that the polyuria was not likely the result of either reduced nephron mass or an obvious structural defect. We were concerned, however, that a subtle structural defect might contribute to the urine concentrating abnormality in our TG animals. For example, mice deficient in AT 1A receptors have a renal concentrating abnormality and a subtle decrease in papillary length that is not detectable using routine histological methods ( 26 ). In mammals, papillary length correlates with urine concentrating capacity ( 2, 32 ). We, therefore, determined papillary length using a high-resolution Q-CT method. These studies revealed a similar papillary length in TG and non-TG mice. As a result, it appears unlikely that papillary length plays an important role in the urinary concentrating defect in our TG mice.

Last, primary polydipsia is sometimes associated with hyponatremia ( 1 ). Compulsive water drinking, however, only results in hyponatremia if water intake exceeds the ability of the kidney to excrete the water load ( 1, 33 ). In this regard, while most patients with primarily polydipsia will have a serum sodium level of <140 mM ( 1 ), patients are not necessarily hyponatremic unless either water intake is massive or solute intake is reduced ( 1, 33 ). In the present study, water intake was increased 40-50% in TG mice compared with non-TG littermate controls. Given the large capacity of the kidney to maintain fluid homeostasis ( 1, 33 ), it is not likely that a 40-50% increase in water intake above normal water consumption rates would cause a change in serum sodium levels. Indeed, baseline serum sodium measurements were similar in TG mice and non-TG controls, indicating that fluid homeostasis was maintained in the TG animals.

In summary, TG mice expressed a constitutively active Gq -subunit in portions of the brain that play a key role in regulating thirst. The phenotype of TG mice was characterized by polyuria, polydipsia, and a urinary concentrating defect that did not appear to be the result of abnormal glucose metabolism, hypercalcemia, insufficient ADH release, a renal structural defect, or an inability of kidneys from TG mice to respond to ADH at the molecular level. The renal concentrating abnormality was corrected by paired water drinking, suggesting that increased urine flow rates and, in turn, a reduction in medullary tonicity reduced maximal urine concentrating ability in the TG mice. These data are consistent with the notion that polyuria in TG mice was the result of primary polydipsia. G q-dependent signaling cascades are, therefore, likely to play an important role in regulating drinking behavior.

GRANTS

These studies were supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01-DK-065956 (R. F. Spurney).

【参考文献】
  Andreoli TE. Water: normal balance, hyponatremia, and hypernatremia. Renal Failure 22: 711-735, 2000.

Bankir L and de Rouffignac C. Urinary concentrating ability: insights from comparative anatomy. Am J Physiol Regul Integr Comp Physiol 249: R643-R666, 1985.

Bumpus FM, Catt KJ, Chiu AT, DeGasparo M, Goodfriend T, Husain A, Peach MJ, Taylor DG, and Timmermans BMWM. Nomenclature for angiotensin receptors: a report of the nomenclature committee of the council for high blood pressure research. Hypertension 17: 720-721, 1991.

Carpenter MB. The hypothalamus. In: Core Text of Neuroanatomy, edited by Carpenter MB. Baltimore, MD: Williams and Wilkins, 1978.

Chen X, Li W, Yoshida H, Tsuchida S, Nishimura H, Takemoto F, Okubo S, Fogo A, Matsusaka T, and Ichikawa I. Targeting deletion of angiotensin type 1B receptor gene in the mouse. Am J Physiol Renal Physiol 272: F299-F304, 1997.

Clarkson MR, McGinty A, Godson C, and Brady HR. Leukotrienes and lipoxins: lipoxygenase-derived modulators of leukocyte recruitment and vascular tone in glomerulonephritis. Nephol Dial Transplant 13: 3043-3051, 1998.

Crowley SD, Gurley SB, Oliverio MI, Pazmino AK, Griffiths R, Flannery PJ, Spurney RF, Kim H, Smithies O, Le TH, and Coffman TM. Distinct roles for the kidney and systemic tissues in blood pressure regulation by the renin-angiotensin system. J Clin Invest 115: 1092-1099, 2005.

Davisson RL, Oliverio MI, Coffman TM, and Sigmund CD. Divergent functions of angiotensin II receptor isoforms in the brain. J Clin Invest 106: 103-106, 2000.

Dohlman HG, Thorner J, Caron MG, and Lefkowitz RJ. Model systems for the study of seven-transmembrane-segment receptors. Annu Rev Biochem 60: 653-688, 1991.

Dunn SR, Qi Z, Bottinger EP, Breyer MD, and Sharma K. Utility of endogenous creatinine clearance as a measure of renal function in mice. Kidney Int 65: 1959-1967, 2004.

Epstein FH, Kleeman CR, and Hendrikx A. The influence of bodily hydration on the renal concentrating process. J Clin Invest 36: 629-634, 1957.

Franci CR, Kozlowski GP, and McCann SM. Water intake in rats subjected to hypothalamic immunoneutralization of angiotensin II, atrial natriuretic peptide, vasopressin, or oxytocin. Proc Natl Acad Sci USA 86: 2952-2956, 1989.

Ito M, Oliverio MI, Mannon PJ, Best CF, Maeda N, Smithies O, and Coffman TM. Regulation of blood pressure by the type 1A angiotensin II receptor gene. Proc Natl Acad Sci USA 92: 3521-3525, 1995.

Iwai N and Inagami T. Identification of two subtypes in the rat type I angiotensin receptor. FEBS Lett 298: 257-260, 1992.

Johnson AK, Cunningham JT, and Thunhorst RL. Integrative role of the lamina terminalis in the regulation of cardiovascular and body fluid homeostasis. Clin Exp Pharmacol Physiol 23: 183-191, 1996.

