Urea transport in the kidney is mediated by a family of transporter proteins that includes renal urea transporters (UT-A) and erythrocyte urea transporters (UT-B). Because newborn rats are not capable of producing concentrated urine, we examined the time of expression and the distribution of UT-A and UT-B in the developing rat kidney by light and electron microscopic immunocytochemistry. Kidneys from 16-, 18-, and 20-day-old fetuses, 1-, 4-, 7-, 14-, and 21-day-old pups, and adult animals were studied. In the adult kidney, UT-A was expressed intensely in the inner medullary collecting duct (IMCD) and terminal portion of the short-loop descending thin limb (DTL) and weakly in long-loop DTL in the outer part of the inner medulla. UT-A immunoreactivity was not present in the fetal kidney but was observed in the IMCD and DTL in 1-day-old pups. The intensity of UT-A immunostaining in the IMCD gradually increased during postnatal development. In 4- and 7-day-old pups, UT-A immunoreactivity was present in the DTL at the border between the outer and inner medulla. In 14- and 21-day-old pups, strong UT-A immunostaining was observed in the terminal part of short-loop DTL in the outer medulla, and weak labeling remained in long-loop DTL descending into the outer part of the inner medulla. In the adult kidney, there was intense staining for UT-B in descending vasa recta (DVR) and weak labeling of glomeruli. In the developing kidney, UT-B was first observed in the DVR of a 20-day-old fetus. After birth there was a striking increase in the number of UT-B-positive DVR, in association with the formation of vascular bundles. The intensity of immunostaining remained strong in the outer medulla but gradually decreased in the inner medulla. We conclude that the expression of urea transporters in short-loop DTL and DVR coincides with the development of the ability to produce a concentrated urine.
- urine concentration
in mammals, urea is the major end product of nitrogen metabolism, and its transport in the kidney is important for the formation of a concentrated urine (1, 9, 13, 23). Urea is a small molecule with a molecular mass of 60 Da. It has a low lipid solubility, and, in the absence of specific transporters, it crosses cell membranes slowly by passive diffusion. In the kidney, the movement of urea through cells of the inner medullary collecting duct (IMCD), descending thin limb (DTL) of Henle's loop, and descending vasa recta (DVR) proceeds by “facilitated” transport mediated by specific transport proteins located in the cell membrane (23). Specific urea transporters are also present in other mammalian cells including erythrocytes and hepatocytes (23).
The urea transporter family includes two main groups, renal urea transporters (UT-A) and erythrocyte urea transporters (UT-B), which are encoded by different genes (23). The cDNA of five isoforms of rat UT-A, UT-A1 (26), UT-A2 (27, 39), UT-A3 (11), UT-A4 (11), and UT-A5 (7) have been cloned. They are thought to be variants of mRNA splicing from a single gene. UT-A1 is the longest isoform, and the others are composed of part of UT-A1 (7, 11). The cDNA of UT-B was first cloned from a human bone marrow library (19) and subsequently isolated from a rat inner medullary library by homology screening (3, 35).
UT-A1–A4 are predominantly expressed in the kidney, whereas UT-A5 is expressed in the testes. UT-A1 and UT-A3 are localized in the IMCD (18, 24, 32), and UT-A2 is located in the DTL of Henle's loop (24, 36). The precise segmental distribution of UT-A4 has not been elucidated. UT-B mRNA and protein are expressed in the endothelial cells of the vasa recta (19, 35, 37).
It is known that newborn rats are not capable of concentrating their urine to adult levels (28). Urine osmolality of neonatal rats rises from 300 mosmol/kgH2O at birth to ∼2,000 mosmol/kgH2O by 3 wk of age (6, 22, 38). The accumulation of urea in the inner medullary interstitium is important for the generation of a concentrated urine. However, little is known about the expression and distribution of urea transporters in the fetal and neonatal kidney.
The purpose of this study was to examine the time of expression and the distribution of UT-A and UT-B in the developing rat kidney by light and electron microscopic immunostaining methods using rabbit polyclonal antibodies to UT-A and UT-B. Because the antibody against UT-A recognizes UT-A1, UT-A2, and UT-A4, it was not possible to distinguish between the different splice variants of UT-A.
