Urea transport in the kidney is mediated by a family of transporter proteins, including renal urea transporters (UT-A) and erythrocyte urea transporters (UT-B). We aimed to determine whether hydration status affects the subcellular distribution of urea transporters. Male Sprague-Dawley rats were divided into three groups: dehydrated rats (WD) given minimum water, hydrated rats (WL) given 3% sucrose in water for 3 days before death, and control rats given free access to water. We labeled kidney sections with antibodies against UT-A1 and UT-A2 (L194), UT-A3 (Q2), and UT-B using preembedding immunoperoxidase and immunogold methods. In control animals, UT-A1 and UT-A3 immunoreactivities were observed throughout the cytoplasm in inner medullary collecting duct (IMCD) cells, and weak labeling was observed on the basolateral plasma membrane. UT-A2 immunoreactivity in the descending thin limbs (DTL) was observed mainly on the apical and basolateral membranes of type I epithelium, and very faint labeling was observed in the long-loop DTL at the border between the outer and inner medulla. UT-A1 immunoreactivity intensity was markedly lower, and UT-A3 immunoreactivity was higher in IMCD of WD vs. controls. UT-A2 immunoreactivity intensities in the plasma membrane and cytoplasm of type I, II, and III epithelia of DTL were greater in WD vs. controls. In contrast, UT-A1 expression was greater and UT-A2 and UT-A3 expressions were lower in WL vs. controls. The subcellular distribution of UT-A in DTL or IMCD did not differ between control and experimental animals. UT-B was expressed in the plasma membrane of the descending vasa recta of both control and experimental animals. UT-B intensity was higher in WD and lower in WL vs. controls. These data indicate that changes in hydration status over 3 days affected urea transporter protein expression without changing its subcellular distribution.
- urea transporter A
- urea transporter B
- hydration status
- subcellular localization
in the mammalian kidney, urea transport is important for the formation of concentrated urine (7, 14, 29, 44). In the kidney and red blood cells (RBC), urea transport is mediated by specific transport proteins, the renal urea transporters (UT-A) and the erythrocyte urea transporter (UT-B). These are encoded by two genes, Slc14a2 (UT-A) and Slc14a1 (UT-B), which occur in tandem on chromosome 18 (3, 10, 11, 33).
To date, the cDNAs of five isoforms of rat UT-A (UT-A1, UT-A2, UT-A3, UT-A4, and UT-A5) have been cloned (12, 21, 44, 47, 48). The rat UT-A gene is very long (∼300 kb) and contains 24 exons (33). It is an atypical gene because it has two promoter elements: promoter I, which is located upstream of exon 1 and drives transcription of UT-A1, UT-A1b, UT-A3, UT-A3b, and UT-A4; and promoter II, which is located within intron 12 and drives transcription of UT-A2 and UT-A2b (4, 33). The cDNA of UT-B was first cloned from a human bone marrow library and subsequently isolated from a rat inner medullary library by homology screening (9, 39, 56) The abundance of these proteins is regulated by vasopressin (28, 46, 62), glucocorticoids (35, 40), and mineralocorticoids (13).
UT-A1 and UT-A3 are localized in the middle and terminal inner medullary collecting duct (IMCD) cells (37). UT-A2 is located in type I and type III epithelia of the loop of Henle (45, 58). Exogenous administration of vasopressin or water restriction decreases the UT-A1 protein abundance, and an increase in UT-A1 protein abundance is associated with several conditions involving reduced urine-concentrating ability. However, most studies have reported no change in mRNA abundance in response to water loading or restriction. In contrast, vasopressin administration or water restriction increases the levels of UT-A2 and UT-A3 mRNA (4, 10, 47, 58). Relatively little is known about UT-A4 compared with other members of the UT-A family. UT-A4 has been detected in the outer medulla by immunoblotting, but its mRNA abundance in the renal medulla is too low to be detected by Northern blot analysis (58).
UT-B protein has been detected by immunohistochemistry in the endothelial cells of human and rat descending vasa recta (DVR) (60) and by immunoblotting in the human RBC membrane (38). In the RBC, the fast urea transport mediated by UT-B is believed to be essential for RBC stability when the cells pass through the hypertonic renal medulla (32). As the result of a countercurrent exchange process, part of the urea escaping the medulla in the ascending vasa recta (AVR) is trapped in the DVR and returns to the medulla. In this way, UT-B helps maintain a high medullary urea concentration and thus the hypertonic medulla required for water reabsorption. Promeneur et al. (41) reported an increase in UT-B mRNA in chronically 1-deamino, 8D-arginine vasopressin-treated Brattleboro rats after 5 days (41).