Johnson AK and Thunhorst RL. The neuroendocrinology of thirst and salt appetite: visceral sensory signals and mechanisms of central integration. Front Neuroendocrinol 18: 292-353, 1997.

Kalinec F, Nazarali AJ, Hermouet S, Xu N, and Gutkind JS. Mutated subunit of Gq protein induces malignant transformation of NIH 3T3 cells. Mol Cell Biol 12: 4687-4693, 1992.

Kohan DE. Endothelins in the kidney: physiology and pathophysiology. Am J Kidney Dis 22: 493-510, 1993.

Levitin HA, Goodman A, Pigeon G, and Epstein FH. Composition of the renal medulla during water diuresis (Abstract). J Clin Invest 41: 1145, 1962.

Lopez-Novoa J. Potential role of platelet activating factor in acute renal failure. Kidney Int 55: 1672-1682, 1999.

Makino H, Tanaka I, Mukoyama M, Sugawara A, Mori K, Muro S, Suganami Y, Yahata K, Ishibashi R, Ocuchida 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.

McKinley MJ, McAllen RM, Davern P, Giles ME, Penschow J, Sunn N, Uschakov A, and Oldfield BJ. The sensory circumventricular organs of the mammalian brain. Adv Anat Embryol Cell Biol 172: 1-122, 2003.

McKinley MJ and Johnson A. The physiological regulation of thirst and fluid intake. News Physiol Sci 19: 1-6, 2004.

Moeller JM, Kovari IA, and Holzman LB. Evaluation of a new tool for exploring podocyte biology: mouse Nphs1 5' flanking region drives LacZ expression in podocytes. J Am Soc Nephrol 11: 2306-2314, 2000.

Murphy TJ, Alexander RW, Griendling KK, Runge MS, and Bernstein KE. Isolation of a cDNA encoding the vascular type-1 angiotensin II receptor. Nature 351: 233-236, 1991.

Oliverio MI, Delnomdedieu M, Best CF, Li P, Morris M, Callahan MF, Johnson GA, Smithies O, and Coffman TM. Abnormal water metabolism in mice lacking the type 1A receptor for ANG II. Am J Physiol Renal Physiol 278: F75-F82, 2000.

Phillips MI. Functions of angiotensin II in the central nervous system. Annu Rev Physiol 49: 413-435, 1987.

Qadri FJ, Culman J, Veltmar A, Mass K, Rascher W, and Unger T. Angiotensin II-induced vasopressin release is mediated through alpha-1 adrenoceptors and angiotensin II AT 1 receptors in the supraoptic nucleus. J Pharmacol Exp Ther 267: 567-574, 1993.

Sandberg K, Ji H, Clark AJ, Shapira H, and Catt KJ. Cloning and expression of a novel angiotensin II receptor subtype. J Biol Chem 267: 9455-9458, 1991.

Sanderberg KJH. Kidney angiotensin receptors, and their role in renal pathophysiology. Semin Nephrol 20: 402-416, 2000.

Sasaki K, Yamano Y, Bardlan S, Iwai N, Murray JJ, Hasegawa M, Matsuda Y, and Inagami T. Cloning and expression of a complementary DNA encoding a bovine adrenal angiotensin II type-1 receptor. Nature 351: 230-233, 1991.

Schmidt-Nielson B and O'Dell R. Structure and concentrating mechanism in the mammalian kidney. Am J Physiol 200: 1119-1124, 1961.

Schrier RW and Berl T. Disorders of water metabolism. In: Renal and Electrolyte Disorders, edited by Schrier RW. Boston, MA: Little, Brown, 1976.

Steiner M and Phillips MI. Renal tubular vasopressin receptors downregulated by dehydration. Am J Physiol Cell Physiol 254: C404-C410, 1988.

Stork JE, Rahman MA, and Dunn MJ. Eicosanoids in experimental human renal disease. Am J Med 80: 34-45, 1986.

Streeten DH, Moses AM, and Miller M. Disorders of the neurohypophysis. In: Harrison's Principal of Internal Medicine (11th ed.), edited by Braunwald E, Isselbacher KJ, Petersdorf RG, Wilson JD, Martin JB, and Fauci AS. New York: McGraw-Hill, 1987.

Terada Y, Tomita K, Nonoguchi H, Yang T, and Marumo F. Different localization and regulation of two types of vasopressin receptor messenger RNA in microdissected rat nephron segments using reverse transcription polymerase chain reaction. J Clin Invest 92: 2339-2345, 1993.

Terashima Y, Kondo K, Mizuno Y, Iwasaki Y, and Oiso Y. Influence of acute elevation of plasma AVP level on rat vasopressin V2 receptor and aquaporin-2 mRNA expression. J Mol Endocrinol 20: 281-285, 1998.

Wang L, Pazmino K, Fields T, Dai Q, Howell DN, Burchette JL, Coffman TM, and Spurney RF. Activation of G q-coupled signaling pathways in glomerular podocytes promotes renal injury. J Am Soc Nephrol 16: 3611-3622, 2005.

Whitesall SE, Hoff JB, Vollmer AP, and Alecy LGD. Comparison of simultaneous measurement of mouse systolic arterial blood pressure by radiotelemetry and tail-cuff methods. Am J Physiol Heart Circ Physiol 286: H2408-H2415, 2004.

作者单位：1 Division of Nephrology, Department of Medicine, Duke University, and Durham Veterans Affairs Medical Centers, Durham, North Carolina; and 2 Division of Nephrology, Department of Medicine, Thomas Jefferson University, Philadelphia, Pennsylvania