Animals and Tissue Preservation
Sprague-Dawley rats were used in all experiments. Prenatal kidneys were obtained from 16-, 18- and 20-day-old fetuses. Postnatal kidneys were obtained the day of birth and from 4-, 7-, 14- and 21-day-old animals. For each age group, animals from two separate litters were used. Kidneys from adult male rats served as a positive reference for the immunohistochemical studies. The kidneys from prenatal and neonatal animals, up to and including those 14 days of age, were preserved by in vivo perfusion through the heart, whereas kidneys from 3-wk-old and adult animals were perfused through the abdominal aorta. The kidneys were initially perfused briefly with PBS to rinse away all blood. This was followed by perfusion with a periodate-lysine-2% paraformaldehyde solution for 10 min. The kidneys were removed and cut into 1- to 2-mm-thick slices that were fixed additionally by immersion in the same fixative for 2 h at room temperature and then overnight at 4°C. Sections of tissue were cut transversely through the entire kidney on a Vibratome (Pelco 101, sectioning series 1000, Technical Products, St. Louis, MO) at a thickness of 50 μm and processed for immunohistochemical studies using the horseradish peroxidase preembedding technique.
To determine the distribution of urea transporters in the developing rat kidney, we used specific rabbit polyclonal antibodies against peptides based on the rat renal urea transporter UT-A (L403; recognizes UT-A1, UT-A2 and UT-A4) (31, 36) and UT-B (33), the human erythrocyte urea transporter. A rabbit polyclonal antibody against aquaporin-1 (AQP1) (30) was used for colocalization with UT-A in the DTL. The antibodies have been characterized in detail in previous studies.
Immunoperoxidase Preembedding Method
Vibratome sections (50 μm thick) were washed with 50 mM NH4Cl in PBS three times for 15 min. Before incubation with the primary antibodies, all tissue sections were incubated for 3 h with PBS containing 1% bovine serum albumin, 0.05% saponin, and 0.2% gelatin (solution A). The tissue sections were then incubated overnight at 4°C in rabbit antisera against UT-A (1:500) and UT-B (1:2,000) in PBS containing 1% bovine serum albumin (solution B). After several washes with solution A, the tissue sections were incubated for 2 h in peroxidase-conjugated donkey anti-rabbit IgG Fab fragment (Jackson ImmunoResearch Laboratories) diluted 1:100 in solution B. The tissues were then rinsed, first in solution A and subsequently in 0.05 M Tris buffer, pH 7.6. For the detection of horseradish peroxidase, the sections were incubated in 0.1% 3,3′-diaminobenzidine in 0.05 M Tris buffer for 5 min, after which H2O2 was added to a final concentration of 0.01% and the incubation was continued for 10 min. After being washed with 0.05 M Tris buffer three times, the sections were dehydrated in a graded series of ethanol and embedded in poly/Bed 812 resin (Polysciences, Warrington, CA). For UT-A and AQP1 double immunostaining, Vibratome sections were labeled with the antibody against UT-A using 3,3′-diaminobenzidine as the chromogen (brown) as described above. The sections were then rinsed with PBS, incubated for 1 h in solution B, and then incubated overnight at 4°C in rabbit antisera against AQP1 (1:500) in solution B. 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) diluted 1:100 insolution B. The tissues were then rinsed in PBS, pH 7.4, and, for the detection of horseradish peroxidase, the sections were incubated in vector SG (blue; Vector Laboratories) in PBS for 5 min, followed by H2O2 incubation for 10 min. After being washed with PBS three times, the sections were dehydrated in a graded series of ethanol and embedded in poly/Bed 812 resin.
For light microscopy, 1-μm sections were cut, the plastic resin was removed by etching with a saturated solution of sodium hydroxide in alcohol, and the sections were stained with hematoxylin. The sections (50 and 1 μm) were examined and photographed with an Olympus photomicroscope equipped with differential-interference contrast optics.
For electron microscopic observation, Vibratome sections were postfixed with 1% glutaraldehyde and 1% osmium tetroxide in 0.1 M phosphate buffer, pH 7.4, before being dehydrated and embedded in poly/Bed 812 resin. Ultrathin sections were stained with uranyl acetate and lead citrate and photographed with a transmission electron microscope (JEOL 1200EX, Tokyo, Japan).
Expression of UT-A
Adult rat kidney.