Until now, most studies have examined global changes in urea transporters by measuring either mRNA or protein levels. Knowledge about the localization of these transporters, especially at the subcellular level, is also required to understand their role in the kidney. However, only a few studies have reported on the subcellular localization of UT-A1 in the IMCD cells (20, 25, 37), and little is known about the subcellular localization of UT-A2, UT-A3, and UT-B in rat kidney.
The purpose of this study was to examine the subcellular localization of UT-As and UT-B in the normal rat kidney and how this might be affected by hydration status using light and electron microscopy (EM).
MATERIALS AND METHODS
Animals and Tissue Preservation
Male Sprague-Dawley rats, weighing ∼250 g, were housed in individual cases in a temperature- and light-controlled environment in the Catholic University Animal Care Facility. The protocol used in these studies was approved by the Catholic University Animal Care Committee. The animals were divided into three groups: control group (n = 6) with free access to water, dehydrated group (n = 6) given a small amount of water mixed with rat chow (∼ 10 ml/day) to cover insensible losses for 3 days, and water-loaded group (n = 6) with free access to 3% sucrose water for 3 days before death. All animals had unrestricted access to standard rat chow (Samyang Oil & Feed, Seoul, Korea; salt content is 1%). Animals were anesthetized with an intraperitoneal injection of 16.5% urethane (10 ml/kg), and blood was collected from the abdominal aorta. The kidneys were preserved by in vivo perfusion through the abdominal aorta. The animals were briefly perfused with 0.01 M PBS, pH 7.4, to rinse out the blood. This was followed by perfusion with a periodate-lysine-paraformaldehyde or 8% paraformaldehyde solution for 10 min. The kidneys were removed and cut into sagittal slices of 1- to 2-mm thickness and postfixed overnight in the periodate-lysine-paraformaldehyde or 2 h in the paraformaldehyde solution at 4°C.
Before death, animals were individually housed in metabolic cages (Tecniplast Gazzda, Beguggiate, Italy) for 24-h urine collection to evaluate urine volume and osmolality; blood samples were obtained to evaluate serum creatinine, blood urea nitrogen, concentration of electrolytes (sodium, potassium), serum osmolality, and aldosterone. Serum creatinine, blood urea nitrogen, and glucose were measured using Hitachi autoanalyzer (Hitachi, Tokyo, Japan). Cubas autoanalyzers (Roche Diagnostics, Nutley, NJ) were used for electrolytes concentration. Osmolalities in urine and serum were measured with an osmometer (The Advanced MicroOsmometer, LabTrader, Vista, CA). Aldosterone levels in serum were measured by radioimmunoassay kit (Diagnostic Systems Laboratories, Webster, TX). The creatinine clearance rate was calculated with standard formula.
To determine the distribution of urea transporters in the rat kidney, we used specific rabbit polyclonal antibodies against peptides based on the rat renal urea transporter UT-A1, UT-A2, and UT-A4 (L194 and L403) (37, 51, 58); UT-A1, UT-A3, and UT-A4 (L446) (58); UT-A3 (Q2) (52); and UT-B (38, 53), the human erythrocyte urea transporter. We identified the thick ascending limb (TAL) of the loop of Henle using rabbit polyclonal antibody against Na-K-ATPase α1 (Upstate Biotechnology, Lake Placid, NY), and the descending thin limb of loop of Henle was marked by use of rabbit polyclonal antibody against aquaporin-1 (AQP1; Chemicon, Temecula, CA). We probed for β-actin protein, a housekeeping gene, using β-actin antibody (Sigma-Aldrich, St. Louis, MO).