Light microscopy of 50-μm sections demonstrated strong UT-A immunoreactivity only in the inner stripe of the outer medulla (ISOM) and inner medulla. There was no labeling in the cortex and outer stripe of the outer medulla (Fig. 1). At higher magnification of 1-μm sections, the strongly labeled tubular profiles in the ISOM were identified as the terminal portion of short-loop DTL by the abrupt transition to the unlabeled thick ascending limb (TAL) (Fig. 2 A). In the inner medulla, there was intense UT-A immunostaining of the middle and terminal IMCD but only weak or an absence of labeling in the initial IMCD (Fig. 2, B and C). Weak labeling for UT-A was also observed in the DTL of long-loop nephrons close to the border between the outer and inner medulla (Figs. 1 and 2 B). To confirm that only the DTL of short-loop nephrons are UT-A positive in the ISOM, electron microscopy was carried out. Strong UT-A labeling was located in the apical and basolateral plasma membrane as well as the cytoplasm of the type I epithelium of the short-loop DTL, which is composed of very flat and noninterdigitating cells without apical microvilli (Fig. 3). There was no labeling of the type II epithelium of the long-loop DTL.
Developing rat kidney.
Before birth, there was no UT-A immunoreactivity in the developing uriniferous tubules, including collecting ducts (Fig.4 A). UT-A immunoreactivity was first observed in the terminal part of medullary collecting ducts and in a few DTL in the base of the renal papilla immediately after birth (Fig. 4). During postnatal renal development, the intensity of UT-A immunolabeling gradually increased in both IMCD and DTL (Figs. 4 and5). Labeling was strong in the terminal IMCD and was observed only in the apical region of IMCD cells throughout development.
By postnatal day 7, loops of Henle with the structural characteristics of short-loop nephrons were present at various levels in the renal medulla. UT-A immunolabeling was observed in the terminal part of DTL of immature loops of Henle located in the base of renal papilla, which corresponds to the ISOM of the adult kidney (Fig.5 A). There was no UT-A immunolabeling in the terminal part of the DTL located in the renal papilla, indicating that the UT-A immunoreactivity disappeared as the loops of Henle descended into the papilla. A striking increase in the number of UT-A-labeled DTL and in the intensity of immunostaining was observed in the ISOM in 14-day-old pups (Fig. 5 B). The UT-A-positive DTL in the ISOM belonged to short-loop nephrons. There was no labeling of the inner stripe portion of many of the long-loop DTL. However, weak UT-A immunostaining was now present in some of the long-loop DTL in the inner medulla at the base of the papilla, indicating that some immunoreactivity remained in this part of the DTL as the loops descended from the outer to the inner medulla. In the kidney of 21-day-old pups, UT-A was increased in short-loop DTL in the ISOM and decreased in long-loop DTL in the inner medulla, and the pattern of immunoreactivity was similar to that observed in adults (Fig. 5 C).
UT-A and AQP1 colocalization.
To determine whether UT-A and AQP1 were expressed in the same DTL segments, double-labeling experiments were performed. In adult kidney, AQP1 was expressed in the DTL of both long- and short-loop nephrons. However, there was no AQP1 immunoreactivity in the UT-A-positive part of short-loop DTL in the ISOM. In the inner medulla, however, DTL with weak UT-A immunoreactivity also expressed AQP1 (Fig.6, A–C). In 4-day-old pups, AQP1 was mainly present in DTL of more mature long-loop nephrons, whereas immature long-loop nephrons located in the area corresponding to the future ISOM had AQP1 only in the proximal part and UT-A in the distal part of the DTL (Fig. 6 D). As the DTL of long-loop nephrons descend toward the inner medulla, UT-A immunoreactivity decreased and became restricted to the border between the outer and inner medulla in the adult kidney whereas AQP1 immunoreactivity increased (Fig. 6, E and F).
Expression of UT-B
Adult rat kidney.
Light microscopy of 50-μm-thick sections demonstrated that UT-B was expressed in DVR and glomeruli (Fig. 7). UT-B labeling in DVR was pronounced in the ISOM and proximal part of inner medulla but was much weaker in the papillary tip. Electron microscopy revealed strong UT-B immunolabeling in the apical membrane and diffuse staining of the cytoplasm in the continuous endothelial cells of DVR (Fig. 8). There was no UT-B immunoreactivity in the pericytes, embedded in the basement membrane of DVR, or in the fenestrated endothelial cells of ascending vasa recta (Fig. 8).
Developing rat kidney.
During early postnatal stages (1-, 4-, and 7-day-old pups), UT-B was strongly expressed in DVR, and there was a gradual increase not only in the number of UT-B-labeled DVR but also in the intensity of UT-B immunoreactivity during renal development (Figs. 9 and 10). In 14-day-old pups (Fig. 9), UT-B-positive DVR formed vascular bundles in the ISOM, and in 21-day-old pups the pattern and intensity of UT-B immunostaining was similar to those observed in adult animals. There was no labeling in developing uriniferous tubules including collecting ducts.