Preembedding Immunoperoxidase Method
As in previous reports (20, 25, 31), vibratome sections (50 μm thick) were used. In brief, before incubation with the primary antibodies, the tissue sections were incubated for 3 h with 1% BSA, 0.05% saponin, and 0.2% gelatin-PBS (solution B). The tissue sections were then incubated overnight at 4°C in rabbit antisera against L194 (1:1,000), L403 (1:1,000), L446 (1:800), Q2 (1:800), and UT-B (1:3,000) in 1% BSA-PBS (solution A). After several washes with 0.1% BSA, 0.05% saponin, and 0.2% gelatin-PBS (solution C), the tissue sections were incubated for 2 h in peroxidase-conjugated donkey anti-rabbit IgG Fab fragment (Jackson ImmunoResearch Laboratories, West Grove, PA). The tissues were then rinsed, first in solution C and subsequently in 0.05 M Tris buffer, pH 7.6. For the detection of horseradish peroxidase, sections were incubated in 0.1% 3,3′-diaminobenzidine and H2O2. The sections were then embedded in poly/Bed 812 resin (Polysciences, Warrington, CA). For EM, sections were postfixed with 1% glutaraldehyde, 1% osmium tetroxide, and 1% uranyl acetate, before being dehydrated and embedded in poly/Bed 812 resin. Ultrathin sections were cut and photographed with a transmission electron microscope (JEOL 1200EX, Tokyo, Japan).
Vibratome sections were labeled with antibody against L194 using 3,3′-diaminobenzidine as the chromogen (brown color) as described above. The sections were incubated for 1 h in solution B and then incubated overnight at 4°C in rabbit antisera against Na-K-ATPase α1 (1:500) or AQP1 (1:500). After several washes with PBS, the tissue sections were incubated for 2 h in horseradish peroxidase-conjugated donkey anti-rabbit IgG Fab fragment (Jackson ImmunoResearch Laboratories). For detection, the sections were incubated in vector SG (blue color; Vector Laboratories). After sections were washed, they were dehydrated in a graded series of ethanol and embedded in poly/Bed 812 resin.
Preembedding Immunogold Method
This method was described in a previous report (20). In brief, vibratome sections were incubated with primary antibodies overnight as described above, washed with solution C, and then washed with 0.8% BSA-0.1% gelatin-2 mM NaN3-PBS and 5% normal goat serum, pH 7.4 (gold buffer). The tissue sections were then incubated overnight at 4°C in NANOGOLD-IgG and Fab′ conjugates (Nanoprobes, Yaphank, NY). After they were washed with PBS, labeled gold was postfixed with 1% glutaraldehyde for 10 min. The tissue sections were enhanced for 7–8 min by HQ silver system (Nanoprobes). The tissue sections were rinsed, postfixed, and embedded for EM as described above.
Semiquantitative Immunoblot Analysis
For immunoblotting analysis, a kidney from each animal was dissected into cortex, outer medulla, and inner medulla or medulla as described previously (31). In brief, the tissues were homogenized in lysis buffer containing 0.3 M sucrose, 25 mM imidazole, and 1 mM EDTA. The homogenates were centrifuged at 4,000 g for 20 min at 4°C. The supernatant was centrifuged at 200,000 g for 1 h (Beckman Instruments). Protein concentrations were determined with a BCA protein assay reagent kit (Pierce, Rockford, IL). Samples were separated by SDS-PAGE (Bio-Rad, Hercules, CA) and then transferred to nitrocellulose membranes by electroblotting. The membranes were blocked with 5% nonfat dried milk in Tris-buffered saline for 1 h and then incubated for 24 h at 4°C with primary antibodies as described above. The membranes were incubated for 1 h with peroxidase-labeled anti-rabbit/mouse IgG (Amersham Biosciences, Buckinghamshire, UK). Samples were visualized after a 1- to 2-min exposure to enhanced chemiluminescence (Amersham Biosciences). Densitometric analysis was performed using the Zero-Dscan software of the Eagle Eye II Still Video system (Stratagene, La Jolla, CA). Optical densities were obtained after three determinations for each band.
Results are presented as means ± SE. Comparisons between three groups were done by ANOVA (Kruskal-Wallis test followed by Tukey’s or Dunnet’s test). The level of statistical significance was accepted as P < 0.05.
Table 1 shows the functional parameters of the control and experimental groups. There was no difference in initial body weight among the experimental rats. At the end of the experimental day, dehydrated animals had lost significantly more body weight than the control or water-loaded animals. Serum osmolality and sodium concentration did not differ between the groups, but serum potassium concentration was slightly lower in the water-loaded animals than in the controls. Urine volume and urine osmolality changed markedly in response to changes in water intake, but there were minimal effects on serum electrolytes. Urine osmolality was 2,089 ± 112 mosmol/kgH2O in control animals, but decreased significantly to 1,008 ± 128 mosmol/kgH2O in the water-loaded animals and increased to 4,628 ± 325 mosmol/kgH2O in the dehydrated animals. Amount of glucose in urine was not different in among groups. Aldosterone concentrations were significantly higher in dehydrated animals than in the control or water-loaded groups.
Light Microscopic Immunohistochemistry
Expression of UT-A in control rat kidney.
To confirm that each subtype of UT-A is expressed in the rat kidney, we used immunohistochemistry with antibodies L194 and L403 (UT-A1, UT-A2, and UT-A4), L446 (UT-A1, UT-A3 and UT-A4), and Q2 (UT-A3). As previously reported (37, 58), L194/L403 antibodies were immunolabeled in the IMCD in the middle and terminal parts of the inner medulla (IM) (UT-A1), in the descending thin limbs (DTL) of the loop of Henle in the inner stripe of the outer medulla (ISOM), and in the initial part of the IM (IMi) (UT-A2) (Fig. 1, A and B). There was no labeling in the cortex and OSOM. L446 (UT-A1, UT-A3, and UT-A4) was localized only in the IMCD in the middle and terminal parts of the IM (Fig. 1C). Q2 immunoreactivity (UT-A3) was also observed in the middle and terminal parts of the IMCD (Fig. 1D). High-magnification images revealed that strong positive DTL immunoreactivities were restricted to the ISOM, whereas weak positive DTL immunoreactivities were observed both in the lower part of ISOM and in the IMi (Fig. 1, E and F). Because no L446 and Q2 immunoreactivities were observed in the ISOM and IMi (Fig. 1, G and H), we conclude that the immunoreactivity for L194/L403 in these regions was due to the presence of UT-A2 protein in the DTL. In deeper regions of the IM, there was no L194/L403 immunoreactivity in the DTL (Fig. 1, I and J).
To determine where UT-A2 is expressed in the ISOM, we performed double-labeling experiments using antibodies directed against UT-A (L194) and Na-K-ATPase α1, which is expressed only in the TAL (Fig. 2A) and the distal convoluted tubules (20). The strongly labeled tubular profiles located in the ISOM were identified as the terminal part of the short-loop DTL by the abrupt transition to the Na-K-ATPase α1-positive TAL, which usually bends at this level (Fig. 2B). At the border between the ISOM and IM, weakly labeled thin limbs were observed, which were not connected to Na-K-ATPase α1-positive TAL but were colabeled with AQP1, which is a marker of long-loop DTL (20) (Fig. 2, E and F).
Expression of UT-A in experimental rat kidney.
UT-A1 immunoreactivity in the IMCD of the IM was markedly decreased in the dehydrated animals but slightly increased in the water-loaded animals (Fig. 3). In control animals, labeling varied substantially from cell to cell within a single tubule; most IMCD cells showed intense immunoreactivity for UT-A1, and a few UT-A1-negative IMCD cells with or without pyknotic nuclei were scattered among the positive cells (Fig. 3D). In the dehydrated animals, however, UT-A1 immunoreactivity was markedly decreased; most of IMCD cells were not immunolabeled for UT-A1, and only a small number of cells showed faint immunoreactivity for UT-A1 (Fig. 3E). In contrast, most IMCD cells had strong UT-A1 immunoreactivity in the water-loaded animals (Fig. 3F).
UT-A2 immunoreactivities in the ISOM and the IMi were significantly increased in the dehydrated animals (Fig. 4B) and slightly decreased in the water-loaded animals compared with controls (Fig. 4C). Double labeling with AQP1 and UT-A2 showed that only a small number of AQP1-positive DTLs in the IMi had UT-A2 immunoreactivity in controls (Fig. 4D). In contrast, in the dehydrated animals, most AQP1-positive DTLs showed UT-A2 immunoreactivity (Fig. 4E). In the water-loaded animals, only a small number of AQP1-positive DTLs had a few UT-A2-positive cells (Fig. 4F).