The present study represents the first detailed description of the expression and distribution of urea transporters in the developing kidney. Urea transport in the kidney and the accumulation of urea in the renal medulla are of critical importance for the urinary concentrating process and the regulation of water excretion (1,13, 14, 23). Newborn rats are unable to produce a concentrated urine but develop this capacity during the first 2–3 wk after birth (4, 22, 34, 38). We therefore determined the time of expression of urea transporters and their pattern of distribution in the developing rat kidney.
Our results demonstrate that UT-B immunoreactivity appears in DVR shortly before birth and gradually increases during the first 2–3 wk after birth. It is noteworthy that the UT-B-positive DVR do not form vascular bundles until the animals are 2 wk old. UT-A first appears in the IMCD and in developing long-loop DTL at the base of the renal papilla right after birth. As the UT-A-positive long-loop DTL descend into the renal papilla, UT-A immunoreactivity decreases and is barely detectable in animals after 3 wk of age. In contrast, the expression of UT-A in short-loop DTL does not occur until 2 wk after birth, the time they reach the ISOM, but the level of expression is very strong and remains so in the adult kidney (Fig.11). UT-A immunoreactivity in the IMCD gradually increased after birth and reached adult levels when the animals were 3 wk of age.
The antibody against UT-A used in this study was raised against a COOH-terminal peptide sequence common to UT-A1, UT-A2, and UT-A4 (31, 36). As reported previously in a detailed study by Wade et al. (36), this antibody labels UT-A1 in the IMCD and UT-A2 in the DTL. The very strong labeling of the IMCD and short-loop DTL in the ISOM observed in our study is similar to the results reported by Wade et al. (36) and also confirms previous observations by Nielsen et al. (18). However, the expression of UT-A2 protein in long-loop DTL has been somewhat controversial. Although UT-A2 mRNA expression was demonstrated in both short-loop DTL in the ISOM and long-loop DTL in the initial inner medulla by RT-PCR of microdissected tubules and by in situ hybridization, previous immunohistochemical studies did not detect any UT-A immunoreactivity in long-loop DTL under normal conditions. The failure to detect UT-A in long-loop DTL is most likely due to the very low abundance of the protein. However, after stimulation of UT-A2 expression by chronic infusion of vasopressin, Wade et al. (36) and Shayakul et al. (25) demonstrated labeling also in long-loop DTL in the initial part of the inner medulla of Brattleboro rats. The high sensitivity of the preembedding method used in our study enabled us to detect UT-A immunoreactivity in long-loop DTL in the base of the renal papilla of normal animals, in a pattern similar to that observed in vasopressin-treated Brattleboro rats (25, 36). The differences in UT-A immunoreactivity along the DTL and the absence of UT-A in the type II epithelium of long-loop DTL in the ISOM, which express high levels of AQP1, are in agreement with results of transport studies demonstrating significant differences in isolated perfused segments of long-loop DTL from chinchilas (2) and hamsters (10). Interestingly, in the developing kidney, the DTL of both long- and short-loop nephrons exhibit strong UT-A immunoreactivity in an area at the base of the renal papilla.
The gradual increase in the expression of UT-A in the IMCD after birth is in agreement with the results of a recent study by Liu et al. (15) demonstrating an increase in urea permeability in the IMCD during postnatal development. Surprisingly, these investigators did not detect UT-A1 mRNA in the IMCD until animals were 14 days of age, although there was an exponential increase in urea permeability from postnatal day 1 until postnatal day 14. These observations raise the possibility that another urea transporter, possibly UT-A4, may be responsible for the UT-A immunoreactivity and urea permeability demonstrated in the neonatal IMCD.
The loops of Henle from juxtamedullary and superficial nephrons descend through the renal medulla at different times during kidney development, from before birth until 2–3 wk after birth (Fig. 11). Loops from juxtamedullary nephrons descend first to form the long loops of Henle, and those from the more superficial nephrons descend subsequently to form the short loops of Henle. Our results demonstrate that UT-A appears in the DTL of both long- and short-loop nephrons when they enter the area corresponding to the future ISOM, located at the border between the outer and inner medulla. As the long loops from juxtamedullary nephrons continue their descent into the inner medulla, UT-A immunoreactivity gradually decreases and finally disappears from long-loop DTL, except for those segments located in the initial part of the inner medulla adjacent to the ISOM. These observations suggest that local factors in the area corresponding to the ISOM play a role in the induction of UT-A2 in DTL in the developing kidney.
There is increasing evidence that urea transporters belonging to the UT-A family are regulated by vasopressin as well as changes in osmolality (17, 21, 23, 25, 36). Recent studies have demonstrated that vasopressin stimulates UT-A2 expression in both long- and short-loop DTL (25, 36). Whether this represents a direct effect of vasopressin on DTL or is secondary to activation of vasopressin receptors in adjacent structures has not been established. There is no evidence so far that vasopressin receptors are expressed in DTL (8, 29). However, V2 vasopressin receptors are expressed in the adjacent TAL (8, 29). Vasopressin is known to stimulate the reabsorption of sodium in the TAL, which leads to increased sodium content in the medullary interstitium and increased interstitial osmolality. Thus it is possible that an increase in interstitial sodium and/or osmolality might play a role in the induction of UT-A2 in the DTL, in particular during development of the renal medulla, when there is a close association between the DTL and the TAL. In this regard, it is noteworthy that at birth all loops of Henle have the structural characteristics of short loops. The DTL continue directly into the TAL, and there are no ascending thin limbs (12). Thus TAL are present throughout the renal papilla in close proximity to the DTL. During the first 2–3 wk of life, cells are deleted by apoptosis from the TAL in the inner medulla and the remaining TAL cells are tranformed into ascending thin limb cells (12). Thus the close association between the DTL and the TAL is lost in the inner medulla. Interestingly, at the same time the expression of UT-A2 decreases or disappears from long-loop DTL in the inner medulla, with the exception of those located close to the ISOM. This observation suggests that the close association with the TAL might be important for the expression of UT-A2.
A recent study by Nakayama et al. (16) demonstrated consensus sequences for the cAMP response element in promoter II of the UT-A2 gene, suggesting that cAMP might play a role in the regulation of UT-A2. In this regard, it is noteworthy that a gradual increase in osmolality has been shown to have a biphasic effect on adenylate cyclase in the papillary collecting duct, increasing adenylate cyclase activity at 800 mosM and inhibiting the activity at 2,000 mosM (5). An increase in osmolality also inhibited cAMP-phosphodiesterase activity (5). Whether changes in osmolality have similar effects on cAMP metabolism in the DTL remains to be established.
The antibody against UT-B used in this study was characterized in detail recently by Timmer et al. (33). As demonstrated by those investigators and confirmed by us, the expression of UT-B in the kidney was restricted to the continuous endothelium of the DVR in the outer and inner medulla, and there was no labeling of the fenestrated endothelium of the ascending vasa recta. These observations are also in agreement with results of previous in situ hybridization and immunohistochemical studies demonstrating expression of UT-B in vasa recta in both outer and inner medulla of the rat kidney (35,37). The results of these studies indicate that UT-B is responsible for the high urea permeability demonstrated in microperfused DVR (20). 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 inner medulla via ascending vasa recta, urea can exit these vessels through the UT-B-negative fenestrated endothelium.
Our studies in the developing kidney revealed that UT-B is expressed in the DVR already before birth and gradually increases during the first 2 wk after birth, indicating that countercurrent exchange between ascending vasa recta and DVR can occur already in the neonatal kidney. Interestingly, the formation of vascular bundles coincided with the appearance of UT-A2-positive short-loop DTL in the ISOM and was followed by a striking increase in the expression of UT-A2 and UT-B and a clear delineation of the ISOM. We conclude that the expression of urea transporters in the neonatal rat renal medulla during the first 2 wk after birth coincides with the development of the ability to produce a concentrated urine. The striking increase in UT-A2 and UT-B immunoreactivity 2 wk after birth is closely associated with the descent of short-loop DTL from superficial nephrons and the formation of vascular bundles in the ISOM.
We gratefully acknowledge the technical assistance of Hee-Duk Rho and Kyung-A Ryu.
These studies were supported by Korea Research Foundation Grant KRF-98–005-F00127. They were presented in part at the annual meeting of the American Society of Nephrology held in Miami, FL, in November 1999 and have been published in abstract form (J Am Soc Nephrol 10: 406A, 1999).
Address for reprint requests and other correspondence: J. Kim, Dept. of Anatomy, Catholic University Medical College, 505 Banpo-Dong, Socho-Ku, Seoul 137-701, Korea (E-mail:).
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 © 2002 the American Physiological Society