Higher UT-A3 (Q2) immunoreactivity was observed in the middle and terminal parts of the IM in the dehydrated animals than in the controls (Fig. 5B), and lower immunoreactivity was observed in the water-loaded animals than in controls (Fig. 5C). Like UT-A1, immunoreactivity for UT-A3 varied from cell to cell even in the single tubule. We observed more strong UT-A3-positive IMCD cells in dehydrated animals (Fig. 5E) and fewer in water-loaded animals (Fig. 5F) than in controls (Fig. 5D).
Expression of UT-B in the control rat kidney.
Light microscopy of 50-μm-thick sections showed strong UT-B immunoreactivity in the endothelial cells of the DVR in the medulla, as previously reported (20, 25). UT-B labeling in the DVR was pronounced in the ISOM (Fig. 6, A and D), and the outer third of IM but was much weaker in the papillary tip.
Electron Microscopic Immunocytochemistry
Subcellular localization of UT-A.
To identify the type of epithelium present in the UT-A2-positive tubules in the ISOM and IM, we used electron microscopic immunocytochemistry with a preembedding immunoperoxidase method. In the upper part of the ISOM, we did not observe UT-A immunoreactivity in the type I short-loop DTL nor in the type II long-loop DTL (Fig. 7, A and B). In the lower part of the ISOM, we found strong UT-A2 labeling in the type I epithelium of the short-loop DTL, which is composed of very flat and noninterdigitating cells without microvilli (Fig. 7, B and C). Most type II (Fig. 7B) and III (Fig. 7E) epithelia of the long-loop DTL were not labeled with UT-A2. However, at the lower part of the ISOM and IMi, some DTLs (type II and III epithelia) of the long-loop nephrons were weakly positive for UT-A2 (Fig. 7D). No UT-A2 immunolabeling was observed in the type IV epithelium or in the principal cells of the collecting duct in the IMi (Fig. 7, D and E).
We used immunogold EM to determine the changes in subcellular localization of UT-A and UT-B in the kidneys of rats with different hydration status. In the control, UT-A2 was located mainly in both the apical and basolateral plasma membranes and sparsely in the cytoplasm of the type I epithelium of the short-loop DTL (Fig. 8A). Greater UT-A2 labeling was observed in both the apical and basolateral plasma membranes of the type I epithelium in the dehydrated animals (Fig. 8B), but less labeling was seen in the water-loaded animals (Fig. 8C). Furthermore, in the dehydrated animals, we observed enhanced UT-A2 immunolabeling in the type II epithelium at the lower part of the ISOM and in the type III epithelium at the IMi (Fig. 9). In controls, UT-A2 was not observed in type II and type III epithelia, which were weakly labeled with light microscopic immunohistochemistry and preembedding immunoperoxidase methods, possibly because the UT-A1 protein may be present in insufficient amounts to detect by immunolabeling with immunogold.
In the control animals, UT-A1 was located mainly throughout the cytoplasm of the IMCD cells in the middle and terminal parts of the IMCD (Fig. 10A). We observed a small amount of labeling of UT-A1 on the basolateral plasma membrane (Fig. 10C), but little or no UT-A1 immunolabeling was found on the apical plasma membrane (Fig. 10B). Most cells in the middle and terminal IMCD exhibited strong UT-A1 immunoreactivity, although a small minority of cells showed no detectable UT-A1 immunoreactivity in the control. Cells with no UT-A1 immunoreactivity occasionally appeared to contain dark cytoplasm and pyknotic nuclei (data not shown). In the dehydrated animals, we observed less UT-A1 immunoreactivity in both the cytoplasm and the basolateral plasma membrane (Fig. 11, A–C). In contrast, in the water-loaded animals, we found markedly greater UT-A1 immunoreactivity in both the cytoplasm and the basolateral plasma membrane (Fig. 11, D–F). Despite the upregulation of UT-A1 in the water-loaded animals, no UT-A1 immunolabeling was detected on the apical plasma membrane (Fig. 11E).
In control animals, we observed UT-A3 labeling throughout the cytoplasm of the IMCD cells in the middle and terminal parts of the IMCD (Fig. 12), and a small amount of basolateral labeling for UT-A3. In contrast, we found no labeling of the apical plasma membrane of the IMCD cells (Fig. 12A). UT-A3 labeling varied considerably from cell to cell within a single tubule, similar to the UT-A1 immunolabeling in IMCD cells. A small minority of cells had no detectable UT-A3 immunoreactivity (Fig. 12, B and C). In the dehydrated animals, we found markedly greater UT-A3 immunoreactivity in both the cytoplasm and the basolateral plasma membrane (Fig. 12, D–F). Despite the upregulation of UT-A3 in the dehydrated animals, there was no UT-A3 immunolabeling on the apical plasma membrane (Fig. 12, D and F). In the water-loaded animals, UT-A3 was markedly decreased in both the cytoplasm and the basolateral plasma membrane (Fig. 12G).
Subcellular localization of UT-B.
We confirmed the exact localization and subcellular distribution pattern for UT-B in the rat kidney. EM revealed strong UT-B immunolabeling of both the apical and basolateral plasma membranes and a little labeling of the cytoplasm in the continuous endothelial cells of the DVR. There was no UT-B immunoreactivity in the pericytes, which are embedded in the basement membrane of the DVR (Fig. 13A). In the AVR, we observed no UT-B immunoreactivity in the control or the experimental groups (data not shown). In the experimental animals, UT-B immunolabeling in the endothelial cells of the DVR was markedly greater in the dehydrated animals (Fig. 13B) and lower in the water-loaded animals (Fig. 13C) than in the controls.
Semiquantitative Analysis of Immunoblotting
We used immunoblotting and densitometric techniques to provide a semiquantitative analysis of UT-A1, UT-A2, UT-A3, and UT-B expression in the membrane fraction. The relative optical densities of the UTs bands in each lane were compared with control values. Immunoblotting analysis of UT-A1 showed two bands at 97 and 117 kDa in the IM (Fig. 14A), the intensities of which were significantly lower in the dehydrated animals (39 ± 4 vs. 100 ± 10%, P < 0.05, n = 6) and greater in the water-loaded animals (144 ± 41 vs. 100 ± 10%, P < 0.05, n = 6) than in the controls. In contrast, the band intensities of UT-A2 (245 ± 4 vs. 100 ± 8%, P < 0.05, n = 6) at 55 kDa in the OM were markedly greater in the dehydrated animals and correspondingly lower in the water-loaded animals (70 ± 14 vs. 100 ± 20%, P < 0.05, n = 6) than in the controls (Fig. 14B). UT-A3 expression revealed bands at 44 and 67 kDa, which were more intense in the IM of dehydrated animals (221 ± 6 vs. 100 ± 3%, P < 0.05, n = 6) than in the controls or water-loaded animals (45 ± 2 vs. 100 ± 3%, P < 0.05, n = 6) (Fig. 14C). Similar to expressions found for UT-A2 and UT-A3, UT-B expression was greater in the medulla of dehydrated animals (160 ± 12 vs. 100 ± 6%, P < 0.05, n = 6) and lower in water-loaded animals (36 ± 6 vs. 100 ± 6%, P < 0.05, n = 6) than in the controls (Fig. 14D). These data are consistent with the results of the immunohistochemical studies.
Our studies using the L194 antibody demonstrated that UT-A1 is expressed exclusively in the middle and terminal IMCD in rat kidneys; UT-A1 was localized in the cytoplasm of IMCD cells but showed little labeling of the basolateral plasma membrane. These observations are consistent with results of previous studies using light microscopy and EM (20, 25, 58). Interestingly, we found that the changes in UT-A1 protein abundance with different hydration status were related to changes in the intensity of staining and the number of strongly UT-A1-positive IMCD cells but not to changes in the subcellular distribution. We note that UT-A1 localization varied considerably from cell to cell within a single IMCD.
Nielsen et al. (37) found UT-A1 immunolabeling in the apical plasma membrane and intracellular vesicles in IMCD cells, suggesting that UT-A1-mediated urea transport in the IMCD might be regulated via an apical shuttling mechanism in these cells. We addressed this hypothesis and showed that UT-A1 is expressed mainly in the cytoplasm, with little labeling of the apical plasma membrane of IMCD cells in both the hyper- and hypotonic medulla. Furthermore, previous studies of the effect of vasopressin on UT-A1 suggested that the regulatory trafficking theory of UT-A1 does not occur in response to vasopressin in the rat IMCD (17). Several recent reports have suggested instead that UT-A1 regulation occurs, in part, via an alteration in the phosphorylation of UT-A1 by vasopressin (21, 36, 49, 59, 63). Our data, however, do not rule out a low level of expression of UT-A1 in the apical plasma membrane. A very low level of apical expression may be sufficient to account for apical urea transport because UT-A1 is believed to have channel-like properties with a very high turnover number per channel unit (26). Thus relatively few copies of the UT-A1 molecule may be responsible for a substantial amount of transport.
Previous reports show that UT-A1 protein abundance decreases in response to increased concentrations of vasopressin and aldosterone (8, 13, 22, 51, 55). Light microscopy, EM, and immunoblot analysis showed that expression of UT-A1 in the IMCD cells decreased dramatically in kidneys of dehydrated rats. UT-A1 expression was greater in rats given 3% sucrose in water for 3 days, whereas expression declined in rats given sucrose in water for 14 days (24, 55). These data are consistent with the observations that UT-A1 expression increases in response to a few days of water diuresis, furosemide diuresis, hypercalcemia, low-protein diet, or adrenalectomy, conditions that are associated with reduced urine-concentrating ability and reduced plasma vasopressin concentration (2, 18, 19, 22, 23, 28, 35, 51). However, dehydration or loading for 3 days in rats may stimulate or suppress other regulating mechanisms such as the sympathetic nervous system or renin-angiotensin-aldosterone system, in addition to stimulating or suppressing vasopressin and aldosterone secretion. Although UT-A1 has been studied extensively, further research is needed to fully understand its role.
As reported previously in a detailed study using antiserum L403 by Wade et al. (58), the L194 antibody labeled UT-A2 in the DTL and UT-A1 in the IMCD (58). In our study, UT-A2 immunoreactivity was strong in the terminal portion of the short-loop DTL (type I epithelium) and was faint in the long-loop DTL (type II and III epithelia). EM demonstrated that UT-A2 is expressed in the apical and basolateral plasma membranes and in the cytoplasm of type I and type III epithelia, as previously reported (25). However, the expression of UT-A2 in the long-loop DTL has been somewhat controversial. Although UT-A2 mRNA expression was demonstrated in both the short-loop DTL in the ISOM and the long-loop DTL in the IMi by RT-PCR of microdissected tubules and by in situ hybridization, several studies have not detected any UT immunoreactivity in the long-loop DTL under normal conditions. The failure to detect UT-A2 in this portion is most likely caused by the very low abundance of the protein. However, after stimulation of UT-A2 expression by chronic infusion of vasopressin, UT-A2 labeling was observed in the long-loop DTL of IMi of Brattleboro rats (46, 58). We demonstrated weak UT-A2 labeling in the long-loop DTL in both the ISOM and IMi using an immunoperoxidase method in control animals. Moreover, UT-A2 immunogold labeling appeared on the type II epithelium at the lower part of the ISOM and on type III epithelium in the IMi in dehydrated animals. This is the first observation of UT-A2 immunolabeling in type II and type III epithelia in the rat kidney using an EM immunogold method. Our demonstration of increased UT-A2 immunoreactivity in the DTL of the IMi in dehydrated animals agrees with previous reports of increased expression after vasopressin treatment (43, 46, 47, 58). UT-A2 expression in the type I epithelium of the short-loop DTL adjacent to the vasa recta bundles facilitates the transfer of urea from the AVR to the short-loop DTL. The existence of this pathway allows recycling of urea from the inner medullary interstitium back to the nephron (6, 30, 57). The observed increase in UT-A2 expression in the apical and basolateral plasma membrane of the DTL in the ISOM and the IM during dehydration permits increased urea recycling through the distal nephron and the collecting duct back to the IM. In contrast, UT-A2 expression in the DTL decreased after water loading, which would tend to decrease urea recycling and consequently lead to reduced medullary tonicity and decreased water reabsorption. Therefore, the urine-concentrating ability might be determined in this region by the regulation of UT-A2 expression and urea reabsorption. The mechanism responsible for the changes in UT-A2 expression during different hydration states is not known with certainty. Although UT-A2 transcription is increased by cAMP (33) and UT-A2 expression is stimulated by vasopressin, it is not known whether this is a direct effect. To date, vasopressin receptors have not been detected in the long-loop DTL, whereas vasopressin V1 binding sites have been detected in the short-loop DTL and may play a role in the modulation of UT-A2 levels (1).
The localization of UT-A3 in normal kidneys is controversial. Terris et al. (52) reported that UT-A3 is expressed in the cytoplasm and the apical region of the IMCD. In contrast, Stewart et al. (50) detected UT-A3 immunoreactivity in the basolateral membrane. Both studies were based on light microscopic studies. Thus we performed immunogold methods to localize UT-A3 more precisely. Our data clearly demonstrated that UT-A3 was abundantly labeled in the cytoplasm and was faintly labeled in the basolateral plasma membrane and that UT-A3 expression varied substantially from cell to cell within a single tubule.
We observed greater UT-A3 expression the cytoplasm and basolateral plasma membrane of the IMCD cell in the dehydrated group than in the controls but observed no differences between groups in the subcellular distribution. In contrast, UT-A3 immunoreactivity was lower in water-loaded rats than in controls, despite no changes in subcellular changes and protein abundance in the IM compared with control (Table 2). These results suggest that UT-A3 contributes to basolateral urea permeability in IMCD cells in response to changes in water balance over 3 days.
Interestingly, changes in hydration status cause opposite changes in the expression of UT-A3 and UT-A1, both of which are located in IMCD cells. Both UT-A1 and UT-A3 are controlled by the 5′-promoter of the UT-A gene (promoter 1), which includes a TonE sequence (5, 34). Thus, during dehydration, increased extracellular tonicity activates transcription of UT-A1 and UT-A3 by stimulating promoter 1. However, our results and other reports show that hydration status causes different effects on the expression of UT-A1 and UT-A3 (4, 16). The responsible mechanisms are unknown at present but may involve posttranscriptional regulation of UT-A isoform expression related to hydration status (4); alternatively, the functional, phosphorylated form of UT-A1, although in low abundance, may increase during dehydration but not during water loading (63).
Our observation that UT-B is localized in the continuous endothelium of the DVR is consistent with previous reports in the human (54), mouse (20), and rat kidneys (42, 53). We demonstrated with immunogold cytochemistry that UT-B is localized on both the apical and basolateral plasma membranes of the continuous endothelium of the DVR, suggesting that UT-B is responsible for the high urea permeability demonstrated in microperfused DVR (15). The presence of UT-B in the plasma membrane of the DVR allows urea in the medullary interstitium to enter the DVR. When blood leaves the IM via the AVR, urea can exit these vessels through the UT-B-negative fenestrated endothelium.
The long-term regulation of UT-B has not been as extensively studied as that of UT-A. In our study, UT-B expression increased in dehydrated rats for 3 days, whereas it decreased in water-loaded rats. Consistent with this, UT-B mRNA abundance in the ISOM and IM increased in Brattleboro rats administered vasopressin or 1-deamino, 8D-arginine vasopressin for 5 days (41). Moreover, studies in UT-B knockout mice underlying the regulation of UT-B in different hydration states have demonstrated decreased urinary osmolality and increased blood urea nitrogen due to a decrease in the countercurrent exchange of urea (61). Therefore, the production of maximally concentrated urine during dehydration appears to require UT-B protein expression in the DVR or RBC or both (32). The downregulation of UT-B protein expression has been studied under several conditions associated with reduced urinary-concentrating ability, such as furosemide treatment, nephrectomy, and lithium treatment (15, 27, 42). Overall, UT-B certainly plays an important role in preserving the urine-concentrating ability. However, further research is needed to clearly identify the mechanisms.
In summary, our data show that 3 days of water restriction increased the expression levels of UT-A2, UT-A3, and UT-B and decreased the expression level of UT-A1 without changing the subcellular distribution of these transporters. These observations suggest that multiple factors may be involved in regulating water balance.
This work was supported by the Korea Science and Engineering Foundation (R13-2002-005-01001-0) through the MRC for Cell Death Disease Research Center at the Catholic University of Korea and was presented at the 2002 annual meeting of American Society of Nephrology (Philadelphia, PA).
We gratefully acknowledge the technical assistance of H.-D. Roh, I.-S. Lee, and H.-L. Kim. We thank D.-H. Park in MSD Korea for assistance in preparing this manuscript.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society